nohup: ignoring input
[2022-12-20 17:18:11,002] [WARNING] [runner.py:179:fetch_hostfile] Unable to find hostfile, will proceed with training with local resources only.
[2022-12-20 17:18:11,035] [INFO] [runner.py:508:main] cmd = /usr/bin/python3 -u -m deepspeed.launcher.launch --world_info=eyJsb2NhbGhvc3QiOiBbMCwgMSwgMiwgM119 --master_addr=127.0.0.1 --master_port=29500 tune_gpt.py --deepspeed deepspeed.json
[2022-12-20 17:18:13,299] [INFO] [launch.py:142:main] WORLD INFO DICT: {'localhost': [0, 1, 2, 3]}
[2022-12-20 17:18:13,299] [INFO] [launch.py:148:main] nnodes=1, num_local_procs=4, node_rank=0
[2022-12-20 17:18:13,299] [INFO] [launch.py:161:main] global_rank_mapping=defaultdict(<class 'list'>, {'localhost': [0, 1, 2, 3]})
[2022-12-20 17:18:13,299] [INFO] [launch.py:162:main] dist_world_size=4
[2022-12-20 17:18:13,299] [INFO] [launch.py:164:main] Setting CUDA_VISIBLE_DEVICES=0,1,2,3

Downloading:   0%|          | 0.00/560 [00:00<?, ?B/s]
Downloading: 100%|██████████| 560/560 [00:00<00:00, 373kB/s]

Downloading:   0%|          | 0.00/1.01k [00:00<?, ?B/s]
Downloading: 100%|██████████| 1.01k/1.01k [00:00<00:00, 662kB/s]

Downloading:   0%|          | 0.00/899k [00:00<?, ?B/s]
Downloading: 100%|██████████| 899k/899k [00:00<00:00, 9.63MB/s]

Downloading:   0%|          | 0.00/456k [00:00<?, ?B/s]
Downloading: 100%|██████████| 456k/456k [00:00<00:00, 6.68MB/s]

Downloading:   0%|          | 0.00/357 [00:00<?, ?B/s]
Downloading: 100%|██████████| 357/357 [00:00<00:00, 243kB/s]

Downloading:   0%|          | 0.00/526M [00:00<?, ?B/s]
Downloading:   1%|          | 3.11M/526M [00:00<00:16, 31.1MB/s]
Downloading:   2%|▏         | 8.22M/526M [00:00<00:12, 42.9MB/s]
Downloading:   3%|▎         | 13.4M/526M [00:00<00:10, 47.1MB/s]
Downloading:   4%|▎         | 18.6M/526M [00:00<00:10, 49.1MB/s]
Downloading:   5%|▍         | 23.9M/526M [00:00<00:09, 50.3MB/s]
Downloading:   6%|▌         | 29.1M/526M [00:00<00:09, 50.9MB/s]
Downloading:   7%|▋         | 34.3M/526M [00:00<00:09, 51.2MB/s]
Downloading:   8%|▊         | 40.0M/526M [00:00<00:09, 53.1MB/s]
Downloading:   9%|▉         | 46.8M/526M [00:00<00:08, 57.7MB/s]
Downloading:  10%|█         | 53.4M/526M [00:01<00:07, 60.5MB/s]
Downloading:  11%|█▏        | 60.0M/526M [00:01<00:07, 62.1MB/s]
Downloading:  13%|█▎        | 66.2M/526M [00:01<00:07, 59.8MB/s]
Downloading:  14%|█▎        | 72.2M/526M [00:01<00:07, 57.0MB/s]
Downloading:  15%|█▍        | 77.9M/526M [00:01<00:07, 56.2MB/s]
Downloading:  16%|█▌        | 83.6M/526M [00:01<00:07, 55.7MB/s]
Downloading:  17%|█▋        | 89.2M/526M [00:01<00:07, 55.1MB/s]
Downloading:  18%|█▊        | 95.2M/526M [00:01<00:07, 56.5MB/s]
Downloading:  19%|█▉        | 102M/526M [00:01<00:06, 60.8MB/s] 
Downloading:  21%|██        | 110M/526M [00:01<00:06, 64.9MB/s]
Downloading:  22%|██▏       | 117M/526M [00:02<00:06, 67.6MB/s]
Downloading:  24%|██▎       | 124M/526M [00:02<00:05, 68.8MB/s]
Downloading:  25%|██▌       | 132M/526M [00:02<00:05, 70.1MB/s]
Downloading:  26%|██▋       | 139M/526M [00:02<00:05, 71.2MB/s]
Downloading:  28%|██▊       | 146M/526M [00:02<00:05, 71.9MB/s]
Downloading:  29%|██▉       | 154M/526M [00:02<00:05, 72.3MB/s]
Downloading:  31%|███       | 161M/526M [00:02<00:05, 71.7MB/s]
Downloading:  32%|███▏      | 168M/526M [00:02<00:04, 72.3MB/s]
Downloading:  33%|███▎      | 176M/526M [00:02<00:04, 73.0MB/s]
Downloading:  35%|███▍      | 183M/526M [00:02<00:04, 73.5MB/s]
Downloading:  36%|███▋      | 191M/526M [00:03<00:04, 74.0MB/s]
Downloading:  38%|███▊      | 198M/526M [00:03<00:04, 71.7MB/s]
Downloading:  39%|███▉      | 205M/526M [00:03<00:04, 68.2MB/s]
Downloading:  40%|████      | 212M/526M [00:03<00:04, 68.7MB/s]
Downloading:  42%|████▏     | 219M/526M [00:03<00:04, 69.3MB/s]
Downloading:  43%|████▎     | 226M/526M [00:03<00:04, 69.4MB/s]
Downloading:  44%|████▍     | 233M/526M [00:03<00:04, 69.8MB/s]
Downloading:  46%|████▌     | 241M/526M [00:03<00:04, 70.0MB/s]
Downloading:  47%|████▋     | 248M/526M [00:03<00:04, 65.7MB/s]
Downloading:  48%|████▊     | 254M/526M [00:03<00:04, 66.8MB/s]
Downloading:  50%|████▉     | 261M/526M [00:04<00:03, 67.2MB/s]
Downloading:  51%|█████     | 268M/526M [00:04<00:03, 67.9MB/s]
Downloading:  52%|█████▏    | 275M/526M [00:04<00:03, 68.7MB/s]
Downloading:  54%|█████▎    | 282M/526M [00:04<00:03, 68.6MB/s]
Downloading:  55%|█████▌    | 289M/526M [00:04<00:03, 69.5MB/s]
Downloading:  56%|█████▋    | 296M/526M [00:04<00:03, 63.1MB/s]
Downloading:  58%|█████▊    | 303M/526M [00:04<00:03, 65.3MB/s]
Downloading:  59%|█████▉    | 311M/526M [00:04<00:03, 67.0MB/s]
Downloading:  60%|██████    | 318M/526M [00:04<00:03, 68.3MB/s]
Downloading:  62%|██████▏   | 325M/526M [00:05<00:02, 68.9MB/s]
Downloading:  63%|██████▎   | 332M/526M [00:05<00:02, 68.8MB/s]
Downloading:  64%|██████▍   | 339M/526M [00:05<00:02, 69.5MB/s]
Downloading:  66%|██████▌   | 346M/526M [00:05<00:02, 70.1MB/s]
Downloading:  67%|██████▋   | 353M/526M [00:05<00:02, 68.0MB/s]
Downloading:  68%|██████▊   | 360M/526M [00:05<00:02, 62.8MB/s]
Downloading:  70%|██████▉   | 367M/526M [00:05<00:02, 64.9MB/s]
Downloading:  71%|███████   | 374M/526M [00:05<00:02, 66.6MB/s]
Downloading:  72%|███████▏  | 381M/526M [00:05<00:02, 68.2MB/s]
Downloading:  74%|███████▍  | 388M/526M [00:05<00:01, 69.3MB/s]
Downloading:  75%|███████▌  | 395M/526M [00:06<00:01, 69.7MB/s]
Downloading:  77%|███████▋  | 402M/526M [00:06<00:01, 69.3MB/s]
Downloading:  78%|███████▊  | 409M/526M [00:06<00:01, 68.6MB/s]
Downloading:  79%|███████▉  | 416M/526M [00:06<00:01, 67.3MB/s]
Downloading:  80%|████████  | 423M/526M [00:06<00:01, 68.2MB/s]
Downloading:  82%|████████▏ | 430M/526M [00:06<00:01, 68.2MB/s]
Downloading:  83%|████████▎ | 437M/526M [00:06<00:01, 68.9MB/s]
Downloading:  84%|████████▍ | 444M/526M [00:06<00:01, 69.4MB/s]
Downloading:  86%|████████▌ | 451M/526M [00:06<00:01, 69.6MB/s]
Downloading:  87%|████████▋ | 458M/526M [00:06<00:00, 69.9MB/s]
Downloading:  88%|████████▊ | 465M/526M [00:07<00:00, 70.0MB/s]
Downloading:  90%|████████▉ | 472M/526M [00:07<00:00, 69.9MB/s]
Downloading:  91%|█████████ | 480M/526M [00:07<00:00, 70.5MB/s]
Downloading:  93%|█████████▎| 487M/526M [00:07<00:00, 70.9MB/s]
Downloading:  94%|█████████▍| 494M/526M [00:07<00:00, 70.8MB/s]
Downloading:  95%|█████████▌| 501M/526M [00:07<00:00, 70.4MB/s]
Downloading:  97%|█████████▋| 508M/526M [00:07<00:00, 70.6MB/s]
Downloading:  98%|█████████▊| 515M/526M [00:07<00:00, 71.0MB/s]
Downloading:  99%|█████████▉| 522M/526M [00:07<00:00, 71.2MB/s]
Downloading: 100%|██████████| 526M/526M [00:07<00:00, 66.3MB/s]

Downloading builder script:   0%|          | 0.00/4.95k [00:00<?, ?B/s]
Downloading builder script: 100%|██████████| 4.95k/4.95k [00:00<00:00, 2.70MB/s]

Downloading builder script:   0%|          | 0.00/4.95k [00:00<?, ?B/s]
Downloading builder script: 100%|██████████| 4.95k/4.95k [00:00<00:00, 1.90MB/s]

Downloading readme:   0%|          | 0.00/5.63k [00:00<?, ?B/s]
Downloading readme: 100%|██████████| 5.63k/5.63k [00:00<00:00, 3.14MB/s]
No config specified, defaulting to: apps/all
No config specified, defaulting to: apps/all
Downloading and preparing dataset apps/all to /home/user/.cache/huggingface/datasets/codeparrot___apps/all/0.0.0/04ac807715d07d6e5cc580f59cdc8213cd7dc4529d0bb819cca72c9f8e8c1aa5...
No config specified, defaulting to: apps/all

Downloading readme:   0%|          | 0.00/5.63k [00:00<?, ?B/s]
Downloading readme: 100%|██████████| 5.63k/5.63k [00:00<00:00, 3.23MB/s]
No config specified, defaulting to: apps/all

Downloading data files:   0%|          | 0/2 [00:00<?, ?it/s]

Downloading data:   0%|          | 0.00/107M [00:00<?, ?B/s]

Downloading data:   3%|▎         | 3.31M/107M [00:00<00:03, 33.1MB/s]

Downloading data:   8%|▊         | 8.95M/107M [00:00<00:02, 46.8MB/s]

Downloading data:  14%|█▎        | 14.7M/107M [00:00<00:01, 51.6MB/s]

Downloading data:  19%|█▉        | 20.4M/107M [00:00<00:01, 53.9MB/s]

Downloading data:  24%|██▍       | 26.2M/107M [00:00<00:01, 55.2MB/s]

Downloading data:  30%|██▉       | 31.9M/107M [00:00<00:01, 55.9MB/s]

Downloading data:  35%|███▌      | 37.6M/107M [00:00<00:01, 56.4MB/s]

Downloading data:  42%|████▏     | 44.6M/107M [00:00<00:01, 60.5MB/s]

Downloading data:  48%|████▊     | 51.8M/107M [00:00<00:00, 64.3MB/s]

Downloading data:  55%|█████▌    | 59.1M/107M [00:01<00:00, 66.8MB/s]

Downloading data:  62%|██████▏   | 66.3M/107M [00:01<00:00, 68.6MB/s]

Downloading data:  69%|██████▊   | 73.5M/107M [00:01<00:00, 69.6MB/s]

Downloading data:  75%|███████▌  | 80.5M/107M [00:01<00:00, 68.3MB/s]

Downloading data:  82%|████████▏ | 87.8M/107M [00:01<00:00, 69.5MB/s]

Downloading data:  89%|████████▊ | 95.0M/107M [00:01<00:00, 69.9MB/s]

Downloading data:  95%|█████████▌| 102M/107M [00:01<00:00, 63.9MB/s] 
Downloading data: 100%|██████████| 107M/107M [00:01<00:00, 62.6MB/s]

Downloading data files:  50%|█████     | 1/2 [00:02<00:02,  2.02s/it]

Downloading data:   0%|          | 0.00/1.29G [00:00<?, ?B/s]

Downloading data:   0%|          | 3.30M/1.29G [00:00<00:39, 33.0MB/s]

Downloading data:   1%|          | 9.89M/1.29G [00:00<00:24, 52.3MB/s]

Downloading data:   1%|▏         | 17.1M/1.29G [00:00<00:20, 61.2MB/s]

Downloading data:   2%|▏         | 24.2M/1.29G [00:00<00:19, 65.1MB/s]

Downloading data:   2%|▏         | 31.4M/1.29G [00:00<00:18, 67.7MB/s]

Downloading data:   3%|▎         | 38.6M/1.29G [00:00<00:18, 69.2MB/s]

Downloading data:   4%|▎         | 45.9M/1.29G [00:00<00:17, 70.2MB/s]

Downloading data:   4%|▍         | 53.1M/1.29G [00:00<00:17, 70.9MB/s]

Downloading data:   5%|▍         | 60.2M/1.29G [00:00<00:17, 70.8MB/s]

Downloading data:   5%|▌         | 67.3M/1.29G [00:01<00:17, 70.9MB/s]

Downloading data:   6%|▌         | 74.5M/1.29G [00:01<00:17, 71.3MB/s]

Downloading data:   6%|▋         | 81.8M/1.29G [00:01<00:16, 71.7MB/s]

Downloading data:   7%|▋         | 89.1M/1.29G [00:01<00:16, 72.1MB/s]

Downloading data:   7%|▋         | 96.3M/1.29G [00:01<00:17, 69.5MB/s]

Downloading data:   8%|▊         | 103M/1.29G [00:01<00:21, 56.0MB/s] 

Downloading data:   9%|▊         | 110M/1.29G [00:01<00:19, 59.7MB/s]

Downloading data:   9%|▉         | 117M/1.29G [00:01<00:22, 52.5MB/s]

Downloading data:  10%|▉         | 124M/1.29G [00:01<00:20, 56.5MB/s]

Downloading data:  10%|█         | 130M/1.29G [00:02<00:19, 59.5MB/s]

Downloading data:  11%|█         | 137M/1.29G [00:02<00:18, 62.3MB/s]

Downloading data:  11%|█         | 144M/1.29G [00:02<00:18, 63.3MB/s]

Downloading data:  12%|█▏        | 151M/1.29G [00:02<00:17, 64.6MB/s]

Downloading data:  12%|█▏        | 158M/1.29G [00:02<00:17, 66.1MB/s]

Downloading data:  13%|█▎        | 165M/1.29G [00:02<00:16, 68.0MB/s]

Downloading data:  13%|█▎        | 172M/1.29G [00:02<00:16, 69.1MB/s]

Downloading data:  14%|█▍        | 179M/1.29G [00:02<00:16, 68.6MB/s]

Downloading data:  14%|█▍        | 186M/1.29G [00:02<00:19, 57.3MB/s]

Downloading data:  15%|█▍        | 193M/1.29G [00:03<00:18, 60.2MB/s]

Downloading data:  15%|█▌        | 200M/1.29G [00:03<00:17, 62.4MB/s]

Downloading data:  16%|█▌        | 207M/1.29G [00:03<00:16, 64.4MB/s]

Downloading data:  16%|█▋        | 213M/1.29G [00:03<00:16, 64.9MB/s]

Downloading data:  17%|█▋        | 220M/1.29G [00:03<00:20, 52.4MB/s]

Downloading data:  17%|█▋        | 226M/1.29G [00:03<00:19, 53.5MB/s]

Downloading data:  18%|█▊        | 232M/1.29G [00:03<00:18, 57.3MB/s]

Downloading data:  19%|█▊        | 239M/1.29G [00:03<00:17, 60.6MB/s]

Downloading data:  19%|█▉        | 246M/1.29G [00:03<00:16, 62.9MB/s]

Downloading data:  20%|█▉        | 253M/1.29G [00:04<00:15, 65.0MB/s]

Downloading data:  20%|██        | 260M/1.29G [00:04<00:15, 66.4MB/s]

Downloading data:  21%|██        | 267M/1.29G [00:04<00:15, 67.4MB/s]

Downloading data:  21%|██        | 274M/1.29G [00:04<00:14, 68.1MB/s]

Downloading data:  22%|██▏       | 281M/1.29G [00:04<00:14, 68.6MB/s]

Downloading data:  22%|██▏       | 288M/1.29G [00:04<00:14, 69.5MB/s]

Downloading data:  23%|██▎       | 295M/1.29G [00:04<00:14, 70.0MB/s]

Downloading data:  23%|██▎       | 302M/1.29G [00:04<00:14, 70.0MB/s]

Downloading data:  24%|██▍       | 309M/1.29G [00:04<00:14, 70.0MB/s]

Downloading data:  24%|██▍       | 316M/1.29G [00:04<00:13, 69.9MB/s]

Downloading data:  25%|██▌       | 323M/1.29G [00:05<00:13, 69.8MB/s]

Downloading data:  26%|██▌       | 330M/1.29G [00:05<00:14, 68.2MB/s]

Downloading data:  26%|██▌       | 337M/1.29G [00:05<00:14, 67.6MB/s]

Downloading data:  27%|██▋       | 344M/1.29G [00:05<00:13, 68.0MB/s]

Downloading data:  27%|██▋       | 351M/1.29G [00:05<00:13, 68.9MB/s]

Downloading data:  28%|██▊       | 358M/1.29G [00:05<00:13, 69.2MB/s]

Downloading data:  28%|██▊       | 365M/1.29G [00:05<00:13, 69.8MB/s]

Downloading data:  29%|██▉       | 372M/1.29G [00:05<00:13, 70.1MB/s]

Downloading data:  29%|██▉       | 380M/1.29G [00:05<00:12, 71.1MB/s]

Downloading data:  30%|██▉       | 387M/1.29G [00:05<00:12, 71.1MB/s]

Downloading data:  30%|███       | 394M/1.29G [00:06<00:12, 70.6MB/s]

Downloading data:  31%|███       | 401M/1.29G [00:06<00:12, 70.6MB/s]

Downloading data:  32%|███▏      | 408M/1.29G [00:06<00:17, 50.6MB/s]

Downloading data:  32%|███▏      | 414M/1.29G [00:06<00:16, 53.2MB/s]

Downloading data:  33%|███▎      | 420M/1.29G [00:06<00:16, 53.9MB/s]

Downloading data:  33%|███▎      | 426M/1.29G [00:06<00:17, 49.0MB/s]

Downloading data:  33%|███▎      | 431M/1.29G [00:06<00:18, 46.4MB/s]

Downloading data:  34%|███▍      | 438M/1.29G [00:06<00:16, 50.7MB/s]

Downloading data:  34%|███▍      | 444M/1.29G [00:07<00:15, 54.3MB/s]

Downloading data:  35%|███▍      | 450M/1.29G [00:07<00:18, 45.6MB/s]

Downloading data:  35%|███▌      | 457M/1.29G [00:07<00:16, 51.8MB/s]

Downloading data:  36%|███▌      | 464M/1.29G [00:07<00:14, 57.3MB/s]

Downloading data:  36%|███▋      | 471M/1.29G [00:07<00:13, 60.8MB/s]

Downloading data:  37%|███▋      | 479M/1.29G [00:07<00:12, 64.4MB/s]

Downloading data:  38%|███▊      | 486M/1.29G [00:07<00:11, 67.2MB/s]

Downloading data:  38%|███▊      | 494M/1.29G [00:07<00:11, 69.6MB/s]

Downloading data:  39%|███▉      | 501M/1.29G [00:07<00:11, 70.8MB/s]

Downloading data:  39%|███▉      | 508M/1.29G [00:08<00:10, 71.7MB/s]

Downloading data:  40%|███▉      | 516M/1.29G [00:08<00:11, 69.9MB/s]

Downloading data:  40%|████      | 523M/1.29G [00:08<00:11, 67.5MB/s]

Downloading data:  41%|████      | 530M/1.29G [00:08<00:11, 68.2MB/s]

Downloading data:  42%|████▏     | 537M/1.29G [00:08<00:10, 69.1MB/s]

Downloading data:  42%|████▏     | 544M/1.29G [00:08<00:11, 65.1MB/s]

Downloading data:  43%|████▎     | 550M/1.29G [00:08<00:11, 64.1MB/s]

Downloading data:  43%|████▎     | 557M/1.29G [00:08<00:11, 64.3MB/s]

Downloading data:  44%|████▎     | 564M/1.29G [00:08<00:11, 64.8MB/s]

Downloading data:  44%|████▍     | 570M/1.29G [00:08<00:11, 64.3MB/s]

Downloading data:  45%|████▍     | 577M/1.29G [00:09<00:11, 64.6MB/s]

Downloading data:  45%|████▌     | 583M/1.29G [00:09<00:10, 65.1MB/s]

Downloading data:  46%|████▌     | 590M/1.29G [00:09<00:10, 65.7MB/s]

Downloading data:  46%|████▌     | 596M/1.29G [00:09<00:10, 65.2MB/s]

Downloading data:  47%|████▋     | 604M/1.29G [00:09<00:10, 67.6MB/s]

Downloading data:  47%|████▋     | 611M/1.29G [00:09<00:10, 63.6MB/s]

Downloading data:  48%|████▊     | 617M/1.29G [00:09<00:10, 63.1MB/s]

Downloading data:  48%|████▊     | 623M/1.29G [00:09<00:12, 54.4MB/s]

Downloading data:  49%|████▉     | 630M/1.29G [00:09<00:11, 58.2MB/s]

Downloading data:  49%|████▉     | 637M/1.29G [00:10<00:10, 62.2MB/s]

Downloading data:  50%|████▉     | 645M/1.29G [00:10<00:09, 64.9MB/s]

Downloading data:  50%|█████     | 651M/1.29G [00:10<00:10, 64.0MB/s]

Downloading data:  51%|█████     | 658M/1.29G [00:10<00:09, 64.2MB/s]

Downloading data:  51%|█████▏    | 665M/1.29G [00:10<00:09, 65.3MB/s]

Downloading data:  52%|█████▏    | 671M/1.29G [00:10<00:10, 59.6MB/s]

Downloading data:  52%|█████▏    | 677M/1.29G [00:10<00:12, 50.0MB/s]

Downloading data:  53%|█████▎    | 684M/1.29G [00:10<00:11, 54.6MB/s]

Downloading data:  53%|█████▎    | 691M/1.29G [00:10<00:10, 59.3MB/s]

Downloading data:  54%|█████▍    | 698M/1.29G [00:11<00:09, 62.3MB/s]

Downloading data:  55%|█████▍    | 705M/1.29G [00:11<00:09, 64.5MB/s]

Downloading data:  55%|█████▌    | 713M/1.29G [00:11<00:08, 66.4MB/s]

Downloading data:  56%|█████▌    | 719M/1.29G [00:11<00:08, 67.1MB/s]

Downloading data:  56%|█████▌    | 726M/1.29G [00:11<00:08, 67.8MB/s]

Downloading data:  57%|█████▋    | 733M/1.29G [00:11<00:09, 61.2MB/s]

Downloading data:  57%|█████▋    | 740M/1.29G [00:11<00:08, 64.0MB/s]

Downloading data:  58%|█████▊    | 748M/1.29G [00:11<00:08, 66.2MB/s]

Downloading data:  58%|█████▊    | 754M/1.29G [00:11<00:08, 62.6MB/s]

Downloading data:  59%|█████▉    | 761M/1.29G [00:12<00:08, 64.2MB/s]

Downloading data:  59%|█████▉    | 768M/1.29G [00:12<00:07, 66.4MB/s]

Downloading data:  60%|█████▉    | 775M/1.29G [00:12<00:08, 57.7MB/s]

Downloading data:  60%|██████    | 781M/1.29G [00:12<00:08, 57.8MB/s]

Downloading data:  61%|██████    | 788M/1.29G [00:12<00:08, 60.6MB/s]

Downloading data:  61%|██████▏   | 794M/1.29G [00:12<00:08, 57.9MB/s]

Downloading data:  62%|██████▏   | 801M/1.29G [00:12<00:07, 61.9MB/s]

Downloading data:  62%|██████▏   | 808M/1.29G [00:12<00:08, 57.7MB/s]

Downloading data:  63%|██████▎   | 815M/1.29G [00:12<00:07, 61.6MB/s]

Downloading data:  64%|██████▎   | 822M/1.29G [00:13<00:07, 64.4MB/s]

Downloading data:  64%|██████▍   | 829M/1.29G [00:13<00:06, 66.4MB/s]

Downloading data:  65%|██████▍   | 836M/1.29G [00:13<00:06, 68.2MB/s]

Downloading data:  65%|██████▌   | 843M/1.29G [00:13<00:06, 68.2MB/s]

Downloading data:  66%|██████▌   | 850M/1.29G [00:13<00:07, 61.9MB/s]

Downloading data:  66%|██████▋   | 858M/1.29G [00:13<00:06, 65.5MB/s]

Downloading data:  67%|██████▋   | 865M/1.29G [00:13<00:06, 67.1MB/s]

Downloading data:  68%|██████▊   | 872M/1.29G [00:13<00:06, 69.6MB/s]

Downloading data:  68%|██████▊   | 879M/1.29G [00:13<00:07, 54.9MB/s]

Downloading data:  69%|██████▊   | 885M/1.29G [00:14<00:07, 54.4MB/s]

Downloading data:  69%|██████▉   | 892M/1.29G [00:14<00:06, 58.5MB/s]

Downloading data:  70%|██████▉   | 900M/1.29G [00:14<00:06, 61.6MB/s]

Downloading data:  70%|███████   | 907M/1.29G [00:14<00:06, 64.0MB/s]

Downloading data:  71%|███████   | 913M/1.29G [00:14<00:05, 63.9MB/s]

Downloading data:  71%|███████   | 920M/1.29G [00:14<00:05, 65.1MB/s]

Downloading data:  72%|███████▏  | 927M/1.29G [00:14<00:05, 67.2MB/s]

Downloading data:  72%|███████▏  | 934M/1.29G [00:14<00:05, 68.7MB/s]

Downloading data:  73%|███████▎  | 942M/1.29G [00:14<00:05, 69.6MB/s]

Downloading data:  73%|███████▎  | 949M/1.29G [00:14<00:04, 71.1MB/s]

Downloading data:  74%|███████▍  | 956M/1.29G [00:15<00:05, 63.6MB/s]

Downloading data:  75%|███████▍  | 964M/1.29G [00:15<00:04, 66.2MB/s]

Downloading data:  75%|███████▌  | 971M/1.29G [00:15<00:04, 67.8MB/s]

Downloading data:  76%|███████▌  | 978M/1.29G [00:15<00:05, 57.1MB/s]

Downloading data:  76%|███████▌  | 985M/1.29G [00:15<00:04, 62.1MB/s]

Downloading data:  77%|███████▋  | 993M/1.29G [00:15<00:04, 65.1MB/s]

Downloading data:  77%|███████▋  | 1.00G/1.29G [00:15<00:04, 66.7MB/s]

Downloading data:  78%|███████▊  | 1.01G/1.29G [00:15<00:04, 63.0MB/s]

Downloading data:  78%|███████▊  | 1.01G/1.29G [00:16<00:04, 61.5MB/s]

Downloading data:  79%|███████▉  | 1.02G/1.29G [00:16<00:04, 60.5MB/s]

Downloading data:  79%|███████▉  | 1.03G/1.29G [00:16<00:04, 60.1MB/s]

Downloading data:  80%|███████▉  | 1.03G/1.29G [00:16<00:04, 59.8MB/s]

Downloading data:  80%|████████  | 1.04G/1.29G [00:16<00:04, 59.5MB/s]

Downloading data:  81%|████████  | 1.04G/1.29G [00:16<00:04, 59.3MB/s]

Downloading data:  81%|████████  | 1.05G/1.29G [00:16<00:04, 59.4MB/s]

Downloading data:  82%|████████▏ | 1.06G/1.29G [00:16<00:03, 59.3MB/s]

Downloading data:  82%|████████▏ | 1.06G/1.29G [00:16<00:03, 59.2MB/s]

Downloading data:  83%|████████▎ | 1.07G/1.29G [00:16<00:03, 59.0MB/s]

Downloading data:  83%|████████▎ | 1.07G/1.29G [00:17<00:03, 58.8MB/s]

Downloading data:  84%|████████▎ | 1.08G/1.29G [00:17<00:03, 59.1MB/s]

Downloading data:  84%|████████▍ | 1.09G/1.29G [00:17<00:03, 59.3MB/s]

Downloading data:  84%|████████▍ | 1.09G/1.29G [00:17<00:03, 59.5MB/s]

Downloading data:  85%|████████▍ | 1.10G/1.29G [00:17<00:03, 59.6MB/s]

Downloading data:  85%|████████▌ | 1.10G/1.29G [00:17<00:03, 59.4MB/s]

Downloading data:  86%|████████▌ | 1.11G/1.29G [00:17<00:03, 59.4MB/s]

Downloading data:  86%|████████▋ | 1.12G/1.29G [00:17<00:03, 59.1MB/s]

Downloading data:  87%|████████▋ | 1.12G/1.29G [00:17<00:02, 59.3MB/s]

Downloading data:  87%|████████▋ | 1.13G/1.29G [00:17<00:02, 59.2MB/s]

Downloading data:  88%|████████▊ | 1.13G/1.29G [00:18<00:02, 59.0MB/s]

Downloading data:  88%|████████▊ | 1.14G/1.29G [00:18<00:02, 59.0MB/s]

Downloading data:  89%|████████▊ | 1.14G/1.29G [00:18<00:02, 58.9MB/s]

Downloading data:  89%|████████▉ | 1.15G/1.29G [00:18<00:02, 59.2MB/s]

Downloading data:  89%|████████▉ | 1.16G/1.29G [00:18<00:02, 58.7MB/s]

Downloading data:  90%|████████▉ | 1.16G/1.29G [00:18<00:02, 58.3MB/s]

Downloading data:  90%|█████████ | 1.17G/1.29G [00:18<00:02, 58.6MB/s]

Downloading data:  91%|█████████ | 1.17G/1.29G [00:18<00:02, 58.7MB/s]

Downloading data:  91%|█████████▏| 1.18G/1.29G [00:18<00:01, 58.6MB/s]

Downloading data:  92%|█████████▏| 1.19G/1.29G [00:18<00:01, 58.5MB/s]

Downloading data:  92%|█████████▏| 1.19G/1.29G [00:19<00:01, 58.3MB/s]

Downloading data:  93%|█████████▎| 1.20G/1.29G [00:19<00:01, 58.1MB/s]

Downloading data:  93%|█████████▎| 1.20G/1.29G [00:19<00:01, 58.4MB/s]

Downloading data:  94%|█████████▎| 1.21G/1.29G [00:19<00:01, 58.4MB/s]

Downloading data:  94%|█████████▍| 1.22G/1.29G [00:19<00:01, 58.5MB/s]

Downloading data:  94%|█████████▍| 1.22G/1.29G [00:19<00:01, 58.3MB/s]

Downloading data:  95%|█████████▍| 1.23G/1.29G [00:19<00:01, 58.1MB/s]

Downloading data:  95%|█████████▌| 1.23G/1.29G [00:19<00:01, 58.5MB/s]

Downloading data:  96%|█████████▌| 1.24G/1.29G [00:19<00:00, 58.8MB/s]

Downloading data:  96%|█████████▋| 1.24G/1.29G [00:19<00:00, 58.9MB/s]

Downloading data:  97%|█████████▋| 1.25G/1.29G [00:20<00:00, 58.9MB/s]

Downloading data:  97%|█████████▋| 1.26G/1.29G [00:20<00:00, 58.9MB/s]

Downloading data:  98%|█████████▊| 1.26G/1.29G [00:20<00:00, 58.6MB/s]

Downloading data:  98%|█████████▊| 1.27G/1.29G [00:20<00:00, 58.6MB/s]

Downloading data:  99%|█████████▊| 1.27G/1.29G [00:20<00:00, 58.7MB/s]

Downloading data:  99%|█████████▉| 1.28G/1.29G [00:20<00:00, 58.5MB/s]

Downloading data:  99%|█████████▉| 1.29G/1.29G [00:20<00:00, 58.5MB/s]

Downloading data: 100%|█████████▉| 1.29G/1.29G [00:20<00:00, 59.6MB/s]
Downloading data: 100%|██████████| 1.29G/1.29G [00:20<00:00, 62.2MB/s]

Downloading data files: 100%|██████████| 2/2 [00:23<00:00, 13.22s/it]
Downloading data files: 100%|██████████| 2/2 [00:23<00:00, 11.54s/it]

Extracting data files:   0%|          | 0/2 [00:00<?, ?it/s]
Extracting data files: 100%|██████████| 2/2 [00:00<00:00, 1382.89it/s]

Generating train split: 0 examples [00:00, ? examples/s]
Generating train split: 314 examples [00:00, 3000.29 examples/s]
Generating train split: 820 examples [00:00, 3246.26 examples/s]
Generating train split: 1305 examples [00:00, 2689.30 examples/s]
Generating train split: 1833 examples [00:00, 3409.15 examples/s]
Generating train split: 2246 examples [00:00, 2620.47 examples/s]
Generating train split: 2607 examples [00:00, 2364.40 examples/s]
Generating train split: 3000 examples [00:01, 2351.00 examples/s]
Generating train split: 3666 examples [00:01, 3252.95 examples/s]
Generating train split: 4363 examples [00:01, 3953.23 examples/s]
Generating train split: 5000 examples [00:01, 4443.61 examples/s]
                                                                 

Generating test split: 0 examples [00:00, ? examples/s]
Generating test split: 271 examples [00:00, 874.49 examples/s]
Generating test split: 456 examples [00:00, 539.42 examples/s]
Generating test split: 542 examples [00:01, 351.31 examples/s]
Generating test split: 588 examples [00:01, 259.08 examples/s]
Generating test split: 802 examples [00:01, 454.86 examples/s]
Generating test split: 930 examples [00:02, 396.07 examples/s]
Generating test split: 1000 examples [00:03, 221.55 examples/s]
Generating test split: 1088 examples [00:04, 150.00 examples/s]
Generating test split: 1174 examples [00:04, 185.92 examples/s]
Generating test split: 1273 examples [00:05, 176.96 examples/s]
Generating test split: 1375 examples [00:05, 168.49 examples/s]
Generating test split: 1427 examples [00:06, 168.06 examples/s]
Generating test split: 1472 examples [00:06, 161.86 examples/s]
Generating test split: 1580 examples [00:06, 225.60 examples/s]
Generating test split: 1716 examples [00:07, 169.13 examples/s]
Generating test split: 1909 examples [00:08, 233.80 examples/s]
Generating test split: 2000 examples [00:09, 123.30 examples/s]
Generating test split: 2170 examples [00:10, 177.12 examples/s]
Generating test split: 2512 examples [00:10, 352.55 examples/s]
Generating test split: 3000 examples [00:10, 622.95 examples/s]
Generating test split: 3725 examples [00:10, 1051.47 examples/s]
Generating test split: 3984 examples [00:13, 373.55 examples/s] 
Generating test split: 4132 examples [00:15, 221.88 examples/s]
Generating test split: 4274 examples [00:16, 214.32 examples/s]
Generating test split: 4362 examples [00:17, 177.12 examples/s]
Generating test split: 4818 examples [00:17, 334.01 examples/s]
Generating test split: 5000 examples [00:18, 285.03 examples/s]
                                                               
Dataset apps downloaded and prepared to /home/user/.cache/huggingface/datasets/codeparrot___apps/all/0.0.0/04ac807715d07d6e5cc580f59cdc8213cd7dc4529d0bb819cca72c9f8e8c1aa5. Subsequent calls will reuse this data.
Found cached dataset apps (/home/user/.cache/huggingface/datasets/codeparrot___apps/all/0.0.0/04ac807715d07d6e5cc580f59cdc8213cd7dc4529d0bb819cca72c9f8e8c1aa5)
Found cached dataset apps (/home/user/.cache/huggingface/datasets/codeparrot___apps/all/0.0.0/04ac807715d07d6e5cc580f59cdc8213cd7dc4529d0bb819cca72c9f8e8c1aa5)
Found cached dataset apps (/home/user/.cache/huggingface/datasets/codeparrot___apps/all/0.0.0/04ac807715d07d6e5cc580f59cdc8213cd7dc4529d0bb819cca72c9f8e8c1aa5)
Max length: 2048
PyTorch: setting up devices
Max length: 2048
PyTorch: setting up devices
[2022-12-20 17:19:24,412] [INFO] [comm.py:654:init_distributed] Initializing TorchBackend in DeepSpeed with backend nccl
Max length: 2048
PyTorch: setting up devices
Max length: 2048
PyTorch: setting up devices
The default value for the training argument `--report_to` will change in v5 (from all installed integrations to none). In v5, you will need to use `--report_to all` to get the same behavior as now. You should start updating your code and make this info disappear :-).
The default value for the training argument `--report_to` will change in v5 (from all installed integrations to none). In v5, you will need to use `--report_to all` to get the same behavior as now. You should start updating your code and make this info disappear :-).
The default value for the training argument `--report_to` will change in v5 (from all installed integrations to none). In v5, you will need to use `--report_to all` to get the same behavior as now. You should start updating your code and make this info disappear :-).
The default value for the training argument `--report_to` will change in v5 (from all installed integrations to none). In v5, you will need to use `--report_to all` to get the same behavior as now. You should start updating your code and make this info disappear :-).
GPU memory occupied: 5758 MB.
GPU memory occupied: 5758 MB.
GPU memory occupied: 5758 MB.
GPU memory occupied: 5758 MB.
Using /home/user/.cache/torch_extensions/py38_cu116 as PyTorch extensions root...
Creating extension directory /home/user/.cache/torch_extensions/py38_cu116/cpu_adam...
Using /home/user/.cache/torch_extensions/py38_cu116 as PyTorch extensions root...
Using /home/user/.cache/torch_extensions/py38_cu116 as PyTorch extensions root...
Using /home/user/.cache/torch_extensions/py38_cu116 as PyTorch extensions root...
Detected CUDA files, patching ldflags
Emitting ninja build file /home/user/.cache/torch_extensions/py38_cu116/cpu_adam/build.ninja...
Building extension module cpu_adam...
Allowing ninja to set a default number of workers... (overridable by setting the environment variable MAX_JOBS=N)
[1/3] /usr/local/cuda/bin/nvcc  -DTORCH_EXTENSION_NAME=cpu_adam -DTORCH_API_INCLUDE_EXTENSION_H -DPYBIND11_COMPILER_TYPE=\"_gcc\" -DPYBIND11_STDLIB=\"_libstdcpp\" -DPYBIND11_BUILD_ABI=\"_cxxabi1011\" -I/home/user/.local/lib/python3.8/site-packages/deepspeed/ops/csrc/includes -I/usr/local/cuda/include -isystem /home/user/.local/lib/python3.8/site-packages/torch/include -isystem /home/user/.local/lib/python3.8/site-packages/torch/include/torch/csrc/api/include -isystem /home/user/.local/lib/python3.8/site-packages/torch/include/TH -isystem /home/user/.local/lib/python3.8/site-packages/torch/include/THC -isystem /usr/local/cuda/include -isystem /usr/include/python3.8 -D_GLIBCXX_USE_CXX11_ABI=0 -D__CUDA_NO_HALF_OPERATORS__ -D__CUDA_NO_HALF_CONVERSIONS__ -D__CUDA_NO_BFLOAT16_CONVERSIONS__ -D__CUDA_NO_HALF2_OPERATORS__ --expt-relaxed-constexpr -gencode=arch=compute_86,code=compute_86 -gencode=arch=compute_86,code=sm_86 --compiler-options '-fPIC' -O3 --use_fast_math -std=c++14 -U__CUDA_NO_HALF_OPERATORS__ -U__CUDA_NO_HALF_CONVERSIONS__ -U__CUDA_NO_HALF2_OPERATORS__ -gencode=arch=compute_86,code=sm_86 -gencode=arch=compute_86,code=compute_86 -c /home/user/.local/lib/python3.8/site-packages/deepspeed/ops/csrc/common/custom_cuda_kernel.cu -o custom_cuda_kernel.cuda.o 
[2/3] c++ -MMD -MF cpu_adam.o.d -DTORCH_EXTENSION_NAME=cpu_adam -DTORCH_API_INCLUDE_EXTENSION_H -DPYBIND11_COMPILER_TYPE=\"_gcc\" -DPYBIND11_STDLIB=\"_libstdcpp\" -DPYBIND11_BUILD_ABI=\"_cxxabi1011\" -I/home/user/.local/lib/python3.8/site-packages/deepspeed/ops/csrc/includes -I/usr/local/cuda/include -isystem /home/user/.local/lib/python3.8/site-packages/torch/include -isystem /home/user/.local/lib/python3.8/site-packages/torch/include/torch/csrc/api/include -isystem /home/user/.local/lib/python3.8/site-packages/torch/include/TH -isystem /home/user/.local/lib/python3.8/site-packages/torch/include/THC -isystem /usr/local/cuda/include -isystem /usr/include/python3.8 -D_GLIBCXX_USE_CXX11_ABI=0 -fPIC -std=c++14 -O3 -std=c++14 -g -Wno-reorder -L/usr/local/cuda/lib64 -lcudart -lcublas -g -march=native -fopenmp -D__AVX256__ -c /home/user/.local/lib/python3.8/site-packages/deepspeed/ops/csrc/adam/cpu_adam.cpp -o cpu_adam.o 
[3/3] c++ cpu_adam.o custom_cuda_kernel.cuda.o -shared -lcurand -L/home/user/.local/lib/python3.8/site-packages/torch/lib -lc10 -lc10_cuda -ltorch_cpu -ltorch_cuda_cu -ltorch_cuda_cpp -ltorch -ltorch_python -L/usr/local/cuda/lib64 -lcudart -o cpu_adam.so
Loading extension module cpu_adam...
Time to load cpu_adam op: 39.87346434593201 seconds
Loading extension module cpu_adam...
Time to load cpu_adam op: 39.953386545181274 seconds
Loading extension module cpu_adam...
Time to load cpu_adam op: 39.9042444229126 seconds
Loading extension module cpu_adam...
Time to load cpu_adam op: 39.96823048591614 seconds
Using /home/user/.cache/torch_extensions/py38_cu116 as PyTorch extensions root...
Creating extension directory /home/user/.cache/torch_extensions/py38_cu116/utils...
Using /home/user/.cache/torch_extensions/py38_cu116 as PyTorch extensions root...
Using /home/user/.cache/torch_extensions/py38_cu116 as PyTorch extensions root...
Using /home/user/.cache/torch_extensions/py38_cu116 as PyTorch extensions root...
Emitting ninja build file /home/user/.cache/torch_extensions/py38_cu116/utils/build.ninja...
Building extension module utils...
Allowing ninja to set a default number of workers... (overridable by setting the environment variable MAX_JOBS=N)
[1/2] c++ -MMD -MF flatten_unflatten.o.d -DTORCH_EXTENSION_NAME=utils -DTORCH_API_INCLUDE_EXTENSION_H -DPYBIND11_COMPILER_TYPE=\"_gcc\" -DPYBIND11_STDLIB=\"_libstdcpp\" -DPYBIND11_BUILD_ABI=\"_cxxabi1011\" -isystem /home/user/.local/lib/python3.8/site-packages/torch/include -isystem /home/user/.local/lib/python3.8/site-packages/torch/include/torch/csrc/api/include -isystem /home/user/.local/lib/python3.8/site-packages/torch/include/TH -isystem /home/user/.local/lib/python3.8/site-packages/torch/include/THC -isystem /usr/include/python3.8 -D_GLIBCXX_USE_CXX11_ABI=0 -fPIC -std=c++14 -c /home/user/.local/lib/python3.8/site-packages/deepspeed/ops/csrc/utils/flatten_unflatten.cpp -o flatten_unflatten.o 
[2/2] c++ flatten_unflatten.o -shared -L/home/user/.local/lib/python3.8/site-packages/torch/lib -lc10 -ltorch_cpu -ltorch -ltorch_python -o utils.so
Loading extension module utils...
Time to load utils op: 23.038707971572876 seconds
Loading extension module utils...
Time to load utils op: 22.962859392166138 seconds
Loading extension module utils...
Time to load utils op: 22.961502075195312 seconds
Loading extension module utils...
Time to load utils op: 23.0585675239563 seconds
Rank: 2 partition count [4] and sizes[(31299648, False)] 
Rank: 3 partition count [4] and sizes[(31299648, False)] 
Rank: 1 partition count [4] and sizes[(31299648, False)] 
Rank: 0 partition count [4] and sizes[(31299648, False)] 
Using /home/user/.cache/torch_extensions/py38_cu116 as PyTorch extensions root...
No modifications detected for re-loaded extension module utils, skipping build step...
Loading extension module utils...
Time to load utils op: 0.0010988712310791016 seconds
Using /home/user/.cache/torch_extensions/py38_cu116 as PyTorch extensions root...
No modifications detected for re-loaded extension module utils, skipping build step...
Loading extension module utils...
Time to load utils op: 0.0008800029754638672 seconds
Using /home/user/.cache/torch_extensions/py38_cu116 as PyTorch extensions root...
No modifications detected for re-loaded extension module utils, skipping build step...
Loading extension module utils...
Time to load utils op: 0.0007402896881103516 seconds
Using /home/user/.cache/torch_extensions/py38_cu116 as PyTorch extensions root...
No modifications detected for re-loaded extension module utils, skipping build step...
Loading extension module utils...
Time to load utils op: 0.0005016326904296875 seconds

  0%|          | 0/24420 [00:00<?, ?it/s]
  0%|          | 1/24420 [00:01<10:02:25,  1.48s/it]
                                                    
{'loss': 6.2666, 'learning_rate': 0.0, 'epoch': 0.0}

  0%|          | 1/24420 [00:01<10:02:25,  1.48s/it]
  0%|          | 2/24420 [00:02<9:57:28,  1.47s/it] 
  0%|          | 3/24420 [00:04<10:12:29,  1.51s/it]
  0%|          | 4/24420 [00:05<10:07:51,  1.49s/it]
  0%|          | 5/24420 [00:07<10:04:20,  1.49s/it]
                                                    
{'loss': 4.304, 'learning_rate': 1.294882868674145e-05, 'epoch': 0.0}

  0%|          | 5/24420 [00:07<10:04:20,  1.49s/it]
  0%|          | 6/24420 [00:08<10:03:48,  1.48s/it]
  0%|          | 7/24420 [00:10<10:00:46,  1.48s/it]
  0%|          | 8/24420 [00:11<9:58:07,  1.47s/it] 
  0%|          | 9/24420 [00:13<9:55:41,  1.46s/it]
  0%|          | 10/24420 [00:14<9:53:20,  1.46s/it]
                                                    
{'loss': 1.7203, 'learning_rate': 1.852558565662928e-05, 'epoch': 0.0}

  0%|          | 10/24420 [00:14<9:53:20,  1.46s/it]
  0%|          | 11/24420 [00:16<9:52:23,  1.46s/it]
  0%|          | 12/24420 [00:17<9:52:40,  1.46s/it]
  0%|          | 13/24420 [00:19<9:52:12,  1.46s/it]
  0%|          | 14/24420 [00:20<9:51:13,  1.45s/it]
  0%|          | 15/24420 [00:21<9:49:59,  1.45s/it]
                                                    
{'loss': 0.587, 'learning_rate': 2.1787779359648994e-05, 'epoch': 0.0}

  0%|          | 15/24420 [00:21<9:49:59,  1.45s/it]
  0%|          | 16/24420 [00:23<9:51:48,  1.46s/it]
  0%|          | 17/24420 [00:24<9:51:49,  1.46s/it]
  0%|          | 18/24420 [00:26<9:51:04,  1.45s/it]
  0%|          | 19/24420 [00:27<9:53:12,  1.46s/it]
  0%|          | 20/24420 [00:29<9:53:16,  1.46s/it]
                                                    
{'loss': 0.5233, 'learning_rate': 2.41023426265171e-05, 'epoch': 0.0}

  0%|          | 20/24420 [00:29<9:53:16,  1.46s/it]
  0%|          | 21/24420 [00:30<9:55:37,  1.46s/it]
  0%|          | 22/24420 [00:32<9:56:08,  1.47s/it]
  0%|          | 23/24420 [00:33<9:55:07,  1.46s/it]
  0%|          | 24/24420 [00:35<9:55:23,  1.46s/it]
  0%|          | 25/24420 [00:36<9:54:27,  1.46s/it]
                                                    
{'loss': 0.558, 'learning_rate': 2.58976573734829e-05, 'epoch': 0.01}

  0%|          | 25/24420 [00:36<9:54:27,  1.46s/it]
  0%|          | 26/24420 [00:38<9:56:07,  1.47s/it]
  0%|          | 27/24420 [00:39<9:56:41,  1.47s/it]
  0%|          | 28/24420 [00:41<9:56:49,  1.47s/it]
  0%|          | 29/24420 [00:42<9:56:22,  1.47s/it]
  0%|          | 30/24420 [00:43<9:56:18,  1.47s/it]
                                                    
{'loss': 0.496, 'learning_rate': 2.7364536329536817e-05, 'epoch': 0.01}

  0%|          | 30/24420 [00:43<9:56:18,  1.47s/it]
  0%|          | 31/24420 [00:45<9:57:17,  1.47s/it]
  0%|          | 32/24420 [00:46<9:58:37,  1.47s/it]
  0%|          | 33/24420 [00:48<9:57:28,  1.47s/it]
  0%|          | 34/24420 [00:49<9:55:31,  1.47s/it]
  0%|          | 35/24420 [00:51<9:54:45,  1.46s/it]
                                                    
{'loss': 0.4932, 'learning_rate': 2.8604764815275082e-05, 'epoch': 0.01}

  0%|          | 35/24420 [00:51<9:54:45,  1.46s/it]
  0%|          | 36/24420 [00:52<9:53:30,  1.46s/it]
  0%|          | 37/24420 [00:54<9:55:22,  1.47s/it]
  0%|          | 38/24420 [00:55<9:55:20,  1.47s/it]
  0%|          | 39/24420 [00:57<9:54:48,  1.46s/it]
  0%|          | 40/24420 [00:58<9:52:57,  1.46s/it]
                                                    
{'loss': 0.4482, 'learning_rate': 2.9679099596404923e-05, 'epoch': 0.01}

  0%|          | 40/24420 [00:58<9:52:57,  1.46s/it]
  0%|          | 41/24420 [01:00<10:52:38,  1.61s/it]
  0%|          | 42/24420 [01:02<10:36:17,  1.57s/it]
  0%|          | 43/24420 [01:03<10:23:01,  1.53s/it]
  0%|          | 44/24420 [01:04<10:13:34,  1.51s/it]
  0%|          | 45/24420 [01:06<10:07:01,  1.49s/it]
                                                     
{'loss': 0.4979, 'learning_rate': 3.0626730032556536e-05, 'epoch': 0.01}

  0%|          | 45/24420 [01:06<10:07:01,  1.49s/it]
  0%|          | 46/24420 [01:07<10:02:00,  1.48s/it]
  0%|          | 47/24420 [01:09<9:58:47,  1.47s/it] 
  0%|          | 48/24420 [01:10<9:55:28,  1.47s/it]
  0%|          | 49/24420 [01:12<9:53:59,  1.46s/it]
  0%|          | 50/24420 [01:13<9:52:53,  1.46s/it]
                                                    
{'loss': 0.4416, 'learning_rate': 3.147441434337073e-05, 'epoch': 0.01}

  0%|          | 50/24420 [01:13<9:52:53,  1.46s/it]
  0%|          | 51/24420 [01:15<9:53:00,  1.46s/it]
  0%|          | 52/24420 [01:16<9:55:49,  1.47s/it]
  0%|          | 53/24420 [01:18<9:54:16,  1.46s/it]
  0%|          | 54/24420 [01:19<9:53:02,  1.46s/it]
  0%|          | 55/24420 [01:20<9:51:51,  1.46s/it]
                                                    
{'loss': 0.4864, 'learning_rate': 3.224123807782732e-05, 'epoch': 0.01}

  0%|          | 55/24420 [01:20<9:51:51,  1.46s/it]
  0%|          | 56/24420 [01:22<9:52:24,  1.46s/it]
  0%|          | 57/24420 [01:23<9:53:22,  1.46s/it]
  0%|          | 58/24420 [01:25<9:53:08,  1.46s/it]
  0%|          | 59/24420 [01:26<9:52:04,  1.46s/it]
  0%|          | 60/24420 [01:28<9:52:46,  1.46s/it]
                                                    
{'loss': 0.4579, 'learning_rate': 3.294129329942464e-05, 'epoch': 0.01}

  0%|          | 60/24420 [01:28<9:52:46,  1.46s/it]
  0%|          | 61/24420 [01:29<9:52:57,  1.46s/it]
  0%|          | 62/24420 [01:31<9:55:26,  1.47s/it]
  0%|          | 63/24420 [01:32<9:55:11,  1.47s/it]
  0%|          | 64/24420 [01:34<9:54:50,  1.47s/it]
  0%|          | 65/24420 [01:35<9:56:23,  1.47s/it]
                                                    
{'loss': 0.4636, 'learning_rate': 3.358528167653452e-05, 'epoch': 0.01}

  0%|          | 65/24420 [01:35<9:56:23,  1.47s/it]
  0%|          | 66/24420 [01:37<9:55:08,  1.47s/it]
  0%|          | 67/24420 [01:38<9:53:13,  1.46s/it]
  0%|          | 68/24420 [01:39<9:51:44,  1.46s/it]
  0%|          | 69/24420 [01:41<9:51:43,  1.46s/it]
  0%|          | 70/24420 [01:42<9:52:07,  1.46s/it]
                                                    
{'loss': 0.4723, 'learning_rate': 3.4181521785162905e-05, 'epoch': 0.01}

  0%|          | 70/24420 [01:42<9:52:07,  1.46s/it]
  0%|          | 71/24420 [01:44<9:51:59,  1.46s/it]
  0%|          | 72/24420 [01:45<9:51:37,  1.46s/it]
  0%|          | 73/24420 [01:47<9:51:19,  1.46s/it]
  0%|          | 74/24420 [01:48<9:51:53,  1.46s/it]
  0%|          | 75/24420 [01:50<9:52:17,  1.46s/it]
                                                    
{'loss': 0.4074, 'learning_rate': 3.473660804639045e-05, 'epoch': 0.02}

  0%|          | 75/24420 [01:50<9:52:17,  1.46s/it]
  0%|          | 76/24420 [01:51<9:52:44,  1.46s/it]
  0%|          | 77/24420 [01:53<9:56:47,  1.47s/it]
  0%|          | 78/24420 [01:54<9:56:20,  1.47s/it]
  0%|          | 79/24420 [01:56<9:55:31,  1.47s/it]
  0%|          | 80/24420 [01:57<9:57:29,  1.47s/it]
                                                    
{'loss': 0.4294, 'learning_rate': 3.525585656629274e-05, 'epoch': 0.02}

  0%|          | 80/24420 [01:57<9:57:29,  1.47s/it]
  0%|          | 81/24420 [01:59<9:58:11,  1.47s/it]
  0%|          | 82/24420 [02:00<9:57:09,  1.47s/it]
  0%|          | 83/24420 [02:01<9:57:30,  1.47s/it]
  0%|          | 84/24420 [02:03<9:56:52,  1.47s/it]
  0%|          | 85/24420 [02:04<9:54:24,  1.47s/it]
                                                    
{'loss': 0.4103, 'learning_rate': 3.574361557584177e-05, 'epoch': 0.02}

  0%|          | 85/24420 [02:04<9:54:24,  1.47s/it]
  0%|          | 86/24420 [02:06<9:54:16,  1.47s/it]
  0%|          | 87/24420 [02:07<9:53:30,  1.46s/it]
  0%|          | 88/24420 [02:09<9:53:22,  1.46s/it]
  0%|          | 89/24420 [02:10<9:53:52,  1.46s/it]
  0%|          | 90/24420 [02:12<9:53:15,  1.46s/it]
                                                    
{'loss': 0.3957, 'learning_rate': 3.620348700244436e-05, 'epoch': 0.02}

  0%|          | 90/24420 [02:12<9:53:15,  1.46s/it]
  0%|          | 91/24420 [02:13<9:55:21,  1.47s/it]
  0%|          | 92/24420 [02:15<9:57:49,  1.47s/it]
  0%|          | 93/24420 [02:16<9:57:34,  1.47s/it]
  0%|          | 94/24420 [02:18<9:56:27,  1.47s/it]
  0%|          | 95/24420 [02:19<9:54:54,  1.47s/it]
                                                    
{'loss': 0.4273, 'learning_rate': 3.6638488054916214e-05, 'epoch': 0.02}

  0%|          | 95/24420 [02:19<9:54:54,  1.47s/it]
  0%|          | 96/24420 [02:21<9:56:39,  1.47s/it]
  0%|          | 97/24420 [02:22<9:55:58,  1.47s/it]
  0%|          | 98/24420 [02:24<9:56:59,  1.47s/it]
  0%|          | 99/24420 [02:25<9:57:52,  1.47s/it]
  0%|          | 100/24420 [02:26<9:56:04,  1.47s/it]
                                                     
{'loss': 0.4358, 'learning_rate': 3.705117131325856e-05, 'epoch': 0.02}

  0%|          | 100/24420 [02:26<9:56:04,  1.47s/it]
  0%|          | 101/24420 [02:28<9:55:59,  1.47s/it]
  0%|          | 102/24420 [02:29<9:54:20,  1.47s/it]
  0%|          | 103/24420 [02:31<9:54:03,  1.47s/it]
  0%|          | 104/24420 [02:32<9:53:35,  1.46s/it]
  0%|          | 105/24420 [02:34<9:52:24,  1.46s/it]
                                                     
{'loss': 0.4061, 'learning_rate': 3.7443715488182624e-05, 'epoch': 0.02}

  0%|          | 105/24420 [02:34<9:52:24,  1.46s/it]
  0%|          | 106/24420 [02:35<9:53:12,  1.46s/it]
  0%|          | 107/24420 [02:37<9:51:56,  1.46s/it]
  0%|          | 108/24420 [02:38<9:52:48,  1.46s/it]
  0%|          | 109/24420 [02:40<9:52:30,  1.46s/it]
  0%|          | 110/24420 [02:41<9:52:00,  1.46s/it]
                                                     
{'loss': 0.3687, 'learning_rate': 3.781799504771514e-05, 'epoch': 0.02}

  0%|          | 110/24420 [02:41<9:52:00,  1.46s/it]
  0%|          | 111/24420 [02:43<9:51:47,  1.46s/it]
  0%|          | 112/24420 [02:44<9:50:49,  1.46s/it]
  0%|          | 113/24420 [02:45<9:51:39,  1.46s/it]
  0%|          | 114/24420 [02:47<9:52:11,  1.46s/it]
  0%|          | 115/24420 [02:48<9:53:06,  1.46s/it]
                                                     
{'loss': 0.4007, 'learning_rate': 3.81756343539018e-05, 'epoch': 0.02}

  0%|          | 115/24420 [02:48<9:53:06,  1.46s/it]
  0%|          | 116/24420 [02:50<9:53:14,  1.46s/it]
  0%|          | 117/24420 [02:51<9:52:24,  1.46s/it]
  0%|          | 118/24420 [02:53<9:51:54,  1.46s/it]
  0%|          | 119/24420 [02:54<9:51:10,  1.46s/it]
  0%|          | 120/24420 [02:56<9:51:13,  1.46s/it]
                                                     
{'loss': 0.4091, 'learning_rate': 3.851805026931246e-05, 'epoch': 0.02}

  0%|          | 120/24420 [02:56<9:51:13,  1.46s/it]
  0%|          | 121/24420 [02:57<9:52:41,  1.46s/it]
  0%|          | 122/24420 [02:59<9:52:33,  1.46s/it]
  1%|          | 123/24420 [03:00<9:52:58,  1.46s/it]
  1%|          | 124/24420 [03:02<9:53:11,  1.46s/it]
  1%|          | 125/24420 [03:03<9:51:57,  1.46s/it]
                                                     
{'loss': 0.3635, 'learning_rate': 3.8846486060224364e-05, 'epoch': 0.03}

  1%|          | 125/24420 [03:03<9:51:57,  1.46s/it]
  1%|          | 126/24420 [03:04<9:52:09,  1.46s/it]
  1%|          | 127/24420 [03:06<9:54:18,  1.47s/it]
  1%|          | 128/24420 [03:07<9:54:04,  1.47s/it]
  1%|          | 129/24420 [03:09<9:53:26,  1.47s/it]
  1%|          | 130/24420 [03:10<9:52:35,  1.46s/it]
                                                     
{'loss': 0.4027, 'learning_rate': 3.916203864642234e-05, 'epoch': 0.03}

  1%|          | 130/24420 [03:10<9:52:35,  1.46s/it]
  1%|          | 131/24420 [03:12<9:53:50,  1.47s/it]
  1%|          | 132/24420 [03:13<9:56:16,  1.47s/it]
  1%|          | 133/24420 [03:15<9:54:42,  1.47s/it]
  1%|          | 134/24420 [03:16<9:53:14,  1.47s/it]
  1%|          | 135/24420 [03:18<9:53:41,  1.47s/it]
                                                     
{'loss': 0.4473, 'learning_rate': 3.946568070546408e-05, 'epoch': 0.03}

  1%|          | 135/24420 [03:18<9:53:41,  1.47s/it]
  1%|          | 136/24420 [03:19<9:54:13,  1.47s/it]
  1%|          | 137/24420 [03:21<9:54:21,  1.47s/it]
  1%|          | 138/24420 [03:22<9:54:40,  1.47s/it]
  1%|          | 139/24420 [03:24<9:53:28,  1.47s/it]
  1%|          | 140/24420 [03:25<9:52:44,  1.46s/it]
                                                     
{'loss': 0.3939, 'learning_rate': 3.975827875505073e-05, 'epoch': 0.03}

  1%|          | 140/24420 [03:25<9:52:44,  1.46s/it]
  1%|          | 141/24420 [03:26<9:52:52,  1.47s/it]
  1%|          | 142/24420 [03:28<9:52:51,  1.47s/it]
  1%|          | 143/24420 [03:29<9:53:01,  1.47s/it]
  1%|          | 144/24420 [03:31<9:52:12,  1.46s/it]
  1%|          | 145/24420 [03:32<9:51:24,  1.46s/it]
                                                     
{'loss': 0.3985, 'learning_rate': 4.004060806090172e-05, 'epoch': 0.03}

  1%|          | 145/24420 [03:32<9:51:24,  1.46s/it]
  1%|          | 146/24420 [03:34<9:51:46,  1.46s/it]
  1%|          | 147/24420 [03:35<9:50:57,  1.46s/it]
  1%|          | 148/24420 [03:37<9:51:04,  1.46s/it]
  1%|          | 149/24420 [03:38<9:51:22,  1.46s/it]
  1%|          | 150/24420 [03:40<9:51:43,  1.46s/it]
                                                     
{'loss': 0.3767, 'learning_rate': 4.031336501627827e-05, 'epoch': 0.03}

  1%|          | 150/24420 [03:40<9:51:43,  1.46s/it]
  1%|          | 151/24420 [03:41<9:51:53,  1.46s/it]
  1%|          | 152/24420 [03:43<9:52:55,  1.47s/it]
  1%|          | 153/24420 [03:44<9:54:04,  1.47s/it]
  1%|          | 154/24420 [03:46<9:53:48,  1.47s/it]
  1%|          | 155/24420 [03:47<9:53:44,  1.47s/it]
                                                     
{'loss': 0.377, 'learning_rate': 4.0577177490884e-05, 'epoch': 0.03}

  1%|          | 155/24420 [03:47<9:53:44,  1.47s/it]
  1%|          | 156/24420 [03:48<9:53:48,  1.47s/it]
  1%|          | 157/24420 [03:50<9:52:54,  1.47s/it]
  1%|          | 158/24420 [03:51<9:52:59,  1.47s/it]
  1%|          | 159/24420 [03:53<9:52:36,  1.47s/it]
  1%|          | 160/24420 [03:54<9:53:23,  1.47s/it]
                                                     
{'loss': 0.3921, 'learning_rate': 4.0832613536180565e-05, 'epoch': 0.03}

  1%|          | 160/24420 [03:54<9:53:23,  1.47s/it]
  1%|          | 161/24420 [03:56<9:55:19,  1.47s/it]
  1%|          | 162/24420 [03:57<9:54:02,  1.47s/it]
  1%|          | 163/24420 [03:59<9:53:27,  1.47s/it]
  1%|          | 164/24420 [04:00<9:52:50,  1.47s/it]
  1%|          | 165/24420 [04:02<9:52:41,  1.47s/it]
                                                     
{'loss': 0.4209, 'learning_rate': 4.1080188750734856e-05, 'epoch': 0.03}

  1%|          | 165/24420 [04:02<9:52:41,  1.47s/it]
  1%|          | 166/24420 [04:03<9:52:45,  1.47s/it]
  1%|          | 167/24420 [04:05<9:53:39,  1.47s/it]
  1%|          | 168/24420 [04:06<9:53:07,  1.47s/it]
  1%|          | 169/24420 [04:08<9:52:30,  1.47s/it]
  1%|          | 170/24420 [04:09<9:54:16,  1.47s/it]
                                                     
{'loss': 0.4281, 'learning_rate': 4.1320372545729594e-05, 'epoch': 0.03}

  1%|          | 170/24420 [04:09<9:54:16,  1.47s/it]
  1%|          | 171/24420 [04:10<9:55:44,  1.47s/it]
  1%|          | 172/24420 [04:12<9:55:30,  1.47s/it]
  1%|          | 173/24420 [04:13<9:55:25,  1.47s/it]
  1%|          | 174/24420 [04:15<9:56:14,  1.48s/it]
  1%|          | 175/24420 [04:16<9:56:15,  1.48s/it]
                                                     
{'loss': 0.406, 'learning_rate': 4.155359350201654e-05, 'epoch': 0.04}

  1%|          | 175/24420 [04:16<9:56:15,  1.48s/it]
  1%|          | 176/24420 [04:18<9:55:09,  1.47s/it]
  1%|          | 177/24420 [04:19<9:54:35,  1.47s/it]
  1%|          | 178/24420 [04:21<9:54:25,  1.47s/it]
  1%|          | 179/24420 [04:22<9:53:37,  1.47s/it]
  1%|          | 180/24420 [04:24<9:52:28,  1.47s/it]
                                                     
{'loss': 0.3291, 'learning_rate': 4.178024397233218e-05, 'epoch': 0.04}

  1%|          | 180/24420 [04:24<9:52:28,  1.47s/it]
  1%|          | 181/24420 [04:25<9:53:37,  1.47s/it]
  1%|          | 182/24420 [04:27<9:54:31,  1.47s/it]
  1%|          | 183/24420 [04:28<9:53:47,  1.47s/it]
  1%|          | 184/24420 [04:30<9:54:04,  1.47s/it]
  1%|          | 185/24420 [04:31<9:53:28,  1.47s/it]
                                                     
{'loss': 0.3794, 'learning_rate': 4.200068405281827e-05, 'epoch': 0.04}

  1%|          | 185/24420 [04:31<9:53:28,  1.47s/it]
  1%|          | 186/24420 [04:33<9:53:08,  1.47s/it]
  1%|          | 187/24420 [04:34<9:54:23,  1.47s/it]
  1%|          | 188/24420 [04:35<9:53:16,  1.47s/it]
  1%|          | 189/24420 [04:37<9:51:53,  1.47s/it]
  1%|          | 190/24420 [04:38<9:51:09,  1.46s/it]
                                                     
{'loss': 0.4035, 'learning_rate': 4.221524502480404e-05, 'epoch': 0.04}

  1%|          | 190/24420 [04:38<9:51:09,  1.46s/it]
  1%|          | 191/24420 [04:40<9:52:42,  1.47s/it]
  1%|          | 192/24420 [04:41<9:53:07,  1.47s/it]
  1%|          | 193/24420 [04:43<9:54:14,  1.47s/it]
  1%|          | 194/24420 [04:44<9:53:28,  1.47s/it]
  1%|          | 195/24420 [04:46<9:53:10,  1.47s/it]
                                                     
{'loss': 0.412, 'learning_rate': 4.242423234944206e-05, 'epoch': 0.04}

  1%|          | 195/24420 [04:46<9:53:10,  1.47s/it]
  1%|          | 196/24420 [04:47<9:53:29,  1.47s/it]
  1%|          | 197/24420 [04:49<9:53:28,  1.47s/it]
  1%|          | 198/24420 [04:50<9:53:49,  1.47s/it]
  1%|          | 199/24420 [04:52<9:53:15,  1.47s/it]
  1%|          | 200/24420 [04:53<9:52:21,  1.47s/it]
                                                     
{'loss': 0.4318, 'learning_rate': 4.262792828314637e-05, 'epoch': 0.04}

  1%|          | 200/24420 [04:53<9:52:21,  1.47s/it]/home/user/.local/lib/python3.8/site-packages/torch/nn/modules/module.py:1432: UserWarning: Positional args are being deprecated, use kwargs instead. Refer to https://pytorch.org/docs/master/generated/torch.nn.Module.html#torch.nn.Module.state_dict for details.
  warnings.warn(
/home/user/.local/lib/python3.8/site-packages/torch/nn/modules/module.py:1432: UserWarning: Positional args are being deprecated, use kwargs instead. Refer to https://pytorch.org/docs/master/generated/torch.nn.Module.html#torch.nn.Module.state_dict for details.
  warnings.warn(
/home/user/.local/lib/python3.8/site-packages/torch/nn/modules/module.py:1432: UserWarning: Positional args are being deprecated, use kwargs instead. Refer to https://pytorch.org/docs/master/generated/torch.nn.Module.html#torch.nn.Module.state_dict for details.
  warnings.warn(
/home/user/.local/lib/python3.8/site-packages/torch/nn/modules/module.py:1432: UserWarning: Positional args are being deprecated, use kwargs instead. Refer to https://pytorch.org/docs/master/generated/torch.nn.Module.html#torch.nn.Module.state_dict for details.
  warnings.warn(

  1%|          | 201/24420 [04:57<15:20:43,  2.28s/it]
  1%|          | 202/24420 [04:59<13:42:03,  2.04s/it]
  1%|          | 203/24420 [05:00<12:32:36,  1.86s/it]
  1%|          | 204/24420 [05:02<11:43:05,  1.74s/it]
  1%|          | 205/24420 [05:03<11:08:20,  1.66s/it]
                                                      
{'loss': 0.3757, 'learning_rate': 4.282659417003183e-05, 'epoch': 0.04}

  1%|          | 205/24420 [05:03<11:08:20,  1.66s/it]
  1%|          | 206/24420 [05:05<10:47:14,  1.60s/it]
  1%|          | 207/24420 [05:06<10:30:06,  1.56s/it]
  1%|          | 208/24420 [05:08<10:17:16,  1.53s/it]
  1%|          | 209/24420 [05:09<10:08:26,  1.51s/it]
  1%|          | 210/24420 [05:10<10:01:34,  1.49s/it]
                                                      
{'loss': 0.3782, 'learning_rate': 4.302047245807045e-05, 'epoch': 0.04}

  1%|          | 210/24420 [05:10<10:01:34,  1.49s/it]
  1%|          | 211/24420 [05:12<9:58:04,  1.48s/it] 
  1%|          | 212/24420 [05:13<9:55:57,  1.48s/it]
  1%|          | 213/24420 [05:15<9:54:19,  1.47s/it]
  1%|          | 214/24420 [05:16<9:52:17,  1.47s/it]
  1%|          | 215/24420 [05:18<9:50:57,  1.46s/it]
                                                     
{'loss': 0.3931, 'learning_rate': 4.320978847798302e-05, 'epoch': 0.04}

  1%|          | 215/24420 [05:18<9:50:57,  1.46s/it]
  1%|          | 216/24420 [05:19<9:51:52,  1.47s/it]
  1%|          | 217/24420 [05:21<9:51:39,  1.47s/it]
  1%|          | 218/24420 [05:22<9:51:25,  1.47s/it]
  1%|          | 219/24420 [05:24<9:51:25,  1.47s/it]
  1%|          | 220/24420 [05:25<9:51:31,  1.47s/it]
                                                     
{'loss': 0.4475, 'learning_rate': 4.3394752017602966e-05, 'epoch': 0.05}

  1%|          | 220/24420 [05:25<9:51:31,  1.47s/it]
  1%|          | 221/24420 [05:27<9:52:09,  1.47s/it]
  1%|          | 222/24420 [05:28<9:52:32,  1.47s/it]
  1%|          | 223/24420 [05:30<9:54:59,  1.48s/it]
  1%|          | 224/24420 [05:31<9:53:57,  1.47s/it]
  1%|          | 225/24420 [05:32<9:53:28,  1.47s/it]
                                                     
{'loss': 0.395, 'learning_rate': 4.357555871929799e-05, 'epoch': 0.05}

  1%|          | 225/24420 [05:32<9:53:28,  1.47s/it]
  1%|          | 226/24420 [05:34<9:53:45,  1.47s/it]
  1%|          | 227/24420 [05:35<9:56:59,  1.48s/it]
  1%|          | 228/24420 [05:37<9:54:58,  1.48s/it]
  1%|          | 229/24420 [05:38<9:53:44,  1.47s/it]
  1%|          | 230/24420 [05:40<9:53:45,  1.47s/it]
                                                     
{'loss': 0.4118, 'learning_rate': 4.375239132378962e-05, 'epoch': 0.05}

  1%|          | 230/24420 [05:40<9:53:45,  1.47s/it]
  1%|          | 231/24420 [05:41<9:54:53,  1.48s/it]
  1%|          | 232/24420 [05:43<9:53:55,  1.47s/it]
  1%|          | 233/24420 [05:44<9:53:38,  1.47s/it]
  1%|          | 234/24420 [05:46<9:52:49,  1.47s/it]
  1%|          | 235/24420 [05:47<9:52:58,  1.47s/it]
                                                     
{'loss': 0.4017, 'learning_rate': 4.392542078019592e-05, 'epoch': 0.05}

  1%|          | 235/24420 [05:47<9:52:58,  1.47s/it]
  1%|          | 236/24420 [05:49<9:53:31,  1.47s/it]
  1%|          | 237/24420 [05:50<9:53:45,  1.47s/it]
  1%|          | 238/24420 [05:52<9:52:39,  1.47s/it]
  1%|          | 239/24420 [05:53<9:53:09,  1.47s/it]
  1%|          | 240/24420 [05:55<9:52:29,  1.47s/it]
                                                     
{'loss': 0.4164, 'learning_rate': 4.4094807239200284e-05, 'epoch': 0.05}

  1%|          | 240/24420 [05:55<9:52:29,  1.47s/it]
  1%|          | 241/24420 [05:56<9:51:41,  1.47s/it]
  1%|          | 242/24420 [05:57<9:51:14,  1.47s/it]
  1%|          | 243/24420 [05:59<9:50:57,  1.47s/it]
  1%|          | 244/24420 [06:00<9:50:16,  1.46s/it]
  1%|          | 245/24420 [06:02<9:51:19,  1.47s/it]
                                                     
{'loss': 0.3915, 'learning_rate': 4.426070094380871e-05, 'epoch': 0.05}

  1%|          | 245/24420 [06:02<9:51:19,  1.47s/it]
  1%|          | 246/24420 [06:03<9:52:22,  1.47s/it]
  1%|          | 247/24420 [06:05<9:52:04,  1.47s/it]
  1%|          | 248/24420 [06:06<9:50:37,  1.47s/it]
  1%|          | 249/24420 [06:08<9:51:00,  1.47s/it]
  1%|          | 250/24420 [06:09<9:51:42,  1.47s/it]
                                                     
{'loss': 0.3841, 'learning_rate': 4.442324303011218e-05, 'epoch': 0.05}

  1%|          | 250/24420 [06:09<9:51:42,  1.47s/it]
  1%|          | 251/24420 [06:11<9:51:05,  1.47s/it]
  1%|          | 252/24420 [06:12<9:51:19,  1.47s/it]
  1%|          | 253/24420 [06:14<9:50:50,  1.47s/it]
  1%|          | 254/24420 [06:15<9:51:29,  1.47s/it]
  1%|          | 255/24420 [06:17<9:51:58,  1.47s/it]
                                                     
{'loss': 0.39, 'learning_rate': 4.458256624874931e-05, 'epoch': 0.05}

  1%|          | 255/24420 [06:17<9:51:58,  1.47s/it]
  1%|          | 256/24420 [06:18<9:52:37,  1.47s/it]
  1%|          | 257/24420 [06:19<9:52:13,  1.47s/it]
  1%|          | 258/24420 [06:21<9:51:40,  1.47s/it]
  1%|          | 259/24420 [06:22<9:51:27,  1.47s/it]
  1%|          | 260/24420 [06:24<9:50:40,  1.47s/it]
                                                     
{'loss': 0.3778, 'learning_rate': 4.4738795616310163e-05, 'epoch': 0.05}

  1%|          | 260/24420 [06:24<9:50:40,  1.47s/it]
  1%|          | 261/24420 [06:25<9:50:22,  1.47s/it]
  1%|          | 262/24420 [06:27<9:50:47,  1.47s/it]
  1%|          | 263/24420 [06:28<9:50:41,  1.47s/it]
  1%|          | 264/24420 [06:30<9:51:09,  1.47s/it]
  1%|          | 265/24420 [06:31<9:50:32,  1.47s/it]
                                                     
{'loss': 0.3456, 'learning_rate': 4.48920490046898e-05, 'epoch': 0.05}

  1%|          | 265/24420 [06:31<9:50:32,  1.47s/it]
  1%|          | 266/24420 [06:33<9:50:38,  1.47s/it]
  1%|          | 267/24420 [06:34<9:51:10,  1.47s/it]
  1%|          | 268/24420 [06:36<9:54:09,  1.48s/it]
  1%|          | 269/24420 [06:37<9:54:05,  1.48s/it]
  1%|          | 270/24420 [06:39<9:52:55,  1.47s/it]
                                                     
{'loss': 0.4019, 'learning_rate': 4.50424376753519e-05, 'epoch': 0.06}

  1%|          | 270/24420 [06:39<9:52:55,  1.47s/it]
  1%|          | 271/24420 [06:40<9:51:29,  1.47s/it]
  1%|          | 272/24420 [06:42<9:51:03,  1.47s/it]
  1%|          | 273/24420 [06:43<9:50:22,  1.47s/it]
  1%|          | 274/24420 [06:44<9:50:54,  1.47s/it]
  1%|          | 275/24420 [06:46<9:52:58,  1.47s/it]
                                                     
{'loss': 0.3863, 'learning_rate': 4.5190066764568774e-05, 'epoch': 0.06}

  1%|          | 275/24420 [06:46<9:52:58,  1.47s/it]
  1%|          | 276/24420 [06:47<9:53:20,  1.47s/it]
  1%|          | 277/24420 [06:49<9:51:55,  1.47s/it]
  1%|          | 278/24420 [06:50<9:52:28,  1.47s/it]
  1%|          | 279/24420 [06:52<9:52:51,  1.47s/it]
  1%|          | 280/24420 [06:53<9:53:07,  1.47s/it]
                                                     
{'loss': 0.3607, 'learning_rate': 4.533503572493855e-05, 'epoch': 0.06}

  1%|          | 280/24420 [06:53<9:53:07,  1.47s/it]
  1%|          | 281/24420 [06:55<9:53:26,  1.48s/it]
  1%|          | 282/24420 [06:56<9:53:14,  1.47s/it]
  1%|          | 283/24420 [06:58<9:53:50,  1.48s/it]
  1%|          | 284/24420 [06:59<9:53:35,  1.48s/it]
  1%|          | 285/24420 [07:01<9:51:45,  1.47s/it]
                                                     
{'loss': 0.3697, 'learning_rate': 4.547743872782376e-05, 'epoch': 0.06}

  1%|          | 285/24420 [07:01<9:51:45,  1.47s/it]
  1%|          | 286/24420 [07:02<9:51:38,  1.47s/it]
  1%|          | 287/24420 [07:04<9:51:09,  1.47s/it]
  1%|          | 288/24420 [07:05<9:52:17,  1.47s/it]
  1%|          | 289/24420 [07:07<9:51:52,  1.47s/it]
  1%|          | 290/24420 [07:08<9:52:13,  1.47s/it]
                                                     
{'loss': 0.361, 'learning_rate': 4.5617365030789545e-05, 'epoch': 0.06}

  1%|          | 290/24420 [07:08<9:52:13,  1.47s/it]
  1%|          | 291/24420 [07:10<9:53:40,  1.48s/it]
  1%|          | 292/24420 [07:11<9:52:21,  1.47s/it]
  1%|          | 293/24420 [07:12<9:51:38,  1.47s/it]
  1%|          | 294/24420 [07:14<9:51:08,  1.47s/it]
  1%|          | 295/24420 [07:15<9:51:47,  1.47s/it]
                                                     
{'loss': 0.3761, 'learning_rate': 4.575489931363226e-05, 'epoch': 0.06}

  1%|          | 295/24420 [07:15<9:51:47,  1.47s/it]
  1%|          | 296/24420 [07:17<9:54:12,  1.48s/it]
  1%|          | 297/24420 [07:18<9:52:33,  1.47s/it]
  1%|          | 298/24420 [07:20<9:51:37,  1.47s/it]
  1%|          | 299/24420 [07:21<9:52:28,  1.47s/it]
  1%|          | 300/24420 [07:23<9:53:01,  1.48s/it]
                                                     
{'loss': 0.3722, 'learning_rate': 4.589012198616609e-05, 'epoch': 0.06}

  1%|          | 300/24420 [07:23<9:53:01,  1.48s/it]
  1%|          | 301/24420 [07:24<9:52:57,  1.48s/it]
  1%|          | 302/24420 [07:26<9:54:01,  1.48s/it]
  1%|          | 303/24420 [07:27<9:54:09,  1.48s/it]
  1%|          | 304/24420 [07:29<9:53:06,  1.48s/it]
  1%|          | 305/24420 [07:30<9:52:11,  1.47s/it]
                                                     
{'loss': 0.3887, 'learning_rate': 4.602310947056923e-05, 'epoch': 0.06}

  1%|          | 305/24420 [07:30<9:52:11,  1.47s/it]
  1%|▏         | 306/24420 [07:32<9:56:09,  1.48s/it]
  1%|▏         | 307/24420 [07:33<9:55:17,  1.48s/it]
  1%|▏         | 308/24420 [07:35<9:55:08,  1.48s/it]
  1%|▏         | 309/24420 [07:36<9:54:56,  1.48s/it]
  1%|▏         | 310/24420 [07:38<9:54:38,  1.48s/it]
                                                     
{'loss': 0.4047, 'learning_rate': 4.615393446077182e-05, 'epoch': 0.06}

  1%|▏         | 310/24420 [07:38<9:54:38,  1.48s/it]
  1%|▏         | 311/24420 [07:39<9:52:54,  1.48s/it]
  1%|▏         | 312/24420 [07:41<9:53:31,  1.48s/it]
  1%|▏         | 313/24420 [07:42<9:53:37,  1.48s/it]
  1%|▏         | 314/24420 [07:43<9:53:05,  1.48s/it]
  1%|▏         | 315/24420 [07:45<9:52:33,  1.47s/it]
                                                     
{'loss': 0.343, 'learning_rate': 4.6282666161090166e-05, 'epoch': 0.06}

  1%|▏         | 315/24420 [07:45<9:52:33,  1.47s/it]
  1%|▏         | 316/24420 [07:46<9:54:02,  1.48s/it]
  1%|▏         | 317/24420 [07:48<9:53:16,  1.48s/it]
  1%|▏         | 318/24420 [07:49<9:52:12,  1.47s/it]
  1%|▏         | 319/24420 [07:51<9:52:14,  1.47s/it]
  1%|▏         | 320/24420 [07:52<9:52:13,  1.47s/it]
                                                     
{'loss': 0.351, 'learning_rate': 4.640937050606839e-05, 'epoch': 0.07}

  1%|▏         | 320/24420 [07:52<9:52:13,  1.47s/it]
  1%|▏         | 321/24420 [07:54<9:51:26,  1.47s/it]
  1%|▏         | 322/24420 [07:55<9:49:53,  1.47s/it]
  1%|▏         | 323/24420 [07:57<9:50:40,  1.47s/it]
  1%|▏         | 324/24420 [07:58<9:52:34,  1.48s/it]
  1%|▏         | 325/24420 [08:00<9:50:57,  1.47s/it]
                                                     
{'loss': 0.3758, 'learning_rate': 4.653411036327597e-05, 'epoch': 0.07}

  1%|▏         | 325/24420 [08:00<9:50:57,  1.47s/it]
  1%|▏         | 326/24420 [08:01<9:50:40,  1.47s/it]
  1%|▏         | 327/24420 [08:03<9:51:05,  1.47s/it]
  1%|▏         | 328/24420 [08:04<9:50:43,  1.47s/it]
  1%|▏         | 329/24420 [08:06<9:50:34,  1.47s/it]
  1%|▏         | 330/24420 [08:07<9:50:38,  1.47s/it]
                                                     
{'loss': 0.3506, 'learning_rate': 4.665694572062267e-05, 'epoch': 0.07}

  1%|▏         | 330/24420 [08:07<9:50:38,  1.47s/it]
  1%|▏         | 331/24420 [08:09<9:50:47,  1.47s/it]
  1%|▏         | 332/24420 [08:10<9:51:26,  1.47s/it]
  1%|▏         | 333/24420 [08:11<9:50:40,  1.47s/it]
  1%|▏         | 334/24420 [08:13<9:50:39,  1.47s/it]
  1%|▏         | 335/24420 [08:14<9:49:52,  1.47s/it]
                                                     
{'loss': 0.3582, 'learning_rate': 4.677793385958802e-05, 'epoch': 0.07}

  1%|▏         | 335/24420 [08:14<9:49:52,  1.47s/it]
  1%|▏         | 336/24420 [08:16<9:49:45,  1.47s/it]
  1%|▏         | 337/24420 [08:17<9:50:15,  1.47s/it]
  1%|▏         | 338/24420 [08:19<9:50:31,  1.47s/it]
  1%|▏         | 339/24420 [08:20<9:50:45,  1.47s/it]
  1%|▏         | 340/24420 [08:22<9:48:56,  1.47s/it]
                                                     
{'loss': 0.3527, 'learning_rate': 4.689712951561742e-05, 'epoch': 0.07}

  1%|▏         | 340/24420 [08:22<9:48:56,  1.47s/it]
  1%|▏         | 341/24420 [08:23<9:48:57,  1.47s/it]
  1%|▏         | 342/24420 [08:25<9:48:54,  1.47s/it]
  1%|▏         | 343/24420 [08:26<9:48:54,  1.47s/it]
  1%|▏         | 344/24420 [08:28<9:50:28,  1.47s/it]
  1%|▏         | 345/24420 [08:29<9:50:11,  1.47s/it]
                                                     
{'loss': 0.4034, 'learning_rate': 4.701458502680934e-05, 'epoch': 0.07}

  1%|▏         | 345/24420 [08:29<9:50:11,  1.47s/it]
  1%|▏         | 346/24420 [08:31<9:49:36,  1.47s/it]
  1%|▏         | 347/24420 [08:32<9:49:23,  1.47s/it]
  1%|▏         | 348/24420 [08:34<9:50:15,  1.47s/it]
  1%|▏         | 349/24420 [08:35<9:49:46,  1.47s/it]
  1%|▏         | 350/24420 [08:36<9:48:51,  1.47s/it]
                                                     
{'loss': 0.4317, 'learning_rate': 4.713035047190436e-05, 'epoch': 0.07}

  1%|▏         | 350/24420 [08:36<9:48:51,  1.47s/it]
  1%|▏         | 351/24420 [08:38<9:49:18,  1.47s/it]
  1%|▏         | 352/24420 [08:39<9:48:33,  1.47s/it]
  1%|▏         | 353/24420 [08:41<9:48:34,  1.47s/it]
  1%|▏         | 354/24420 [08:42<9:48:37,  1.47s/it]
  1%|▏         | 355/24420 [08:44<9:48:09,  1.47s/it]
                                                     
{'loss': 0.3733, 'learning_rate': 4.7244473798486756e-05, 'epoch': 0.07}

  1%|▏         | 355/24420 [08:44<9:48:09,  1.47s/it]
  1%|▏         | 356/24420 [08:45<9:49:39,  1.47s/it]
  1%|▏         | 357/24420 [08:47<9:51:12,  1.47s/it]
  1%|▏         | 358/24420 [08:48<9:51:23,  1.47s/it]
  1%|▏         | 359/24420 [08:50<9:50:43,  1.47s/it]
  1%|▏         | 360/24420 [08:51<9:50:57,  1.47s/it]
                                                     
{'loss': 0.3894, 'learning_rate': 4.735700094222001e-05, 'epoch': 0.07}

  1%|▏         | 360/24420 [08:51<9:50:57,  1.47s/it]
  1%|▏         | 361/24420 [08:53<9:50:32,  1.47s/it]
  1%|▏         | 362/24420 [08:54<9:49:28,  1.47s/it]
  1%|▏         | 363/24420 [08:56<9:48:40,  1.47s/it]
  1%|▏         | 364/24420 [08:57<9:48:39,  1.47s/it]
  1%|▏         | 365/24420 [08:58<9:49:04,  1.47s/it]
                                                     
{'loss': 0.3961, 'learning_rate': 4.746797593785841e-05, 'epoch': 0.07}

  1%|▏         | 365/24420 [08:58<9:49:04,  1.47s/it]
  1%|▏         | 366/24420 [09:00<9:49:46,  1.47s/it]
  2%|▏         | 367/24420 [09:01<9:51:09,  1.47s/it]
  2%|▏         | 368/24420 [09:03<9:50:44,  1.47s/it]
  2%|▏         | 369/24420 [09:04<9:49:54,  1.47s/it]
  2%|▏         | 370/24420 [09:06<9:48:53,  1.47s/it]
                                                     
{'loss': 0.4013, 'learning_rate': 4.7577441022706095e-05, 'epoch': 0.08}

  2%|▏         | 370/24420 [09:06<9:48:53,  1.47s/it]
  2%|▏         | 371/24420 [09:07<9:47:55,  1.47s/it]
  2%|▏         | 372/24420 [09:09<9:48:08,  1.47s/it]
  2%|▏         | 373/24420 [09:10<9:49:01,  1.47s/it]
  2%|▏         | 374/24420 [09:12<9:49:42,  1.47s/it]
  2%|▏         | 375/24420 [09:13<9:50:07,  1.47s/it]
                                                     
{'loss': 0.3634, 'learning_rate': 4.76854367331319e-05, 'epoch': 0.08}

  2%|▏         | 375/24420 [09:13<9:50:07,  1.47s/it]
  2%|▏         | 376/24420 [09:15<9:50:10,  1.47s/it]
  2%|▏         | 377/24420 [09:16<9:49:49,  1.47s/it]
  2%|▏         | 378/24420 [09:18<9:50:10,  1.47s/it]
  2%|▏         | 379/24420 [09:19<9:49:33,  1.47s/it]
  2%|▏         | 380/24420 [09:21<9:48:37,  1.47s/it]
                                                     
{'loss': 0.358, 'learning_rate': 4.7792001994691866e-05, 'epoch': 0.08}

  2%|▏         | 380/24420 [09:21<9:48:37,  1.47s/it]
  2%|▏         | 381/24420 [09:22<9:48:20,  1.47s/it]
  2%|▏         | 382/24420 [09:23<9:48:17,  1.47s/it]
  2%|▏         | 383/24420 [09:25<9:47:44,  1.47s/it]
  2%|▏         | 384/24420 [09:26<9:46:36,  1.46s/it]
  2%|▏         | 385/24420 [09:28<9:46:11,  1.46s/it]
                                                     
{'loss': 0.3685, 'learning_rate': 4.7897174206360945e-05, 'epoch': 0.08}

  2%|▏         | 385/24420 [09:28<9:46:11,  1.46s/it]
  2%|▏         | 386/24420 [09:29<9:46:11,  1.46s/it]
  2%|▏         | 387/24420 [09:31<9:46:47,  1.46s/it]
  2%|▏         | 388/24420 [09:32<9:47:49,  1.47s/it]
  2%|▏         | 389/24420 [09:34<9:49:03,  1.47s/it]
  2%|▏         | 390/24420 [09:35<9:48:19,  1.47s/it]
                                                     
{'loss': 0.3448, 'learning_rate': 4.800098931932989e-05, 'epoch': 0.08}

  2%|▏         | 390/24420 [09:35<9:48:19,  1.47s/it]
  2%|▏         | 391/24420 [09:37<9:49:26,  1.47s/it]
  2%|▏         | 392/24420 [09:38<9:49:38,  1.47s/it]
  2%|▏         | 393/24420 [09:40<9:49:45,  1.47s/it]
  2%|▏         | 394/24420 [09:41<9:48:33,  1.47s/it]
  2%|▏         | 395/24420 [09:43<9:48:02,  1.47s/it]
                                                     
{'loss': 0.3617, 'learning_rate': 4.810348191078279e-05, 'epoch': 0.08}

  2%|▏         | 395/24420 [09:43<9:48:02,  1.47s/it]
  2%|▏         | 396/24420 [09:44<9:48:08,  1.47s/it]
  2%|▏         | 397/24420 [09:46<9:47:26,  1.47s/it]
  2%|▏         | 398/24420 [09:47<9:47:44,  1.47s/it]
  2%|▏         | 399/24420 [09:48<9:46:50,  1.47s/it]
  2%|▏         | 400/24420 [09:50<9:46:16,  1.46s/it]
                                                     
{'loss': 0.3189, 'learning_rate': 4.82046852530342e-05, 'epoch': 0.08}

  2%|▏         | 400/24420 [09:50<9:46:16,  1.46s/it]
  2%|▏         | 401/24420 [09:54<15:37:54,  2.34s/it]
  2%|▏         | 402/24420 [09:56<13:51:05,  2.08s/it]
  2%|▏         | 403/24420 [09:57<12:44:51,  1.91s/it]
  2%|▏         | 404/24420 [09:59<11:51:12,  1.78s/it]
  2%|▏         | 405/24420 [10:00<11:12:49,  1.68s/it]
                                                      
{'loss': 0.3415, 'learning_rate': 4.830463137837162e-05, 'epoch': 0.08}

  2%|▏         | 405/24420 [10:00<11:12:49,  1.68s/it]
  2%|▏         | 406/24420 [10:02<10:47:27,  1.62s/it]
  2%|▏         | 407/24420 [10:03<10:28:48,  1.57s/it]
  2%|▏         | 408/24420 [10:05<10:16:08,  1.54s/it]
  2%|▏         | 409/24420 [10:06<10:08:14,  1.52s/it]
  2%|▏         | 410/24420 [10:08<10:02:28,  1.51s/it]
                                                      
{'loss': 0.3917, 'learning_rate': 4.8403351139919656e-05, 'epoch': 0.08}

  2%|▏         | 410/24420 [10:08<10:02:28,  1.51s/it]
  2%|▏         | 411/24420 [10:09<9:58:14,  1.50s/it] 
  2%|▏         | 412/24420 [10:10<9:54:41,  1.49s/it]
  2%|▏         | 413/24420 [10:12<9:53:19,  1.48s/it]
  2%|▏         | 414/24420 [10:13<9:50:49,  1.48s/it]
  2%|▏         | 415/24420 [10:15<9:50:46,  1.48s/it]
                                                     
{'loss': 0.3584, 'learning_rate': 4.850087426881512e-05, 'epoch': 0.08}

  2%|▏         | 415/24420 [10:15<9:50:46,  1.48s/it]
  2%|▏         | 416/24420 [10:16<9:51:29,  1.48s/it]
  2%|▏         | 417/24420 [10:18<9:50:02,  1.47s/it]
  2%|▏         | 418/24420 [10:19<9:48:41,  1.47s/it]
  2%|▏         | 419/24420 [10:21<9:48:42,  1.47s/it]
  2%|▏         | 420/24420 [10:22<9:49:53,  1.47s/it]
                                                     
{'loss': 0.3367, 'learning_rate': 4.859722942795827e-05, 'epoch': 0.09}

  2%|▏         | 420/24420 [10:22<9:49:53,  1.47s/it]
  2%|▏         | 421/24420 [10:24<9:51:05,  1.48s/it]
  2%|▏         | 422/24420 [10:25<9:51:19,  1.48s/it]
  2%|▏         | 423/24420 [10:27<9:49:53,  1.47s/it]
  2%|▏         | 424/24420 [10:28<9:49:24,  1.47s/it]
  2%|▏         | 425/24420 [10:30<9:49:30,  1.47s/it]
                                                     
{'loss': 0.3694, 'learning_rate': 4.8692444262583224e-05, 'epoch': 0.09}

  2%|▏         | 425/24420 [10:30<9:49:30,  1.47s/it]
  2%|▏         | 426/24420 [10:31<9:49:21,  1.47s/it]
  2%|▏         | 427/24420 [10:33<9:49:46,  1.47s/it]
  2%|▏         | 428/24420 [10:34<9:49:56,  1.48s/it]
  2%|▏         | 429/24420 [10:36<9:50:07,  1.48s/it]
  2%|▏         | 430/24420 [10:37<9:49:31,  1.47s/it]
                                                     
{'loss': 0.3636, 'learning_rate': 4.8786545447870833e-05, 'epoch': 0.09}

  2%|▏         | 430/24420 [10:37<9:49:31,  1.47s/it]
  2%|▏         | 431/24420 [10:38<9:49:28,  1.47s/it]
  2%|▏         | 432/24420 [10:40<9:49:35,  1.47s/it]
  2%|▏         | 433/24420 [10:41<9:49:00,  1.47s/it]
  2%|▏         | 434/24420 [10:43<9:49:17,  1.47s/it]
  2%|▏         | 435/24420 [10:44<9:48:47,  1.47s/it]
                                                     
{'loss': 0.3147, 'learning_rate': 4.8879558733809264e-05, 'epoch': 0.09}

  2%|▏         | 435/24420 [10:44<9:48:47,  1.47s/it]
  2%|▏         | 436/24420 [10:46<9:49:05,  1.47s/it]
  2%|▏         | 437/24420 [10:47<9:48:58,  1.47s/it]
  2%|▏         | 438/24420 [10:49<9:49:58,  1.48s/it]
  2%|▏         | 439/24420 [10:50<9:50:04,  1.48s/it]
  2%|▏         | 440/24420 [10:52<9:50:44,  1.48s/it]
                                                     
{'loss': 0.3436, 'learning_rate': 4.897150898749078e-05, 'epoch': 0.09}

  2%|▏         | 440/24420 [10:52<9:50:44,  1.48s/it]
  2%|▏         | 441/24420 [10:53<9:52:11,  1.48s/it]
  2%|▏         | 442/24420 [10:55<9:51:15,  1.48s/it]
  2%|▏         | 443/24420 [10:56<9:50:09,  1.48s/it]
  2%|▏         | 444/24420 [10:58<9:50:27,  1.48s/it]
  2%|▏         | 445/24420 [10:59<9:50:38,  1.48s/it]
                                                     
{'loss': 0.3808, 'learning_rate': 4.9062420233018894e-05, 'epoch': 0.09}

  2%|▏         | 445/24420 [10:59<9:50:38,  1.48s/it]
  2%|▏         | 446/24420 [11:01<9:51:02,  1.48s/it]
  2%|▏         | 447/24420 [11:02<9:49:40,  1.48s/it]
  2%|▏         | 448/24420 [11:04<9:49:27,  1.48s/it]
  2%|▏         | 449/24420 [11:05<9:51:43,  1.48s/it]
  2%|▏         | 450/24420 [11:07<9:49:02,  1.47s/it]
                                                     
{'loss': 0.326, 'learning_rate': 4.915231568918581e-05, 'epoch': 0.09}

  2%|▏         | 450/24420 [11:07<9:49:02,  1.47s/it]
  2%|▏         | 451/24420 [11:08<9:49:04,  1.47s/it]
  2%|▏         | 452/24420 [11:10<9:54:14,  1.49s/it]
  2%|▏         | 453/24420 [11:11<9:54:27,  1.49s/it]
  2%|▏         | 454/24420 [11:13<9:54:23,  1.49s/it]
  2%|▏         | 455/24420 [11:14<9:53:53,  1.49s/it]
                                                     
{'loss': 0.3527, 'learning_rate': 4.924121780506815e-05, 'epoch': 0.09}

  2%|▏         | 455/24420 [11:14<9:53:53,  1.49s/it]
  2%|▏         | 456/24420 [11:15<9:53:06,  1.49s/it]
  2%|▏         | 457/24420 [11:17<9:51:28,  1.48s/it]
  2%|▏         | 458/24420 [11:18<9:50:31,  1.48s/it]
  2%|▏         | 459/24420 [11:20<9:50:14,  1.48s/it]
  2%|▏         | 460/24420 [11:21<9:51:20,  1.48s/it]
                                                     
{'loss': 0.3319, 'learning_rate': 4.932914829367744e-05, 'epoch': 0.09}

  2%|▏         | 460/24420 [11:21<9:51:20,  1.48s/it]
  2%|▏         | 461/24420 [11:23<9:52:13,  1.48s/it]
  2%|▏         | 462/24420 [11:24<9:51:58,  1.48s/it]
  2%|▏         | 463/24420 [11:26<9:51:51,  1.48s/it]
  2%|▏         | 464/24420 [11:27<9:53:18,  1.49s/it]
  2%|▏         | 465/24420 [11:29<9:51:52,  1.48s/it]
                                                     
{'loss': 0.3662, 'learning_rate': 4.941612816379154e-05, 'epoch': 0.1}

  2%|▏         | 465/24420 [11:29<9:51:52,  1.48s/it]
  2%|▏         | 466/24420 [11:30<9:51:15,  1.48s/it]
  2%|▏         | 467/24420 [11:32<9:49:52,  1.48s/it]
  2%|▏         | 468/24420 [11:33<9:49:40,  1.48s/it]
  2%|▏         | 469/24420 [11:35<9:50:43,  1.48s/it]
  2%|▏         | 470/24420 [11:36<9:51:04,  1.48s/it]
                                                     
{'loss': 0.3433, 'learning_rate': 4.950217775008374e-05, 'epoch': 0.1}

  2%|▏         | 470/24420 [11:36<9:51:04,  1.48s/it]
  2%|▏         | 471/24420 [11:38<9:50:40,  1.48s/it]
  2%|▏         | 472/24420 [11:39<9:49:04,  1.48s/it]
  2%|▏         | 473/24420 [11:41<9:48:28,  1.47s/it]
  2%|▏         | 474/24420 [11:42<9:47:47,  1.47s/it]
  2%|▏         | 475/24420 [11:44<9:46:41,  1.47s/it]
                                                     
{'loss': 0.3392, 'learning_rate': 4.958731674165767e-05, 'epoch': 0.1}

  2%|▏         | 475/24420 [11:44<9:46:41,  1.47s/it]
  2%|▏         | 476/24420 [11:45<9:46:19,  1.47s/it]
  2%|▏         | 477/24420 [11:46<9:46:26,  1.47s/it]
  2%|▏         | 478/24420 [11:48<9:47:07,  1.47s/it]
  2%|▏         | 479/24420 [11:49<9:46:46,  1.47s/it]
  2%|▏         | 480/24420 [11:51<9:46:33,  1.47s/it]
                                                     
{'loss': 0.3569, 'learning_rate': 4.967156420908811e-05, 'epoch': 0.1}

  2%|▏         | 480/24420 [11:51<9:46:33,  1.47s/it]
  2%|▏         | 481/24420 [11:52<9:46:23,  1.47s/it]
  2%|▏         | 482/24420 [11:54<9:46:55,  1.47s/it]
  2%|▏         | 483/24420 [11:55<9:46:58,  1.47s/it]
  2%|▏         | 484/24420 [11:57<9:48:37,  1.48s/it]
  2%|▏         | 485/24420 [11:58<9:49:27,  1.48s/it]
                                                     
{'loss': 0.3795, 'learning_rate': 4.975493863006067e-05, 'epoch': 0.1}

  2%|▏         | 485/24420 [11:58<9:49:27,  1.48s/it]
  2%|▏         | 486/24420 [12:00<9:50:07,  1.48s/it]
  2%|▏         | 487/24420 [12:01<9:48:54,  1.48s/it]
  2%|▏         | 488/24420 [12:03<9:49:34,  1.48s/it]
  2%|▏         | 489/24420 [12:04<9:48:08,  1.47s/it]
  2%|▏         | 490/24420 [12:06<9:47:31,  1.47s/it]
                                                     
{'loss': 0.3499, 'learning_rate': 4.9837457913696535e-05, 'epoch': 0.1}

  2%|▏         | 490/24420 [12:06<9:47:31,  1.47s/it]
  2%|▏         | 491/24420 [12:07<9:47:40,  1.47s/it]
  2%|▏         | 492/24420 [12:09<9:47:16,  1.47s/it]
  2%|▏         | 493/24420 [12:10<9:46:09,  1.47s/it]
  2%|▏         | 494/24420 [12:12<9:46:05,  1.47s/it]
  2%|▏         | 495/24420 [12:13<9:45:23,  1.47s/it]
                                                     
{'loss': 0.3757, 'learning_rate': 4.99191394236424e-05, 'epoch': 0.1}

  2%|▏         | 495/24420 [12:13<9:45:23,  1.47s/it]
  2%|▏         | 496/24420 [12:14<9:46:10,  1.47s/it]
  2%|▏         | 497/24420 [12:16<9:46:15,  1.47s/it]
  2%|▏         | 498/24420 [12:17<9:45:10,  1.47s/it]
  2%|▏         | 499/24420 [12:19<9:46:23,  1.47s/it]
  2%|▏         | 500/24420 [12:20<9:46:46,  1.47s/it]
                                                     
{'loss': 0.3545, 'learning_rate': 5e-05, 'epoch': 0.1}

  2%|▏         | 500/24420 [12:20<9:46:46,  1.47s/it]
  2%|▏         | 501/24420 [12:22<9:47:20,  1.47s/it]
  2%|▏         | 502/24420 [12:23<9:50:39,  1.48s/it]
  2%|▏         | 503/24420 [12:25<9:49:56,  1.48s/it]
  2%|▏         | 504/24420 [12:26<9:49:29,  1.48s/it]
  2%|▏         | 505/24420 [12:28<9:48:39,  1.48s/it]
                                                     
{'loss': 0.3111, 'learning_rate': 5e-05, 'epoch': 0.1}

  2%|▏         | 505/24420 [12:28<9:48:39,  1.48s/it]
  2%|▏         | 506/24420 [12:29<9:48:48,  1.48s/it]
  2%|▏         | 507/24420 [12:31<9:48:59,  1.48s/it]
  2%|▏         | 508/24420 [12:32<9:47:56,  1.48s/it]
  2%|▏         | 509/24420 [12:34<9:48:37,  1.48s/it]
  2%|▏         | 510/24420 [12:35<9:48:42,  1.48s/it]
                                                     
{'loss': 0.3516, 'learning_rate': 5e-05, 'epoch': 0.1}

  2%|▏         | 510/24420 [12:35<9:48:42,  1.48s/it]
  2%|▏         | 511/24420 [12:37<9:47:45,  1.47s/it]
  2%|▏         | 512/24420 [12:38<9:47:05,  1.47s/it]
  2%|▏         | 513/24420 [12:40<9:46:43,  1.47s/it]
  2%|▏         | 514/24420 [12:41<9:47:13,  1.47s/it]
  2%|▏         | 515/24420 [12:42<9:47:48,  1.48s/it]
                                                     
{'loss': 0.3211, 'learning_rate': 5e-05, 'epoch': 0.11}

  2%|▏         | 515/24420 [12:43<9:47:48,  1.48s/it]
  2%|▏         | 516/24420 [12:44<9:49:08,  1.48s/it]
  2%|▏         | 517/24420 [12:45<9:52:17,  1.49s/it]
  2%|▏         | 518/24420 [12:47<9:51:55,  1.49s/it]
  2%|▏         | 519/24420 [12:48<9:49:44,  1.48s/it]
  2%|▏         | 520/24420 [12:50<9:49:50,  1.48s/it]
                                                     
{'loss': 0.342, 'learning_rate': 5e-05, 'epoch': 0.11}

  2%|▏         | 520/24420 [12:50<9:49:50,  1.48s/it]
  2%|▏         | 521/24420 [12:51<9:50:20,  1.48s/it]
  2%|▏         | 522/24420 [12:53<9:48:31,  1.48s/it]
  2%|▏         | 523/24420 [12:54<9:47:32,  1.48s/it]
  2%|▏         | 524/24420 [12:56<9:47:16,  1.47s/it]
  2%|▏         | 525/24420 [12:57<9:46:42,  1.47s/it]
                                                     
{'loss': 0.3216, 'learning_rate': 5e-05, 'epoch': 0.11}

  2%|▏         | 525/24420 [12:57<9:46:42,  1.47s/it]
  2%|▏         | 526/24420 [12:59<9:45:52,  1.47s/it]
  2%|▏         | 527/24420 [13:00<9:47:15,  1.47s/it]
  2%|▏         | 528/24420 [13:02<9:48:13,  1.48s/it]
  2%|▏         | 529/24420 [13:03<9:47:07,  1.47s/it]
  2%|▏         | 530/24420 [13:05<9:47:11,  1.47s/it]
                                                     
{'loss': 0.3076, 'learning_rate': 5e-05, 'epoch': 0.11}

  2%|▏         | 530/24420 [13:05<9:47:11,  1.47s/it]
  2%|▏         | 531/24420 [13:06<9:46:36,  1.47s/it]
  2%|▏         | 532/24420 [13:08<9:45:08,  1.47s/it]
  2%|▏         | 533/24420 [13:09<9:45:34,  1.47s/it]
  2%|▏         | 534/24420 [13:11<9:45:44,  1.47s/it]
  2%|▏         | 535/24420 [13:12<9:46:38,  1.47s/it]
                                                     
{'loss': 0.3308, 'learning_rate': 5e-05, 'epoch': 0.11}

  2%|▏         | 535/24420 [13:12<9:46:38,  1.47s/it]
  2%|▏         | 536/24420 [13:14<9:47:29,  1.48s/it]
  2%|▏         | 537/24420 [13:15<9:46:58,  1.47s/it]
  2%|▏         | 538/24420 [13:16<9:46:53,  1.47s/it]
  2%|▏         | 539/24420 [13:18<9:46:20,  1.47s/it]
  2%|▏         | 540/24420 [13:19<9:44:28,  1.47s/it]
                                                     
{'loss': 0.3662, 'learning_rate': 5e-05, 'epoch': 0.11}

  2%|▏         | 540/24420 [13:19<9:44:28,  1.47s/it]
  2%|▏         | 541/24420 [13:21<9:45:07,  1.47s/it]
  2%|▏         | 542/24420 [13:22<9:45:18,  1.47s/it]
  2%|▏         | 543/24420 [13:24<9:44:51,  1.47s/it]
  2%|▏         | 544/24420 [13:25<9:44:58,  1.47s/it]
  2%|▏         | 545/24420 [13:27<9:45:16,  1.47s/it]
                                                     
{'loss': 0.3298, 'learning_rate': 5e-05, 'epoch': 0.11}

  2%|▏         | 545/24420 [13:27<9:45:16,  1.47s/it]
  2%|▏         | 546/24420 [13:28<9:46:20,  1.47s/it]
  2%|▏         | 547/24420 [13:30<9:46:50,  1.47s/it]
  2%|▏         | 548/24420 [13:31<9:46:31,  1.47s/it]
  2%|▏         | 549/24420 [13:33<9:46:05,  1.47s/it]
  2%|▏         | 550/24420 [13:34<9:52:10,  1.49s/it]
                                                     
{'loss': 0.3414, 'learning_rate': 5e-05, 'epoch': 0.11}

  2%|▏         | 550/24420 [13:34<9:52:10,  1.49s/it]
  2%|▏         | 551/24420 [13:36<9:49:59,  1.48s/it]
  2%|▏         | 552/24420 [13:37<9:48:32,  1.48s/it]
  2%|▏         | 553/24420 [13:39<9:47:22,  1.48s/it]
  2%|▏         | 554/24420 [13:40<9:45:33,  1.47s/it]
  2%|▏         | 555/24420 [13:42<9:46:21,  1.47s/it]
                                                     
{'loss': 0.3214, 'learning_rate': 5e-05, 'epoch': 0.11}

  2%|▏         | 555/24420 [13:42<9:46:21,  1.47s/it]
  2%|▏         | 556/24420 [13:43<9:47:07,  1.48s/it]
  2%|▏         | 557/24420 [13:44<9:46:54,  1.48s/it]
  2%|▏         | 558/24420 [13:46<9:45:07,  1.47s/it]
  2%|▏         | 559/24420 [13:47<9:45:42,  1.47s/it]
  2%|▏         | 560/24420 [13:49<9:45:44,  1.47s/it]
                                                     
{'loss': 0.3061, 'learning_rate': 5e-05, 'epoch': 0.11}

  2%|▏         | 560/24420 [13:49<9:45:44,  1.47s/it]
  2%|▏         | 561/24420 [13:50<9:45:53,  1.47s/it]
  2%|▏         | 562/24420 [13:52<9:46:09,  1.47s/it]
  2%|▏         | 563/24420 [13:53<9:46:22,  1.47s/it]
  2%|▏         | 564/24420 [13:55<9:45:11,  1.47s/it]
  2%|▏         | 565/24420 [13:56<9:44:56,  1.47s/it]
                                                     
{'loss': 0.3408, 'learning_rate': 5e-05, 'epoch': 0.12}

  2%|▏         | 565/24420 [13:56<9:44:56,  1.47s/it]
  2%|▏         | 566/24420 [13:58<9:47:50,  1.48s/it]
  2%|▏         | 567/24420 [13:59<9:48:26,  1.48s/it]
  2%|▏         | 568/24420 [14:01<9:47:08,  1.48s/it]
  2%|▏         | 569/24420 [14:02<9:45:19,  1.47s/it]
  2%|▏         | 570/24420 [14:04<9:45:39,  1.47s/it]
                                                     
{'loss': 0.2838, 'learning_rate': 5e-05, 'epoch': 0.12}

  2%|▏         | 570/24420 [14:04<9:45:39,  1.47s/it]
  2%|▏         | 571/24420 [14:05<9:46:40,  1.48s/it]
  2%|▏         | 572/24420 [14:07<9:46:26,  1.48s/it]
  2%|▏         | 573/24420 [14:08<9:45:37,  1.47s/it]
  2%|▏         | 574/24420 [14:10<9:45:39,  1.47s/it]
  2%|▏         | 575/24420 [14:11<9:45:11,  1.47s/it]
                                                     
{'loss': 0.3027, 'learning_rate': 5e-05, 'epoch': 0.12}

  2%|▏         | 575/24420 [14:11<9:45:11,  1.47s/it]
  2%|▏         | 576/24420 [14:12<9:46:04,  1.47s/it]
  2%|▏         | 577/24420 [14:14<9:46:09,  1.48s/it]
  2%|▏         | 578/24420 [14:15<9:45:45,  1.47s/it]
  2%|▏         | 579/24420 [14:17<9:45:13,  1.47s/it]
  2%|▏         | 580/24420 [14:18<9:45:47,  1.47s/it]
                                                     
{'loss': 0.2984, 'learning_rate': 5e-05, 'epoch': 0.12}

  2%|▏         | 580/24420 [14:18<9:45:47,  1.47s/it]
  2%|▏         | 581/24420 [14:20<9:46:34,  1.48s/it]
  2%|▏         | 582/24420 [14:21<9:46:24,  1.48s/it]
  2%|▏         | 583/24420 [14:23<9:46:40,  1.48s/it]
  2%|▏         | 584/24420 [14:24<9:45:36,  1.47s/it]
  2%|▏         | 585/24420 [14:26<9:45:52,  1.47s/it]
                                                     
{'loss': 0.3014, 'learning_rate': 5e-05, 'epoch': 0.12}

  2%|▏         | 585/24420 [14:26<9:45:52,  1.47s/it]
  2%|▏         | 586/24420 [14:27<9:46:48,  1.48s/it]
  2%|▏         | 587/24420 [14:29<9:46:22,  1.48s/it]
  2%|▏         | 588/24420 [14:30<9:45:36,  1.47s/it]
  2%|▏         | 589/24420 [14:32<9:46:25,  1.48s/it]
  2%|▏         | 590/24420 [14:33<9:46:14,  1.48s/it]
                                                     
{'loss': 0.3216, 'learning_rate': 5e-05, 'epoch': 0.12}

  2%|▏         | 590/24420 [14:33<9:46:14,  1.48s/it]
  2%|▏         | 591/24420 [14:35<9:46:30,  1.48s/it]
  2%|▏         | 592/24420 [14:36<9:45:58,  1.48s/it]
  2%|▏         | 593/24420 [14:38<9:45:50,  1.48s/it]
  2%|▏         | 594/24420 [14:39<9:45:59,  1.48s/it]
  2%|▏         | 595/24420 [14:41<9:45:35,  1.47s/it]
                                                     
{'loss': 0.3158, 'learning_rate': 5e-05, 'epoch': 0.12}

  2%|▏         | 595/24420 [14:41<9:45:35,  1.47s/it]
  2%|▏         | 596/24420 [14:42<9:45:05,  1.47s/it]
  2%|▏         | 597/24420 [14:43<9:46:34,  1.48s/it]
  2%|▏         | 598/24420 [14:45<9:45:50,  1.48s/it]
  2%|▏         | 599/24420 [14:46<9:48:17,  1.48s/it]
  2%|▏         | 600/24420 [14:48<9:47:58,  1.48s/it]
                                                     
{'loss': 0.3583, 'learning_rate': 5e-05, 'epoch': 0.12}

  2%|▏         | 600/24420 [14:48<9:47:58,  1.48s/it]
  2%|▏         | 601/24420 [14:53<16:21:18,  2.47s/it]
  2%|▏         | 602/24420 [14:54<14:25:19,  2.18s/it]
  2%|▏         | 603/24420 [14:56<13:04:42,  1.98s/it]
  2%|▏         | 604/24420 [14:57<12:10:12,  1.84s/it]
  2%|▏         | 605/24420 [14:59<11:25:47,  1.73s/it]
                                                      
{'loss': 0.3694, 'learning_rate': 5e-05, 'epoch': 0.12}

  2%|▏         | 605/24420 [14:59<11:25:47,  1.73s/it]
  2%|▏         | 606/24420 [15:00<10:54:23,  1.65s/it]
  2%|▏         | 607/24420 [15:02<10:35:04,  1.60s/it]
  2%|▏         | 608/24420 [15:03<10:20:55,  1.56s/it]
  2%|▏         | 609/24420 [15:05<10:11:51,  1.54s/it]
  2%|▏         | 610/24420 [15:06<10:01:20,  1.52s/it]
                                                      
{'loss': 0.3198, 'learning_rate': 5e-05, 'epoch': 0.12}

  2%|▏         | 610/24420 [15:06<10:01:20,  1.52s/it]
  3%|▎         | 611/24420 [15:08<9:56:17,  1.50s/it] 
  3%|▎         | 612/24420 [15:09<9:53:39,  1.50s/it]
  3%|▎         | 613/24420 [15:10<9:50:35,  1.49s/it]
  3%|▎         | 614/24420 [15:12<9:48:14,  1.48s/it]
  3%|▎         | 615/24420 [15:13<9:47:18,  1.48s/it]
                                                     
{'loss': 0.3439, 'learning_rate': 5e-05, 'epoch': 0.13}

  3%|▎         | 615/24420 [15:13<9:47:18,  1.48s/it]
  3%|▎         | 616/24420 [15:15<9:46:49,  1.48s/it]
  3%|▎         | 617/24420 [15:16<9:45:46,  1.48s/it]
  3%|▎         | 618/24420 [15:18<9:45:21,  1.48s/it]
  3%|▎         | 619/24420 [15:19<9:45:14,  1.48s/it]
  3%|▎         | 620/24420 [15:21<9:44:42,  1.47s/it]
                                                     
{'loss': 0.3304, 'learning_rate': 5e-05, 'epoch': 0.13}

  3%|▎         | 620/24420 [15:21<9:44:42,  1.47s/it]
  3%|▎         | 621/24420 [15:22<9:44:03,  1.47s/it]
  3%|▎         | 622/24420 [15:24<9:43:11,  1.47s/it]
  3%|▎         | 623/24420 [15:25<9:43:29,  1.47s/it]
  3%|▎         | 624/24420 [15:27<9:45:05,  1.48s/it]
  3%|▎         | 625/24420 [15:28<9:45:01,  1.48s/it]
                                                     
{'loss': 0.3174, 'learning_rate': 5e-05, 'epoch': 0.13}

  3%|▎         | 625/24420 [15:28<9:45:01,  1.48s/it]
  3%|▎         | 626/24420 [15:30<9:44:54,  1.47s/it]
  3%|▎         | 627/24420 [15:31<9:44:07,  1.47s/it]
  3%|▎         | 628/24420 [15:33<9:44:55,  1.48s/it]
  3%|▎         | 629/24420 [15:34<9:44:02,  1.47s/it]
  3%|▎         | 630/24420 [15:36<9:44:26,  1.47s/it]
                                                     
{'loss': 0.3669, 'learning_rate': 5e-05, 'epoch': 0.13}

  3%|▎         | 630/24420 [15:36<9:44:26,  1.47s/it]
  3%|▎         | 631/24420 [15:37<9:44:39,  1.47s/it]
  3%|▎         | 632/24420 [15:38<9:44:08,  1.47s/it]
  3%|▎         | 633/24420 [15:40<9:45:15,  1.48s/it]
  3%|▎         | 634/24420 [15:41<9:45:57,  1.48s/it]
  3%|▎         | 635/24420 [15:43<9:46:47,  1.48s/it]
                                                     
{'loss': 0.3423, 'learning_rate': 5e-05, 'epoch': 0.13}

  3%|▎         | 635/24420 [15:43<9:46:47,  1.48s/it]
  3%|▎         | 636/24420 [15:44<9:46:54,  1.48s/it]
  3%|▎         | 637/24420 [15:46<9:46:18,  1.48s/it]
  3%|▎         | 638/24420 [15:47<9:46:44,  1.48s/it]
  3%|▎         | 639/24420 [15:49<9:46:27,  1.48s/it]
  3%|▎         | 640/24420 [15:50<9:45:28,  1.48s/it]
                                                     
{'loss': 0.3082, 'learning_rate': 5e-05, 'epoch': 0.13}

  3%|▎         | 640/24420 [15:50<9:45:28,  1.48s/it]
  3%|▎         | 641/24420 [15:52<9:47:28,  1.48s/it]
  3%|▎         | 642/24420 [15:53<9:45:50,  1.48s/it]
  3%|▎         | 643/24420 [15:55<9:45:30,  1.48s/it]
  3%|▎         | 644/24420 [15:56<9:44:17,  1.47s/it]
  3%|▎         | 645/24420 [15:58<9:43:33,  1.47s/it]
                                                     
{'loss': 0.3171, 'learning_rate': 5e-05, 'epoch': 0.13}

  3%|▎         | 645/24420 [15:58<9:43:33,  1.47s/it]
  3%|▎         | 646/24420 [15:59<9:43:12,  1.47s/it]
  3%|▎         | 647/24420 [16:01<9:42:49,  1.47s/it]
  3%|▎         | 648/24420 [16:02<9:44:26,  1.48s/it]
  3%|▎         | 649/24420 [16:04<9:44:23,  1.48s/it]
  3%|▎         | 650/24420 [16:05<9:43:04,  1.47s/it]
                                                     
{'loss': 0.3514, 'learning_rate': 5e-05, 'epoch': 0.13}

  3%|▎         | 650/24420 [16:05<9:43:04,  1.47s/it]
  3%|▎         | 651/24420 [16:07<9:43:00,  1.47s/it]
  3%|▎         | 652/24420 [16:08<9:42:59,  1.47s/it]
  3%|▎         | 653/24420 [16:09<9:42:51,  1.47s/it]
  3%|▎         | 654/24420 [16:11<9:42:24,  1.47s/it]
  3%|▎         | 655/24420 [16:12<9:42:02,  1.47s/it]
                                                     
{'loss': 0.325, 'learning_rate': 5e-05, 'epoch': 0.13}

  3%|▎         | 655/24420 [16:12<9:42:02,  1.47s/it]
  3%|▎         | 656/24420 [16:14<9:43:09,  1.47s/it]
  3%|▎         | 657/24420 [16:15<9:42:47,  1.47s/it]
  3%|▎         | 658/24420 [16:17<9:42:22,  1.47s/it]
  3%|▎         | 659/24420 [16:18<9:42:39,  1.47s/it]
  3%|▎         | 660/24420 [16:20<9:43:23,  1.47s/it]
                                                     
{'loss': 0.3285, 'learning_rate': 5e-05, 'epoch': 0.14}

  3%|▎         | 660/24420 [16:20<9:43:23,  1.47s/it]
  3%|▎         | 661/24420 [16:21<9:44:00,  1.47s/it]
  3%|▎         | 662/24420 [16:23<9:44:24,  1.48s/it]
  3%|▎         | 663/24420 [16:24<9:45:07,  1.48s/it]
  3%|▎         | 664/24420 [16:26<9:44:43,  1.48s/it]
  3%|▎         | 665/24420 [16:27<9:43:40,  1.47s/it]
                                                     
{'loss': 0.3328, 'learning_rate': 5e-05, 'epoch': 0.14}

  3%|▎         | 665/24420 [16:27<9:43:40,  1.47s/it]
  3%|▎         | 666/24420 [16:29<9:46:00,  1.48s/it]
  3%|▎         | 667/24420 [16:30<9:44:24,  1.48s/it]
  3%|▎         | 668/24420 [16:32<9:43:28,  1.47s/it]
  3%|▎         | 669/24420 [16:33<9:43:22,  1.47s/it]
  3%|▎         | 670/24420 [16:35<9:43:33,  1.47s/it]
                                                     
{'loss': 0.3433, 'learning_rate': 5e-05, 'epoch': 0.14}

  3%|▎         | 670/24420 [16:35<9:43:33,  1.47s/it]
  3%|▎         | 671/24420 [16:36<9:43:53,  1.48s/it]
  3%|▎         | 672/24420 [16:37<9:44:40,  1.48s/it]
  3%|▎         | 673/24420 [16:39<9:44:51,  1.48s/it]
  3%|▎         | 674/24420 [16:40<9:44:25,  1.48s/it]
  3%|▎         | 675/24420 [16:42<9:42:57,  1.47s/it]
                                                     
{'loss': 0.3275, 'learning_rate': 5e-05, 'epoch': 0.14}

  3%|▎         | 675/24420 [16:42<9:42:57,  1.47s/it]
  3%|▎         | 676/24420 [16:43<9:42:55,  1.47s/it]
  3%|▎         | 677/24420 [16:45<9:42:30,  1.47s/it]
  3%|▎         | 678/24420 [16:46<9:42:01,  1.47s/it]
  3%|▎         | 679/24420 [16:48<9:42:58,  1.47s/it]
  3%|▎         | 680/24420 [16:49<9:44:36,  1.48s/it]
                                                     
{'loss': 0.3258, 'learning_rate': 5e-05, 'epoch': 0.14}

  3%|▎         | 680/24420 [16:49<9:44:36,  1.48s/it]
  3%|▎         | 681/24420 [16:51<9:44:20,  1.48s/it]
  3%|▎         | 682/24420 [16:52<9:45:17,  1.48s/it]
  3%|▎         | 683/24420 [16:54<9:45:39,  1.48s/it]
  3%|▎         | 684/24420 [16:55<9:45:11,  1.48s/it]
  3%|▎         | 685/24420 [16:57<10:49:48,  1.64s/it]
                                                      
{'loss': 0.3089, 'learning_rate': 5e-05, 'epoch': 0.14}

  3%|▎         | 685/24420 [16:57<10:49:48,  1.64s/it]
  3%|▎         | 686/24420 [16:59<10:30:04,  1.59s/it]
  3%|▎         | 687/24420 [17:00<10:16:43,  1.56s/it]
  3%|▎         | 688/24420 [17:02<10:06:53,  1.53s/it]
  3%|▎         | 689/24420 [17:03<9:59:18,  1.52s/it] 
  3%|▎         | 690/24420 [17:05<9:56:18,  1.51s/it]
                                                     
{'loss': 0.2906, 'learning_rate': 5e-05, 'epoch': 0.14}

  3%|▎         | 690/24420 [17:05<9:56:18,  1.51s/it]
  3%|▎         | 691/24420 [17:06<9:52:11,  1.50s/it]
  3%|▎         | 692/24420 [17:08<9:49:33,  1.49s/it]
  3%|▎         | 693/24420 [17:09<9:47:10,  1.48s/it]
  3%|▎         | 694/24420 [17:11<9:45:07,  1.48s/it]
  3%|▎         | 695/24420 [17:12<9:45:00,  1.48s/it]
                                                     
{'loss': 0.3108, 'learning_rate': 5e-05, 'epoch': 0.14}

  3%|▎         | 695/24420 [17:12<9:45:00,  1.48s/it]
  3%|▎         | 696/24420 [17:13<9:44:47,  1.48s/it]
  3%|▎         | 697/24420 [17:15<9:44:14,  1.48s/it]
  3%|▎         | 698/24420 [17:16<9:43:04,  1.47s/it]
  3%|▎         | 699/24420 [17:18<9:42:18,  1.47s/it]
  3%|▎         | 700/24420 [17:19<9:42:21,  1.47s/it]
                                                     
{'loss': 0.3375, 'learning_rate': 5e-05, 'epoch': 0.14}

  3%|▎         | 700/24420 [17:19<9:42:21,  1.47s/it]
  3%|▎         | 701/24420 [17:21<9:44:14,  1.48s/it]
  3%|▎         | 702/24420 [17:22<9:44:16,  1.48s/it]
  3%|▎         | 703/24420 [17:24<9:44:09,  1.48s/it]
  3%|▎         | 704/24420 [17:25<9:44:22,  1.48s/it]
  3%|▎         | 705/24420 [17:27<9:43:28,  1.48s/it]
                                                     
{'loss': 0.2871, 'learning_rate': 5e-05, 'epoch': 0.14}

  3%|▎         | 705/24420 [17:27<9:43:28,  1.48s/it]
  3%|▎         | 706/24420 [17:28<9:43:30,  1.48s/it]
  3%|▎         | 707/24420 [17:30<9:42:08,  1.47s/it]
  3%|▎         | 708/24420 [17:31<9:42:04,  1.47s/it]
  3%|▎         | 709/24420 [17:33<9:42:01,  1.47s/it]
  3%|▎         | 710/24420 [17:34<9:41:52,  1.47s/it]
                                                     
{'loss': 0.33, 'learning_rate': 5e-05, 'epoch': 0.15}

  3%|▎         | 710/24420 [17:34<9:41:52,  1.47s/it]
  3%|▎         | 711/24420 [17:36<9:40:40,  1.47s/it]
  3%|▎         | 712/24420 [17:37<9:40:40,  1.47s/it]
  3%|▎         | 713/24420 [17:39<9:40:24,  1.47s/it]
  3%|▎         | 714/24420 [17:40<9:40:51,  1.47s/it]
  3%|▎         | 715/24420 [17:41<9:41:27,  1.47s/it]
                                                     
{'loss': 0.3167, 'learning_rate': 5e-05, 'epoch': 0.15}

  3%|▎         | 715/24420 [17:41<9:41:27,  1.47s/it]
  3%|▎         | 716/24420 [17:43<9:41:02,  1.47s/it]
  3%|▎         | 717/24420 [17:44<9:41:37,  1.47s/it]
  3%|▎         | 718/24420 [17:46<9:41:48,  1.47s/it]
  3%|▎         | 719/24420 [17:47<9:40:43,  1.47s/it]
  3%|▎         | 720/24420 [17:49<9:40:56,  1.47s/it]
                                                     
{'loss': 0.3093, 'learning_rate': 5e-05, 'epoch': 0.15}

  3%|▎         | 720/24420 [17:49<9:40:56,  1.47s/it]
  3%|▎         | 721/24420 [17:50<9:41:47,  1.47s/it]
  3%|▎         | 722/24420 [17:52<9:41:09,  1.47s/it]
  3%|▎         | 723/24420 [17:53<9:41:02,  1.47s/it]
  3%|▎         | 724/24420 [17:55<9:42:29,  1.47s/it]
  3%|▎         | 725/24420 [17:56<9:42:29,  1.47s/it]
                                                     
{'loss': 0.2909, 'learning_rate': 5e-05, 'epoch': 0.15}

  3%|▎         | 725/24420 [17:56<9:42:29,  1.47s/it]
  3%|▎         | 726/24420 [17:58<9:43:27,  1.48s/it]
  3%|▎         | 727/24420 [17:59<9:42:40,  1.48s/it]
  3%|▎         | 728/24420 [18:01<9:42:23,  1.47s/it]
  3%|▎         | 729/24420 [18:02<9:41:11,  1.47s/it]
  3%|▎         | 730/24420 [18:04<9:42:08,  1.47s/it]
                                                     
{'loss': 0.322, 'learning_rate': 5e-05, 'epoch': 0.15}

  3%|▎         | 730/24420 [18:04<9:42:08,  1.47s/it]
  3%|▎         | 731/24420 [18:05<9:42:13,  1.47s/it]
  3%|▎         | 732/24420 [18:07<9:42:18,  1.47s/it]
  3%|▎         | 733/24420 [18:08<9:42:08,  1.47s/it]
  3%|▎         | 734/24420 [18:09<9:41:40,  1.47s/it]
  3%|▎         | 735/24420 [18:11<9:42:55,  1.48s/it]
                                                     
{'loss': 0.3315, 'learning_rate': 5e-05, 'epoch': 0.15}

  3%|▎         | 735/24420 [18:11<9:42:55,  1.48s/it]
  3%|▎         | 736/24420 [18:12<9:42:20,  1.48s/it]
  3%|▎         | 737/24420 [18:14<9:42:24,  1.48s/it]
  3%|▎         | 738/24420 [18:15<9:42:29,  1.48s/it]
  3%|▎         | 739/24420 [18:17<9:43:37,  1.48s/it]
  3%|▎         | 740/24420 [18:18<9:43:25,  1.48s/it]
                                                     
{'loss': 0.312, 'learning_rate': 5e-05, 'epoch': 0.15}

  3%|▎         | 740/24420 [18:18<9:43:25,  1.48s/it]
  3%|▎         | 741/24420 [18:20<9:44:39,  1.48s/it]
  3%|▎         | 742/24420 [18:21<9:43:27,  1.48s/it]
  3%|▎         | 743/24420 [18:23<9:42:26,  1.48s/it]
  3%|▎         | 744/24420 [18:24<9:42:21,  1.48s/it]
  3%|▎         | 745/24420 [18:26<9:42:19,  1.48s/it]
                                                     
{'loss': 0.3214, 'learning_rate': 5e-05, 'epoch': 0.15}

  3%|▎         | 745/24420 [18:26<9:42:19,  1.48s/it]
  3%|▎         | 746/24420 [18:27<9:41:59,  1.47s/it]
  3%|▎         | 747/24420 [18:29<9:41:03,  1.47s/it]
  3%|▎         | 748/24420 [18:30<9:41:01,  1.47s/it]
  3%|▎         | 749/24420 [18:32<9:40:37,  1.47s/it]
  3%|▎         | 750/24420 [18:33<9:40:05,  1.47s/it]
                                                     
{'loss': 0.3292, 'learning_rate': 5e-05, 'epoch': 0.15}

  3%|▎         | 750/24420 [18:33<9:40:05,  1.47s/it]
  3%|▎         | 751/24420 [18:35<9:40:08,  1.47s/it]
  3%|▎         | 752/24420 [18:36<9:40:01,  1.47s/it]
  3%|▎         | 753/24420 [18:37<9:41:15,  1.47s/it]
  3%|▎         | 754/24420 [18:39<9:41:20,  1.47s/it]
  3%|▎         | 755/24420 [18:40<9:41:56,  1.48s/it]
                                                     
{'loss': 0.3413, 'learning_rate': 5e-05, 'epoch': 0.15}

  3%|▎         | 755/24420 [18:40<9:41:56,  1.48s/it]
  3%|▎         | 756/24420 [18:42<9:42:11,  1.48s/it]
  3%|▎         | 757/24420 [18:43<9:42:17,  1.48s/it]
  3%|▎         | 758/24420 [18:45<9:42:07,  1.48s/it]
  3%|▎         | 759/24420 [18:46<9:42:13,  1.48s/it]
  3%|▎         | 760/24420 [18:48<9:42:05,  1.48s/it]
                                                     
{'loss': 0.3192, 'learning_rate': 5e-05, 'epoch': 0.16}

  3%|▎         | 760/24420 [18:48<9:42:05,  1.48s/it]
  3%|▎         | 761/24420 [18:49<9:43:23,  1.48s/it]
  3%|▎         | 762/24420 [18:51<9:43:32,  1.48s/it]
  3%|▎         | 763/24420 [18:52<9:41:47,  1.48s/it]
  3%|▎         | 764/24420 [18:54<9:42:14,  1.48s/it]
  3%|▎         | 765/24420 [18:55<9:42:31,  1.48s/it]
                                                     
{'loss': 0.394, 'learning_rate': 5e-05, 'epoch': 0.16}

  3%|▎         | 765/24420 [18:55<9:42:31,  1.48s/it]
  3%|▎         | 766/24420 [18:57<9:41:49,  1.48s/it]
  3%|▎         | 767/24420 [18:58<9:40:59,  1.47s/it]
  3%|▎         | 768/24420 [19:00<9:41:01,  1.47s/it]
  3%|▎         | 769/24420 [19:01<9:41:03,  1.47s/it]
  3%|▎         | 770/24420 [19:03<9:40:55,  1.47s/it]
                                                     
{'loss': 0.3246, 'learning_rate': 5e-05, 'epoch': 0.16}

  3%|▎         | 770/24420 [19:03<9:40:55,  1.47s/it]
  3%|▎         | 771/24420 [19:04<9:41:15,  1.47s/it]
  3%|▎         | 772/24420 [19:06<9:41:06,  1.47s/it]
  3%|▎         | 773/24420 [19:07<9:41:44,  1.48s/it]
  3%|▎         | 774/24420 [19:08<9:42:46,  1.48s/it]
  3%|▎         | 775/24420 [19:10<9:42:25,  1.48s/it]
                                                     
{'loss': 0.3426, 'learning_rate': 5e-05, 'epoch': 0.16}

  3%|▎         | 775/24420 [19:10<9:42:25,  1.48s/it]
  3%|▎         | 776/24420 [19:11<9:42:00,  1.48s/it]
  3%|▎         | 777/24420 [19:13<9:41:32,  1.48s/it]
  3%|▎         | 778/24420 [19:14<9:42:01,  1.48s/it]
  3%|▎         | 779/24420 [19:16<9:42:30,  1.48s/it]
  3%|▎         | 780/24420 [19:17<9:42:38,  1.48s/it]
                                                     
{'loss': 0.3202, 'learning_rate': 5e-05, 'epoch': 0.16}

  3%|▎         | 780/24420 [19:17<9:42:38,  1.48s/it]
  3%|▎         | 781/24420 [19:19<9:42:16,  1.48s/it]
  3%|▎         | 782/24420 [19:20<9:41:47,  1.48s/it]
  3%|▎         | 783/24420 [19:22<9:41:01,  1.47s/it]
  3%|▎         | 784/24420 [19:23<9:40:35,  1.47s/it]
  3%|▎         | 785/24420 [19:25<9:40:53,  1.47s/it]
                                                     
{'loss': 0.3052, 'learning_rate': 5e-05, 'epoch': 0.16}

  3%|▎         | 785/24420 [19:25<9:40:53,  1.47s/it]
  3%|▎         | 786/24420 [19:26<9:43:09,  1.48s/it]
  3%|▎         | 787/24420 [19:28<9:42:56,  1.48s/it]
  3%|▎         | 788/24420 [19:29<9:42:17,  1.48s/it]
  3%|▎         | 789/24420 [19:31<9:42:10,  1.48s/it]
  3%|▎         | 790/24420 [19:32<9:43:01,  1.48s/it]
                                                     
{'loss': 0.2478, 'learning_rate': 5e-05, 'epoch': 0.16}

  3%|▎         | 790/24420 [19:32<9:43:01,  1.48s/it]
  3%|▎         | 791/24420 [19:34<9:43:10,  1.48s/it]
  3%|▎         | 792/24420 [19:35<9:43:33,  1.48s/it]
  3%|▎         | 793/24420 [19:37<9:44:01,  1.48s/it]
  3%|▎         | 794/24420 [19:38<9:43:44,  1.48s/it]
  3%|▎         | 795/24420 [19:40<9:41:51,  1.48s/it]
                                                     
{'loss': 0.3349, 'learning_rate': 5e-05, 'epoch': 0.16}

  3%|▎         | 795/24420 [19:40<9:41:51,  1.48s/it]
  3%|▎         | 796/24420 [19:41<9:41:22,  1.48s/it]
  3%|▎         | 797/24420 [19:42<9:41:48,  1.48s/it]
  3%|▎         | 798/24420 [19:44<9:40:42,  1.47s/it]
  3%|▎         | 799/24420 [19:45<9:41:00,  1.48s/it]
  3%|▎         | 800/24420 [19:47<9:40:08,  1.47s/it]
                                                     
{'loss': 0.3273, 'learning_rate': 5e-05, 'epoch': 0.16}

  3%|▎         | 800/24420 [19:47<9:40:08,  1.47s/it]
  3%|▎         | 801/24420 [19:51<15:47:22,  2.41s/it]
  3%|▎         | 802/24420 [19:53<13:56:35,  2.13s/it]
  3%|▎         | 803/24420 [19:54<12:39:51,  1.93s/it]
  3%|▎         | 804/24420 [19:56<11:45:58,  1.79s/it]
  3%|▎         | 805/24420 [19:57<11:08:46,  1.70s/it]
                                                      
{'loss': 0.3127, 'learning_rate': 5e-05, 'epoch': 0.16}

  3%|▎         | 805/24420 [19:57<11:08:46,  1.70s/it]
  3%|▎         | 806/24420 [19:59<10:42:53,  1.63s/it]
  3%|▎         | 807/24420 [20:00<10:23:23,  1.58s/it]
  3%|▎         | 808/24420 [20:02<10:10:04,  1.55s/it]
  3%|▎         | 809/24420 [20:03<9:59:59,  1.52s/it] 
  3%|▎         | 810/24420 [20:05<9:52:41,  1.51s/it]
                                                     
{'loss': 0.3093, 'learning_rate': 5e-05, 'epoch': 0.17}

  3%|▎         | 810/24420 [20:05<9:52:41,  1.51s/it]
  3%|▎         | 811/24420 [20:06<9:48:18,  1.50s/it]
  3%|▎         | 812/24420 [20:08<9:46:02,  1.49s/it]
  3%|▎         | 813/24420 [20:09<9:43:01,  1.48s/it]
  3%|▎         | 814/24420 [20:11<9:42:08,  1.48s/it]
  3%|▎         | 815/24420 [20:12<9:41:17,  1.48s/it]
                                                     
{'loss': 0.2677, 'learning_rate': 5e-05, 'epoch': 0.17}

  3%|▎         | 815/24420 [20:12<9:41:17,  1.48s/it]
  3%|▎         | 816/24420 [20:14<9:40:08,  1.47s/it]
  3%|▎         | 817/24420 [20:15<9:39:11,  1.47s/it]
  3%|▎         | 818/24420 [20:16<9:38:50,  1.47s/it]
  3%|▎         | 819/24420 [20:18<9:39:23,  1.47s/it]
  3%|▎         | 820/24420 [20:19<9:39:23,  1.47s/it]
                                                     
{'loss': 0.3068, 'learning_rate': 5e-05, 'epoch': 0.17}

  3%|▎         | 820/24420 [20:19<9:39:23,  1.47s/it]
  3%|▎         | 821/24420 [20:21<9:39:56,  1.47s/it]
  3%|▎         | 822/24420 [20:22<9:39:46,  1.47s/it]
  3%|▎         | 823/24420 [20:24<9:40:04,  1.47s/it]
  3%|▎         | 824/24420 [20:25<9:39:41,  1.47s/it]
  3%|▎         | 825/24420 [20:27<9:39:25,  1.47s/it]
                                                     
{'loss': 0.345, 'learning_rate': 5e-05, 'epoch': 0.17}

  3%|▎         | 825/24420 [20:27<9:39:25,  1.47s/it]
  3%|▎         | 826/24420 [20:28<9:39:02,  1.47s/it]
  3%|▎         | 827/24420 [20:30<9:38:29,  1.47s/it]
  3%|▎         | 828/24420 [20:31<9:37:22,  1.47s/it]
  3%|▎         | 829/24420 [20:33<9:37:45,  1.47s/it]
  3%|▎         | 830/24420 [20:34<9:36:36,  1.47s/it]
                                                     
{'loss': 0.2935, 'learning_rate': 5e-05, 'epoch': 0.17}

  3%|▎         | 830/24420 [20:34<9:36:36,  1.47s/it]
  3%|▎         | 831/24420 [20:36<9:38:09,  1.47s/it]
  3%|▎         | 832/24420 [20:37<9:37:30,  1.47s/it]
  3%|▎         | 833/24420 [20:39<9:38:12,  1.47s/it]
  3%|▎         | 834/24420 [20:40<9:37:58,  1.47s/it]
  3%|▎         | 835/24420 [20:42<9:38:20,  1.47s/it]
                                                     
{'loss': 0.3186, 'learning_rate': 5e-05, 'epoch': 0.17}

  3%|▎         | 835/24420 [20:42<9:38:20,  1.47s/it]
  3%|▎         | 836/24420 [20:43<9:38:03,  1.47s/it]
  3%|▎         | 837/24420 [20:44<9:38:28,  1.47s/it]
  3%|▎         | 838/24420 [20:46<9:37:23,  1.47s/it]
  3%|▎         | 839/24420 [20:47<9:38:04,  1.47s/it]
  3%|▎         | 840/24420 [20:49<9:38:32,  1.47s/it]
                                                     
{'loss': 0.2834, 'learning_rate': 5e-05, 'epoch': 0.17}

  3%|▎         | 840/24420 [20:49<9:38:32,  1.47s/it]
  3%|▎         | 841/24420 [20:50<9:38:31,  1.47s/it]
  3%|▎         | 842/24420 [20:52<9:38:49,  1.47s/it]
  3%|▎         | 843/24420 [20:53<9:38:26,  1.47s/it]
  3%|▎         | 844/24420 [20:55<9:37:12,  1.47s/it]
  3%|▎         | 845/24420 [20:56<9:36:59,  1.47s/it]
                                                     
{'loss': 0.3061, 'learning_rate': 5e-05, 'epoch': 0.17}

  3%|▎         | 845/24420 [20:56<9:36:59,  1.47s/it]
  3%|▎         | 846/24420 [20:58<9:36:56,  1.47s/it]
  3%|▎         | 847/24420 [20:59<9:36:42,  1.47s/it]
  3%|▎         | 848/24420 [21:01<9:37:12,  1.47s/it]
  3%|▎         | 849/24420 [21:02<9:37:02,  1.47s/it]
  3%|▎         | 850/24420 [21:04<9:38:00,  1.47s/it]
                                                     
{'loss': 0.2804, 'learning_rate': 5e-05, 'epoch': 0.17}

  3%|▎         | 850/24420 [21:04<9:38:00,  1.47s/it]
  3%|▎         | 851/24420 [21:05<9:38:48,  1.47s/it]
  3%|▎         | 852/24420 [21:07<9:37:55,  1.47s/it]
  3%|▎         | 853/24420 [21:08<9:37:23,  1.47s/it]
  3%|▎         | 854/24420 [21:09<9:37:18,  1.47s/it]
  4%|▎         | 855/24420 [21:11<9:37:16,  1.47s/it]
                                                     
{'loss': 0.2772, 'learning_rate': 5e-05, 'epoch': 0.18}

  4%|▎         | 855/24420 [21:11<9:37:16,  1.47s/it]
  4%|▎         | 856/24420 [21:12<9:36:19,  1.47s/it]
  4%|▎         | 857/24420 [21:14<9:36:52,  1.47s/it]
  4%|▎         | 858/24420 [21:15<9:37:21,  1.47s/it]
  4%|▎         | 859/24420 [21:17<9:37:43,  1.47s/it]
  4%|▎         | 860/24420 [21:18<9:38:47,  1.47s/it]
                                                     
{'loss': 0.2997, 'learning_rate': 5e-05, 'epoch': 0.18}

  4%|▎         | 860/24420 [21:18<9:38:47,  1.47s/it]
  4%|▎         | 861/24420 [21:20<9:39:35,  1.48s/it]
  4%|▎         | 862/24420 [21:21<9:40:24,  1.48s/it]
  4%|▎         | 863/24420 [21:23<9:40:22,  1.48s/it]
  4%|▎         | 864/24420 [21:24<9:40:37,  1.48s/it]
  4%|▎         | 865/24420 [21:26<9:41:47,  1.48s/it]
                                                     
{'loss': 0.2813, 'learning_rate': 5e-05, 'epoch': 0.18}

  4%|▎         | 865/24420 [21:26<9:41:47,  1.48s/it]
  4%|▎         | 866/24420 [21:27<9:40:40,  1.48s/it]
  4%|▎         | 867/24420 [21:29<9:40:19,  1.48s/it]
  4%|▎         | 868/24420 [21:30<9:38:51,  1.47s/it]
  4%|▎         | 869/24420 [21:32<9:38:27,  1.47s/it]
  4%|▎         | 870/24420 [21:33<9:37:42,  1.47s/it]
                                                     
{'loss': 0.3031, 'learning_rate': 5e-05, 'epoch': 0.18}

  4%|▎         | 870/24420 [21:33<9:37:42,  1.47s/it]
  4%|▎         | 871/24420 [21:35<9:37:28,  1.47s/it]
  4%|▎         | 872/24420 [21:36<9:36:42,  1.47s/it]
  4%|▎         | 873/24420 [21:37<9:37:31,  1.47s/it]
  4%|▎         | 874/24420 [21:39<9:38:07,  1.47s/it]
  4%|▎         | 875/24420 [21:40<9:37:34,  1.47s/it]
                                                     
{'loss': 0.3064, 'learning_rate': 5e-05, 'epoch': 0.18}

  4%|▎         | 875/24420 [21:40<9:37:34,  1.47s/it]
  4%|▎         | 876/24420 [21:42<9:37:05,  1.47s/it]
  4%|▎         | 877/24420 [21:43<9:37:38,  1.47s/it]
  4%|▎         | 878/24420 [21:45<9:38:00,  1.47s/it]
  4%|▎         | 879/24420 [21:46<9:38:46,  1.48s/it]
  4%|▎         | 880/24420 [21:48<9:40:00,  1.48s/it]
                                                     
{'loss': 0.2915, 'learning_rate': 5e-05, 'epoch': 0.18}

  4%|▎         | 880/24420 [21:48<9:40:00,  1.48s/it]
  4%|▎         | 881/24420 [21:49<9:40:23,  1.48s/it]
  4%|▎         | 882/24420 [21:51<9:40:11,  1.48s/it]
  4%|▎         | 883/24420 [21:52<9:39:46,  1.48s/it]
  4%|▎         | 884/24420 [21:54<9:39:14,  1.48s/it]
  4%|▎         | 885/24420 [21:55<9:38:06,  1.47s/it]
                                                     
{'loss': 0.3091, 'learning_rate': 5e-05, 'epoch': 0.18}

  4%|▎         | 885/24420 [21:55<9:38:06,  1.47s/it]
  4%|▎         | 886/24420 [21:57<9:38:18,  1.47s/it]
  4%|▎         | 887/24420 [21:58<9:37:15,  1.47s/it]
  4%|▎         | 888/24420 [22:00<9:37:23,  1.47s/it]
  4%|▎         | 889/24420 [22:01<9:36:50,  1.47s/it]
  4%|▎         | 890/24420 [22:03<9:36:42,  1.47s/it]
                                                     
{'loss': 0.3146, 'learning_rate': 5e-05, 'epoch': 0.18}

  4%|▎         | 890/24420 [22:03<9:36:42,  1.47s/it]
  4%|▎         | 891/24420 [22:04<9:36:50,  1.47s/it]
  4%|▎         | 892/24420 [22:05<9:36:14,  1.47s/it]
  4%|▎         | 893/24420 [22:07<9:36:18,  1.47s/it]
  4%|▎         | 894/24420 [22:08<9:35:28,  1.47s/it]
  4%|▎         | 895/24420 [22:10<9:35:59,  1.47s/it]
                                                     
{'loss': 0.2902, 'learning_rate': 5e-05, 'epoch': 0.18}

  4%|▎         | 895/24420 [22:10<9:35:59,  1.47s/it]
  4%|▎         | 896/24420 [22:11<9:37:30,  1.47s/it]
  4%|▎         | 897/24420 [22:13<9:36:58,  1.47s/it]
  4%|▎         | 898/24420 [22:14<9:36:31,  1.47s/it]
  4%|▎         | 899/24420 [22:16<9:37:02,  1.47s/it]
  4%|▎         | 900/24420 [22:17<9:36:26,  1.47s/it]
                                                     
{'loss': 0.2727, 'learning_rate': 5e-05, 'epoch': 0.18}

  4%|▎         | 900/24420 [22:17<9:36:26,  1.47s/it]
  4%|▎         | 901/24420 [22:19<9:36:50,  1.47s/it]
  4%|▎         | 902/24420 [22:20<9:36:46,  1.47s/it]
  4%|▎         | 903/24420 [22:22<9:37:10,  1.47s/it]
  4%|▎         | 904/24420 [22:23<9:37:14,  1.47s/it]
  4%|▎         | 905/24420 [22:25<9:36:03,  1.47s/it]
                                                     
{'loss': 0.2739, 'learning_rate': 5e-05, 'epoch': 0.19}

  4%|▎         | 905/24420 [22:25<9:36:03,  1.47s/it]
  4%|▎         | 906/24420 [22:26<9:35:33,  1.47s/it]
  4%|▎         | 907/24420 [22:27<9:34:57,  1.47s/it]
  4%|▎         | 908/24420 [22:29<9:34:49,  1.47s/it]
  4%|▎         | 909/24420 [22:30<9:35:06,  1.47s/it]
  4%|▎         | 910/24420 [22:32<9:36:01,  1.47s/it]
                                                     
{'loss': 0.3088, 'learning_rate': 5e-05, 'epoch': 0.19}

  4%|▎         | 910/24420 [22:32<9:36:01,  1.47s/it]
  4%|▎         | 911/24420 [22:33<9:37:17,  1.47s/it]
  4%|▎         | 912/24420 [22:35<9:36:55,  1.47s/it]
  4%|▎         | 913/24420 [22:36<9:35:51,  1.47s/it]
  4%|▎         | 914/24420 [22:38<9:35:19,  1.47s/it]
  4%|▎         | 915/24420 [22:39<9:35:12,  1.47s/it]
                                                     
{'loss': 0.3228, 'learning_rate': 5e-05, 'epoch': 0.19}

  4%|▎         | 915/24420 [22:39<9:35:12,  1.47s/it]
  4%|▍         | 916/24420 [22:41<9:35:59,  1.47s/it]
  4%|▍         | 917/24420 [22:42<9:36:49,  1.47s/it]
  4%|▍         | 918/24420 [22:44<9:36:41,  1.47s/it]
  4%|▍         | 919/24420 [22:45<9:37:02,  1.47s/it]
  4%|▍         | 920/24420 [22:47<9:36:08,  1.47s/it]
                                                     
{'loss': 0.3177, 'learning_rate': 5e-05, 'epoch': 0.19}

  4%|▍         | 920/24420 [22:47<9:36:08,  1.47s/it]
  4%|▍         | 921/24420 [22:48<9:37:07,  1.47s/it]
  4%|▍         | 922/24420 [22:50<9:37:24,  1.47s/it]
  4%|▍         | 923/24420 [22:51<9:37:53,  1.48s/it]
  4%|▍         | 924/24420 [22:53<9:38:21,  1.48s/it]
  4%|▍         | 925/24420 [22:54<9:38:11,  1.48s/it]
                                                     
{'loss': 0.2883, 'learning_rate': 5e-05, 'epoch': 0.19}

  4%|▍         | 925/24420 [22:54<9:38:11,  1.48s/it]
  4%|▍         | 926/24420 [22:55<9:37:09,  1.47s/it]
  4%|▍         | 927/24420 [22:57<9:37:07,  1.47s/it]
  4%|▍         | 928/24420 [22:58<9:37:08,  1.47s/it]
  4%|▍         | 929/24420 [23:00<9:36:10,  1.47s/it]
  4%|▍         | 930/24420 [23:01<9:35:35,  1.47s/it]
                                                     
{'loss': 0.299, 'learning_rate': 5e-05, 'epoch': 0.19}

  4%|▍         | 930/24420 [23:01<9:35:35,  1.47s/it]
  4%|▍         | 931/24420 [23:03<9:38:56,  1.48s/it]
  4%|▍         | 932/24420 [23:04<9:37:28,  1.48s/it]
  4%|▍         | 933/24420 [23:06<9:37:03,  1.47s/it]
  4%|▍         | 934/24420 [23:07<9:36:24,  1.47s/it]
  4%|▍         | 935/24420 [23:09<9:36:12,  1.47s/it]
                                                     
{'loss': 0.2757, 'learning_rate': 5e-05, 'epoch': 0.19}

  4%|▍         | 935/24420 [23:09<9:36:12,  1.47s/it]
  4%|▍         | 936/24420 [23:10<9:36:29,  1.47s/it]
  4%|▍         | 937/24420 [23:12<9:36:34,  1.47s/it]
  4%|▍         | 938/24420 [23:13<9:36:08,  1.47s/it]
  4%|▍         | 939/24420 [23:15<9:36:10,  1.47s/it]
  4%|▍         | 940/24420 [23:16<9:36:30,  1.47s/it]
                                                     
{'loss': 0.2734, 'learning_rate': 5e-05, 'epoch': 0.19}

  4%|▍         | 940/24420 [23:16<9:36:30,  1.47s/it]
  4%|▍         | 941/24420 [23:18<9:37:23,  1.48s/it]
  4%|▍         | 942/24420 [23:19<9:36:21,  1.47s/it]
  4%|▍         | 943/24420 [23:21<9:36:28,  1.47s/it]
  4%|▍         | 944/24420 [23:22<9:37:08,  1.48s/it]
  4%|▍         | 945/24420 [23:23<9:36:26,  1.47s/it]
                                                     
{'loss': 0.3084, 'learning_rate': 5e-05, 'epoch': 0.19}

  4%|▍         | 945/24420 [23:23<9:36:26,  1.47s/it]
  4%|▍         | 946/24420 [23:25<9:36:25,  1.47s/it]
  4%|▍         | 947/24420 [23:26<9:36:35,  1.47s/it]
  4%|▍         | 948/24420 [23:28<9:36:40,  1.47s/it]
  4%|▍         | 949/24420 [23:29<9:36:49,  1.47s/it]
  4%|▍         | 950/24420 [23:31<9:36:55,  1.47s/it]
                                                     
{'loss': 0.2849, 'learning_rate': 5e-05, 'epoch': 0.19}

  4%|▍         | 950/24420 [23:31<9:36:55,  1.47s/it]
  4%|▍         | 951/24420 [23:32<9:37:20,  1.48s/it]
  4%|▍         | 952/24420 [23:34<9:36:20,  1.47s/it]
  4%|▍         | 953/24420 [23:35<9:36:02,  1.47s/it]
  4%|▍         | 954/24420 [23:37<9:35:05,  1.47s/it]
  4%|▍         | 955/24420 [23:38<9:34:55,  1.47s/it]
                                                     
{'loss': 0.2798, 'learning_rate': 5e-05, 'epoch': 0.2}

  4%|▍         | 955/24420 [23:38<9:34:55,  1.47s/it]
  4%|▍         | 956/24420 [23:40<9:35:18,  1.47s/it]
  4%|▍         | 957/24420 [23:41<9:35:14,  1.47s/it]
  4%|▍         | 958/24420 [23:43<9:34:52,  1.47s/it]
  4%|▍         | 959/24420 [23:44<9:34:26,  1.47s/it]
  4%|▍         | 960/24420 [23:46<9:33:54,  1.47s/it]
                                                     
{'loss': 0.3177, 'learning_rate': 5e-05, 'epoch': 0.2}

  4%|▍         | 960/24420 [23:46<9:33:54,  1.47s/it]
  4%|▍         | 961/24420 [23:47<9:34:41,  1.47s/it]
  4%|▍         | 962/24420 [23:49<9:36:01,  1.47s/it]
  4%|▍         | 963/24420 [23:50<9:35:29,  1.47s/it]
  4%|▍         | 964/24420 [23:51<9:35:17,  1.47s/it]
  4%|▍         | 965/24420 [23:53<9:36:02,  1.47s/it]
                                                     
{'loss': 0.2927, 'learning_rate': 5e-05, 'epoch': 0.2}

  4%|▍         | 965/24420 [23:53<9:36:02,  1.47s/it]
  4%|▍         | 966/24420 [23:54<9:37:20,  1.48s/it]
  4%|▍         | 967/24420 [23:56<9:48:44,  1.51s/it]
  4%|▍         | 968/24420 [23:57<9:44:23,  1.50s/it]
  4%|▍         | 969/24420 [23:59<9:41:43,  1.49s/it]
  4%|▍         | 970/24420 [24:00<9:39:50,  1.48s/it]
                                                     
{'loss': 0.3056, 'learning_rate': 5e-05, 'epoch': 0.2}

  4%|▍         | 970/24420 [24:00<9:39:50,  1.48s/it]
  4%|▍         | 971/24420 [24:02<9:38:27,  1.48s/it]
  4%|▍         | 972/24420 [24:03<9:37:11,  1.48s/it]
  4%|▍         | 973/24420 [24:05<9:36:46,  1.48s/it]
  4%|▍         | 974/24420 [24:06<9:34:46,  1.47s/it]
  4%|▍         | 975/24420 [24:08<9:34:59,  1.47s/it]
                                                     
{'loss': 0.2921, 'learning_rate': 5e-05, 'epoch': 0.2}

  4%|▍         | 975/24420 [24:08<9:34:59,  1.47s/it]
  4%|▍         | 976/24420 [24:09<9:35:36,  1.47s/it]
  4%|▍         | 977/24420 [24:11<9:34:41,  1.47s/it]
  4%|▍         | 978/24420 [24:12<9:34:20,  1.47s/it]
  4%|▍         | 979/24420 [24:14<9:34:19,  1.47s/it]
  4%|▍         | 980/24420 [24:15<9:34:36,  1.47s/it]
                                                     
{'loss': 0.303, 'learning_rate': 5e-05, 'epoch': 0.2}

  4%|▍         | 980/24420 [24:15<9:34:36,  1.47s/it]
  4%|▍         | 981/24420 [24:17<9:34:01,  1.47s/it]
  4%|▍         | 982/24420 [24:18<9:35:10,  1.47s/it]
  4%|▍         | 983/24420 [24:20<9:35:54,  1.47s/it]
  4%|▍         | 984/24420 [24:21<9:37:01,  1.48s/it]
  4%|▍         | 985/24420 [24:22<9:35:53,  1.47s/it]
                                                     
{'loss': 0.3001, 'learning_rate': 5e-05, 'epoch': 0.2}

  4%|▍         | 985/24420 [24:22<9:35:53,  1.47s/it]
  4%|▍         | 986/24420 [24:24<9:34:55,  1.47s/it]
  4%|▍         | 987/24420 [24:25<9:36:46,  1.48s/it]
  4%|▍         | 988/24420 [24:27<9:37:18,  1.48s/it]
  4%|▍         | 989/24420 [24:28<9:37:11,  1.48s/it]
  4%|▍         | 990/24420 [24:30<9:37:02,  1.48s/it]
                                                     
{'loss': 0.2754, 'learning_rate': 5e-05, 'epoch': 0.2}

  4%|▍         | 990/24420 [24:30<9:37:02,  1.48s/it]
  4%|▍         | 991/24420 [24:31<9:37:29,  1.48s/it]
  4%|▍         | 992/24420 [24:33<9:35:51,  1.47s/it]
  4%|▍         | 993/24420 [24:34<9:36:01,  1.48s/it]
  4%|▍         | 994/24420 [24:36<9:35:48,  1.47s/it]
  4%|▍         | 995/24420 [24:37<9:35:24,  1.47s/it]
                                                     
{'loss': 0.2996, 'learning_rate': 5e-05, 'epoch': 0.2}

  4%|▍         | 995/24420 [24:37<9:35:24,  1.47s/it]
  4%|▍         | 996/24420 [24:39<9:35:24,  1.47s/it]
  4%|▍         | 997/24420 [24:40<9:35:25,  1.47s/it]
  4%|▍         | 998/24420 [24:42<9:34:52,  1.47s/it]
  4%|▍         | 999/24420 [24:43<9:34:40,  1.47s/it]
  4%|▍         | 1000/24420 [24:45<9:35:28,  1.47s/it]
                                                      
{'loss': 0.2837, 'learning_rate': 5e-05, 'epoch': 0.2}

  4%|▍         | 1000/24420 [24:45<9:35:28,  1.47s/it]
  4%|▍         | 1001/24420 [24:49<15:47:57,  2.43s/it]
  4%|▍         | 1002/24420 [24:51<13:55:48,  2.14s/it]
  4%|▍         | 1003/24420 [24:52<12:37:39,  1.94s/it]
  4%|▍         | 1004/24420 [24:54<11:44:32,  1.81s/it]
  4%|▍         | 1005/24420 [24:55<11:05:29,  1.71s/it]
                                                       
{'loss': 0.28, 'learning_rate': 5e-05, 'epoch': 0.21}

  4%|▍         | 1005/24420 [24:55<11:05:29,  1.71s/it]
  4%|▍         | 1006/24420 [24:57<10:38:33,  1.64s/it]
  4%|▍         | 1007/24420 [24:58<10:18:39,  1.59s/it]
  4%|▍         | 1008/24420 [25:00<10:05:35,  1.55s/it]
  4%|▍         | 1009/24420 [25:01<9:55:55,  1.53s/it] 
  4%|▍         | 1010/24420 [25:03<9:48:54,  1.51s/it]
                                                      
{'loss': 0.2569, 'learning_rate': 5e-05, 'epoch': 0.21}

  4%|▍         | 1010/24420 [25:03<9:48:54,  1.51s/it]
  4%|▍         | 1011/24420 [25:04<9:44:52,  1.50s/it]
  4%|▍         | 1012/24420 [25:05<9:40:32,  1.49s/it]
  4%|▍         | 1013/24420 [25:07<9:37:48,  1.48s/it]
  4%|▍         | 1014/24420 [25:08<9:37:02,  1.48s/it]
  4%|▍         | 1015/24420 [25:10<9:34:59,  1.47s/it]
                                                      
{'loss': 0.3024, 'learning_rate': 5e-05, 'epoch': 0.21}

  4%|▍         | 1015/24420 [25:10<9:34:59,  1.47s/it]
  4%|▍         | 1016/24420 [25:11<9:33:59,  1.47s/it]
  4%|▍         | 1017/24420 [25:13<9:33:29,  1.47s/it]
  4%|▍         | 1018/24420 [25:14<9:33:20,  1.47s/it]
  4%|▍         | 1019/24420 [25:16<9:32:41,  1.47s/it]
  4%|▍         | 1020/24420 [25:17<9:32:33,  1.47s/it]
                                                      
{'loss': 0.3372, 'learning_rate': 5e-05, 'epoch': 0.21}

  4%|▍         | 1020/24420 [25:17<9:32:33,  1.47s/it]
  4%|▍         | 1021/24420 [25:19<9:32:38,  1.47s/it]
  4%|▍         | 1022/24420 [25:20<9:32:29,  1.47s/it]
  4%|▍         | 1023/24420 [25:22<9:32:04,  1.47s/it]
  4%|▍         | 1024/24420 [25:23<9:32:26,  1.47s/it]
  4%|▍         | 1025/24420 [25:25<9:32:55,  1.47s/it]
                                                      
{'loss': 0.2974, 'learning_rate': 5e-05, 'epoch': 0.21}

  4%|▍         | 1025/24420 [25:25<9:32:55,  1.47s/it]
  4%|▍         | 1026/24420 [25:26<9:33:31,  1.47s/it]
  4%|▍         | 1027/24420 [25:27<9:34:07,  1.47s/it]
  4%|▍         | 1028/24420 [25:29<9:33:08,  1.47s/it]
  4%|▍         | 1029/24420 [25:30<9:32:27,  1.47s/it]
  4%|▍         | 1030/24420 [25:32<9:32:02,  1.47s/it]
                                                      
{'loss': 0.2759, 'learning_rate': 5e-05, 'epoch': 0.21}

  4%|▍         | 1030/24420 [25:32<9:32:02,  1.47s/it]
  4%|▍         | 1031/24420 [25:33<9:33:04,  1.47s/it]
  4%|▍         | 1032/24420 [25:35<9:33:11,  1.47s/it]
  4%|▍         | 1033/24420 [25:36<9:32:26,  1.47s/it]
  4%|▍         | 1034/24420 [25:38<9:35:06,  1.48s/it]
  4%|▍         | 1035/24420 [25:39<9:34:39,  1.47s/it]
                                                      
{'loss': 0.2839, 'learning_rate': 5e-05, 'epoch': 0.21}

  4%|▍         | 1035/24420 [25:39<9:34:39,  1.47s/it]
  4%|▍         | 1036/24420 [25:41<9:33:35,  1.47s/it]
  4%|▍         | 1037/24420 [25:42<9:32:44,  1.47s/it]
  4%|▍         | 1038/24420 [25:44<9:32:21,  1.47s/it]
  4%|▍         | 1039/24420 [25:45<9:33:18,  1.47s/it]
  4%|▍         | 1040/24420 [25:47<9:32:50,  1.47s/it]
                                                      
{'loss': 0.2764, 'learning_rate': 5e-05, 'epoch': 0.21}

  4%|▍         | 1040/24420 [25:47<9:32:50,  1.47s/it]
  4%|▍         | 1041/24420 [25:48<9:32:34,  1.47s/it]
  4%|▍         | 1042/24420 [25:50<9:32:38,  1.47s/it]
  4%|▍         | 1043/24420 [25:51<9:32:11,  1.47s/it]
  4%|▍         | 1044/24420 [25:52<9:32:30,  1.47s/it]
  4%|▍         | 1045/24420 [25:54<9:32:05,  1.47s/it]
                                                      
{'loss': 0.2955, 'learning_rate': 5e-05, 'epoch': 0.21}

  4%|▍         | 1045/24420 [25:54<9:32:05,  1.47s/it]
  4%|▍         | 1046/24420 [25:55<9:31:33,  1.47s/it]
  4%|▍         | 1047/24420 [25:57<9:31:38,  1.47s/it]
  4%|▍         | 1048/24420 [25:58<9:31:31,  1.47s/it]
  4%|▍         | 1049/24420 [26:00<9:31:09,  1.47s/it]
  4%|▍         | 1050/24420 [26:01<9:31:37,  1.47s/it]
                                                      
{'loss': 0.2835, 'learning_rate': 5e-05, 'epoch': 0.21}

  4%|▍         | 1050/24420 [26:01<9:31:37,  1.47s/it]
  4%|▍         | 1051/24420 [26:03<9:32:05,  1.47s/it]
  4%|▍         | 1052/24420 [26:04<9:31:39,  1.47s/it]
  4%|▍         | 1053/24420 [26:06<9:31:25,  1.47s/it]
  4%|▍         | 1054/24420 [26:07<9:30:46,  1.47s/it]
  4%|▍         | 1055/24420 [26:09<9:31:21,  1.47s/it]
                                                      
{'loss': 0.2527, 'learning_rate': 5e-05, 'epoch': 0.22}

  4%|▍         | 1055/24420 [26:09<9:31:21,  1.47s/it]
  4%|▍         | 1056/24420 [26:10<9:31:20,  1.47s/it]
  4%|▍         | 1057/24420 [26:12<9:32:01,  1.47s/it]
  4%|▍         | 1058/24420 [26:13<9:32:19,  1.47s/it]
  4%|▍         | 1059/24420 [26:14<9:31:22,  1.47s/it]
  4%|▍         | 1060/24420 [26:16<9:31:32,  1.47s/it]
                                                      
{'loss': 0.2705, 'learning_rate': 5e-05, 'epoch': 0.22}

  4%|▍         | 1060/24420 [26:16<9:31:32,  1.47s/it]
  4%|▍         | 1061/24420 [26:17<9:32:39,  1.47s/it]
  4%|▍         | 1062/24420 [26:19<9:32:02,  1.47s/it]
  4%|▍         | 1063/24420 [26:20<9:32:29,  1.47s/it]
  4%|▍         | 1064/24420 [26:22<9:31:28,  1.47s/it]
  4%|▍         | 1065/24420 [26:23<9:31:40,  1.47s/it]
                                                      
{'loss': 0.2824, 'learning_rate': 5e-05, 'epoch': 0.22}

  4%|▍         | 1065/24420 [26:23<9:31:40,  1.47s/it]
  4%|▍         | 1066/24420 [26:25<9:31:25,  1.47s/it]
  4%|▍         | 1067/24420 [26:26<9:30:41,  1.47s/it]
  4%|▍         | 1068/24420 [26:28<9:30:16,  1.47s/it]
  4%|▍         | 1069/24420 [26:29<9:30:29,  1.47s/it]
  4%|▍         | 1070/24420 [26:31<9:30:22,  1.47s/it]
                                                      
{'loss': 0.2612, 'learning_rate': 5e-05, 'epoch': 0.22}

  4%|▍         | 1070/24420 [26:31<9:30:22,  1.47s/it]
  4%|▍         | 1071/24420 [26:32<9:30:55,  1.47s/it]
  4%|▍         | 1072/24420 [26:34<9:30:59,  1.47s/it]
  4%|▍         | 1073/24420 [26:35<9:32:01,  1.47s/it]
  4%|▍         | 1074/24420 [26:37<9:32:33,  1.47s/it]
  4%|▍         | 1075/24420 [26:38<9:32:10,  1.47s/it]
                                                      
{'loss': 0.2579, 'learning_rate': 5e-05, 'epoch': 0.22}

  4%|▍         | 1075/24420 [26:38<9:32:10,  1.47s/it]
  4%|▍         | 1076/24420 [26:39<9:32:03,  1.47s/it]
  4%|▍         | 1077/24420 [26:41<9:31:18,  1.47s/it]
  4%|▍         | 1078/24420 [26:42<9:31:53,  1.47s/it]
  4%|▍         | 1079/24420 [26:44<9:31:45,  1.47s/it]
  4%|▍         | 1080/24420 [26:45<9:31:33,  1.47s/it]
                                                      
{'loss': 0.281, 'learning_rate': 5e-05, 'epoch': 0.22}

  4%|▍         | 1080/24420 [26:45<9:31:33,  1.47s/it]
  4%|▍         | 1081/24420 [26:47<9:31:07,  1.47s/it]
  4%|▍         | 1082/24420 [26:48<9:30:17,  1.47s/it]
  4%|▍         | 1083/24420 [26:50<9:29:59,  1.47s/it]
  4%|▍         | 1084/24420 [26:51<9:31:13,  1.47s/it]
  4%|▍         | 1085/24420 [26:53<9:31:03,  1.47s/it]
                                                      
{'loss': 0.2812, 'learning_rate': 5e-05, 'epoch': 0.22}

  4%|▍         | 1085/24420 [26:53<9:31:03,  1.47s/it]
  4%|▍         | 1086/24420 [26:54<9:30:45,  1.47s/it]
  4%|▍         | 1087/24420 [26:56<9:30:41,  1.47s/it]
  4%|▍         | 1088/24420 [26:57<9:32:12,  1.47s/it]
  4%|▍         | 1089/24420 [26:59<9:32:17,  1.47s/it]
  4%|▍         | 1090/24420 [27:00<9:31:59,  1.47s/it]
                                                      
{'loss': 0.3327, 'learning_rate': 5e-05, 'epoch': 0.22}

  4%|▍         | 1090/24420 [27:00<9:31:59,  1.47s/it]
  4%|▍         | 1091/24420 [27:01<9:31:23,  1.47s/it]
  4%|▍         | 1092/24420 [27:03<9:31:12,  1.47s/it]
  4%|▍         | 1093/24420 [27:04<9:30:05,  1.47s/it]
  4%|▍         | 1094/24420 [27:06<9:29:40,  1.47s/it]
  4%|▍         | 1095/24420 [27:07<9:30:16,  1.47s/it]
                                                      
{'loss': 0.2649, 'learning_rate': 5e-05, 'epoch': 0.22}

  4%|▍         | 1095/24420 [27:07<9:30:16,  1.47s/it]
  4%|▍         | 1096/24420 [27:09<9:30:23,  1.47s/it]
  4%|▍         | 1097/24420 [27:10<9:30:12,  1.47s/it]
  4%|▍         | 1098/24420 [27:12<9:30:10,  1.47s/it]
  5%|▍         | 1099/24420 [27:13<9:29:47,  1.47s/it]
  5%|▍         | 1100/24420 [27:15<9:29:09,  1.46s/it]
                                                      
{'loss': 0.2614, 'learning_rate': 5e-05, 'epoch': 0.23}

  5%|▍         | 1100/24420 [27:15<9:29:09,  1.46s/it]
  5%|▍         | 1101/24420 [27:16<9:29:58,  1.47s/it]
  5%|▍         | 1102/24420 [27:18<9:30:02,  1.47s/it]
  5%|▍         | 1103/24420 [27:19<9:29:46,  1.47s/it]
  5%|▍         | 1104/24420 [27:21<9:30:45,  1.47s/it]
  5%|▍         | 1105/24420 [27:22<9:30:10,  1.47s/it]
                                                      
{'loss': 0.2708, 'learning_rate': 5e-05, 'epoch': 0.23}

  5%|▍         | 1105/24420 [27:22<9:30:10,  1.47s/it]
  5%|▍         | 1106/24420 [27:23<9:30:52,  1.47s/it]
  5%|▍         | 1107/24420 [27:25<9:30:26,  1.47s/it]
  5%|▍         | 1108/24420 [27:26<9:30:58,  1.47s/it]
  5%|▍         | 1109/24420 [27:28<9:31:42,  1.47s/it]
  5%|▍         | 1110/24420 [27:29<9:31:27,  1.47s/it]
                                                      
{'loss': 0.2765, 'learning_rate': 5e-05, 'epoch': 0.23}

  5%|▍         | 1110/24420 [27:29<9:31:27,  1.47s/it]
  5%|▍         | 1111/24420 [27:31<9:31:47,  1.47s/it]
  5%|▍         | 1112/24420 [27:32<9:31:05,  1.47s/it]
  5%|▍         | 1113/24420 [27:34<9:32:11,  1.47s/it]
  5%|▍         | 1114/24420 [27:35<9:31:06,  1.47s/it]
  5%|▍         | 1115/24420 [27:37<9:31:01,  1.47s/it]
                                                      
{'loss': 0.3124, 'learning_rate': 5e-05, 'epoch': 0.23}

  5%|▍         | 1115/24420 [27:37<9:31:01,  1.47s/it]
  5%|▍         | 1116/24420 [27:38<9:30:40,  1.47s/it]
  5%|▍         | 1117/24420 [27:40<9:31:21,  1.47s/it]
  5%|▍         | 1118/24420 [27:41<9:31:34,  1.47s/it]
  5%|▍         | 1119/24420 [27:43<9:31:17,  1.47s/it]
  5%|▍         | 1120/24420 [27:44<9:31:47,  1.47s/it]
                                                      
{'loss': 0.29, 'learning_rate': 5e-05, 'epoch': 0.23}

  5%|▍         | 1120/24420 [27:44<9:31:47,  1.47s/it]
  5%|▍         | 1121/24420 [27:46<9:32:20,  1.47s/it]
  5%|▍         | 1122/24420 [27:47<9:31:34,  1.47s/it]
  5%|▍         | 1123/24420 [27:49<9:32:27,  1.47s/it]
  5%|▍         | 1124/24420 [27:50<9:31:52,  1.47s/it]
  5%|▍         | 1125/24420 [27:51<9:30:59,  1.47s/it]
                                                      
{'loss': 0.2725, 'learning_rate': 5e-05, 'epoch': 0.23}

  5%|▍         | 1125/24420 [27:51<9:30:59,  1.47s/it]
  5%|▍         | 1126/24420 [27:53<9:30:54,  1.47s/it]
  5%|▍         | 1127/24420 [27:54<9:32:01,  1.47s/it]
  5%|▍         | 1128/24420 [27:56<9:31:00,  1.47s/it]
  5%|▍         | 1129/24420 [27:57<9:30:44,  1.47s/it]
  5%|▍         | 1130/24420 [27:59<9:31:09,  1.47s/it]
                                                      
{'loss': 0.2483, 'learning_rate': 5e-05, 'epoch': 0.23}

  5%|▍         | 1130/24420 [27:59<9:31:09,  1.47s/it]
  5%|▍         | 1131/24420 [28:00<9:31:45,  1.47s/it]
  5%|▍         | 1132/24420 [28:02<9:31:39,  1.47s/it]
  5%|▍         | 1133/24420 [28:03<9:30:27,  1.47s/it]
  5%|▍         | 1134/24420 [28:05<9:30:24,  1.47s/it]
  5%|▍         | 1135/24420 [28:06<9:31:31,  1.47s/it]
                                                      
{'loss': 0.329, 'learning_rate': 5e-05, 'epoch': 0.23}

  5%|▍         | 1135/24420 [28:06<9:31:31,  1.47s/it]
  5%|▍         | 1136/24420 [28:08<9:31:56,  1.47s/it]
  5%|▍         | 1137/24420 [28:09<9:31:15,  1.47s/it]
  5%|▍         | 1138/24420 [28:11<9:29:54,  1.47s/it]
  5%|▍         | 1139/24420 [28:12<9:29:17,  1.47s/it]
  5%|▍         | 1140/24420 [28:13<9:28:48,  1.47s/it]
                                                      
{'loss': 0.259, 'learning_rate': 5e-05, 'epoch': 0.23}

  5%|▍         | 1140/24420 [28:13<9:28:48,  1.47s/it]
  5%|▍         | 1141/24420 [28:15<9:29:29,  1.47s/it]
  5%|▍         | 1142/24420 [28:16<9:30:25,  1.47s/it]
  5%|▍         | 1143/24420 [28:18<9:30:42,  1.47s/it]
  5%|▍         | 1144/24420 [28:19<9:30:43,  1.47s/it]
  5%|▍         | 1145/24420 [28:21<9:29:57,  1.47s/it]
                                                      
{'loss': 0.2863, 'learning_rate': 5e-05, 'epoch': 0.23}

  5%|▍         | 1145/24420 [28:21<9:29:57,  1.47s/it]
  5%|▍         | 1146/24420 [28:22<9:29:57,  1.47s/it]
  5%|▍         | 1147/24420 [28:24<9:29:37,  1.47s/it]
  5%|▍         | 1148/24420 [28:25<9:29:41,  1.47s/it]
  5%|▍         | 1149/24420 [28:27<9:30:49,  1.47s/it]
  5%|▍         | 1150/24420 [28:28<9:30:08,  1.47s/it]
                                                      
{'loss': 0.2753, 'learning_rate': 5e-05, 'epoch': 0.24}

  5%|▍         | 1150/24420 [28:28<9:30:08,  1.47s/it]
  5%|▍         | 1151/24420 [28:30<9:30:58,  1.47s/it]
  5%|▍         | 1152/24420 [28:31<9:30:51,  1.47s/it]
  5%|▍         | 1153/24420 [28:33<9:29:47,  1.47s/it]
  5%|▍         | 1154/24420 [28:34<9:30:08,  1.47s/it]
  5%|▍         | 1155/24420 [28:36<9:29:44,  1.47s/it]
                                                      
{'loss': 0.24, 'learning_rate': 5e-05, 'epoch': 0.24}

  5%|▍         | 1155/24420 [28:36<9:29:44,  1.47s/it]
  5%|▍         | 1156/24420 [28:37<9:29:56,  1.47s/it]
  5%|▍         | 1157/24420 [28:38<9:29:40,  1.47s/it]
  5%|▍         | 1158/24420 [28:40<9:43:05,  1.50s/it]
  5%|▍         | 1159/24420 [28:42<9:38:52,  1.49s/it]
  5%|▍         | 1160/24420 [28:43<9:35:53,  1.49s/it]
                                                      
{'loss': 0.2579, 'learning_rate': 5e-05, 'epoch': 0.24}

  5%|▍         | 1160/24420 [28:43<9:35:53,  1.49s/it]
  5%|▍         | 1161/24420 [28:44<9:33:53,  1.48s/it]
  5%|▍         | 1162/24420 [28:46<9:32:04,  1.48s/it]
  5%|▍         | 1163/24420 [28:47<9:32:18,  1.48s/it]
  5%|▍         | 1164/24420 [28:49<9:31:59,  1.48s/it]
  5%|▍         | 1165/24420 [28:50<9:30:38,  1.47s/it]
                                                      
{'loss': 0.2808, 'learning_rate': 5e-05, 'epoch': 0.24}

  5%|▍         | 1165/24420 [28:50<9:30:38,  1.47s/it]
  5%|▍         | 1166/24420 [28:52<9:31:46,  1.48s/it]
  5%|▍         | 1167/24420 [28:53<9:31:05,  1.47s/it]
  5%|▍         | 1168/24420 [28:55<9:31:27,  1.47s/it]
  5%|▍         | 1169/24420 [28:56<9:30:27,  1.47s/it]
  5%|▍         | 1170/24420 [28:58<9:29:32,  1.47s/it]
                                                      
{'loss': 0.3037, 'learning_rate': 5e-05, 'epoch': 0.24}

  5%|▍         | 1170/24420 [28:58<9:29:32,  1.47s/it]
  5%|▍         | 1171/24420 [28:59<9:29:31,  1.47s/it]
  5%|▍         | 1172/24420 [29:01<9:29:18,  1.47s/it]
  5%|▍         | 1173/24420 [29:02<9:29:49,  1.47s/it]
  5%|▍         | 1174/24420 [29:04<9:29:40,  1.47s/it]
  5%|▍         | 1175/24420 [29:05<9:29:14,  1.47s/it]
                                                      
{'loss': 0.2912, 'learning_rate': 5e-05, 'epoch': 0.24}

  5%|▍         | 1175/24420 [29:05<9:29:14,  1.47s/it]
  5%|▍         | 1176/24420 [29:07<9:29:31,  1.47s/it]
  5%|▍         | 1177/24420 [29:08<9:28:55,  1.47s/it]
  5%|▍         | 1178/24420 [29:09<9:28:29,  1.47s/it]
  5%|▍         | 1179/24420 [29:11<9:27:49,  1.47s/it]
  5%|▍         | 1180/24420 [29:12<9:28:08,  1.47s/it]
                                                      
{'loss': 0.2184, 'learning_rate': 5e-05, 'epoch': 0.24}

  5%|▍         | 1180/24420 [29:12<9:28:08,  1.47s/it]
  5%|▍         | 1181/24420 [29:14<9:28:33,  1.47s/it]
  5%|▍         | 1182/24420 [29:15<9:28:45,  1.47s/it]
  5%|▍         | 1183/24420 [29:17<9:29:29,  1.47s/it]
  5%|▍         | 1184/24420 [29:18<9:29:50,  1.47s/it]
  5%|▍         | 1185/24420 [29:20<9:29:31,  1.47s/it]
                                                      
{'loss': 0.3175, 'learning_rate': 5e-05, 'epoch': 0.24}

  5%|▍         | 1185/24420 [29:20<9:29:31,  1.47s/it]
  5%|▍         | 1186/24420 [29:21<9:29:35,  1.47s/it]
  5%|▍         | 1187/24420 [29:23<9:29:26,  1.47s/it]
  5%|▍         | 1188/24420 [29:24<9:29:19,  1.47s/it]
  5%|▍         | 1189/24420 [29:26<9:29:59,  1.47s/it]
  5%|▍         | 1190/24420 [29:27<9:29:55,  1.47s/it]
                                                      
{'loss': 0.202, 'learning_rate': 5e-05, 'epoch': 0.24}

  5%|▍         | 1190/24420 [29:27<9:29:55,  1.47s/it]
  5%|▍         | 1191/24420 [29:29<9:29:33,  1.47s/it]
  5%|▍         | 1192/24420 [29:30<9:28:25,  1.47s/it]
  5%|▍         | 1193/24420 [29:32<9:28:49,  1.47s/it]
  5%|▍         | 1194/24420 [29:33<9:28:34,  1.47s/it]
  5%|▍         | 1195/24420 [29:34<9:28:17,  1.47s/it]
                                                      
{'loss': 0.2754, 'learning_rate': 5e-05, 'epoch': 0.24}

  5%|▍         | 1195/24420 [29:34<9:28:17,  1.47s/it]
  5%|▍         | 1196/24420 [29:36<9:29:23,  1.47s/it]
  5%|▍         | 1197/24420 [29:37<9:30:57,  1.48s/it]
  5%|▍         | 1198/24420 [29:39<9:34:00,  1.48s/it]
  5%|▍         | 1199/24420 [29:40<9:32:27,  1.48s/it]
  5%|▍         | 1200/24420 [29:42<9:31:57,  1.48s/it]
                                                      
{'loss': 0.2688, 'learning_rate': 5e-05, 'epoch': 0.25}

  5%|▍         | 1200/24420 [29:42<9:31:57,  1.48s/it]
  5%|▍         | 1201/24420 [29:48<18:44:16,  2.91s/it]
  5%|▍         | 1202/24420 [29:50<15:56:42,  2.47s/it]
  5%|▍         | 1203/24420 [29:51<14:00:25,  2.17s/it]
  5%|▍         | 1204/24420 [29:53<12:39:18,  1.96s/it]
  5%|▍         | 1205/24420 [29:54<11:41:43,  1.81s/it]
                                                       
{'loss': 0.2579, 'learning_rate': 5e-05, 'epoch': 0.25}

  5%|▍         | 1205/24420 [29:54<11:41:43,  1.81s/it]
  5%|▍         | 1206/24420 [29:55<11:03:20,  1.71s/it]
  5%|▍         | 1207/24420 [29:57<10:34:49,  1.64s/it]
  5%|▍         | 1208/24420 [29:58<10:15:29,  1.59s/it]
  5%|▍         | 1209/24420 [30:00<10:00:33,  1.55s/it]
  5%|▍         | 1210/24420 [30:01<9:50:35,  1.53s/it] 
                                                      
{'loss': 0.2951, 'learning_rate': 5e-05, 'epoch': 0.25}

  5%|▍         | 1210/24420 [30:01<9:50:35,  1.53s/it]
  5%|▍         | 1211/24420 [30:03<9:44:34,  1.51s/it]
  5%|▍         | 1212/24420 [30:04<9:39:14,  1.50s/it]
  5%|▍         | 1213/24420 [30:06<9:37:19,  1.49s/it]
  5%|▍         | 1214/24420 [30:07<9:33:53,  1.48s/it]
  5%|▍         | 1215/24420 [30:09<9:32:43,  1.48s/it]
                                                      
{'loss': 0.2739, 'learning_rate': 5e-05, 'epoch': 0.25}

  5%|▍         | 1215/24420 [30:09<9:32:43,  1.48s/it]
  5%|▍         | 1216/24420 [30:10<9:32:02,  1.48s/it]
  5%|▍         | 1217/24420 [30:12<9:30:20,  1.47s/it]
  5%|▍         | 1218/24420 [30:13<9:30:39,  1.48s/it]
  5%|▍         | 1219/24420 [30:15<9:29:31,  1.47s/it]
  5%|▍         | 1220/24420 [30:16<9:28:57,  1.47s/it]
                                                      
{'loss': 0.2754, 'learning_rate': 5e-05, 'epoch': 0.25}

  5%|▍         | 1220/24420 [30:16<9:28:57,  1.47s/it]
  5%|▌         | 1221/24420 [30:18<9:29:03,  1.47s/it]
  5%|▌         | 1222/24420 [30:19<9:29:03,  1.47s/it]
  5%|▌         | 1223/24420 [30:20<9:28:53,  1.47s/it]
  5%|▌         | 1224/24420 [30:22<9:28:04,  1.47s/it]
  5%|▌         | 1225/24420 [30:23<9:27:19,  1.47s/it]
                                                      
{'loss': 0.2557, 'learning_rate': 5e-05, 'epoch': 0.25}

  5%|▌         | 1225/24420 [30:23<9:27:19,  1.47s/it]
  5%|▌         | 1226/24420 [30:25<9:29:18,  1.47s/it]
  5%|▌         | 1227/24420 [30:26<9:30:17,  1.48s/it]
  5%|▌         | 1228/24420 [30:28<9:30:06,  1.47s/it]
  5%|▌         | 1229/24420 [30:29<9:29:35,  1.47s/it]
  5%|▌         | 1230/24420 [30:31<9:29:40,  1.47s/it]
                                                      
{'loss': 0.2894, 'learning_rate': 5e-05, 'epoch': 0.25}

  5%|▌         | 1230/24420 [30:31<9:29:40,  1.47s/it]
  5%|▌         | 1231/24420 [30:32<9:31:15,  1.48s/it]
  5%|▌         | 1232/24420 [30:34<9:30:29,  1.48s/it]
  5%|▌         | 1233/24420 [30:35<9:29:34,  1.47s/it]
  5%|▌         | 1234/24420 [30:37<9:28:53,  1.47s/it]
  5%|▌         | 1235/24420 [30:38<9:29:09,  1.47s/it]
                                                      
{'loss': 0.2559, 'learning_rate': 5e-05, 'epoch': 0.25}

  5%|▌         | 1235/24420 [30:38<9:29:09,  1.47s/it]
  5%|▌         | 1236/24420 [30:40<9:29:03,  1.47s/it]
  5%|▌         | 1237/24420 [30:41<9:29:30,  1.47s/it]
  5%|▌         | 1238/24420 [30:43<9:28:45,  1.47s/it]
  5%|▌         | 1239/24420 [30:44<9:28:08,  1.47s/it]
  5%|▌         | 1240/24420 [30:46<9:28:53,  1.47s/it]
                                                      
{'loss': 0.243, 'learning_rate': 5e-05, 'epoch': 0.25}

  5%|▌         | 1240/24420 [30:46<9:28:53,  1.47s/it]
  5%|▌         | 1241/24420 [30:47<9:29:25,  1.47s/it]
  5%|▌         | 1242/24420 [30:48<9:28:28,  1.47s/it]
  5%|▌         | 1243/24420 [30:50<9:29:02,  1.47s/it]
  5%|▌         | 1244/24420 [30:51<9:30:04,  1.48s/it]
  5%|▌         | 1245/24420 [30:53<9:29:47,  1.48s/it]
                                                      
{'loss': 0.2456, 'learning_rate': 5e-05, 'epoch': 0.25}

  5%|▌         | 1245/24420 [30:53<9:29:47,  1.48s/it]
  5%|▌         | 1246/24420 [30:54<9:29:19,  1.47s/it]
  5%|▌         | 1247/24420 [30:56<9:29:15,  1.47s/it]
  5%|▌         | 1248/24420 [30:57<9:30:10,  1.48s/it]
  5%|▌         | 1249/24420 [30:59<9:30:21,  1.48s/it]
  5%|▌         | 1250/24420 [31:00<9:29:13,  1.47s/it]
                                                      
{'loss': 0.3231, 'learning_rate': 5e-05, 'epoch': 0.26}

  5%|▌         | 1250/24420 [31:00<9:29:13,  1.47s/it]
  5%|▌         | 1251/24420 [31:02<9:28:51,  1.47s/it]
  5%|▌         | 1252/24420 [31:03<9:29:15,  1.47s/it]
  5%|▌         | 1253/24420 [31:05<9:32:01,  1.48s/it]
  5%|▌         | 1254/24420 [31:06<9:30:56,  1.48s/it]
  5%|▌         | 1255/24420 [31:08<9:29:18,  1.47s/it]
                                                      
{'loss': 0.2295, 'learning_rate': 5e-05, 'epoch': 0.26}

  5%|▌         | 1255/24420 [31:08<9:29:18,  1.47s/it]
  5%|▌         | 1256/24420 [31:09<9:28:39,  1.47s/it]
  5%|▌         | 1257/24420 [31:11<9:28:29,  1.47s/it]
  5%|▌         | 1258/24420 [31:12<9:27:59,  1.47s/it]
  5%|▌         | 1259/24420 [31:14<9:46:21,  1.52s/it]
  5%|▌         | 1260/24420 [31:15<9:40:25,  1.50s/it]
                                                      
{'loss': 0.3158, 'learning_rate': 5e-05, 'epoch': 0.26}

  5%|▌         | 1260/24420 [31:15<9:40:25,  1.50s/it]
  5%|▌         | 1261/24420 [31:17<9:37:17,  1.50s/it]
  5%|▌         | 1262/24420 [31:18<9:34:56,  1.49s/it]
  5%|▌         | 1263/24420 [31:20<9:33:54,  1.49s/it]
  5%|▌         | 1264/24420 [31:21<9:31:32,  1.48s/it]
  5%|▌         | 1265/24420 [31:23<9:29:43,  1.48s/it]
                                                      
{'loss': 0.2281, 'learning_rate': 5e-05, 'epoch': 0.26}

  5%|▌         | 1265/24420 [31:23<9:29:43,  1.48s/it]
  5%|▌         | 1266/24420 [31:24<9:29:20,  1.48s/it]
  5%|▌         | 1267/24420 [31:25<9:27:49,  1.47s/it]
  5%|▌         | 1268/24420 [31:27<9:28:23,  1.47s/it]
  5%|▌         | 1269/24420 [31:28<9:29:04,  1.47s/it]
  5%|▌         | 1270/24420 [31:30<9:28:47,  1.47s/it]
                                                      
{'loss': 0.2526, 'learning_rate': 5e-05, 'epoch': 0.26}

  5%|▌         | 1270/24420 [31:30<9:28:47,  1.47s/it]
  5%|▌         | 1271/24420 [31:31<9:28:12,  1.47s/it]
  5%|▌         | 1272/24420 [31:33<9:29:04,  1.48s/it]
  5%|▌         | 1273/24420 [31:34<9:28:28,  1.47s/it]
  5%|▌         | 1274/24420 [31:36<9:27:59,  1.47s/it]
  5%|▌         | 1275/24420 [31:37<9:27:37,  1.47s/it]
                                                      
{'loss': 0.2462, 'learning_rate': 5e-05, 'epoch': 0.26}

  5%|▌         | 1275/24420 [31:37<9:27:37,  1.47s/it]
  5%|▌         | 1276/24420 [31:39<9:27:33,  1.47s/it]
  5%|▌         | 1277/24420 [31:40<9:27:49,  1.47s/it]
  5%|▌         | 1278/24420 [31:42<9:26:29,  1.47s/it]
  5%|▌         | 1279/24420 [31:43<9:26:53,  1.47s/it]
  5%|▌         | 1280/24420 [31:45<9:27:22,  1.47s/it]
                                                      
{'loss': 0.2464, 'learning_rate': 5e-05, 'epoch': 0.26}

  5%|▌         | 1280/24420 [31:45<9:27:22,  1.47s/it]
  5%|▌         | 1281/24420 [31:46<9:27:16,  1.47s/it]
  5%|▌         | 1282/24420 [31:48<9:27:14,  1.47s/it]
  5%|▌         | 1283/24420 [31:49<9:27:19,  1.47s/it]
  5%|▌         | 1284/24420 [31:50<9:27:30,  1.47s/it]
  5%|▌         | 1285/24420 [31:52<9:27:26,  1.47s/it]
                                                      
{'loss': 0.2602, 'learning_rate': 5e-05, 'epoch': 0.26}

  5%|▌         | 1285/24420 [31:52<9:27:26,  1.47s/it]
  5%|▌         | 1286/24420 [31:53<9:27:30,  1.47s/it]
  5%|▌         | 1287/24420 [31:55<9:27:41,  1.47s/it]
  5%|▌         | 1288/24420 [31:56<9:26:48,  1.47s/it]
  5%|▌         | 1289/24420 [31:58<9:27:59,  1.47s/it]
  5%|▌         | 1290/24420 [31:59<9:27:16,  1.47s/it]
                                                      
{'loss': 0.3167, 'learning_rate': 5e-05, 'epoch': 0.26}

  5%|▌         | 1290/24420 [31:59<9:27:16,  1.47s/it]
  5%|▌         | 1291/24420 [32:01<9:27:59,  1.47s/it]
  5%|▌         | 1292/24420 [32:02<9:27:53,  1.47s/it]
  5%|▌         | 1293/24420 [32:04<9:27:43,  1.47s/it]
  5%|▌         | 1294/24420 [32:05<9:29:00,  1.48s/it]
  5%|▌         | 1295/24420 [32:07<9:28:56,  1.48s/it]
                                                      
{'loss': 0.2821, 'learning_rate': 5e-05, 'epoch': 0.27}

  5%|▌         | 1295/24420 [32:07<9:28:56,  1.48s/it]
  5%|▌         | 1296/24420 [32:08<9:29:36,  1.48s/it]
  5%|▌         | 1297/24420 [32:10<9:29:31,  1.48s/it]
  5%|▌         | 1298/24420 [32:11<9:29:40,  1.48s/it]
  5%|▌         | 1299/24420 [32:13<9:28:49,  1.48s/it]
  5%|▌         | 1300/24420 [32:14<9:28:07,  1.47s/it]
                                                      
{'loss': 0.269, 'learning_rate': 5e-05, 'epoch': 0.27}

  5%|▌         | 1300/24420 [32:14<9:28:07,  1.47s/it]
  5%|▌         | 1301/24420 [32:16<9:28:13,  1.47s/it]
  5%|▌         | 1302/24420 [32:17<9:28:02,  1.47s/it]
  5%|▌         | 1303/24420 [32:18<9:27:55,  1.47s/it]
  5%|▌         | 1304/24420 [32:20<9:27:32,  1.47s/it]
  5%|▌         | 1305/24420 [32:21<9:26:52,  1.47s/it]
                                                      
{'loss': 0.2565, 'learning_rate': 5e-05, 'epoch': 0.27}

  5%|▌         | 1305/24420 [32:21<9:26:52,  1.47s/it]
  5%|▌         | 1306/24420 [32:23<9:27:30,  1.47s/it]
  5%|▌         | 1307/24420 [32:24<9:28:53,  1.48s/it]
  5%|▌         | 1308/24420 [32:26<9:28:45,  1.48s/it]
  5%|▌         | 1309/24420 [32:27<9:28:26,  1.48s/it]
  5%|▌         | 1310/24420 [32:29<9:27:17,  1.47s/it]
                                                      
{'loss': 0.2554, 'learning_rate': 5e-05, 'epoch': 0.27}

  5%|▌         | 1310/24420 [32:29<9:27:17,  1.47s/it]
  5%|▌         | 1311/24420 [32:30<9:27:02,  1.47s/it]
  5%|▌         | 1312/24420 [32:32<9:27:11,  1.47s/it]
  5%|▌         | 1313/24420 [32:33<9:27:19,  1.47s/it]
  5%|▌         | 1314/24420 [32:35<9:28:14,  1.48s/it]
  5%|▌         | 1315/24420 [32:36<9:28:32,  1.48s/it]
                                                      
{'loss': 0.249, 'learning_rate': 5e-05, 'epoch': 0.27}

  5%|▌         | 1315/24420 [32:36<9:28:32,  1.48s/it]
  5%|▌         | 1316/24420 [32:38<9:28:13,  1.48s/it]
  5%|▌         | 1317/24420 [32:39<9:26:41,  1.47s/it]
  5%|▌         | 1318/24420 [32:41<9:26:13,  1.47s/it]
  5%|▌         | 1319/24420 [32:42<9:26:28,  1.47s/it]
  5%|▌         | 1320/24420 [32:44<9:25:57,  1.47s/it]
                                                      
{'loss': 0.2422, 'learning_rate': 5e-05, 'epoch': 0.27}

  5%|▌         | 1320/24420 [32:44<9:25:57,  1.47s/it]
  5%|▌         | 1321/24420 [32:45<9:27:24,  1.47s/it]
  5%|▌         | 1322/24420 [32:46<9:27:20,  1.47s/it]
  5%|▌         | 1323/24420 [32:48<9:28:32,  1.48s/it]
  5%|▌         | 1324/24420 [32:49<9:29:27,  1.48s/it]
  5%|▌         | 1325/24420 [32:51<9:29:35,  1.48s/it]
                                                      
{'loss': 0.2476, 'learning_rate': 5e-05, 'epoch': 0.27}

  5%|▌         | 1325/24420 [32:51<9:29:35,  1.48s/it]
  5%|▌         | 1326/24420 [32:52<9:28:55,  1.48s/it]
  5%|▌         | 1327/24420 [32:54<9:28:37,  1.48s/it]
  5%|▌         | 1328/24420 [32:55<9:27:57,  1.48s/it]
  5%|▌         | 1329/24420 [32:57<9:27:54,  1.48s/it]
  5%|▌         | 1330/24420 [32:58<9:27:39,  1.48s/it]
                                                      
{'loss': 0.2738, 'learning_rate': 5e-05, 'epoch': 0.27}

  5%|▌         | 1330/24420 [32:58<9:27:39,  1.48s/it]
  5%|▌         | 1331/24420 [33:00<9:27:28,  1.47s/it]
  5%|▌         | 1332/24420 [33:01<9:27:24,  1.47s/it]
  5%|▌         | 1333/24420 [33:03<9:27:53,  1.48s/it]
  5%|▌         | 1334/24420 [33:04<9:27:52,  1.48s/it]
  5%|▌         | 1335/24420 [33:06<9:26:15,  1.47s/it]
                                                      
{'loss': 0.2701, 'learning_rate': 5e-05, 'epoch': 0.27}

  5%|▌         | 1335/24420 [33:06<9:26:15,  1.47s/it]
  5%|▌         | 1336/24420 [33:07<9:26:28,  1.47s/it]
  5%|▌         | 1337/24420 [33:09<9:26:42,  1.47s/it]
  5%|▌         | 1338/24420 [33:10<9:26:04,  1.47s/it]
  5%|▌         | 1339/24420 [33:12<9:26:29,  1.47s/it]
  5%|▌         | 1340/24420 [33:13<9:28:43,  1.48s/it]
                                                      
{'loss': 0.2614, 'learning_rate': 5e-05, 'epoch': 0.27}

  5%|▌         | 1340/24420 [33:13<9:28:43,  1.48s/it]
  5%|▌         | 1341/24420 [33:15<9:30:11,  1.48s/it]
  5%|▌         | 1342/24420 [33:16<9:30:02,  1.48s/it]
  5%|▌         | 1343/24420 [33:17<9:28:12,  1.48s/it]
  6%|▌         | 1344/24420 [33:19<9:27:49,  1.48s/it]
  6%|▌         | 1345/24420 [33:20<9:28:26,  1.48s/it]
                                                      
{'loss': 0.2793, 'learning_rate': 5e-05, 'epoch': 0.28}

  6%|▌         | 1345/24420 [33:20<9:28:26,  1.48s/it]
  6%|▌         | 1346/24420 [33:22<9:28:30,  1.48s/it]
  6%|▌         | 1347/24420 [33:23<9:29:52,  1.48s/it]
  6%|▌         | 1348/24420 [33:25<9:28:49,  1.48s/it]
  6%|▌         | 1349/24420 [33:26<9:28:04,  1.48s/it]
  6%|▌         | 1350/24420 [33:28<9:27:47,  1.48s/it]
                                                      
{'loss': 0.2378, 'learning_rate': 5e-05, 'epoch': 0.28}

  6%|▌         | 1350/24420 [33:28<9:27:47,  1.48s/it]
  6%|▌         | 1351/24420 [33:29<9:35:04,  1.50s/it]
  6%|▌         | 1352/24420 [33:31<9:31:55,  1.49s/it]
  6%|▌         | 1353/24420 [33:32<9:29:58,  1.48s/it]
  6%|▌         | 1354/24420 [33:34<9:29:13,  1.48s/it]
  6%|▌         | 1355/24420 [33:35<9:27:33,  1.48s/it]
                                                      
{'loss': 0.2688, 'learning_rate': 5e-05, 'epoch': 0.28}

  6%|▌         | 1355/24420 [33:35<9:27:33,  1.48s/it]
  6%|▌         | 1356/24420 [33:37<9:27:57,  1.48s/it]
  6%|▌         | 1357/24420 [33:38<9:28:30,  1.48s/it]
  6%|▌         | 1358/24420 [33:40<9:28:52,  1.48s/it]
  6%|▌         | 1359/24420 [33:41<9:27:48,  1.48s/it]
  6%|▌         | 1360/24420 [33:43<9:26:05,  1.47s/it]
                                                      
{'loss': 0.2408, 'learning_rate': 5e-05, 'epoch': 0.28}

  6%|▌         | 1360/24420 [33:43<9:26:05,  1.47s/it]
  6%|▌         | 1361/24420 [33:44<9:26:01,  1.47s/it]
  6%|▌         | 1362/24420 [33:46<9:25:49,  1.47s/it]
  6%|▌         | 1363/24420 [33:47<9:25:45,  1.47s/it]
  6%|▌         | 1364/24420 [33:49<9:26:41,  1.47s/it]
  6%|▌         | 1365/24420 [33:50<9:25:36,  1.47s/it]
                                                      
{'loss': 0.2226, 'learning_rate': 5e-05, 'epoch': 0.28}

  6%|▌         | 1365/24420 [33:50<9:25:36,  1.47s/it]
  6%|▌         | 1366/24420 [33:51<9:25:39,  1.47s/it]
  6%|▌         | 1367/24420 [33:53<9:26:49,  1.48s/it]
  6%|▌         | 1368/24420 [33:54<9:26:31,  1.47s/it]
  6%|▌         | 1369/24420 [33:56<9:26:44,  1.48s/it]
  6%|▌         | 1370/24420 [33:57<9:26:42,  1.48s/it]
                                                      
{'loss': 0.2443, 'learning_rate': 5e-05, 'epoch': 0.28}

  6%|▌         | 1370/24420 [33:57<9:26:42,  1.48s/it]
  6%|▌         | 1371/24420 [33:59<9:27:02,  1.48s/it]
  6%|▌         | 1372/24420 [34:00<9:26:51,  1.48s/it]
  6%|▌         | 1373/24420 [34:02<9:26:13,  1.47s/it]
  6%|▌         | 1374/24420 [34:03<9:25:17,  1.47s/it]
  6%|▌         | 1375/24420 [34:05<9:25:03,  1.47s/it]
                                                      
{'loss': 0.2474, 'learning_rate': 5e-05, 'epoch': 0.28}

  6%|▌         | 1375/24420 [34:05<9:25:03,  1.47s/it]
  6%|▌         | 1376/24420 [34:06<9:24:45,  1.47s/it]
  6%|▌         | 1377/24420 [34:08<9:25:42,  1.47s/it]
  6%|▌         | 1378/24420 [34:09<9:26:09,  1.47s/it]
  6%|▌         | 1379/24420 [34:11<9:25:29,  1.47s/it]
  6%|▌         | 1380/24420 [34:12<9:24:38,  1.47s/it]
                                                      
{'loss': 0.2272, 'learning_rate': 5e-05, 'epoch': 0.28}

  6%|▌         | 1380/24420 [34:12<9:24:38,  1.47s/it]
  6%|▌         | 1381/24420 [34:14<9:24:31,  1.47s/it]
  6%|▌         | 1382/24420 [34:15<9:25:03,  1.47s/it]
  6%|▌         | 1383/24420 [34:17<9:25:07,  1.47s/it]
  6%|▌         | 1384/24420 [34:18<9:26:30,  1.48s/it]
  6%|▌         | 1385/24420 [34:19<9:26:44,  1.48s/it]
                                                      
{'loss': 0.2488, 'learning_rate': 5e-05, 'epoch': 0.28}

  6%|▌         | 1385/24420 [34:19<9:26:44,  1.48s/it]
  6%|▌         | 1386/24420 [34:21<9:27:34,  1.48s/it]
  6%|▌         | 1387/24420 [34:22<9:26:29,  1.48s/it]
  6%|▌         | 1388/24420 [34:24<9:26:45,  1.48s/it]
  6%|▌         | 1389/24420 [34:25<9:26:27,  1.48s/it]
  6%|▌         | 1390/24420 [34:27<9:26:27,  1.48s/it]
                                                      
{'loss': 0.2791, 'learning_rate': 5e-05, 'epoch': 0.28}

  6%|▌         | 1390/24420 [34:27<9:26:27,  1.48s/it]
  6%|▌         | 1391/24420 [34:28<9:25:49,  1.47s/it]
  6%|▌         | 1392/24420 [34:30<9:26:01,  1.47s/it]
  6%|▌         | 1393/24420 [34:31<9:26:24,  1.48s/it]
  6%|▌         | 1394/24420 [34:33<9:26:46,  1.48s/it]
  6%|▌         | 1395/24420 [34:34<9:27:16,  1.48s/it]
                                                      
{'loss': 0.2543, 'learning_rate': 5e-05, 'epoch': 0.29}

  6%|▌         | 1395/24420 [34:34<9:27:16,  1.48s/it]
  6%|▌         | 1396/24420 [34:36<9:27:50,  1.48s/it]
  6%|▌         | 1397/24420 [34:37<9:28:05,  1.48s/it]
  6%|▌         | 1398/24420 [34:39<9:26:51,  1.48s/it]
  6%|▌         | 1399/24420 [34:40<9:25:47,  1.47s/it]
  6%|▌         | 1400/24420 [34:42<9:25:32,  1.47s/it]
                                                      
{'loss': 0.2502, 'learning_rate': 5e-05, 'epoch': 0.29}

  6%|▌         | 1400/24420 [34:42<9:25:32,  1.47s/it]
  6%|▌         | 1401/24420 [34:46<15:24:43,  2.41s/it]
  6%|▌         | 1402/24420 [34:48<13:38:24,  2.13s/it]
  6%|▌         | 1403/24420 [34:49<12:22:32,  1.94s/it]
  6%|▌         | 1404/24420 [34:51<11:30:35,  1.80s/it]
  6%|▌         | 1405/24420 [34:52<10:53:15,  1.70s/it]
                                                       
{'loss': 0.2556, 'learning_rate': 5e-05, 'epoch': 0.29}

  6%|▌         | 1405/24420 [34:52<10:53:15,  1.70s/it]
  6%|▌         | 1406/24420 [34:54<10:27:13,  1.64s/it]
  6%|▌         | 1407/24420 [34:55<10:09:09,  1.59s/it]
  6%|▌         | 1408/24420 [34:57<9:55:43,  1.55s/it] 
  6%|▌         | 1409/24420 [34:58<9:47:33,  1.53s/it]
  6%|▌         | 1410/24420 [35:00<9:40:50,  1.51s/it]
                                                      
{'loss': 0.2665, 'learning_rate': 5e-05, 'epoch': 0.29}

  6%|▌         | 1410/24420 [35:00<9:40:50,  1.51s/it]
  6%|▌         | 1411/24420 [35:01<9:37:06,  1.50s/it]
  6%|▌         | 1412/24420 [35:02<9:32:31,  1.49s/it]
  6%|▌         | 1413/24420 [35:04<9:30:52,  1.49s/it]
  6%|▌         | 1414/24420 [35:05<9:28:11,  1.48s/it]
  6%|▌         | 1415/24420 [35:07<9:27:13,  1.48s/it]
                                                      
{'loss': 0.2558, 'learning_rate': 5e-05, 'epoch': 0.29}

  6%|▌         | 1415/24420 [35:07<9:27:13,  1.48s/it]
  6%|▌         | 1416/24420 [35:08<9:27:25,  1.48s/it]
  6%|▌         | 1417/24420 [35:10<9:25:42,  1.48s/it]
  6%|▌         | 1418/24420 [35:11<9:25:51,  1.48s/it]
  6%|▌         | 1419/24420 [35:13<9:26:35,  1.48s/it]
  6%|▌         | 1420/24420 [35:14<9:26:59,  1.48s/it]
                                                      
{'loss': 0.2745, 'learning_rate': 5e-05, 'epoch': 0.29}

  6%|▌         | 1420/24420 [35:14<9:26:59,  1.48s/it]
  6%|▌         | 1421/24420 [35:16<9:26:02,  1.48s/it]
  6%|▌         | 1422/24420 [35:17<9:26:31,  1.48s/it]
  6%|▌         | 1423/24420 [35:19<9:27:52,  1.48s/it]
  6%|▌         | 1424/24420 [35:20<9:26:16,  1.48s/it]
  6%|▌         | 1425/24420 [35:22<9:26:42,  1.48s/it]
                                                      
{'loss': 0.2886, 'learning_rate': 5e-05, 'epoch': 0.29}

  6%|▌         | 1425/24420 [35:22<9:26:42,  1.48s/it]
  6%|▌         | 1426/24420 [35:23<9:26:47,  1.48s/it]
  6%|▌         | 1427/24420 [35:25<9:26:32,  1.48s/it]
  6%|▌         | 1428/24420 [35:26<9:26:30,  1.48s/it]
  6%|▌         | 1429/24420 [35:28<9:24:47,  1.47s/it]
  6%|▌         | 1430/24420 [35:29<9:23:56,  1.47s/it]
                                                      
{'loss': 0.2268, 'learning_rate': 5e-05, 'epoch': 0.29}

  6%|▌         | 1430/24420 [35:29<9:23:56,  1.47s/it]
  6%|▌         | 1431/24420 [35:31<9:23:34,  1.47s/it]
  6%|▌         | 1432/24420 [35:32<9:23:41,  1.47s/it]
  6%|▌         | 1433/24420 [35:33<9:24:28,  1.47s/it]
  6%|▌         | 1434/24420 [35:35<9:24:18,  1.47s/it]
  6%|▌         | 1435/24420 [35:36<9:25:17,  1.48s/it]
                                                      
{'loss': 0.2794, 'learning_rate': 5e-05, 'epoch': 0.29}

  6%|▌         | 1435/24420 [35:36<9:25:17,  1.48s/it]
  6%|▌         | 1436/24420 [35:38<9:27:10,  1.48s/it]
  6%|▌         | 1437/24420 [35:39<9:26:47,  1.48s/it]
  6%|▌         | 1438/24420 [35:41<9:25:52,  1.48s/it]
  6%|▌         | 1439/24420 [35:42<9:24:51,  1.47s/it]
  6%|▌         | 1440/24420 [35:44<9:25:35,  1.48s/it]
                                                      
{'loss': 0.2012, 'learning_rate': 5e-05, 'epoch': 0.29}

  6%|▌         | 1440/24420 [35:44<9:25:35,  1.48s/it]
  6%|▌         | 1441/24420 [35:45<9:25:05,  1.48s/it]
  6%|▌         | 1442/24420 [35:47<9:24:35,  1.47s/it]
  6%|▌         | 1443/24420 [35:48<9:24:29,  1.47s/it]
  6%|▌         | 1444/24420 [35:50<9:24:44,  1.47s/it]
  6%|▌         | 1445/24420 [35:51<9:25:46,  1.48s/it]
                                                      
{'loss': 0.257, 'learning_rate': 5e-05, 'epoch': 0.3}

  6%|▌         | 1445/24420 [35:51<9:25:46,  1.48s/it]
  6%|▌         | 1446/24420 [35:53<9:25:47,  1.48s/it]
  6%|▌         | 1447/24420 [35:54<9:24:21,  1.47s/it]
  6%|▌         | 1448/24420 [35:56<9:25:10,  1.48s/it]
  6%|▌         | 1449/24420 [35:57<9:25:43,  1.48s/it]
  6%|▌         | 1450/24420 [35:59<9:25:37,  1.48s/it]
                                                      
{'loss': 0.275, 'learning_rate': 5e-05, 'epoch': 0.3}

  6%|▌         | 1450/24420 [35:59<9:25:37,  1.48s/it]
  6%|▌         | 1451/24420 [36:00<9:25:20,  1.48s/it]
  6%|▌         | 1452/24420 [36:02<9:26:40,  1.48s/it]
  6%|▌         | 1453/24420 [36:03<9:26:15,  1.48s/it]
  6%|▌         | 1454/24420 [36:04<9:25:50,  1.48s/it]
  6%|▌         | 1455/24420 [36:06<9:24:41,  1.48s/it]
                                                      
{'loss': 0.2422, 'learning_rate': 5e-05, 'epoch': 0.3}

  6%|▌         | 1455/24420 [36:06<9:24:41,  1.48s/it]
  6%|▌         | 1456/24420 [36:07<9:24:26,  1.47s/it]
  6%|▌         | 1457/24420 [36:09<9:24:52,  1.48s/it]
  6%|▌         | 1458/24420 [36:10<9:25:00,  1.48s/it]
  6%|▌         | 1459/24420 [36:12<9:24:30,  1.48s/it]
  6%|▌         | 1460/24420 [36:13<9:23:32,  1.47s/it]
                                                      
{'loss': 0.2112, 'learning_rate': 5e-05, 'epoch': 0.3}

  6%|▌         | 1460/24420 [36:13<9:23:32,  1.47s/it]
  6%|▌         | 1461/24420 [36:15<9:23:24,  1.47s/it]
  6%|▌         | 1462/24420 [36:16<9:22:48,  1.47s/it]
  6%|▌         | 1463/24420 [36:18<9:23:04,  1.47s/it]
  6%|▌         | 1464/24420 [36:19<9:22:52,  1.47s/it]
  6%|▌         | 1465/24420 [36:21<9:23:14,  1.47s/it]
                                                      
{'loss': 0.2483, 'learning_rate': 5e-05, 'epoch': 0.3}

  6%|▌         | 1465/24420 [36:21<9:23:14,  1.47s/it]
  6%|▌         | 1466/24420 [36:22<9:22:58,  1.47s/it]
  6%|▌         | 1467/24420 [36:24<9:22:47,  1.47s/it]
  6%|▌         | 1468/24420 [36:25<9:23:17,  1.47s/it]
  6%|▌         | 1469/24420 [36:27<9:23:25,  1.47s/it]
  6%|▌         | 1470/24420 [36:28<9:23:43,  1.47s/it]
                                                      
{'loss': 0.2346, 'learning_rate': 5e-05, 'epoch': 0.3}

  6%|▌         | 1470/24420 [36:28<9:23:43,  1.47s/it]
  6%|▌         | 1471/24420 [36:30<9:25:12,  1.48s/it]
  6%|▌         | 1472/24420 [36:31<9:25:23,  1.48s/it]
  6%|▌         | 1473/24420 [36:32<9:25:49,  1.48s/it]
  6%|▌         | 1474/24420 [36:34<9:26:15,  1.48s/it]
  6%|▌         | 1475/24420 [36:35<9:25:29,  1.48s/it]
                                                      
{'loss': 0.2374, 'learning_rate': 5e-05, 'epoch': 0.3}

  6%|▌         | 1475/24420 [36:35<9:25:29,  1.48s/it]
  6%|▌         | 1476/24420 [36:37<9:24:14,  1.48s/it]
  6%|▌         | 1477/24420 [36:38<9:24:04,  1.48s/it]
  6%|▌         | 1478/24420 [36:40<9:23:37,  1.47s/it]
  6%|▌         | 1479/24420 [36:41<9:23:58,  1.48s/it]
  6%|▌         | 1480/24420 [36:43<9:24:01,  1.48s/it]
                                                      
{'loss': 0.2144, 'learning_rate': 5e-05, 'epoch': 0.3}

  6%|▌         | 1480/24420 [36:43<9:24:01,  1.48s/it]
  6%|▌         | 1481/24420 [36:44<9:23:36,  1.47s/it]
  6%|▌         | 1482/24420 [36:46<9:23:26,  1.47s/it]
  6%|▌         | 1483/24420 [36:47<9:23:54,  1.48s/it]
  6%|▌         | 1484/24420 [36:49<9:23:44,  1.47s/it]
  6%|▌         | 1485/24420 [36:50<9:23:46,  1.47s/it]
                                                      
{'loss': 0.2364, 'learning_rate': 5e-05, 'epoch': 0.3}

  6%|▌         | 1485/24420 [36:50<9:23:46,  1.47s/it]
  6%|▌         | 1486/24420 [36:52<9:24:03,  1.48s/it]
  6%|▌         | 1487/24420 [36:53<9:23:37,  1.47s/it]
  6%|▌         | 1488/24420 [36:55<9:38:19,  1.51s/it]
  6%|▌         | 1489/24420 [36:56<9:34:18,  1.50s/it]
  6%|▌         | 1490/24420 [36:58<9:31:42,  1.50s/it]
                                                      
{'loss': 0.2561, 'learning_rate': 5e-05, 'epoch': 0.31}

  6%|▌         | 1490/24420 [36:58<9:31:42,  1.50s/it]
  6%|▌         | 1491/24420 [36:59<9:30:10,  1.49s/it]
  6%|▌         | 1492/24420 [37:01<9:28:51,  1.49s/it]
  6%|▌         | 1493/24420 [37:02<9:28:24,  1.49s/it]
  6%|▌         | 1494/24420 [37:04<9:27:24,  1.48s/it]
  6%|▌         | 1495/24420 [37:05<9:26:53,  1.48s/it]
                                                      
{'loss': 0.2271, 'learning_rate': 5e-05, 'epoch': 0.31}

  6%|▌         | 1495/24420 [37:05<9:26:53,  1.48s/it]
  6%|▌         | 1496/24420 [37:07<9:26:07,  1.48s/it]
  6%|▌         | 1497/24420 [37:08<9:24:42,  1.48s/it]
  6%|▌         | 1498/24420 [37:10<9:24:07,  1.48s/it]
  6%|▌         | 1499/24420 [37:11<9:23:51,  1.48s/it]
  6%|▌         | 1500/24420 [37:12<9:24:27,  1.48s/it]
                                                      
{'loss': 0.2078, 'learning_rate': 5e-05, 'epoch': 0.31}

  6%|▌         | 1500/24420 [37:12<9:24:27,  1.48s/it]
  6%|▌         | 1501/24420 [37:14<9:25:29,  1.48s/it]
  6%|▌         | 1502/24420 [37:15<9:24:37,  1.48s/it]
  6%|▌         | 1503/24420 [37:17<9:24:01,  1.48s/it]
  6%|▌         | 1504/24420 [37:18<9:24:36,  1.48s/it]
  6%|▌         | 1505/24420 [37:20<9:24:35,  1.48s/it]
                                                      
{'loss': 0.2501, 'learning_rate': 5e-05, 'epoch': 0.31}

  6%|▌         | 1505/24420 [37:20<9:24:35,  1.48s/it]
  6%|▌         | 1506/24420 [37:21<9:25:03,  1.48s/it]
  6%|▌         | 1507/24420 [37:23<9:26:15,  1.48s/it]
  6%|▌         | 1508/24420 [37:24<9:26:54,  1.48s/it]
  6%|▌         | 1509/24420 [37:26<9:26:36,  1.48s/it]
  6%|▌         | 1510/24420 [37:27<9:27:48,  1.49s/it]
                                                      
{'loss': 0.2819, 'learning_rate': 5e-05, 'epoch': 0.31}

  6%|▌         | 1510/24420 [37:27<9:27:48,  1.49s/it]
  6%|▌         | 1511/24420 [37:29<9:28:36,  1.49s/it]
  6%|▌         | 1512/24420 [37:30<9:28:52,  1.49s/it]
  6%|▌         | 1513/24420 [37:32<9:27:37,  1.49s/it]
  6%|▌         | 1514/24420 [37:33<9:25:35,  1.48s/it]
  6%|▌         | 1515/24420 [37:35<9:25:17,  1.48s/it]
                                                      
{'loss': 0.2676, 'learning_rate': 5e-05, 'epoch': 0.31}

  6%|▌         | 1515/24420 [37:35<9:25:17,  1.48s/it]
  6%|▌         | 1516/24420 [37:36<9:25:03,  1.48s/it]
  6%|▌         | 1517/24420 [37:38<9:24:46,  1.48s/it]
  6%|▌         | 1518/24420 [37:39<9:24:20,  1.48s/it]
  6%|▌         | 1519/24420 [37:41<9:23:13,  1.48s/it]
  6%|▌         | 1520/24420 [37:42<9:23:30,  1.48s/it]
                                                      
{'loss': 0.2082, 'learning_rate': 5e-05, 'epoch': 0.31}

  6%|▌         | 1520/24420 [37:42<9:23:30,  1.48s/it]
  6%|▌         | 1521/24420 [37:44<9:23:23,  1.48s/it]
  6%|▌         | 1522/24420 [37:45<9:22:55,  1.48s/it]
  6%|▌         | 1523/24420 [37:47<9:23:24,  1.48s/it]
  6%|▌         | 1524/24420 [37:48<9:23:14,  1.48s/it]
  6%|▌         | 1525/24420 [37:49<9:23:12,  1.48s/it]
                                                      
{'loss': 0.2684, 'learning_rate': 5e-05, 'epoch': 0.31}

  6%|▌         | 1525/24420 [37:49<9:23:12,  1.48s/it]
  6%|▌         | 1526/24420 [37:51<9:23:45,  1.48s/it]
  6%|▋         | 1527/24420 [37:52<9:23:49,  1.48s/it]
  6%|▋         | 1528/24420 [37:54<9:23:02,  1.48s/it]
  6%|▋         | 1529/24420 [37:55<9:23:35,  1.48s/it]
  6%|▋         | 1530/24420 [37:57<9:22:45,  1.48s/it]
                                                      
{'loss': 0.2237, 'learning_rate': 5e-05, 'epoch': 0.31}

  6%|▋         | 1530/24420 [37:57<9:22:45,  1.48s/it]
  6%|▋         | 1531/24420 [37:58<9:23:05,  1.48s/it]
  6%|▋         | 1532/24420 [38:00<9:23:31,  1.48s/it]
  6%|▋         | 1533/24420 [38:01<9:22:44,  1.48s/it]
  6%|▋         | 1534/24420 [38:03<9:22:05,  1.47s/it]
  6%|▋         | 1535/24420 [38:04<9:22:38,  1.48s/it]
                                                      
{'loss': 0.2576, 'learning_rate': 5e-05, 'epoch': 0.31}

  6%|▋         | 1535/24420 [38:04<9:22:38,  1.48s/it]
  6%|▋         | 1536/24420 [38:06<9:22:17,  1.47s/it]
  6%|▋         | 1537/24420 [38:07<9:22:34,  1.48s/it]
  6%|▋         | 1538/24420 [38:09<9:22:32,  1.48s/it]
  6%|▋         | 1539/24420 [38:10<9:21:44,  1.47s/it]
  6%|▋         | 1540/24420 [38:12<9:21:20,  1.47s/it]
                                                      
{'loss': 0.2427, 'learning_rate': 5e-05, 'epoch': 0.32}

  6%|▋         | 1540/24420 [38:12<9:21:20,  1.47s/it]
  6%|▋         | 1541/24420 [38:13<9:21:41,  1.47s/it]
  6%|▋         | 1542/24420 [38:15<9:21:56,  1.47s/it]
  6%|▋         | 1543/24420 [38:16<9:22:37,  1.48s/it]
  6%|▋         | 1544/24420 [38:18<9:23:46,  1.48s/it]
  6%|▋         | 1545/24420 [38:19<9:23:19,  1.48s/it]
                                                      
{'loss': 0.2747, 'learning_rate': 5e-05, 'epoch': 0.32}

  6%|▋         | 1545/24420 [38:19<9:23:19,  1.48s/it]
  6%|▋         | 1546/24420 [38:20<9:22:59,  1.48s/it]
  6%|▋         | 1547/24420 [38:22<9:23:22,  1.48s/it]
  6%|▋         | 1548/24420 [38:23<9:22:45,  1.48s/it]
  6%|▋         | 1549/24420 [38:25<9:21:51,  1.47s/it]
  6%|▋         | 1550/24420 [38:26<9:23:36,  1.48s/it]
                                                      
{'loss': 0.2769, 'learning_rate': 5e-05, 'epoch': 0.32}

  6%|▋         | 1550/24420 [38:26<9:23:36,  1.48s/it]
  6%|▋         | 1551/24420 [38:28<9:24:41,  1.48s/it]
  6%|▋         | 1552/24420 [38:29<9:23:22,  1.48s/it]
  6%|▋         | 1553/24420 [38:31<9:23:28,  1.48s/it]
  6%|▋         | 1554/24420 [38:32<9:22:36,  1.48s/it]
  6%|▋         | 1555/24420 [38:34<9:22:27,  1.48s/it]
                                                      
{'loss': 0.2379, 'learning_rate': 5e-05, 'epoch': 0.32}

  6%|▋         | 1555/24420 [38:34<9:22:27,  1.48s/it]
  6%|▋         | 1556/24420 [38:35<9:23:13,  1.48s/it]
  6%|▋         | 1557/24420 [38:37<9:22:34,  1.48s/it]
  6%|▋         | 1558/24420 [38:38<9:23:00,  1.48s/it]
  6%|▋         | 1559/24420 [38:40<9:22:32,  1.48s/it]
  6%|▋         | 1560/24420 [38:41<9:40:48,  1.52s/it]
                                                      
{'loss': 0.2243, 'learning_rate': 5e-05, 'epoch': 0.32}

  6%|▋         | 1560/24420 [38:41<9:40:48,  1.52s/it]
  6%|▋         | 1561/24420 [38:43<9:36:25,  1.51s/it]
  6%|▋         | 1562/24420 [38:44<9:32:47,  1.50s/it]
  6%|▋         | 1563/24420 [38:46<9:29:27,  1.49s/it]
  6%|▋         | 1564/24420 [38:47<9:27:59,  1.49s/it]
  6%|▋         | 1565/24420 [38:49<9:27:59,  1.49s/it]
                                                      
{'loss': 0.2347, 'learning_rate': 5e-05, 'epoch': 0.32}

  6%|▋         | 1565/24420 [38:49<9:27:59,  1.49s/it]
  6%|▋         | 1566/24420 [38:50<9:27:15,  1.49s/it]
  6%|▋         | 1567/24420 [38:52<9:25:00,  1.48s/it]
  6%|▋         | 1568/24420 [38:53<9:25:54,  1.49s/it]
  6%|▋         | 1569/24420 [38:55<9:24:45,  1.48s/it]
  6%|▋         | 1570/24420 [38:56<9:23:43,  1.48s/it]
                                                      
{'loss': 0.2362, 'learning_rate': 5e-05, 'epoch': 0.32}

  6%|▋         | 1570/24420 [38:56<9:23:43,  1.48s/it]
  6%|▋         | 1571/24420 [38:58<9:23:56,  1.48s/it]
  6%|▋         | 1572/24420 [38:59<9:22:20,  1.48s/it]
  6%|▋         | 1573/24420 [39:01<9:22:22,  1.48s/it]
  6%|▋         | 1574/24420 [39:02<9:22:51,  1.48s/it]
  6%|▋         | 1575/24420 [39:04<9:22:56,  1.48s/it]
                                                      
{'loss': 0.2249, 'learning_rate': 5e-05, 'epoch': 0.32}

  6%|▋         | 1575/24420 [39:04<9:22:56,  1.48s/it]
  6%|▋         | 1576/24420 [39:05<9:22:16,  1.48s/it]
  6%|▋         | 1577/24420 [39:06<9:22:49,  1.48s/it]
  6%|▋         | 1578/24420 [39:08<9:21:50,  1.48s/it]
  6%|▋         | 1579/24420 [39:09<9:23:11,  1.48s/it]
  6%|▋         | 1580/24420 [39:11<9:22:08,  1.48s/it]
                                                      
{'loss': 0.2561, 'learning_rate': 5e-05, 'epoch': 0.32}

  6%|▋         | 1580/24420 [39:11<9:22:08,  1.48s/it]
  6%|▋         | 1581/24420 [39:12<9:22:54,  1.48s/it]
  6%|▋         | 1582/24420 [39:14<9:22:31,  1.48s/it]
  6%|▋         | 1583/24420 [39:15<9:21:49,  1.48s/it]
  6%|▋         | 1584/24420 [39:17<9:21:11,  1.47s/it]
  6%|▋         | 1585/24420 [39:18<9:20:42,  1.47s/it]
                                                      
{'loss': 0.2148, 'learning_rate': 5e-05, 'epoch': 0.32}

  6%|▋         | 1585/24420 [39:18<9:20:42,  1.47s/it]
  6%|▋         | 1586/24420 [39:20<9:20:26,  1.47s/it]
  6%|▋         | 1587/24420 [39:21<9:20:18,  1.47s/it]
  7%|▋         | 1588/24420 [39:23<9:19:41,  1.47s/it]
  7%|▋         | 1589/24420 [39:24<9:19:42,  1.47s/it]
  7%|▋         | 1590/24420 [39:26<9:20:32,  1.47s/it]
                                                      
{'loss': 0.2157, 'learning_rate': 5e-05, 'epoch': 0.33}

  7%|▋         | 1590/24420 [39:26<9:20:32,  1.47s/it]
  7%|▋         | 1591/24420 [39:27<9:21:21,  1.48s/it]
  7%|▋         | 1592/24420 [39:29<9:20:57,  1.47s/it]
  7%|▋         | 1593/24420 [39:30<9:21:14,  1.48s/it]
  7%|▋         | 1594/24420 [39:32<9:20:39,  1.47s/it]
  7%|▋         | 1595/24420 [39:33<9:20:20,  1.47s/it]
                                                      
{'loss': 0.264, 'learning_rate': 5e-05, 'epoch': 0.33}

  7%|▋         | 1595/24420 [39:33<9:20:20,  1.47s/it]
  7%|▋         | 1596/24420 [39:34<9:20:17,  1.47s/it]
  7%|▋         | 1597/24420 [39:36<9:19:48,  1.47s/it]
  7%|▋         | 1598/24420 [39:37<9:20:39,  1.47s/it]
  7%|▋         | 1599/24420 [39:39<9:20:22,  1.47s/it]
  7%|▋         | 1600/24420 [39:40<9:21:17,  1.48s/it]
                                                      
{'loss': 0.2187, 'learning_rate': 5e-05, 'epoch': 0.33}

  7%|▋         | 1600/24420 [39:40<9:21:17,  1.48s/it]
  7%|▋         | 1601/24420 [39:45<15:30:50,  2.45s/it]
  7%|▋         | 1602/24420 [39:47<13:40:00,  2.16s/it]
  7%|▋         | 1603/24420 [39:48<12:21:23,  1.95s/it]
  7%|▋         | 1604/24420 [39:50<11:27:30,  1.81s/it]
  7%|▋         | 1605/24420 [39:51<10:49:15,  1.71s/it]
                                                       
{'loss': 0.2457, 'learning_rate': 5e-05, 'epoch': 0.33}

  7%|▋         | 1605/24420 [39:51<10:49:15,  1.71s/it]
  7%|▋         | 1606/24420 [39:52<10:22:08,  1.64s/it]
  7%|▋         | 1607/24420 [39:54<10:03:44,  1.59s/it]
  7%|▋         | 1608/24420 [39:55<9:51:13,  1.56s/it] 
  7%|▋         | 1609/24420 [39:57<9:42:16,  1.53s/it]
  7%|▋         | 1610/24420 [39:58<9:35:07,  1.51s/it]
                                                      
{'loss': 0.2215, 'learning_rate': 5e-05, 'epoch': 0.33}

  7%|▋         | 1610/24420 [39:58<9:35:07,  1.51s/it]
  7%|▋         | 1611/24420 [40:00<9:30:11,  1.50s/it]
  7%|▋         | 1612/24420 [40:01<9:27:07,  1.49s/it]
  7%|▋         | 1613/24420 [40:03<9:25:11,  1.49s/it]
  7%|▋         | 1614/24420 [40:04<9:23:46,  1.48s/it]
  7%|▋         | 1615/24420 [40:06<9:22:21,  1.48s/it]
                                                      
{'loss': 0.2351, 'learning_rate': 5e-05, 'epoch': 0.33}

  7%|▋         | 1615/24420 [40:06<9:22:21,  1.48s/it]
  7%|▋         | 1616/24420 [40:07<9:22:18,  1.48s/it]
  7%|▋         | 1617/24420 [40:09<9:21:25,  1.48s/it]
  7%|▋         | 1618/24420 [40:10<9:20:58,  1.48s/it]
  7%|▋         | 1619/24420 [40:12<9:20:42,  1.48s/it]
  7%|▋         | 1620/24420 [40:13<9:19:49,  1.47s/it]
                                                      
{'loss': 0.2594, 'learning_rate': 5e-05, 'epoch': 0.33}

  7%|▋         | 1620/24420 [40:13<9:19:49,  1.47s/it]
  7%|▋         | 1621/24420 [40:15<9:20:17,  1.47s/it]
  7%|▋         | 1622/24420 [40:16<9:19:07,  1.47s/it]
  7%|▋         | 1623/24420 [40:17<9:19:10,  1.47s/it]
  7%|▋         | 1624/24420 [40:19<9:19:54,  1.47s/it]
  7%|▋         | 1625/24420 [40:20<9:19:46,  1.47s/it]
                                                      
{'loss': 0.2423, 'learning_rate': 5e-05, 'epoch': 0.33}

  7%|▋         | 1625/24420 [40:20<9:19:46,  1.47s/it]
  7%|▋         | 1626/24420 [40:22<9:19:49,  1.47s/it]
  7%|▋         | 1627/24420 [40:23<9:20:04,  1.47s/it]
  7%|▋         | 1628/24420 [40:25<9:19:06,  1.47s/it]
  7%|▋         | 1629/24420 [40:26<9:19:46,  1.47s/it]
  7%|▋         | 1630/24420 [40:28<9:19:39,  1.47s/it]
                                                      
{'loss': 0.2544, 'learning_rate': 5e-05, 'epoch': 0.33}

  7%|▋         | 1630/24420 [40:28<9:19:39,  1.47s/it]
  7%|▋         | 1631/24420 [40:29<9:20:36,  1.48s/it]
  7%|▋         | 1632/24420 [40:31<9:20:06,  1.47s/it]
  7%|▋         | 1633/24420 [40:32<9:19:20,  1.47s/it]
  7%|▋         | 1634/24420 [40:34<9:19:03,  1.47s/it]
  7%|▋         | 1635/24420 [40:35<9:19:25,  1.47s/it]
                                                      
{'loss': 0.2286, 'learning_rate': 5e-05, 'epoch': 0.33}

  7%|▋         | 1635/24420 [40:35<9:19:25,  1.47s/it]
  7%|▋         | 1636/24420 [40:37<9:20:20,  1.48s/it]
  7%|▋         | 1637/24420 [40:38<9:20:51,  1.48s/it]
  7%|▋         | 1638/24420 [40:40<9:20:20,  1.48s/it]
  7%|▋         | 1639/24420 [40:41<9:19:35,  1.47s/it]
  7%|▋         | 1640/24420 [40:43<9:18:37,  1.47s/it]
                                                      
{'loss': 0.2312, 'learning_rate': 5e-05, 'epoch': 0.34}

  7%|▋         | 1640/24420 [40:43<9:18:37,  1.47s/it]
  7%|▋         | 1641/24420 [40:44<9:19:59,  1.48s/it]
  7%|▋         | 1642/24420 [40:46<9:19:24,  1.47s/it]
  7%|▋         | 1643/24420 [40:47<9:19:28,  1.47s/it]
  7%|▋         | 1644/24420 [40:48<9:19:04,  1.47s/it]
  7%|▋         | 1645/24420 [40:50<9:19:09,  1.47s/it]
                                                      
{'loss': 0.2458, 'learning_rate': 5e-05, 'epoch': 0.34}

  7%|▋         | 1645/24420 [40:50<9:19:09,  1.47s/it]
  7%|▋         | 1646/24420 [40:51<9:19:32,  1.47s/it]
  7%|▋         | 1647/24420 [40:53<9:19:42,  1.47s/it]
  7%|▋         | 1648/24420 [40:54<9:20:07,  1.48s/it]
  7%|▋         | 1649/24420 [40:56<9:20:40,  1.48s/it]
  7%|▋         | 1650/24420 [40:57<9:20:45,  1.48s/it]
                                                      
{'loss': 0.2286, 'learning_rate': 5e-05, 'epoch': 0.34}

  7%|▋         | 1650/24420 [40:57<9:20:45,  1.48s/it]
  7%|▋         | 1651/24420 [40:59<9:21:45,  1.48s/it]
  7%|▋         | 1652/24420 [41:00<9:21:43,  1.48s/it]
  7%|▋         | 1653/24420 [41:02<9:20:40,  1.48s/it]
  7%|▋         | 1654/24420 [41:03<9:19:36,  1.47s/it]
  7%|▋         | 1655/24420 [41:05<9:18:28,  1.47s/it]
                                                      
{'loss': 0.2359, 'learning_rate': 5e-05, 'epoch': 0.34}

  7%|▋         | 1655/24420 [41:05<9:18:28,  1.47s/it]
  7%|▋         | 1656/24420 [41:06<9:19:39,  1.48s/it]
  7%|▋         | 1657/24420 [41:08<9:18:59,  1.47s/it]
  7%|▋         | 1658/24420 [41:09<9:18:12,  1.47s/it]
  7%|▋         | 1659/24420 [41:11<9:19:05,  1.47s/it]
  7%|▋         | 1660/24420 [41:12<9:19:29,  1.47s/it]
                                                      
{'loss': 0.2255, 'learning_rate': 5e-05, 'epoch': 0.34}

  7%|▋         | 1660/24420 [41:12<9:19:29,  1.47s/it]
  7%|▋         | 1661/24420 [41:14<9:19:47,  1.48s/it]
  7%|▋         | 1662/24420 [41:15<9:20:01,  1.48s/it]
  7%|▋         | 1663/24420 [41:16<9:20:03,  1.48s/it]
  7%|▋         | 1664/24420 [41:18<9:19:45,  1.48s/it]
  7%|▋         | 1665/24420 [41:19<9:19:52,  1.48s/it]
                                                      
{'loss': 0.2265, 'learning_rate': 5e-05, 'epoch': 0.34}

  7%|▋         | 1665/24420 [41:19<9:19:52,  1.48s/it]
  7%|▋         | 1666/24420 [41:21<9:21:05,  1.48s/it]
  7%|▋         | 1667/24420 [41:22<9:21:16,  1.48s/it]
  7%|▋         | 1668/24420 [41:24<9:21:42,  1.48s/it]
  7%|▋         | 1669/24420 [41:25<9:21:58,  1.48s/it]
  7%|▋         | 1670/24420 [41:27<9:21:40,  1.48s/it]
                                                      
{'loss': 0.264, 'learning_rate': 5e-05, 'epoch': 0.34}

  7%|▋         | 1670/24420 [41:27<9:21:40,  1.48s/it]
  7%|▋         | 1671/24420 [41:28<9:20:43,  1.48s/it]
  7%|▋         | 1672/24420 [41:30<9:20:34,  1.48s/it]
  7%|▋         | 1673/24420 [41:31<9:19:49,  1.48s/it]
  7%|▋         | 1674/24420 [41:33<9:19:30,  1.48s/it]
  7%|▋         | 1675/24420 [41:34<9:18:51,  1.47s/it]
                                                      
{'loss': 0.2411, 'learning_rate': 5e-05, 'epoch': 0.34}

  7%|▋         | 1675/24420 [41:34<9:18:51,  1.47s/it]
  7%|▋         | 1676/24420 [41:36<9:18:56,  1.47s/it]
  7%|▋         | 1677/24420 [41:37<9:20:12,  1.48s/it]
  7%|▋         | 1678/24420 [41:39<9:19:53,  1.48s/it]
  7%|▋         | 1679/24420 [41:40<9:20:22,  1.48s/it]
  7%|▋         | 1680/24420 [41:42<9:21:17,  1.48s/it]
                                                      
{'loss': 0.2415, 'learning_rate': 5e-05, 'epoch': 0.34}

  7%|▋         | 1680/24420 [41:42<9:21:17,  1.48s/it]
  7%|▋         | 1681/24420 [41:43<9:20:42,  1.48s/it]
  7%|▋         | 1682/24420 [41:45<9:19:43,  1.48s/it]
  7%|▋         | 1683/24420 [41:46<9:18:56,  1.47s/it]
  7%|▋         | 1684/24420 [41:48<9:19:09,  1.48s/it]
  7%|▋         | 1685/24420 [41:49<9:18:22,  1.47s/it]
                                                      
{'loss': 0.2374, 'learning_rate': 5e-05, 'epoch': 0.35}

  7%|▋         | 1685/24420 [41:49<9:18:22,  1.47s/it]
  7%|▋         | 1686/24420 [41:50<9:19:11,  1.48s/it]
  7%|▋         | 1687/24420 [41:52<9:19:44,  1.48s/it]
  7%|▋         | 1688/24420 [41:53<9:20:30,  1.48s/it]
  7%|▋         | 1689/24420 [41:55<9:20:38,  1.48s/it]
  7%|▋         | 1690/24420 [41:56<9:20:30,  1.48s/it]
                                                      
{'loss': 0.2229, 'learning_rate': 5e-05, 'epoch': 0.35}

  7%|▋         | 1690/24420 [41:56<9:20:30,  1.48s/it]
  7%|▋         | 1691/24420 [41:58<9:20:44,  1.48s/it]
  7%|▋         | 1692/24420 [41:59<9:20:59,  1.48s/it]
  7%|▋         | 1693/24420 [42:01<9:19:52,  1.48s/it]
  7%|▋         | 1694/24420 [42:02<9:18:40,  1.48s/it]
  7%|▋         | 1695/24420 [42:04<9:18:51,  1.48s/it]
                                                      
{'loss': 0.2325, 'learning_rate': 5e-05, 'epoch': 0.35}

  7%|▋         | 1695/24420 [42:04<9:18:51,  1.48s/it]
  7%|▋         | 1696/24420 [42:05<9:19:05,  1.48s/it]
  7%|▋         | 1697/24420 [42:07<9:18:39,  1.48s/it]
  7%|▋         | 1698/24420 [42:08<9:19:34,  1.48s/it]
  7%|▋         | 1699/24420 [42:10<9:20:18,  1.48s/it]
  7%|▋         | 1700/24420 [42:11<9:19:43,  1.48s/it]
                                                      
{'loss': 0.2499, 'learning_rate': 5e-05, 'epoch': 0.35}

  7%|▋         | 1700/24420 [42:11<9:19:43,  1.48s/it]
  7%|▋         | 1701/24420 [42:13<9:19:27,  1.48s/it]
  7%|▋         | 1702/24420 [42:14<9:20:10,  1.48s/it]
  7%|▋         | 1703/24420 [42:16<9:19:14,  1.48s/it]
  7%|▋         | 1704/24420 [42:17<9:18:48,  1.48s/it]
  7%|▋         | 1705/24420 [42:19<9:19:06,  1.48s/it]
                                                      
{'loss': 0.2193, 'learning_rate': 5e-05, 'epoch': 0.35}

  7%|▋         | 1705/24420 [42:19<9:19:06,  1.48s/it]
  7%|▋         | 1706/24420 [42:20<9:19:05,  1.48s/it]
  7%|▋         | 1707/24420 [42:22<9:18:22,  1.48s/it]
  7%|▋         | 1708/24420 [42:23<9:18:02,  1.47s/it]
  7%|▋         | 1709/24420 [42:24<9:18:08,  1.47s/it]
  7%|▋         | 1710/24420 [42:26<9:17:25,  1.47s/it]
                                                      
{'loss': 0.2371, 'learning_rate': 5e-05, 'epoch': 0.35}

  7%|▋         | 1710/24420 [42:26<9:17:25,  1.47s/it]
  7%|▋         | 1711/24420 [42:27<9:18:21,  1.48s/it]
  7%|▋         | 1712/24420 [42:29<9:18:13,  1.47s/it]
  7%|▋         | 1713/24420 [42:30<9:17:57,  1.47s/it]
  7%|▋         | 1714/24420 [42:32<9:17:25,  1.47s/it]
  7%|▋         | 1715/24420 [42:33<9:18:08,  1.47s/it]
                                                      
{'loss': 0.2188, 'learning_rate': 5e-05, 'epoch': 0.35}

  7%|▋         | 1715/24420 [42:33<9:18:08,  1.47s/it]
  7%|▋         | 1716/24420 [42:35<9:18:38,  1.48s/it]
  7%|▋         | 1717/24420 [42:36<9:17:43,  1.47s/it]
  7%|▋         | 1718/24420 [42:38<9:16:56,  1.47s/it]
  7%|▋         | 1719/24420 [42:39<9:17:25,  1.47s/it]
  7%|▋         | 1720/24420 [42:41<9:18:03,  1.48s/it]
                                                      
{'loss': 0.25, 'learning_rate': 5e-05, 'epoch': 0.35}

  7%|▋         | 1720/24420 [42:41<9:18:03,  1.48s/it]
  7%|▋         | 1721/24420 [42:42<9:18:00,  1.47s/it]
  7%|▋         | 1722/24420 [42:44<9:17:46,  1.47s/it]
  7%|▋         | 1723/24420 [42:45<9:17:03,  1.47s/it]
  7%|▋         | 1724/24420 [42:47<9:17:04,  1.47s/it]
  7%|▋         | 1725/24420 [42:48<9:16:49,  1.47s/it]
                                                      
{'loss': 0.2415, 'learning_rate': 5e-05, 'epoch': 0.35}

  7%|▋         | 1725/24420 [42:48<9:16:49,  1.47s/it]
  7%|▋         | 1726/24420 [42:50<9:18:22,  1.48s/it]
  7%|▋         | 1727/24420 [42:51<9:18:24,  1.48s/it]
  7%|▋         | 1728/24420 [42:52<9:17:46,  1.47s/it]
  7%|▋         | 1729/24420 [42:54<9:18:06,  1.48s/it]
  7%|▋         | 1730/24420 [42:55<9:17:20,  1.47s/it]
                                                      
{'loss': 0.2637, 'learning_rate': 5e-05, 'epoch': 0.35}

  7%|▋         | 1730/24420 [42:55<9:17:20,  1.47s/it]
  7%|▋         | 1731/24420 [42:57<9:17:27,  1.47s/it]
  7%|▋         | 1732/24420 [42:58<9:17:02,  1.47s/it]
  7%|▋         | 1733/24420 [43:00<9:16:59,  1.47s/it]
  7%|▋         | 1734/24420 [43:01<9:17:52,  1.48s/it]
  7%|▋         | 1735/24420 [43:03<9:18:17,  1.48s/it]
                                                      
{'loss': 0.254, 'learning_rate': 5e-05, 'epoch': 0.36}

  7%|▋         | 1735/24420 [43:03<9:18:17,  1.48s/it]
  7%|▋         | 1736/24420 [43:04<9:18:00,  1.48s/it]
  7%|▋         | 1737/24420 [43:06<9:18:08,  1.48s/it]
  7%|▋         | 1738/24420 [43:07<9:18:19,  1.48s/it]
  7%|▋         | 1739/24420 [43:09<9:17:12,  1.47s/it]
  7%|▋         | 1740/24420 [43:10<9:18:51,  1.48s/it]
                                                      
{'loss': 0.2594, 'learning_rate': 5e-05, 'epoch': 0.36}

  7%|▋         | 1740/24420 [43:10<9:18:51,  1.48s/it]
  7%|▋         | 1741/24420 [43:12<9:17:54,  1.48s/it]
  7%|▋         | 1742/24420 [43:13<9:17:06,  1.47s/it]
  7%|▋         | 1743/24420 [43:15<9:16:20,  1.47s/it]
  7%|▋         | 1744/24420 [43:16<9:17:18,  1.47s/it]
  7%|▋         | 1745/24420 [43:18<9:16:39,  1.47s/it]
                                                      
{'loss': 0.2366, 'learning_rate': 5e-05, 'epoch': 0.36}

  7%|▋         | 1745/24420 [43:18<9:16:39,  1.47s/it]
  7%|▋         | 1746/24420 [43:19<9:16:28,  1.47s/it]
  7%|▋         | 1747/24420 [43:20<9:16:31,  1.47s/it]
  7%|▋         | 1748/24420 [43:22<9:16:48,  1.47s/it]
  7%|▋         | 1749/24420 [43:23<9:16:48,  1.47s/it]
  7%|▋         | 1750/24420 [43:25<9:16:06,  1.47s/it]
                                                      
{'loss': 0.2384, 'learning_rate': 5e-05, 'epoch': 0.36}

  7%|▋         | 1750/24420 [43:25<9:16:06,  1.47s/it]
  7%|▋         | 1751/24420 [43:26<9:16:25,  1.47s/it]
  7%|▋         | 1752/24420 [43:28<9:17:44,  1.48s/it]
  7%|▋         | 1753/24420 [43:29<9:18:14,  1.48s/it]
  7%|▋         | 1754/24420 [43:31<9:17:28,  1.48s/it]
  7%|▋         | 1755/24420 [43:32<9:17:29,  1.48s/it]
                                                      
{'loss': 0.2516, 'learning_rate': 5e-05, 'epoch': 0.36}

  7%|▋         | 1755/24420 [43:32<9:17:29,  1.48s/it]
  7%|▋         | 1756/24420 [43:34<9:19:01,  1.48s/it]
  7%|▋         | 1757/24420 [43:35<9:19:55,  1.48s/it]
  7%|▋         | 1758/24420 [43:37<9:18:30,  1.48s/it]
  7%|▋         | 1759/24420 [43:38<9:23:36,  1.49s/it]
  7%|▋         | 1760/24420 [43:40<9:21:21,  1.49s/it]
                                                      
{'loss': 0.2556, 'learning_rate': 5e-05, 'epoch': 0.36}

  7%|▋         | 1760/24420 [43:40<9:21:21,  1.49s/it]
  7%|▋         | 1761/24420 [43:41<9:20:32,  1.48s/it]
  7%|▋         | 1762/24420 [43:43<9:23:10,  1.49s/it]
  7%|▋         | 1763/24420 [43:44<9:21:00,  1.49s/it]
  7%|▋         | 1764/24420 [43:46<9:20:22,  1.48s/it]
  7%|▋         | 1765/24420 [43:47<9:18:40,  1.48s/it]
                                                      
{'loss': 0.2742, 'learning_rate': 5e-05, 'epoch': 0.36}

  7%|▋         | 1765/24420 [43:47<9:18:40,  1.48s/it]
  7%|▋         | 1766/24420 [43:49<9:19:21,  1.48s/it]
  7%|▋         | 1767/24420 [43:50<9:19:14,  1.48s/it]
  7%|▋         | 1768/24420 [43:52<9:18:19,  1.48s/it]
  7%|▋         | 1769/24420 [43:53<9:17:59,  1.48s/it]
  7%|▋         | 1770/24420 [43:55<9:17:55,  1.48s/it]
                                                      
{'loss': 0.2246, 'learning_rate': 5e-05, 'epoch': 0.36}

  7%|▋         | 1770/24420 [43:55<9:17:55,  1.48s/it]
  7%|▋         | 1771/24420 [43:56<9:17:19,  1.48s/it]
  7%|▋         | 1772/24420 [43:57<9:16:14,  1.47s/it]
  7%|▋         | 1773/24420 [43:59<9:16:28,  1.47s/it]
  7%|▋         | 1774/24420 [44:00<9:16:36,  1.47s/it]
  7%|▋         | 1775/24420 [44:02<9:16:30,  1.47s/it]
                                                      
{'loss': 0.2086, 'learning_rate': 5e-05, 'epoch': 0.36}

  7%|▋         | 1775/24420 [44:02<9:16:30,  1.47s/it]
  7%|▋         | 1776/24420 [44:03<9:16:08,  1.47s/it]
  7%|▋         | 1777/24420 [44:05<9:16:03,  1.47s/it]
  7%|▋         | 1778/24420 [44:06<9:16:07,  1.47s/it]
  7%|▋         | 1779/24420 [44:08<9:16:13,  1.47s/it]
  7%|▋         | 1780/24420 [44:09<9:16:23,  1.47s/it]
                                                      
{'loss': 0.2172, 'learning_rate': 5e-05, 'epoch': 0.36}

  7%|▋         | 1780/24420 [44:09<9:16:23,  1.47s/it]
  7%|▋         | 1781/24420 [44:11<9:16:53,  1.48s/it]
  7%|▋         | 1782/24420 [44:12<9:17:22,  1.48s/it]
  7%|▋         | 1783/24420 [44:14<9:19:56,  1.48s/it]
  7%|▋         | 1784/24420 [44:15<9:17:32,  1.48s/it]
  7%|▋         | 1785/24420 [44:17<9:16:51,  1.48s/it]
                                                      
{'loss': 0.206, 'learning_rate': 5e-05, 'epoch': 0.37}

  7%|▋         | 1785/24420 [44:17<9:16:51,  1.48s/it]
  7%|▋         | 1786/24420 [44:18<9:16:36,  1.48s/it]
  7%|▋         | 1787/24420 [44:20<9:16:07,  1.47s/it]
  7%|▋         | 1788/24420 [44:21<9:15:52,  1.47s/it]
  7%|▋         | 1789/24420 [44:23<9:15:34,  1.47s/it]
  7%|▋         | 1790/24420 [44:24<9:15:21,  1.47s/it]
                                                      
{'loss': 0.2153, 'learning_rate': 5e-05, 'epoch': 0.37}

  7%|▋         | 1790/24420 [44:24<9:15:21,  1.47s/it]
  7%|▋         | 1791/24420 [44:25<9:15:48,  1.47s/it]
  7%|▋         | 1792/24420 [44:27<9:15:18,  1.47s/it]
  7%|▋         | 1793/24420 [44:28<9:14:51,  1.47s/it]
  7%|▋         | 1794/24420 [44:30<9:15:24,  1.47s/it]
  7%|▋         | 1795/24420 [44:31<9:16:04,  1.47s/it]
                                                      
{'loss': 0.2361, 'learning_rate': 5e-05, 'epoch': 0.37}

  7%|▋         | 1795/24420 [44:31<9:16:04,  1.47s/it]
  7%|▋         | 1796/24420 [44:33<9:17:17,  1.48s/it]
  7%|▋         | 1797/24420 [44:34<9:16:41,  1.48s/it]
  7%|▋         | 1798/24420 [44:36<9:16:04,  1.47s/it]
  7%|▋         | 1799/24420 [44:37<9:15:05,  1.47s/it]
  7%|▋         | 1800/24420 [44:39<9:14:48,  1.47s/it]
                                                      
{'loss': 0.2589, 'learning_rate': 5e-05, 'epoch': 0.37}

  7%|▋         | 1800/24420 [44:39<9:14:48,  1.47s/it]
  7%|▋         | 1801/24420 [44:43<14:58:28,  2.38s/it]
  7%|▋         | 1802/24420 [44:45<13:15:06,  2.11s/it]
  7%|▋         | 1803/24420 [44:46<12:03:18,  1.92s/it]
  7%|▋         | 1804/24420 [44:48<11:12:19,  1.78s/it]
  7%|▋         | 1805/24420 [44:49<10:36:24,  1.69s/it]
                                                       
{'loss': 0.2523, 'learning_rate': 5e-05, 'epoch': 0.37}

  7%|▋         | 1805/24420 [44:49<10:36:24,  1.69s/it]
  7%|▋         | 1806/24420 [44:51<10:12:07,  1.62s/it]
  7%|▋         | 1807/24420 [44:52<9:54:51,  1.58s/it] 
  7%|▋         | 1808/24420 [44:54<9:42:15,  1.54s/it]
  7%|▋         | 1809/24420 [44:55<9:33:18,  1.52s/it]
  7%|▋         | 1810/24420 [44:56<9:26:52,  1.50s/it]
                                                      
{'loss': 0.2082, 'learning_rate': 5e-05, 'epoch': 0.37}

  7%|▋         | 1810/24420 [44:56<9:26:52,  1.50s/it]
  7%|▋         | 1811/24420 [44:58<9:22:30,  1.49s/it]
  7%|▋         | 1812/24420 [44:59<9:19:47,  1.49s/it]
  7%|▋         | 1813/24420 [45:01<9:17:28,  1.48s/it]
  7%|▋         | 1814/24420 [45:02<9:16:32,  1.48s/it]
  7%|▋         | 1815/24420 [45:04<9:15:36,  1.47s/it]
                                                      
{'loss': 0.2021, 'learning_rate': 5e-05, 'epoch': 0.37}

  7%|▋         | 1815/24420 [45:04<9:15:36,  1.47s/it]
  7%|▋         | 1816/24420 [45:05<9:15:21,  1.47s/it]
  7%|▋         | 1817/24420 [45:07<9:14:58,  1.47s/it]
  7%|▋         | 1818/24420 [45:08<9:15:39,  1.48s/it]
  7%|▋         | 1819/24420 [45:10<9:15:30,  1.47s/it]
  7%|▋         | 1820/24420 [45:11<9:16:39,  1.48s/it]
                                                      
{'loss': 0.2466, 'learning_rate': 5e-05, 'epoch': 0.37}

  7%|▋         | 1820/24420 [45:11<9:16:39,  1.48s/it]
  7%|▋         | 1821/24420 [45:13<9:17:14,  1.48s/it]
  7%|▋         | 1822/24420 [45:14<9:15:55,  1.48s/it]
  7%|▋         | 1823/24420 [45:16<9:15:12,  1.47s/it]
  7%|▋         | 1824/24420 [45:17<9:15:38,  1.48s/it]
  7%|▋         | 1825/24420 [45:19<9:14:56,  1.47s/it]
                                                      
{'loss': 0.1922, 'learning_rate': 5e-05, 'epoch': 0.37}

  7%|▋         | 1825/24420 [45:19<9:14:56,  1.47s/it]
  7%|▋         | 1826/24420 [45:20<9:14:33,  1.47s/it]
  7%|▋         | 1827/24420 [45:22<9:14:50,  1.47s/it]
  7%|▋         | 1828/24420 [45:23<9:14:08,  1.47s/it]
  7%|▋         | 1829/24420 [45:24<9:13:32,  1.47s/it]
  7%|▋         | 1830/24420 [45:26<9:13:48,  1.47s/it]
                                                      
{'loss': 0.2358, 'learning_rate': 5e-05, 'epoch': 0.37}

  7%|▋         | 1830/24420 [45:26<9:13:48,  1.47s/it]
  7%|▋         | 1831/24420 [45:27<9:14:06,  1.47s/it]
  8%|▊         | 1832/24420 [45:29<9:14:03,  1.47s/it]
  8%|▊         | 1833/24420 [45:30<9:13:20,  1.47s/it]
  8%|▊         | 1834/24420 [45:32<9:14:07,  1.47s/it]
  8%|▊         | 1835/24420 [45:33<9:14:43,  1.47s/it]
                                                      
{'loss': 0.2286, 'learning_rate': 5e-05, 'epoch': 0.38}

  8%|▊         | 1835/24420 [45:33<9:14:43,  1.47s/it]
  8%|▊         | 1836/24420 [45:35<9:15:16,  1.48s/it]
  8%|▊         | 1837/24420 [45:36<9:15:45,  1.48s/it]
  8%|▊         | 1838/24420 [45:38<9:15:10,  1.48s/it]
  8%|▊         | 1839/24420 [45:39<9:14:44,  1.47s/it]
  8%|▊         | 1840/24420 [45:41<9:13:51,  1.47s/it]
                                                      
{'loss': 0.224, 'learning_rate': 5e-05, 'epoch': 0.38}

  8%|▊         | 1840/24420 [45:41<9:13:51,  1.47s/it]
  8%|▊         | 1841/24420 [45:42<9:13:32,  1.47s/it]
  8%|▊         | 1842/24420 [45:44<9:13:36,  1.47s/it]
  8%|▊         | 1843/24420 [45:45<9:14:09,  1.47s/it]
  8%|▊         | 1844/24420 [45:47<9:14:55,  1.47s/it]
  8%|▊         | 1845/24420 [45:48<9:13:59,  1.47s/it]
                                                      
{'loss': 0.224, 'learning_rate': 5e-05, 'epoch': 0.38}

  8%|▊         | 1845/24420 [45:48<9:13:59,  1.47s/it]
  8%|▊         | 1846/24420 [45:50<9:15:02,  1.48s/it]
  8%|▊         | 1847/24420 [45:51<9:14:41,  1.47s/it]
  8%|▊         | 1848/24420 [45:52<9:13:59,  1.47s/it]
  8%|▊         | 1849/24420 [45:54<9:14:22,  1.47s/it]
  8%|▊         | 1850/24420 [45:55<9:14:06,  1.47s/it]
                                                      
{'loss': 0.2461, 'learning_rate': 5e-05, 'epoch': 0.38}

  8%|▊         | 1850/24420 [45:55<9:14:06,  1.47s/it]
  8%|▊         | 1851/24420 [45:57<9:14:41,  1.47s/it]
  8%|▊         | 1852/24420 [45:58<9:14:04,  1.47s/it]
  8%|▊         | 1853/24420 [46:00<9:13:53,  1.47s/it]
  8%|▊         | 1854/24420 [46:01<9:14:17,  1.47s/it]
  8%|▊         | 1855/24420 [46:03<9:13:13,  1.47s/it]
                                                      
{'loss': 0.221, 'learning_rate': 5e-05, 'epoch': 0.38}

  8%|▊         | 1855/24420 [46:03<9:13:13,  1.47s/it]
  8%|▊         | 1856/24420 [46:04<9:13:53,  1.47s/it]
  8%|▊         | 1857/24420 [46:06<9:13:53,  1.47s/it]
  8%|▊         | 1858/24420 [46:07<9:14:17,  1.47s/it]
  8%|▊         | 1859/24420 [46:09<9:13:55,  1.47s/it]
  8%|▊         | 1860/24420 [46:10<9:13:47,  1.47s/it]
                                                      
{'loss': 0.262, 'learning_rate': 5e-05, 'epoch': 0.38}

  8%|▊         | 1860/24420 [46:10<9:13:47,  1.47s/it]
  8%|▊         | 1861/24420 [46:12<9:14:30,  1.47s/it]
  8%|▊         | 1862/24420 [46:13<9:14:14,  1.47s/it]
  8%|▊         | 1863/24420 [46:15<9:13:30,  1.47s/it]
  8%|▊         | 1864/24420 [46:16<9:12:52,  1.47s/it]
  8%|▊         | 1865/24420 [46:17<9:12:19,  1.47s/it]
                                                      
{'loss': 0.2015, 'learning_rate': 5e-05, 'epoch': 0.38}

  8%|▊         | 1865/24420 [46:17<9:12:19,  1.47s/it]
  8%|▊         | 1866/24420 [46:19<9:12:40,  1.47s/it]
  8%|▊         | 1867/24420 [46:20<9:12:48,  1.47s/it]
  8%|▊         | 1868/24420 [46:22<9:13:18,  1.47s/it]
  8%|▊         | 1869/24420 [46:23<9:13:57,  1.47s/it]
  8%|▊         | 1870/24420 [46:25<9:12:56,  1.47s/it]
                                                      
{'loss': 0.2159, 'learning_rate': 5e-05, 'epoch': 0.38}

  8%|▊         | 1870/24420 [46:25<9:12:56,  1.47s/it]
  8%|▊         | 1871/24420 [46:26<9:13:55,  1.47s/it]
  8%|▊         | 1872/24420 [46:28<9:13:44,  1.47s/it]
  8%|▊         | 1873/24420 [46:29<9:14:13,  1.47s/it]
  8%|▊         | 1874/24420 [46:31<9:13:48,  1.47s/it]
  8%|▊         | 1875/24420 [46:32<9:14:19,  1.48s/it]
                                                      
{'loss': 0.2412, 'learning_rate': 5e-05, 'epoch': 0.38}

  8%|▊         | 1875/24420 [46:32<9:14:19,  1.48s/it]
  8%|▊         | 1876/24420 [46:34<9:14:44,  1.48s/it]
  8%|▊         | 1877/24420 [46:35<9:14:43,  1.48s/it]
  8%|▊         | 1878/24420 [46:37<9:13:21,  1.47s/it]
  8%|▊         | 1879/24420 [46:38<9:13:04,  1.47s/it]
  8%|▊         | 1880/24420 [46:40<9:12:57,  1.47s/it]
                                                      
{'loss': 0.2015, 'learning_rate': 5e-05, 'epoch': 0.38}

  8%|▊         | 1880/24420 [46:40<9:12:57,  1.47s/it]
  8%|▊         | 1881/24420 [46:41<9:12:51,  1.47s/it]
  8%|▊         | 1882/24420 [46:43<9:12:28,  1.47s/it]
  8%|▊         | 1883/24420 [46:44<9:13:16,  1.47s/it]
  8%|▊         | 1884/24420 [46:45<9:13:48,  1.47s/it]
  8%|▊         | 1885/24420 [46:47<9:13:11,  1.47s/it]
                                                      
{'loss': 0.2328, 'learning_rate': 5e-05, 'epoch': 0.39}

  8%|▊         | 1885/24420 [46:47<9:13:11,  1.47s/it]
  8%|▊         | 1886/24420 [46:48<9:12:56,  1.47s/it]
  8%|▊         | 1887/24420 [46:50<9:13:45,  1.47s/it]
  8%|▊         | 1888/24420 [46:51<9:13:20,  1.47s/it]
  8%|▊         | 1889/24420 [46:53<9:13:10,  1.47s/it]
  8%|▊         | 1890/24420 [46:54<9:13:01,  1.47s/it]
                                                      
{'loss': 0.2384, 'learning_rate': 5e-05, 'epoch': 0.39}

  8%|▊         | 1890/24420 [46:54<9:13:01,  1.47s/it]
  8%|▊         | 1891/24420 [46:56<9:13:20,  1.47s/it]
  8%|▊         | 1892/24420 [46:57<9:12:52,  1.47s/it]
  8%|▊         | 1893/24420 [46:59<9:12:56,  1.47s/it]
  8%|▊         | 1894/24420 [47:00<9:13:00,  1.47s/it]
  8%|▊         | 1895/24420 [47:02<9:13:54,  1.48s/it]
                                                      
{'loss': 0.2342, 'learning_rate': 5e-05, 'epoch': 0.39}

  8%|▊         | 1895/24420 [47:02<9:13:54,  1.48s/it]
  8%|▊         | 1896/24420 [47:03<9:13:19,  1.47s/it]
  8%|▊         | 1897/24420 [47:05<9:12:20,  1.47s/it]
  8%|▊         | 1898/24420 [47:06<9:12:36,  1.47s/it]
  8%|▊         | 1899/24420 [47:08<9:12:20,  1.47s/it]
  8%|▊         | 1900/24420 [47:09<9:13:00,  1.47s/it]
                                                      
{'loss': 0.2272, 'learning_rate': 5e-05, 'epoch': 0.39}

  8%|▊         | 1900/24420 [47:09<9:13:00,  1.47s/it]
  8%|▊         | 1901/24420 [47:11<9:12:52,  1.47s/it]
  8%|▊         | 1902/24420 [47:12<9:12:54,  1.47s/it]
  8%|▊         | 1903/24420 [47:13<9:13:20,  1.47s/it]
  8%|▊         | 1904/24420 [47:15<9:12:43,  1.47s/it]
  8%|▊         | 1905/24420 [47:16<9:13:14,  1.47s/it]
                                                      
{'loss': 0.241, 'learning_rate': 5e-05, 'epoch': 0.39}

  8%|▊         | 1905/24420 [47:16<9:13:14,  1.47s/it]
  8%|▊         | 1906/24420 [47:18<9:13:22,  1.47s/it]
  8%|▊         | 1907/24420 [47:19<9:13:43,  1.48s/it]
  8%|▊         | 1908/24420 [47:21<9:13:30,  1.48s/it]
  8%|▊         | 1909/24420 [47:22<9:13:06,  1.47s/it]
  8%|▊         | 1910/24420 [47:24<9:12:07,  1.47s/it]
                                                      
{'loss': 0.2538, 'learning_rate': 5e-05, 'epoch': 0.39}

  8%|▊         | 1910/24420 [47:24<9:12:07,  1.47s/it]
  8%|▊         | 1911/24420 [47:25<9:12:09,  1.47s/it]
  8%|▊         | 1912/24420 [47:27<9:11:24,  1.47s/it]
  8%|▊         | 1913/24420 [47:28<9:11:07,  1.47s/it]
  8%|▊         | 1914/24420 [47:30<9:12:50,  1.47s/it]
  8%|▊         | 1915/24420 [47:31<9:14:33,  1.48s/it]
                                                      
{'loss': 0.2412, 'learning_rate': 5e-05, 'epoch': 0.39}

  8%|▊         | 1915/24420 [47:31<9:14:33,  1.48s/it]
  8%|▊         | 1916/24420 [47:33<9:14:34,  1.48s/it]
  8%|▊         | 1917/24420 [47:34<9:14:20,  1.48s/it]
  8%|▊         | 1918/24420 [47:36<9:13:42,  1.48s/it]
  8%|▊         | 1919/24420 [47:37<9:13:27,  1.48s/it]
  8%|▊         | 1920/24420 [47:39<9:14:05,  1.48s/it]
                                                      
{'loss': 0.2372, 'learning_rate': 5e-05, 'epoch': 0.39}

  8%|▊         | 1920/24420 [47:39<9:14:05,  1.48s/it]
  8%|▊         | 1921/24420 [47:40<9:13:39,  1.48s/it]
  8%|▊         | 1922/24420 [47:41<9:12:57,  1.47s/it]
  8%|▊         | 1923/24420 [47:43<9:12:46,  1.47s/it]
  8%|▊         | 1924/24420 [47:44<9:12:32,  1.47s/it]
  8%|▊         | 1925/24420 [47:46<9:11:47,  1.47s/it]
                                                      
{'loss': 0.2155, 'learning_rate': 5e-05, 'epoch': 0.39}

  8%|▊         | 1925/24420 [47:46<9:11:47,  1.47s/it]
  8%|▊         | 1926/24420 [47:47<9:11:43,  1.47s/it]
  8%|▊         | 1927/24420 [47:49<9:11:14,  1.47s/it]
  8%|▊         | 1928/24420 [47:50<9:11:04,  1.47s/it]
  8%|▊         | 1929/24420 [47:52<9:11:30,  1.47s/it]
  8%|▊         | 1930/24420 [47:53<9:12:07,  1.47s/it]
                                                      
{'loss': 0.2107, 'learning_rate': 5e-05, 'epoch': 0.4}

  8%|▊         | 1930/24420 [47:53<9:12:07,  1.47s/it]
  8%|▊         | 1931/24420 [47:55<9:11:57,  1.47s/it]
  8%|▊         | 1932/24420 [47:56<9:11:49,  1.47s/it]
  8%|▊         | 1933/24420 [47:58<9:11:56,  1.47s/it]
  8%|▊         | 1934/24420 [47:59<9:12:35,  1.47s/it]
  8%|▊         | 1935/24420 [48:01<9:12:12,  1.47s/it]
                                                      
{'loss': 0.218, 'learning_rate': 5e-05, 'epoch': 0.4}

  8%|▊         | 1935/24420 [48:01<9:12:12,  1.47s/it]
  8%|▊         | 1936/24420 [48:02<9:11:27,  1.47s/it]
  8%|▊         | 1937/24420 [48:04<9:12:32,  1.47s/it]
  8%|▊         | 1938/24420 [48:05<9:12:43,  1.48s/it]
  8%|▊         | 1939/24420 [48:07<9:13:08,  1.48s/it]
  8%|▊         | 1940/24420 [48:08<9:12:16,  1.47s/it]
                                                      
{'loss': 0.2291, 'learning_rate': 5e-05, 'epoch': 0.4}

  8%|▊         | 1940/24420 [48:08<9:12:16,  1.47s/it]
  8%|▊         | 1941/24420 [48:09<9:12:33,  1.47s/it]
  8%|▊         | 1942/24420 [48:11<9:12:24,  1.47s/it]
  8%|▊         | 1943/24420 [48:12<9:12:22,  1.47s/it]
  8%|▊         | 1944/24420 [48:14<9:11:47,  1.47s/it]
  8%|▊         | 1945/24420 [48:15<9:10:59,  1.47s/it]
                                                      
{'loss': 0.2221, 'learning_rate': 5e-05, 'epoch': 0.4}

  8%|▊         | 1945/24420 [48:15<9:10:59,  1.47s/it]
  8%|▊         | 1946/24420 [48:17<9:10:16,  1.47s/it]
  8%|▊         | 1947/24420 [48:18<9:09:37,  1.47s/it]
  8%|▊         | 1948/24420 [48:20<9:09:23,  1.47s/it]
  8%|▊         | 1949/24420 [48:21<9:10:27,  1.47s/it]
  8%|▊         | 1950/24420 [48:23<9:10:31,  1.47s/it]
                                                      
{'loss': 0.2072, 'learning_rate': 5e-05, 'epoch': 0.4}

  8%|▊         | 1950/24420 [48:23<9:10:31,  1.47s/it]
  8%|▊         | 1951/24420 [48:24<9:10:33,  1.47s/it]
  8%|▊         | 1952/24420 [48:26<9:10:03,  1.47s/it]
  8%|▊         | 1953/24420 [48:27<9:10:40,  1.47s/it]
  8%|▊         | 1954/24420 [48:29<9:10:49,  1.47s/it]
  8%|▊         | 1955/24420 [48:30<9:12:13,  1.47s/it]
                                                      
{'loss': 0.2286, 'learning_rate': 5e-05, 'epoch': 0.4}

  8%|▊         | 1955/24420 [48:30<9:12:13,  1.47s/it]
  8%|▊         | 1956/24420 [48:32<9:12:46,  1.48s/it]
  8%|▊         | 1957/24420 [48:33<9:12:31,  1.48s/it]
  8%|▊         | 1958/24420 [48:35<9:11:59,  1.47s/it]
  8%|▊         | 1959/24420 [48:36<9:11:01,  1.47s/it]
  8%|▊         | 1960/24420 [48:37<9:10:31,  1.47s/it]
                                                      
{'loss': 0.2217, 'learning_rate': 5e-05, 'epoch': 0.4}

  8%|▊         | 1960/24420 [48:37<9:10:31,  1.47s/it]
  8%|▊         | 1961/24420 [48:39<9:09:54,  1.47s/it]
  8%|▊         | 1962/24420 [48:40<9:10:00,  1.47s/it]
  8%|▊         | 1963/24420 [48:42<9:11:25,  1.47s/it]
  8%|▊         | 1964/24420 [48:43<9:11:06,  1.47s/it]
  8%|▊         | 1965/24420 [48:45<9:11:32,  1.47s/it]
                                                      
{'loss': 0.2102, 'learning_rate': 5e-05, 'epoch': 0.4}

  8%|▊         | 1965/24420 [48:45<9:11:32,  1.47s/it]
  8%|▊         | 1966/24420 [48:46<9:12:07,  1.48s/it]
  8%|▊         | 1967/24420 [48:48<9:12:46,  1.48s/it]
  8%|▊         | 1968/24420 [48:49<9:12:11,  1.48s/it]
  8%|▊         | 1969/24420 [48:51<9:13:10,  1.48s/it]
  8%|▊         | 1970/24420 [48:52<9:12:26,  1.48s/it]
                                                      
{'loss': 0.2508, 'learning_rate': 5e-05, 'epoch': 0.4}

  8%|▊         | 1970/24420 [48:52<9:12:26,  1.48s/it]
  8%|▊         | 1971/24420 [48:54<9:11:44,  1.47s/it]
  8%|▊         | 1972/24420 [48:55<9:11:21,  1.47s/it]
  8%|▊         | 1973/24420 [48:57<9:10:37,  1.47s/it]
  8%|▊         | 1974/24420 [48:58<9:09:58,  1.47s/it]
  8%|▊         | 1975/24420 [49:00<9:11:06,  1.47s/it]
                                                      
{'loss': 0.2274, 'learning_rate': 5e-05, 'epoch': 0.4}

  8%|▊         | 1975/24420 [49:00<9:11:06,  1.47s/it]
  8%|▊         | 1976/24420 [49:01<9:11:19,  1.47s/it]
  8%|▊         | 1977/24420 [49:02<9:10:51,  1.47s/it]
  8%|▊         | 1978/24420 [49:04<9:10:51,  1.47s/it]
  8%|▊         | 1979/24420 [49:05<9:11:14,  1.47s/it]
  8%|▊         | 1980/24420 [49:07<9:11:10,  1.47s/it]
                                                      
{'loss': 0.1963, 'learning_rate': 5e-05, 'epoch': 0.41}

  8%|▊         | 1980/24420 [49:07<9:11:10,  1.47s/it]
  8%|▊         | 1981/24420 [49:08<9:11:14,  1.47s/it]
  8%|▊         | 1982/24420 [49:10<9:10:52,  1.47s/it]
  8%|▊         | 1983/24420 [49:11<9:10:00,  1.47s/it]
  8%|▊         | 1984/24420 [49:13<9:09:34,  1.47s/it]
  8%|▊         | 1985/24420 [49:14<9:09:25,  1.47s/it]
                                                      
{'loss': 0.2006, 'learning_rate': 5e-05, 'epoch': 0.41}

  8%|▊         | 1985/24420 [49:14<9:09:25,  1.47s/it]
  8%|▊         | 1986/24420 [49:16<9:09:32,  1.47s/it]
  8%|▊         | 1987/24420 [49:17<9:09:33,  1.47s/it]
  8%|▊         | 1988/24420 [49:19<9:09:54,  1.47s/it]
  8%|▊         | 1989/24420 [49:20<9:09:34,  1.47s/it]
  8%|▊         | 1990/24420 [49:22<9:10:11,  1.47s/it]
                                                      
{'loss': 0.2203, 'learning_rate': 5e-05, 'epoch': 0.41}

  8%|▊         | 1990/24420 [49:22<9:10:11,  1.47s/it]
  8%|▊         | 1991/24420 [49:23<9:09:31,  1.47s/it]
  8%|▊         | 1992/24420 [49:25<9:09:49,  1.47s/it]
  8%|▊         | 1993/24420 [49:26<9:10:19,  1.47s/it]
  8%|▊         | 1994/24420 [49:28<9:10:30,  1.47s/it]
  8%|▊         | 1995/24420 [49:29<9:10:51,  1.47s/it]
                                                      
{'loss': 0.2323, 'learning_rate': 5e-05, 'epoch': 0.41}

  8%|▊         | 1995/24420 [49:29<9:10:51,  1.47s/it]
  8%|▊         | 1996/24420 [49:30<9:10:53,  1.47s/it]
  8%|▊         | 1997/24420 [49:32<9:10:41,  1.47s/it]
  8%|▊         | 1998/24420 [49:33<9:11:17,  1.48s/it]
  8%|▊         | 1999/24420 [49:35<9:10:36,  1.47s/it]
  8%|▊         | 2000/24420 [49:36<9:09:36,  1.47s/it]
                                                      
{'loss': 0.2549, 'learning_rate': 5e-05, 'epoch': 0.41}

  8%|▊         | 2000/24420 [49:36<9:09:36,  1.47s/it]
  8%|▊         | 2001/24420 [49:41<14:50:56,  2.38s/it]
  8%|▊         | 2002/24420 [49:42<13:08:38,  2.11s/it]
  8%|▊         | 2003/24420 [49:44<11:58:28,  1.92s/it]
  8%|▊         | 2004/24420 [49:45<11:07:48,  1.79s/it]
  8%|▊         | 2005/24420 [49:47<10:32:12,  1.69s/it]
                                                       
{'loss': 0.2149, 'learning_rate': 5e-05, 'epoch': 0.41}

  8%|▊         | 2005/24420 [49:47<10:32:12,  1.69s/it]
  8%|▊         | 2006/24420 [49:48<10:07:32,  1.63s/it]
  8%|▊         | 2007/24420 [49:50<9:50:27,  1.58s/it] 
  8%|▊         | 2008/24420 [49:51<9:39:24,  1.55s/it]
  8%|▊         | 2009/24420 [49:53<9:30:42,  1.53s/it]
  8%|▊         | 2010/24420 [49:54<9:24:07,  1.51s/it]
                                                      
{'loss': 0.2191, 'learning_rate': 5e-05, 'epoch': 0.41}

  8%|▊         | 2010/24420 [49:54<9:24:07,  1.51s/it]
  8%|▊         | 2011/24420 [49:56<9:19:00,  1.50s/it]
  8%|▊         | 2012/24420 [49:57<9:16:20,  1.49s/it]
  8%|▊         | 2013/24420 [49:59<9:15:17,  1.49s/it]
  8%|▊         | 2014/24420 [50:00<9:13:56,  1.48s/it]
  8%|▊         | 2015/24420 [50:01<9:12:30,  1.48s/it]
                                                      
{'loss': 0.23, 'learning_rate': 5e-05, 'epoch': 0.41}

  8%|▊         | 2015/24420 [50:01<9:12:30,  1.48s/it]
  8%|▊         | 2016/24420 [50:03<9:12:01,  1.48s/it]
  8%|▊         | 2017/24420 [50:04<9:11:45,  1.48s/it]
  8%|▊         | 2018/24420 [50:06<9:10:41,  1.47s/it]
  8%|▊         | 2019/24420 [50:07<9:10:35,  1.47s/it]
  8%|▊         | 2020/24420 [50:09<9:10:19,  1.47s/it]
                                                      
{'loss': 0.2197, 'learning_rate': 5e-05, 'epoch': 0.41}

  8%|▊         | 2020/24420 [50:09<9:10:19,  1.47s/it]
  8%|▊         | 2021/24420 [50:10<9:12:04,  1.48s/it]
  8%|▊         | 2022/24420 [50:12<9:11:49,  1.48s/it]
  8%|▊         | 2023/24420 [50:13<9:10:49,  1.48s/it]
  8%|▊         | 2024/24420 [50:15<9:10:52,  1.48s/it]
  8%|▊         | 2025/24420 [50:16<9:10:11,  1.47s/it]
                                                      
{'loss': 0.2092, 'learning_rate': 5e-05, 'epoch': 0.41}

  8%|▊         | 2025/24420 [50:16<9:10:11,  1.47s/it]
  8%|▊         | 2026/24420 [50:18<9:09:56,  1.47s/it]
  8%|▊         | 2027/24420 [50:19<9:09:35,  1.47s/it]
  8%|▊         | 2028/24420 [50:21<9:11:53,  1.48s/it]
  8%|▊         | 2029/24420 [50:22<9:10:25,  1.47s/it]
  8%|▊         | 2030/24420 [50:24<9:10:06,  1.47s/it]
                                                      
{'loss': 0.21, 'learning_rate': 5e-05, 'epoch': 0.42}

  8%|▊         | 2030/24420 [50:24<9:10:06,  1.47s/it]
  8%|▊         | 2031/24420 [50:25<9:09:57,  1.47s/it]
  8%|▊         | 2032/24420 [50:27<9:08:49,  1.47s/it]
  8%|▊         | 2033/24420 [50:28<9:09:59,  1.47s/it]
  8%|▊         | 2034/24420 [50:30<9:09:17,  1.47s/it]
  8%|▊         | 2035/24420 [50:31<9:09:10,  1.47s/it]
                                                      
{'loss': 0.2282, 'learning_rate': 5e-05, 'epoch': 0.42}

  8%|▊         | 2035/24420 [50:31<9:09:10,  1.47s/it]
  8%|▊         | 2036/24420 [50:32<9:08:40,  1.47s/it]
  8%|▊         | 2037/24420 [50:34<9:09:07,  1.47s/it]
  8%|▊         | 2038/24420 [50:35<9:09:52,  1.47s/it]
  8%|▊         | 2039/24420 [50:37<9:09:39,  1.47s/it]
  8%|▊         | 2040/24420 [50:38<9:08:56,  1.47s/it]
                                                      
{'loss': 0.2276, 'learning_rate': 5e-05, 'epoch': 0.42}

  8%|▊         | 2040/24420 [50:38<9:08:56,  1.47s/it]
  8%|▊         | 2041/24420 [50:40<9:10:20,  1.48s/it]
  8%|▊         | 2042/24420 [50:41<9:11:09,  1.48s/it]
  8%|▊         | 2043/24420 [50:43<9:10:26,  1.48s/it]
  8%|▊         | 2044/24420 [50:44<9:09:43,  1.47s/it]
  8%|▊         | 2045/24420 [50:46<9:09:49,  1.47s/it]
                                                      
{'loss': 0.2129, 'learning_rate': 5e-05, 'epoch': 0.42}

  8%|▊         | 2045/24420 [50:46<9:09:49,  1.47s/it]
  8%|▊         | 2046/24420 [50:47<9:09:39,  1.47s/it]
  8%|▊         | 2047/24420 [50:49<9:08:47,  1.47s/it]
  8%|▊         | 2048/24420 [50:50<9:09:05,  1.47s/it]
  8%|▊         | 2049/24420 [50:52<9:09:38,  1.47s/it]
  8%|▊         | 2050/24420 [50:53<9:08:53,  1.47s/it]
                                                      
{'loss': 0.199, 'learning_rate': 5e-05, 'epoch': 0.42}

  8%|▊         | 2050/24420 [50:53<9:08:53,  1.47s/it]
  8%|▊         | 2051/24420 [50:55<9:08:41,  1.47s/it]
  8%|▊         | 2052/24420 [50:56<9:08:51,  1.47s/it]
  8%|▊         | 2053/24420 [50:57<9:08:49,  1.47s/it]
  8%|▊         | 2054/24420 [50:59<9:09:15,  1.47s/it]
  8%|▊         | 2055/24420 [51:00<9:09:46,  1.47s/it]
                                                      
{'loss': 0.2545, 'learning_rate': 5e-05, 'epoch': 0.42}

  8%|▊         | 2055/24420 [51:00<9:09:46,  1.47s/it]
  8%|▊         | 2056/24420 [51:02<9:10:02,  1.48s/it]
  8%|▊         | 2057/24420 [51:03<9:10:03,  1.48s/it]
  8%|▊         | 2058/24420 [51:05<9:09:14,  1.47s/it]
  8%|▊         | 2059/24420 [51:06<9:08:58,  1.47s/it]
  8%|▊         | 2060/24420 [51:08<9:09:37,  1.47s/it]
                                                      
{'loss': 0.226, 'learning_rate': 5e-05, 'epoch': 0.42}

  8%|▊         | 2060/24420 [51:08<9:09:37,  1.47s/it]
  8%|▊         | 2061/24420 [51:09<9:10:20,  1.48s/it]
  8%|▊         | 2062/24420 [51:11<9:09:12,  1.47s/it]
  8%|▊         | 2063/24420 [51:12<9:08:30,  1.47s/it]
  8%|▊         | 2064/24420 [51:14<9:08:04,  1.47s/it]
  8%|▊         | 2065/24420 [51:15<9:08:21,  1.47s/it]
                                                      
{'loss': 0.2383, 'learning_rate': 5e-05, 'epoch': 0.42}

  8%|▊         | 2065/24420 [51:15<9:08:21,  1.47s/it]
  8%|▊         | 2066/24420 [51:17<9:09:24,  1.47s/it]
  8%|▊         | 2067/24420 [51:18<9:08:40,  1.47s/it]
  8%|▊         | 2068/24420 [51:20<9:08:37,  1.47s/it]
  8%|▊         | 2069/24420 [51:21<9:09:02,  1.47s/it]
  8%|▊         | 2070/24420 [51:23<9:08:51,  1.47s/it]
                                                      
{'loss': 0.2271, 'learning_rate': 5e-05, 'epoch': 0.42}

  8%|▊         | 2070/24420 [51:23<9:08:51,  1.47s/it]
  8%|▊         | 2071/24420 [51:24<9:08:45,  1.47s/it]
  8%|▊         | 2072/24420 [51:26<9:11:51,  1.48s/it]
  8%|▊         | 2073/24420 [51:27<9:11:34,  1.48s/it]
  8%|▊         | 2074/24420 [51:28<9:10:36,  1.48s/it]
  8%|▊         | 2075/24420 [51:30<9:09:03,  1.47s/it]
                                                      
{'loss': 0.2322, 'learning_rate': 5e-05, 'epoch': 0.42}

  8%|▊         | 2075/24420 [51:30<9:09:03,  1.47s/it]
  9%|▊         | 2076/24420 [51:31<9:08:55,  1.47s/it]
  9%|▊         | 2077/24420 [51:33<9:09:59,  1.48s/it]
  9%|▊         | 2078/24420 [51:34<9:09:32,  1.48s/it]
  9%|▊         | 2079/24420 [51:36<9:08:13,  1.47s/it]
  9%|▊         | 2080/24420 [51:37<9:08:34,  1.47s/it]
                                                      
{'loss': 0.2309, 'learning_rate': 5e-05, 'epoch': 0.43}

  9%|▊         | 2080/24420 [51:37<9:08:34,  1.47s/it]
  9%|▊         | 2081/24420 [51:39<9:14:23,  1.49s/it]
  9%|▊         | 2082/24420 [51:40<9:11:54,  1.48s/it]
  9%|▊         | 2083/24420 [51:42<9:10:45,  1.48s/it]
  9%|▊         | 2084/24420 [51:43<9:09:18,  1.48s/it]
  9%|▊         | 2085/24420 [51:45<9:08:11,  1.47s/it]
                                                      
{'loss': 0.2177, 'learning_rate': 5e-05, 'epoch': 0.43}

  9%|▊         | 2085/24420 [51:45<9:08:11,  1.47s/it]
  9%|▊         | 2086/24420 [51:46<9:07:53,  1.47s/it]
  9%|▊         | 2087/24420 [51:48<9:07:57,  1.47s/it]
  9%|▊         | 2088/24420 [51:49<9:07:29,  1.47s/it]
  9%|▊         | 2089/24420 [51:51<9:06:54,  1.47s/it]
  9%|▊         | 2090/24420 [51:52<9:07:09,  1.47s/it]
                                                      
{'loss': 0.2384, 'learning_rate': 5e-05, 'epoch': 0.43}

  9%|▊         | 2090/24420 [51:52<9:07:09,  1.47s/it]
  9%|▊         | 2091/24420 [51:54<9:07:23,  1.47s/it]
  9%|▊         | 2092/24420 [51:55<9:07:39,  1.47s/it]
  9%|▊         | 2093/24420 [51:56<9:07:26,  1.47s/it]
  9%|▊         | 2094/24420 [51:58<9:06:53,  1.47s/it]
  9%|▊         | 2095/24420 [51:59<9:07:43,  1.47s/it]
                                                      
{'loss': 0.2221, 'learning_rate': 5e-05, 'epoch': 0.43}

  9%|▊         | 2095/24420 [51:59<9:07:43,  1.47s/it]
  9%|▊         | 2096/24420 [52:01<9:07:34,  1.47s/it]
  9%|▊         | 2097/24420 [52:02<9:07:36,  1.47s/it]
  9%|▊         | 2098/24420 [52:04<9:08:23,  1.47s/it]
  9%|▊         | 2099/24420 [52:05<9:07:15,  1.47s/it]
  9%|▊         | 2100/24420 [52:07<9:08:43,  1.48s/it]
                                                      
{'loss': 0.2085, 'learning_rate': 5e-05, 'epoch': 0.43}

  9%|▊         | 2100/24420 [52:07<9:08:43,  1.48s/it]
  9%|▊         | 2101/24420 [52:08<9:08:39,  1.47s/it]
  9%|▊         | 2102/24420 [52:10<9:08:22,  1.47s/it]
  9%|▊         | 2103/24420 [52:11<9:08:27,  1.47s/it]
  9%|▊         | 2104/24420 [52:13<9:08:53,  1.48s/it]
  9%|▊         | 2105/24420 [52:14<9:08:58,  1.48s/it]
                                                      
{'loss': 0.2133, 'learning_rate': 5e-05, 'epoch': 0.43}

  9%|▊         | 2105/24420 [52:14<9:08:58,  1.48s/it]
  9%|▊         | 2106/24420 [52:16<9:08:44,  1.48s/it]
  9%|▊         | 2107/24420 [52:17<9:08:53,  1.48s/it]
  9%|▊         | 2108/24420 [52:19<9:13:31,  1.49s/it]
  9%|▊         | 2109/24420 [52:20<9:11:33,  1.48s/it]
  9%|▊         | 2110/24420 [52:22<9:10:30,  1.48s/it]
                                                      
{'loss': 0.1812, 'learning_rate': 5e-05, 'epoch': 0.43}

  9%|▊         | 2110/24420 [52:22<9:10:30,  1.48s/it]
  9%|▊         | 2111/24420 [52:23<9:09:49,  1.48s/it]
  9%|▊         | 2112/24420 [52:25<9:08:54,  1.48s/it]
  9%|▊         | 2113/24420 [52:26<9:11:27,  1.48s/it]
  9%|▊         | 2114/24420 [52:27<9:09:46,  1.48s/it]
  9%|▊         | 2115/24420 [52:29<9:08:08,  1.47s/it]
                                                      
{'loss': 0.2018, 'learning_rate': 5e-05, 'epoch': 0.43}

  9%|▊         | 2115/24420 [52:29<9:08:08,  1.47s/it]
  9%|▊         | 2116/24420 [52:30<9:07:13,  1.47s/it]
  9%|▊         | 2117/24420 [52:32<9:06:16,  1.47s/it]
  9%|▊         | 2118/24420 [52:33<9:05:54,  1.47s/it]
  9%|▊         | 2119/24420 [52:35<9:06:52,  1.47s/it]
  9%|▊         | 2120/24420 [52:36<9:06:25,  1.47s/it]
                                                      
{'loss': 0.1914, 'learning_rate': 5e-05, 'epoch': 0.43}

  9%|▊         | 2120/24420 [52:36<9:06:25,  1.47s/it]
  9%|▊         | 2121/24420 [52:38<9:06:53,  1.47s/it]
  9%|▊         | 2122/24420 [52:39<9:06:24,  1.47s/it]
  9%|▊         | 2123/24420 [52:41<9:06:20,  1.47s/it]
  9%|▊         | 2124/24420 [52:42<9:07:14,  1.47s/it]
  9%|▊         | 2125/24420 [52:44<9:08:29,  1.48s/it]
                                                      
{'loss': 0.2328, 'learning_rate': 5e-05, 'epoch': 0.44}

  9%|▊         | 2125/24420 [52:44<9:08:29,  1.48s/it]
  9%|▊         | 2126/24420 [52:45<9:08:25,  1.48s/it]
  9%|▊         | 2127/24420 [52:47<9:08:18,  1.48s/it]
  9%|▊         | 2128/24420 [52:48<9:07:12,  1.47s/it]
  9%|▊         | 2129/24420 [52:50<9:06:57,  1.47s/it]
  9%|▊         | 2130/24420 [52:51<9:06:38,  1.47s/it]
                                                      
{'loss': 0.21, 'learning_rate': 5e-05, 'epoch': 0.44}

  9%|▊         | 2130/24420 [52:51<9:06:38,  1.47s/it]
  9%|▊         | 2131/24420 [52:53<9:07:11,  1.47s/it]
  9%|▊         | 2132/24420 [52:54<9:06:29,  1.47s/it]
  9%|▊         | 2133/24420 [52:55<9:06:56,  1.47s/it]
  9%|▊         | 2134/24420 [52:57<9:07:10,  1.47s/it]
  9%|▊         | 2135/24420 [52:58<9:06:33,  1.47s/it]
                                                      
{'loss': 0.2125, 'learning_rate': 5e-05, 'epoch': 0.44}

  9%|▊         | 2135/24420 [52:58<9:06:33,  1.47s/it]
  9%|▊         | 2136/24420 [53:00<9:07:52,  1.48s/it]
  9%|▉         | 2137/24420 [53:01<9:07:38,  1.47s/it]
  9%|▉         | 2138/24420 [53:03<9:07:19,  1.47s/it]
  9%|▉         | 2139/24420 [53:04<9:07:30,  1.47s/it]
  9%|▉         | 2140/24420 [53:06<9:07:51,  1.48s/it]
                                                      
{'loss': 0.2499, 'learning_rate': 5e-05, 'epoch': 0.44}

  9%|▉         | 2140/24420 [53:06<9:07:51,  1.48s/it]
  9%|▉         | 2141/24420 [53:07<9:07:45,  1.48s/it]
  9%|▉         | 2142/24420 [53:09<9:07:01,  1.47s/it]
  9%|▉         | 2143/24420 [53:10<9:06:18,  1.47s/it]
  9%|▉         | 2144/24420 [53:12<9:06:40,  1.47s/it]
  9%|▉         | 2145/24420 [53:13<9:08:13,  1.48s/it]
                                                      
{'loss': 0.231, 'learning_rate': 5e-05, 'epoch': 0.44}

  9%|▉         | 2145/24420 [53:13<9:08:13,  1.48s/it]
  9%|▉         | 2146/24420 [53:15<9:08:18,  1.48s/it]
  9%|▉         | 2147/24420 [53:16<9:07:58,  1.48s/it]
  9%|▉         | 2148/24420 [53:18<9:07:21,  1.47s/it]
  9%|▉         | 2149/24420 [53:19<9:07:07,  1.47s/it]
  9%|▉         | 2150/24420 [53:21<9:07:01,  1.47s/it]
                                                      
{'loss': 0.1945, 'learning_rate': 5e-05, 'epoch': 0.44}

  9%|▉         | 2150/24420 [53:21<9:07:01,  1.47s/it]
  9%|▉         | 2151/24420 [53:22<9:08:09,  1.48s/it]
  9%|▉         | 2152/24420 [53:23<9:08:31,  1.48s/it]
  9%|▉         | 2153/24420 [53:25<9:08:16,  1.48s/it]
  9%|▉         | 2154/24420 [53:26<9:08:21,  1.48s/it]
  9%|▉         | 2155/24420 [53:28<9:08:54,  1.48s/it]
                                                      
{'loss': 0.194, 'learning_rate': 5e-05, 'epoch': 0.44}

  9%|▉         | 2155/24420 [53:28<9:08:54,  1.48s/it]
  9%|▉         | 2156/24420 [53:29<9:09:46,  1.48s/it]
  9%|▉         | 2157/24420 [53:31<9:09:01,  1.48s/it]
  9%|▉         | 2158/24420 [53:32<9:08:22,  1.48s/it]
  9%|▉         | 2159/24420 [53:34<9:08:17,  1.48s/it]
  9%|▉         | 2160/24420 [53:35<9:08:21,  1.48s/it]
                                                      
{'loss': 0.2733, 'learning_rate': 5e-05, 'epoch': 0.44}

  9%|▉         | 2160/24420 [53:35<9:08:21,  1.48s/it]
  9%|▉         | 2161/24420 [53:37<9:07:54,  1.48s/it]
  9%|▉         | 2162/24420 [53:38<9:08:18,  1.48s/it]
  9%|▉         | 2163/24420 [53:40<9:07:32,  1.48s/it]
  9%|▉         | 2164/24420 [53:41<9:07:01,  1.47s/it]
  9%|▉         | 2165/24420 [53:43<9:06:28,  1.47s/it]
                                                      
{'loss': 0.2614, 'learning_rate': 5e-05, 'epoch': 0.44}

  9%|▉         | 2165/24420 [53:43<9:06:28,  1.47s/it]
  9%|▉         | 2166/24420 [53:44<9:06:55,  1.47s/it]
  9%|▉         | 2167/24420 [53:46<9:06:04,  1.47s/it]
  9%|▉         | 2168/24420 [53:47<9:06:17,  1.47s/it]
  9%|▉         | 2169/24420 [53:49<9:06:09,  1.47s/it]
  9%|▉         | 2170/24420 [53:50<9:05:31,  1.47s/it]
                                                      
{'loss': 0.2294, 'learning_rate': 5e-05, 'epoch': 0.44}

  9%|▉         | 2170/24420 [53:50<9:05:31,  1.47s/it]
  9%|▉         | 2171/24420 [53:52<9:04:49,  1.47s/it]
  9%|▉         | 2172/24420 [53:53<9:04:15,  1.47s/it]
  9%|▉         | 2173/24420 [53:54<9:04:11,  1.47s/it]
  9%|▉         | 2174/24420 [53:56<9:05:46,  1.47s/it]
  9%|▉         | 2175/24420 [53:57<9:05:13,  1.47s/it]
                                                      
{'loss': 0.2179, 'learning_rate': 5e-05, 'epoch': 0.45}

  9%|▉         | 2175/24420 [53:57<9:05:13,  1.47s/it]
  9%|▉         | 2176/24420 [53:59<9:04:49,  1.47s/it]
  9%|▉         | 2177/24420 [54:00<9:05:01,  1.47s/it]
  9%|▉         | 2178/24420 [54:02<9:05:16,  1.47s/it]
  9%|▉         | 2179/24420 [54:03<9:05:09,  1.47s/it]
  9%|▉         | 2180/24420 [54:05<9:05:16,  1.47s/it]
                                                      
{'loss': 0.2134, 'learning_rate': 5e-05, 'epoch': 0.45}

  9%|▉         | 2180/24420 [54:05<9:05:16,  1.47s/it]
  9%|▉         | 2181/24420 [54:06<9:05:28,  1.47s/it]
  9%|▉         | 2182/24420 [54:08<9:05:27,  1.47s/it]
  9%|▉         | 2183/24420 [54:09<9:05:11,  1.47s/it]
  9%|▉         | 2184/24420 [54:11<9:05:07,  1.47s/it]
  9%|▉         | 2185/24420 [54:12<9:05:15,  1.47s/it]
                                                      
{'loss': 0.2143, 'learning_rate': 5e-05, 'epoch': 0.45}

  9%|▉         | 2185/24420 [54:12<9:05:15,  1.47s/it]
  9%|▉         | 2186/24420 [54:14<9:05:22,  1.47s/it]
  9%|▉         | 2187/24420 [54:15<9:05:25,  1.47s/it]
  9%|▉         | 2188/24420 [54:17<9:05:14,  1.47s/it]
  9%|▉         | 2189/24420 [54:18<9:05:14,  1.47s/it]
  9%|▉         | 2190/24420 [54:19<9:04:43,  1.47s/it]
                                                      
{'loss': 0.1833, 'learning_rate': 5e-05, 'epoch': 0.45}

  9%|▉         | 2190/24420 [54:19<9:04:43,  1.47s/it]
  9%|▉         | 2191/24420 [54:21<9:05:20,  1.47s/it]
  9%|▉         | 2192/24420 [54:22<9:05:14,  1.47s/it]
  9%|▉         | 2193/24420 [54:24<9:05:09,  1.47s/it]
  9%|▉         | 2194/24420 [54:25<9:05:29,  1.47s/it]
  9%|▉         | 2195/24420 [54:27<9:06:04,  1.47s/it]
                                                      
{'loss': 0.1786, 'learning_rate': 5e-05, 'epoch': 0.45}

  9%|▉         | 2195/24420 [54:27<9:06:04,  1.47s/it]
  9%|▉         | 2196/24420 [54:28<9:05:33,  1.47s/it]
  9%|▉         | 2197/24420 [54:30<9:06:41,  1.48s/it]
  9%|▉         | 2198/24420 [54:31<9:06:05,  1.47s/it]
  9%|▉         | 2199/24420 [54:33<9:05:41,  1.47s/it]
  9%|▉         | 2200/24420 [54:34<9:06:32,  1.48s/it]
                                                      
{'loss': 0.2037, 'learning_rate': 5e-05, 'epoch': 0.45}

  9%|▉         | 2200/24420 [54:34<9:06:32,  1.48s/it]
  9%|▉         | 2201/24420 [54:39<14:41:43,  2.38s/it]
  9%|▉         | 2202/24420 [54:40<13:00:24,  2.11s/it]
  9%|▉         | 2203/24420 [54:42<11:48:48,  1.91s/it]
  9%|▉         | 2204/24420 [54:43<10:59:36,  1.78s/it]
  9%|▉         | 2205/24420 [54:45<10:25:16,  1.69s/it]
                                                       
{'loss': 0.2546, 'learning_rate': 5e-05, 'epoch': 0.45}

  9%|▉         | 2205/24420 [54:45<10:25:16,  1.69s/it]
  9%|▉         | 2206/24420 [54:46<10:01:07,  1.62s/it]
  9%|▉         | 2207/24420 [54:48<9:43:50,  1.58s/it] 
  9%|▉         | 2208/24420 [54:49<9:31:57,  1.54s/it]
  9%|▉         | 2209/24420 [54:50<9:24:32,  1.53s/it]
  9%|▉         | 2210/24420 [54:52<9:19:08,  1.51s/it]
                                                      
{'loss': 0.2136, 'learning_rate': 5e-05, 'epoch': 0.45}

  9%|▉         | 2210/24420 [54:52<9:19:08,  1.51s/it]
  9%|▉         | 2211/24420 [54:53<9:15:21,  1.50s/it]
  9%|▉         | 2212/24420 [54:55<9:12:02,  1.49s/it]
  9%|▉         | 2213/24420 [54:56<9:09:51,  1.49s/it]
  9%|▉         | 2214/24420 [54:58<9:07:39,  1.48s/it]
  9%|▉         | 2215/24420 [54:59<9:06:29,  1.48s/it]
                                                      
{'loss': 0.193, 'learning_rate': 5e-05, 'epoch': 0.45}

  9%|▉         | 2215/24420 [54:59<9:06:29,  1.48s/it]
  9%|▉         | 2216/24420 [55:01<9:05:36,  1.47s/it]
  9%|▉         | 2217/24420 [55:02<9:05:18,  1.47s/it]
  9%|▉         | 2218/24420 [55:04<9:05:10,  1.47s/it]
  9%|▉         | 2219/24420 [55:05<9:05:15,  1.47s/it]
  9%|▉         | 2220/24420 [55:07<9:04:57,  1.47s/it]
                                                      
{'loss': 0.1837, 'learning_rate': 5e-05, 'epoch': 0.45}

  9%|▉         | 2220/24420 [55:07<9:04:57,  1.47s/it]
  9%|▉         | 2221/24420 [55:08<9:05:15,  1.47s/it]
  9%|▉         | 2222/24420 [55:10<9:04:47,  1.47s/it]
  9%|▉         | 2223/24420 [55:11<9:04:18,  1.47s/it]
  9%|▉         | 2224/24420 [55:13<9:04:13,  1.47s/it]
  9%|▉         | 2225/24420 [55:14<9:04:16,  1.47s/it]
                                                      
{'loss': 0.163, 'learning_rate': 5e-05, 'epoch': 0.46}

  9%|▉         | 2225/24420 [55:14<9:04:16,  1.47s/it]
  9%|▉         | 2226/24420 [55:15<9:05:19,  1.47s/it]
  9%|▉         | 2227/24420 [55:17<9:05:03,  1.47s/it]
  9%|▉         | 2228/24420 [55:18<9:05:06,  1.47s/it]
  9%|▉         | 2229/24420 [55:20<9:04:27,  1.47s/it]
  9%|▉         | 2230/24420 [55:21<9:04:34,  1.47s/it]
                                                      
{'loss': 0.2013, 'learning_rate': 5e-05, 'epoch': 0.46}

  9%|▉         | 2230/24420 [55:21<9:04:34,  1.47s/it]
  9%|▉         | 2231/24420 [55:23<9:04:14,  1.47s/it]
  9%|▉         | 2232/24420 [55:24<9:03:51,  1.47s/it]
  9%|▉         | 2233/24420 [55:26<9:04:36,  1.47s/it]
  9%|▉         | 2234/24420 [55:27<9:05:10,  1.47s/it]
  9%|▉         | 2235/24420 [55:29<9:06:00,  1.48s/it]
                                                      
{'loss': 0.194, 'learning_rate': 5e-05, 'epoch': 0.46}

  9%|▉         | 2235/24420 [55:29<9:06:00,  1.48s/it]
  9%|▉         | 2236/24420 [55:30<9:05:44,  1.48s/it]
  9%|▉         | 2237/24420 [55:32<9:05:12,  1.47s/it]
  9%|▉         | 2238/24420 [55:33<9:05:18,  1.47s/it]
  9%|▉         | 2239/24420 [55:35<9:04:10,  1.47s/it]
  9%|▉         | 2240/24420 [55:36<9:04:14,  1.47s/it]
                                                      
{'loss': 0.2091, 'learning_rate': 5e-05, 'epoch': 0.46}

  9%|▉         | 2240/24420 [55:36<9:04:14,  1.47s/it]
  9%|▉         | 2241/24420 [55:38<9:05:51,  1.48s/it]
  9%|▉         | 2242/24420 [55:39<9:05:39,  1.48s/it]
  9%|▉         | 2243/24420 [55:41<9:05:10,  1.47s/it]
  9%|▉         | 2244/24420 [55:42<9:05:59,  1.48s/it]
  9%|▉         | 2245/24420 [55:44<9:06:04,  1.48s/it]
                                                      
{'loss': 0.245, 'learning_rate': 5e-05, 'epoch': 0.46}

  9%|▉         | 2245/24420 [55:44<9:06:04,  1.48s/it]
  9%|▉         | 2246/24420 [55:45<9:05:31,  1.48s/it]
  9%|▉         | 2247/24420 [55:46<9:04:26,  1.47s/it]
  9%|▉         | 2248/24420 [55:48<9:05:23,  1.48s/it]
  9%|▉         | 2249/24420 [55:49<9:05:47,  1.48s/it]
  9%|▉         | 2250/24420 [55:51<9:05:18,  1.48s/it]
                                                      
{'loss': 0.1865, 'learning_rate': 5e-05, 'epoch': 0.46}

  9%|▉         | 2250/24420 [55:51<9:05:18,  1.48s/it]
  9%|▉         | 2251/24420 [55:52<9:05:40,  1.48s/it]
  9%|▉         | 2252/24420 [55:54<9:05:06,  1.48s/it]
  9%|▉         | 2253/24420 [55:55<9:06:33,  1.48s/it]
  9%|▉         | 2254/24420 [55:57<9:05:33,  1.48s/it]
  9%|▉         | 2255/24420 [55:58<9:04:37,  1.47s/it]
                                                      
{'loss': 0.2274, 'learning_rate': 5e-05, 'epoch': 0.46}

  9%|▉         | 2255/24420 [55:58<9:04:37,  1.47s/it]
  9%|▉         | 2256/24420 [56:00<9:04:13,  1.47s/it]
  9%|▉         | 2257/24420 [56:01<9:03:50,  1.47s/it]
  9%|▉         | 2258/24420 [56:03<9:04:17,  1.47s/it]
  9%|▉         | 2259/24420 [56:04<9:03:44,  1.47s/it]
  9%|▉         | 2260/24420 [56:06<9:03:24,  1.47s/it]
                                                      
{'loss': 0.2017, 'learning_rate': 5e-05, 'epoch': 0.46}

  9%|▉         | 2260/24420 [56:06<9:03:24,  1.47s/it]
  9%|▉         | 2261/24420 [56:07<9:03:53,  1.47s/it]
  9%|▉         | 2262/24420 [56:09<9:03:57,  1.47s/it]
  9%|▉         | 2263/24420 [56:10<9:03:01,  1.47s/it]
  9%|▉         | 2264/24420 [56:11<9:02:28,  1.47s/it]
  9%|▉         | 2265/24420 [56:13<9:02:49,  1.47s/it]
                                                      
{'loss': 0.1933, 'learning_rate': 5e-05, 'epoch': 0.46}

  9%|▉         | 2265/24420 [56:13<9:02:49,  1.47s/it]
  9%|▉         | 2266/24420 [56:14<9:04:32,  1.47s/it]
  9%|▉         | 2267/24420 [56:16<9:04:24,  1.47s/it]
  9%|▉         | 2268/24420 [56:17<9:04:02,  1.47s/it]
  9%|▉         | 2269/24420 [56:19<9:03:35,  1.47s/it]
  9%|▉         | 2270/24420 [56:20<9:04:11,  1.47s/it]
                                                      
{'loss': 0.263, 'learning_rate': 5e-05, 'epoch': 0.46}

  9%|▉         | 2270/24420 [56:20<9:04:11,  1.47s/it]
  9%|▉         | 2271/24420 [56:22<9:05:01,  1.48s/it]
  9%|▉         | 2272/24420 [56:23<9:04:57,  1.48s/it]
  9%|▉         | 2273/24420 [56:25<9:03:57,  1.47s/it]
  9%|▉         | 2274/24420 [56:26<9:04:08,  1.47s/it]
  9%|▉         | 2275/24420 [56:28<9:03:40,  1.47s/it]
                                                      
{'loss': 0.2048, 'learning_rate': 5e-05, 'epoch': 0.47}

  9%|▉         | 2275/24420 [56:28<9:03:40,  1.47s/it]
  9%|▉         | 2276/24420 [56:29<9:03:11,  1.47s/it]
  9%|▉         | 2277/24420 [56:31<9:03:17,  1.47s/it]
  9%|▉         | 2278/24420 [56:32<9:02:55,  1.47s/it]
  9%|▉         | 2279/24420 [56:34<9:02:02,  1.47s/it]
  9%|▉         | 2280/24420 [56:35<9:01:56,  1.47s/it]
                                                      
{'loss': 0.2042, 'learning_rate': 5e-05, 'epoch': 0.47}

  9%|▉         | 2280/24420 [56:35<9:01:56,  1.47s/it]
  9%|▉         | 2281/24420 [56:37<9:01:59,  1.47s/it]
  9%|▉         | 2282/24420 [56:38<9:02:41,  1.47s/it]
  9%|▉         | 2283/24420 [56:39<9:02:46,  1.47s/it]
  9%|▉         | 2284/24420 [56:41<9:03:15,  1.47s/it]
  9%|▉         | 2285/24420 [56:42<9:03:27,  1.47s/it]
                                                      
{'loss': 0.2228, 'learning_rate': 5e-05, 'epoch': 0.47}

  9%|▉         | 2285/24420 [56:42<9:03:27,  1.47s/it]
  9%|▉         | 2286/24420 [56:44<9:03:37,  1.47s/it]
  9%|▉         | 2287/24420 [56:45<9:02:57,  1.47s/it]
  9%|▉         | 2288/24420 [56:47<9:02:44,  1.47s/it]
  9%|▉         | 2289/24420 [56:48<9:02:31,  1.47s/it]
  9%|▉         | 2290/24420 [56:50<9:02:36,  1.47s/it]
                                                      
{'loss': 0.2031, 'learning_rate': 5e-05, 'epoch': 0.47}

  9%|▉         | 2290/24420 [56:50<9:02:36,  1.47s/it]
  9%|▉         | 2291/24420 [56:51<9:04:06,  1.48s/it]
  9%|▉         | 2292/24420 [56:53<9:03:28,  1.47s/it]
  9%|▉         | 2293/24420 [56:54<9:03:07,  1.47s/it]
  9%|▉         | 2294/24420 [56:56<9:03:47,  1.47s/it]
  9%|▉         | 2295/24420 [56:57<9:02:53,  1.47s/it]
                                                      
{'loss': 0.23, 'learning_rate': 5e-05, 'epoch': 0.47}

  9%|▉         | 2295/24420 [56:57<9:02:53,  1.47s/it]
  9%|▉         | 2296/24420 [56:59<9:02:31,  1.47s/it]
  9%|▉         | 2297/24420 [57:00<9:02:44,  1.47s/it]
  9%|▉         | 2298/24420 [57:02<9:02:23,  1.47s/it]
  9%|▉         | 2299/24420 [57:03<9:02:57,  1.47s/it]
  9%|▉         | 2300/24420 [57:05<9:03:17,  1.47s/it]
                                                      
{'loss': 0.2394, 'learning_rate': 5e-05, 'epoch': 0.47}

  9%|▉         | 2300/24420 [57:05<9:03:17,  1.47s/it]
  9%|▉         | 2301/24420 [57:06<9:03:41,  1.47s/it]
  9%|▉         | 2302/24420 [57:07<9:03:58,  1.48s/it]
  9%|▉         | 2303/24420 [57:09<9:03:13,  1.47s/it]
  9%|▉         | 2304/24420 [57:10<9:03:22,  1.47s/it]
  9%|▉         | 2305/24420 [57:12<9:02:25,  1.47s/it]
                                                      
{'loss': 0.2254, 'learning_rate': 5e-05, 'epoch': 0.47}

  9%|▉         | 2305/24420 [57:12<9:02:25,  1.47s/it]
  9%|▉         | 2306/24420 [57:13<9:02:03,  1.47s/it]
  9%|▉         | 2307/24420 [57:15<9:01:50,  1.47s/it]
  9%|▉         | 2308/24420 [57:16<9:01:00,  1.47s/it]
  9%|▉         | 2309/24420 [57:18<9:01:04,  1.47s/it]
  9%|▉         | 2310/24420 [57:19<9:02:12,  1.47s/it]
                                                      
{'loss': 0.2111, 'learning_rate': 5e-05, 'epoch': 0.47}

  9%|▉         | 2310/24420 [57:19<9:02:12,  1.47s/it]
  9%|▉         | 2311/24420 [57:21<9:03:03,  1.47s/it]
  9%|▉         | 2312/24420 [57:22<9:03:06,  1.47s/it]
  9%|▉         | 2313/24420 [57:24<9:02:48,  1.47s/it]
  9%|▉         | 2314/24420 [57:25<9:02:29,  1.47s/it]
  9%|▉         | 2315/24420 [57:27<9:02:10,  1.47s/it]
                                                      
{'loss': 0.2007, 'learning_rate': 5e-05, 'epoch': 0.47}

  9%|▉         | 2315/24420 [57:27<9:02:10,  1.47s/it]
  9%|▉         | 2316/24420 [57:28<9:02:43,  1.47s/it]
  9%|▉         | 2317/24420 [57:30<9:02:37,  1.47s/it]
  9%|▉         | 2318/24420 [57:31<9:02:43,  1.47s/it]
  9%|▉         | 2319/24420 [57:32<9:02:26,  1.47s/it]
 10%|▉         | 2320/24420 [57:34<9:02:26,  1.47s/it]
                                                      
{'loss': 0.2068, 'learning_rate': 5e-05, 'epoch': 0.48}

 10%|▉         | 2320/24420 [57:34<9:02:26,  1.47s/it]
 10%|▉         | 2321/24420 [57:35<9:02:23,  1.47s/it]
 10%|▉         | 2322/24420 [57:37<9:02:13,  1.47s/it]
 10%|▉         | 2323/24420 [57:38<9:02:08,  1.47s/it]
 10%|▉         | 2324/24420 [57:40<9:02:31,  1.47s/it]
 10%|▉         | 2325/24420 [57:41<9:02:02,  1.47s/it]
                                                      
{'loss': 0.2097, 'learning_rate': 5e-05, 'epoch': 0.48}

 10%|▉         | 2325/24420 [57:41<9:02:02,  1.47s/it]
 10%|▉         | 2326/24420 [57:43<9:02:44,  1.47s/it]
 10%|▉         | 2327/24420 [57:44<9:02:14,  1.47s/it]
 10%|▉         | 2328/24420 [57:46<9:02:03,  1.47s/it]
 10%|▉         | 2329/24420 [57:47<9:02:33,  1.47s/it]
 10%|▉         | 2330/24420 [57:49<9:02:51,  1.47s/it]
                                                      
{'loss': 0.2215, 'learning_rate': 5e-05, 'epoch': 0.48}

 10%|▉         | 2330/24420 [57:49<9:02:51,  1.47s/it]
 10%|▉         | 2331/24420 [57:50<9:02:20,  1.47s/it]
 10%|▉         | 2332/24420 [57:52<9:01:59,  1.47s/it]
 10%|▉         | 2333/24420 [57:53<9:02:17,  1.47s/it]
 10%|▉         | 2334/24420 [57:55<9:02:16,  1.47s/it]
 10%|▉         | 2335/24420 [57:56<9:02:01,  1.47s/it]
                                                      
{'loss': 0.2187, 'learning_rate': 5e-05, 'epoch': 0.48}

 10%|▉         | 2335/24420 [57:56<9:02:01,  1.47s/it]
 10%|▉         | 2336/24420 [57:58<9:02:01,  1.47s/it]
 10%|▉         | 2337/24420 [57:59<9:02:24,  1.47s/it]
 10%|▉         | 2338/24420 [58:00<9:02:29,  1.47s/it]
 10%|▉         | 2339/24420 [58:02<9:01:35,  1.47s/it]
 10%|▉         | 2340/24420 [58:03<9:02:05,  1.47s/it]
                                                      
{'loss': 0.196, 'learning_rate': 5e-05, 'epoch': 0.48}

 10%|▉         | 2340/24420 [58:03<9:02:05,  1.47s/it]
 10%|▉         | 2341/24420 [58:05<9:02:38,  1.47s/it]
 10%|▉         | 2342/24420 [58:06<9:02:03,  1.47s/it]
 10%|▉         | 2343/24420 [58:08<9:01:32,  1.47s/it]
 10%|▉         | 2344/24420 [58:09<9:01:50,  1.47s/it]
 10%|▉         | 2345/24420 [58:11<9:02:01,  1.47s/it]
                                                      
{'loss': 0.2151, 'learning_rate': 5e-05, 'epoch': 0.48}

 10%|▉         | 2345/24420 [58:11<9:02:01,  1.47s/it]
 10%|▉         | 2346/24420 [58:12<9:02:01,  1.47s/it]
 10%|▉         | 2347/24420 [58:14<9:01:35,  1.47s/it]
 10%|▉         | 2348/24420 [58:15<9:01:20,  1.47s/it]
 10%|▉         | 2349/24420 [58:17<9:01:22,  1.47s/it]
 10%|▉         | 2350/24420 [58:18<9:01:35,  1.47s/it]
                                                      
{'loss': 0.2124, 'learning_rate': 5e-05, 'epoch': 0.48}

 10%|▉         | 2350/24420 [58:18<9:01:35,  1.47s/it]
 10%|▉         | 2351/24420 [58:20<9:02:00,  1.47s/it]
 10%|▉         | 2352/24420 [58:21<9:01:26,  1.47s/it]
 10%|▉         | 2353/24420 [58:23<9:01:32,  1.47s/it]
 10%|▉         | 2354/24420 [58:24<9:01:30,  1.47s/it]
 10%|▉         | 2355/24420 [58:26<9:01:15,  1.47s/it]
                                                      
{'loss': 0.2238, 'learning_rate': 5e-05, 'epoch': 0.48}

 10%|▉         | 2355/24420 [58:26<9:01:15,  1.47s/it]
 10%|▉         | 2356/24420 [58:27<9:01:04,  1.47s/it]
 10%|▉         | 2357/24420 [58:28<9:01:22,  1.47s/it]
 10%|▉         | 2358/24420 [58:30<9:00:39,  1.47s/it]
 10%|▉         | 2359/24420 [58:31<9:02:06,  1.47s/it]
 10%|▉         | 2360/24420 [58:33<9:03:19,  1.48s/it]
                                                      
{'loss': 0.2064, 'learning_rate': 5e-05, 'epoch': 0.48}

 10%|▉         | 2360/24420 [58:33<9:03:19,  1.48s/it]
 10%|▉         | 2361/24420 [58:34<9:02:55,  1.48s/it]
 10%|▉         | 2362/24420 [58:36<9:01:59,  1.47s/it]
 10%|▉         | 2363/24420 [58:37<9:02:16,  1.48s/it]
 10%|▉         | 2364/24420 [58:39<9:02:47,  1.48s/it]
 10%|▉         | 2365/24420 [58:40<9:03:08,  1.48s/it]
                                                      
{'loss': 0.1957, 'learning_rate': 5e-05, 'epoch': 0.48}

 10%|▉         | 2365/24420 [58:40<9:03:08,  1.48s/it]
 10%|▉         | 2366/24420 [58:42<9:03:34,  1.48s/it]
 10%|▉         | 2367/24420 [58:43<9:03:09,  1.48s/it]
 10%|▉         | 2368/24420 [58:45<9:01:48,  1.47s/it]
 10%|▉         | 2369/24420 [58:46<9:01:59,  1.47s/it]
 10%|▉         | 2370/24420 [58:48<9:03:34,  1.48s/it]
                                                      
{'loss': 0.2328, 'learning_rate': 5e-05, 'epoch': 0.49}

 10%|▉         | 2370/24420 [58:48<9:03:34,  1.48s/it]
 10%|▉         | 2371/24420 [58:49<9:03:13,  1.48s/it]
 10%|▉         | 2372/24420 [58:51<9:02:12,  1.48s/it]
 10%|▉         | 2373/24420 [58:52<9:02:56,  1.48s/it]
 10%|▉         | 2374/24420 [58:54<9:02:31,  1.48s/it]
 10%|▉         | 2375/24420 [58:55<9:01:12,  1.47s/it]
                                                      
{'loss': 0.2032, 'learning_rate': 5e-05, 'epoch': 0.49}

 10%|▉         | 2375/24420 [58:55<9:01:12,  1.47s/it]
 10%|▉         | 2376/24420 [58:56<9:01:11,  1.47s/it]
 10%|▉         | 2377/24420 [58:58<9:00:32,  1.47s/it]
 10%|▉         | 2378/24420 [58:59<9:01:21,  1.47s/it]
 10%|▉         | 2379/24420 [59:01<9:01:32,  1.47s/it]
 10%|▉         | 2380/24420 [59:02<9:00:41,  1.47s/it]
                                                      
{'loss': 0.2006, 'learning_rate': 5e-05, 'epoch': 0.49}

 10%|▉         | 2380/24420 [59:02<9:00:41,  1.47s/it]
 10%|▉         | 2381/24420 [59:04<9:01:01,  1.47s/it]
 10%|▉         | 2382/24420 [59:05<9:00:50,  1.47s/it]
 10%|▉         | 2383/24420 [59:07<9:01:01,  1.47s/it]
 10%|▉         | 2384/24420 [59:08<9:01:22,  1.47s/it]
 10%|▉         | 2385/24420 [59:10<9:01:17,  1.47s/it]
                                                      
{'loss': 0.2159, 'learning_rate': 5e-05, 'epoch': 0.49}

 10%|▉         | 2385/24420 [59:10<9:01:17,  1.47s/it]
 10%|▉         | 2386/24420 [59:11<9:01:51,  1.48s/it]
 10%|▉         | 2387/24420 [59:13<9:01:24,  1.47s/it]
 10%|▉         | 2388/24420 [59:14<9:00:59,  1.47s/it]
 10%|▉         | 2389/24420 [59:16<9:01:56,  1.48s/it]
 10%|▉         | 2390/24420 [59:17<9:02:14,  1.48s/it]
                                                      
{'loss': 0.2156, 'learning_rate': 5e-05, 'epoch': 0.49}

 10%|▉         | 2390/24420 [59:17<9:02:14,  1.48s/it]
 10%|▉         | 2391/24420 [59:19<9:02:04,  1.48s/it]
 10%|▉         | 2392/24420 [59:20<9:01:07,  1.47s/it]
 10%|▉         | 2393/24420 [59:22<9:01:49,  1.48s/it]
 10%|▉         | 2394/24420 [59:23<9:01:24,  1.47s/it]
 10%|▉         | 2395/24420 [59:25<9:01:11,  1.47s/it]
                                                      
{'loss': 0.2316, 'learning_rate': 5e-05, 'epoch': 0.49}

 10%|▉         | 2395/24420 [59:25<9:01:11,  1.47s/it]
 10%|▉         | 2396/24420 [59:26<9:02:55,  1.48s/it]
 10%|▉         | 2397/24420 [59:27<9:02:45,  1.48s/it]
 10%|▉         | 2398/24420 [59:29<9:02:20,  1.48s/it]
 10%|▉         | 2399/24420 [59:30<9:03:14,  1.48s/it]
 10%|▉         | 2400/24420 [59:32<9:03:19,  1.48s/it]
                                                      
{'loss': 0.1959, 'learning_rate': 5e-05, 'epoch': 0.49}

 10%|▉         | 2400/24420 [59:32<9:03:19,  1.48s/it]
 10%|▉         | 2401/24420 [59:38<17:43:44,  2.90s/it]
 10%|▉         | 2402/24420 [59:40<15:06:09,  2.47s/it]
 10%|▉         | 2403/24420 [59:41<13:16:05,  2.17s/it]
 10%|▉         | 2404/24420 [59:43<11:59:28,  1.96s/it]
 10%|▉         | 2405/24420 [59:44<11:05:27,  1.81s/it]
                                                       
{'loss': 0.2118, 'learning_rate': 5e-05, 'epoch': 0.49}

 10%|▉         | 2405/24420 [59:44<11:05:27,  1.81s/it]
 10%|▉         | 2406/24420 [59:45<10:27:48,  1.71s/it]
 10%|▉         | 2407/24420 [59:47<10:01:46,  1.64s/it]
 10%|▉         | 2408/24420 [59:48<9:44:21,  1.59s/it] 
 10%|▉         | 2409/24420 [59:50<9:30:21,  1.55s/it]
 10%|▉         | 2410/24420 [59:51<9:22:05,  1.53s/it]
                                                      
{'loss': 0.2246, 'learning_rate': 5e-05, 'epoch': 0.49}

 10%|▉         | 2410/24420 [59:51<9:22:05,  1.53s/it]
 10%|▉         | 2411/24420 [59:53<9:15:30,  1.51s/it]
 10%|▉         | 2412/24420 [59:54<9:12:06,  1.51s/it]
 10%|▉         | 2413/24420 [59:56<9:08:41,  1.50s/it]
 10%|▉         | 2414/24420 [59:57<9:07:07,  1.49s/it]
 10%|▉         | 2415/24420 [59:59<9:05:52,  1.49s/it]
                                                      
{'loss': 0.1822, 'learning_rate': 5e-05, 'epoch': 0.49}

 10%|▉         | 2415/24420 [59:59<9:05:52,  1.49s/it]
 10%|▉         | 2416/24420 [1:00:00<9:05:44,  1.49s/it]
 10%|▉         | 2417/24420 [1:00:02<9:04:11,  1.48s/it]
 10%|▉         | 2418/24420 [1:00:03<9:02:15,  1.48s/it]
 10%|▉         | 2419/24420 [1:00:05<9:00:48,  1.47s/it]
 10%|▉         | 2420/24420 [1:00:06<9:01:08,  1.48s/it]
                                                        
{'loss': 0.191, 'learning_rate': 5e-05, 'epoch': 0.5}

 10%|▉         | 2420/24420 [1:00:06<9:01:08,  1.48s/it]
 10%|▉         | 2421/24420 [1:00:08<9:01:40,  1.48s/it]
 10%|▉         | 2422/24420 [1:00:09<9:01:09,  1.48s/it]
 10%|▉         | 2423/24420 [1:00:11<9:01:02,  1.48s/it]
 10%|▉         | 2424/24420 [1:00:12<9:01:39,  1.48s/it]
 10%|▉         | 2425/24420 [1:00:14<9:01:01,  1.48s/it]
                                                        
{'loss': 0.2705, 'learning_rate': 5e-05, 'epoch': 0.5}

 10%|▉         | 2425/24420 [1:00:14<9:01:01,  1.48s/it]
 10%|▉         | 2426/24420 [1:00:15<9:00:28,  1.47s/it]
 10%|▉         | 2427/24420 [1:00:16<8:59:47,  1.47s/it]
 10%|▉         | 2428/24420 [1:00:18<8:59:16,  1.47s/it]
 10%|▉         | 2429/24420 [1:00:19<9:00:27,  1.47s/it]
 10%|▉         | 2430/24420 [1:00:21<9:02:37,  1.48s/it]
                                                        
{'loss': 0.1936, 'learning_rate': 5e-05, 'epoch': 0.5}

 10%|▉         | 2430/24420 [1:00:21<9:02:37,  1.48s/it]
 10%|▉         | 2431/24420 [1:00:22<9:01:48,  1.48s/it]
 10%|▉         | 2432/24420 [1:00:25<10:13:39,  1.67s/it]
 10%|▉         | 2433/24420 [1:00:26<9:57:08,  1.63s/it] 
 10%|▉         | 2434/24420 [1:00:28<9:39:54,  1.58s/it]
 10%|▉         | 2435/24420 [1:00:29<9:27:26,  1.55s/it]
                                                        
{'loss': 0.2754, 'learning_rate': 5e-05, 'epoch': 0.5}

 10%|▉         | 2435/24420 [1:00:29<9:27:26,  1.55s/it]
 10%|▉         | 2436/24420 [1:00:30<9:19:21,  1.53s/it]
 10%|▉         | 2437/24420 [1:00:32<9:13:23,  1.51s/it]
 10%|▉         | 2438/24420 [1:00:33<9:07:58,  1.50s/it]
 10%|▉         | 2439/24420 [1:00:35<9:05:42,  1.49s/it]
 10%|▉         | 2440/24420 [1:00:36<9:04:00,  1.48s/it]
                                                        
{'loss': 0.1977, 'learning_rate': 5e-05, 'epoch': 0.5}

 10%|▉         | 2440/24420 [1:00:36<9:04:00,  1.48s/it]
 10%|▉         | 2441/24420 [1:00:38<9:03:09,  1.48s/it]
 10%|█         | 2442/24420 [1:00:39<9:02:14,  1.48s/it]
 10%|█         | 2443/24420 [1:00:41<9:01:45,  1.48s/it]
 10%|█         | 2444/24420 [1:00:42<9:01:16,  1.48s/it]
 10%|█         | 2445/24420 [1:00:44<9:00:50,  1.48s/it]
                                                        
{'loss': 0.2203, 'learning_rate': 5e-05, 'epoch': 0.5}

 10%|█         | 2445/24420 [1:00:44<9:00:50,  1.48s/it]
 10%|█         | 2446/24420 [1:00:45<9:00:14,  1.48s/it]
 10%|█         | 2447/24420 [1:00:47<8:59:23,  1.47s/it]
 10%|█         | 2448/24420 [1:00:48<8:58:41,  1.47s/it]
 10%|█         | 2449/24420 [1:00:50<8:59:13,  1.47s/it]
 10%|█         | 2450/24420 [1:00:51<8:59:57,  1.47s/it]
                                                        
{'loss': 0.1959, 'learning_rate': 5e-05, 'epoch': 0.5}

 10%|█         | 2450/24420 [1:00:51<8:59:57,  1.47s/it]
 10%|█         | 2451/24420 [1:00:53<9:00:02,  1.47s/it]
 10%|█         | 2452/24420 [1:00:54<8:59:56,  1.47s/it]
 10%|█         | 2453/24420 [1:00:56<9:00:03,  1.48s/it]
 10%|█         | 2454/24420 [1:00:57<8:58:52,  1.47s/it]
 10%|█         | 2455/24420 [1:00:58<8:59:01,  1.47s/it]
                                                        
{'loss': 0.1938, 'learning_rate': 5e-05, 'epoch': 0.5}

 10%|█         | 2455/24420 [1:00:58<8:59:01,  1.47s/it]
 10%|█         | 2456/24420 [1:01:00<9:00:19,  1.48s/it]
 10%|█         | 2457/24420 [1:01:01<9:00:43,  1.48s/it]
 10%|█         | 2458/24420 [1:01:03<9:00:31,  1.48s/it]
 10%|█         | 2459/24420 [1:01:04<9:00:17,  1.48s/it]
 10%|█         | 2460/24420 [1:01:06<9:00:56,  1.48s/it]
                                                        
{'loss': 0.2013, 'learning_rate': 5e-05, 'epoch': 0.5}

 10%|█         | 2460/24420 [1:01:06<9:00:56,  1.48s/it]
 10%|█         | 2461/24420 [1:01:07<9:01:23,  1.48s/it]
 10%|█         | 2462/24420 [1:01:09<9:00:42,  1.48s/it]
 10%|█         | 2463/24420 [1:01:10<8:59:45,  1.47s/it]
 10%|█         | 2464/24420 [1:01:12<8:59:03,  1.47s/it]
 10%|█         | 2465/24420 [1:01:13<8:59:09,  1.47s/it]
                                                        
{'loss': 0.1976, 'learning_rate': 5e-05, 'epoch': 0.5}

 10%|█         | 2465/24420 [1:01:13<8:59:09,  1.47s/it]
 10%|█         | 2466/24420 [1:01:15<8:59:45,  1.48s/it]
 10%|█         | 2467/24420 [1:01:16<9:02:07,  1.48s/it]
 10%|█         | 2468/24420 [1:01:18<9:01:18,  1.48s/it]
 10%|█         | 2469/24420 [1:01:19<9:00:12,  1.48s/it]
 10%|█         | 2470/24420 [1:01:21<9:00:27,  1.48s/it]
                                                        
{'loss': 0.2196, 'learning_rate': 5e-05, 'epoch': 0.51}

 10%|█         | 2470/24420 [1:01:21<9:00:27,  1.48s/it]
 10%|█         | 2471/24420 [1:01:22<9:00:09,  1.48s/it]
 10%|█         | 2472/24420 [1:01:24<8:59:38,  1.48s/it]
 10%|█         | 2473/24420 [1:01:25<8:59:21,  1.47s/it]
 10%|█         | 2474/24420 [1:01:27<8:58:59,  1.47s/it]
 10%|█         | 2475/24420 [1:01:28<8:59:05,  1.47s/it]
                                                        
{'loss': 0.2443, 'learning_rate': 5e-05, 'epoch': 0.51}

 10%|█         | 2475/24420 [1:01:28<8:59:05,  1.47s/it]
 10%|█         | 2476/24420 [1:01:29<8:59:07,  1.47s/it]
 10%|█         | 2477/24420 [1:01:31<8:58:58,  1.47s/it]
 10%|█         | 2478/24420 [1:01:32<8:58:43,  1.47s/it]
 10%|█         | 2479/24420 [1:01:34<8:58:27,  1.47s/it]
 10%|█         | 2480/24420 [1:01:35<8:58:45,  1.47s/it]
                                                        
{'loss': 0.2003, 'learning_rate': 5e-05, 'epoch': 0.51}

 10%|█         | 2480/24420 [1:01:35<8:58:45,  1.47s/it]
 10%|█         | 2481/24420 [1:01:37<8:58:57,  1.47s/it]
 10%|█         | 2482/24420 [1:01:38<8:59:08,  1.47s/it]
 10%|█         | 2483/24420 [1:01:40<8:59:24,  1.48s/it]
 10%|█         | 2484/24420 [1:01:41<8:58:57,  1.47s/it]
 10%|█         | 2485/24420 [1:01:43<8:58:35,  1.47s/it]
                                                        
{'loss': 0.1977, 'learning_rate': 5e-05, 'epoch': 0.51}

 10%|█         | 2485/24420 [1:01:43<8:58:35,  1.47s/it]
 10%|█         | 2486/24420 [1:01:44<8:57:54,  1.47s/it]
 10%|█         | 2487/24420 [1:01:46<8:58:25,  1.47s/it]
 10%|█         | 2488/24420 [1:01:47<8:58:01,  1.47s/it]
 10%|█         | 2489/24420 [1:01:49<8:58:03,  1.47s/it]
 10%|█         | 2490/24420 [1:01:50<8:57:47,  1.47s/it]
                                                        
{'loss': 0.2013, 'learning_rate': 5e-05, 'epoch': 0.51}

 10%|█         | 2490/24420 [1:01:50<8:57:47,  1.47s/it]
 10%|█         | 2491/24420 [1:01:52<8:57:42,  1.47s/it]
 10%|█         | 2492/24420 [1:01:53<8:57:26,  1.47s/it]
 10%|█         | 2493/24420 [1:01:54<8:57:27,  1.47s/it]
 10%|█         | 2494/24420 [1:01:56<8:57:19,  1.47s/it]
 10%|█         | 2495/24420 [1:01:57<8:58:08,  1.47s/it]
                                                        
{'loss': 0.1952, 'learning_rate': 5e-05, 'epoch': 0.51}

 10%|█         | 2495/24420 [1:01:57<8:58:08,  1.47s/it]
 10%|█         | 2496/24420 [1:01:59<8:59:02,  1.48s/it]
 10%|█         | 2497/24420 [1:02:00<8:58:11,  1.47s/it]
 10%|█         | 2498/24420 [1:02:02<8:58:52,  1.47s/it]
 10%|█         | 2499/24420 [1:02:03<8:59:10,  1.48s/it]
 10%|█         | 2500/24420 [1:02:05<8:59:54,  1.48s/it]
                                                        
{'loss': 0.1973, 'learning_rate': 5e-05, 'epoch': 0.51}

 10%|█         | 2500/24420 [1:02:05<8:59:54,  1.48s/it]
 10%|█         | 2501/24420 [1:02:06<9:00:01,  1.48s/it]
 10%|█         | 2502/24420 [1:02:08<8:59:09,  1.48s/it]
 10%|█         | 2503/24420 [1:02:09<8:59:27,  1.48s/it]
 10%|█         | 2504/24420 [1:02:11<9:00:07,  1.48s/it]
 10%|█         | 2505/24420 [1:02:12<9:00:13,  1.48s/it]
                                                        
{'loss': 0.2158, 'learning_rate': 5e-05, 'epoch': 0.51}

 10%|█         | 2505/24420 [1:02:12<9:00:13,  1.48s/it]
 10%|█         | 2506/24420 [1:02:14<9:00:58,  1.48s/it]
 10%|█         | 2507/24420 [1:02:15<8:59:34,  1.48s/it]
 10%|█         | 2508/24420 [1:02:17<8:58:42,  1.48s/it]
 10%|█         | 2509/24420 [1:02:18<8:59:18,  1.48s/it]
 10%|█         | 2510/24420 [1:02:20<8:59:29,  1.48s/it]
                                                        
{'loss': 0.1819, 'learning_rate': 5e-05, 'epoch': 0.51}

 10%|█         | 2510/24420 [1:02:20<8:59:29,  1.48s/it]
 10%|█         | 2511/24420 [1:02:21<9:00:44,  1.48s/it]
 10%|█         | 2512/24420 [1:02:23<8:59:51,  1.48s/it]
 10%|█         | 2513/24420 [1:02:24<8:58:50,  1.48s/it]
 10%|█         | 2514/24420 [1:02:25<8:58:40,  1.48s/it]
 10%|█         | 2515/24420 [1:02:27<8:59:11,  1.48s/it]
                                                        
{'loss': 0.1937, 'learning_rate': 5e-05, 'epoch': 0.51}

 10%|█         | 2515/24420 [1:02:27<8:59:11,  1.48s/it]
 10%|█         | 2516/24420 [1:02:28<8:58:45,  1.48s/it]
 10%|█         | 2517/24420 [1:02:30<8:59:24,  1.48s/it]
 10%|█         | 2518/24420 [1:02:31<9:00:16,  1.48s/it]
 10%|█         | 2519/24420 [1:02:33<9:00:56,  1.48s/it]
 10%|█         | 2520/24420 [1:02:34<9:00:11,  1.48s/it]
                                                        
{'loss': 0.2034, 'learning_rate': 5e-05, 'epoch': 0.52}

 10%|█         | 2520/24420 [1:02:34<9:00:11,  1.48s/it]
 10%|█         | 2521/24420 [1:02:36<8:59:40,  1.48s/it]
 10%|█         | 2522/24420 [1:02:37<8:59:23,  1.48s/it]
 10%|█         | 2523/24420 [1:02:39<8:59:11,  1.48s/it]
 10%|█         | 2524/24420 [1:02:40<8:59:03,  1.48s/it]
 10%|█         | 2525/24420 [1:02:42<8:58:47,  1.48s/it]
                                                        
{'loss': 0.2332, 'learning_rate': 5e-05, 'epoch': 0.52}

 10%|█         | 2525/24420 [1:02:42<8:58:47,  1.48s/it]
 10%|█         | 2526/24420 [1:02:43<8:57:57,  1.47s/it]
 10%|█         | 2527/24420 [1:02:45<8:57:34,  1.47s/it]
 10%|█         | 2528/24420 [1:02:46<8:57:30,  1.47s/it]
 10%|█         | 2529/24420 [1:02:48<8:58:34,  1.48s/it]
 10%|█         | 2530/24420 [1:02:49<8:59:08,  1.48s/it]
                                                        
{'loss': 0.2181, 'learning_rate': 5e-05, 'epoch': 0.52}

 10%|█         | 2530/24420 [1:02:49<8:59:08,  1.48s/it]
 10%|█         | 2531/24420 [1:02:51<8:58:41,  1.48s/it]
 10%|█         | 2532/24420 [1:02:52<8:59:42,  1.48s/it]
 10%|█         | 2533/24420 [1:02:54<8:59:34,  1.48s/it]
 10%|█         | 2534/24420 [1:02:55<8:58:56,  1.48s/it]
 10%|█         | 2535/24420 [1:02:57<8:58:24,  1.48s/it]
                                                        
{'loss': 0.2482, 'learning_rate': 5e-05, 'epoch': 0.52}

 10%|█         | 2535/24420 [1:02:57<8:58:24,  1.48s/it]
 10%|█         | 2536/24420 [1:02:58<8:57:35,  1.47s/it]
 10%|█         | 2537/24420 [1:02:59<8:57:28,  1.47s/it]
 10%|█         | 2538/24420 [1:03:01<8:57:58,  1.48s/it]
 10%|█         | 2539/24420 [1:03:02<8:57:42,  1.47s/it]
 10%|█         | 2540/24420 [1:03:04<8:57:36,  1.47s/it]
                                                        
{'loss': 0.2018, 'learning_rate': 5e-05, 'epoch': 0.52}

 10%|█         | 2540/24420 [1:03:04<8:57:36,  1.47s/it]
 10%|█         | 2541/24420 [1:03:05<8:59:04,  1.48s/it]
 10%|█         | 2542/24420 [1:03:07<8:58:51,  1.48s/it]
 10%|█         | 2543/24420 [1:03:08<8:58:18,  1.48s/it]
 10%|█         | 2544/24420 [1:03:10<8:58:02,  1.48s/it]
 10%|█         | 2545/24420 [1:03:11<8:57:22,  1.47s/it]
                                                        
{'loss': 0.1971, 'learning_rate': 5e-05, 'epoch': 0.52}

 10%|█         | 2545/24420 [1:03:11<8:57:22,  1.47s/it]
 10%|█         | 2546/24420 [1:03:13<8:57:49,  1.48s/it]
 10%|█         | 2547/24420 [1:03:14<8:57:47,  1.48s/it]
 10%|█         | 2548/24420 [1:03:16<8:57:00,  1.47s/it]
 10%|█         | 2549/24420 [1:03:17<8:56:55,  1.47s/it]
 10%|█         | 2550/24420 [1:03:19<8:56:19,  1.47s/it]
                                                        
{'loss': 0.2121, 'learning_rate': 5e-05, 'epoch': 0.52}

 10%|█         | 2550/24420 [1:03:19<8:56:19,  1.47s/it]
 10%|█         | 2551/24420 [1:03:20<8:56:47,  1.47s/it]
 10%|█         | 2552/24420 [1:03:22<8:56:08,  1.47s/it]
 10%|█         | 2553/24420 [1:03:23<8:56:05,  1.47s/it]
 10%|█         | 2554/24420 [1:03:25<8:56:35,  1.47s/it]
 10%|█         | 2555/24420 [1:03:26<8:56:22,  1.47s/it]
                                                        
{'loss': 0.192, 'learning_rate': 5e-05, 'epoch': 0.52}

 10%|█         | 2555/24420 [1:03:26<8:56:22,  1.47s/it]
 10%|█         | 2556/24420 [1:03:27<8:57:04,  1.47s/it]
 10%|█         | 2557/24420 [1:03:29<8:57:00,  1.47s/it]
 10%|█         | 2558/24420 [1:03:30<8:56:54,  1.47s/it]
 10%|█         | 2559/24420 [1:03:32<8:55:46,  1.47s/it]
 10%|█         | 2560/24420 [1:03:33<8:56:34,  1.47s/it]
                                                        
{'loss': 0.1714, 'learning_rate': 5e-05, 'epoch': 0.52}

 10%|█         | 2560/24420 [1:03:33<8:56:34,  1.47s/it]
 10%|█         | 2561/24420 [1:03:35<8:58:07,  1.48s/it]
 10%|█         | 2562/24420 [1:03:36<8:57:51,  1.48s/it]
 10%|█         | 2563/24420 [1:03:38<8:57:37,  1.48s/it]
 10%|█         | 2564/24420 [1:03:39<8:57:29,  1.48s/it]
 11%|█         | 2565/24420 [1:03:41<8:57:27,  1.48s/it]
                                                        
{'loss': 0.2039, 'learning_rate': 5e-05, 'epoch': 0.53}

 11%|█         | 2565/24420 [1:03:41<8:57:27,  1.48s/it]
 11%|█         | 2566/24420 [1:03:42<8:56:47,  1.47s/it]
 11%|█         | 2567/24420 [1:03:44<8:57:31,  1.48s/it]
 11%|█         | 2568/24420 [1:03:45<8:58:01,  1.48s/it]
 11%|█         | 2569/24420 [1:03:47<8:58:20,  1.48s/it]
 11%|█         | 2570/24420 [1:03:48<8:57:38,  1.48s/it]
                                                        
{'loss': 0.2117, 'learning_rate': 5e-05, 'epoch': 0.53}

 11%|█         | 2570/24420 [1:03:48<8:57:38,  1.48s/it]
 11%|█         | 2571/24420 [1:03:50<8:57:45,  1.48s/it]
 11%|█         | 2572/24420 [1:03:51<8:58:17,  1.48s/it]
 11%|█         | 2573/24420 [1:03:53<8:57:48,  1.48s/it]
 11%|█         | 2574/24420 [1:03:54<8:57:41,  1.48s/it]
 11%|█         | 2575/24420 [1:03:56<8:58:04,  1.48s/it]
                                                        
{'loss': 0.1964, 'learning_rate': 5e-05, 'epoch': 0.53}

 11%|█         | 2575/24420 [1:03:56<8:58:04,  1.48s/it]
 11%|█         | 2576/24420 [1:03:57<8:59:40,  1.48s/it]
 11%|█         | 2577/24420 [1:03:58<8:59:17,  1.48s/it]
 11%|█         | 2578/24420 [1:04:00<8:58:40,  1.48s/it]
 11%|█         | 2579/24420 [1:04:01<8:58:14,  1.48s/it]
 11%|█         | 2580/24420 [1:04:03<8:57:29,  1.48s/it]
                                                        
{'loss': 0.202, 'learning_rate': 5e-05, 'epoch': 0.53}

 11%|█         | 2580/24420 [1:04:03<8:57:29,  1.48s/it]
 11%|█         | 2581/24420 [1:04:04<9:04:04,  1.49s/it]
 11%|█         | 2582/24420 [1:04:06<9:02:35,  1.49s/it]
 11%|█         | 2583/24420 [1:04:07<9:01:00,  1.49s/it]
 11%|█         | 2584/24420 [1:04:09<8:59:23,  1.48s/it]
 11%|█         | 2585/24420 [1:04:10<8:57:37,  1.48s/it]
                                                        
{'loss': 0.2076, 'learning_rate': 5e-05, 'epoch': 0.53}

 11%|█         | 2585/24420 [1:04:10<8:57:37,  1.48s/it]
 11%|█         | 2586/24420 [1:04:12<8:57:06,  1.48s/it]
 11%|█         | 2587/24420 [1:04:13<8:56:14,  1.47s/it]
 11%|█         | 2588/24420 [1:04:15<8:55:47,  1.47s/it]
 11%|█         | 2589/24420 [1:04:16<8:55:34,  1.47s/it]
 11%|█         | 2590/24420 [1:04:18<8:55:21,  1.47s/it]
                                                        
{'loss': 0.2033, 'learning_rate': 5e-05, 'epoch': 0.53}

 11%|█         | 2590/24420 [1:04:18<8:55:21,  1.47s/it]
 11%|█         | 2591/24420 [1:04:19<8:55:37,  1.47s/it]
 11%|█         | 2592/24420 [1:04:21<8:55:38,  1.47s/it]
 11%|█         | 2593/24420 [1:04:22<8:56:13,  1.47s/it]
 11%|█         | 2594/24420 [1:04:24<8:55:43,  1.47s/it]
 11%|█         | 2595/24420 [1:04:25<8:55:52,  1.47s/it]
                                                        
{'loss': 0.1841, 'learning_rate': 5e-05, 'epoch': 0.53}

 11%|█         | 2595/24420 [1:04:25<8:55:52,  1.47s/it]
 11%|█         | 2596/24420 [1:04:27<8:55:49,  1.47s/it]
 11%|█         | 2597/24420 [1:04:28<8:56:16,  1.47s/it]
 11%|█         | 2598/24420 [1:04:29<8:55:45,  1.47s/it]
 11%|█         | 2599/24420 [1:04:31<8:55:42,  1.47s/it]
 11%|█         | 2600/24420 [1:04:32<8:55:58,  1.47s/it]
                                                        
{'loss': 0.2035, 'learning_rate': 5e-05, 'epoch': 0.53}

 11%|█         | 2600/24420 [1:04:32<8:55:58,  1.47s/it]
 11%|█         | 2601/24420 [1:04:37<14:47:53,  2.44s/it]
 11%|█         | 2602/24420 [1:04:39<13:02:23,  2.15s/it]
 11%|█         | 2603/24420 [1:04:40<11:48:19,  1.95s/it]
 11%|█         | 2604/24420 [1:04:42<10:55:43,  1.80s/it]
 11%|█         | 2605/24420 [1:04:43<10:18:34,  1.70s/it]
                                                         
{'loss': 0.1899, 'learning_rate': 5e-05, 'epoch': 0.53}

 11%|█         | 2605/24420 [1:04:43<10:18:34,  1.70s/it]
 11%|█         | 2606/24420 [1:04:44<9:53:35,  1.63s/it] 
 11%|█         | 2607/24420 [1:04:46<9:35:42,  1.58s/it]
 11%|█         | 2608/24420 [1:04:47<9:23:11,  1.55s/it]
 11%|█         | 2609/24420 [1:04:49<9:14:57,  1.53s/it]
 11%|█         | 2610/24420 [1:04:50<9:10:40,  1.51s/it]
                                                        
{'loss': 0.1965, 'learning_rate': 5e-05, 'epoch': 0.53}

 11%|█         | 2610/24420 [1:04:50<9:10:40,  1.51s/it]
 11%|█         | 2611/24420 [1:04:52<9:05:55,  1.50s/it]
 11%|█         | 2612/24420 [1:04:53<9:02:46,  1.49s/it]
 11%|█         | 2613/24420 [1:04:55<9:01:16,  1.49s/it]
 11%|█         | 2614/24420 [1:04:56<8:59:25,  1.48s/it]
 11%|█         | 2615/24420 [1:04:58<8:58:26,  1.48s/it]
                                                        
{'loss': 0.2158, 'learning_rate': 5e-05, 'epoch': 0.54}

 11%|█         | 2615/24420 [1:04:58<8:58:26,  1.48s/it]
 11%|█         | 2616/24420 [1:04:59<8:57:30,  1.48s/it]
 11%|█         | 2617/24420 [1:05:01<8:57:16,  1.48s/it]
 11%|█         | 2618/24420 [1:05:02<8:56:46,  1.48s/it]
 11%|█         | 2619/24420 [1:05:04<8:56:16,  1.48s/it]
 11%|█         | 2620/24420 [1:05:05<8:55:42,  1.47s/it]
                                                        
{'loss': 0.2032, 'learning_rate': 5e-05, 'epoch': 0.54}

 11%|█         | 2620/24420 [1:05:05<8:55:42,  1.47s/it]
 11%|█         | 2621/24420 [1:05:07<8:56:02,  1.48s/it]
 11%|█         | 2622/24420 [1:05:08<8:56:08,  1.48s/it]
 11%|█         | 2623/24420 [1:05:10<8:56:15,  1.48s/it]
 11%|█         | 2624/24420 [1:05:11<8:55:59,  1.48s/it]
 11%|█         | 2625/24420 [1:05:13<9:05:42,  1.50s/it]
                                                        
{'loss': 0.1938, 'learning_rate': 5e-05, 'epoch': 0.54}

 11%|█         | 2625/24420 [1:05:13<9:05:42,  1.50s/it]
 11%|█         | 2626/24420 [1:05:14<9:02:13,  1.49s/it]
 11%|█         | 2627/24420 [1:05:16<8:59:17,  1.48s/it]
 11%|█         | 2628/24420 [1:05:17<8:57:21,  1.48s/it]
 11%|█         | 2629/24420 [1:05:18<8:56:56,  1.48s/it]
 11%|█         | 2630/24420 [1:05:20<8:56:18,  1.48s/it]
                                                        
{'loss': 0.2155, 'learning_rate': 5e-05, 'epoch': 0.54}

 11%|█         | 2630/24420 [1:05:20<8:56:18,  1.48s/it]
 11%|█         | 2631/24420 [1:05:21<8:56:07,  1.48s/it]
 11%|█         | 2632/24420 [1:05:23<8:55:27,  1.47s/it]
 11%|█         | 2633/24420 [1:05:24<8:54:48,  1.47s/it]
 11%|█         | 2634/24420 [1:05:26<8:55:59,  1.48s/it]
 11%|█         | 2635/24420 [1:05:27<8:54:41,  1.47s/it]
                                                        
{'loss': 0.2167, 'learning_rate': 5e-05, 'epoch': 0.54}

 11%|█         | 2635/24420 [1:05:27<8:54:41,  1.47s/it]
 11%|█         | 2636/24420 [1:05:29<8:55:40,  1.48s/it]
 11%|█         | 2637/24420 [1:05:30<8:55:37,  1.48s/it]
 11%|█         | 2638/24420 [1:05:32<8:55:18,  1.47s/it]
 11%|█         | 2639/24420 [1:05:33<8:54:32,  1.47s/it]
 11%|█         | 2640/24420 [1:05:35<8:55:19,  1.47s/it]
                                                        
{'loss': 0.2153, 'learning_rate': 5e-05, 'epoch': 0.54}

 11%|█         | 2640/24420 [1:05:35<8:55:19,  1.47s/it]
 11%|█         | 2641/24420 [1:05:36<8:55:50,  1.48s/it]
 11%|█         | 2642/24420 [1:05:38<8:55:04,  1.47s/it]
 11%|█         | 2643/24420 [1:05:39<8:56:27,  1.48s/it]
 11%|█         | 2644/24420 [1:05:41<8:56:15,  1.48s/it]
 11%|█         | 2645/24420 [1:05:42<8:56:03,  1.48s/it]
                                                        
{'loss': 0.2061, 'learning_rate': 5e-05, 'epoch': 0.54}

 11%|█         | 2645/24420 [1:05:42<8:56:03,  1.48s/it]
 11%|█         | 2646/24420 [1:05:44<8:55:40,  1.48s/it]
 11%|█         | 2647/24420 [1:05:45<8:54:50,  1.47s/it]
 11%|█         | 2648/24420 [1:05:46<8:55:00,  1.47s/it]
 11%|█         | 2649/24420 [1:05:48<8:54:37,  1.47s/it]
 11%|█         | 2650/24420 [1:05:49<8:54:47,  1.47s/it]
                                                        
{'loss': 0.1823, 'learning_rate': 5e-05, 'epoch': 0.54}

 11%|█         | 2650/24420 [1:05:49<8:54:47,  1.47s/it]
 11%|█         | 2651/24420 [1:05:51<8:55:35,  1.48s/it]
 11%|█         | 2652/24420 [1:05:52<8:55:04,  1.47s/it]
 11%|█         | 2653/24420 [1:05:54<8:54:17,  1.47s/it]
 11%|█         | 2654/24420 [1:05:55<8:54:19,  1.47s/it]
 11%|█         | 2655/24420 [1:05:57<8:54:39,  1.47s/it]
                                                        
{'loss': 0.2436, 'learning_rate': 5e-05, 'epoch': 0.54}

 11%|█         | 2655/24420 [1:05:57<8:54:39,  1.47s/it]
 11%|█         | 2656/24420 [1:05:58<8:54:34,  1.47s/it]
 11%|█         | 2657/24420 [1:06:00<8:56:14,  1.48s/it]
 11%|█         | 2658/24420 [1:06:01<8:54:34,  1.47s/it]
 11%|█         | 2659/24420 [1:06:03<8:55:29,  1.48s/it]
 11%|█         | 2660/24420 [1:06:04<8:54:58,  1.48s/it]
                                                        
{'loss': 0.1748, 'learning_rate': 5e-05, 'epoch': 0.54}

 11%|█         | 2660/24420 [1:06:04<8:54:58,  1.48s/it]
 11%|█         | 2661/24420 [1:06:06<8:55:03,  1.48s/it]
 11%|█         | 2662/24420 [1:06:07<8:54:40,  1.47s/it]
 11%|█         | 2663/24420 [1:06:09<8:54:47,  1.47s/it]
 11%|█         | 2664/24420 [1:06:10<8:55:35,  1.48s/it]
 11%|█         | 2665/24420 [1:06:12<8:55:34,  1.48s/it]
                                                        
{'loss': 0.1882, 'learning_rate': 5e-05, 'epoch': 0.55}

 11%|█         | 2665/24420 [1:06:12<8:55:34,  1.48s/it]
 11%|█         | 2666/24420 [1:06:13<8:55:29,  1.48s/it]
 11%|█         | 2667/24420 [1:06:15<8:54:37,  1.47s/it]
 11%|█         | 2668/24420 [1:06:16<8:54:34,  1.47s/it]
 11%|█         | 2669/24420 [1:06:17<8:54:24,  1.47s/it]
 11%|█         | 2670/24420 [1:06:19<8:55:25,  1.48s/it]
                                                        
{'loss': 0.2371, 'learning_rate': 5e-05, 'epoch': 0.55}

 11%|█         | 2670/24420 [1:06:19<8:55:25,  1.48s/it]
 11%|█         | 2671/24420 [1:06:20<8:57:17,  1.48s/it]
 11%|█         | 2672/24420 [1:06:22<8:57:19,  1.48s/it]
 11%|█         | 2673/24420 [1:06:23<8:56:45,  1.48s/it]
 11%|█         | 2674/24420 [1:06:25<8:55:25,  1.48s/it]
 11%|█         | 2675/24420 [1:06:26<8:55:09,  1.48s/it]
                                                        
{'loss': 0.1905, 'learning_rate': 5e-05, 'epoch': 0.55}

 11%|█         | 2675/24420 [1:06:26<8:55:09,  1.48s/it]
 11%|█         | 2676/24420 [1:06:28<8:55:42,  1.48s/it]
 11%|█         | 2677/24420 [1:06:29<8:55:58,  1.48s/it]
 11%|█         | 2678/24420 [1:06:31<8:56:40,  1.48s/it]
 11%|█         | 2679/24420 [1:06:32<8:56:11,  1.48s/it]
 11%|█         | 2680/24420 [1:06:34<8:55:07,  1.48s/it]
                                                        
{'loss': 0.1816, 'learning_rate': 5e-05, 'epoch': 0.55}

 11%|█         | 2680/24420 [1:06:34<8:55:07,  1.48s/it]
 11%|█         | 2681/24420 [1:06:35<8:55:33,  1.48s/it]
 11%|█         | 2682/24420 [1:06:37<8:55:25,  1.48s/it]
 11%|█         | 2683/24420 [1:06:38<8:55:27,  1.48s/it]
 11%|█         | 2684/24420 [1:06:40<8:55:18,  1.48s/it]
 11%|█         | 2685/24420 [1:06:41<8:55:19,  1.48s/it]
                                                        
{'loss': 0.1974, 'learning_rate': 5e-05, 'epoch': 0.55}

 11%|█         | 2685/24420 [1:06:41<8:55:19,  1.48s/it]
 11%|█         | 2686/24420 [1:06:43<8:57:11,  1.48s/it]
 11%|█         | 2687/24420 [1:06:44<8:57:28,  1.48s/it]
 11%|█         | 2688/24420 [1:06:46<8:57:23,  1.48s/it]
 11%|█         | 2689/24420 [1:06:47<8:55:53,  1.48s/it]
 11%|█         | 2690/24420 [1:06:49<8:55:33,  1.48s/it]
                                                        
{'loss': 0.1931, 'learning_rate': 5e-05, 'epoch': 0.55}

 11%|█         | 2690/24420 [1:06:49<8:55:33,  1.48s/it]
 11%|█         | 2691/24420 [1:06:50<8:55:15,  1.48s/it]
 11%|█         | 2692/24420 [1:06:52<8:54:53,  1.48s/it]
 11%|█         | 2693/24420 [1:06:53<8:55:03,  1.48s/it]
 11%|█         | 2694/24420 [1:06:54<8:54:34,  1.48s/it]
 11%|█         | 2695/24420 [1:06:56<8:54:37,  1.48s/it]
                                                        
{'loss': 0.1919, 'learning_rate': 5e-05, 'epoch': 0.55}

 11%|█         | 2695/24420 [1:06:56<8:54:37,  1.48s/it]
 11%|█         | 2696/24420 [1:06:57<8:54:29,  1.48s/it]
 11%|█         | 2697/24420 [1:06:59<8:54:27,  1.48s/it]
 11%|█         | 2698/24420 [1:07:00<8:53:44,  1.47s/it]
 11%|█         | 2699/24420 [1:07:02<8:53:38,  1.47s/it]
 11%|█         | 2700/24420 [1:07:03<8:54:46,  1.48s/it]
                                                        
{'loss': 0.1804, 'learning_rate': 5e-05, 'epoch': 0.55}

 11%|█         | 2700/24420 [1:07:03<8:54:46,  1.48s/it]
 11%|█         | 2701/24420 [1:07:05<8:54:31,  1.48s/it]
 11%|█         | 2702/24420 [1:07:06<8:54:56,  1.48s/it]
 11%|█         | 2703/24420 [1:07:08<8:54:27,  1.48s/it]
 11%|█         | 2704/24420 [1:07:09<8:54:19,  1.48s/it]
 11%|█         | 2705/24420 [1:07:11<8:54:26,  1.48s/it]
                                                        
{'loss': 0.1884, 'learning_rate': 5e-05, 'epoch': 0.55}

 11%|█         | 2705/24420 [1:07:11<8:54:26,  1.48s/it]
 11%|█         | 2706/24420 [1:07:12<8:54:56,  1.48s/it]
 11%|█         | 2707/24420 [1:07:14<8:55:38,  1.48s/it]
 11%|█         | 2708/24420 [1:07:15<9:01:45,  1.50s/it]
 11%|█         | 2709/24420 [1:07:17<9:00:28,  1.49s/it]
 11%|█         | 2710/24420 [1:07:18<8:58:16,  1.49s/it]
                                                        
{'loss': 0.2267, 'learning_rate': 5e-05, 'epoch': 0.55}

 11%|█         | 2710/24420 [1:07:18<8:58:16,  1.49s/it]
 11%|█         | 2711/24420 [1:07:20<8:56:57,  1.48s/it]
 11%|█         | 2712/24420 [1:07:21<8:55:13,  1.48s/it]
 11%|█         | 2713/24420 [1:07:23<8:55:16,  1.48s/it]
 11%|█         | 2714/24420 [1:07:24<8:55:39,  1.48s/it]
 11%|█         | 2715/24420 [1:07:26<8:54:58,  1.48s/it]
                                                        
{'loss': 0.1781, 'learning_rate': 5e-05, 'epoch': 0.56}

 11%|█         | 2715/24420 [1:07:26<8:54:58,  1.48s/it]
 11%|█         | 2716/24420 [1:07:27<8:55:03,  1.48s/it]
 11%|█         | 2717/24420 [1:07:29<8:56:13,  1.48s/it]
 11%|█         | 2718/24420 [1:07:30<8:55:39,  1.48s/it]
 11%|█         | 2719/24420 [1:07:31<8:55:04,  1.48s/it]
 11%|█         | 2720/24420 [1:07:33<8:54:42,  1.48s/it]
                                                        
{'loss': 0.1879, 'learning_rate': 5e-05, 'epoch': 0.56}

 11%|█         | 2720/24420 [1:07:33<8:54:42,  1.48s/it]
 11%|█         | 2721/24420 [1:07:34<8:54:34,  1.48s/it]
 11%|█         | 2722/24420 [1:07:36<8:54:13,  1.48s/it]
 11%|█         | 2723/24420 [1:07:37<8:53:43,  1.48s/it]
 11%|█         | 2724/24420 [1:07:39<8:53:04,  1.47s/it]
 11%|█         | 2725/24420 [1:07:40<8:52:22,  1.47s/it]
                                                        
{'loss': 0.1758, 'learning_rate': 5e-05, 'epoch': 0.56}

 11%|█         | 2725/24420 [1:07:40<8:52:22,  1.47s/it]
 11%|█         | 2726/24420 [1:07:42<8:52:58,  1.47s/it]
 11%|█         | 2727/24420 [1:07:43<8:53:05,  1.47s/it]
 11%|█         | 2728/24420 [1:07:45<8:53:53,  1.48s/it]
 11%|█         | 2729/24420 [1:07:46<8:54:18,  1.48s/it]
 11%|█         | 2730/24420 [1:07:48<8:53:40,  1.48s/it]
                                                        
{'loss': 0.2057, 'learning_rate': 5e-05, 'epoch': 0.56}

 11%|█         | 2730/24420 [1:07:48<8:53:40,  1.48s/it]
 11%|█         | 2731/24420 [1:07:49<8:54:47,  1.48s/it]
 11%|█         | 2732/24420 [1:07:51<8:54:03,  1.48s/it]
 11%|█         | 2733/24420 [1:07:52<8:53:21,  1.48s/it]
 11%|█         | 2734/24420 [1:07:54<8:52:28,  1.47s/it]
 11%|█         | 2735/24420 [1:07:55<8:53:03,  1.47s/it]
                                                        
{'loss': 0.2455, 'learning_rate': 5e-05, 'epoch': 0.56}

 11%|█         | 2735/24420 [1:07:55<8:53:03,  1.47s/it]
 11%|█         | 2736/24420 [1:07:57<8:52:35,  1.47s/it]
 11%|█         | 2737/24420 [1:07:58<8:53:31,  1.48s/it]
 11%|█         | 2738/24420 [1:07:59<8:53:14,  1.48s/it]
 11%|█         | 2739/24420 [1:08:01<8:52:31,  1.47s/it]
 11%|█         | 2740/24420 [1:08:02<8:52:49,  1.47s/it]
                                                        
{'loss': 0.225, 'learning_rate': 5e-05, 'epoch': 0.56}

 11%|█         | 2740/24420 [1:08:02<8:52:49,  1.47s/it]
 11%|█         | 2741/24420 [1:08:04<8:53:13,  1.48s/it]
 11%|█         | 2742/24420 [1:08:05<8:53:53,  1.48s/it]
 11%|█         | 2743/24420 [1:08:07<8:53:33,  1.48s/it]
 11%|█         | 2744/24420 [1:08:08<8:52:50,  1.47s/it]
 11%|█         | 2745/24420 [1:08:10<8:53:50,  1.48s/it]
                                                        
{'loss': 0.1925, 'learning_rate': 5e-05, 'epoch': 0.56}

 11%|█         | 2745/24420 [1:08:10<8:53:50,  1.48s/it]
 11%|█         | 2746/24420 [1:08:11<8:53:51,  1.48s/it]
 11%|█         | 2747/24420 [1:08:13<8:54:33,  1.48s/it]
 11%|█▏        | 2748/24420 [1:08:14<8:54:47,  1.48s/it]
 11%|█▏        | 2749/24420 [1:08:16<8:54:04,  1.48s/it]
 11%|█▏        | 2750/24420 [1:08:17<8:53:52,  1.48s/it]
                                                        
{'loss': 0.1949, 'learning_rate': 5e-05, 'epoch': 0.56}

 11%|█▏        | 2750/24420 [1:08:17<8:53:52,  1.48s/it]
 11%|█▏        | 2751/24420 [1:08:19<8:53:18,  1.48s/it]
 11%|█▏        | 2752/24420 [1:08:20<8:53:34,  1.48s/it]
 11%|█▏        | 2753/24420 [1:08:22<8:53:59,  1.48s/it]
 11%|█▏        | 2754/24420 [1:08:23<8:53:19,  1.48s/it]
 11%|█▏        | 2755/24420 [1:08:25<8:52:22,  1.47s/it]
                                                        
{'loss': 0.1673, 'learning_rate': 5e-05, 'epoch': 0.56}

 11%|█▏        | 2755/24420 [1:08:25<8:52:22,  1.47s/it]
 11%|█▏        | 2756/24420 [1:08:26<8:53:07,  1.48s/it]
 11%|█▏        | 2757/24420 [1:08:28<8:53:25,  1.48s/it]
 11%|█▏        | 2758/24420 [1:08:29<8:53:13,  1.48s/it]
 11%|█▏        | 2759/24420 [1:08:31<8:52:47,  1.48s/it]
 11%|█▏        | 2760/24420 [1:08:32<8:52:56,  1.48s/it]
                                                        
{'loss': 0.213, 'learning_rate': 5e-05, 'epoch': 0.57}

 11%|█▏        | 2760/24420 [1:08:32<8:52:56,  1.48s/it]
 11%|█▏        | 2761/24420 [1:08:33<8:54:04,  1.48s/it]
 11%|█▏        | 2762/24420 [1:08:35<8:54:03,  1.48s/it]
 11%|█▏        | 2763/24420 [1:08:36<8:53:08,  1.48s/it]
 11%|█▏        | 2764/24420 [1:08:38<8:53:27,  1.48s/it]
 11%|█▏        | 2765/24420 [1:08:39<8:54:16,  1.48s/it]
                                                        
{'loss': 0.2136, 'learning_rate': 5e-05, 'epoch': 0.57}

 11%|█▏        | 2765/24420 [1:08:39<8:54:16,  1.48s/it]
 11%|█▏        | 2766/24420 [1:08:41<8:53:57,  1.48s/it]
 11%|█▏        | 2767/24420 [1:08:42<8:53:30,  1.48s/it]
 11%|█▏        | 2768/24420 [1:08:44<8:54:24,  1.48s/it]
 11%|█▏        | 2769/24420 [1:08:45<8:53:39,  1.48s/it]
 11%|█▏        | 2770/24420 [1:08:47<8:53:26,  1.48s/it]
                                                        
{'loss': 0.1728, 'learning_rate': 5e-05, 'epoch': 0.57}

 11%|█▏        | 2770/24420 [1:08:47<8:53:26,  1.48s/it]
 11%|█▏        | 2771/24420 [1:08:48<8:54:15,  1.48s/it]
 11%|█▏        | 2772/24420 [1:08:50<8:53:04,  1.48s/it]
 11%|█▏        | 2773/24420 [1:08:51<8:53:05,  1.48s/it]
 11%|█▏        | 2774/24420 [1:08:53<8:52:26,  1.48s/it]
 11%|█▏        | 2775/24420 [1:08:54<8:52:43,  1.48s/it]
                                                        
{'loss': 0.1819, 'learning_rate': 5e-05, 'epoch': 0.57}

 11%|█▏        | 2775/24420 [1:08:54<8:52:43,  1.48s/it]
 11%|█▏        | 2776/24420 [1:08:56<8:52:51,  1.48s/it]
 11%|█▏        | 2777/24420 [1:08:57<8:52:20,  1.48s/it]
 11%|█▏        | 2778/24420 [1:08:59<8:52:09,  1.48s/it]
 11%|█▏        | 2779/24420 [1:09:00<8:52:46,  1.48s/it]
 11%|█▏        | 2780/24420 [1:09:02<8:51:37,  1.47s/it]
                                                        
{'loss': 0.2287, 'learning_rate': 5e-05, 'epoch': 0.57}

 11%|█▏        | 2780/24420 [1:09:02<8:51:37,  1.47s/it]
 11%|█▏        | 2781/24420 [1:09:03<8:51:30,  1.47s/it]
 11%|█▏        | 2782/24420 [1:09:04<8:51:41,  1.47s/it]
 11%|█▏        | 2783/24420 [1:09:06<8:51:27,  1.47s/it]
 11%|█▏        | 2784/24420 [1:09:07<8:51:36,  1.47s/it]
 11%|█▏        | 2785/24420 [1:09:09<8:50:57,  1.47s/it]
                                                        
{'loss': 0.1954, 'learning_rate': 5e-05, 'epoch': 0.57}

 11%|█▏        | 2785/24420 [1:09:09<8:50:57,  1.47s/it]
 11%|█▏        | 2786/24420 [1:09:10<8:51:21,  1.47s/it]
 11%|█▏        | 2787/24420 [1:09:12<8:52:19,  1.48s/it]
 11%|█▏        | 2788/24420 [1:09:13<8:51:49,  1.48s/it]
 11%|█▏        | 2789/24420 [1:09:15<8:51:03,  1.47s/it]
 11%|█▏        | 2790/24420 [1:09:16<8:51:01,  1.47s/it]
                                                        
{'loss': 0.1893, 'learning_rate': 5e-05, 'epoch': 0.57}

 11%|█▏        | 2790/24420 [1:09:16<8:51:01,  1.47s/it]
 11%|█▏        | 2791/24420 [1:09:18<8:51:30,  1.47s/it]
 11%|█▏        | 2792/24420 [1:09:19<8:50:56,  1.47s/it]
 11%|█▏        | 2793/24420 [1:09:21<8:51:08,  1.47s/it]
 11%|█▏        | 2794/24420 [1:09:22<8:51:39,  1.48s/it]
 11%|█▏        | 2795/24420 [1:09:24<8:51:28,  1.47s/it]
                                                        
{'loss': 0.198, 'learning_rate': 5e-05, 'epoch': 0.57}

 11%|█▏        | 2795/24420 [1:09:24<8:51:28,  1.47s/it]
 11%|█▏        | 2796/24420 [1:09:25<8:51:22,  1.47s/it]
 11%|█▏        | 2797/24420 [1:09:27<8:51:13,  1.47s/it]
 11%|█▏        | 2798/24420 [1:09:28<8:51:36,  1.48s/it]
 11%|█▏        | 2799/24420 [1:09:30<8:50:52,  1.47s/it]
 11%|█▏        | 2800/24420 [1:09:31<8:50:54,  1.47s/it]
                                                        
{'loss': 0.214, 'learning_rate': 5e-05, 'epoch': 0.57}

 11%|█▏        | 2800/24420 [1:09:31<8:50:54,  1.47s/it]
 11%|█▏        | 2801/24420 [1:09:36<14:41:37,  2.45s/it]
 11%|█▏        | 2802/24420 [1:09:37<12:56:14,  2.15s/it]
 11%|█▏        | 2803/24420 [1:09:39<11:41:49,  1.95s/it]
 11%|█▏        | 2804/24420 [1:09:40<10:50:39,  1.81s/it]
 11%|█▏        | 2805/24420 [1:09:42<10:15:47,  1.71s/it]
                                                         
{'loss': 0.235, 'learning_rate': 5e-05, 'epoch': 0.57}

 11%|█▏        | 2805/24420 [1:09:42<10:15:47,  1.71s/it]
 11%|█▏        | 2806/24420 [1:09:43<9:51:18,  1.64s/it] 
 11%|█▏        | 2807/24420 [1:09:45<9:33:12,  1.59s/it]
 11%|█▏        | 2808/24420 [1:09:46<9:20:15,  1.56s/it]
 12%|█▏        | 2809/24420 [1:09:48<9:11:38,  1.53s/it]
 12%|█▏        | 2810/24420 [1:09:49<9:04:54,  1.51s/it]
                                                        
{'loss': 0.1768, 'learning_rate': 5e-05, 'epoch': 0.58}

 12%|█▏        | 2810/24420 [1:09:49<9:04:54,  1.51s/it]
 12%|█▏        | 2811/24420 [1:09:50<9:00:33,  1.50s/it]
 12%|█▏        | 2812/24420 [1:09:52<8:58:39,  1.50s/it]
 12%|█▏        | 2813/24420 [1:09:53<8:56:52,  1.49s/it]
 12%|█▏        | 2814/24420 [1:09:55<8:55:30,  1.49s/it]
 12%|█▏        | 2815/24420 [1:09:56<8:54:24,  1.48s/it]
                                                        
{'loss': 0.185, 'learning_rate': 5e-05, 'epoch': 0.58}

 12%|█▏        | 2815/24420 [1:09:56<8:54:24,  1.48s/it]
 12%|█▏        | 2816/24420 [1:09:58<8:53:29,  1.48s/it]
 12%|█▏        | 2817/24420 [1:09:59<8:52:42,  1.48s/it]
 12%|█▏        | 2818/24420 [1:10:01<8:51:26,  1.48s/it]
 12%|█▏        | 2819/24420 [1:10:02<8:50:31,  1.47s/it]
 12%|█▏        | 2820/24420 [1:10:04<8:50:05,  1.47s/it]
                                                        
{'loss': 0.185, 'learning_rate': 5e-05, 'epoch': 0.58}

 12%|█▏        | 2820/24420 [1:10:04<8:50:05,  1.47s/it]
 12%|█▏        | 2821/24420 [1:10:05<8:50:12,  1.47s/it]
 12%|█▏        | 2822/24420 [1:10:07<8:50:38,  1.47s/it]
 12%|█▏        | 2823/24420 [1:10:08<8:50:07,  1.47s/it]
 12%|█▏        | 2824/24420 [1:10:10<8:50:22,  1.47s/it]
 12%|█▏        | 2825/24420 [1:10:11<8:50:49,  1.47s/it]
                                                        
{'loss': 0.1827, 'learning_rate': 5e-05, 'epoch': 0.58}

 12%|█▏        | 2825/24420 [1:10:11<8:50:49,  1.47s/it]
 12%|█▏        | 2826/24420 [1:10:13<8:49:58,  1.47s/it]
 12%|█▏        | 2827/24420 [1:10:14<8:50:24,  1.47s/it]
 12%|█▏        | 2828/24420 [1:10:16<8:49:45,  1.47s/it]
 12%|█▏        | 2829/24420 [1:10:17<8:50:15,  1.47s/it]
 12%|█▏        | 2830/24420 [1:10:19<8:50:30,  1.47s/it]
                                                        
{'loss': 0.158, 'learning_rate': 5e-05, 'epoch': 0.58}

 12%|█▏        | 2830/24420 [1:10:19<8:50:30,  1.47s/it]
 12%|█▏        | 2831/24420 [1:10:20<8:50:36,  1.47s/it]
 12%|█▏        | 2832/24420 [1:10:21<8:50:33,  1.47s/it]
 12%|█▏        | 2833/24420 [1:10:23<8:50:49,  1.48s/it]
 12%|█▏        | 2834/24420 [1:10:24<8:49:53,  1.47s/it]
 12%|█▏        | 2835/24420 [1:10:26<8:49:41,  1.47s/it]
                                                        
{'loss': 0.1609, 'learning_rate': 5e-05, 'epoch': 0.58}

 12%|█▏        | 2835/24420 [1:10:26<8:49:41,  1.47s/it]
 12%|█▏        | 2836/24420 [1:10:27<8:49:23,  1.47s/it]
 12%|█▏        | 2837/24420 [1:10:29<8:50:05,  1.47s/it]
 12%|█▏        | 2838/24420 [1:10:30<8:50:49,  1.48s/it]
 12%|█▏        | 2839/24420 [1:10:32<8:50:09,  1.47s/it]
 12%|█▏        | 2840/24420 [1:10:33<8:50:26,  1.47s/it]
                                                        
{'loss': 0.1884, 'learning_rate': 5e-05, 'epoch': 0.58}

 12%|█▏        | 2840/24420 [1:10:33<8:50:26,  1.47s/it]
 12%|█▏        | 2841/24420 [1:10:35<8:49:38,  1.47s/it]
 12%|█▏        | 2842/24420 [1:10:36<8:49:41,  1.47s/it]
 12%|█▏        | 2843/24420 [1:10:38<8:49:56,  1.47s/it]
 12%|█▏        | 2844/24420 [1:10:39<8:50:04,  1.47s/it]
 12%|█▏        | 2845/24420 [1:10:41<8:50:49,  1.48s/it]
                                                        
{'loss': 0.204, 'learning_rate': 5e-05, 'epoch': 0.58}

 12%|█▏        | 2845/24420 [1:10:41<8:50:49,  1.48s/it]
 12%|█▏        | 2846/24420 [1:10:42<8:50:11,  1.47s/it]
 12%|█▏        | 2847/24420 [1:10:44<8:49:59,  1.47s/it]
 12%|█▏        | 2848/24420 [1:10:45<8:52:15,  1.48s/it]
 12%|█▏        | 2849/24420 [1:10:47<8:51:35,  1.48s/it]
 12%|█▏        | 2850/24420 [1:10:48<8:51:22,  1.48s/it]
                                                        
{'loss': 0.1896, 'learning_rate': 5e-05, 'epoch': 0.58}

 12%|█▏        | 2850/24420 [1:10:48<8:51:22,  1.48s/it]
 12%|█▏        | 2851/24420 [1:10:49<8:51:20,  1.48s/it]
 12%|█▏        | 2852/24420 [1:10:51<8:50:45,  1.48s/it]
 12%|█▏        | 2853/24420 [1:10:52<8:50:43,  1.48s/it]
 12%|█▏        | 2854/24420 [1:10:54<8:50:05,  1.47s/it]
 12%|█▏        | 2855/24420 [1:10:55<8:50:09,  1.48s/it]
                                                        
{'loss': 0.2387, 'learning_rate': 5e-05, 'epoch': 0.58}

 12%|█▏        | 2855/24420 [1:10:55<8:50:09,  1.48s/it]
 12%|█▏        | 2856/24420 [1:10:57<8:50:29,  1.48s/it]
 12%|█▏        | 2857/24420 [1:10:58<8:50:18,  1.48s/it]
 12%|█▏        | 2858/24420 [1:11:00<8:50:27,  1.48s/it]
 12%|█▏        | 2859/24420 [1:11:01<8:50:29,  1.48s/it]
 12%|█▏        | 2860/24420 [1:11:03<8:50:06,  1.48s/it]
                                                        
{'loss': 0.1803, 'learning_rate': 5e-05, 'epoch': 0.59}

 12%|█▏        | 2860/24420 [1:11:03<8:50:06,  1.48s/it]
 12%|█▏        | 2861/24420 [1:11:04<8:50:24,  1.48s/it]
 12%|█▏        | 2862/24420 [1:11:06<8:50:48,  1.48s/it]
 12%|█▏        | 2863/24420 [1:11:07<8:50:39,  1.48s/it]
 12%|█▏        | 2864/24420 [1:11:09<8:50:37,  1.48s/it]
 12%|█▏        | 2865/24420 [1:11:10<8:50:36,  1.48s/it]
                                                        
{'loss': 0.1917, 'learning_rate': 5e-05, 'epoch': 0.59}

 12%|█▏        | 2865/24420 [1:11:10<8:50:36,  1.48s/it]
 12%|█▏        | 2866/24420 [1:11:12<8:50:59,  1.48s/it]
 12%|█▏        | 2867/24420 [1:11:13<8:50:08,  1.48s/it]
 12%|█▏        | 2868/24420 [1:11:15<8:49:38,  1.47s/it]
 12%|█▏        | 2869/24420 [1:11:16<8:49:21,  1.47s/it]
 12%|█▏        | 2870/24420 [1:11:18<8:49:30,  1.47s/it]
                                                        
{'loss': 0.1994, 'learning_rate': 5e-05, 'epoch': 0.59}

 12%|█▏        | 2870/24420 [1:11:18<8:49:30,  1.47s/it]
 12%|█▏        | 2871/24420 [1:11:19<8:49:38,  1.47s/it]
 12%|█▏        | 2872/24420 [1:11:20<8:49:22,  1.47s/it]
 12%|█▏        | 2873/24420 [1:11:22<8:49:14,  1.47s/it]
 12%|█▏        | 2874/24420 [1:11:23<8:49:21,  1.47s/it]
 12%|█▏        | 2875/24420 [1:11:25<8:49:56,  1.48s/it]
                                                        
{'loss': 0.1604, 'learning_rate': 5e-05, 'epoch': 0.59}

 12%|█▏        | 2875/24420 [1:11:25<8:49:56,  1.48s/it]
 12%|█▏        | 2876/24420 [1:11:26<8:50:17,  1.48s/it]
 12%|█▏        | 2877/24420 [1:11:28<8:50:12,  1.48s/it]
 12%|█▏        | 2878/24420 [1:11:29<8:50:16,  1.48s/it]
 12%|█▏        | 2879/24420 [1:11:31<8:50:26,  1.48s/it]
 12%|█▏        | 2880/24420 [1:11:32<8:49:59,  1.48s/it]
                                                        
{'loss': 0.181, 'learning_rate': 5e-05, 'epoch': 0.59}

 12%|█▏        | 2880/24420 [1:11:32<8:49:59,  1.48s/it]
 12%|█▏        | 2881/24420 [1:11:34<8:50:15,  1.48s/it]
 12%|█▏        | 2882/24420 [1:11:35<8:50:54,  1.48s/it]
 12%|█▏        | 2883/24420 [1:11:37<8:49:20,  1.47s/it]
 12%|█▏        | 2884/24420 [1:11:38<8:48:38,  1.47s/it]
 12%|█▏        | 2885/24420 [1:11:40<8:49:05,  1.47s/it]
                                                        
{'loss': 0.1846, 'learning_rate': 5e-05, 'epoch': 0.59}

 12%|█▏        | 2885/24420 [1:11:40<8:49:05,  1.47s/it]
 12%|█▏        | 2886/24420 [1:11:41<8:49:23,  1.48s/it]
 12%|█▏        | 2887/24420 [1:11:43<8:49:17,  1.47s/it]
 12%|█▏        | 2888/24420 [1:11:44<8:49:19,  1.47s/it]
 12%|█▏        | 2889/24420 [1:11:46<8:49:41,  1.48s/it]
 12%|█▏        | 2890/24420 [1:11:47<8:49:12,  1.47s/it]
                                                        
{'loss': 0.2133, 'learning_rate': 5e-05, 'epoch': 0.59}

 12%|█▏        | 2890/24420 [1:11:47<8:49:12,  1.47s/it]
 12%|█▏        | 2891/24420 [1:11:49<8:49:21,  1.48s/it]
 12%|█▏        | 2892/24420 [1:11:50<8:49:12,  1.47s/it]
 12%|█▏        | 2893/24420 [1:11:51<8:49:13,  1.48s/it]
 12%|█▏        | 2894/24420 [1:11:53<8:50:03,  1.48s/it]
 12%|█▏        | 2895/24420 [1:11:54<8:49:23,  1.48s/it]
                                                        
{'loss': 0.2052, 'learning_rate': 5e-05, 'epoch': 0.59}

 12%|█▏        | 2895/24420 [1:11:54<8:49:23,  1.48s/it]
 12%|█▏        | 2896/24420 [1:11:56<8:49:52,  1.48s/it]
 12%|█▏        | 2897/24420 [1:11:57<8:48:51,  1.47s/it]
 12%|█▏        | 2898/24420 [1:11:59<8:48:43,  1.47s/it]
 12%|█▏        | 2899/24420 [1:12:00<8:48:05,  1.47s/it]
 12%|█▏        | 2900/24420 [1:12:02<8:48:25,  1.47s/it]
                                                        
{'loss': 0.1713, 'learning_rate': 5e-05, 'epoch': 0.59}

 12%|█▏        | 2900/24420 [1:12:02<8:48:25,  1.47s/it]
 12%|█▏        | 2901/24420 [1:12:03<8:50:48,  1.48s/it]
 12%|█▏        | 2902/24420 [1:12:05<8:50:38,  1.48s/it]
 12%|█▏        | 2903/24420 [1:12:06<8:49:51,  1.48s/it]
 12%|█▏        | 2904/24420 [1:12:08<8:49:25,  1.48s/it]
 12%|█▏        | 2905/24420 [1:12:09<8:49:33,  1.48s/it]
                                                        
{'loss': 0.1929, 'learning_rate': 5e-05, 'epoch': 0.59}

 12%|█▏        | 2905/24420 [1:12:09<8:49:33,  1.48s/it]
 12%|█▏        | 2906/24420 [1:12:11<8:49:01,  1.48s/it]
 12%|█▏        | 2907/24420 [1:12:12<8:49:12,  1.48s/it]
 12%|█▏        | 2908/24420 [1:12:14<8:48:21,  1.47s/it]
 12%|█▏        | 2909/24420 [1:12:15<8:48:14,  1.47s/it]
 12%|█▏        | 2910/24420 [1:12:17<8:47:59,  1.47s/it]
                                                        
{'loss': 0.1848, 'learning_rate': 5e-05, 'epoch': 0.6}

 12%|█▏        | 2910/24420 [1:12:17<8:47:59,  1.47s/it]
 12%|█▏        | 2911/24420 [1:12:18<8:48:54,  1.48s/it]
 12%|█▏        | 2912/24420 [1:12:19<8:48:39,  1.47s/it]
 12%|█▏        | 2913/24420 [1:12:21<8:48:40,  1.47s/it]
 12%|█▏        | 2914/24420 [1:12:22<8:48:32,  1.47s/it]
 12%|█▏        | 2915/24420 [1:12:24<8:48:02,  1.47s/it]
                                                        
{'loss': 0.1948, 'learning_rate': 5e-05, 'epoch': 0.6}

 12%|█▏        | 2915/24420 [1:12:24<8:48:02,  1.47s/it]
 12%|█▏        | 2916/24420 [1:12:25<8:48:31,  1.47s/it]
 12%|█▏        | 2917/24420 [1:12:27<8:49:08,  1.48s/it]
 12%|█▏        | 2918/24420 [1:12:28<8:49:37,  1.48s/it]
 12%|█▏        | 2919/24420 [1:12:30<8:50:05,  1.48s/it]
 12%|█▏        | 2920/24420 [1:12:31<8:49:17,  1.48s/it]
                                                        
{'loss': 0.2016, 'learning_rate': 5e-05, 'epoch': 0.6}

 12%|█▏        | 2920/24420 [1:12:31<8:49:17,  1.48s/it]
 12%|█▏        | 2921/24420 [1:12:33<8:49:27,  1.48s/it]
 12%|█▏        | 2922/24420 [1:12:34<8:48:38,  1.48s/it]
 12%|█▏        | 2923/24420 [1:12:36<8:48:31,  1.48s/it]
 12%|█▏        | 2924/24420 [1:12:37<8:48:05,  1.47s/it]
 12%|█▏        | 2925/24420 [1:12:39<8:48:42,  1.48s/it]
                                                        
{'loss': 0.1938, 'learning_rate': 5e-05, 'epoch': 0.6}

 12%|█▏        | 2925/24420 [1:12:39<8:48:42,  1.48s/it]
 12%|█▏        | 2926/24420 [1:12:40<8:48:48,  1.48s/it]
 12%|█▏        | 2927/24420 [1:12:42<8:48:37,  1.48s/it]
 12%|█▏        | 2928/24420 [1:12:43<8:47:55,  1.47s/it]
 12%|█▏        | 2929/24420 [1:12:45<8:49:17,  1.48s/it]
 12%|█▏        | 2930/24420 [1:12:46<8:49:29,  1.48s/it]
                                                        
{'loss': 0.2089, 'learning_rate': 5e-05, 'epoch': 0.6}

 12%|█▏        | 2930/24420 [1:12:46<8:49:29,  1.48s/it]
 12%|█▏        | 2931/24420 [1:12:48<8:49:31,  1.48s/it]
 12%|█▏        | 2932/24420 [1:12:49<8:49:10,  1.48s/it]
 12%|█▏        | 2933/24420 [1:12:50<8:48:48,  1.48s/it]
 12%|█▏        | 2934/24420 [1:12:52<8:48:34,  1.48s/it]
 12%|█▏        | 2935/24420 [1:12:53<8:48:08,  1.47s/it]
                                                        
{'loss': 0.1804, 'learning_rate': 5e-05, 'epoch': 0.6}

 12%|█▏        | 2935/24420 [1:12:53<8:48:08,  1.47s/it]
 12%|█▏        | 2936/24420 [1:12:55<8:48:47,  1.48s/it]
 12%|█▏        | 2937/24420 [1:12:56<8:48:18,  1.48s/it]
 12%|█▏        | 2938/24420 [1:12:58<8:49:06,  1.48s/it]
 12%|█▏        | 2939/24420 [1:12:59<8:48:40,  1.48s/it]
 12%|█▏        | 2940/24420 [1:13:01<8:47:36,  1.47s/it]
                                                        
{'loss': 0.1814, 'learning_rate': 5e-05, 'epoch': 0.6}

 12%|█▏        | 2940/24420 [1:13:01<8:47:36,  1.47s/it]
 12%|█▏        | 2941/24420 [1:13:02<8:48:13,  1.48s/it]
 12%|█▏        | 2942/24420 [1:13:04<8:47:52,  1.47s/it]
 12%|█▏        | 2943/24420 [1:13:05<8:48:35,  1.48s/it]
 12%|█▏        | 2944/24420 [1:13:07<8:48:04,  1.48s/it]
 12%|█▏        | 2945/24420 [1:13:08<8:47:26,  1.47s/it]
                                                        
{'loss': 0.2017, 'learning_rate': 5e-05, 'epoch': 0.6}

 12%|█▏        | 2945/24420 [1:13:08<8:47:26,  1.47s/it]
 12%|█▏        | 2946/24420 [1:13:10<8:47:03,  1.47s/it]
 12%|█▏        | 2947/24420 [1:13:11<8:48:26,  1.48s/it]
 12%|█▏        | 2948/24420 [1:13:13<8:48:37,  1.48s/it]
 12%|█▏        | 2949/24420 [1:13:14<8:48:38,  1.48s/it]
 12%|█▏        | 2950/24420 [1:13:16<8:48:13,  1.48s/it]
                                                        
{'loss': 0.2203, 'learning_rate': 5e-05, 'epoch': 0.6}

 12%|█▏        | 2950/24420 [1:13:16<8:48:13,  1.48s/it]
 12%|█▏        | 2951/24420 [1:13:17<8:48:01,  1.48s/it]
 12%|█▏        | 2952/24420 [1:13:19<8:47:50,  1.48s/it]
 12%|█▏        | 2953/24420 [1:13:20<8:49:21,  1.48s/it]
 12%|█▏        | 2954/24420 [1:13:21<8:49:19,  1.48s/it]
 12%|█▏        | 2955/24420 [1:13:23<8:49:09,  1.48s/it]
                                                        
{'loss': 0.1658, 'learning_rate': 5e-05, 'epoch': 0.61}

 12%|█▏        | 2955/24420 [1:13:23<8:49:09,  1.48s/it]
 12%|█▏        | 2956/24420 [1:13:24<8:48:37,  1.48s/it]
 12%|█▏        | 2957/24420 [1:13:26<8:47:58,  1.48s/it]
 12%|█▏        | 2958/24420 [1:13:27<8:47:33,  1.47s/it]
 12%|█▏        | 2959/24420 [1:13:29<8:47:05,  1.47s/it]
 12%|█▏        | 2960/24420 [1:13:30<8:47:28,  1.47s/it]
                                                        
{'loss': 0.169, 'learning_rate': 5e-05, 'epoch': 0.61}

 12%|█▏        | 2960/24420 [1:13:30<8:47:28,  1.47s/it]
 12%|█▏        | 2961/24420 [1:13:32<8:47:32,  1.48s/it]
 12%|█▏        | 2962/24420 [1:13:33<8:47:43,  1.48s/it]
 12%|█▏        | 2963/24420 [1:13:35<8:48:14,  1.48s/it]
 12%|█▏        | 2964/24420 [1:13:36<8:48:00,  1.48s/it]
 12%|█▏        | 2965/24420 [1:13:38<8:47:10,  1.47s/it]
                                                        
{'loss': 0.1791, 'learning_rate': 5e-05, 'epoch': 0.61}

 12%|█▏        | 2965/24420 [1:13:38<8:47:10,  1.47s/it]
 12%|█▏        | 2966/24420 [1:13:39<8:47:40,  1.48s/it]
 12%|█▏        | 2967/24420 [1:13:41<8:47:15,  1.47s/it]
 12%|█▏        | 2968/24420 [1:13:42<8:46:22,  1.47s/it]
 12%|█▏        | 2969/24420 [1:13:44<8:47:51,  1.48s/it]
 12%|█▏        | 2970/24420 [1:13:45<8:47:24,  1.48s/it]
                                                        
{'loss': 0.1835, 'learning_rate': 5e-05, 'epoch': 0.61}

 12%|█▏        | 2970/24420 [1:13:45<8:47:24,  1.48s/it]
 12%|█▏        | 2971/24420 [1:13:47<8:47:47,  1.48s/it]
 12%|█▏        | 2972/24420 [1:13:48<8:46:58,  1.47s/it]
 12%|█▏        | 2973/24420 [1:13:50<8:47:33,  1.48s/it]
 12%|█▏        | 2974/24420 [1:13:51<8:46:57,  1.47s/it]
 12%|█▏        | 2975/24420 [1:13:52<8:46:46,  1.47s/it]
                                                        
{'loss': 0.1664, 'learning_rate': 5e-05, 'epoch': 0.61}

 12%|█▏        | 2975/24420 [1:13:52<8:46:46,  1.47s/it]
 12%|█▏        | 2976/24420 [1:13:54<8:48:17,  1.48s/it]
 12%|█▏        | 2977/24420 [1:13:55<8:48:12,  1.48s/it]
 12%|█▏        | 2978/24420 [1:13:57<8:48:14,  1.48s/it]
 12%|█▏        | 2979/24420 [1:13:58<8:48:14,  1.48s/it]
 12%|█▏        | 2980/24420 [1:14:00<8:47:41,  1.48s/it]
                                                        
{'loss': 0.1902, 'learning_rate': 5e-05, 'epoch': 0.61}

 12%|█▏        | 2980/24420 [1:14:00<8:47:41,  1.48s/it]
 12%|█▏        | 2981/24420 [1:14:01<8:47:41,  1.48s/it]
 12%|█▏        | 2982/24420 [1:14:03<8:48:15,  1.48s/it]
 12%|█▏        | 2983/24420 [1:14:04<8:47:33,  1.48s/it]
 12%|█▏        | 2984/24420 [1:14:06<8:47:15,  1.48s/it]
 12%|█▏        | 2985/24420 [1:14:07<8:47:17,  1.48s/it]
                                                        
{'loss': 0.1589, 'learning_rate': 5e-05, 'epoch': 0.61}

 12%|█▏        | 2985/24420 [1:14:07<8:47:17,  1.48s/it]
 12%|█▏        | 2986/24420 [1:14:09<8:47:45,  1.48s/it]
 12%|█▏        | 2987/24420 [1:14:10<8:48:18,  1.48s/it]
 12%|█▏        | 2988/24420 [1:14:12<8:48:25,  1.48s/it]
 12%|█▏        | 2989/24420 [1:14:13<8:48:39,  1.48s/it]
 12%|█▏        | 2990/24420 [1:14:15<8:48:45,  1.48s/it]
                                                        
{'loss': 0.1762, 'learning_rate': 5e-05, 'epoch': 0.61}

 12%|█▏        | 2990/24420 [1:14:15<8:48:45,  1.48s/it]
 12%|█▏        | 2991/24420 [1:14:16<8:48:21,  1.48s/it]
 12%|█▏        | 2992/24420 [1:14:18<8:47:56,  1.48s/it]
 12%|█▏        | 2993/24420 [1:14:19<8:48:23,  1.48s/it]
 12%|█▏        | 2994/24420 [1:14:21<8:47:43,  1.48s/it]
 12%|█▏        | 2995/24420 [1:14:22<8:47:41,  1.48s/it]
                                                        
{'loss': 0.1808, 'learning_rate': 5e-05, 'epoch': 0.61}

 12%|█▏        | 2995/24420 [1:14:22<8:47:41,  1.48s/it]
 12%|█▏        | 2996/24420 [1:14:24<8:47:54,  1.48s/it]
 12%|█▏        | 2997/24420 [1:14:25<8:47:51,  1.48s/it]
 12%|█▏        | 2998/24420 [1:14:26<8:49:05,  1.48s/it]
 12%|█▏        | 2999/24420 [1:14:28<8:50:26,  1.49s/it]
 12%|█▏        | 3000/24420 [1:14:29<8:49:35,  1.48s/it]
                                                        
{'loss': 0.1814, 'learning_rate': 5e-05, 'epoch': 0.61}

 12%|█▏        | 3000/24420 [1:14:29<8:49:35,  1.48s/it]
 12%|█▏        | 3001/24420 [1:14:34<14:27:10,  2.43s/it]
 12%|█▏        | 3002/24420 [1:14:36<12:45:10,  2.14s/it]
 12%|█▏        | 3003/24420 [1:14:37<11:35:14,  1.95s/it]
 12%|█▏        | 3004/24420 [1:14:39<10:44:35,  1.81s/it]
 12%|█▏        | 3005/24420 [1:14:40<10:09:21,  1.71s/it]
                                                         
{'loss': 0.2258, 'learning_rate': 5e-05, 'epoch': 0.62}

 12%|█▏        | 3005/24420 [1:14:40<10:09:21,  1.71s/it]
 12%|█▏        | 3006/24420 [1:14:41<9:44:56,  1.64s/it] 
 12%|█▏        | 3007/24420 [1:14:43<9:28:08,  1.59s/it]
 12%|█▏        | 3008/24420 [1:14:44<9:15:36,  1.56s/it]
 12%|█▏        | 3009/24420 [1:14:46<9:07:29,  1.53s/it]
 12%|█▏        | 3010/24420 [1:14:47<9:01:04,  1.52s/it]
                                                        
{'loss': 0.1884, 'learning_rate': 5e-05, 'epoch': 0.62}

 12%|█▏        | 3010/24420 [1:14:47<9:01:04,  1.52s/it]
 12%|█▏        | 3011/24420 [1:14:49<8:57:31,  1.51s/it]
 12%|█▏        | 3012/24420 [1:14:50<8:55:20,  1.50s/it]
 12%|█▏        | 3013/24420 [1:14:52<8:54:19,  1.50s/it]
 12%|█▏        | 3014/24420 [1:14:53<8:52:23,  1.49s/it]
 12%|█▏        | 3015/24420 [1:14:55<8:50:21,  1.49s/it]
                                                        
{'loss': 0.2081, 'learning_rate': 5e-05, 'epoch': 0.62}

 12%|█▏        | 3015/24420 [1:14:55<8:50:21,  1.49s/it]
 12%|█▏        | 3016/24420 [1:14:56<8:49:30,  1.48s/it]
 12%|█▏        | 3017/24420 [1:14:58<8:48:22,  1.48s/it]
 12%|█▏        | 3018/24420 [1:14:59<8:47:49,  1.48s/it]
 12%|█▏        | 3019/24420 [1:15:01<8:47:47,  1.48s/it]
 12%|█▏        | 3020/24420 [1:15:02<8:47:42,  1.48s/it]
                                                        
{'loss': 0.1662, 'learning_rate': 5e-05, 'epoch': 0.62}

 12%|█▏        | 3020/24420 [1:15:02<8:47:42,  1.48s/it]
 12%|█▏        | 3021/24420 [1:15:04<8:48:34,  1.48s/it]
 12%|█▏        | 3022/24420 [1:15:05<8:48:12,  1.48s/it]
 12%|█▏        | 3023/24420 [1:15:07<8:48:39,  1.48s/it]
 12%|█▏        | 3024/24420 [1:15:08<8:47:36,  1.48s/it]
 12%|█▏        | 3025/24420 [1:15:10<8:46:47,  1.48s/it]
                                                        
{'loss': 0.2059, 'learning_rate': 5e-05, 'epoch': 0.62}

 12%|█▏        | 3025/24420 [1:15:10<8:46:47,  1.48s/it]
 12%|█▏        | 3026/24420 [1:15:11<8:47:30,  1.48s/it]
 12%|█▏        | 3027/24420 [1:15:13<8:47:05,  1.48s/it]
 12%|█▏        | 3028/24420 [1:15:14<8:47:28,  1.48s/it]
 12%|█▏        | 3029/24420 [1:15:16<8:47:47,  1.48s/it]
 12%|█▏        | 3030/24420 [1:15:17<8:48:57,  1.48s/it]
                                                        
{'loss': 0.1843, 'learning_rate': 5e-05, 'epoch': 0.62}

 12%|█▏        | 3030/24420 [1:15:17<8:48:57,  1.48s/it]
 12%|█▏        | 3031/24420 [1:15:19<8:48:41,  1.48s/it]
 12%|█▏        | 3032/24420 [1:15:20<8:48:12,  1.48s/it]
 12%|█▏        | 3033/24420 [1:15:21<8:48:09,  1.48s/it]
 12%|█▏        | 3034/24420 [1:15:23<8:48:11,  1.48s/it]
 12%|█▏        | 3035/24420 [1:15:24<8:48:15,  1.48s/it]
                                                        
{'loss': 0.2048, 'learning_rate': 5e-05, 'epoch': 0.62}

 12%|█▏        | 3035/24420 [1:15:24<8:48:15,  1.48s/it]
 12%|█▏        | 3036/24420 [1:15:26<8:49:02,  1.48s/it]
 12%|█▏        | 3037/24420 [1:15:27<8:49:57,  1.49s/it]
 12%|█▏        | 3038/24420 [1:15:29<8:49:52,  1.49s/it]
 12%|█▏        | 3039/24420 [1:15:30<8:48:45,  1.48s/it]
 12%|█▏        | 3040/24420 [1:15:32<8:48:48,  1.48s/it]
                                                        
{'loss': 0.2003, 'learning_rate': 5e-05, 'epoch': 0.62}

 12%|█▏        | 3040/24420 [1:15:32<8:48:48,  1.48s/it]
 12%|█▏        | 3041/24420 [1:15:33<8:48:10,  1.48s/it]
 12%|█▏        | 3042/24420 [1:15:35<8:49:48,  1.49s/it]
 12%|█▏        | 3043/24420 [1:15:36<8:49:55,  1.49s/it]
 12%|█▏        | 3044/24420 [1:15:38<8:48:53,  1.48s/it]
 12%|█▏        | 3045/24420 [1:15:39<8:47:54,  1.48s/it]
                                                        
{'loss': 0.1897, 'learning_rate': 5e-05, 'epoch': 0.62}

 12%|█▏        | 3045/24420 [1:15:39<8:47:54,  1.48s/it]
 12%|█▏        | 3046/24420 [1:15:41<8:48:31,  1.48s/it]
 12%|█▏        | 3047/24420 [1:15:42<8:49:41,  1.49s/it]
 12%|█▏        | 3048/24420 [1:15:44<8:49:14,  1.49s/it]
 12%|█▏        | 3049/24420 [1:15:45<8:49:29,  1.49s/it]
 12%|█▏        | 3050/24420 [1:15:47<8:48:46,  1.48s/it]
                                                        
{'loss': 0.2069, 'learning_rate': 5e-05, 'epoch': 0.62}

 12%|█▏        | 3050/24420 [1:15:47<8:48:46,  1.48s/it]
 12%|█▏        | 3051/24420 [1:15:48<8:47:21,  1.48s/it]
 12%|█▏        | 3052/24420 [1:15:50<8:46:31,  1.48s/it]
 13%|█▎        | 3053/24420 [1:15:51<8:45:53,  1.48s/it]
 13%|█▎        | 3054/24420 [1:15:53<8:45:27,  1.48s/it]
 13%|█▎        | 3055/24420 [1:15:54<8:47:12,  1.48s/it]
                                                        
{'loss': 0.2012, 'learning_rate': 5e-05, 'epoch': 0.63}

 13%|█▎        | 3055/24420 [1:15:54<8:47:12,  1.48s/it]
 13%|█▎        | 3056/24420 [1:15:56<8:46:58,  1.48s/it]
 13%|█▎        | 3057/24420 [1:15:57<8:46:56,  1.48s/it]
 13%|█▎        | 3058/24420 [1:15:59<8:47:13,  1.48s/it]
 13%|█▎        | 3059/24420 [1:16:00<8:46:28,  1.48s/it]
 13%|█▎        | 3060/24420 [1:16:01<8:46:41,  1.48s/it]
                                                        
{'loss': 0.2021, 'learning_rate': 5e-05, 'epoch': 0.63}

 13%|█▎        | 3060/24420 [1:16:01<8:46:41,  1.48s/it]
 13%|█▎        | 3061/24420 [1:16:03<8:46:50,  1.48s/it]
 13%|█▎        | 3062/24420 [1:16:04<8:46:00,  1.48s/it]
 13%|█▎        | 3063/24420 [1:16:06<8:46:38,  1.48s/it]
 13%|█▎        | 3064/24420 [1:16:07<8:46:21,  1.48s/it]
 13%|█▎        | 3065/24420 [1:16:09<8:46:49,  1.48s/it]
                                                        
{'loss': 0.1808, 'learning_rate': 5e-05, 'epoch': 0.63}

 13%|█▎        | 3065/24420 [1:16:09<8:46:49,  1.48s/it]
 13%|█▎        | 3066/24420 [1:16:10<8:46:43,  1.48s/it]
 13%|█▎        | 3067/24420 [1:16:12<8:45:50,  1.48s/it]
 13%|█▎        | 3068/24420 [1:16:13<8:44:57,  1.48s/it]
 13%|█▎        | 3069/24420 [1:16:15<8:46:02,  1.48s/it]
 13%|█▎        | 3070/24420 [1:16:16<8:45:52,  1.48s/it]
                                                        
{'loss': 0.203, 'learning_rate': 5e-05, 'epoch': 0.63}

 13%|█▎        | 3070/24420 [1:16:16<8:45:52,  1.48s/it]
 13%|█▎        | 3071/24420 [1:16:18<8:46:41,  1.48s/it]
 13%|█▎        | 3072/24420 [1:16:19<8:47:21,  1.48s/it]
 13%|█▎        | 3073/24420 [1:16:21<8:46:31,  1.48s/it]
 13%|█▎        | 3074/24420 [1:16:22<8:46:42,  1.48s/it]
 13%|█▎        | 3075/24420 [1:16:24<8:46:12,  1.48s/it]
                                                        
{'loss': 0.2024, 'learning_rate': 5e-05, 'epoch': 0.63}

 13%|█▎        | 3075/24420 [1:16:24<8:46:12,  1.48s/it]
 13%|█▎        | 3076/24420 [1:16:25<8:46:20,  1.48s/it]
 13%|█▎        | 3077/24420 [1:16:27<8:46:22,  1.48s/it]
 13%|█▎        | 3078/24420 [1:16:28<8:46:04,  1.48s/it]
 13%|█▎        | 3079/24420 [1:16:30<8:46:52,  1.48s/it]
 13%|█▎        | 3080/24420 [1:16:31<8:47:11,  1.48s/it]
                                                        
{'loss': 0.1765, 'learning_rate': 5e-05, 'epoch': 0.63}

 13%|█▎        | 3080/24420 [1:16:31<8:47:11,  1.48s/it]
 13%|█▎        | 3081/24420 [1:16:33<8:46:36,  1.48s/it]
 13%|█▎        | 3082/24420 [1:16:34<8:45:33,  1.48s/it]
 13%|█▎        | 3083/24420 [1:16:36<8:45:42,  1.48s/it]
 13%|█▎        | 3084/24420 [1:16:37<8:45:44,  1.48s/it]
 13%|█▎        | 3085/24420 [1:16:38<8:46:30,  1.48s/it]
                                                        
{'loss': 0.164, 'learning_rate': 5e-05, 'epoch': 0.63}

 13%|█▎        | 3085/24420 [1:16:38<8:46:30,  1.48s/it]
 13%|█▎        | 3086/24420 [1:16:40<8:46:54,  1.48s/it]
 13%|█▎        | 3087/24420 [1:16:41<8:46:28,  1.48s/it]
 13%|█▎        | 3088/24420 [1:16:43<8:46:28,  1.48s/it]
 13%|█▎        | 3089/24420 [1:16:44<8:46:45,  1.48s/it]
 13%|█▎        | 3090/24420 [1:16:46<8:46:08,  1.48s/it]
                                                        
{'loss': 0.1823, 'learning_rate': 5e-05, 'epoch': 0.63}

 13%|█▎        | 3090/24420 [1:16:46<8:46:08,  1.48s/it]
 13%|█▎        | 3091/24420 [1:16:47<8:48:37,  1.49s/it]
 13%|█▎        | 3092/24420 [1:16:49<8:46:48,  1.48s/it]
 13%|█▎        | 3093/24420 [1:16:50<8:46:29,  1.48s/it]
 13%|█▎        | 3094/24420 [1:16:52<8:47:09,  1.48s/it]
 13%|█▎        | 3095/24420 [1:16:53<8:45:43,  1.48s/it]
                                                        
{'loss': 0.187, 'learning_rate': 5e-05, 'epoch': 0.63}

 13%|█▎        | 3095/24420 [1:16:53<8:45:43,  1.48s/it]
 13%|█▎        | 3096/24420 [1:16:55<8:45:26,  1.48s/it]
 13%|█▎        | 3097/24420 [1:16:56<8:45:39,  1.48s/it]
 13%|█▎        | 3098/24420 [1:16:58<8:46:19,  1.48s/it]
 13%|█▎        | 3099/24420 [1:16:59<8:47:02,  1.48s/it]
 13%|█▎        | 3100/24420 [1:17:01<8:47:24,  1.48s/it]
                                                        
{'loss': 0.1967, 'learning_rate': 5e-05, 'epoch': 0.63}

 13%|█▎        | 3100/24420 [1:17:01<8:47:24,  1.48s/it]
 13%|█▎        | 3101/24420 [1:17:02<8:47:36,  1.48s/it]
 13%|█▎        | 3102/24420 [1:17:04<8:46:08,  1.48s/it]
 13%|█▎        | 3103/24420 [1:17:05<8:47:02,  1.48s/it]
 13%|█▎        | 3104/24420 [1:17:07<8:46:58,  1.48s/it]
 13%|█▎        | 3105/24420 [1:17:08<8:46:44,  1.48s/it]
                                                        
{'loss': 0.1535, 'learning_rate': 5e-05, 'epoch': 0.64}

 13%|█▎        | 3105/24420 [1:17:08<8:46:44,  1.48s/it]
 13%|█▎        | 3106/24420 [1:17:10<8:47:28,  1.48s/it]
 13%|█▎        | 3107/24420 [1:17:11<8:46:42,  1.48s/it]
 13%|█▎        | 3108/24420 [1:17:13<8:46:14,  1.48s/it]
 13%|█▎        | 3109/24420 [1:17:14<8:46:23,  1.48s/it]
 13%|█▎        | 3110/24420 [1:17:16<8:45:37,  1.48s/it]
                                                        
{'loss': 0.1764, 'learning_rate': 5e-05, 'epoch': 0.64}

 13%|█▎        | 3110/24420 [1:17:16<8:45:37,  1.48s/it]
 13%|█▎        | 3111/24420 [1:17:17<8:45:40,  1.48s/it]
 13%|█▎        | 3112/24420 [1:17:18<8:44:39,  1.48s/it]
 13%|█▎        | 3113/24420 [1:17:20<8:45:00,  1.48s/it]
 13%|█▎        | 3114/24420 [1:17:21<8:45:22,  1.48s/it]
 13%|█▎        | 3115/24420 [1:17:23<8:45:30,  1.48s/it]
                                                        
{'loss': 0.1872, 'learning_rate': 5e-05, 'epoch': 0.64}

 13%|█▎        | 3115/24420 [1:17:23<8:45:30,  1.48s/it]
 13%|█▎        | 3116/24420 [1:17:24<8:45:46,  1.48s/it]
 13%|█▎        | 3117/24420 [1:17:26<8:44:56,  1.48s/it]
 13%|█▎        | 3118/24420 [1:17:27<8:44:20,  1.48s/it]
 13%|█▎        | 3119/24420 [1:17:29<8:46:26,  1.48s/it]
 13%|█▎        | 3120/24420 [1:17:30<8:49:22,  1.49s/it]
                                                        
{'loss': 0.1694, 'learning_rate': 5e-05, 'epoch': 0.64}

 13%|█▎        | 3120/24420 [1:17:30<8:49:22,  1.49s/it]
 13%|█▎        | 3121/24420 [1:17:32<8:47:47,  1.49s/it]
 13%|█▎        | 3122/24420 [1:17:33<8:46:10,  1.48s/it]
 13%|█▎        | 3123/24420 [1:17:35<8:45:23,  1.48s/it]
 13%|█▎        | 3124/24420 [1:17:36<8:44:50,  1.48s/it]
 13%|█▎        | 3125/24420 [1:17:38<8:43:42,  1.48s/it]
                                                        
{'loss': 0.1769, 'learning_rate': 5e-05, 'epoch': 0.64}

 13%|█▎        | 3125/24420 [1:17:38<8:43:42,  1.48s/it]
 13%|█▎        | 3126/24420 [1:17:39<8:44:37,  1.48s/it]
 13%|█▎        | 3127/24420 [1:17:41<8:44:18,  1.48s/it]
 13%|█▎        | 3128/24420 [1:17:42<8:44:36,  1.48s/it]
 13%|█▎        | 3129/24420 [1:17:44<8:44:09,  1.48s/it]
 13%|█▎        | 3130/24420 [1:17:45<8:44:09,  1.48s/it]
                                                        
{'loss': 0.1973, 'learning_rate': 5e-05, 'epoch': 0.64}

 13%|█▎        | 3130/24420 [1:17:45<8:44:09,  1.48s/it]
 13%|█▎        | 3131/24420 [1:17:47<8:45:55,  1.48s/it]
 13%|█▎        | 3132/24420 [1:17:48<8:46:48,  1.48s/it]
 13%|█▎        | 3133/24420 [1:17:50<8:45:40,  1.48s/it]
 13%|█▎        | 3134/24420 [1:17:51<8:45:18,  1.48s/it]
 13%|█▎        | 3135/24420 [1:17:53<8:45:10,  1.48s/it]
                                                        
{'loss': 0.1604, 'learning_rate': 5e-05, 'epoch': 0.64}

 13%|█▎        | 3135/24420 [1:17:53<8:45:10,  1.48s/it]
 13%|█▎        | 3136/24420 [1:17:54<8:44:50,  1.48s/it]
 13%|█▎        | 3137/24420 [1:17:55<8:44:06,  1.48s/it]
 13%|█▎        | 3138/24420 [1:17:57<8:43:07,  1.47s/it]
 13%|█▎        | 3139/24420 [1:17:58<8:43:09,  1.48s/it]
 13%|█▎        | 3140/24420 [1:18:00<8:43:42,  1.48s/it]
                                                        
{'loss': 0.1909, 'learning_rate': 5e-05, 'epoch': 0.64}

 13%|█▎        | 3140/24420 [1:18:00<8:43:42,  1.48s/it]
 13%|█▎        | 3141/24420 [1:18:01<8:43:54,  1.48s/it]
 13%|█▎        | 3142/24420 [1:18:03<8:43:42,  1.48s/it]
 13%|█▎        | 3143/24420 [1:18:04<8:43:36,  1.48s/it]
 13%|█▎        | 3144/24420 [1:18:06<8:44:48,  1.48s/it]
 13%|█▎        | 3145/24420 [1:18:07<8:44:30,  1.48s/it]
                                                        
{'loss': 0.152, 'learning_rate': 5e-05, 'epoch': 0.64}

 13%|█▎        | 3145/24420 [1:18:07<8:44:30,  1.48s/it]
 13%|█▎        | 3146/24420 [1:18:09<8:44:26,  1.48s/it]
 13%|█▎        | 3147/24420 [1:18:10<8:43:37,  1.48s/it]
 13%|█▎        | 3148/24420 [1:18:12<8:43:58,  1.48s/it]
 13%|█▎        | 3149/24420 [1:18:13<8:43:10,  1.48s/it]
 13%|█▎        | 3150/24420 [1:18:15<8:43:24,  1.48s/it]
                                                        
{'loss': 0.1772, 'learning_rate': 5e-05, 'epoch': 0.64}

 13%|█▎        | 3150/24420 [1:18:15<8:43:24,  1.48s/it]
 13%|█▎        | 3151/24420 [1:18:16<8:44:19,  1.48s/it]
 13%|█▎        | 3152/24420 [1:18:18<8:44:01,  1.48s/it]
 13%|█▎        | 3153/24420 [1:18:19<8:43:35,  1.48s/it]
 13%|█▎        | 3154/24420 [1:18:21<8:43:18,  1.48s/it]
 13%|█▎        | 3155/24420 [1:18:22<8:42:52,  1.48s/it]
                                                        
{'loss': 0.1799, 'learning_rate': 5e-05, 'epoch': 0.65}

 13%|█▎        | 3155/24420 [1:18:22<8:42:52,  1.48s/it]
 13%|█▎        | 3156/24420 [1:18:24<8:43:21,  1.48s/it]
 13%|█▎        | 3157/24420 [1:18:25<8:42:51,  1.48s/it]
 13%|█▎        | 3158/24420 [1:18:27<8:44:27,  1.48s/it]
 13%|█▎        | 3159/24420 [1:18:28<8:44:09,  1.48s/it]
 13%|█▎        | 3160/24420 [1:18:29<8:43:40,  1.48s/it]
                                                        
{'loss': 0.183, 'learning_rate': 5e-05, 'epoch': 0.65}

 13%|█▎        | 3160/24420 [1:18:29<8:43:40,  1.48s/it]
 13%|█▎        | 3161/24420 [1:18:31<8:44:21,  1.48s/it]
 13%|█▎        | 3162/24420 [1:18:32<8:44:14,  1.48s/it]
 13%|█▎        | 3163/24420 [1:18:34<8:44:35,  1.48s/it]
 13%|█▎        | 3164/24420 [1:18:35<8:44:30,  1.48s/it]
 13%|█▎        | 3165/24420 [1:18:37<8:43:31,  1.48s/it]
                                                        
{'loss': 0.1694, 'learning_rate': 5e-05, 'epoch': 0.65}

 13%|█▎        | 3165/24420 [1:18:37<8:43:31,  1.48s/it]
 13%|█▎        | 3166/24420 [1:18:38<8:43:34,  1.48s/it]
 13%|█▎        | 3167/24420 [1:18:40<8:43:53,  1.48s/it]
 13%|█▎        | 3168/24420 [1:18:41<8:43:56,  1.48s/it]
 13%|█▎        | 3169/24420 [1:18:43<8:43:20,  1.48s/it]
 13%|█▎        | 3170/24420 [1:18:44<8:42:43,  1.48s/it]
                                                        
{'loss': 0.1812, 'learning_rate': 5e-05, 'epoch': 0.65}

 13%|█▎        | 3170/24420 [1:18:44<8:42:43,  1.48s/it]
 13%|█▎        | 3171/24420 [1:18:46<8:42:13,  1.47s/it]
 13%|█▎        | 3172/24420 [1:18:47<8:42:31,  1.48s/it]
 13%|█▎        | 3173/24420 [1:18:49<8:42:39,  1.48s/it]
 13%|█▎        | 3174/24420 [1:18:50<8:42:35,  1.48s/it]
 13%|█▎        | 3175/24420 [1:18:52<8:42:48,  1.48s/it]
                                                        
{'loss': 0.1873, 'learning_rate': 5e-05, 'epoch': 0.65}

 13%|█▎        | 3175/24420 [1:18:52<8:42:48,  1.48s/it]
 13%|█▎        | 3176/24420 [1:18:53<8:43:25,  1.48s/it]
 13%|█▎        | 3177/24420 [1:18:55<8:43:29,  1.48s/it]
 13%|█▎        | 3178/24420 [1:18:56<8:44:23,  1.48s/it]
 13%|█▎        | 3179/24420 [1:18:58<8:44:26,  1.48s/it]
 13%|█▎        | 3180/24420 [1:18:59<8:43:47,  1.48s/it]
                                                        
{'loss': 0.1935, 'learning_rate': 5e-05, 'epoch': 0.65}

 13%|█▎        | 3180/24420 [1:18:59<8:43:47,  1.48s/it]
 13%|█▎        | 3181/24420 [1:19:01<8:44:20,  1.48s/it]
 13%|█▎        | 3182/24420 [1:19:02<8:43:40,  1.48s/it]
 13%|█▎        | 3183/24420 [1:19:03<8:43:49,  1.48s/it]
 13%|█▎        | 3184/24420 [1:19:05<8:43:30,  1.48s/it]
 13%|█▎        | 3185/24420 [1:19:06<8:43:08,  1.48s/it]
                                                        
{'loss': 0.1738, 'learning_rate': 5e-05, 'epoch': 0.65}

 13%|█▎        | 3185/24420 [1:19:06<8:43:08,  1.48s/it]
 13%|█▎        | 3186/24420 [1:19:08<8:43:25,  1.48s/it]
 13%|█▎        | 3187/24420 [1:19:09<8:42:37,  1.48s/it]
 13%|█▎        | 3188/24420 [1:19:11<8:42:26,  1.48s/it]
 13%|█▎        | 3189/24420 [1:19:12<8:42:39,  1.48s/it]
 13%|█▎        | 3190/24420 [1:19:14<8:42:22,  1.48s/it]
                                                        
{'loss': 0.1678, 'learning_rate': 5e-05, 'epoch': 0.65}

 13%|█▎        | 3190/24420 [1:19:14<8:42:22,  1.48s/it]
 13%|█▎        | 3191/24420 [1:19:15<8:42:13,  1.48s/it]
 13%|█▎        | 3192/24420 [1:19:17<8:42:13,  1.48s/it]
 13%|█▎        | 3193/24420 [1:19:18<8:42:12,  1.48s/it]
 13%|█▎        | 3194/24420 [1:19:20<8:41:31,  1.47s/it]
 13%|█▎        | 3195/24420 [1:19:21<8:41:57,  1.48s/it]
                                                        
{'loss': 0.1607, 'learning_rate': 5e-05, 'epoch': 0.65}

 13%|█▎        | 3195/24420 [1:19:21<8:41:57,  1.48s/it]
 13%|█▎        | 3196/24420 [1:19:23<8:42:27,  1.48s/it]
 13%|█▎        | 3197/24420 [1:19:24<8:44:11,  1.48s/it]
 13%|█▎        | 3198/24420 [1:19:26<8:43:23,  1.48s/it]
 13%|█▎        | 3199/24420 [1:19:27<8:42:57,  1.48s/it]
 13%|█▎        | 3200/24420 [1:19:29<8:42:42,  1.48s/it]
                                                        
{'loss': 0.1747, 'learning_rate': 5e-05, 'epoch': 0.66}

 13%|█▎        | 3200/24420 [1:19:29<8:42:42,  1.48s/it]
 13%|█▎        | 3201/24420 [1:19:33<14:06:43,  2.39s/it]
 13%|█▎        | 3202/24420 [1:19:35<12:30:10,  2.12s/it]
 13%|█▎        | 3203/24420 [1:19:36<11:22:46,  1.93s/it]
 13%|█▎        | 3204/24420 [1:19:38<10:33:57,  1.79s/it]
 13%|█▎        | 3205/24420 [1:19:39<10:00:27,  1.70s/it]
                                                         
{'loss': 0.1914, 'learning_rate': 5e-05, 'epoch': 0.66}

 13%|█▎        | 3205/24420 [1:19:39<10:00:27,  1.70s/it]
 13%|█▎        | 3206/24420 [1:19:41<9:36:45,  1.63s/it] 
 13%|█▎        | 3207/24420 [1:19:42<9:20:32,  1.59s/it]
 13%|█▎        | 3208/24420 [1:19:43<9:10:07,  1.56s/it]
 13%|█▎        | 3209/24420 [1:19:45<9:02:56,  1.54s/it]
 13%|█▎        | 3210/24420 [1:19:46<8:57:09,  1.52s/it]
                                                        
{'loss': 0.181, 'learning_rate': 5e-05, 'epoch': 0.66}

 13%|█▎        | 3210/24420 [1:19:46<8:57:09,  1.52s/it]
 13%|█▎        | 3211/24420 [1:19:48<8:52:49,  1.51s/it]
 13%|█▎        | 3212/24420 [1:19:49<8:49:44,  1.50s/it]
 13%|█▎        | 3213/24420 [1:19:51<8:46:37,  1.49s/it]
 13%|█▎        | 3214/24420 [1:19:52<8:45:25,  1.49s/it]
 13%|█▎        | 3215/24420 [1:19:54<8:44:05,  1.48s/it]
                                                        
{'loss': 0.1738, 'learning_rate': 5e-05, 'epoch': 0.66}

 13%|█▎        | 3215/24420 [1:19:54<8:44:05,  1.48s/it]
 13%|█▎        | 3216/24420 [1:19:55<8:44:47,  1.48s/it]
 13%|█▎        | 3217/24420 [1:19:57<8:46:22,  1.49s/it]
 13%|█▎        | 3218/24420 [1:19:58<8:44:51,  1.49s/it]
 13%|█▎        | 3219/24420 [1:20:00<8:43:58,  1.48s/it]
 13%|█▎        | 3220/24420 [1:20:01<8:43:19,  1.48s/it]
                                                        
{'loss': 0.1913, 'learning_rate': 5e-05, 'epoch': 0.66}

 13%|█▎        | 3220/24420 [1:20:01<8:43:19,  1.48s/it]
 13%|█▎        | 3221/24420 [1:20:03<8:43:00,  1.48s/it]
 13%|█▎        | 3222/24420 [1:20:04<8:42:25,  1.48s/it]
 13%|█▎        | 3223/24420 [1:20:06<8:42:10,  1.48s/it]
 13%|█▎        | 3224/24420 [1:20:07<8:41:43,  1.48s/it]
 13%|█▎        | 3225/24420 [1:20:09<8:43:18,  1.48s/it]
                                                        
{'loss': 0.1616, 'learning_rate': 5e-05, 'epoch': 0.66}

 13%|█▎        | 3225/24420 [1:20:09<8:43:18,  1.48s/it]
 13%|█▎        | 3226/24420 [1:20:10<8:42:36,  1.48s/it]
 13%|█▎        | 3227/24420 [1:20:12<8:42:49,  1.48s/it]
 13%|█▎        | 3228/24420 [1:20:13<8:41:56,  1.48s/it]
 13%|█▎        | 3229/24420 [1:20:15<8:43:27,  1.48s/it]
 13%|█▎        | 3230/24420 [1:20:16<8:42:32,  1.48s/it]
                                                        
{'loss': 0.2089, 'learning_rate': 5e-05, 'epoch': 0.66}

 13%|█▎        | 3230/24420 [1:20:16<8:42:32,  1.48s/it]
 13%|█▎        | 3231/24420 [1:20:18<8:42:31,  1.48s/it]
 13%|█▎        | 3232/24420 [1:20:19<8:43:40,  1.48s/it]
 13%|█▎        | 3233/24420 [1:20:20<8:43:08,  1.48s/it]
 13%|█▎        | 3234/24420 [1:20:22<8:42:08,  1.48s/it]
 13%|█▎        | 3235/24420 [1:20:23<8:41:46,  1.48s/it]
                                                        
{'loss': 0.1989, 'learning_rate': 5e-05, 'epoch': 0.66}

 13%|█▎        | 3235/24420 [1:20:23<8:41:46,  1.48s/it]
 13%|█▎        | 3236/24420 [1:20:25<8:42:46,  1.48s/it]
 13%|█▎        | 3237/24420 [1:20:26<8:42:42,  1.48s/it]
 13%|█▎        | 3238/24420 [1:20:28<8:43:15,  1.48s/it]
 13%|█▎        | 3239/24420 [1:20:29<8:42:35,  1.48s/it]
 13%|█▎        | 3240/24420 [1:20:31<8:41:45,  1.48s/it]
                                                        
{'loss': 0.1792, 'learning_rate': 5e-05, 'epoch': 0.66}

 13%|█▎        | 3240/24420 [1:20:31<8:41:45,  1.48s/it]
 13%|█▎        | 3241/24420 [1:20:32<8:43:37,  1.48s/it]
 13%|█▎        | 3242/24420 [1:20:34<8:43:05,  1.48s/it]
 13%|█▎        | 3243/24420 [1:20:35<8:43:50,  1.48s/it]
 13%|█▎        | 3244/24420 [1:20:37<8:43:17,  1.48s/it]
 13%|█▎        | 3245/24420 [1:20:38<8:43:13,  1.48s/it]
                                                        
{'loss': 0.1985, 'learning_rate': 5e-05, 'epoch': 0.66}

 13%|█▎        | 3245/24420 [1:20:38<8:43:13,  1.48s/it]
 13%|█▎        | 3246/24420 [1:20:40<8:42:31,  1.48s/it]
 13%|█▎        | 3247/24420 [1:20:41<9:10:44,  1.56s/it]
 13%|█▎        | 3248/24420 [1:20:43<9:01:14,  1.53s/it]
 13%|█▎        | 3249/24420 [1:20:44<8:55:22,  1.52s/it]
 13%|█▎        | 3250/24420 [1:20:46<8:50:53,  1.50s/it]
                                                        
{'loss': 0.2211, 'learning_rate': 5e-05, 'epoch': 0.67}

 13%|█▎        | 3250/24420 [1:20:46<8:50:53,  1.50s/it]
 13%|█▎        | 3251/24420 [1:20:47<8:48:17,  1.50s/it]
 13%|█▎        | 3252/24420 [1:20:49<8:46:07,  1.49s/it]
 13%|█▎        | 3253/24420 [1:20:50<8:44:38,  1.49s/it]
 13%|█▎        | 3254/24420 [1:20:52<8:43:42,  1.48s/it]
 13%|█▎        | 3255/24420 [1:20:53<8:42:25,  1.48s/it]
                                                        
{'loss': 0.195, 'learning_rate': 5e-05, 'epoch': 0.67}

 13%|█▎        | 3255/24420 [1:20:53<8:42:25,  1.48s/it]
 13%|█▎        | 3256/24420 [1:20:55<8:42:58,  1.48s/it]
 13%|█▎        | 3257/24420 [1:20:56<8:42:39,  1.48s/it]
 13%|█▎        | 3258/24420 [1:20:58<8:41:50,  1.48s/it]
 13%|█▎        | 3259/24420 [1:20:59<8:41:37,  1.48s/it]
 13%|█▎        | 3260/24420 [1:21:01<8:40:27,  1.48s/it]
                                                        
{'loss': 0.1881, 'learning_rate': 5e-05, 'epoch': 0.67}

 13%|█▎        | 3260/24420 [1:21:01<8:40:27,  1.48s/it]
 13%|█▎        | 3261/24420 [1:21:02<8:41:35,  1.48s/it]
 13%|█▎        | 3262/24420 [1:21:04<8:41:02,  1.48s/it]
 13%|█▎        | 3263/24420 [1:21:05<8:43:23,  1.48s/it]
 13%|█▎        | 3264/24420 [1:21:07<8:43:08,  1.48s/it]
 13%|█▎        | 3265/24420 [1:21:08<8:42:03,  1.48s/it]
                                                        
{'loss': 0.2113, 'learning_rate': 5e-05, 'epoch': 0.67}

 13%|█▎        | 3265/24420 [1:21:08<8:42:03,  1.48s/it]
 13%|█▎        | 3266/24420 [1:21:10<8:41:47,  1.48s/it]
 13%|█▎        | 3267/24420 [1:21:11<8:41:13,  1.48s/it]
 13%|█▎        | 3268/24420 [1:21:13<8:40:59,  1.48s/it]
 13%|█▎        | 3269/24420 [1:21:14<8:40:51,  1.48s/it]
 13%|█▎        | 3270/24420 [1:21:15<8:40:46,  1.48s/it]
                                                        
{'loss': 0.1909, 'learning_rate': 5e-05, 'epoch': 0.67}

 13%|█▎        | 3270/24420 [1:21:15<8:40:46,  1.48s/it]
 13%|█▎        | 3271/24420 [1:21:17<8:42:29,  1.48s/it]
 13%|█▎        | 3272/24420 [1:21:18<8:42:01,  1.48s/it]
 13%|█▎        | 3273/24420 [1:21:20<8:41:17,  1.48s/it]
 13%|█▎        | 3274/24420 [1:21:21<8:40:37,  1.48s/it]
 13%|█▎        | 3275/24420 [1:21:23<8:43:00,  1.48s/it]
                                                        
{'loss': 0.1871, 'learning_rate': 5e-05, 'epoch': 0.67}

 13%|█▎        | 3275/24420 [1:21:23<8:43:00,  1.48s/it]
 13%|█▎        | 3276/24420 [1:21:24<8:44:03,  1.49s/it]
 13%|█▎        | 3277/24420 [1:21:26<8:43:20,  1.49s/it]
 13%|█▎        | 3278/24420 [1:21:27<8:42:17,  1.48s/it]
 13%|█▎        | 3279/24420 [1:21:29<8:42:36,  1.48s/it]
 13%|█▎        | 3280/24420 [1:21:30<8:42:06,  1.48s/it]
                                                        
{'loss': 0.1955, 'learning_rate': 5e-05, 'epoch': 0.67}

 13%|█▎        | 3280/24420 [1:21:30<8:42:06,  1.48s/it]
 13%|█▎        | 3281/24420 [1:21:32<8:41:59,  1.48s/it]
 13%|█▎        | 3282/24420 [1:21:33<8:42:32,  1.48s/it]
 13%|█▎        | 3283/24420 [1:21:35<8:42:55,  1.48s/it]
 13%|█▎        | 3284/24420 [1:21:36<8:43:37,  1.49s/it]
 13%|█▎        | 3285/24420 [1:21:38<8:42:59,  1.48s/it]
                                                        
{'loss': 0.1785, 'learning_rate': 5e-05, 'epoch': 0.67}

 13%|█▎        | 3285/24420 [1:21:38<8:42:59,  1.48s/it]
 13%|█▎        | 3286/24420 [1:21:39<8:43:25,  1.49s/it]
 13%|█▎        | 3287/24420 [1:21:41<8:43:42,  1.49s/it]
 13%|█▎        | 3288/24420 [1:21:42<8:43:11,  1.49s/it]
 13%|█▎        | 3289/24420 [1:21:44<8:42:14,  1.48s/it]
 13%|█▎        | 3290/24420 [1:21:45<8:41:39,  1.48s/it]
                                                        
{'loss': 0.1816, 'learning_rate': 5e-05, 'epoch': 0.67}

 13%|█▎        | 3290/24420 [1:21:45<8:41:39,  1.48s/it]
 13%|█▎        | 3291/24420 [1:21:47<8:41:05,  1.48s/it]
 13%|█▎        | 3292/24420 [1:21:48<8:41:48,  1.48s/it]
 13%|█▎        | 3293/24420 [1:21:50<8:41:27,  1.48s/it]
 13%|█▎        | 3294/24420 [1:21:51<8:42:19,  1.48s/it]
 13%|█▎        | 3295/24420 [1:21:53<8:41:38,  1.48s/it]
                                                        
{'loss': 0.203, 'learning_rate': 5e-05, 'epoch': 0.67}

 13%|█▎        | 3295/24420 [1:21:53<8:41:38,  1.48s/it]
 13%|█▎        | 3296/24420 [1:21:54<8:42:19,  1.48s/it]
 14%|█▎        | 3297/24420 [1:21:56<8:42:26,  1.48s/it]
 14%|█▎        | 3298/24420 [1:21:57<8:42:37,  1.48s/it]
 14%|█▎        | 3299/24420 [1:21:59<8:42:05,  1.48s/it]
 14%|█▎        | 3300/24420 [1:22:00<8:41:18,  1.48s/it]
                                                        
{'loss': 0.1912, 'learning_rate': 5e-05, 'epoch': 0.68}

 14%|█▎        | 3300/24420 [1:22:00<8:41:18,  1.48s/it]
 14%|█▎        | 3301/24420 [1:22:01<8:41:40,  1.48s/it]
 14%|█▎        | 3302/24420 [1:22:03<8:41:04,  1.48s/it]
 14%|█▎        | 3303/24420 [1:22:04<8:41:04,  1.48s/it]
 14%|█▎        | 3304/24420 [1:22:06<8:40:49,  1.48s/it]
 14%|█▎        | 3305/24420 [1:22:07<8:40:37,  1.48s/it]
                                                        
{'loss': 0.163, 'learning_rate': 5e-05, 'epoch': 0.68}

 14%|█▎        | 3305/24420 [1:22:07<8:40:37,  1.48s/it]
 14%|█▎        | 3306/24420 [1:22:09<8:39:29,  1.48s/it]
 14%|█▎        | 3307/24420 [1:22:10<8:39:38,  1.48s/it]
 14%|█▎        | 3308/24420 [1:22:12<8:40:08,  1.48s/it]
 14%|█▎        | 3309/24420 [1:22:13<8:40:42,  1.48s/it]
 14%|█▎        | 3310/24420 [1:22:15<8:39:55,  1.48s/it]
                                                        
{'loss': 0.2034, 'learning_rate': 5e-05, 'epoch': 0.68}

 14%|█▎        | 3310/24420 [1:22:15<8:39:55,  1.48s/it]
 14%|█▎        | 3311/24420 [1:22:16<8:40:07,  1.48s/it]
 14%|█▎        | 3312/24420 [1:22:18<8:39:33,  1.48s/it]
 14%|█▎        | 3313/24420 [1:22:19<8:39:31,  1.48s/it]
 14%|█▎        | 3314/24420 [1:22:21<8:40:20,  1.48s/it]
 14%|█▎        | 3315/24420 [1:22:22<8:40:48,  1.48s/it]
                                                        
{'loss': 0.1909, 'learning_rate': 5e-05, 'epoch': 0.68}

 14%|█▎        | 3315/24420 [1:22:22<8:40:48,  1.48s/it]
 14%|█▎        | 3316/24420 [1:22:24<8:41:36,  1.48s/it]
 14%|█▎        | 3317/24420 [1:22:25<8:40:26,  1.48s/it]
 14%|█▎        | 3318/24420 [1:22:27<8:40:49,  1.48s/it]
 14%|█▎        | 3319/24420 [1:22:28<8:39:55,  1.48s/it]
 14%|█▎        | 3320/24420 [1:22:30<8:40:03,  1.48s/it]
                                                        
{'loss': 0.1708, 'learning_rate': 5e-05, 'epoch': 0.68}

 14%|█▎        | 3320/24420 [1:22:30<8:40:03,  1.48s/it]
 14%|█▎        | 3321/24420 [1:22:31<8:39:55,  1.48s/it]
 14%|█▎        | 3322/24420 [1:22:33<8:39:49,  1.48s/it]
 14%|█▎        | 3323/24420 [1:22:34<8:38:53,  1.48s/it]
 14%|█▎        | 3324/24420 [1:22:35<8:38:59,  1.48s/it]
 14%|█▎        | 3325/24420 [1:22:37<8:39:43,  1.48s/it]
                                                        
{'loss': 0.174, 'learning_rate': 5e-05, 'epoch': 0.68}

 14%|█▎        | 3325/24420 [1:22:37<8:39:43,  1.48s/it]
 14%|█▎        | 3326/24420 [1:22:38<8:39:29,  1.48s/it]
 14%|█▎        | 3327/24420 [1:22:40<8:40:58,  1.48s/it]
 14%|█▎        | 3328/24420 [1:22:41<8:41:12,  1.48s/it]
 14%|█▎        | 3329/24420 [1:22:43<8:39:53,  1.48s/it]
 14%|█▎        | 3330/24420 [1:22:44<8:39:41,  1.48s/it]
                                                        
{'loss': 0.2315, 'learning_rate': 5e-05, 'epoch': 0.68}

 14%|█▎        | 3330/24420 [1:22:44<8:39:41,  1.48s/it]
 14%|█▎        | 3331/24420 [1:22:46<8:39:19,  1.48s/it]
 14%|█▎        | 3332/24420 [1:22:47<8:39:18,  1.48s/it]
 14%|█▎        | 3333/24420 [1:22:49<8:39:50,  1.48s/it]
 14%|█▎        | 3334/24420 [1:22:50<8:39:39,  1.48s/it]
 14%|█▎        | 3335/24420 [1:22:52<8:39:18,  1.48s/it]
                                                        
{'loss': 0.1785, 'learning_rate': 5e-05, 'epoch': 0.68}

 14%|█▎        | 3335/24420 [1:22:52<8:39:18,  1.48s/it]
 14%|█▎        | 3336/24420 [1:22:53<8:39:26,  1.48s/it]
 14%|█▎        | 3337/24420 [1:22:55<8:39:01,  1.48s/it]
 14%|█▎        | 3338/24420 [1:22:56<8:39:07,  1.48s/it]
 14%|█▎        | 3339/24420 [1:22:58<8:39:37,  1.48s/it]
 14%|█▎        | 3340/24420 [1:22:59<8:40:08,  1.48s/it]
                                                        
{'loss': 0.1805, 'learning_rate': 5e-05, 'epoch': 0.68}

 14%|█▎        | 3340/24420 [1:22:59<8:40:08,  1.48s/it]
 14%|█▎        | 3341/24420 [1:23:01<8:40:07,  1.48s/it]
 14%|█▎        | 3342/24420 [1:23:02<8:39:23,  1.48s/it]
 14%|█▎        | 3343/24420 [1:23:04<8:39:15,  1.48s/it]
 14%|█▎        | 3344/24420 [1:23:05<8:39:24,  1.48s/it]
 14%|█▎        | 3345/24420 [1:23:07<8:39:28,  1.48s/it]
                                                        
{'loss': 0.1547, 'learning_rate': 5e-05, 'epoch': 0.68}

 14%|█▎        | 3345/24420 [1:23:07<8:39:28,  1.48s/it]
 14%|█▎        | 3346/24420 [1:23:08<8:39:24,  1.48s/it]
 14%|█▎        | 3347/24420 [1:23:09<8:38:55,  1.48s/it]
 14%|█▎        | 3348/24420 [1:23:11<8:38:45,  1.48s/it]
 14%|█▎        | 3349/24420 [1:23:12<8:39:38,  1.48s/it]
 14%|█▎        | 3350/24420 [1:23:14<8:38:56,  1.48s/it]
                                                        
{'loss': 0.1866, 'learning_rate': 5e-05, 'epoch': 0.69}

 14%|█▎        | 3350/24420 [1:23:14<8:38:56,  1.48s/it]
 14%|█▎        | 3351/24420 [1:23:15<8:39:40,  1.48s/it]
 14%|█▎        | 3352/24420 [1:23:17<8:39:48,  1.48s/it]
 14%|█▎        | 3353/24420 [1:23:18<8:39:44,  1.48s/it]
 14%|█▎        | 3354/24420 [1:23:20<8:39:17,  1.48s/it]
 14%|█▎        | 3355/24420 [1:23:21<8:39:29,  1.48s/it]
                                                        
{'loss': 0.1941, 'learning_rate': 5e-05, 'epoch': 0.69}

 14%|█▎        | 3355/24420 [1:23:21<8:39:29,  1.48s/it]
 14%|█▎        | 3356/24420 [1:23:23<8:39:07,  1.48s/it]
 14%|█▎        | 3357/24420 [1:23:24<8:38:36,  1.48s/it]
 14%|█▍        | 3358/24420 [1:23:26<8:38:43,  1.48s/it]
 14%|█▍        | 3359/24420 [1:23:27<8:38:14,  1.48s/it]
 14%|█▍        | 3360/24420 [1:23:29<8:38:07,  1.48s/it]
                                                        
{'loss': 0.1445, 'learning_rate': 5e-05, 'epoch': 0.69}

 14%|█▍        | 3360/24420 [1:23:29<8:38:07,  1.48s/it]
 14%|█▍        | 3361/24420 [1:23:30<8:38:40,  1.48s/it]
 14%|█▍        | 3362/24420 [1:23:32<8:38:45,  1.48s/it]
 14%|█▍        | 3363/24420 [1:23:33<8:38:22,  1.48s/it]
 14%|█▍        | 3364/24420 [1:23:35<8:38:38,  1.48s/it]
 14%|█▍        | 3365/24420 [1:23:36<8:38:17,  1.48s/it]
                                                        
{'loss': 0.1695, 'learning_rate': 5e-05, 'epoch': 0.69}

 14%|█▍        | 3365/24420 [1:23:36<8:38:17,  1.48s/it]
 14%|█▍        | 3366/24420 [1:23:38<8:38:22,  1.48s/it]
 14%|█▍        | 3367/24420 [1:23:39<8:36:53,  1.47s/it]
 14%|█▍        | 3368/24420 [1:23:41<8:36:29,  1.47s/it]
 14%|█▍        | 3369/24420 [1:23:42<8:36:41,  1.47s/it]
 14%|█▍        | 3370/24420 [1:23:43<8:37:07,  1.47s/it]
                                                        
{'loss': 0.1764, 'learning_rate': 5e-05, 'epoch': 0.69}

 14%|█▍        | 3370/24420 [1:23:43<8:37:07,  1.47s/it]
 14%|█▍        | 3371/24420 [1:23:45<8:38:17,  1.48s/it]
 14%|█▍        | 3372/24420 [1:23:46<8:38:34,  1.48s/it]
 14%|█▍        | 3373/24420 [1:23:48<8:38:59,  1.48s/it]
 14%|█▍        | 3374/24420 [1:23:49<8:40:22,  1.48s/it]
 14%|█▍        | 3375/24420 [1:23:51<8:39:57,  1.48s/it]
                                                        
{'loss': 0.2209, 'learning_rate': 5e-05, 'epoch': 0.69}

 14%|█▍        | 3375/24420 [1:23:51<8:39:57,  1.48s/it]
 14%|█▍        | 3376/24420 [1:23:52<8:39:38,  1.48s/it]
 14%|█▍        | 3377/24420 [1:23:54<8:39:50,  1.48s/it]
 14%|█▍        | 3378/24420 [1:23:55<8:40:03,  1.48s/it]
 14%|█▍        | 3379/24420 [1:23:57<8:39:33,  1.48s/it]
 14%|█▍        | 3380/24420 [1:23:58<8:38:41,  1.48s/it]
                                                        
{'loss': 0.1793, 'learning_rate': 5e-05, 'epoch': 0.69}

 14%|█▍        | 3380/24420 [1:23:58<8:38:41,  1.48s/it]
 14%|█▍        | 3381/24420 [1:24:00<8:38:48,  1.48s/it]
 14%|█▍        | 3382/24420 [1:24:01<8:38:22,  1.48s/it]
 14%|█▍        | 3383/24420 [1:24:03<8:38:39,  1.48s/it]
 14%|█▍        | 3384/24420 [1:24:04<8:38:45,  1.48s/it]
 14%|█▍        | 3385/24420 [1:24:06<8:37:49,  1.48s/it]
                                                        
{'loss': 0.2221, 'learning_rate': 5e-05, 'epoch': 0.69}

 14%|█▍        | 3385/24420 [1:24:06<8:37:49,  1.48s/it]
 14%|█▍        | 3386/24420 [1:24:07<8:38:09,  1.48s/it]
 14%|█▍        | 3387/24420 [1:24:09<8:38:25,  1.48s/it]
 14%|█▍        | 3388/24420 [1:24:10<8:38:06,  1.48s/it]
 14%|█▍        | 3389/24420 [1:24:12<8:37:33,  1.48s/it]
 14%|█▍        | 3390/24420 [1:24:13<8:37:35,  1.48s/it]
                                                        
{'loss': 0.2049, 'learning_rate': 5e-05, 'epoch': 0.69}

 14%|█▍        | 3390/24420 [1:24:13<8:37:35,  1.48s/it]
 14%|█▍        | 3391/24420 [1:24:15<8:37:24,  1.48s/it]
 14%|█▍        | 3392/24420 [1:24:16<8:37:13,  1.48s/it]
 14%|█▍        | 3393/24420 [1:24:17<8:37:04,  1.48s/it]
 14%|█▍        | 3394/24420 [1:24:19<8:36:58,  1.48s/it]
 14%|█▍        | 3395/24420 [1:24:20<8:36:35,  1.47s/it]
                                                        
{'loss': 0.1704, 'learning_rate': 5e-05, 'epoch': 0.7}

 14%|█▍        | 3395/24420 [1:24:20<8:36:35,  1.47s/it]
 14%|█▍        | 3396/24420 [1:24:22<8:36:52,  1.48s/it]
 14%|█▍        | 3397/24420 [1:24:23<8:38:01,  1.48s/it]
 14%|█▍        | 3398/24420 [1:24:25<8:37:09,  1.48s/it]
 14%|█▍        | 3399/24420 [1:24:26<8:36:26,  1.47s/it]
 14%|█▍        | 3400/24420 [1:24:28<8:36:17,  1.47s/it]
                                                        
{'loss': 0.1856, 'learning_rate': 5e-05, 'epoch': 0.7}

 14%|█▍        | 3400/24420 [1:24:28<8:36:17,  1.47s/it]
 14%|█▍        | 3401/24420 [1:24:32<13:57:39,  2.39s/it]
 14%|█▍        | 3402/24420 [1:24:34<12:22:27,  2.12s/it]
 14%|█▍        | 3403/24420 [1:24:35<11:14:17,  1.93s/it]
 14%|█▍        | 3404/24420 [1:24:37<10:26:40,  1.79s/it]
 14%|█▍        | 3405/24420 [1:24:38<9:53:34,  1.69s/it] 
                                                        
{'loss': 0.1675, 'learning_rate': 5e-05, 'epoch': 0.7}

 14%|█▍        | 3405/24420 [1:24:38<9:53:34,  1.69s/it]
 14%|█▍        | 3406/24420 [1:24:40<9:30:36,  1.63s/it]
 14%|█▍        | 3407/24420 [1:24:41<9:14:48,  1.58s/it]
 14%|█▍        | 3408/24420 [1:24:43<9:02:46,  1.55s/it]
 14%|█▍        | 3409/24420 [1:24:44<8:54:17,  1.53s/it]
 14%|█▍        | 3410/24420 [1:24:46<8:49:30,  1.51s/it]
                                                        
{'loss': 0.181, 'learning_rate': 5e-05, 'epoch': 0.7}

 14%|█▍        | 3410/24420 [1:24:46<8:49:30,  1.51s/it]
 14%|█▍        | 3411/24420 [1:24:47<8:45:40,  1.50s/it]
 14%|█▍        | 3412/24420 [1:24:49<8:43:36,  1.50s/it]
 14%|█▍        | 3413/24420 [1:24:50<8:40:57,  1.49s/it]
 14%|█▍        | 3414/24420 [1:24:52<8:39:20,  1.48s/it]
 14%|█▍        | 3415/24420 [1:24:53<8:38:14,  1.48s/it]
                                                        
{'loss': 0.1492, 'learning_rate': 5e-05, 'epoch': 0.7}

 14%|█▍        | 3415/24420 [1:24:53<8:38:14,  1.48s/it]
 14%|█▍        | 3416/24420 [1:24:54<8:37:57,  1.48s/it]
 14%|█▍        | 3417/24420 [1:24:56<8:37:51,  1.48s/it]
 14%|█▍        | 3418/24420 [1:24:58<8:52:46,  1.52s/it]
 14%|█▍        | 3419/24420 [1:24:59<8:48:00,  1.51s/it]
 14%|█▍        | 3420/24420 [1:25:01<8:45:45,  1.50s/it]
                                                        
{'loss': 0.2022, 'learning_rate': 5e-05, 'epoch': 0.7}

 14%|█▍        | 3420/24420 [1:25:01<8:45:45,  1.50s/it]
 14%|█▍        | 3421/24420 [1:25:02<8:43:08,  1.49s/it]
 14%|█▍        | 3422/24420 [1:25:03<8:41:45,  1.49s/it]
 14%|█▍        | 3423/24420 [1:25:05<8:40:47,  1.49s/it]
 14%|█▍        | 3424/24420 [1:25:06<8:39:20,  1.48s/it]
 14%|█▍        | 3425/24420 [1:25:08<8:38:27,  1.48s/it]
                                                        
{'loss': 0.1891, 'learning_rate': 5e-05, 'epoch': 0.7}

 14%|█▍        | 3425/24420 [1:25:08<8:38:27,  1.48s/it]
 14%|█▍        | 3426/24420 [1:25:09<8:38:18,  1.48s/it]
 14%|█▍        | 3427/24420 [1:25:11<8:38:04,  1.48s/it]
 14%|█▍        | 3428/24420 [1:25:12<8:38:10,  1.48s/it]
 14%|█▍        | 3429/24420 [1:25:14<8:37:56,  1.48s/it]
 14%|█▍        | 3430/24420 [1:25:15<8:37:34,  1.48s/it]
                                                        
{'loss': 0.2519, 'learning_rate': 5e-05, 'epoch': 0.7}

 14%|█▍        | 3430/24420 [1:25:15<8:37:34,  1.48s/it]
 14%|█▍        | 3431/24420 [1:25:17<8:38:17,  1.48s/it]
 14%|█▍        | 3432/24420 [1:25:18<8:37:59,  1.48s/it]
 14%|█▍        | 3433/24420 [1:25:20<8:38:20,  1.48s/it]
 14%|█▍        | 3434/24420 [1:25:21<8:37:01,  1.48s/it]
 14%|█▍        | 3435/24420 [1:25:23<8:36:52,  1.48s/it]
                                                        
{'loss': 0.2048, 'learning_rate': 5e-05, 'epoch': 0.7}

 14%|█▍        | 3435/24420 [1:25:23<8:36:52,  1.48s/it]
 14%|█▍        | 3436/24420 [1:25:24<8:37:07,  1.48s/it]
 14%|█▍        | 3437/24420 [1:25:26<8:36:56,  1.48s/it]
 14%|█▍        | 3438/24420 [1:25:27<8:37:15,  1.48s/it]
 14%|█▍        | 3439/24420 [1:25:29<8:36:58,  1.48s/it]
 14%|█▍        | 3440/24420 [1:25:30<8:37:31,  1.48s/it]
                                                        
{'loss': 0.2191, 'learning_rate': 5e-05, 'epoch': 0.7}

 14%|█▍        | 3440/24420 [1:25:30<8:37:31,  1.48s/it]
 14%|█▍        | 3441/24420 [1:25:32<8:37:53,  1.48s/it]
 14%|█▍        | 3442/24420 [1:25:33<8:37:37,  1.48s/it]
 14%|█▍        | 3443/24420 [1:25:35<8:39:51,  1.49s/it]
 14%|█▍        | 3444/24420 [1:25:36<8:39:13,  1.49s/it]
 14%|█▍        | 3445/24420 [1:25:38<8:38:30,  1.48s/it]
                                                        
{'loss': 0.1792, 'learning_rate': 5e-05, 'epoch': 0.71}

 14%|█▍        | 3445/24420 [1:25:38<8:38:30,  1.48s/it]
 14%|█▍        | 3446/24420 [1:25:39<8:38:14,  1.48s/it]
 14%|█▍        | 3447/24420 [1:25:40<8:36:51,  1.48s/it]
 14%|█▍        | 3448/24420 [1:25:42<8:36:47,  1.48s/it]
 14%|█▍        | 3449/24420 [1:25:43<8:36:36,  1.48s/it]
 14%|█▍        | 3450/24420 [1:25:45<8:35:54,  1.48s/it]
                                                        
{'loss': 0.1804, 'learning_rate': 5e-05, 'epoch': 0.71}

 14%|█▍        | 3450/24420 [1:25:45<8:35:54,  1.48s/it]
 14%|█▍        | 3451/24420 [1:25:46<8:35:38,  1.48s/it]
 14%|█▍        | 3452/24420 [1:25:48<8:35:32,  1.48s/it]
 14%|█▍        | 3453/24420 [1:25:49<8:35:11,  1.47s/it]
 14%|█▍        | 3454/24420 [1:25:51<8:36:08,  1.48s/it]
 14%|█▍        | 3455/24420 [1:25:52<8:37:16,  1.48s/it]
                                                        
{'loss': 0.1647, 'learning_rate': 5e-05, 'epoch': 0.71}

 14%|█▍        | 3455/24420 [1:25:52<8:37:16,  1.48s/it]
 14%|█▍        | 3456/24420 [1:25:54<8:37:17,  1.48s/it]
 14%|█▍        | 3457/24420 [1:25:55<8:36:34,  1.48s/it]
 14%|█▍        | 3458/24420 [1:25:57<8:35:44,  1.48s/it]
 14%|█▍        | 3459/24420 [1:25:58<8:35:45,  1.48s/it]
 14%|█▍        | 3460/24420 [1:26:00<8:35:55,  1.48s/it]
                                                        
{'loss': 0.1985, 'learning_rate': 5e-05, 'epoch': 0.71}

 14%|█▍        | 3460/24420 [1:26:00<8:35:55,  1.48s/it]
 14%|█▍        | 3461/24420 [1:26:01<8:37:01,  1.48s/it]
 14%|█▍        | 3462/24420 [1:26:03<8:36:18,  1.48s/it]
 14%|█▍        | 3463/24420 [1:26:04<8:36:09,  1.48s/it]
 14%|█▍        | 3464/24420 [1:26:06<8:36:18,  1.48s/it]
 14%|█▍        | 3465/24420 [1:26:07<8:36:03,  1.48s/it]
                                                        
{'loss': 0.167, 'learning_rate': 5e-05, 'epoch': 0.71}

 14%|█▍        | 3465/24420 [1:26:07<8:36:03,  1.48s/it]
 14%|█▍        | 3466/24420 [1:26:09<8:37:06,  1.48s/it]
 14%|█▍        | 3467/24420 [1:26:10<8:36:52,  1.48s/it]
 14%|█▍        | 3468/24420 [1:26:12<8:36:44,  1.48s/it]
 14%|█▍        | 3469/24420 [1:26:13<8:37:12,  1.48s/it]
 14%|█▍        | 3470/24420 [1:26:14<8:36:40,  1.48s/it]
                                                        
{'loss': 0.2003, 'learning_rate': 5e-05, 'epoch': 0.71}

 14%|█▍        | 3470/24420 [1:26:14<8:36:40,  1.48s/it]
 14%|█▍        | 3471/24420 [1:26:16<8:35:59,  1.48s/it]
 14%|█▍        | 3472/24420 [1:26:17<8:36:30,  1.48s/it]
 14%|█▍        | 3473/24420 [1:26:19<8:37:18,  1.48s/it]
 14%|█▍        | 3474/24420 [1:26:20<8:36:46,  1.48s/it]
 14%|█▍        | 3475/24420 [1:26:22<8:36:14,  1.48s/it]
                                                        
{'loss': 0.1911, 'learning_rate': 5e-05, 'epoch': 0.71}

 14%|█▍        | 3475/24420 [1:26:22<8:36:14,  1.48s/it]
 14%|█▍        | 3476/24420 [1:26:23<8:36:31,  1.48s/it]
 14%|█▍        | 3477/24420 [1:26:25<8:36:59,  1.48s/it]
 14%|█▍        | 3478/24420 [1:26:26<8:37:04,  1.48s/it]
 14%|█▍        | 3479/24420 [1:26:28<8:37:42,  1.48s/it]
 14%|█▍        | 3480/24420 [1:26:29<8:38:03,  1.48s/it]
                                                        
{'loss': 0.1864, 'learning_rate': 5e-05, 'epoch': 0.71}

 14%|█▍        | 3480/24420 [1:26:29<8:38:03,  1.48s/it]
 14%|█▍        | 3481/24420 [1:26:31<8:38:22,  1.49s/it]
 14%|█▍        | 3482/24420 [1:26:32<8:38:18,  1.49s/it]
 14%|█▍        | 3483/24420 [1:26:34<8:37:03,  1.48s/it]
 14%|█▍        | 3484/24420 [1:26:35<8:36:30,  1.48s/it]
 14%|█▍        | 3485/24420 [1:26:37<8:35:46,  1.48s/it]
                                                        
{'loss': 0.1909, 'learning_rate': 5e-05, 'epoch': 0.71}

 14%|█▍        | 3485/24420 [1:26:37<8:35:46,  1.48s/it]
 14%|█▍        | 3486/24420 [1:26:38<8:36:00,  1.48s/it]
 14%|█▍        | 3487/24420 [1:26:40<8:36:41,  1.48s/it]
 14%|█▍        | 3488/24420 [1:26:41<8:36:16,  1.48s/it]
 14%|█▍        | 3489/24420 [1:26:43<8:37:32,  1.48s/it]
 14%|█▍        | 3490/24420 [1:26:44<8:36:57,  1.48s/it]
                                                        
{'loss': 0.1788, 'learning_rate': 5e-05, 'epoch': 0.71}

 14%|█▍        | 3490/24420 [1:26:44<8:36:57,  1.48s/it]
 14%|█▍        | 3491/24420 [1:26:46<8:36:39,  1.48s/it]
 14%|█▍        | 3492/24420 [1:26:47<8:35:54,  1.48s/it]
 14%|█▍        | 3493/24420 [1:26:49<8:35:59,  1.48s/it]
 14%|█▍        | 3494/24420 [1:26:50<8:35:41,  1.48s/it]
 14%|█▍        | 3495/24420 [1:26:52<8:36:35,  1.48s/it]
                                                        
{'loss': 0.212, 'learning_rate': 5e-05, 'epoch': 0.72}

 14%|█▍        | 3495/24420 [1:26:52<8:36:35,  1.48s/it]
 14%|█▍        | 3496/24420 [1:26:53<8:35:58,  1.48s/it]
 14%|█▍        | 3497/24420 [1:26:54<8:35:43,  1.48s/it]
 14%|█▍        | 3498/24420 [1:26:56<8:36:18,  1.48s/it]
 14%|█▍        | 3499/24420 [1:26:57<8:35:43,  1.48s/it]
 14%|█▍        | 3500/24420 [1:26:59<8:35:17,  1.48s/it]
                                                        
{'loss': 0.1935, 'learning_rate': 5e-05, 'epoch': 0.72}

 14%|█▍        | 3500/24420 [1:26:59<8:35:17,  1.48s/it]
 14%|█▍        | 3501/24420 [1:27:00<8:35:20,  1.48s/it]
 14%|█▍        | 3502/24420 [1:27:02<8:36:35,  1.48s/it]
 14%|█▍        | 3503/24420 [1:27:03<8:35:41,  1.48s/it]
 14%|█▍        | 3504/24420 [1:27:05<8:35:54,  1.48s/it]
 14%|█▍        | 3505/24420 [1:27:06<8:36:43,  1.48s/it]
                                                        
{'loss': 0.1678, 'learning_rate': 5e-05, 'epoch': 0.72}

 14%|█▍        | 3505/24420 [1:27:06<8:36:43,  1.48s/it]
 14%|█▍        | 3506/24420 [1:27:08<8:36:20,  1.48s/it]
 14%|█▍        | 3507/24420 [1:27:09<8:35:22,  1.48s/it]
 14%|█▍        | 3508/24420 [1:27:11<8:35:27,  1.48s/it]
 14%|█▍        | 3509/24420 [1:27:12<8:35:28,  1.48s/it]
 14%|█▍        | 3510/24420 [1:27:14<8:36:08,  1.48s/it]
                                                        
{'loss': 0.1838, 'learning_rate': 5e-05, 'epoch': 0.72}

 14%|█▍        | 3510/24420 [1:27:14<8:36:08,  1.48s/it]
 14%|█▍        | 3511/24420 [1:27:15<8:35:09,  1.48s/it]
 14%|█▍        | 3512/24420 [1:27:17<8:35:05,  1.48s/it]
 14%|█▍        | 3513/24420 [1:27:18<8:35:29,  1.48s/it]
 14%|█▍        | 3514/24420 [1:27:20<8:35:32,  1.48s/it]
 14%|█▍        | 3515/24420 [1:27:21<8:35:25,  1.48s/it]
                                                        
{'loss': 0.1773, 'learning_rate': 5e-05, 'epoch': 0.72}

 14%|█▍        | 3515/24420 [1:27:21<8:35:25,  1.48s/it]
 14%|█▍        | 3516/24420 [1:27:23<8:35:21,  1.48s/it]
 14%|█▍        | 3517/24420 [1:27:24<8:36:10,  1.48s/it]
 14%|█▍        | 3518/24420 [1:27:26<8:36:18,  1.48s/it]
 14%|█▍        | 3519/24420 [1:27:27<8:36:20,  1.48s/it]
 14%|█▍        | 3520/24420 [1:27:29<8:36:28,  1.48s/it]
                                                        
{'loss': 0.1678, 'learning_rate': 5e-05, 'epoch': 0.72}

 14%|█▍        | 3520/24420 [1:27:29<8:36:28,  1.48s/it]
 14%|█▍        | 3521/24420 [1:27:30<8:36:41,  1.48s/it]
 14%|█▍        | 3522/24420 [1:27:31<8:35:51,  1.48s/it]
 14%|█▍        | 3523/24420 [1:27:33<8:36:06,  1.48s/it]
 14%|█▍        | 3524/24420 [1:27:34<8:35:36,  1.48s/it]
 14%|█▍        | 3525/24420 [1:27:36<8:35:39,  1.48s/it]
                                                        
{'loss': 0.1645, 'learning_rate': 5e-05, 'epoch': 0.72}

 14%|█▍        | 3525/24420 [1:27:36<8:35:39,  1.48s/it]
 14%|█▍        | 3526/24420 [1:27:37<8:35:42,  1.48s/it]
 14%|█▍        | 3527/24420 [1:27:39<8:35:39,  1.48s/it]
 14%|█▍        | 3528/24420 [1:27:40<8:35:34,  1.48s/it]
 14%|█▍        | 3529/24420 [1:27:42<8:36:40,  1.48s/it]
 14%|█▍        | 3530/24420 [1:27:43<8:37:13,  1.49s/it]
                                                        
{'loss': 0.1938, 'learning_rate': 5e-05, 'epoch': 0.72}

 14%|█▍        | 3530/24420 [1:27:43<8:37:13,  1.49s/it]
 14%|█▍        | 3531/24420 [1:27:45<8:36:04,  1.48s/it]
 14%|█▍        | 3532/24420 [1:27:46<8:36:16,  1.48s/it]
 14%|█▍        | 3533/24420 [1:27:48<8:36:43,  1.48s/it]
 14%|█▍        | 3534/24420 [1:27:49<8:35:52,  1.48s/it]
 14%|█▍        | 3535/24420 [1:27:51<8:36:22,  1.48s/it]
                                                        
{'loss': 0.1788, 'learning_rate': 5e-05, 'epoch': 0.72}

 14%|█▍        | 3535/24420 [1:27:51<8:36:22,  1.48s/it]
 14%|█▍        | 3536/24420 [1:27:52<8:36:28,  1.48s/it]
 14%|█▍        | 3537/24420 [1:27:54<8:39:21,  1.49s/it]
 14%|█▍        | 3538/24420 [1:27:55<8:40:02,  1.49s/it]
 14%|█▍        | 3539/24420 [1:27:57<8:39:08,  1.49s/it]
 14%|█▍        | 3540/24420 [1:27:58<8:37:13,  1.49s/it]
                                                        
{'loss': 0.168, 'learning_rate': 5e-05, 'epoch': 0.72}

 14%|█▍        | 3540/24420 [1:27:58<8:37:13,  1.49s/it]
 15%|█▍        | 3541/24420 [1:28:00<8:36:23,  1.48s/it]
 15%|█▍        | 3542/24420 [1:28:01<8:35:25,  1.48s/it]
 15%|█▍        | 3543/24420 [1:28:03<8:36:05,  1.48s/it]
 15%|█▍        | 3544/24420 [1:28:04<8:37:42,  1.49s/it]
 15%|█▍        | 3545/24420 [1:28:06<8:36:58,  1.49s/it]
                                                        
{'loss': 0.174, 'learning_rate': 5e-05, 'epoch': 0.73}

 15%|█▍        | 3545/24420 [1:28:06<8:36:58,  1.49s/it]
 15%|█▍        | 3546/24420 [1:28:07<8:37:25,  1.49s/it]
 15%|█▍        | 3547/24420 [1:28:09<8:36:24,  1.48s/it]
 15%|█▍        | 3548/24420 [1:28:10<8:36:13,  1.48s/it]
 15%|█▍        | 3549/24420 [1:28:12<8:36:22,  1.48s/it]
 15%|█▍        | 3550/24420 [1:28:13<8:35:35,  1.48s/it]
                                                        
{'loss': 0.1663, 'learning_rate': 5e-05, 'epoch': 0.73}

 15%|█▍        | 3550/24420 [1:28:13<8:35:35,  1.48s/it]
 15%|█▍        | 3551/24420 [1:28:15<8:35:34,  1.48s/it]
 15%|█▍        | 3552/24420 [1:28:16<8:34:59,  1.48s/it]
 15%|█▍        | 3553/24420 [1:28:17<8:34:25,  1.48s/it]
 15%|█▍        | 3554/24420 [1:28:19<8:33:53,  1.48s/it]
 15%|█▍        | 3555/24420 [1:28:20<8:35:56,  1.48s/it]
                                                        
{'loss': 0.1936, 'learning_rate': 5e-05, 'epoch': 0.73}

 15%|█▍        | 3555/24420 [1:28:20<8:35:56,  1.48s/it]
 15%|█▍        | 3556/24420 [1:28:22<8:36:20,  1.48s/it]
 15%|█▍        | 3557/24420 [1:28:23<8:35:28,  1.48s/it]
 15%|█▍        | 3558/24420 [1:28:25<8:35:19,  1.48s/it]
 15%|█▍        | 3559/24420 [1:28:26<8:35:21,  1.48s/it]
 15%|█▍        | 3560/24420 [1:28:28<8:34:27,  1.48s/it]
                                                        
{'loss': 0.1786, 'learning_rate': 5e-05, 'epoch': 0.73}

 15%|█▍        | 3560/24420 [1:28:28<8:34:27,  1.48s/it]
 15%|█▍        | 3561/24420 [1:28:29<8:33:56,  1.48s/it]
 15%|█▍        | 3562/24420 [1:28:31<8:33:36,  1.48s/it]
 15%|█▍        | 3563/24420 [1:28:32<8:33:15,  1.48s/it]
 15%|█▍        | 3564/24420 [1:28:34<8:33:19,  1.48s/it]
 15%|█▍        | 3565/24420 [1:28:35<8:33:52,  1.48s/it]
                                                        
{'loss': 0.1729, 'learning_rate': 5e-05, 'epoch': 0.73}

 15%|█▍        | 3565/24420 [1:28:35<8:33:52,  1.48s/it]
 15%|█▍        | 3566/24420 [1:28:37<8:34:17,  1.48s/it]
 15%|█▍        | 3567/24420 [1:28:38<8:34:31,  1.48s/it]
 15%|█▍        | 3568/24420 [1:28:40<8:33:36,  1.48s/it]
 15%|█▍        | 3569/24420 [1:28:41<8:32:56,  1.48s/it]
 15%|█▍        | 3570/24420 [1:28:43<8:33:02,  1.48s/it]
                                                        
{'loss': 0.1937, 'learning_rate': 5e-05, 'epoch': 0.73}

 15%|█▍        | 3570/24420 [1:28:43<8:33:02,  1.48s/it]
 15%|█▍        | 3571/24420 [1:28:44<8:33:45,  1.48s/it]
 15%|█▍        | 3572/24420 [1:28:46<8:32:54,  1.48s/it]
 15%|█▍        | 3573/24420 [1:28:47<8:34:10,  1.48s/it]
 15%|█▍        | 3574/24420 [1:28:49<8:33:38,  1.48s/it]
 15%|█▍        | 3575/24420 [1:28:50<8:32:44,  1.48s/it]
                                                        
{'loss': 0.173, 'learning_rate': 5e-05, 'epoch': 0.73}

 15%|█▍        | 3575/24420 [1:28:50<8:32:44,  1.48s/it]
 15%|█▍        | 3576/24420 [1:28:51<8:32:48,  1.48s/it]
 15%|█▍        | 3577/24420 [1:28:53<8:32:48,  1.48s/it]
 15%|█▍        | 3578/24420 [1:28:54<8:33:32,  1.48s/it]
 15%|█▍        | 3579/24420 [1:28:56<8:33:16,  1.48s/it]
 15%|█▍        | 3580/24420 [1:28:57<8:33:38,  1.48s/it]
                                                        
{'loss': 0.1898, 'learning_rate': 5e-05, 'epoch': 0.73}

 15%|█▍        | 3580/24420 [1:28:57<8:33:38,  1.48s/it]
 15%|█▍        | 3581/24420 [1:28:59<8:33:16,  1.48s/it]
 15%|█▍        | 3582/24420 [1:29:00<8:32:38,  1.48s/it]
 15%|█▍        | 3583/24420 [1:29:02<8:33:07,  1.48s/it]
 15%|█▍        | 3584/24420 [1:29:03<8:32:36,  1.48s/it]
 15%|█▍        | 3585/24420 [1:29:05<8:32:34,  1.48s/it]
                                                        
{'loss': 0.1733, 'learning_rate': 5e-05, 'epoch': 0.73}

 15%|█▍        | 3585/24420 [1:29:05<8:32:34,  1.48s/it]
 15%|█▍        | 3586/24420 [1:29:06<8:32:34,  1.48s/it]
 15%|█▍        | 3587/24420 [1:29:08<8:33:35,  1.48s/it]
 15%|█▍        | 3588/24420 [1:29:09<8:34:02,  1.48s/it]
 15%|█▍        | 3589/24420 [1:29:11<8:34:19,  1.48s/it]
 15%|█▍        | 3590/24420 [1:29:12<8:34:38,  1.48s/it]
                                                        
{'loss': 0.1819, 'learning_rate': 5e-05, 'epoch': 0.74}

 15%|█▍        | 3590/24420 [1:29:12<8:34:38,  1.48s/it]
 15%|█▍        | 3591/24420 [1:29:14<8:36:04,  1.49s/it]
 15%|█▍        | 3592/24420 [1:29:15<8:34:58,  1.48s/it]
 15%|█▍        | 3593/24420 [1:29:17<8:33:23,  1.48s/it]
 15%|█▍        | 3594/24420 [1:29:18<8:33:32,  1.48s/it]
 15%|█▍        | 3595/24420 [1:29:20<8:33:11,  1.48s/it]
                                                        
{'loss': 0.1815, 'learning_rate': 5e-05, 'epoch': 0.74}

 15%|█▍        | 3595/24420 [1:29:20<8:33:11,  1.48s/it]
 15%|█▍        | 3596/24420 [1:29:21<8:33:09,  1.48s/it]
 15%|█▍        | 3597/24420 [1:29:23<8:33:41,  1.48s/it]
 15%|█▍        | 3598/24420 [1:29:24<8:33:13,  1.48s/it]
 15%|█▍        | 3599/24420 [1:29:26<8:32:39,  1.48s/it]
 15%|█▍        | 3600/24420 [1:29:27<8:32:41,  1.48s/it]
                                                        
{'loss': 0.18, 'learning_rate': 5e-05, 'epoch': 0.74}

 15%|█▍        | 3600/24420 [1:29:27<8:32:41,  1.48s/it]
 15%|█▍        | 3601/24420 [1:29:32<14:10:39,  2.45s/it]
 15%|█▍        | 3602/24420 [1:29:33<12:30:24,  2.16s/it]
 15%|█▍        | 3603/24420 [1:29:35<11:18:40,  1.96s/it]
 15%|█▍        | 3604/24420 [1:29:36<10:28:59,  1.81s/it]
 15%|█▍        | 3605/24420 [1:29:38<9:54:22,  1.71s/it] 
                                                        
{'loss': 0.2057, 'learning_rate': 5e-05, 'epoch': 0.74}

 15%|█▍        | 3605/24420 [1:29:38<9:54:22,  1.71s/it]
 15%|█▍        | 3606/24420 [1:29:39<9:30:20,  1.64s/it]
 15%|█▍        | 3607/24420 [1:29:41<9:11:28,  1.59s/it]
 15%|█▍        | 3608/24420 [1:29:42<8:59:09,  1.55s/it]
 15%|█▍        | 3609/24420 [1:29:44<8:50:10,  1.53s/it]
 15%|█▍        | 3610/24420 [1:29:45<8:45:10,  1.51s/it]
                                                        
{'loss': 0.1724, 'learning_rate': 5e-05, 'epoch': 0.74}

 15%|█▍        | 3610/24420 [1:29:45<8:45:10,  1.51s/it]
 15%|█▍        | 3611/24420 [1:29:46<8:40:58,  1.50s/it]
 15%|█▍        | 3612/24420 [1:29:48<8:37:59,  1.49s/it]
 15%|█▍        | 3613/24420 [1:29:49<8:35:47,  1.49s/it]
 15%|█▍        | 3614/24420 [1:29:51<8:34:31,  1.48s/it]
 15%|█▍        | 3615/24420 [1:29:52<8:33:48,  1.48s/it]
                                                        
{'loss': 0.1488, 'learning_rate': 5e-05, 'epoch': 0.74}

 15%|█▍        | 3615/24420 [1:29:52<8:33:48,  1.48s/it]
 15%|█▍        | 3616/24420 [1:29:54<8:32:56,  1.48s/it]
 15%|█▍        | 3617/24420 [1:29:55<8:32:25,  1.48s/it]
 15%|█▍        | 3618/24420 [1:29:57<8:32:41,  1.48s/it]
 15%|█▍        | 3619/24420 [1:29:58<8:32:09,  1.48s/it]
 15%|█▍        | 3620/24420 [1:30:00<8:31:53,  1.48s/it]
                                                        
{'loss': 0.1971, 'learning_rate': 5e-05, 'epoch': 0.74}

 15%|█▍        | 3620/24420 [1:30:00<8:31:53,  1.48s/it]
 15%|█▍        | 3621/24420 [1:30:01<8:32:04,  1.48s/it]
 15%|█▍        | 3622/24420 [1:30:03<8:31:37,  1.48s/it]
 15%|█▍        | 3623/24420 [1:30:04<8:30:58,  1.47s/it]
 15%|█▍        | 3624/24420 [1:30:06<8:31:24,  1.48s/it]
 15%|█▍        | 3625/24420 [1:30:07<8:32:11,  1.48s/it]
                                                        
{'loss': 0.203, 'learning_rate': 5e-05, 'epoch': 0.74}

 15%|█▍        | 3625/24420 [1:30:07<8:32:11,  1.48s/it]
 15%|█▍        | 3626/24420 [1:30:09<8:34:06,  1.48s/it]
 15%|█▍        | 3627/24420 [1:30:10<8:33:15,  1.48s/it]
 15%|█▍        | 3628/24420 [1:30:12<8:33:48,  1.48s/it]
 15%|█▍        | 3629/24420 [1:30:13<8:33:27,  1.48s/it]
 15%|█▍        | 3630/24420 [1:30:15<8:32:36,  1.48s/it]
                                                        
{'loss': 0.1689, 'learning_rate': 5e-05, 'epoch': 0.74}

 15%|█▍        | 3630/24420 [1:30:15<8:32:36,  1.48s/it]
 15%|█▍        | 3631/24420 [1:30:16<8:32:19,  1.48s/it]
 15%|█▍        | 3632/24420 [1:30:18<8:33:19,  1.48s/it]
 15%|█▍        | 3633/24420 [1:30:19<8:33:15,  1.48s/it]
 15%|█▍        | 3634/24420 [1:30:20<8:33:26,  1.48s/it]
 15%|█▍        | 3635/24420 [1:30:22<8:33:08,  1.48s/it]
                                                        
{'loss': 0.1868, 'learning_rate': 5e-05, 'epoch': 0.74}

 15%|█▍        | 3635/24420 [1:30:22<8:33:08,  1.48s/it]
 15%|█▍        | 3636/24420 [1:30:23<8:32:05,  1.48s/it]
 15%|█▍        | 3637/24420 [1:30:25<8:31:20,  1.48s/it]
 15%|█▍        | 3638/24420 [1:30:26<8:31:14,  1.48s/it]
 15%|█▍        | 3639/24420 [1:30:28<8:32:03,  1.48s/it]
 15%|█▍        | 3640/24420 [1:30:29<8:31:59,  1.48s/it]
                                                        
{'loss': 0.1621, 'learning_rate': 5e-05, 'epoch': 0.75}

 15%|█▍        | 3640/24420 [1:30:29<8:31:59,  1.48s/it]
 15%|█▍        | 3641/24420 [1:30:31<8:31:34,  1.48s/it]
 15%|█▍        | 3642/24420 [1:30:32<8:31:06,  1.48s/it]
 15%|█▍        | 3643/24420 [1:30:34<8:31:04,  1.48s/it]
 15%|█▍        | 3644/24420 [1:30:35<8:31:03,  1.48s/it]
 15%|█▍        | 3645/24420 [1:30:37<8:30:19,  1.47s/it]
                                                        
{'loss': 0.1781, 'learning_rate': 5e-05, 'epoch': 0.75}

 15%|█▍        | 3645/24420 [1:30:37<8:30:19,  1.47s/it]
 15%|█▍        | 3646/24420 [1:30:38<8:30:41,  1.48s/it]
 15%|█▍        | 3647/24420 [1:30:40<8:30:42,  1.48s/it]
 15%|█▍        | 3648/24420 [1:30:41<8:30:40,  1.48s/it]
 15%|█▍        | 3649/24420 [1:30:43<8:30:50,  1.48s/it]
 15%|█▍        | 3650/24420 [1:30:44<8:30:43,  1.48s/it]
                                                        
{'loss': 0.1892, 'learning_rate': 5e-05, 'epoch': 0.75}

 15%|█▍        | 3650/24420 [1:30:44<8:30:43,  1.48s/it]
 15%|█▍        | 3651/24420 [1:30:46<8:30:20,  1.47s/it]
 15%|█▍        | 3652/24420 [1:30:47<8:30:26,  1.47s/it]
 15%|█▍        | 3653/24420 [1:30:49<8:30:25,  1.47s/it]
 15%|█▍        | 3654/24420 [1:30:50<8:30:54,  1.48s/it]
 15%|█▍        | 3655/24420 [1:30:51<8:31:22,  1.48s/it]
                                                        
{'loss': 0.1528, 'learning_rate': 5e-05, 'epoch': 0.75}

 15%|█▍        | 3655/24420 [1:30:51<8:31:22,  1.48s/it]
 15%|█▍        | 3656/24420 [1:30:53<8:32:11,  1.48s/it]
 15%|█▍        | 3657/24420 [1:30:54<8:31:49,  1.48s/it]
 15%|█▍        | 3658/24420 [1:30:56<8:31:27,  1.48s/it]
 15%|█▍        | 3659/24420 [1:30:57<8:31:18,  1.48s/it]
 15%|█▍        | 3660/24420 [1:30:59<8:32:39,  1.48s/it]
                                                        
{'loss': 0.1681, 'learning_rate': 5e-05, 'epoch': 0.75}

 15%|█▍        | 3660/24420 [1:30:59<8:32:39,  1.48s/it]
 15%|█▍        | 3661/24420 [1:31:00<8:32:27,  1.48s/it]
 15%|█▍        | 3662/24420 [1:31:02<8:32:29,  1.48s/it]
 15%|█▌        | 3663/24420 [1:31:03<8:32:30,  1.48s/it]
 15%|█▌        | 3664/24420 [1:31:05<8:32:14,  1.48s/it]
 15%|█▌        | 3665/24420 [1:31:06<8:31:49,  1.48s/it]
                                                        
{'loss': 0.1941, 'learning_rate': 5e-05, 'epoch': 0.75}

 15%|█▌        | 3665/24420 [1:31:06<8:31:49,  1.48s/it]
 15%|█▌        | 3666/24420 [1:31:08<8:31:38,  1.48s/it]
 15%|█▌        | 3667/24420 [1:31:09<8:31:24,  1.48s/it]
 15%|█▌        | 3668/24420 [1:31:11<8:31:10,  1.48s/it]
 15%|█▌        | 3669/24420 [1:31:12<8:31:05,  1.48s/it]
 15%|█▌        | 3670/24420 [1:31:14<8:31:08,  1.48s/it]
                                                        
{'loss': 0.1651, 'learning_rate': 5e-05, 'epoch': 0.75}

 15%|█▌        | 3670/24420 [1:31:14<8:31:08,  1.48s/it]
 15%|█▌        | 3671/24420 [1:31:15<8:31:29,  1.48s/it]
 15%|█▌        | 3672/24420 [1:31:17<8:30:24,  1.48s/it]
 15%|█▌        | 3673/24420 [1:31:18<8:30:39,  1.48s/it]
 15%|█▌        | 3674/24420 [1:31:20<8:31:10,  1.48s/it]
 15%|█▌        | 3675/24420 [1:31:21<8:31:44,  1.48s/it]
                                                        
{'loss': 0.1432, 'learning_rate': 5e-05, 'epoch': 0.75}

 15%|█▌        | 3675/24420 [1:31:21<8:31:44,  1.48s/it]
 15%|█▌        | 3676/24420 [1:31:23<8:32:13,  1.48s/it]
 15%|█▌        | 3677/24420 [1:31:24<8:32:16,  1.48s/it]
 15%|█▌        | 3678/24420 [1:31:25<8:31:42,  1.48s/it]
 15%|█▌        | 3679/24420 [1:31:27<8:31:45,  1.48s/it]
 15%|█▌        | 3680/24420 [1:31:28<8:31:21,  1.48s/it]
                                                        
{'loss': 0.1683, 'learning_rate': 5e-05, 'epoch': 0.75}

 15%|█▌        | 3680/24420 [1:31:28<8:31:21,  1.48s/it]
 15%|█▌        | 3681/24420 [1:31:30<8:30:48,  1.48s/it]
 15%|█▌        | 3682/24420 [1:31:31<8:30:25,  1.48s/it]
 15%|█▌        | 3683/24420 [1:31:33<8:29:51,  1.48s/it]
 15%|█▌        | 3684/24420 [1:31:34<8:31:07,  1.48s/it]
 15%|█▌        | 3685/24420 [1:31:36<8:31:03,  1.48s/it]
                                                        
{'loss': 0.2027, 'learning_rate': 5e-05, 'epoch': 0.75}

 15%|█▌        | 3685/24420 [1:31:36<8:31:03,  1.48s/it]
 15%|█▌        | 3686/24420 [1:31:37<8:31:06,  1.48s/it]
 15%|█▌        | 3687/24420 [1:31:39<8:30:46,  1.48s/it]
 15%|█▌        | 3688/24420 [1:31:40<8:30:55,  1.48s/it]
 15%|█▌        | 3689/24420 [1:31:42<8:31:19,  1.48s/it]
 15%|█▌        | 3690/24420 [1:31:43<8:30:57,  1.48s/it]
                                                        
{'loss': 0.1563, 'learning_rate': 5e-05, 'epoch': 0.76}

 15%|█▌        | 3690/24420 [1:31:43<8:30:57,  1.48s/it]
 15%|█▌        | 3691/24420 [1:31:45<8:30:31,  1.48s/it]
 15%|█▌        | 3692/24420 [1:31:46<8:29:42,  1.48s/it]
 15%|█▌        | 3693/24420 [1:31:48<8:29:36,  1.48s/it]
 15%|█▌        | 3694/24420 [1:31:49<8:29:34,  1.48s/it]
 15%|█▌        | 3695/24420 [1:31:51<8:29:59,  1.48s/it]
                                                        
{'loss': 0.1686, 'learning_rate': 5e-05, 'epoch': 0.76}

 15%|█▌        | 3695/24420 [1:31:51<8:29:59,  1.48s/it]
 15%|█▌        | 3696/24420 [1:31:52<8:30:07,  1.48s/it]
 15%|█▌        | 3697/24420 [1:31:54<8:31:07,  1.48s/it]
 15%|█▌        | 3698/24420 [1:31:55<8:31:07,  1.48s/it]
 15%|█▌        | 3699/24420 [1:31:57<8:31:31,  1.48s/it]
 15%|█▌        | 3700/24420 [1:31:58<8:30:29,  1.48s/it]
                                                        
{'loss': 0.1368, 'learning_rate': 5e-05, 'epoch': 0.76}

 15%|█▌        | 3700/24420 [1:31:58<8:30:29,  1.48s/it]
 15%|█▌        | 3701/24420 [1:32:00<8:35:43,  1.49s/it]
 15%|█▌        | 3702/24420 [1:32:01<8:33:42,  1.49s/it]
 15%|█▌        | 3703/24420 [1:32:02<8:33:01,  1.49s/it]
 15%|█▌        | 3704/24420 [1:32:04<8:31:46,  1.48s/it]
 15%|█▌        | 3705/24420 [1:32:05<8:30:48,  1.48s/it]
                                                        
{'loss': 0.1851, 'learning_rate': 5e-05, 'epoch': 0.76}

 15%|█▌        | 3705/24420 [1:32:05<8:30:48,  1.48s/it]
 15%|█▌        | 3706/24420 [1:32:07<8:31:02,  1.48s/it]
 15%|█▌        | 3707/24420 [1:32:08<8:31:53,  1.48s/it]
 15%|█▌        | 3708/24420 [1:32:10<8:31:22,  1.48s/it]
 15%|█▌        | 3709/24420 [1:32:11<8:31:40,  1.48s/it]
 15%|█▌        | 3710/24420 [1:32:13<8:31:17,  1.48s/it]
                                                        
{'loss': 0.1767, 'learning_rate': 5e-05, 'epoch': 0.76}

 15%|█▌        | 3710/24420 [1:32:13<8:31:17,  1.48s/it]
 15%|█▌        | 3711/24420 [1:32:14<8:31:27,  1.48s/it]
 15%|█▌        | 3712/24420 [1:32:16<8:30:45,  1.48s/it]
 15%|█▌        | 3713/24420 [1:32:17<8:30:49,  1.48s/it]
 15%|█▌        | 3714/24420 [1:32:19<8:31:40,  1.48s/it]
 15%|█▌        | 3715/24420 [1:32:20<8:31:20,  1.48s/it]
                                                        
{'loss': 0.1576, 'learning_rate': 5e-05, 'epoch': 0.76}

 15%|█▌        | 3715/24420 [1:32:20<8:31:20,  1.48s/it]
 15%|█▌        | 3716/24420 [1:32:22<8:31:39,  1.48s/it]
 15%|█▌        | 3717/24420 [1:32:23<8:30:46,  1.48s/it]
 15%|█▌        | 3718/24420 [1:32:25<8:30:14,  1.48s/it]
 15%|█▌        | 3719/24420 [1:32:26<8:30:17,  1.48s/it]
 15%|█▌        | 3720/24420 [1:32:28<8:30:46,  1.48s/it]
                                                        
{'loss': 0.1605, 'learning_rate': 5e-05, 'epoch': 0.76}

 15%|█▌        | 3720/24420 [1:32:28<8:30:46,  1.48s/it]
 15%|█▌        | 3721/24420 [1:32:29<8:31:09,  1.48s/it]
 15%|█▌        | 3722/24420 [1:32:31<8:30:35,  1.48s/it]
 15%|█▌        | 3723/24420 [1:32:32<8:30:37,  1.48s/it]
 15%|█▌        | 3724/24420 [1:32:34<8:30:10,  1.48s/it]
 15%|█▌        | 3725/24420 [1:32:35<8:30:01,  1.48s/it]
                                                        
{'loss': 0.1924, 'learning_rate': 5e-05, 'epoch': 0.76}

 15%|█▌        | 3725/24420 [1:32:35<8:30:01,  1.48s/it]
 15%|█▌        | 3726/24420 [1:32:37<8:31:10,  1.48s/it]
 15%|█▌        | 3727/24420 [1:32:38<8:31:09,  1.48s/it]
 15%|█▌        | 3728/24420 [1:32:40<8:30:47,  1.48s/it]
 15%|█▌        | 3729/24420 [1:32:41<8:30:48,  1.48s/it]
 15%|█▌        | 3730/24420 [1:32:42<8:30:46,  1.48s/it]
                                                        
{'loss': 0.1739, 'learning_rate': 5e-05, 'epoch': 0.76}

 15%|█▌        | 3730/24420 [1:32:42<8:30:46,  1.48s/it]
 15%|█▌        | 3731/24420 [1:32:44<8:30:15,  1.48s/it]
 15%|█▌        | 3732/24420 [1:32:45<8:30:05,  1.48s/it]
 15%|█▌        | 3733/24420 [1:32:47<8:29:20,  1.48s/it]
 15%|█▌        | 3734/24420 [1:32:48<8:29:38,  1.48s/it]
 15%|█▌        | 3735/24420 [1:32:50<8:29:22,  1.48s/it]
                                                        
{'loss': 0.1592, 'learning_rate': 5e-05, 'epoch': 0.76}

 15%|█▌        | 3735/24420 [1:32:50<8:29:22,  1.48s/it]
 15%|█▌        | 3736/24420 [1:32:51<8:30:01,  1.48s/it]
 15%|█▌        | 3737/24420 [1:32:53<8:30:30,  1.48s/it]
 15%|█▌        | 3738/24420 [1:32:54<8:30:22,  1.48s/it]
 15%|█▌        | 3739/24420 [1:32:56<8:30:26,  1.48s/it]
 15%|█▌        | 3740/24420 [1:32:57<8:30:14,  1.48s/it]
                                                        
{'loss': 0.193, 'learning_rate': 5e-05, 'epoch': 0.77}

 15%|█▌        | 3740/24420 [1:32:57<8:30:14,  1.48s/it]
 15%|█▌        | 3741/24420 [1:32:59<8:29:25,  1.48s/it]
 15%|█▌        | 3742/24420 [1:33:00<8:29:35,  1.48s/it]
 15%|█▌        | 3743/24420 [1:33:02<8:30:40,  1.48s/it]
 15%|█▌        | 3744/24420 [1:33:03<8:30:01,  1.48s/it]
 15%|█▌        | 3745/24420 [1:33:05<8:29:59,  1.48s/it]
                                                        
{'loss': 0.1654, 'learning_rate': 5e-05, 'epoch': 0.77}

 15%|█▌        | 3745/24420 [1:33:05<8:29:59,  1.48s/it]
 15%|█▌        | 3746/24420 [1:33:06<8:29:38,  1.48s/it]
 15%|█▌        | 3747/24420 [1:33:08<8:30:05,  1.48s/it]
 15%|█▌        | 3748/24420 [1:33:09<8:30:05,  1.48s/it]
 15%|█▌        | 3749/24420 [1:33:11<8:30:29,  1.48s/it]
 15%|█▌        | 3750/24420 [1:33:12<8:30:22,  1.48s/it]
                                                        
{'loss': 0.1734, 'learning_rate': 5e-05, 'epoch': 0.77}

 15%|█▌        | 3750/24420 [1:33:12<8:30:22,  1.48s/it]
 15%|█▌        | 3751/24420 [1:33:14<8:31:16,  1.48s/it]
 15%|█▌        | 3752/24420 [1:33:15<8:31:10,  1.48s/it]
 15%|█▌        | 3753/24420 [1:33:17<8:31:33,  1.49s/it]
 15%|█▌        | 3754/24420 [1:33:18<8:31:06,  1.48s/it]
 15%|█▌        | 3755/24420 [1:33:20<8:31:45,  1.49s/it]
                                                        
{'loss': 0.1581, 'learning_rate': 5e-05, 'epoch': 0.77}

 15%|█▌        | 3755/24420 [1:33:20<8:31:45,  1.49s/it]
 15%|█▌        | 3756/24420 [1:33:21<8:32:07,  1.49s/it]
 15%|█▌        | 3757/24420 [1:33:22<8:31:14,  1.48s/it]
 15%|█▌        | 3758/24420 [1:33:24<8:30:46,  1.48s/it]
 15%|█▌        | 3759/24420 [1:33:25<8:31:12,  1.48s/it]
 15%|█▌        | 3760/24420 [1:33:27<8:31:31,  1.49s/it]
                                                        
{'loss': 0.1608, 'learning_rate': 5e-05, 'epoch': 0.77}

 15%|█▌        | 3760/24420 [1:33:27<8:31:31,  1.49s/it]
 15%|█▌        | 3761/24420 [1:33:28<8:31:24,  1.49s/it]
 15%|█▌        | 3762/24420 [1:33:30<8:31:17,  1.49s/it]
 15%|█▌        | 3763/24420 [1:33:31<8:30:51,  1.48s/it]
 15%|█▌        | 3764/24420 [1:33:33<8:31:14,  1.49s/it]
 15%|█▌        | 3765/24420 [1:33:34<8:32:05,  1.49s/it]
                                                        
{'loss': 0.1675, 'learning_rate': 5e-05, 'epoch': 0.77}

 15%|█▌        | 3765/24420 [1:33:34<8:32:05,  1.49s/it]
 15%|█▌        | 3766/24420 [1:33:36<8:33:30,  1.49s/it]
 15%|█▌        | 3767/24420 [1:33:37<8:32:58,  1.49s/it]
 15%|█▌        | 3768/24420 [1:33:39<8:32:26,  1.49s/it]
 15%|█▌        | 3769/24420 [1:33:40<8:31:46,  1.49s/it]
 15%|█▌        | 3770/24420 [1:33:42<8:31:37,  1.49s/it]
                                                        
{'loss': 0.1659, 'learning_rate': 5e-05, 'epoch': 0.77}

 15%|█▌        | 3770/24420 [1:33:42<8:31:37,  1.49s/it]
 15%|█▌        | 3771/24420 [1:33:43<8:33:29,  1.49s/it]
 15%|█▌        | 3772/24420 [1:33:45<8:31:31,  1.49s/it]
 15%|█▌        | 3773/24420 [1:33:46<8:31:12,  1.49s/it]
 15%|█▌        | 3774/24420 [1:33:48<8:31:01,  1.49s/it]
 15%|█▌        | 3775/24420 [1:33:49<8:30:36,  1.48s/it]
                                                        
{'loss': 0.1636, 'learning_rate': 5e-05, 'epoch': 0.77}

 15%|█▌        | 3775/24420 [1:33:49<8:30:36,  1.48s/it]
 15%|█▌        | 3776/24420 [1:33:51<8:30:05,  1.48s/it]
 15%|█▌        | 3777/24420 [1:33:52<8:29:49,  1.48s/it]
 15%|█▌        | 3778/24420 [1:33:54<8:29:21,  1.48s/it]
 15%|█▌        | 3779/24420 [1:33:55<8:29:00,  1.48s/it]
 15%|█▌        | 3780/24420 [1:33:57<8:29:08,  1.48s/it]
                                                        
{'loss': 0.2285, 'learning_rate': 5e-05, 'epoch': 0.77}

 15%|█▌        | 3780/24420 [1:33:57<8:29:08,  1.48s/it]
 15%|█▌        | 3781/24420 [1:33:58<8:29:28,  1.48s/it]
 15%|█▌        | 3782/24420 [1:34:00<8:29:34,  1.48s/it]
 15%|█▌        | 3783/24420 [1:34:01<8:30:06,  1.48s/it]
 15%|█▌        | 3784/24420 [1:34:03<8:29:57,  1.48s/it]
 15%|█▌        | 3785/24420 [1:34:04<8:29:11,  1.48s/it]
                                                        
{'loss': 0.1508, 'learning_rate': 5e-05, 'epoch': 0.77}

 15%|█▌        | 3785/24420 [1:34:04<8:29:11,  1.48s/it]
 16%|█▌        | 3786/24420 [1:34:06<8:29:00,  1.48s/it]
 16%|█▌        | 3787/24420 [1:34:07<8:28:17,  1.48s/it]
 16%|█▌        | 3788/24420 [1:34:08<8:27:50,  1.48s/it]
 16%|█▌        | 3789/24420 [1:34:10<8:27:57,  1.48s/it]
 16%|█▌        | 3790/24420 [1:34:11<8:28:45,  1.48s/it]
                                                        
{'loss': 0.1645, 'learning_rate': 5e-05, 'epoch': 0.78}

 16%|█▌        | 3790/24420 [1:34:11<8:28:45,  1.48s/it]
 16%|█▌        | 3791/24420 [1:34:13<8:29:58,  1.48s/it]
 16%|█▌        | 3792/24420 [1:34:14<8:30:14,  1.48s/it]
 16%|█▌        | 3793/24420 [1:34:16<8:29:29,  1.48s/it]
 16%|█▌        | 3794/24420 [1:34:17<8:28:46,  1.48s/it]
 16%|█▌        | 3795/24420 [1:34:19<8:28:14,  1.48s/it]
                                                        
{'loss': 0.1669, 'learning_rate': 5e-05, 'epoch': 0.78}

 16%|█▌        | 3795/24420 [1:34:19<8:28:14,  1.48s/it]
 16%|█▌        | 3796/24420 [1:34:20<8:27:59,  1.48s/it]
 16%|█▌        | 3797/24420 [1:34:22<8:28:35,  1.48s/it]
 16%|█▌        | 3798/24420 [1:34:23<8:28:52,  1.48s/it]
 16%|█▌        | 3799/24420 [1:34:25<8:28:52,  1.48s/it]
 16%|█▌        | 3800/24420 [1:34:26<8:28:10,  1.48s/it]
                                                        
{'loss': 0.1886, 'learning_rate': 5e-05, 'epoch': 0.78}

 16%|█▌        | 3800/24420 [1:34:26<8:28:10,  1.48s/it]
 16%|█▌        | 3801/24420 [1:34:31<13:55:17,  2.43s/it]
 16%|█▌        | 3802/24420 [1:34:32<12:16:50,  2.14s/it]
 16%|█▌        | 3803/24420 [1:34:34<11:08:39,  1.95s/it]
 16%|█▌        | 3804/24420 [1:34:35<10:20:59,  1.81s/it]
 16%|█▌        | 3805/24420 [1:34:37<9:47:25,  1.71s/it] 
                                                        
{'loss': 0.1734, 'learning_rate': 5e-05, 'epoch': 0.78}

 16%|█▌        | 3805/24420 [1:34:37<9:47:25,  1.71s/it]
 16%|█▌        | 3806/24420 [1:34:38<9:24:10,  1.64s/it]
 16%|█▌        | 3807/24420 [1:34:40<9:07:14,  1.59s/it]
 16%|█▌        | 3808/24420 [1:34:41<8:55:08,  1.56s/it]
 16%|█▌        | 3809/24420 [1:34:43<8:48:09,  1.54s/it]
 16%|█▌        | 3810/24420 [1:34:44<8:42:39,  1.52s/it]
                                                        
{'loss': 0.1617, 'learning_rate': 5e-05, 'epoch': 0.78}

 16%|█▌        | 3810/24420 [1:34:44<8:42:39,  1.52s/it]
 16%|█▌        | 3811/24420 [1:34:46<8:38:42,  1.51s/it]
 16%|█▌        | 3812/24420 [1:34:47<8:35:23,  1.50s/it]
 16%|█▌        | 3813/24420 [1:34:49<8:34:01,  1.50s/it]
 16%|█▌        | 3814/24420 [1:34:50<8:31:57,  1.49s/it]
 16%|█▌        | 3815/24420 [1:34:52<8:30:15,  1.49s/it]
                                                        
{'loss': 0.1711, 'learning_rate': 5e-05, 'epoch': 0.78}

 16%|█▌        | 3815/24420 [1:34:52<8:30:15,  1.49s/it]
 16%|█▌        | 3816/24420 [1:34:53<8:30:00,  1.49s/it]
 16%|█▌        | 3817/24420 [1:34:55<8:29:08,  1.48s/it]
 16%|█▌        | 3818/24420 [1:34:56<8:28:13,  1.48s/it]
 16%|█▌        | 3819/24420 [1:34:58<8:27:09,  1.48s/it]
 16%|█▌        | 3820/24420 [1:34:59<8:27:46,  1.48s/it]
                                                        
{'loss': 0.1799, 'learning_rate': 5e-05, 'epoch': 0.78}

 16%|█▌        | 3820/24420 [1:34:59<8:27:46,  1.48s/it]
 16%|█▌        | 3821/24420 [1:35:00<8:28:28,  1.48s/it]
 16%|█▌        | 3822/24420 [1:35:02<8:28:29,  1.48s/it]
 16%|█▌        | 3823/24420 [1:35:03<8:28:47,  1.48s/it]
 16%|█▌        | 3824/24420 [1:35:05<8:28:02,  1.48s/it]
 16%|█▌        | 3825/24420 [1:35:06<8:28:55,  1.48s/it]
                                                        
{'loss': 0.1745, 'learning_rate': 5e-05, 'epoch': 0.78}

 16%|█▌        | 3825/24420 [1:35:06<8:28:55,  1.48s/it]
 16%|█▌        | 3826/24420 [1:35:08<8:28:30,  1.48s/it]
 16%|█▌        | 3827/24420 [1:35:09<8:28:56,  1.48s/it]
 16%|█▌        | 3828/24420 [1:35:11<8:30:22,  1.49s/it]
 16%|█▌        | 3829/24420 [1:35:12<8:30:18,  1.49s/it]
 16%|█▌        | 3830/24420 [1:35:14<8:30:10,  1.49s/it]
                                                        
{'loss': 0.1444, 'learning_rate': 5e-05, 'epoch': 0.78}

 16%|█▌        | 3830/24420 [1:35:14<8:30:10,  1.49s/it]
 16%|█▌        | 3831/24420 [1:35:15<8:29:56,  1.49s/it]
 16%|█▌        | 3832/24420 [1:35:17<8:29:10,  1.48s/it]
 16%|█▌        | 3833/24420 [1:35:18<8:28:50,  1.48s/it]
 16%|█▌        | 3834/24420 [1:35:20<8:27:42,  1.48s/it]
 16%|█▌        | 3835/24420 [1:35:21<8:27:22,  1.48s/it]
                                                        
{'loss': 0.168, 'learning_rate': 5e-05, 'epoch': 0.79}

 16%|█▌        | 3835/24420 [1:35:21<8:27:22,  1.48s/it]
 16%|█▌        | 3836/24420 [1:35:23<8:28:31,  1.48s/it]
 16%|█▌        | 3837/24420 [1:35:24<8:28:18,  1.48s/it]
 16%|█▌        | 3838/24420 [1:35:26<8:28:13,  1.48s/it]
 16%|█▌        | 3839/24420 [1:35:27<8:28:29,  1.48s/it]
 16%|█▌        | 3840/24420 [1:35:29<8:29:26,  1.49s/it]
                                                        
{'loss': 0.2005, 'learning_rate': 5e-05, 'epoch': 0.79}

 16%|█▌        | 3840/24420 [1:35:29<8:29:26,  1.49s/it]
 16%|█▌        | 3841/24420 [1:35:30<8:29:14,  1.48s/it]
 16%|█▌        | 3842/24420 [1:35:32<8:29:29,  1.49s/it]
 16%|█▌        | 3843/24420 [1:35:33<8:29:31,  1.49s/it]
 16%|█▌        | 3844/24420 [1:35:35<8:28:45,  1.48s/it]
 16%|█▌        | 3845/24420 [1:35:36<8:29:04,  1.48s/it]
                                                        
{'loss': 0.1917, 'learning_rate': 5e-05, 'epoch': 0.79}

 16%|█▌        | 3845/24420 [1:35:36<8:29:04,  1.48s/it]
 16%|█▌        | 3846/24420 [1:35:38<8:29:23,  1.49s/it]
 16%|█▌        | 3847/24420 [1:35:39<8:28:50,  1.48s/it]
 16%|█▌        | 3848/24420 [1:35:41<8:27:59,  1.48s/it]
 16%|█▌        | 3849/24420 [1:35:42<8:26:51,  1.48s/it]
 16%|█▌        | 3850/24420 [1:35:43<8:26:50,  1.48s/it]
                                                        
{'loss': 0.1812, 'learning_rate': 5e-05, 'epoch': 0.79}

 16%|█▌        | 3850/24420 [1:35:43<8:26:50,  1.48s/it]
 16%|█▌        | 3851/24420 [1:35:45<8:27:41,  1.48s/it]
 16%|█▌        | 3852/24420 [1:35:46<8:27:19,  1.48s/it]
 16%|█▌        | 3853/24420 [1:35:48<8:27:44,  1.48s/it]
 16%|█▌        | 3854/24420 [1:35:49<8:27:32,  1.48s/it]
 16%|█▌        | 3855/24420 [1:35:51<8:27:41,  1.48s/it]
                                                        
{'loss': 0.2068, 'learning_rate': 5e-05, 'epoch': 0.79}

 16%|█▌        | 3855/24420 [1:35:51<8:27:41,  1.48s/it]
 16%|█▌        | 3856/24420 [1:35:52<8:27:56,  1.48s/it]
 16%|█▌        | 3857/24420 [1:35:54<8:27:26,  1.48s/it]
 16%|█▌        | 3858/24420 [1:35:55<8:26:24,  1.48s/it]
 16%|█▌        | 3859/24420 [1:35:57<8:25:59,  1.48s/it]
 16%|█▌        | 3860/24420 [1:35:58<8:26:10,  1.48s/it]
                                                        
{'loss': 0.1633, 'learning_rate': 5e-05, 'epoch': 0.79}

 16%|█▌        | 3860/24420 [1:35:58<8:26:10,  1.48s/it]
 16%|█▌        | 3861/24420 [1:36:00<8:27:30,  1.48s/it]
 16%|█▌        | 3862/24420 [1:36:01<8:26:52,  1.48s/it]
 16%|█▌        | 3863/24420 [1:36:03<8:26:56,  1.48s/it]
 16%|█▌        | 3864/24420 [1:36:04<8:27:06,  1.48s/it]
 16%|█▌        | 3865/24420 [1:36:06<8:27:56,  1.48s/it]
                                                        
{'loss': 0.1693, 'learning_rate': 5e-05, 'epoch': 0.79}

 16%|█▌        | 3865/24420 [1:36:06<8:27:56,  1.48s/it]
 16%|█▌        | 3866/24420 [1:36:07<8:29:22,  1.49s/it]
 16%|█▌        | 3867/24420 [1:36:09<8:28:55,  1.49s/it]
 16%|█▌        | 3868/24420 [1:36:10<8:28:05,  1.48s/it]
 16%|█▌        | 3869/24420 [1:36:12<8:27:08,  1.48s/it]
 16%|█▌        | 3870/24420 [1:36:13<8:26:29,  1.48s/it]
                                                        
{'loss': 0.1522, 'learning_rate': 5e-05, 'epoch': 0.79}

 16%|█▌        | 3870/24420 [1:36:13<8:26:29,  1.48s/it]
 16%|█▌        | 3871/24420 [1:36:15<8:26:51,  1.48s/it]
 16%|█▌        | 3872/24420 [1:36:16<8:26:31,  1.48s/it]
 16%|█▌        | 3873/24420 [1:36:18<8:27:45,  1.48s/it]
 16%|█▌        | 3874/24420 [1:36:19<8:27:55,  1.48s/it]
 16%|█▌        | 3875/24420 [1:36:21<8:28:33,  1.49s/it]
                                                        
{'loss': 0.1652, 'learning_rate': 5e-05, 'epoch': 0.79}

 16%|█▌        | 3875/24420 [1:36:21<8:28:33,  1.49s/it]
 16%|█▌        | 3876/24420 [1:36:22<8:28:11,  1.48s/it]
 16%|█▌        | 3877/24420 [1:36:23<8:27:06,  1.48s/it]
 16%|█▌        | 3878/24420 [1:36:25<8:26:51,  1.48s/it]
 16%|█▌        | 3879/24420 [1:36:26<8:26:43,  1.48s/it]
 16%|█▌        | 3880/24420 [1:36:28<8:26:44,  1.48s/it]
                                                        
{'loss': 0.129, 'learning_rate': 5e-05, 'epoch': 0.79}

 16%|█▌        | 3880/24420 [1:36:28<8:26:44,  1.48s/it]
 16%|█▌        | 3881/24420 [1:36:29<8:27:35,  1.48s/it]
 16%|█▌        | 3882/24420 [1:36:31<8:27:28,  1.48s/it]
 16%|█▌        | 3883/24420 [1:36:32<8:27:30,  1.48s/it]
 16%|█▌        | 3884/24420 [1:36:34<8:27:12,  1.48s/it]
 16%|█▌        | 3885/24420 [1:36:35<8:26:38,  1.48s/it]
                                                        
{'loss': 0.1577, 'learning_rate': 5e-05, 'epoch': 0.8}

 16%|█▌        | 3885/24420 [1:36:35<8:26:38,  1.48s/it]
 16%|█▌        | 3886/24420 [1:36:37<8:26:41,  1.48s/it]
 16%|█▌        | 3887/24420 [1:36:38<8:27:04,  1.48s/it]
 16%|█▌        | 3888/24420 [1:36:40<8:26:19,  1.48s/it]
 16%|█▌        | 3889/24420 [1:36:41<8:27:18,  1.48s/it]
 16%|█▌        | 3890/24420 [1:36:43<8:26:38,  1.48s/it]
                                                        
{'loss': 0.1713, 'learning_rate': 5e-05, 'epoch': 0.8}

 16%|█▌        | 3890/24420 [1:36:43<8:26:38,  1.48s/it]
 16%|█▌        | 3891/24420 [1:36:44<8:26:27,  1.48s/it]
 16%|█▌        | 3892/24420 [1:36:46<8:27:31,  1.48s/it]
 16%|█▌        | 3893/24420 [1:36:47<8:28:12,  1.49s/it]
 16%|█▌        | 3894/24420 [1:36:49<8:27:21,  1.48s/it]
 16%|█▌        | 3895/24420 [1:36:50<8:26:14,  1.48s/it]
                                                        
{'loss': 0.168, 'learning_rate': 5e-05, 'epoch': 0.8}

 16%|█▌        | 3895/24420 [1:36:50<8:26:14,  1.48s/it]
 16%|█▌        | 3896/24420 [1:36:52<8:27:41,  1.48s/it]
 16%|█▌        | 3897/24420 [1:36:53<8:26:54,  1.48s/it]
 16%|█▌        | 3898/24420 [1:36:55<8:26:28,  1.48s/it]
 16%|█▌        | 3899/24420 [1:36:56<8:25:38,  1.48s/it]
 16%|█▌        | 3900/24420 [1:36:58<8:25:46,  1.48s/it]
                                                        
{'loss': 0.1623, 'learning_rate': 5e-05, 'epoch': 0.8}

 16%|█▌        | 3900/24420 [1:36:58<8:25:46,  1.48s/it]
 16%|█▌        | 3901/24420 [1:36:59<8:27:17,  1.48s/it]
 16%|█▌        | 3902/24420 [1:37:01<8:26:03,  1.48s/it]
 16%|█▌        | 3903/24420 [1:37:02<8:26:21,  1.48s/it]
 16%|█▌        | 3904/24420 [1:37:03<8:26:38,  1.48s/it]
 16%|█▌        | 3905/24420 [1:37:05<8:26:26,  1.48s/it]
                                                        
{'loss': 0.1579, 'learning_rate': 5e-05, 'epoch': 0.8}

 16%|█▌        | 3905/24420 [1:37:05<8:26:26,  1.48s/it]
 16%|█▌        | 3906/24420 [1:37:06<8:26:47,  1.48s/it]
 16%|█▌        | 3907/24420 [1:37:08<8:26:27,  1.48s/it]
 16%|█▌        | 3908/24420 [1:37:09<8:26:04,  1.48s/it]
 16%|█▌        | 3909/24420 [1:37:11<8:26:09,  1.48s/it]
 16%|█▌        | 3910/24420 [1:37:12<8:26:03,  1.48s/it]
                                                        
{'loss': 0.1616, 'learning_rate': 5e-05, 'epoch': 0.8}

 16%|█▌        | 3910/24420 [1:37:12<8:26:03,  1.48s/it]
 16%|█▌        | 3911/24420 [1:37:14<8:26:07,  1.48s/it]
 16%|█▌        | 3912/24420 [1:37:15<8:26:38,  1.48s/it]
 16%|█▌        | 3913/24420 [1:37:17<8:26:02,  1.48s/it]
 16%|█▌        | 3914/24420 [1:37:18<8:25:11,  1.48s/it]
 16%|█▌        | 3915/24420 [1:37:20<8:25:31,  1.48s/it]
                                                        
{'loss': 0.1962, 'learning_rate': 5e-05, 'epoch': 0.8}

 16%|█▌        | 3915/24420 [1:37:20<8:25:31,  1.48s/it]
 16%|█▌        | 3916/24420 [1:37:21<8:25:13,  1.48s/it]
 16%|█▌        | 3917/24420 [1:37:23<8:24:45,  1.48s/it]
 16%|█▌        | 3918/24420 [1:37:24<8:24:32,  1.48s/it]
 16%|█▌        | 3919/24420 [1:37:26<8:25:04,  1.48s/it]
 16%|█▌        | 3920/24420 [1:37:27<8:25:02,  1.48s/it]
                                                        
{'loss': 0.1702, 'learning_rate': 5e-05, 'epoch': 0.8}

 16%|█▌        | 3920/24420 [1:37:27<8:25:02,  1.48s/it]
 16%|█▌        | 3921/24420 [1:37:29<8:26:20,  1.48s/it]
 16%|█▌        | 3922/24420 [1:37:30<8:26:40,  1.48s/it]
 16%|█▌        | 3923/24420 [1:37:32<8:26:31,  1.48s/it]
 16%|█▌        | 3924/24420 [1:37:33<8:26:10,  1.48s/it]
 16%|█▌        | 3925/24420 [1:37:35<8:25:35,  1.48s/it]
                                                        
{'loss': 0.1787, 'learning_rate': 5e-05, 'epoch': 0.8}

 16%|█▌        | 3925/24420 [1:37:35<8:25:35,  1.48s/it]
 16%|█▌        | 3926/24420 [1:37:36<8:25:28,  1.48s/it]
 16%|█▌        | 3927/24420 [1:37:38<8:25:45,  1.48s/it]
 16%|█▌        | 3928/24420 [1:37:39<8:25:29,  1.48s/it]
 16%|█▌        | 3929/24420 [1:37:40<8:25:05,  1.48s/it]
 16%|█▌        | 3930/24420 [1:37:42<8:24:59,  1.48s/it]
                                                        
{'loss': 0.1819, 'learning_rate': 5e-05, 'epoch': 0.8}

 16%|█▌        | 3930/24420 [1:37:42<8:24:59,  1.48s/it]
 16%|█▌        | 3931/24420 [1:37:43<8:26:03,  1.48s/it]
 16%|█▌        | 3932/24420 [1:37:45<8:26:08,  1.48s/it]
 16%|█▌        | 3933/24420 [1:37:46<8:25:14,  1.48s/it]
 16%|█▌        | 3934/24420 [1:37:48<8:24:39,  1.48s/it]
 16%|█▌        | 3935/24420 [1:37:49<8:24:21,  1.48s/it]
                                                        
{'loss': 0.1675, 'learning_rate': 5e-05, 'epoch': 0.81}

 16%|█▌        | 3935/24420 [1:37:49<8:24:21,  1.48s/it]
 16%|█▌        | 3936/24420 [1:37:51<8:24:08,  1.48s/it]
 16%|█▌        | 3937/24420 [1:37:52<8:23:44,  1.48s/it]
 16%|█▌        | 3938/24420 [1:37:54<8:23:38,  1.48s/it]
 16%|█▌        | 3939/24420 [1:37:55<8:23:43,  1.48s/it]
 16%|█▌        | 3940/24420 [1:37:57<8:24:36,  1.48s/it]
                                                        
{'loss': 0.1689, 'learning_rate': 5e-05, 'epoch': 0.81}

 16%|█▌        | 3940/24420 [1:37:57<8:24:36,  1.48s/it]
 16%|█▌        | 3941/24420 [1:37:58<8:25:38,  1.48s/it]
 16%|█▌        | 3942/24420 [1:38:00<8:26:08,  1.48s/it]
 16%|█▌        | 3943/24420 [1:38:01<8:26:03,  1.48s/it]
 16%|█▌        | 3944/24420 [1:38:03<8:25:12,  1.48s/it]
 16%|█▌        | 3945/24420 [1:38:04<8:25:26,  1.48s/it]
                                                        
{'loss': 0.1565, 'learning_rate': 5e-05, 'epoch': 0.81}

 16%|█▌        | 3945/24420 [1:38:04<8:25:26,  1.48s/it]
 16%|█▌        | 3946/24420 [1:38:06<8:25:19,  1.48s/it]
 16%|█▌        | 3947/24420 [1:38:07<8:25:40,  1.48s/it]
 16%|█▌        | 3948/24420 [1:38:09<8:27:10,  1.49s/it]
 16%|█▌        | 3949/24420 [1:38:10<8:26:00,  1.48s/it]
 16%|█▌        | 3950/24420 [1:38:12<8:25:13,  1.48s/it]
                                                        
{'loss': 0.1628, 'learning_rate': 5e-05, 'epoch': 0.81}

 16%|█▌        | 3950/24420 [1:38:12<8:25:13,  1.48s/it]
 16%|█▌        | 3951/24420 [1:38:13<8:24:51,  1.48s/it]
 16%|█▌        | 3952/24420 [1:38:15<8:24:31,  1.48s/it]
 16%|█▌        | 3953/24420 [1:38:16<8:24:18,  1.48s/it]
 16%|█▌        | 3954/24420 [1:38:17<8:25:34,  1.48s/it]
 16%|█▌        | 3955/24420 [1:38:19<8:27:14,  1.49s/it]
                                                        
{'loss': 0.1698, 'learning_rate': 5e-05, 'epoch': 0.81}

 16%|█▌        | 3955/24420 [1:38:19<8:27:14,  1.49s/it]
 16%|█▌        | 3956/24420 [1:38:20<8:26:33,  1.49s/it]
 16%|█▌        | 3957/24420 [1:38:22<8:25:14,  1.48s/it]
 16%|█▌        | 3958/24420 [1:38:23<8:25:26,  1.48s/it]
 16%|█▌        | 3959/24420 [1:38:25<8:25:07,  1.48s/it]
 16%|█▌        | 3960/24420 [1:38:26<8:26:01,  1.48s/it]
                                                        
{'loss': 0.1674, 'learning_rate': 5e-05, 'epoch': 0.81}

 16%|█▌        | 3960/24420 [1:38:26<8:26:01,  1.48s/it]
 16%|█▌        | 3961/24420 [1:38:28<8:24:56,  1.48s/it]
 16%|█▌        | 3962/24420 [1:38:29<8:24:00,  1.48s/it]
 16%|█▌        | 3963/24420 [1:38:31<8:24:49,  1.48s/it]
 16%|█▌        | 3964/24420 [1:38:32<8:24:31,  1.48s/it]
 16%|█▌        | 3965/24420 [1:38:34<8:23:59,  1.48s/it]
                                                        
{'loss': 0.2163, 'learning_rate': 5e-05, 'epoch': 0.81}

 16%|█▌        | 3965/24420 [1:38:34<8:23:59,  1.48s/it]
 16%|█▌        | 3966/24420 [1:38:35<8:23:53,  1.48s/it]
 16%|█▌        | 3967/24420 [1:38:37<8:23:23,  1.48s/it]
 16%|█▌        | 3968/24420 [1:38:38<8:23:58,  1.48s/it]
 16%|█▋        | 3969/24420 [1:38:40<8:24:07,  1.48s/it]
 16%|█▋        | 3970/24420 [1:38:41<8:24:42,  1.48s/it]
                                                        
{'loss': 0.1615, 'learning_rate': 5e-05, 'epoch': 0.81}

 16%|█▋        | 3970/24420 [1:38:41<8:24:42,  1.48s/it]
 16%|█▋        | 3971/24420 [1:38:43<8:24:09,  1.48s/it]
 16%|█▋        | 3972/24420 [1:38:44<8:23:49,  1.48s/it]
 16%|█▋        | 3973/24420 [1:38:46<8:24:13,  1.48s/it]
 16%|█▋        | 3974/24420 [1:38:47<8:23:48,  1.48s/it]
 16%|█▋        | 3975/24420 [1:38:49<8:24:06,  1.48s/it]
                                                        
{'loss': 0.1699, 'learning_rate': 5e-05, 'epoch': 0.81}

 16%|█▋        | 3975/24420 [1:38:49<8:24:06,  1.48s/it]
 16%|█▋        | 3976/24420 [1:38:50<8:23:20,  1.48s/it]
 16%|█▋        | 3977/24420 [1:38:52<8:23:34,  1.48s/it]
 16%|█▋        | 3978/24420 [1:38:53<8:22:54,  1.48s/it]
 16%|█▋        | 3979/24420 [1:38:54<8:22:39,  1.48s/it]
 16%|█▋        | 3980/24420 [1:38:56<8:22:16,  1.47s/it]
                                                        
{'loss': 0.168, 'learning_rate': 5e-05, 'epoch': 0.81}

 16%|█▋        | 3980/24420 [1:38:56<8:22:16,  1.47s/it]
 16%|█▋        | 3981/24420 [1:38:57<8:22:40,  1.48s/it]
 16%|█▋        | 3982/24420 [1:38:59<8:22:35,  1.48s/it]
 16%|█▋        | 3983/24420 [1:39:00<8:22:21,  1.47s/it]
 16%|█▋        | 3984/24420 [1:39:02<8:22:09,  1.47s/it]
 16%|█▋        | 3985/24420 [1:39:03<8:22:15,  1.47s/it]
                                                        
{'loss': 0.1607, 'learning_rate': 5e-05, 'epoch': 0.82}

 16%|█▋        | 3985/24420 [1:39:03<8:22:15,  1.47s/it]
 16%|█▋        | 3986/24420 [1:39:05<8:22:13,  1.47s/it]
 16%|█▋        | 3987/24420 [1:39:06<8:23:30,  1.48s/it]
 16%|█▋        | 3988/24420 [1:39:08<8:22:45,  1.48s/it]
 16%|█▋        | 3989/24420 [1:39:09<8:23:13,  1.48s/it]
 16%|█▋        | 3990/24420 [1:39:11<8:24:06,  1.48s/it]
                                                        
{'loss': 0.1675, 'learning_rate': 5e-05, 'epoch': 0.82}

 16%|█▋        | 3990/24420 [1:39:11<8:24:06,  1.48s/it]
 16%|█▋        | 3991/24420 [1:39:12<8:24:08,  1.48s/it]
 16%|█▋        | 3992/24420 [1:39:14<8:23:51,  1.48s/it]
 16%|█▋        | 3993/24420 [1:39:15<8:22:28,  1.48s/it]
 16%|█▋        | 3994/24420 [1:39:17<8:22:00,  1.47s/it]
 16%|█▋        | 3995/24420 [1:39:18<8:21:38,  1.47s/it]
                                                        
{'loss': 0.16, 'learning_rate': 5e-05, 'epoch': 0.82}

 16%|█▋        | 3995/24420 [1:39:18<8:21:38,  1.47s/it]
 16%|█▋        | 3996/24420 [1:39:20<8:22:04,  1.47s/it]
 16%|█▋        | 3997/24420 [1:39:21<8:24:07,  1.48s/it]
 16%|█▋        | 3998/24420 [1:39:23<8:24:22,  1.48s/it]
 16%|█▋        | 3999/24420 [1:39:24<8:23:44,  1.48s/it]
 16%|█▋        | 4000/24420 [1:39:26<8:24:26,  1.48s/it]
                                                        
{'loss': 0.1672, 'learning_rate': 5e-05, 'epoch': 0.82}

 16%|█▋        | 4000/24420 [1:39:26<8:24:26,  1.48s/it]
 16%|█▋        | 4001/24420 [1:39:30<13:34:55,  2.39s/it]
 16%|█▋        | 4002/24420 [1:39:32<12:00:43,  2.12s/it]
 16%|█▋        | 4003/24420 [1:39:33<10:55:23,  1.93s/it]
 16%|█▋        | 4004/24420 [1:39:34<10:09:30,  1.79s/it]
 16%|█▋        | 4005/24420 [1:39:36<9:37:32,  1.70s/it] 
                                                        
{'loss': 0.1531, 'learning_rate': 5e-05, 'epoch': 0.82}

 16%|█▋        | 4005/24420 [1:39:36<9:37:32,  1.70s/it]
 16%|█▋        | 4006/24420 [1:39:37<9:15:53,  1.63s/it]
 16%|█▋        | 4007/24420 [1:39:39<9:00:51,  1.59s/it]
 16%|█▋        | 4008/24420 [1:39:40<8:50:20,  1.56s/it]
 16%|█▋        | 4009/24420 [1:39:42<8:41:50,  1.53s/it]
 16%|█▋        | 4010/24420 [1:39:43<8:36:31,  1.52s/it]
                                                        
{'loss': 0.1895, 'learning_rate': 5e-05, 'epoch': 0.82}

 16%|█▋        | 4010/24420 [1:39:43<8:36:31,  1.52s/it]
 16%|█▋        | 4011/24420 [1:39:45<8:33:00,  1.51s/it]
 16%|█▋        | 4012/24420 [1:39:46<8:30:46,  1.50s/it]
 16%|█▋        | 4013/24420 [1:39:48<8:28:27,  1.49s/it]
 16%|█▋        | 4014/24420 [1:39:49<8:26:40,  1.49s/it]
 16%|█▋        | 4015/24420 [1:39:51<8:25:15,  1.49s/it]
                                                        
{'loss': 0.1576, 'learning_rate': 5e-05, 'epoch': 0.82}

 16%|█▋        | 4015/24420 [1:39:51<8:25:15,  1.49s/it]
 16%|█▋        | 4016/24420 [1:39:52<8:24:22,  1.48s/it]
 16%|█▋        | 4017/24420 [1:39:54<8:24:22,  1.48s/it]
 16%|█▋        | 4018/24420 [1:39:55<8:24:17,  1.48s/it]
 16%|█▋        | 4019/24420 [1:39:57<8:24:35,  1.48s/it]
 16%|█▋        | 4020/24420 [1:39:58<8:23:56,  1.48s/it]
                                                        
{'loss': 0.1747, 'learning_rate': 5e-05, 'epoch': 0.82}

 16%|█▋        | 4020/24420 [1:39:58<8:23:56,  1.48s/it]
 16%|█▋        | 4021/24420 [1:40:00<8:24:25,  1.48s/it]
 16%|█▋        | 4022/24420 [1:40:01<8:24:31,  1.48s/it]
 16%|█▋        | 4023/24420 [1:40:03<8:23:45,  1.48s/it]
 16%|█▋        | 4024/24420 [1:40:04<8:23:10,  1.48s/it]
 16%|█▋        | 4025/24420 [1:40:06<8:25:09,  1.49s/it]
                                                        
{'loss': 0.1727, 'learning_rate': 5e-05, 'epoch': 0.82}

 16%|█▋        | 4025/24420 [1:40:06<8:25:09,  1.49s/it]
 16%|█▋        | 4026/24420 [1:40:07<8:24:58,  1.49s/it]
 16%|█▋        | 4027/24420 [1:40:09<8:24:10,  1.48s/it]
 16%|█▋        | 4028/24420 [1:40:10<8:23:08,  1.48s/it]
 16%|█▋        | 4029/24420 [1:40:12<8:22:01,  1.48s/it]
 17%|█▋        | 4030/24420 [1:40:13<8:23:14,  1.48s/it]
                                                        
{'loss': 0.2049, 'learning_rate': 5e-05, 'epoch': 0.83}

 17%|█▋        | 4030/24420 [1:40:13<8:23:14,  1.48s/it]
 17%|█▋        | 4031/24420 [1:40:14<8:22:51,  1.48s/it]
 17%|█▋        | 4032/24420 [1:40:16<8:23:07,  1.48s/it]
 17%|█▋        | 4033/24420 [1:40:17<8:23:32,  1.48s/it]
 17%|█▋        | 4034/24420 [1:40:19<8:23:23,  1.48s/it]
 17%|█▋        | 4035/24420 [1:40:20<8:23:44,  1.48s/it]
                                                        
{'loss': 0.1779, 'learning_rate': 5e-05, 'epoch': 0.83}

 17%|█▋        | 4035/24420 [1:40:20<8:23:44,  1.48s/it]
 17%|█▋        | 4036/24420 [1:40:22<8:24:17,  1.48s/it]
 17%|█▋        | 4037/24420 [1:40:23<8:23:03,  1.48s/it]
 17%|█▋        | 4038/24420 [1:40:25<8:22:27,  1.48s/it]
 17%|█▋        | 4039/24420 [1:40:26<8:22:04,  1.48s/it]
 17%|█▋        | 4040/24420 [1:40:28<8:22:17,  1.48s/it]
                                                        
{'loss': 0.1878, 'learning_rate': 5e-05, 'epoch': 0.83}

 17%|█▋        | 4040/24420 [1:40:28<8:22:17,  1.48s/it]
 17%|█▋        | 4041/24420 [1:40:29<8:22:03,  1.48s/it]
 17%|█▋        | 4042/24420 [1:40:31<8:21:29,  1.48s/it]
 17%|█▋        | 4043/24420 [1:40:32<8:21:07,  1.48s/it]
 17%|█▋        | 4044/24420 [1:40:34<8:22:08,  1.48s/it]
 17%|█▋        | 4045/24420 [1:40:35<8:22:28,  1.48s/it]
                                                        
{'loss': 0.1658, 'learning_rate': 5e-05, 'epoch': 0.83}

 17%|█▋        | 4045/24420 [1:40:35<8:22:28,  1.48s/it]
 17%|█▋        | 4046/24420 [1:40:37<8:22:28,  1.48s/it]
 17%|█▋        | 4047/24420 [1:40:38<8:22:50,  1.48s/it]
 17%|█▋        | 4048/24420 [1:40:40<8:22:16,  1.48s/it]
 17%|█▋        | 4049/24420 [1:40:41<8:23:07,  1.48s/it]
 17%|█▋        | 4050/24420 [1:40:43<8:22:15,  1.48s/it]
                                                        
{'loss': 0.1975, 'learning_rate': 5e-05, 'epoch': 0.83}

 17%|█▋        | 4050/24420 [1:40:43<8:22:15,  1.48s/it]
 17%|█▋        | 4051/24420 [1:40:44<8:22:53,  1.48s/it]
 17%|█▋        | 4052/24420 [1:40:46<8:22:33,  1.48s/it]
 17%|█▋        | 4053/24420 [1:40:47<8:21:50,  1.48s/it]
 17%|█▋        | 4054/24420 [1:40:49<8:23:17,  1.48s/it]
 17%|█▋        | 4055/24420 [1:40:50<8:23:39,  1.48s/it]
                                                        
{'loss': 0.1697, 'learning_rate': 5e-05, 'epoch': 0.83}

 17%|█▋        | 4055/24420 [1:40:50<8:23:39,  1.48s/it]
 17%|█▋        | 4056/24420 [1:40:51<8:22:55,  1.48s/it]
 17%|█▋        | 4057/24420 [1:40:53<8:21:56,  1.48s/it]
 17%|█▋        | 4058/24420 [1:40:54<8:22:37,  1.48s/it]
 17%|█▋        | 4059/24420 [1:40:56<8:22:07,  1.48s/it]
 17%|█▋        | 4060/24420 [1:40:57<8:21:20,  1.48s/it]
                                                        
{'loss': 0.1683, 'learning_rate': 5e-05, 'epoch': 0.83}

 17%|█▋        | 4060/24420 [1:40:57<8:21:20,  1.48s/it]
 17%|█▋        | 4061/24420 [1:40:59<8:21:30,  1.48s/it]
 17%|█▋        | 4062/24420 [1:41:00<8:22:17,  1.48s/it]
 17%|█▋        | 4063/24420 [1:41:02<8:22:37,  1.48s/it]
 17%|█▋        | 4064/24420 [1:41:03<8:22:41,  1.48s/it]
 17%|█▋        | 4065/24420 [1:41:05<8:22:15,  1.48s/it]
                                                        
{'loss': 0.1821, 'learning_rate': 5e-05, 'epoch': 0.83}

 17%|█▋        | 4065/24420 [1:41:05<8:22:15,  1.48s/it]
 17%|█▋        | 4066/24420 [1:41:06<8:22:54,  1.48s/it]
 17%|█▋        | 4067/24420 [1:41:08<8:23:09,  1.48s/it]
 17%|█▋        | 4068/24420 [1:41:09<8:22:28,  1.48s/it]
 17%|█▋        | 4069/24420 [1:41:11<8:21:33,  1.48s/it]
 17%|█▋        | 4070/24420 [1:41:12<8:21:35,  1.48s/it]
                                                        
{'loss': 0.1789, 'learning_rate': 5e-05, 'epoch': 0.83}

 17%|█▋        | 4070/24420 [1:41:12<8:21:35,  1.48s/it]
 17%|█▋        | 4071/24420 [1:41:14<8:22:42,  1.48s/it]
 17%|█▋        | 4072/24420 [1:41:15<8:22:30,  1.48s/it]
 17%|█▋        | 4073/24420 [1:41:17<8:21:16,  1.48s/it]
 17%|█▋        | 4074/24420 [1:41:18<8:20:50,  1.48s/it]
 17%|█▋        | 4075/24420 [1:41:20<8:21:13,  1.48s/it]
                                                        
{'loss': 0.1703, 'learning_rate': 5e-05, 'epoch': 0.83}

 17%|█▋        | 4075/24420 [1:41:20<8:21:13,  1.48s/it]
 17%|█▋        | 4076/24420 [1:41:21<8:22:11,  1.48s/it]
 17%|█▋        | 4077/24420 [1:41:23<8:21:18,  1.48s/it]
 17%|█▋        | 4078/24420 [1:41:24<8:21:05,  1.48s/it]
 17%|█▋        | 4079/24420 [1:41:26<8:20:46,  1.48s/it]
 17%|█▋        | 4080/24420 [1:41:27<8:21:12,  1.48s/it]
                                                        
{'loss': 0.1786, 'learning_rate': 5e-05, 'epoch': 0.84}

 17%|█▋        | 4080/24420 [1:41:27<8:21:12,  1.48s/it]
 17%|█▋        | 4081/24420 [1:41:28<8:21:01,  1.48s/it]
 17%|█▋        | 4082/24420 [1:41:30<8:20:48,  1.48s/it]
 17%|█▋        | 4083/24420 [1:41:31<8:20:32,  1.48s/it]
 17%|█▋        | 4084/24420 [1:41:33<8:20:45,  1.48s/it]
 17%|█▋        | 4085/24420 [1:41:34<8:20:31,  1.48s/it]
                                                        
{'loss': 0.1354, 'learning_rate': 5e-05, 'epoch': 0.84}

 17%|█▋        | 4085/24420 [1:41:34<8:20:31,  1.48s/it]
 17%|█▋        | 4086/24420 [1:41:36<8:20:55,  1.48s/it]
 17%|█▋        | 4087/24420 [1:41:37<8:20:33,  1.48s/it]
 17%|█▋        | 4088/24420 [1:41:39<8:20:34,  1.48s/it]
 17%|█▋        | 4089/24420 [1:41:40<8:19:38,  1.47s/it]
 17%|█▋        | 4090/24420 [1:41:42<8:20:30,  1.48s/it]
                                                        
{'loss': 0.1681, 'learning_rate': 5e-05, 'epoch': 0.84}

 17%|█▋        | 4090/24420 [1:41:42<8:20:30,  1.48s/it]
 17%|█▋        | 4091/24420 [1:41:43<8:20:45,  1.48s/it]
 17%|█▋        | 4092/24420 [1:41:45<8:21:50,  1.48s/it]
 17%|█▋        | 4093/24420 [1:41:46<8:21:10,  1.48s/it]
 17%|█▋        | 4094/24420 [1:41:48<8:20:46,  1.48s/it]
 17%|█▋        | 4095/24420 [1:41:49<8:21:40,  1.48s/it]
                                                        
{'loss': 0.1679, 'learning_rate': 5e-05, 'epoch': 0.84}

 17%|█▋        | 4095/24420 [1:41:49<8:21:40,  1.48s/it]
 17%|█▋        | 4096/24420 [1:41:51<8:21:38,  1.48s/it]
 17%|█▋        | 4097/24420 [1:41:52<8:20:46,  1.48s/it]
 17%|█▋        | 4098/24420 [1:41:54<8:21:10,  1.48s/it]
 17%|█▋        | 4099/24420 [1:41:55<8:21:13,  1.48s/it]
 17%|█▋        | 4100/24420 [1:41:57<8:20:34,  1.48s/it]
                                                        
{'loss': 0.164, 'learning_rate': 5e-05, 'epoch': 0.84}

 17%|█▋        | 4100/24420 [1:41:57<8:20:34,  1.48s/it]
 17%|█▋        | 4101/24420 [1:41:58<8:20:20,  1.48s/it]
 17%|█▋        | 4102/24420 [1:42:00<8:20:03,  1.48s/it]
 17%|█▋        | 4103/24420 [1:42:01<8:20:07,  1.48s/it]
 17%|█▋        | 4104/24420 [1:42:02<8:20:09,  1.48s/it]
 17%|█▋        | 4105/24420 [1:42:04<8:20:12,  1.48s/it]
                                                        
{'loss': 0.1603, 'learning_rate': 5e-05, 'epoch': 0.84}

 17%|█▋        | 4105/24420 [1:42:04<8:20:12,  1.48s/it]
 17%|█▋        | 4106/24420 [1:42:05<8:20:39,  1.48s/it]
 17%|█▋        | 4107/24420 [1:42:07<8:21:02,  1.48s/it]
 17%|█▋        | 4108/24420 [1:42:08<8:20:03,  1.48s/it]
 17%|█▋        | 4109/24420 [1:42:10<8:20:38,  1.48s/it]
 17%|█▋        | 4110/24420 [1:42:11<8:21:13,  1.48s/it]
                                                        
{'loss': 0.17, 'learning_rate': 5e-05, 'epoch': 0.84}

 17%|█▋        | 4110/24420 [1:42:11<8:21:13,  1.48s/it]
 17%|█▋        | 4111/24420 [1:42:13<8:21:10,  1.48s/it]
 17%|█▋        | 4112/24420 [1:42:14<8:20:44,  1.48s/it]
 17%|█▋        | 4113/24420 [1:42:16<8:21:34,  1.48s/it]
 17%|█▋        | 4114/24420 [1:42:17<8:21:38,  1.48s/it]
 17%|█▋        | 4115/24420 [1:42:19<8:20:58,  1.48s/it]
                                                        
{'loss': 0.1506, 'learning_rate': 5e-05, 'epoch': 0.84}

 17%|█▋        | 4115/24420 [1:42:19<8:20:58,  1.48s/it]
 17%|█▋        | 4116/24420 [1:42:20<8:21:30,  1.48s/it]
 17%|█▋        | 4117/24420 [1:42:22<8:21:20,  1.48s/it]
 17%|█▋        | 4118/24420 [1:42:23<8:20:39,  1.48s/it]
 17%|█▋        | 4119/24420 [1:42:25<8:20:25,  1.48s/it]
 17%|█▋        | 4120/24420 [1:42:26<8:20:31,  1.48s/it]
                                                        
{'loss': 0.1629, 'learning_rate': 5e-05, 'epoch': 0.84}

 17%|█▋        | 4120/24420 [1:42:26<8:20:31,  1.48s/it]
 17%|█▋        | 4121/24420 [1:42:28<8:20:20,  1.48s/it]
 17%|█▋        | 4122/24420 [1:42:29<8:19:42,  1.48s/it]
 17%|█▋        | 4123/24420 [1:42:31<8:19:51,  1.48s/it]
 17%|█▋        | 4124/24420 [1:42:32<8:20:16,  1.48s/it]
 17%|█▋        | 4125/24420 [1:42:34<8:20:31,  1.48s/it]
                                                        
{'loss': 0.1695, 'learning_rate': 5e-05, 'epoch': 0.84}

 17%|█▋        | 4125/24420 [1:42:34<8:20:31,  1.48s/it]
 17%|█▋        | 4126/24420 [1:42:35<8:20:06,  1.48s/it]
 17%|█▋        | 4127/24420 [1:42:36<8:20:12,  1.48s/it]
 17%|█▋        | 4128/24420 [1:42:38<8:20:10,  1.48s/it]
 17%|█▋        | 4129/24420 [1:42:39<8:20:24,  1.48s/it]
 17%|█▋        | 4130/24420 [1:42:41<8:20:56,  1.48s/it]
                                                        
{'loss': 0.1344, 'learning_rate': 5e-05, 'epoch': 0.85}

 17%|█▋        | 4130/24420 [1:42:41<8:20:56,  1.48s/it]
 17%|█▋        | 4131/24420 [1:42:42<8:20:04,  1.48s/it]
 17%|█▋        | 4132/24420 [1:42:44<8:19:36,  1.48s/it]
 17%|█▋        | 4133/24420 [1:42:45<8:19:30,  1.48s/it]
 17%|█▋        | 4134/24420 [1:42:47<8:19:44,  1.48s/it]
 17%|█▋        | 4135/24420 [1:42:48<8:20:11,  1.48s/it]
                                                        
{'loss': 0.1462, 'learning_rate': 5e-05, 'epoch': 0.85}

 17%|█▋        | 4135/24420 [1:42:48<8:20:11,  1.48s/it]
 17%|█▋        | 4136/24420 [1:42:50<8:19:48,  1.48s/it]
 17%|█▋        | 4137/24420 [1:42:51<8:21:27,  1.48s/it]
 17%|█▋        | 4138/24420 [1:42:53<8:20:52,  1.48s/it]
 17%|█▋        | 4139/24420 [1:42:54<8:21:35,  1.48s/it]
 17%|█▋        | 4140/24420 [1:42:56<8:21:52,  1.48s/it]
                                                        
{'loss': 0.1561, 'learning_rate': 5e-05, 'epoch': 0.85}

 17%|█▋        | 4140/24420 [1:42:56<8:21:52,  1.48s/it]
 17%|█▋        | 4141/24420 [1:42:57<8:21:24,  1.48s/it]
 17%|█▋        | 4142/24420 [1:42:59<8:21:03,  1.48s/it]
 17%|█▋        | 4143/24420 [1:43:00<8:20:32,  1.48s/it]
 17%|█▋        | 4144/24420 [1:43:02<8:20:19,  1.48s/it]
 17%|█▋        | 4145/24420 [1:43:03<8:20:21,  1.48s/it]
                                                        
{'loss': 0.1977, 'learning_rate': 5e-05, 'epoch': 0.85}

 17%|█▋        | 4145/24420 [1:43:03<8:20:21,  1.48s/it]
 17%|█▋        | 4146/24420 [1:43:05<8:20:46,  1.48s/it]
 17%|█▋        | 4147/24420 [1:43:06<8:21:46,  1.49s/it]
 17%|█▋        | 4148/24420 [1:43:08<8:22:01,  1.49s/it]
 17%|█▋        | 4149/24420 [1:43:09<8:20:48,  1.48s/it]
 17%|█▋        | 4150/24420 [1:43:11<8:20:24,  1.48s/it]
                                                        
{'loss': 0.1563, 'learning_rate': 5e-05, 'epoch': 0.85}

 17%|█▋        | 4150/24420 [1:43:11<8:20:24,  1.48s/it]
 17%|█▋        | 4151/24420 [1:43:12<8:21:03,  1.48s/it]
 17%|█▋        | 4152/24420 [1:43:14<8:20:34,  1.48s/it]
 17%|█▋        | 4153/24420 [1:43:15<8:19:50,  1.48s/it]
 17%|█▋        | 4154/24420 [1:43:16<8:20:06,  1.48s/it]
 17%|█▋        | 4155/24420 [1:43:18<8:20:13,  1.48s/it]
                                                        
{'loss': 0.1512, 'learning_rate': 5e-05, 'epoch': 0.85}

 17%|█▋        | 4155/24420 [1:43:18<8:20:13,  1.48s/it]
 17%|█▋        | 4156/24420 [1:43:19<8:20:30,  1.48s/it]
 17%|█▋        | 4157/24420 [1:43:21<8:20:49,  1.48s/it]
 17%|█▋        | 4158/24420 [1:43:22<8:20:53,  1.48s/it]
 17%|█▋        | 4159/24420 [1:43:24<8:20:00,  1.48s/it]
 17%|█▋        | 4160/24420 [1:43:25<8:20:13,  1.48s/it]
                                                        
{'loss': 0.1683, 'learning_rate': 5e-05, 'epoch': 0.85}

 17%|█▋        | 4160/24420 [1:43:25<8:20:13,  1.48s/it]
 17%|█▋        | 4161/24420 [1:43:27<8:20:47,  1.48s/it]
 17%|█▋        | 4162/24420 [1:43:28<8:21:06,  1.48s/it]
 17%|█▋        | 4163/24420 [1:43:30<8:20:31,  1.48s/it]
 17%|█▋        | 4164/24420 [1:43:31<8:19:56,  1.48s/it]
 17%|█▋        | 4165/24420 [1:43:33<8:20:05,  1.48s/it]
                                                        
{'loss': 0.1865, 'learning_rate': 5e-05, 'epoch': 0.85}

 17%|█▋        | 4165/24420 [1:43:33<8:20:05,  1.48s/it]
 17%|█▋        | 4166/24420 [1:43:34<8:19:35,  1.48s/it]
 17%|█▋        | 4167/24420 [1:43:36<8:20:37,  1.48s/it]
 17%|█▋        | 4168/24420 [1:43:37<8:20:58,  1.48s/it]
 17%|█▋        | 4169/24420 [1:43:39<8:20:37,  1.48s/it]
 17%|█▋        | 4170/24420 [1:43:40<8:23:07,  1.49s/it]
                                                        
{'loss': 0.1832, 'learning_rate': 5e-05, 'epoch': 0.85}

 17%|█▋        | 4170/24420 [1:43:40<8:23:07,  1.49s/it]
 17%|█▋        | 4171/24420 [1:43:42<8:22:31,  1.49s/it]
 17%|█▋        | 4172/24420 [1:43:43<8:21:50,  1.49s/it]
 17%|█▋        | 4173/24420 [1:43:45<8:21:11,  1.49s/it]
 17%|█▋        | 4174/24420 [1:43:46<8:20:05,  1.48s/it]
 17%|█▋        | 4175/24420 [1:43:48<8:19:39,  1.48s/it]
                                                        
{'loss': 0.1778, 'learning_rate': 5e-05, 'epoch': 0.85}

 17%|█▋        | 4175/24420 [1:43:48<8:19:39,  1.48s/it]
 17%|█▋        | 4176/24420 [1:43:49<8:20:25,  1.48s/it]
 17%|█▋        | 4177/24420 [1:43:51<8:19:13,  1.48s/it]
 17%|█▋        | 4178/24420 [1:43:52<8:20:13,  1.48s/it]
 17%|█▋        | 4179/24420 [1:43:54<8:20:55,  1.48s/it]
 17%|█▋        | 4180/24420 [1:43:55<8:21:03,  1.49s/it]
                                                        
{'loss': 0.1756, 'learning_rate': 5e-05, 'epoch': 0.86}

 17%|█▋        | 4180/24420 [1:43:55<8:21:03,  1.49s/it]
 17%|█▋        | 4181/24420 [1:43:57<8:21:54,  1.49s/it]
 17%|█▋        | 4182/24420 [1:43:58<8:21:01,  1.49s/it]
 17%|█▋        | 4183/24420 [1:44:00<8:21:55,  1.49s/it]
 17%|█▋        | 4184/24420 [1:44:01<8:21:52,  1.49s/it]
 17%|█▋        | 4185/24420 [1:44:03<8:21:29,  1.49s/it]
                                                        
{'loss': 0.1435, 'learning_rate': 5e-05, 'epoch': 0.86}

 17%|█▋        | 4185/24420 [1:44:03<8:21:29,  1.49s/it]
 17%|█▋        | 4186/24420 [1:44:04<8:21:11,  1.49s/it]
 17%|█▋        | 4187/24420 [1:44:05<8:20:11,  1.48s/it]
 17%|█▋        | 4188/24420 [1:44:07<8:19:26,  1.48s/it]
 17%|█▋        | 4189/24420 [1:44:08<8:19:06,  1.48s/it]
 17%|█▋        | 4190/24420 [1:44:10<8:18:56,  1.48s/it]
                                                        
{'loss': 0.1648, 'learning_rate': 5e-05, 'epoch': 0.86}

 17%|█▋        | 4190/24420 [1:44:10<8:18:56,  1.48s/it]
 17%|█▋        | 4191/24420 [1:44:11<8:19:38,  1.48s/it]
 17%|█▋        | 4192/24420 [1:44:13<8:19:30,  1.48s/it]
 17%|█▋        | 4193/24420 [1:44:14<8:19:03,  1.48s/it]
 17%|█▋        | 4194/24420 [1:44:16<8:19:01,  1.48s/it]
 17%|█▋        | 4195/24420 [1:44:17<8:18:49,  1.48s/it]
                                                        
{'loss': 0.1811, 'learning_rate': 5e-05, 'epoch': 0.86}

 17%|█▋        | 4195/24420 [1:44:17<8:18:49,  1.48s/it]
 17%|█▋        | 4196/24420 [1:44:19<8:18:48,  1.48s/it]
 17%|█▋        | 4197/24420 [1:44:20<8:18:13,  1.48s/it]
 17%|█▋        | 4198/24420 [1:44:22<8:18:24,  1.48s/it]
 17%|█▋        | 4199/24420 [1:44:23<8:18:22,  1.48s/it]
 17%|█▋        | 4200/24420 [1:44:25<8:17:46,  1.48s/it]
                                                        
{'loss': 0.1585, 'learning_rate': 5e-05, 'epoch': 0.86}

 17%|█▋        | 4200/24420 [1:44:25<8:17:46,  1.48s/it]
 17%|█▋        | 4201/24420 [1:44:30<13:58:26,  2.49s/it]
 17%|█▋        | 4202/24420 [1:44:31<12:16:34,  2.19s/it]
 17%|█▋        | 4203/24420 [1:44:32<11:05:09,  1.97s/it]
 17%|█▋        | 4204/24420 [1:44:34<10:15:07,  1.83s/it]
 17%|█▋        | 4205/24420 [1:44:35<9:39:58,  1.72s/it] 
                                                        
{'loss': 0.2436, 'learning_rate': 5e-05, 'epoch': 0.86}

 17%|█▋        | 4205/24420 [1:44:35<9:39:58,  1.72s/it]
 17%|█▋        | 4206/24420 [1:44:37<9:15:10,  1.65s/it]
 17%|█▋        | 4207/24420 [1:44:38<8:57:42,  1.60s/it]
 17%|█▋        | 4208/24420 [1:44:40<8:45:11,  1.56s/it]
 17%|█▋        | 4209/24420 [1:44:41<8:36:59,  1.53s/it]
 17%|█▋        | 4210/24420 [1:44:43<8:30:59,  1.52s/it]
                                                        
{'loss': 0.1744, 'learning_rate': 5e-05, 'epoch': 0.86}

 17%|█▋        | 4210/24420 [1:44:43<8:30:59,  1.52s/it]
 17%|█▋        | 4211/24420 [1:44:44<8:26:38,  1.50s/it]
 17%|█▋        | 4212/24420 [1:44:46<8:23:06,  1.49s/it]
 17%|█▋        | 4213/24420 [1:44:47<8:22:05,  1.49s/it]
 17%|█▋        | 4214/24420 [1:44:49<8:20:35,  1.49s/it]
 17%|█▋        | 4215/24420 [1:44:50<8:19:37,  1.48s/it]
                                                        
{'loss': 0.1697, 'learning_rate': 5e-05, 'epoch': 0.86}

 17%|█▋        | 4215/24420 [1:44:50<8:19:37,  1.48s/it]
 17%|█▋        | 4216/24420 [1:44:52<8:19:17,  1.48s/it]
 17%|█▋        | 4217/24420 [1:44:53<8:18:02,  1.48s/it]
 17%|█▋        | 4218/24420 [1:44:55<8:17:12,  1.48s/it]
 17%|█▋        | 4219/24420 [1:44:56<8:17:20,  1.48s/it]
 17%|█▋        | 4220/24420 [1:44:58<8:18:20,  1.48s/it]
                                                        
{'loss': 0.1407, 'learning_rate': 5e-05, 'epoch': 0.86}

 17%|█▋        | 4220/24420 [1:44:58<8:18:20,  1.48s/it]
 17%|█▋        | 4221/24420 [1:44:59<8:17:51,  1.48s/it]
 17%|█▋        | 4222/24420 [1:45:01<8:20:38,  1.49s/it]
 17%|█▋        | 4223/24420 [1:45:02<8:20:13,  1.49s/it]
 17%|█▋        | 4224/24420 [1:45:04<8:20:23,  1.49s/it]
 17%|█▋        | 4225/24420 [1:45:05<8:19:52,  1.49s/it]
                                                        
{'loss': 0.2023, 'learning_rate': 5e-05, 'epoch': 0.87}

 17%|█▋        | 4225/24420 [1:45:05<8:19:52,  1.49s/it]
 17%|█▋        | 4226/24420 [1:45:07<8:18:46,  1.48s/it]
 17%|█▋        | 4227/24420 [1:45:08<8:18:23,  1.48s/it]
 17%|█▋        | 4228/24420 [1:45:09<8:19:02,  1.48s/it]
 17%|█▋        | 4229/24420 [1:45:11<8:18:38,  1.48s/it]
 17%|█▋        | 4230/24420 [1:45:12<8:17:54,  1.48s/it]
                                                        
{'loss': 0.1711, 'learning_rate': 5e-05, 'epoch': 0.87}

 17%|█▋        | 4230/24420 [1:45:12<8:17:54,  1.48s/it]
 17%|█▋        | 4231/24420 [1:45:14<8:17:15,  1.48s/it]
 17%|█▋        | 4232/24420 [1:45:15<8:18:04,  1.48s/it]
 17%|█▋        | 4233/24420 [1:45:17<8:17:31,  1.48s/it]
 17%|█▋        | 4234/24420 [1:45:18<8:17:51,  1.48s/it]
 17%|█▋        | 4235/24420 [1:45:20<8:17:03,  1.48s/it]
                                                        
{'loss': 0.1521, 'learning_rate': 5e-05, 'epoch': 0.87}

 17%|█▋        | 4235/24420 [1:45:20<8:17:03,  1.48s/it]
 17%|█▋        | 4236/24420 [1:45:21<8:17:19,  1.48s/it]
 17%|█▋        | 4237/24420 [1:45:23<8:16:49,  1.48s/it]
 17%|█▋        | 4238/24420 [1:45:24<8:16:15,  1.48s/it]
 17%|█▋        | 4239/24420 [1:45:26<8:16:13,  1.48s/it]
 17%|█▋        | 4240/24420 [1:45:27<8:16:58,  1.48s/it]
                                                        
{'loss': 0.1725, 'learning_rate': 5e-05, 'epoch': 0.87}

 17%|█▋        | 4240/24420 [1:45:27<8:16:58,  1.48s/it]
 17%|█▋        | 4241/24420 [1:45:29<8:16:42,  1.48s/it]
 17%|█▋        | 4242/24420 [1:45:30<8:16:37,  1.48s/it]
 17%|█▋        | 4243/24420 [1:45:32<8:15:56,  1.47s/it]
 17%|█▋        | 4244/24420 [1:45:33<8:15:58,  1.47s/it]
 17%|█▋        | 4245/24420 [1:45:35<8:15:58,  1.48s/it]
                                                        
{'loss': 0.1527, 'learning_rate': 5e-05, 'epoch': 0.87}

 17%|█▋        | 4245/24420 [1:45:35<8:15:58,  1.48s/it]
 17%|█▋        | 4246/24420 [1:45:36<8:15:39,  1.47s/it]
 17%|█▋        | 4247/24420 [1:45:38<8:15:24,  1.47s/it]
 17%|█▋        | 4248/24420 [1:45:39<8:16:34,  1.48s/it]
 17%|█▋        | 4249/24420 [1:45:40<8:17:01,  1.48s/it]
 17%|█▋        | 4250/24420 [1:45:42<8:16:54,  1.48s/it]
                                                        
{'loss': 0.1742, 'learning_rate': 5e-05, 'epoch': 0.87}

 17%|█▋        | 4250/24420 [1:45:42<8:16:54,  1.48s/it]
 17%|█▋        | 4251/24420 [1:45:43<8:17:28,  1.48s/it]
 17%|█▋        | 4252/24420 [1:45:45<8:17:23,  1.48s/it]
 17%|█▋        | 4253/24420 [1:45:46<8:16:58,  1.48s/it]
 17%|█▋        | 4254/24420 [1:45:48<8:16:40,  1.48s/it]
 17%|█▋        | 4255/24420 [1:45:49<8:17:28,  1.48s/it]
                                                        
{'loss': 0.1772, 'learning_rate': 5e-05, 'epoch': 0.87}

 17%|█▋        | 4255/24420 [1:45:49<8:17:28,  1.48s/it]
 17%|█▋        | 4256/24420 [1:45:51<8:17:57,  1.48s/it]
 17%|█▋        | 4257/24420 [1:45:52<8:16:48,  1.48s/it]
 17%|█▋        | 4258/24420 [1:45:54<8:16:27,  1.48s/it]
 17%|█▋        | 4259/24420 [1:45:55<8:16:47,  1.48s/it]
 17%|█▋        | 4260/24420 [1:45:57<8:16:25,  1.48s/it]
                                                        
{'loss': 0.1786, 'learning_rate': 5e-05, 'epoch': 0.87}

 17%|█▋        | 4260/24420 [1:45:57<8:16:25,  1.48s/it]
 17%|█▋        | 4261/24420 [1:45:58<8:16:57,  1.48s/it]
 17%|█▋        | 4262/24420 [1:46:00<8:17:03,  1.48s/it]
 17%|█▋        | 4263/24420 [1:46:01<8:16:58,  1.48s/it]
 17%|█▋        | 4264/24420 [1:46:03<8:16:50,  1.48s/it]
 17%|█▋        | 4265/24420 [1:46:04<8:17:07,  1.48s/it]
                                                        
{'loss': 0.1968, 'learning_rate': 5e-05, 'epoch': 0.87}

 17%|█▋        | 4265/24420 [1:46:04<8:17:07,  1.48s/it]
 17%|█▋        | 4266/24420 [1:46:06<8:17:01,  1.48s/it]
 17%|█▋        | 4267/24420 [1:46:07<8:16:16,  1.48s/it]
 17%|█▋        | 4268/24420 [1:46:09<8:16:10,  1.48s/it]
 17%|█▋        | 4269/24420 [1:46:10<8:16:24,  1.48s/it]
 17%|█▋        | 4270/24420 [1:46:12<8:16:08,  1.48s/it]
                                                        
{'loss': 0.1455, 'learning_rate': 5e-05, 'epoch': 0.87}

 17%|█▋        | 4270/24420 [1:46:12<8:16:08,  1.48s/it]
 17%|█▋        | 4271/24420 [1:46:13<8:15:43,  1.48s/it]
 17%|█▋        | 4272/24420 [1:46:14<8:16:32,  1.48s/it]
 17%|█▋        | 4273/24420 [1:46:16<8:16:45,  1.48s/it]
 18%|█▊        | 4274/24420 [1:46:17<8:17:09,  1.48s/it]
 18%|█▊        | 4275/24420 [1:46:19<8:17:07,  1.48s/it]
                                                        
{'loss': 0.1532, 'learning_rate': 5e-05, 'epoch': 0.88}

 18%|█▊        | 4275/24420 [1:46:19<8:17:07,  1.48s/it]
 18%|█▊        | 4276/24420 [1:46:20<8:16:53,  1.48s/it]
 18%|█▊        | 4277/24420 [1:46:22<8:17:01,  1.48s/it]
 18%|█▊        | 4278/24420 [1:46:23<8:17:38,  1.48s/it]
 18%|█▊        | 4279/24420 [1:46:25<8:17:00,  1.48s/it]
 18%|█▊        | 4280/24420 [1:46:26<8:16:21,  1.48s/it]
                                                        
{'loss': 0.1697, 'learning_rate': 5e-05, 'epoch': 0.88}

 18%|█▊        | 4280/24420 [1:46:26<8:16:21,  1.48s/it]
 18%|█▊        | 4281/24420 [1:46:28<8:17:03,  1.48s/it]
 18%|█▊        | 4282/24420 [1:46:29<8:17:56,  1.48s/it]
 18%|█▊        | 4283/24420 [1:46:31<8:17:18,  1.48s/it]
 18%|█▊        | 4284/24420 [1:46:32<8:16:29,  1.48s/it]
 18%|█▊        | 4285/24420 [1:46:34<8:17:00,  1.48s/it]
                                                        
{'loss': 0.1509, 'learning_rate': 5e-05, 'epoch': 0.88}

 18%|█▊        | 4285/24420 [1:46:34<8:17:00,  1.48s/it]
 18%|█▊        | 4286/24420 [1:46:35<8:17:31,  1.48s/it]
 18%|█▊        | 4287/24420 [1:46:37<8:16:53,  1.48s/it]
 18%|█▊        | 4288/24420 [1:46:38<8:17:13,  1.48s/it]
 18%|█▊        | 4289/24420 [1:46:40<8:17:06,  1.48s/it]
 18%|█▊        | 4290/24420 [1:46:41<8:17:27,  1.48s/it]
                                                        
{'loss': 0.1375, 'learning_rate': 5e-05, 'epoch': 0.88}

 18%|█▊        | 4290/24420 [1:46:41<8:17:27,  1.48s/it]
 18%|█▊        | 4291/24420 [1:46:43<8:16:46,  1.48s/it]
 18%|█▊        | 4292/24420 [1:46:44<8:17:14,  1.48s/it]
 18%|█▊        | 4293/24420 [1:46:46<8:16:43,  1.48s/it]
 18%|█▊        | 4294/24420 [1:46:47<8:16:57,  1.48s/it]
 18%|█▊        | 4295/24420 [1:46:49<8:17:41,  1.48s/it]
                                                        
{'loss': 0.1823, 'learning_rate': 5e-05, 'epoch': 0.88}

 18%|█▊        | 4295/24420 [1:46:49<8:17:41,  1.48s/it]
 18%|█▊        | 4296/24420 [1:46:50<8:18:07,  1.49s/it]
 18%|█▊        | 4297/24420 [1:46:52<8:17:10,  1.48s/it]
 18%|█▊        | 4298/24420 [1:46:53<8:17:29,  1.48s/it]
 18%|█▊        | 4299/24420 [1:46:55<8:17:09,  1.48s/it]
 18%|█▊        | 4300/24420 [1:46:56<8:18:19,  1.49s/it]
                                                        
{'loss': 0.1701, 'learning_rate': 5e-05, 'epoch': 0.88}

 18%|█▊        | 4300/24420 [1:46:56<8:18:19,  1.49s/it]
 18%|█▊        | 4301/24420 [1:46:57<8:17:33,  1.48s/it]
 18%|█▊        | 4302/24420 [1:46:59<8:17:51,  1.48s/it]
 18%|█▊        | 4303/24420 [1:47:00<8:17:09,  1.48s/it]
 18%|█▊        | 4304/24420 [1:47:02<8:17:34,  1.48s/it]
 18%|█▊        | 4305/24420 [1:47:03<8:17:01,  1.48s/it]
                                                        
{'loss': 0.1706, 'learning_rate': 5e-05, 'epoch': 0.88}

 18%|█▊        | 4305/24420 [1:47:03<8:17:01,  1.48s/it]
 18%|█▊        | 4306/24420 [1:47:05<8:16:44,  1.48s/it]
 18%|█▊        | 4307/24420 [1:47:06<8:16:53,  1.48s/it]
 18%|█▊        | 4308/24420 [1:47:08<8:18:27,  1.49s/it]
 18%|█▊        | 4309/24420 [1:47:09<8:17:51,  1.49s/it]
 18%|█▊        | 4310/24420 [1:47:11<8:16:49,  1.48s/it]
                                                        
{'loss': 0.1574, 'learning_rate': 5e-05, 'epoch': 0.88}

 18%|█▊        | 4310/24420 [1:47:11<8:16:49,  1.48s/it]
 18%|█▊        | 4311/24420 [1:47:12<8:18:36,  1.49s/it]
 18%|█▊        | 4312/24420 [1:47:14<8:18:12,  1.49s/it]
 18%|█▊        | 4313/24420 [1:47:15<8:18:50,  1.49s/it]
 18%|█▊        | 4314/24420 [1:47:17<8:22:02,  1.50s/it]
 18%|█▊        | 4315/24420 [1:47:18<8:19:39,  1.49s/it]
                                                        
{'loss': 0.1497, 'learning_rate': 5e-05, 'epoch': 0.88}

 18%|█▊        | 4315/24420 [1:47:18<8:19:39,  1.49s/it]
 18%|█▊        | 4316/24420 [1:47:20<8:18:57,  1.49s/it]
 18%|█▊        | 4317/24420 [1:47:21<8:18:09,  1.49s/it]
 18%|█▊        | 4318/24420 [1:47:23<8:18:15,  1.49s/it]
 18%|█▊        | 4319/24420 [1:47:24<8:17:35,  1.49s/it]
 18%|█▊        | 4320/24420 [1:47:26<8:17:21,  1.48s/it]
                                                        
{'loss': 0.1415, 'learning_rate': 5e-05, 'epoch': 0.88}

 18%|█▊        | 4320/24420 [1:47:26<8:17:21,  1.48s/it]
 18%|█▊        | 4321/24420 [1:47:27<8:17:19,  1.48s/it]
 18%|█▊        | 4322/24420 [1:47:29<8:16:25,  1.48s/it]
 18%|█▊        | 4323/24420 [1:47:30<8:16:26,  1.48s/it]
 18%|█▊        | 4324/24420 [1:47:32<8:16:12,  1.48s/it]
 18%|█▊        | 4325/24420 [1:47:33<8:15:37,  1.48s/it]
                                                        
{'loss': 0.1462, 'learning_rate': 5e-05, 'epoch': 0.89}

 18%|█▊        | 4325/24420 [1:47:33<8:15:37,  1.48s/it]
 18%|█▊        | 4326/24420 [1:47:35<8:15:47,  1.48s/it]
 18%|█▊        | 4327/24420 [1:47:36<8:15:46,  1.48s/it]
 18%|█▊        | 4328/24420 [1:47:38<8:16:29,  1.48s/it]
 18%|█▊        | 4329/24420 [1:47:39<8:16:31,  1.48s/it]
 18%|█▊        | 4330/24420 [1:47:41<8:16:17,  1.48s/it]
                                                        
{'loss': 0.1608, 'learning_rate': 5e-05, 'epoch': 0.89}

 18%|█▊        | 4330/24420 [1:47:41<8:16:17,  1.48s/it]
 18%|█▊        | 4331/24420 [1:47:42<8:15:50,  1.48s/it]
 18%|█▊        | 4332/24420 [1:47:43<8:15:21,  1.48s/it]
 18%|█▊        | 4333/24420 [1:47:45<8:15:31,  1.48s/it]
 18%|█▊        | 4334/24420 [1:47:46<8:17:26,  1.49s/it]
 18%|█▊        | 4335/24420 [1:47:48<8:17:26,  1.49s/it]
                                                        
{'loss': 0.166, 'learning_rate': 5e-05, 'epoch': 0.89}

 18%|█▊        | 4335/24420 [1:47:48<8:17:26,  1.49s/it]
 18%|█▊        | 4336/24420 [1:47:49<8:16:51,  1.48s/it]
 18%|█▊        | 4337/24420 [1:47:51<8:16:02,  1.48s/it]
 18%|█▊        | 4338/24420 [1:47:52<8:15:27,  1.48s/it]
 18%|█▊        | 4339/24420 [1:47:54<8:15:16,  1.48s/it]
 18%|█▊        | 4340/24420 [1:47:55<8:15:33,  1.48s/it]
                                                        
{'loss': 0.1482, 'learning_rate': 5e-05, 'epoch': 0.89}

 18%|█▊        | 4340/24420 [1:47:55<8:15:33,  1.48s/it]
 18%|█▊        | 4341/24420 [1:47:57<8:16:07,  1.48s/it]
 18%|█▊        | 4342/24420 [1:47:58<8:15:40,  1.48s/it]
 18%|█▊        | 4343/24420 [1:48:00<8:15:03,  1.48s/it]
 18%|█▊        | 4344/24420 [1:48:01<8:15:25,  1.48s/it]
 18%|█▊        | 4345/24420 [1:48:03<8:15:04,  1.48s/it]
                                                        
{'loss': 0.1996, 'learning_rate': 5e-05, 'epoch': 0.89}

 18%|█▊        | 4345/24420 [1:48:03<8:15:04,  1.48s/it]
 18%|█▊        | 4346/24420 [1:48:04<8:15:26,  1.48s/it]
 18%|█▊        | 4347/24420 [1:48:06<8:15:01,  1.48s/it]
 18%|█▊        | 4348/24420 [1:48:07<8:14:56,  1.48s/it]
 18%|█▊        | 4349/24420 [1:48:09<8:15:32,  1.48s/it]
 18%|█▊        | 4350/24420 [1:48:10<8:16:04,  1.48s/it]
                                                        
{'loss': 0.1792, 'learning_rate': 5e-05, 'epoch': 0.89}

 18%|█▊        | 4350/24420 [1:48:10<8:16:04,  1.48s/it]
 18%|█▊        | 4351/24420 [1:48:12<8:16:31,  1.48s/it]
 18%|█▊        | 4352/24420 [1:48:13<8:15:57,  1.48s/it]
 18%|█▊        | 4353/24420 [1:48:15<8:14:43,  1.48s/it]
 18%|█▊        | 4354/24420 [1:48:16<8:15:14,  1.48s/it]
 18%|█▊        | 4355/24420 [1:48:18<8:14:31,  1.48s/it]
                                                        
{'loss': 0.1503, 'learning_rate': 5e-05, 'epoch': 0.89}

 18%|█▊        | 4355/24420 [1:48:18<8:14:31,  1.48s/it]
 18%|█▊        | 4356/24420 [1:48:19<8:15:36,  1.48s/it]
 18%|█▊        | 4357/24420 [1:48:21<8:15:20,  1.48s/it]
 18%|█▊        | 4358/24420 [1:48:22<8:14:53,  1.48s/it]
 18%|█▊        | 4359/24420 [1:48:23<8:14:32,  1.48s/it]
 18%|█▊        | 4360/24420 [1:48:25<8:13:58,  1.48s/it]
                                                        
{'loss': 0.1548, 'learning_rate': 5e-05, 'epoch': 0.89}

 18%|█▊        | 4360/24420 [1:48:25<8:13:58,  1.48s/it]
 18%|█▊        | 4361/24420 [1:48:26<8:13:35,  1.48s/it]
 18%|█▊        | 4362/24420 [1:48:28<8:13:46,  1.48s/it]
 18%|█▊        | 4363/24420 [1:48:29<8:13:23,  1.48s/it]
 18%|█▊        | 4364/24420 [1:48:31<8:13:09,  1.48s/it]
 18%|█▊        | 4365/24420 [1:48:32<8:13:18,  1.48s/it]
                                                        
{'loss': 0.1702, 'learning_rate': 5e-05, 'epoch': 0.89}

 18%|█▊        | 4365/24420 [1:48:32<8:13:18,  1.48s/it]
 18%|█▊        | 4366/24420 [1:48:34<8:13:40,  1.48s/it]
 18%|█▊        | 4367/24420 [1:48:35<8:13:03,  1.48s/it]
 18%|█▊        | 4368/24420 [1:48:37<8:12:50,  1.47s/it]
 18%|█▊        | 4369/24420 [1:48:38<8:13:37,  1.48s/it]
 18%|█▊        | 4370/24420 [1:48:40<8:13:45,  1.48s/it]
                                                        
{'loss': 0.1525, 'learning_rate': 5e-05, 'epoch': 0.89}

 18%|█▊        | 4370/24420 [1:48:40<8:13:45,  1.48s/it]
 18%|█▊        | 4371/24420 [1:48:41<8:14:17,  1.48s/it]
 18%|█▊        | 4372/24420 [1:48:43<8:14:22,  1.48s/it]
 18%|█▊        | 4373/24420 [1:48:44<8:14:37,  1.48s/it]
 18%|█▊        | 4374/24420 [1:48:46<8:13:50,  1.48s/it]
 18%|█▊        | 4375/24420 [1:48:47<8:14:10,  1.48s/it]
                                                        
{'loss': 0.1485, 'learning_rate': 5e-05, 'epoch': 0.9}

 18%|█▊        | 4375/24420 [1:48:47<8:14:10,  1.48s/it]
 18%|█▊        | 4376/24420 [1:48:49<8:14:59,  1.48s/it]
 18%|█▊        | 4377/24420 [1:48:50<8:15:59,  1.48s/it]
 18%|█▊        | 4378/24420 [1:48:52<8:14:14,  1.48s/it]
 18%|█▊        | 4379/24420 [1:48:53<8:14:56,  1.48s/it]
 18%|█▊        | 4380/24420 [1:48:55<8:15:16,  1.48s/it]
                                                        
{'loss': 0.1792, 'learning_rate': 5e-05, 'epoch': 0.9}

 18%|█▊        | 4380/24420 [1:48:55<8:15:16,  1.48s/it]
 18%|█▊        | 4381/24420 [1:48:56<8:14:52,  1.48s/it]
 18%|█▊        | 4382/24420 [1:48:57<8:14:29,  1.48s/it]
 18%|█▊        | 4383/24420 [1:48:59<8:15:01,  1.48s/it]
 18%|█▊        | 4384/24420 [1:49:00<8:14:55,  1.48s/it]
 18%|█▊        | 4385/24420 [1:49:02<8:16:56,  1.49s/it]
                                                        
{'loss': 0.1599, 'learning_rate': 5e-05, 'epoch': 0.9}

 18%|█▊        | 4385/24420 [1:49:02<8:16:56,  1.49s/it]
 18%|█▊        | 4386/24420 [1:49:03<8:17:29,  1.49s/it]
 18%|█▊        | 4387/24420 [1:49:05<8:16:01,  1.49s/it]
 18%|█▊        | 4388/24420 [1:49:06<8:17:42,  1.49s/it]
 18%|█▊        | 4389/24420 [1:49:08<8:17:03,  1.49s/it]
 18%|█▊        | 4390/24420 [1:49:09<8:16:10,  1.49s/it]
                                                        
{'loss': 0.1554, 'learning_rate': 5e-05, 'epoch': 0.9}

 18%|█▊        | 4390/24420 [1:49:09<8:16:10,  1.49s/it]
 18%|█▊        | 4391/24420 [1:49:11<8:15:20,  1.48s/it]
 18%|█▊        | 4392/24420 [1:49:12<8:16:45,  1.49s/it]
 18%|█▊        | 4393/24420 [1:49:14<8:16:53,  1.49s/it]
 18%|█▊        | 4394/24420 [1:49:15<8:16:06,  1.49s/it]
 18%|█▊        | 4395/24420 [1:49:17<8:15:51,  1.49s/it]
                                                        
{'loss': 0.1744, 'learning_rate': 5e-05, 'epoch': 0.9}

 18%|█▊        | 4395/24420 [1:49:17<8:15:51,  1.49s/it]
 18%|█▊        | 4396/24420 [1:49:18<8:14:38,  1.48s/it]
 18%|█▊        | 4397/24420 [1:49:20<8:14:11,  1.48s/it]
 18%|█▊        | 4398/24420 [1:49:21<8:14:56,  1.48s/it]
 18%|█▊        | 4399/24420 [1:49:23<8:14:29,  1.48s/it]
 18%|█▊        | 4400/24420 [1:49:24<8:13:39,  1.48s/it]
                                                        
{'loss': 0.1672, 'learning_rate': 5e-05, 'epoch': 0.9}

 18%|█▊        | 4400/24420 [1:49:24<8:13:39,  1.48s/it]
 18%|█▊        | 4401/24420 [1:49:29<13:20:17,  2.40s/it]
 18%|█▊        | 4402/24420 [1:49:30<11:48:01,  2.12s/it]
 18%|█▊        | 4403/24420 [1:49:32<10:45:05,  1.93s/it]
 18%|█▊        | 4404/24420 [1:49:33<10:00:12,  1.80s/it]
 18%|█▊        | 4405/24420 [1:49:35<9:28:51,  1.71s/it] 
                                                        
{'loss': 0.174, 'learning_rate': 5e-05, 'epoch': 0.9}

 18%|█▊        | 4405/24420 [1:49:35<9:28:51,  1.71s/it]
 18%|█▊        | 4406/24420 [1:49:36<9:07:28,  1.64s/it]
 18%|█▊        | 4407/24420 [1:49:38<8:51:20,  1.59s/it]
 18%|█▊        | 4408/24420 [1:49:39<8:40:24,  1.56s/it]
 18%|█▊        | 4409/24420 [1:49:41<8:34:00,  1.54s/it]
 18%|█▊        | 4410/24420 [1:49:42<8:27:28,  1.52s/it]
                                                        
{'loss': 0.163, 'learning_rate': 5e-05, 'epoch': 0.9}

 18%|█▊        | 4410/24420 [1:49:42<8:27:28,  1.52s/it]
 18%|█▊        | 4411/24420 [1:49:44<8:23:00,  1.51s/it]
 18%|█▊        | 4412/24420 [1:49:45<8:20:22,  1.50s/it]
 18%|█▊        | 4413/24420 [1:49:47<8:18:17,  1.49s/it]
 18%|█▊        | 4414/24420 [1:49:48<8:16:24,  1.49s/it]
 18%|█▊        | 4415/24420 [1:49:50<8:16:22,  1.49s/it]
                                                        
{'loss': 0.1516, 'learning_rate': 5e-05, 'epoch': 0.9}

 18%|█▊        | 4415/24420 [1:49:50<8:16:22,  1.49s/it]
 18%|█▊        | 4416/24420 [1:49:51<8:15:37,  1.49s/it]
 18%|█▊        | 4417/24420 [1:49:53<8:15:13,  1.49s/it]
 18%|█▊        | 4418/24420 [1:49:54<8:14:23,  1.48s/it]
 18%|█▊        | 4419/24420 [1:49:55<8:15:29,  1.49s/it]
 18%|█▊        | 4420/24420 [1:49:57<8:15:28,  1.49s/it]
                                                        
{'loss': 0.152, 'learning_rate': 5e-05, 'epoch': 0.9}

 18%|█▊        | 4420/24420 [1:49:57<8:15:28,  1.49s/it]
 18%|█▊        | 4421/24420 [1:49:58<8:15:07,  1.49s/it]
 18%|█▊        | 4422/24420 [1:50:00<8:14:43,  1.48s/it]
 18%|█▊        | 4423/24420 [1:50:01<8:13:47,  1.48s/it]
 18%|█▊        | 4424/24420 [1:50:03<8:13:21,  1.48s/it]
 18%|█▊        | 4425/24420 [1:50:04<8:13:03,  1.48s/it]
                                                        
{'loss': 0.1536, 'learning_rate': 5e-05, 'epoch': 0.91}

 18%|█▊        | 4425/24420 [1:50:04<8:13:03,  1.48s/it]
 18%|█▊        | 4426/24420 [1:50:06<8:12:48,  1.48s/it]
 18%|█▊        | 4427/24420 [1:50:07<8:13:00,  1.48s/it]
 18%|█▊        | 4428/24420 [1:50:09<8:13:19,  1.48s/it]
 18%|█▊        | 4429/24420 [1:50:10<8:13:35,  1.48s/it]
 18%|█▊        | 4430/24420 [1:50:12<8:12:57,  1.48s/it]
                                                        
{'loss': 0.1527, 'learning_rate': 5e-05, 'epoch': 0.91}

 18%|█▊        | 4430/24420 [1:50:12<8:12:57,  1.48s/it]
 18%|█▊        | 4431/24420 [1:50:13<8:13:04,  1.48s/it]
 18%|█▊        | 4432/24420 [1:50:15<8:12:40,  1.48s/it]
 18%|█▊        | 4433/24420 [1:50:16<8:12:14,  1.48s/it]
 18%|█▊        | 4434/24420 [1:50:18<8:12:14,  1.48s/it]
 18%|█▊        | 4435/24420 [1:50:19<8:12:06,  1.48s/it]
                                                        
{'loss': 0.1297, 'learning_rate': 5e-05, 'epoch': 0.91}

 18%|█▊        | 4435/24420 [1:50:19<8:12:06,  1.48s/it]
 18%|█▊        | 4436/24420 [1:50:21<8:12:24,  1.48s/it]
 18%|█▊        | 4437/24420 [1:50:22<8:13:03,  1.48s/it]
 18%|█▊        | 4438/24420 [1:50:24<8:13:07,  1.48s/it]
 18%|█▊        | 4439/24420 [1:50:25<8:12:48,  1.48s/it]
 18%|█▊        | 4440/24420 [1:50:27<8:13:21,  1.48s/it]
                                                        
{'loss': 0.1703, 'learning_rate': 5e-05, 'epoch': 0.91}

 18%|█▊        | 4440/24420 [1:50:27<8:13:21,  1.48s/it]
 18%|█▊        | 4441/24420 [1:50:28<8:12:54,  1.48s/it]
 18%|█▊        | 4442/24420 [1:50:30<8:12:25,  1.48s/it]
 18%|█▊        | 4443/24420 [1:50:31<8:12:56,  1.48s/it]
 18%|█▊        | 4444/24420 [1:50:32<8:12:18,  1.48s/it]
 18%|█▊        | 4445/24420 [1:50:34<8:12:25,  1.48s/it]
                                                        
{'loss': 0.1705, 'learning_rate': 5e-05, 'epoch': 0.91}

 18%|█▊        | 4445/24420 [1:50:34<8:12:25,  1.48s/it]
 18%|█▊        | 4446/24420 [1:50:35<8:15:11,  1.49s/it]
 18%|█▊        | 4447/24420 [1:50:37<8:14:17,  1.48s/it]
 18%|█▊        | 4448/24420 [1:50:38<8:14:09,  1.48s/it]
 18%|█▊        | 4449/24420 [1:50:40<8:13:16,  1.48s/it]
 18%|█▊        | 4450/24420 [1:50:41<8:13:23,  1.48s/it]
                                                        
{'loss': 0.1558, 'learning_rate': 5e-05, 'epoch': 0.91}

 18%|█▊        | 4450/24420 [1:50:41<8:13:23,  1.48s/it]
 18%|█▊        | 4451/24420 [1:50:43<8:13:13,  1.48s/it]
 18%|█▊        | 4452/24420 [1:50:44<8:12:17,  1.48s/it]
 18%|█▊        | 4453/24420 [1:50:46<8:12:40,  1.48s/it]
 18%|█▊        | 4454/24420 [1:50:47<8:13:28,  1.48s/it]
 18%|█▊        | 4455/24420 [1:50:49<8:13:12,  1.48s/it]
                                                        
{'loss': 0.1434, 'learning_rate': 5e-05, 'epoch': 0.91}

 18%|█▊        | 4455/24420 [1:50:49<8:13:12,  1.48s/it]
 18%|█▊        | 4456/24420 [1:50:50<8:13:07,  1.48s/it]
 18%|█▊        | 4457/24420 [1:50:52<8:12:19,  1.48s/it]
 18%|█▊        | 4458/24420 [1:50:53<8:11:59,  1.48s/it]
 18%|█▊        | 4459/24420 [1:50:55<8:12:24,  1.48s/it]
 18%|█▊        | 4460/24420 [1:50:56<8:12:29,  1.48s/it]
                                                        
{'loss': 0.1913, 'learning_rate': 5e-05, 'epoch': 0.91}

 18%|█▊        | 4460/24420 [1:50:56<8:12:29,  1.48s/it]
 18%|█▊        | 4461/24420 [1:50:58<8:11:42,  1.48s/it]
 18%|█▊        | 4462/24420 [1:50:59<8:11:24,  1.48s/it]
 18%|█▊        | 4463/24420 [1:51:01<8:11:29,  1.48s/it]
 18%|█▊        | 4464/24420 [1:51:02<8:12:03,  1.48s/it]
 18%|█▊        | 4465/24420 [1:51:04<8:12:37,  1.48s/it]
                                                        
{'loss': 0.155, 'learning_rate': 5e-05, 'epoch': 0.91}

 18%|█▊        | 4465/24420 [1:51:04<8:12:37,  1.48s/it]
 18%|█▊        | 4466/24420 [1:51:05<8:12:35,  1.48s/it]
 18%|█▊        | 4467/24420 [1:51:07<8:12:27,  1.48s/it]
 18%|█▊        | 4468/24420 [1:51:08<8:13:03,  1.48s/it]
 18%|█▊        | 4469/24420 [1:51:10<8:13:43,  1.48s/it]
 18%|█▊        | 4470/24420 [1:51:11<8:12:44,  1.48s/it]
                                                        
{'loss': 0.145, 'learning_rate': 5e-05, 'epoch': 0.92}

 18%|█▊        | 4470/24420 [1:51:11<8:12:44,  1.48s/it]
 18%|█▊        | 4471/24420 [1:51:12<8:11:55,  1.48s/it]
 18%|█▊        | 4472/24420 [1:51:14<8:11:10,  1.48s/it]
 18%|█▊        | 4473/24420 [1:51:15<8:11:18,  1.48s/it]
 18%|█▊        | 4474/24420 [1:51:17<8:11:09,  1.48s/it]
 18%|█▊        | 4475/24420 [1:51:18<8:11:25,  1.48s/it]
                                                        
{'loss': 0.1288, 'learning_rate': 5e-05, 'epoch': 0.92}

 18%|█▊        | 4475/24420 [1:51:18<8:11:25,  1.48s/it]
 18%|█▊        | 4476/24420 [1:51:20<8:12:10,  1.48s/it]
 18%|█▊        | 4477/24420 [1:51:21<8:11:46,  1.48s/it]
 18%|█▊        | 4478/24420 [1:51:23<8:10:48,  1.48s/it]
 18%|█▊        | 4479/24420 [1:51:24<8:10:50,  1.48s/it]
 18%|█▊        | 4480/24420 [1:51:26<8:11:15,  1.48s/it]
                                                        
{'loss': 0.1879, 'learning_rate': 5e-05, 'epoch': 0.92}

 18%|█▊        | 4480/24420 [1:51:26<8:11:15,  1.48s/it]
 18%|█▊        | 4481/24420 [1:51:27<8:11:53,  1.48s/it]
 18%|█▊        | 4482/24420 [1:51:29<8:11:14,  1.48s/it]
 18%|█▊        | 4483/24420 [1:51:30<8:11:02,  1.48s/it]
 18%|█▊        | 4484/24420 [1:51:32<8:11:17,  1.48s/it]
 18%|█▊        | 4485/24420 [1:51:33<8:11:44,  1.48s/it]
                                                        
{'loss': 0.1461, 'learning_rate': 5e-05, 'epoch': 0.92}

 18%|█▊        | 4485/24420 [1:51:33<8:11:44,  1.48s/it]
 18%|█▊        | 4486/24420 [1:51:35<8:11:58,  1.48s/it]
 18%|█▊        | 4487/24420 [1:51:36<8:12:00,  1.48s/it]
 18%|█▊        | 4488/24420 [1:51:38<8:11:23,  1.48s/it]
 18%|█▊        | 4489/24420 [1:51:39<8:10:48,  1.48s/it]
 18%|█▊        | 4490/24420 [1:51:41<8:12:19,  1.48s/it]
                                                        
{'loss': 0.1622, 'learning_rate': 5e-05, 'epoch': 0.92}

 18%|█▊        | 4490/24420 [1:51:41<8:12:19,  1.48s/it]
 18%|█▊        | 4491/24420 [1:51:42<8:12:50,  1.48s/it]
 18%|█▊        | 4492/24420 [1:51:44<8:12:24,  1.48s/it]
 18%|█▊        | 4493/24420 [1:51:45<8:11:37,  1.48s/it]
 18%|█▊        | 4494/24420 [1:51:46<8:11:10,  1.48s/it]
 18%|█▊        | 4495/24420 [1:51:48<8:10:56,  1.48s/it]
                                                        
{'loss': 0.1604, 'learning_rate': 5e-05, 'epoch': 0.92}

 18%|█▊        | 4495/24420 [1:51:48<8:10:56,  1.48s/it]
 18%|█▊        | 4496/24420 [1:51:49<8:10:23,  1.48s/it]
 18%|█▊        | 4497/24420 [1:51:51<8:11:19,  1.48s/it]
 18%|█▊        | 4498/24420 [1:51:52<8:11:31,  1.48s/it]
 18%|█▊        | 4499/24420 [1:51:54<8:10:37,  1.48s/it]
 18%|█▊        | 4500/24420 [1:51:55<8:10:46,  1.48s/it]
                                                        
{'loss': 0.1574, 'learning_rate': 5e-05, 'epoch': 0.92}

 18%|█▊        | 4500/24420 [1:51:55<8:10:46,  1.48s/it]
 18%|█▊        | 4501/24420 [1:51:57<8:11:43,  1.48s/it]
 18%|█▊        | 4502/24420 [1:51:58<8:11:17,  1.48s/it]
 18%|█▊        | 4503/24420 [1:52:00<8:11:36,  1.48s/it]
 18%|█▊        | 4504/24420 [1:52:01<8:11:28,  1.48s/it]
 18%|█▊        | 4505/24420 [1:52:03<8:11:12,  1.48s/it]
                                                        
{'loss': 0.1321, 'learning_rate': 5e-05, 'epoch': 0.92}

 18%|█▊        | 4505/24420 [1:52:03<8:11:12,  1.48s/it]
 18%|█▊        | 4506/24420 [1:52:04<8:11:25,  1.48s/it]
 18%|█▊        | 4507/24420 [1:52:06<8:11:07,  1.48s/it]
 18%|█▊        | 4508/24420 [1:52:07<8:11:07,  1.48s/it]
 18%|█▊        | 4509/24420 [1:52:09<8:11:12,  1.48s/it]
 18%|█▊        | 4510/24420 [1:52:10<8:10:20,  1.48s/it]
                                                        
{'loss': 0.1865, 'learning_rate': 5e-05, 'epoch': 0.92}

 18%|█▊        | 4510/24420 [1:52:10<8:10:20,  1.48s/it]
 18%|█▊        | 4511/24420 [1:52:12<8:10:30,  1.48s/it]
 18%|█▊        | 4512/24420 [1:52:13<8:10:22,  1.48s/it]
 18%|█▊        | 4513/24420 [1:52:15<8:10:21,  1.48s/it]
 18%|█▊        | 4514/24420 [1:52:16<8:10:05,  1.48s/it]
 18%|█▊        | 4515/24420 [1:52:18<8:10:29,  1.48s/it]
                                                        
{'loss': 0.1329, 'learning_rate': 5e-05, 'epoch': 0.92}

 18%|█▊        | 4515/24420 [1:52:18<8:10:29,  1.48s/it]
 18%|█▊        | 4516/24420 [1:52:19<8:09:50,  1.48s/it]
 18%|█▊        | 4517/24420 [1:52:20<8:09:19,  1.48s/it]
 19%|█▊        | 4518/24420 [1:52:22<8:15:27,  1.49s/it]
 19%|█▊        | 4519/24420 [1:52:24<8:13:34,  1.49s/it]
 19%|█▊        | 4520/24420 [1:52:25<8:12:15,  1.48s/it]
                                                        
{'loss': 0.1742, 'learning_rate': 5e-05, 'epoch': 0.93}

 19%|█▊        | 4520/24420 [1:52:25<8:12:15,  1.48s/it]
 19%|█▊        | 4521/24420 [1:52:26<8:10:55,  1.48s/it]
 19%|█▊        | 4522/24420 [1:52:28<8:11:15,  1.48s/it]
 19%|█▊        | 4523/24420 [1:52:29<8:10:54,  1.48s/it]
 19%|█▊        | 4524/24420 [1:52:31<8:10:32,  1.48s/it]
 19%|█▊        | 4525/24420 [1:52:32<8:10:01,  1.48s/it]
                                                        
{'loss': 0.1518, 'learning_rate': 5e-05, 'epoch': 0.93}

 19%|█▊        | 4525/24420 [1:52:32<8:10:01,  1.48s/it]
 19%|█▊        | 4526/24420 [1:52:34<8:10:20,  1.48s/it]
 19%|█▊        | 4527/24420 [1:52:35<8:09:57,  1.48s/it]
 19%|█▊        | 4528/24420 [1:52:37<8:10:07,  1.48s/it]
 19%|█▊        | 4529/24420 [1:52:38<8:10:39,  1.48s/it]
 19%|█▊        | 4530/24420 [1:52:40<8:11:05,  1.48s/it]
                                                        
{'loss': 0.18, 'learning_rate': 5e-05, 'epoch': 0.93}

 19%|█▊        | 4530/24420 [1:52:40<8:11:05,  1.48s/it]
 19%|█▊        | 4531/24420 [1:52:41<8:10:35,  1.48s/it]
 19%|█▊        | 4532/24420 [1:52:43<8:09:58,  1.48s/it]
 19%|█▊        | 4533/24420 [1:52:44<8:10:08,  1.48s/it]
 19%|█▊        | 4534/24420 [1:52:46<8:10:03,  1.48s/it]
 19%|█▊        | 4535/24420 [1:52:47<8:09:29,  1.48s/it]
                                                        
{'loss': 0.1661, 'learning_rate': 5e-05, 'epoch': 0.93}

 19%|█▊        | 4535/24420 [1:52:47<8:09:29,  1.48s/it]
 19%|█▊        | 4536/24420 [1:52:49<8:10:20,  1.48s/it]
 19%|█▊        | 4537/24420 [1:52:50<8:10:24,  1.48s/it]
 19%|█▊        | 4538/24420 [1:52:52<8:10:15,  1.48s/it]
 19%|█▊        | 4539/24420 [1:52:53<8:10:14,  1.48s/it]
 19%|█▊        | 4540/24420 [1:52:55<8:11:30,  1.48s/it]
                                                        
{'loss': 0.1559, 'learning_rate': 5e-05, 'epoch': 0.93}

 19%|█▊        | 4540/24420 [1:52:55<8:11:30,  1.48s/it]
 19%|█▊        | 4541/24420 [1:52:56<8:11:21,  1.48s/it]
 19%|█▊        | 4542/24420 [1:52:58<8:10:05,  1.48s/it]
 19%|█▊        | 4543/24420 [1:52:59<8:09:27,  1.48s/it]
 19%|█▊        | 4544/24420 [1:53:00<8:09:34,  1.48s/it]
 19%|█▊        | 4545/24420 [1:53:02<8:09:33,  1.48s/it]
                                                        
{'loss': 0.188, 'learning_rate': 5e-05, 'epoch': 0.93}

 19%|█▊        | 4545/24420 [1:53:02<8:09:33,  1.48s/it]
 19%|█▊        | 4546/24420 [1:53:03<8:09:26,  1.48s/it]
 19%|█▊        | 4547/24420 [1:53:05<8:09:38,  1.48s/it]
 19%|█▊        | 4548/24420 [1:53:06<8:10:31,  1.48s/it]
 19%|█▊        | 4549/24420 [1:53:08<8:10:07,  1.48s/it]
 19%|█▊        | 4550/24420 [1:53:09<8:10:20,  1.48s/it]
                                                        
{'loss': 0.1841, 'learning_rate': 5e-05, 'epoch': 0.93}

 19%|█▊        | 4550/24420 [1:53:09<8:10:20,  1.48s/it]
 19%|█▊        | 4551/24420 [1:53:11<8:10:10,  1.48s/it]
 19%|█▊        | 4552/24420 [1:53:12<8:09:32,  1.48s/it]
 19%|█▊        | 4553/24420 [1:53:14<8:09:12,  1.48s/it]
 19%|█▊        | 4554/24420 [1:53:15<8:08:43,  1.48s/it]
 19%|█▊        | 4555/24420 [1:53:17<8:08:23,  1.48s/it]
                                                        
{'loss': 0.1448, 'learning_rate': 5e-05, 'epoch': 0.93}

 19%|█▊        | 4555/24420 [1:53:17<8:08:23,  1.48s/it]
 19%|█▊        | 4556/24420 [1:53:18<8:08:49,  1.48s/it]
 19%|█▊        | 4557/24420 [1:53:20<8:08:38,  1.48s/it]
 19%|█▊        | 4558/24420 [1:53:21<8:09:23,  1.48s/it]
 19%|█▊        | 4559/24420 [1:53:23<8:09:40,  1.48s/it]
 19%|█▊        | 4560/24420 [1:53:24<8:10:05,  1.48s/it]
                                                        
{'loss': 0.1752, 'learning_rate': 5e-05, 'epoch': 0.93}

 19%|█▊        | 4560/24420 [1:53:24<8:10:05,  1.48s/it]
 19%|█▊        | 4561/24420 [1:53:26<8:11:00,  1.48s/it]
 19%|█▊        | 4562/24420 [1:53:27<8:09:46,  1.48s/it]
 19%|█▊        | 4563/24420 [1:53:29<8:10:24,  1.48s/it]
 19%|█▊        | 4564/24420 [1:53:30<8:09:21,  1.48s/it]
 19%|█▊        | 4565/24420 [1:53:32<8:08:47,  1.48s/it]
                                                        
{'loss': 0.1631, 'learning_rate': 5e-05, 'epoch': 0.93}

 19%|█▊        | 4565/24420 [1:53:32<8:08:47,  1.48s/it]
 19%|█▊        | 4566/24420 [1:53:33<8:08:43,  1.48s/it]
 19%|█▊        | 4567/24420 [1:53:34<8:08:17,  1.48s/it]
 19%|█▊        | 4568/24420 [1:53:36<8:07:57,  1.47s/it]
 19%|█▊        | 4569/24420 [1:53:37<8:09:18,  1.48s/it]
 19%|█▊        | 4570/24420 [1:53:39<8:08:40,  1.48s/it]
                                                        
{'loss': 0.1358, 'learning_rate': 5e-05, 'epoch': 0.94}

 19%|█▊        | 4570/24420 [1:53:39<8:08:40,  1.48s/it]
 19%|█▊        | 4571/24420 [1:53:40<8:08:02,  1.48s/it]
 19%|█▊        | 4572/24420 [1:53:42<8:07:44,  1.47s/it]
 19%|█▊        | 4573/24420 [1:53:43<8:08:36,  1.48s/it]
 19%|█▊        | 4574/24420 [1:53:45<8:08:58,  1.48s/it]
 19%|█▊        | 4575/24420 [1:53:46<8:09:38,  1.48s/it]
                                                        
{'loss': 0.159, 'learning_rate': 5e-05, 'epoch': 0.94}

 19%|█▊        | 4575/24420 [1:53:46<8:09:38,  1.48s/it]
 19%|█▊        | 4576/24420 [1:53:48<8:09:58,  1.48s/it]
 19%|█▊        | 4577/24420 [1:53:49<8:09:57,  1.48s/it]
 19%|█▊        | 4578/24420 [1:53:51<8:09:17,  1.48s/it]
 19%|█▉        | 4579/24420 [1:53:52<8:09:34,  1.48s/it]
 19%|█▉        | 4580/24420 [1:53:54<8:09:46,  1.48s/it]
                                                        
{'loss': 0.164, 'learning_rate': 5e-05, 'epoch': 0.94}

 19%|█▉        | 4580/24420 [1:53:54<8:09:46,  1.48s/it]
 19%|█▉        | 4581/24420 [1:53:55<8:09:24,  1.48s/it]
 19%|█▉        | 4582/24420 [1:53:57<8:08:57,  1.48s/it]
 19%|█▉        | 4583/24420 [1:53:58<8:09:14,  1.48s/it]
 19%|█▉        | 4584/24420 [1:54:00<8:09:57,  1.48s/it]
 19%|█▉        | 4585/24420 [1:54:01<8:10:55,  1.49s/it]
                                                        
{'loss': 0.1516, 'learning_rate': 5e-05, 'epoch': 0.94}

 19%|█▉        | 4585/24420 [1:54:01<8:10:55,  1.49s/it]
 19%|█▉        | 4586/24420 [1:54:03<8:10:35,  1.48s/it]
 19%|█▉        | 4587/24420 [1:54:05<9:31:40,  1.73s/it]
 19%|█▉        | 4588/24420 [1:54:06<9:06:23,  1.65s/it]
 19%|█▉        | 4589/24420 [1:54:08<8:49:37,  1.60s/it]
 19%|█▉        | 4590/24420 [1:54:09<8:36:40,  1.56s/it]
                                                        
{'loss': 0.1449, 'learning_rate': 5e-05, 'epoch': 0.94}

 19%|█▉        | 4590/24420 [1:54:09<8:36:40,  1.56s/it]
 19%|█▉        | 4591/24420 [1:54:11<8:28:42,  1.54s/it]
 19%|█▉        | 4592/24420 [1:54:12<8:21:58,  1.52s/it]
 19%|█▉        | 4593/24420 [1:54:14<8:17:30,  1.51s/it]
 19%|█▉        | 4594/24420 [1:54:15<8:15:13,  1.50s/it]
 19%|█▉        | 4595/24420 [1:54:17<8:13:12,  1.49s/it]
                                                        
{'loss': 0.1428, 'learning_rate': 5e-05, 'epoch': 0.94}

 19%|█▉        | 4595/24420 [1:54:17<8:13:12,  1.49s/it]
 19%|█▉        | 4596/24420 [1:54:18<8:11:57,  1.49s/it]
 19%|█▉        | 4597/24420 [1:54:20<8:10:32,  1.48s/it]
 19%|█▉        | 4598/24420 [1:54:21<8:10:12,  1.48s/it]
 19%|█▉        | 4599/24420 [1:54:23<8:10:31,  1.48s/it]
 19%|█▉        | 4600/24420 [1:54:24<8:09:56,  1.48s/it]
                                                        
{'loss': 0.1588, 'learning_rate': 5e-05, 'epoch': 0.94}

 19%|█▉        | 4600/24420 [1:54:24<8:09:56,  1.48s/it]
 19%|█▉        | 4601/24420 [1:54:29<13:11:41,  2.40s/it]
 19%|█▉        | 4602/24420 [1:54:30<11:40:16,  2.12s/it]
 19%|█▉        | 4603/24420 [1:54:32<10:36:32,  1.93s/it]
 19%|█▉        | 4604/24420 [1:54:33<9:52:17,  1.79s/it] 
 19%|█▉        | 4605/24420 [1:54:35<9:20:34,  1.70s/it]
                                                        
{'loss': 0.1519, 'learning_rate': 5e-05, 'epoch': 0.94}

 19%|█▉        | 4605/24420 [1:54:35<9:20:34,  1.70s/it]
 19%|█▉        | 4606/24420 [1:54:36<8:59:03,  1.63s/it]
 19%|█▉        | 4607/24420 [1:54:38<8:43:21,  1.58s/it]
 19%|█▉        | 4608/24420 [1:54:39<8:32:27,  1.55s/it]
 19%|█▉        | 4609/24420 [1:54:40<8:25:29,  1.53s/it]
 19%|█▉        | 4610/24420 [1:54:42<8:20:19,  1.52s/it]
                                                        
{'loss': 0.1363, 'learning_rate': 5e-05, 'epoch': 0.94}

 19%|█▉        | 4610/24420 [1:54:42<8:20:19,  1.52s/it]
 19%|█▉        | 4611/24420 [1:54:43<8:17:17,  1.51s/it]
 19%|█▉        | 4612/24420 [1:54:45<8:14:49,  1.50s/it]
 19%|█▉        | 4613/24420 [1:54:46<8:12:58,  1.49s/it]
 19%|█▉        | 4614/24420 [1:54:48<8:11:27,  1.49s/it]
 19%|█▉        | 4615/24420 [1:54:49<8:10:10,  1.48s/it]
                                                        
{'loss': 0.1737, 'learning_rate': 5e-05, 'epoch': 0.94}

 19%|█▉        | 4615/24420 [1:54:49<8:10:10,  1.48s/it]
 19%|█▉        | 4616/24420 [1:54:51<8:10:09,  1.49s/it]
 19%|█▉        | 4617/24420 [1:54:52<8:09:14,  1.48s/it]
 19%|█▉        | 4618/24420 [1:54:54<8:07:58,  1.48s/it]
 19%|█▉        | 4619/24420 [1:54:55<8:08:08,  1.48s/it]
 19%|█▉        | 4620/24420 [1:54:57<8:07:31,  1.48s/it]
                                                        
{'loss': 0.1529, 'learning_rate': 5e-05, 'epoch': 0.95}

 19%|█▉        | 4620/24420 [1:54:57<8:07:31,  1.48s/it]
 19%|█▉        | 4621/24420 [1:54:58<8:08:20,  1.48s/it]
 19%|█▉        | 4622/24420 [1:55:00<8:09:32,  1.48s/it]
 19%|█▉        | 4623/24420 [1:55:01<8:09:21,  1.48s/it]
 19%|█▉        | 4624/24420 [1:55:03<8:09:24,  1.48s/it]
 19%|█▉        | 4625/24420 [1:55:04<8:08:54,  1.48s/it]
                                                        
{'loss': 0.1549, 'learning_rate': 5e-05, 'epoch': 0.95}

 19%|█▉        | 4625/24420 [1:55:04<8:08:54,  1.48s/it]
 19%|█▉        | 4626/24420 [1:55:06<8:08:21,  1.48s/it]
 19%|█▉        | 4627/24420 [1:55:07<8:09:22,  1.48s/it]
 19%|█▉        | 4628/24420 [1:55:09<8:08:44,  1.48s/it]
 19%|█▉        | 4629/24420 [1:55:10<8:08:21,  1.48s/it]
 19%|█▉        | 4630/24420 [1:55:12<8:08:59,  1.48s/it]
                                                        
{'loss': 0.1769, 'learning_rate': 5e-05, 'epoch': 0.95}

 19%|█▉        | 4630/24420 [1:55:12<8:08:59,  1.48s/it]
 19%|█▉        | 4631/24420 [1:55:13<8:08:22,  1.48s/it]
 19%|█▉        | 4632/24420 [1:55:15<8:07:18,  1.48s/it]
 19%|█▉        | 4633/24420 [1:55:16<8:08:17,  1.48s/it]
 19%|█▉        | 4634/24420 [1:55:17<8:07:31,  1.48s/it]
 19%|█▉        | 4635/24420 [1:55:19<8:07:33,  1.48s/it]
                                                        
{'loss': 0.1508, 'learning_rate': 5e-05, 'epoch': 0.95}

 19%|█▉        | 4635/24420 [1:55:19<8:07:33,  1.48s/it]
 19%|█▉        | 4636/24420 [1:55:20<8:07:30,  1.48s/it]
 19%|█▉        | 4637/24420 [1:55:22<8:07:06,  1.48s/it]
 19%|█▉        | 4638/24420 [1:55:23<8:07:07,  1.48s/it]
 19%|█▉        | 4639/24420 [1:55:25<8:06:49,  1.48s/it]
 19%|█▉        | 4640/24420 [1:55:26<8:07:15,  1.48s/it]
                                                        
{'loss': 0.1971, 'learning_rate': 5e-05, 'epoch': 0.95}

 19%|█▉        | 4640/24420 [1:55:26<8:07:15,  1.48s/it]
 19%|█▉        | 4641/24420 [1:55:28<8:08:07,  1.48s/it]
 19%|█▉        | 4642/24420 [1:55:29<8:07:36,  1.48s/it]
 19%|█▉        | 4643/24420 [1:55:31<8:07:38,  1.48s/it]
 19%|█▉        | 4644/24420 [1:55:32<8:08:12,  1.48s/it]
 19%|█▉        | 4645/24420 [1:55:34<8:08:38,  1.48s/it]
                                                        
{'loss': 0.2165, 'learning_rate': 5e-05, 'epoch': 0.95}

 19%|█▉        | 4645/24420 [1:55:34<8:08:38,  1.48s/it]
 19%|█▉        | 4646/24420 [1:55:35<8:07:45,  1.48s/it]
 19%|█▉        | 4647/24420 [1:55:37<8:07:28,  1.48s/it]
 19%|█▉        | 4648/24420 [1:55:38<8:07:56,  1.48s/it]
 19%|█▉        | 4649/24420 [1:55:40<8:07:05,  1.48s/it]
 19%|█▉        | 4650/24420 [1:55:41<8:06:45,  1.48s/it]
                                                        
{'loss': 0.1646, 'learning_rate': 5e-05, 'epoch': 0.95}

 19%|█▉        | 4650/24420 [1:55:41<8:06:45,  1.48s/it]
 19%|█▉        | 4651/24420 [1:55:43<8:07:58,  1.48s/it]
 19%|█▉        | 4652/24420 [1:55:44<8:07:49,  1.48s/it]
 19%|█▉        | 4653/24420 [1:55:46<8:07:16,  1.48s/it]
 19%|█▉        | 4654/24420 [1:55:47<8:06:57,  1.48s/it]
 19%|█▉        | 4655/24420 [1:55:49<8:06:32,  1.48s/it]
                                                        
{'loss': 0.1258, 'learning_rate': 5e-05, 'epoch': 0.95}

 19%|█▉        | 4655/24420 [1:55:49<8:06:32,  1.48s/it]
 19%|█▉        | 4656/24420 [1:55:50<8:07:05,  1.48s/it]
 19%|█▉        | 4657/24420 [1:55:52<8:07:05,  1.48s/it]
 19%|█▉        | 4658/24420 [1:55:53<8:07:37,  1.48s/it]
 19%|█▉        | 4659/24420 [1:55:54<8:06:37,  1.48s/it]
 19%|█▉        | 4660/24420 [1:55:56<8:06:14,  1.48s/it]
                                                        
{'loss': 0.1516, 'learning_rate': 5e-05, 'epoch': 0.95}

 19%|█▉        | 4660/24420 [1:55:56<8:06:14,  1.48s/it]
 19%|█▉        | 4661/24420 [1:55:57<8:06:33,  1.48s/it]
 19%|█▉        | 4662/24420 [1:55:59<8:06:42,  1.48s/it]
 19%|█▉        | 4663/24420 [1:56:00<8:07:51,  1.48s/it]
 19%|█▉        | 4664/24420 [1:56:02<8:07:46,  1.48s/it]
 19%|█▉        | 4665/24420 [1:56:03<8:07:32,  1.48s/it]
                                                        
{'loss': 0.1545, 'learning_rate': 5e-05, 'epoch': 0.96}

 19%|█▉        | 4665/24420 [1:56:03<8:07:32,  1.48s/it]
 19%|█▉        | 4666/24420 [1:56:05<8:07:10,  1.48s/it]
 19%|█▉        | 4667/24420 [1:56:06<8:06:58,  1.48s/it]
 19%|█▉        | 4668/24420 [1:56:08<8:06:39,  1.48s/it]
 19%|█▉        | 4669/24420 [1:56:09<8:06:50,  1.48s/it]
 19%|█▉        | 4670/24420 [1:56:11<8:06:40,  1.48s/it]
                                                        
{'loss': 0.1618, 'learning_rate': 5e-05, 'epoch': 0.96}

 19%|█▉        | 4670/24420 [1:56:11<8:06:40,  1.48s/it]
 19%|█▉        | 4671/24420 [1:56:12<8:06:53,  1.48s/it]
 19%|█▉        | 4672/24420 [1:56:14<8:07:26,  1.48s/it]
 19%|█▉        | 4673/24420 [1:56:15<8:07:11,  1.48s/it]
 19%|█▉        | 4674/24420 [1:56:17<8:07:34,  1.48s/it]
 19%|█▉        | 4675/24420 [1:56:18<8:07:18,  1.48s/it]
                                                        
{'loss': 0.1394, 'learning_rate': 5e-05, 'epoch': 0.96}

 19%|█▉        | 4675/24420 [1:56:18<8:07:18,  1.48s/it]
 19%|█▉        | 4676/24420 [1:56:20<8:08:04,  1.48s/it]
 19%|█▉        | 4677/24420 [1:56:21<8:08:03,  1.48s/it]
 19%|█▉        | 4678/24420 [1:56:23<8:08:04,  1.48s/it]
 19%|█▉        | 4679/24420 [1:56:24<8:08:10,  1.48s/it]
 19%|█▉        | 4680/24420 [1:56:26<8:08:18,  1.48s/it]
                                                        
{'loss': 0.1821, 'learning_rate': 5e-05, 'epoch': 0.96}

 19%|█▉        | 4680/24420 [1:56:26<8:08:18,  1.48s/it]
 19%|█▉        | 4681/24420 [1:56:27<8:07:29,  1.48s/it]
 19%|█▉        | 4682/24420 [1:56:29<8:06:28,  1.48s/it]
 19%|█▉        | 4683/24420 [1:56:30<8:06:43,  1.48s/it]
 19%|█▉        | 4684/24420 [1:56:31<8:07:21,  1.48s/it]
 19%|█▉        | 4685/24420 [1:56:33<8:07:11,  1.48s/it]
                                                        
{'loss': 0.1325, 'learning_rate': 5e-05, 'epoch': 0.96}

 19%|█▉        | 4685/24420 [1:56:33<8:07:11,  1.48s/it]
 19%|█▉        | 4686/24420 [1:56:34<8:07:05,  1.48s/it]
 19%|█▉        | 4687/24420 [1:56:36<8:06:51,  1.48s/it]
 19%|█▉        | 4688/24420 [1:56:37<8:07:21,  1.48s/it]
 19%|█▉        | 4689/24420 [1:56:39<8:08:21,  1.49s/it]
 19%|█▉        | 4690/24420 [1:56:40<8:08:01,  1.48s/it]
                                                        
{'loss': 0.1728, 'learning_rate': 5e-05, 'epoch': 0.96}

 19%|█▉        | 4690/24420 [1:56:40<8:08:01,  1.48s/it]
 19%|█▉        | 4691/24420 [1:56:42<8:08:09,  1.48s/it]
 19%|█▉        | 4692/24420 [1:56:43<8:08:42,  1.49s/it]
 19%|█▉        | 4693/24420 [1:56:45<8:08:18,  1.49s/it]
 19%|█▉        | 4694/24420 [1:56:46<8:07:12,  1.48s/it]
 19%|█▉        | 4695/24420 [1:56:48<8:07:22,  1.48s/it]
                                                        
{'loss': 0.1421, 'learning_rate': 5e-05, 'epoch': 0.96}

 19%|█▉        | 4695/24420 [1:56:48<8:07:22,  1.48s/it]
 19%|█▉        | 4696/24420 [1:56:49<8:08:41,  1.49s/it]
 19%|█▉        | 4697/24420 [1:56:51<8:08:17,  1.49s/it]
 19%|█▉        | 4698/24420 [1:56:52<8:09:23,  1.49s/it]
 19%|█▉        | 4699/24420 [1:56:54<8:08:13,  1.49s/it]
 19%|█▉        | 4700/24420 [1:56:55<8:08:32,  1.49s/it]
                                                        
{'loss': 0.181, 'learning_rate': 5e-05, 'epoch': 0.96}

 19%|█▉        | 4700/24420 [1:56:55<8:08:32,  1.49s/it]
 19%|█▉        | 4701/24420 [1:56:57<8:07:45,  1.48s/it]
 19%|█▉        | 4702/24420 [1:56:58<8:08:04,  1.49s/it]
 19%|█▉        | 4703/24420 [1:57:00<8:07:18,  1.48s/it]
 19%|█▉        | 4704/24420 [1:57:01<8:07:12,  1.48s/it]
 19%|█▉        | 4705/24420 [1:57:03<8:09:37,  1.49s/it]
                                                        
{'loss': 0.1586, 'learning_rate': 5e-05, 'epoch': 0.96}

 19%|█▉        | 4705/24420 [1:57:03<8:09:37,  1.49s/it]
 19%|█▉        | 4706/24420 [1:57:04<8:10:13,  1.49s/it]
 19%|█▉        | 4707/24420 [1:57:06<8:09:50,  1.49s/it]
 19%|█▉        | 4708/24420 [1:57:07<8:08:13,  1.49s/it]
 19%|█▉        | 4709/24420 [1:57:09<8:07:00,  1.48s/it]
 19%|█▉        | 4710/24420 [1:57:10<8:08:11,  1.49s/it]
                                                        
{'loss': 0.1419, 'learning_rate': 5e-05, 'epoch': 0.96}

 19%|█▉        | 4710/24420 [1:57:10<8:08:11,  1.49s/it]
 19%|█▉        | 4711/24420 [1:57:12<8:07:46,  1.48s/it]
 19%|█▉        | 4712/24420 [1:57:13<8:07:40,  1.48s/it]
 19%|█▉        | 4713/24420 [1:57:15<8:06:52,  1.48s/it]
 19%|█▉        | 4714/24420 [1:57:16<8:06:57,  1.48s/it]
 19%|█▉        | 4715/24420 [1:57:18<8:06:42,  1.48s/it]
                                                        
{'loss': 0.1609, 'learning_rate': 5e-05, 'epoch': 0.97}

 19%|█▉        | 4715/24420 [1:57:18<8:06:42,  1.48s/it]
 19%|█▉        | 4716/24420 [1:57:19<8:05:55,  1.48s/it]
 19%|█▉        | 4717/24420 [1:57:20<8:05:53,  1.48s/it]
 19%|█▉        | 4718/24420 [1:57:22<8:04:57,  1.48s/it]
 19%|█▉        | 4719/24420 [1:57:23<8:05:46,  1.48s/it]
 19%|█▉        | 4720/24420 [1:57:25<8:05:33,  1.48s/it]
                                                        
{'loss': 0.1347, 'learning_rate': 5e-05, 'epoch': 0.97}

 19%|█▉        | 4720/24420 [1:57:25<8:05:33,  1.48s/it]
 19%|█▉        | 4721/24420 [1:57:26<8:05:03,  1.48s/it]
 19%|█▉        | 4722/24420 [1:57:28<8:05:07,  1.48s/it]
 19%|█▉        | 4723/24420 [1:57:29<8:05:23,  1.48s/it]
 19%|█▉        | 4724/24420 [1:57:31<8:05:37,  1.48s/it]
 19%|█▉        | 4725/24420 [1:57:32<8:06:24,  1.48s/it]
                                                        
{'loss': 0.135, 'learning_rate': 5e-05, 'epoch': 0.97}

 19%|█▉        | 4725/24420 [1:57:32<8:06:24,  1.48s/it]
 19%|█▉        | 4726/24420 [1:57:34<8:06:43,  1.48s/it]
 19%|█▉        | 4727/24420 [1:57:35<8:07:12,  1.48s/it]
 19%|█▉        | 4728/24420 [1:57:37<8:08:01,  1.49s/it]
 19%|█▉        | 4729/24420 [1:57:38<8:07:09,  1.48s/it]
 19%|█▉        | 4730/24420 [1:57:40<8:07:12,  1.48s/it]
                                                        
{'loss': 0.1423, 'learning_rate': 5e-05, 'epoch': 0.97}

 19%|█▉        | 4730/24420 [1:57:40<8:07:12,  1.48s/it]
 19%|█▉        | 4731/24420 [1:57:41<8:07:16,  1.48s/it]
 19%|█▉        | 4732/24420 [1:57:43<8:07:20,  1.49s/it]
 19%|█▉        | 4733/24420 [1:57:44<8:06:40,  1.48s/it]
 19%|█▉        | 4734/24420 [1:57:46<8:08:08,  1.49s/it]
 19%|█▉        | 4735/24420 [1:57:47<8:07:05,  1.48s/it]
                                                        
{'loss': 0.1613, 'learning_rate': 5e-05, 'epoch': 0.97}

 19%|█▉        | 4735/24420 [1:57:47<8:07:05,  1.48s/it]
 19%|█▉        | 4736/24420 [1:57:49<8:06:28,  1.48s/it]
 19%|█▉        | 4737/24420 [1:57:50<8:05:39,  1.48s/it]
 19%|█▉        | 4738/24420 [1:57:52<8:06:21,  1.48s/it]
 19%|█▉        | 4739/24420 [1:57:53<8:05:55,  1.48s/it]
 19%|█▉        | 4740/24420 [1:57:55<8:05:49,  1.48s/it]
                                                        
{'loss': 0.1567, 'learning_rate': 5e-05, 'epoch': 0.97}

 19%|█▉        | 4740/24420 [1:57:55<8:05:49,  1.48s/it]
 19%|█▉        | 4741/24420 [1:57:56<8:05:39,  1.48s/it]
 19%|█▉        | 4742/24420 [1:57:58<8:05:22,  1.48s/it]
 19%|█▉        | 4743/24420 [1:57:59<8:05:25,  1.48s/it]
 19%|█▉        | 4744/24420 [1:58:00<8:05:59,  1.48s/it]
 19%|█▉        | 4745/24420 [1:58:02<8:05:38,  1.48s/it]
                                                        
{'loss': 0.1419, 'learning_rate': 5e-05, 'epoch': 0.97}

 19%|█▉        | 4745/24420 [1:58:02<8:05:38,  1.48s/it]
 19%|█▉        | 4746/24420 [1:58:03<8:05:57,  1.48s/it]
 19%|█▉        | 4747/24420 [1:58:05<8:05:18,  1.48s/it]
 19%|█▉        | 4748/24420 [1:58:06<8:05:47,  1.48s/it]
 19%|█▉        | 4749/24420 [1:58:08<8:05:51,  1.48s/it]
 19%|█▉        | 4750/24420 [1:58:09<8:05:49,  1.48s/it]
                                                        
{'loss': 0.1526, 'learning_rate': 5e-05, 'epoch': 0.97}

 19%|█▉        | 4750/24420 [1:58:09<8:05:49,  1.48s/it]
 19%|█▉        | 4751/24420 [1:58:11<8:05:58,  1.48s/it]
 19%|█▉        | 4752/24420 [1:58:12<8:06:29,  1.48s/it]
 19%|█▉        | 4753/24420 [1:58:14<8:06:08,  1.48s/it]
 19%|█▉        | 4754/24420 [1:58:15<8:06:29,  1.48s/it]
 19%|█▉        | 4755/24420 [1:58:17<8:07:04,  1.49s/it]
                                                        
{'loss': 0.1518, 'learning_rate': 5e-05, 'epoch': 0.97}

 19%|█▉        | 4755/24420 [1:58:17<8:07:04,  1.49s/it]
 19%|█▉        | 4756/24420 [1:58:18<8:07:07,  1.49s/it]
 19%|█▉        | 4757/24420 [1:58:20<8:06:45,  1.49s/it]
 19%|█▉        | 4758/24420 [1:58:21<8:07:15,  1.49s/it]
 19%|█▉        | 4759/24420 [1:58:23<8:08:03,  1.49s/it]
 19%|█▉        | 4760/24420 [1:58:24<8:07:58,  1.49s/it]
                                                        
{'loss': 0.1328, 'learning_rate': 5e-05, 'epoch': 0.97}

 19%|█▉        | 4760/24420 [1:58:24<8:07:58,  1.49s/it]
 19%|█▉        | 4761/24420 [1:58:26<8:08:18,  1.49s/it]
 20%|█▉        | 4762/24420 [1:58:27<8:08:36,  1.49s/it]
 20%|█▉        | 4763/24420 [1:58:29<8:07:56,  1.49s/it]
 20%|█▉        | 4764/24420 [1:58:30<8:07:35,  1.49s/it]
 20%|█▉        | 4765/24420 [1:58:32<8:07:05,  1.49s/it]
                                                        
{'loss': 0.1449, 'learning_rate': 5e-05, 'epoch': 0.98}

 20%|█▉        | 4765/24420 [1:58:32<8:07:05,  1.49s/it]
 20%|█▉        | 4766/24420 [1:58:33<8:07:59,  1.49s/it]
 20%|█▉        | 4767/24420 [1:58:35<8:07:07,  1.49s/it]
 20%|█▉        | 4768/24420 [1:58:36<8:06:58,  1.49s/it]
 20%|█▉        | 4769/24420 [1:58:38<8:06:12,  1.48s/it]
 20%|█▉        | 4770/24420 [1:58:39<8:06:23,  1.49s/it]
                                                        
{'loss': 0.1412, 'learning_rate': 5e-05, 'epoch': 0.98}

 20%|█▉        | 4770/24420 [1:58:39<8:06:23,  1.49s/it]
 20%|█▉        | 4771/24420 [1:58:41<8:06:00,  1.48s/it]
 20%|█▉        | 4772/24420 [1:58:42<8:05:33,  1.48s/it]
 20%|█▉        | 4773/24420 [1:58:44<8:05:18,  1.48s/it]
 20%|█▉        | 4774/24420 [1:58:45<8:04:57,  1.48s/it]
 20%|█▉        | 4775/24420 [1:58:47<8:03:51,  1.48s/it]
                                                        
{'loss': 0.1179, 'learning_rate': 5e-05, 'epoch': 0.98}

 20%|█▉        | 4775/24420 [1:58:47<8:03:51,  1.48s/it]
 20%|█▉        | 4776/24420 [1:58:48<8:04:04,  1.48s/it]
 20%|█▉        | 4777/24420 [1:58:49<8:04:17,  1.48s/it]
 20%|█▉        | 4778/24420 [1:58:51<8:04:48,  1.48s/it]
 20%|█▉        | 4779/24420 [1:58:52<8:04:20,  1.48s/it]
 20%|█▉        | 4780/24420 [1:58:54<8:04:11,  1.48s/it]
                                                        
{'loss': 0.1574, 'learning_rate': 5e-05, 'epoch': 0.98}

 20%|█▉        | 4780/24420 [1:58:54<8:04:11,  1.48s/it]
 20%|█▉        | 4781/24420 [1:58:55<8:04:48,  1.48s/it]
 20%|█▉        | 4782/24420 [1:58:57<8:04:22,  1.48s/it]
 20%|█▉        | 4783/24420 [1:58:58<8:04:05,  1.48s/it]
 20%|█▉        | 4784/24420 [1:59:00<8:05:22,  1.48s/it]
 20%|█▉        | 4785/24420 [1:59:01<8:05:31,  1.48s/it]
                                                        
{'loss': 0.1593, 'learning_rate': 5e-05, 'epoch': 0.98}

 20%|█▉        | 4785/24420 [1:59:01<8:05:31,  1.48s/it]
 20%|█▉        | 4786/24420 [1:59:03<8:05:33,  1.48s/it]
 20%|█▉        | 4787/24420 [1:59:04<8:04:15,  1.48s/it]
 20%|█▉        | 4788/24420 [1:59:06<8:04:46,  1.48s/it]
 20%|█▉        | 4789/24420 [1:59:07<8:04:37,  1.48s/it]
 20%|█▉        | 4790/24420 [1:59:09<8:05:02,  1.48s/it]
                                                        
{'loss': 0.1813, 'learning_rate': 5e-05, 'epoch': 0.98}

 20%|█▉        | 4790/24420 [1:59:09<8:05:02,  1.48s/it]
 20%|█▉        | 4791/24420 [1:59:10<8:04:45,  1.48s/it]
 20%|█▉        | 4792/24420 [1:59:12<8:04:32,  1.48s/it]
 20%|█▉        | 4793/24420 [1:59:13<8:03:59,  1.48s/it]
 20%|█▉        | 4794/24420 [1:59:15<8:03:36,  1.48s/it]
 20%|█▉        | 4795/24420 [1:59:16<8:04:19,  1.48s/it]
                                                        
{'loss': 0.1521, 'learning_rate': 5e-05, 'epoch': 0.98}

 20%|█▉        | 4795/24420 [1:59:16<8:04:19,  1.48s/it]
 20%|█▉        | 4796/24420 [1:59:18<8:03:19,  1.48s/it]
 20%|█▉        | 4797/24420 [1:59:19<8:03:33,  1.48s/it]
 20%|█▉        | 4798/24420 [1:59:21<8:03:34,  1.48s/it]
 20%|█▉        | 4799/24420 [1:59:22<8:03:51,  1.48s/it]
 20%|█▉        | 4800/24420 [1:59:24<8:05:06,  1.48s/it]
                                                        
{'loss': 0.1584, 'learning_rate': 5e-05, 'epoch': 0.98}

 20%|█▉        | 4800/24420 [1:59:24<8:05:06,  1.48s/it]
 20%|█▉        | 4801/24420 [1:59:30<15:38:26,  2.87s/it]
 20%|█▉        | 4802/24420 [1:59:31<13:22:16,  2.45s/it]
 20%|█▉        | 4803/24420 [1:59:33<11:46:52,  2.16s/it]
 20%|█▉        | 4804/24420 [1:59:34<10:39:34,  1.96s/it]
 20%|█▉        | 4805/24420 [1:59:36<9:52:56,  1.81s/it] 
                                                        
{'loss': 0.1555, 'learning_rate': 5e-05, 'epoch': 0.98}

 20%|█▉        | 4805/24420 [1:59:36<9:52:56,  1.81s/it]
 20%|█▉        | 4806/24420 [1:59:37<9:21:05,  1.72s/it]
 20%|█▉        | 4807/24420 [1:59:39<8:57:39,  1.64s/it]
 20%|█▉        | 4808/24420 [1:59:40<8:41:41,  1.60s/it]
 20%|█▉        | 4809/24420 [1:59:41<8:30:28,  1.56s/it]
 20%|█▉        | 4810/24420 [1:59:43<8:21:50,  1.54s/it]
                                                        
{'loss': 0.167, 'learning_rate': 5e-05, 'epoch': 0.98}

 20%|█▉        | 4810/24420 [1:59:43<8:21:50,  1.54s/it]
 20%|█▉        | 4811/24420 [1:59:44<8:15:18,  1.52s/it]
 20%|█▉        | 4812/24420 [1:59:46<8:11:50,  1.51s/it]
 20%|█▉        | 4813/24420 [1:59:47<8:08:49,  1.50s/it]
 20%|█▉        | 4814/24420 [1:59:49<8:06:52,  1.49s/it]
 20%|█▉        | 4815/24420 [1:59:50<8:04:40,  1.48s/it]
                                                        
{'loss': 0.1485, 'learning_rate': 5e-05, 'epoch': 0.99}

 20%|█▉        | 4815/24420 [1:59:50<8:04:40,  1.48s/it]
 20%|█▉        | 4816/24420 [1:59:52<8:03:49,  1.48s/it]
 20%|█▉        | 4817/24420 [1:59:53<8:04:02,  1.48s/it]
 20%|█▉        | 4818/24420 [1:59:55<8:03:53,  1.48s/it]
 20%|█▉        | 4819/24420 [1:59:56<8:03:37,  1.48s/it]
 20%|█▉        | 4820/24420 [1:59:58<8:03:10,  1.48s/it]
                                                        
{'loss': 0.1367, 'learning_rate': 5e-05, 'epoch': 0.99}

 20%|█▉        | 4820/24420 [1:59:58<8:03:10,  1.48s/it]
 20%|█▉        | 4821/24420 [1:59:59<8:03:30,  1.48s/it]
 20%|█▉        | 4822/24420 [2:00:01<8:04:19,  1.48s/it]
 20%|█▉        | 4823/24420 [2:00:02<8:04:38,  1.48s/it]
 20%|█▉        | 4824/24420 [2:00:04<8:04:40,  1.48s/it]
 20%|█▉        | 4825/24420 [2:00:05<8:04:31,  1.48s/it]
                                                        
{'loss': 0.1462, 'learning_rate': 5e-05, 'epoch': 0.99}

 20%|█▉        | 4825/24420 [2:00:05<8:04:31,  1.48s/it]
 20%|█▉        | 4826/24420 [2:00:07<8:05:22,  1.49s/it]
 20%|█▉        | 4827/24420 [2:00:08<8:04:38,  1.48s/it]
 20%|█▉        | 4828/24420 [2:00:10<8:04:58,  1.49s/it]
 20%|█▉        | 4829/24420 [2:00:11<8:05:16,  1.49s/it]
 20%|█▉        | 4830/24420 [2:00:13<8:04:45,  1.48s/it]
                                                        
{'loss': 0.1569, 'learning_rate': 5e-05, 'epoch': 0.99}

 20%|█▉        | 4830/24420 [2:00:13<8:04:45,  1.48s/it]
 20%|█▉        | 4831/24420 [2:00:14<8:04:40,  1.48s/it]
 20%|█▉        | 4832/24420 [2:00:16<8:04:05,  1.48s/it]
 20%|█▉        | 4833/24420 [2:00:17<8:04:22,  1.48s/it]
 20%|█▉        | 4834/24420 [2:00:19<8:04:56,  1.49s/it]
 20%|█▉        | 4835/24420 [2:00:20<8:04:28,  1.48s/it]
                                                        
{'loss': 0.1564, 'learning_rate': 5e-05, 'epoch': 0.99}

 20%|█▉        | 4835/24420 [2:00:20<8:04:28,  1.48s/it]
 20%|█▉        | 4836/24420 [2:00:21<8:05:14,  1.49s/it]
 20%|█▉        | 4837/24420 [2:00:23<8:04:46,  1.49s/it]
 20%|█▉        | 4838/24420 [2:00:24<8:04:40,  1.49s/it]
 20%|█▉        | 4839/24420 [2:00:26<8:04:23,  1.48s/it]
 20%|█▉        | 4840/24420 [2:00:27<8:04:38,  1.49s/it]
                                                        
{'loss': 0.1449, 'learning_rate': 5e-05, 'epoch': 0.99}

 20%|█▉        | 4840/24420 [2:00:27<8:04:38,  1.49s/it]
 20%|█▉        | 4841/24420 [2:00:29<8:04:38,  1.49s/it]
 20%|█▉        | 4842/24420 [2:00:30<8:04:35,  1.49s/it]
 20%|█▉        | 4843/24420 [2:00:32<8:05:22,  1.49s/it]
 20%|█▉        | 4844/24420 [2:00:33<8:04:53,  1.49s/it]
 20%|█▉        | 4845/24420 [2:00:35<8:04:36,  1.49s/it]
                                                        
{'loss': 0.1469, 'learning_rate': 5e-05, 'epoch': 0.99}

 20%|█▉        | 4845/24420 [2:00:35<8:04:36,  1.49s/it]
 20%|█▉        | 4846/24420 [2:00:36<8:04:01,  1.48s/it]
 20%|█▉        | 4847/24420 [2:00:38<8:03:19,  1.48s/it]
 20%|█▉        | 4848/24420 [2:00:39<8:03:22,  1.48s/it]
 20%|█▉        | 4849/24420 [2:00:41<8:02:28,  1.48s/it]
 20%|█▉        | 4850/24420 [2:00:42<8:02:15,  1.48s/it]
                                                        
{'loss': 0.135, 'learning_rate': 5e-05, 'epoch': 0.99}

 20%|█▉        | 4850/24420 [2:00:42<8:02:15,  1.48s/it]
 20%|█▉        | 4851/24420 [2:00:44<8:02:36,  1.48s/it]
 20%|█▉        | 4852/24420 [2:00:45<8:02:24,  1.48s/it]
 20%|█▉        | 4853/24420 [2:00:47<8:02:56,  1.48s/it]
 20%|█▉        | 4854/24420 [2:00:48<8:02:21,  1.48s/it]
 20%|█▉        | 4855/24420 [2:00:50<8:02:55,  1.48s/it]
                                                        
{'loss': 0.145, 'learning_rate': 5e-05, 'epoch': 0.99}

 20%|█▉        | 4855/24420 [2:00:50<8:02:55,  1.48s/it]
 20%|█▉        | 4856/24420 [2:00:51<8:03:48,  1.48s/it]
 20%|█▉        | 4857/24420 [2:00:53<8:03:47,  1.48s/it]
 20%|█▉        | 4858/24420 [2:00:54<8:03:20,  1.48s/it]
 20%|█▉        | 4859/24420 [2:00:56<8:02:37,  1.48s/it]
 20%|█▉        | 4860/24420 [2:00:57<8:02:31,  1.48s/it]
                                                        
{'loss': 0.1528, 'learning_rate': 5e-05, 'epoch': 1.0}

 20%|█▉        | 4860/24420 [2:00:57<8:02:31,  1.48s/it]
 20%|█▉        | 4861/24420 [2:00:59<8:02:27,  1.48s/it]
 20%|█▉        | 4862/24420 [2:01:00<8:02:32,  1.48s/it]
 20%|█▉        | 4863/24420 [2:01:02<8:02:12,  1.48s/it]
 20%|█▉        | 4864/24420 [2:01:03<8:02:49,  1.48s/it]
 20%|█▉        | 4865/24420 [2:01:04<8:03:03,  1.48s/it]
                                                        
{'loss': 0.1447, 'learning_rate': 5e-05, 'epoch': 1.0}

 20%|█▉        | 4865/24420 [2:01:04<8:03:03,  1.48s/it]
 20%|█▉        | 4866/24420 [2:01:06<8:03:30,  1.48s/it]
 20%|█▉        | 4867/24420 [2:01:07<8:03:33,  1.48s/it]
 20%|█▉        | 4868/24420 [2:01:09<8:04:47,  1.49s/it]
 20%|█▉        | 4869/24420 [2:01:10<8:04:51,  1.49s/it]
 20%|█▉        | 4870/24420 [2:01:12<8:04:41,  1.49s/it]
                                                        
{'loss': 0.1586, 'learning_rate': 5e-05, 'epoch': 1.0}

 20%|█▉        | 4870/24420 [2:01:12<8:04:41,  1.49s/it]
 20%|█▉        | 4871/24420 [2:01:13<8:04:21,  1.49s/it]
 20%|█▉        | 4872/24420 [2:01:15<8:03:57,  1.49s/it]
 20%|█▉        | 4873/24420 [2:01:16<8:03:45,  1.48s/it]
 20%|█▉        | 4874/24420 [2:01:18<8:04:12,  1.49s/it]
 20%|█▉        | 4875/24420 [2:01:19<8:04:05,  1.49s/it]
                                                        
{'loss': 0.1531, 'learning_rate': 5e-05, 'epoch': 1.0}

 20%|█▉        | 4875/24420 [2:01:19<8:04:05,  1.49s/it]
 20%|█▉        | 4876/24420 [2:01:21<8:04:40,  1.49s/it]
 20%|█▉        | 4877/24420 [2:01:22<8:03:13,  1.48s/it]
 20%|█▉        | 4878/24420 [2:01:24<8:04:18,  1.49s/it]
 20%|█▉        | 4879/24420 [2:01:25<8:04:26,  1.49s/it]
 20%|█▉        | 4880/24420 [2:01:27<8:04:07,  1.49s/it]
                                                        
{'loss': 0.157, 'learning_rate': 5e-05, 'epoch': 1.0}

 20%|█▉        | 4880/24420 [2:01:27<8:04:07,  1.49s/it]
 20%|█▉        | 4881/24420 [2:01:28<8:05:20,  1.49s/it]
 20%|█▉        | 4882/24420 [2:01:30<8:04:41,  1.49s/it]
 20%|█▉        | 4883/24420 [2:01:31<8:03:38,  1.49s/it]
 20%|██        | 4884/24420 [2:01:33<8:03:49,  1.49s/it]
 20%|██        | 4885/24420 [2:01:34<8:07:08,  1.50s/it]
                                                        
{'loss': 0.1519, 'learning_rate': 5e-05, 'epoch': 1.0}

 20%|██        | 4885/24420 [2:01:34<8:07:08,  1.50s/it]
 20%|██        | 4886/24420 [2:01:36<8:07:26,  1.50s/it]
 20%|██        | 4887/24420 [2:01:37<8:06:23,  1.49s/it]
 20%|██        | 4888/24420 [2:01:39<8:06:16,  1.49s/it]
 20%|██        | 4889/24420 [2:01:40<8:04:27,  1.49s/it]
 20%|██        | 4890/24420 [2:01:42<8:02:50,  1.48s/it]
                                                        
{'loss': 0.1516, 'learning_rate': 5e-05, 'epoch': 1.0}

 20%|██        | 4890/24420 [2:01:42<8:02:50,  1.48s/it]
 20%|██        | 4891/24420 [2:01:43<8:02:30,  1.48s/it]
 20%|██        | 4892/24420 [2:01:45<8:02:40,  1.48s/it]
 20%|██        | 4893/24420 [2:01:46<8:03:24,  1.49s/it]
 20%|██        | 4894/24420 [2:01:48<8:02:47,  1.48s/it]
 20%|██        | 4895/24420 [2:01:49<8:02:20,  1.48s/it]
                                                        
{'loss': 0.1331, 'learning_rate': 5e-05, 'epoch': 1.0}

 20%|██        | 4895/24420 [2:01:49<8:02:20,  1.48s/it]
 20%|██        | 4896/24420 [2:01:51<8:02:40,  1.48s/it]
 20%|██        | 4897/24420 [2:01:52<8:01:46,  1.48s/it]
 20%|██        | 4898/24420 [2:01:54<8:02:32,  1.48s/it]
 20%|██        | 4899/24420 [2:01:55<8:02:44,  1.48s/it]
 20%|██        | 4900/24420 [2:01:57<8:03:17,  1.49s/it]
                                                        
{'loss': 0.1364, 'learning_rate': 5e-05, 'epoch': 1.0}

 20%|██        | 4900/24420 [2:01:57<8:03:17,  1.49s/it]
 20%|██        | 4901/24420 [2:01:58<8:04:50,  1.49s/it]
 20%|██        | 4902/24420 [2:01:59<8:04:19,  1.49s/it]
 20%|██        | 4903/24420 [2:02:01<8:04:03,  1.49s/it]
 20%|██        | 4904/24420 [2:02:02<8:03:21,  1.49s/it]
 20%|██        | 4905/24420 [2:02:04<8:03:30,  1.49s/it]
                                                        
{'loss': 0.1379, 'learning_rate': 5e-05, 'epoch': 1.0}

 20%|██        | 4905/24420 [2:02:04<8:03:30,  1.49s/it]
 20%|██        | 4906/24420 [2:02:05<8:03:21,  1.49s/it]
 20%|██        | 4907/24420 [2:02:07<8:03:54,  1.49s/it]
 20%|██        | 4908/24420 [2:02:08<8:03:40,  1.49s/it]
 20%|██        | 4909/24420 [2:02:10<8:02:35,  1.48s/it]
 20%|██        | 4910/24420 [2:02:11<8:02:23,  1.48s/it]
                                                        
{'loss': 0.1451, 'learning_rate': 5e-05, 'epoch': 1.01}

 20%|██        | 4910/24420 [2:02:11<8:02:23,  1.48s/it]
 20%|██        | 4911/24420 [2:02:13<8:02:01,  1.48s/it]
 20%|██        | 4912/24420 [2:02:14<8:01:55,  1.48s/it]
 20%|██        | 4913/24420 [2:02:16<8:02:09,  1.48s/it]
 20%|██        | 4914/24420 [2:02:17<8:02:07,  1.48s/it]
 20%|██        | 4915/24420 [2:02:19<8:01:47,  1.48s/it]
                                                        
{'loss': 0.1574, 'learning_rate': 5e-05, 'epoch': 1.01}

 20%|██        | 4915/24420 [2:02:19<8:01:47,  1.48s/it]
 20%|██        | 4916/24420 [2:02:20<8:02:04,  1.48s/it]
 20%|██        | 4917/24420 [2:02:22<8:02:18,  1.48s/it]
 20%|██        | 4918/24420 [2:02:23<8:01:12,  1.48s/it]
 20%|██        | 4919/24420 [2:02:25<8:00:41,  1.48s/it]
 20%|██        | 4920/24420 [2:02:26<8:02:34,  1.48s/it]
                                                        
{'loss': 0.1633, 'learning_rate': 5e-05, 'epoch': 1.01}

 20%|██        | 4920/24420 [2:02:26<8:02:34,  1.48s/it]
 20%|██        | 4921/24420 [2:02:28<8:02:44,  1.49s/it]
 20%|██        | 4922/24420 [2:02:29<8:02:18,  1.48s/it]
 20%|██        | 4923/24420 [2:02:31<8:01:34,  1.48s/it]
 20%|██        | 4924/24420 [2:02:32<8:02:00,  1.48s/it]
 20%|██        | 4925/24420 [2:02:34<8:01:02,  1.48s/it]
                                                        
{'loss': 0.1497, 'learning_rate': 5e-05, 'epoch': 1.01}

 20%|██        | 4925/24420 [2:02:34<8:01:02,  1.48s/it]
 20%|██        | 4926/24420 [2:02:35<8:01:26,  1.48s/it]
 20%|██        | 4927/24420 [2:02:37<8:01:38,  1.48s/it]
 20%|██        | 4928/24420 [2:02:38<8:01:28,  1.48s/it]
 20%|██        | 4929/24420 [2:02:40<8:01:40,  1.48s/it]
 20%|██        | 4930/24420 [2:02:41<8:01:15,  1.48s/it]
                                                        
{'loss': 0.1301, 'learning_rate': 5e-05, 'epoch': 1.01}

 20%|██        | 4930/24420 [2:02:41<8:01:15,  1.48s/it]
 20%|██        | 4931/24420 [2:02:43<8:01:45,  1.48s/it]
 20%|██        | 4932/24420 [2:02:44<8:00:48,  1.48s/it]
 20%|██        | 4933/24420 [2:02:45<8:01:06,  1.48s/it]
 20%|██        | 4934/24420 [2:02:47<8:01:37,  1.48s/it]
 20%|██        | 4935/24420 [2:02:48<8:02:55,  1.49s/it]
                                                        
{'loss': 0.1648, 'learning_rate': 5e-05, 'epoch': 1.01}

 20%|██        | 4935/24420 [2:02:48<8:02:55,  1.49s/it]
 20%|██        | 4936/24420 [2:02:50<8:02:50,  1.49s/it]
 20%|██        | 4937/24420 [2:02:51<8:02:06,  1.48s/it]
 20%|██        | 4938/24420 [2:02:53<8:02:42,  1.49s/it]
 20%|██        | 4939/24420 [2:02:54<8:02:04,  1.48s/it]
 20%|██        | 4940/24420 [2:02:56<8:01:56,  1.48s/it]
                                                        
{'loss': 0.1471, 'learning_rate': 5e-05, 'epoch': 1.01}

 20%|██        | 4940/24420 [2:02:56<8:01:56,  1.48s/it]
 20%|██        | 4941/24420 [2:02:57<8:02:40,  1.49s/it]
 20%|██        | 4942/24420 [2:02:59<8:02:03,  1.48s/it]
 20%|██        | 4943/24420 [2:03:00<8:01:40,  1.48s/it]
 20%|██        | 4944/24420 [2:03:02<8:01:17,  1.48s/it]
 20%|██        | 4945/24420 [2:03:03<8:02:02,  1.49s/it]
                                                        
{'loss': 0.1306, 'learning_rate': 5e-05, 'epoch': 1.01}

 20%|██        | 4945/24420 [2:03:03<8:02:02,  1.49s/it]
 20%|██        | 4946/24420 [2:03:05<8:03:57,  1.49s/it]
 20%|██        | 4947/24420 [2:03:06<8:03:25,  1.49s/it]
 20%|██        | 4948/24420 [2:03:08<8:01:42,  1.48s/it]
 20%|██        | 4949/24420 [2:03:09<8:01:40,  1.48s/it]
 20%|██        | 4950/24420 [2:03:11<8:01:21,  1.48s/it]
                                                        
{'loss': 0.1378, 'learning_rate': 5e-05, 'epoch': 1.01}

 20%|██        | 4950/24420 [2:03:11<8:01:21,  1.48s/it]
 20%|██        | 4951/24420 [2:03:12<8:01:42,  1.48s/it]
 20%|██        | 4952/24420 [2:03:14<8:01:00,  1.48s/it]
 20%|██        | 4953/24420 [2:03:15<8:01:07,  1.48s/it]
 20%|██        | 4954/24420 [2:03:17<8:00:45,  1.48s/it]
 20%|██        | 4955/24420 [2:03:18<8:00:37,  1.48s/it]
                                                        
{'loss': 0.1353, 'learning_rate': 5e-05, 'epoch': 1.01}

 20%|██        | 4955/24420 [2:03:18<8:00:37,  1.48s/it]
 20%|██        | 4956/24420 [2:03:20<8:00:56,  1.48s/it]
 20%|██        | 4957/24420 [2:03:21<8:01:00,  1.48s/it]
 20%|██        | 4958/24420 [2:03:23<8:00:52,  1.48s/it]
 20%|██        | 4959/24420 [2:03:24<8:01:05,  1.48s/it]
 20%|██        | 4960/24420 [2:03:26<8:01:11,  1.48s/it]
                                                        
{'loss': 0.1642, 'learning_rate': 5e-05, 'epoch': 1.02}

 20%|██        | 4960/24420 [2:03:26<8:01:11,  1.48s/it]
 20%|██        | 4961/24420 [2:03:27<8:00:49,  1.48s/it]
 20%|██        | 4962/24420 [2:03:29<8:00:43,  1.48s/it]
 20%|██        | 4963/24420 [2:03:30<8:01:06,  1.48s/it]
 20%|██        | 4964/24420 [2:03:31<8:01:29,  1.48s/it]
 20%|██        | 4965/24420 [2:03:33<8:01:40,  1.49s/it]
                                                        
{'loss': 0.1238, 'learning_rate': 5e-05, 'epoch': 1.02}

 20%|██        | 4965/24420 [2:03:33<8:01:40,  1.49s/it]
 20%|██        | 4966/24420 [2:03:34<8:01:13,  1.48s/it]
 20%|██        | 4967/24420 [2:03:36<8:01:30,  1.49s/it]
 20%|██        | 4968/24420 [2:03:37<8:00:39,  1.48s/it]
 20%|██        | 4969/24420 [2:03:39<8:00:05,  1.48s/it]
 20%|██        | 4970/24420 [2:03:40<7:59:22,  1.48s/it]
                                                        
{'loss': 0.1515, 'learning_rate': 5e-05, 'epoch': 1.02}

 20%|██        | 4970/24420 [2:03:40<7:59:22,  1.48s/it]
 20%|██        | 4971/24420 [2:03:42<8:00:26,  1.48s/it]
 20%|██        | 4972/24420 [2:03:43<8:00:50,  1.48s/it]
 20%|██        | 4973/24420 [2:03:45<8:00:30,  1.48s/it]
 20%|██        | 4974/24420 [2:03:46<8:00:41,  1.48s/it]
 20%|██        | 4975/24420 [2:03:48<8:01:14,  1.48s/it]
                                                        
{'loss': 0.143, 'learning_rate': 5e-05, 'epoch': 1.02}

 20%|██        | 4975/24420 [2:03:48<8:01:14,  1.48s/it]
 20%|██        | 4976/24420 [2:03:49<8:03:11,  1.49s/it]
 20%|██        | 4977/24420 [2:03:51<8:02:46,  1.49s/it]
 20%|██        | 4978/24420 [2:03:52<8:01:50,  1.49s/it]
 20%|██        | 4979/24420 [2:03:54<8:00:46,  1.48s/it]
 20%|██        | 4980/24420 [2:03:55<8:00:52,  1.48s/it]
                                                        
{'loss': 0.1427, 'learning_rate': 5e-05, 'epoch': 1.02}

 20%|██        | 4980/24420 [2:03:55<8:00:52,  1.48s/it]
 20%|██        | 4981/24420 [2:03:57<8:00:50,  1.48s/it]
 20%|██        | 4982/24420 [2:03:58<8:00:54,  1.48s/it]
 20%|██        | 4983/24420 [2:04:00<8:01:37,  1.49s/it]
 20%|██        | 4984/24420 [2:04:01<8:01:17,  1.49s/it]
 20%|██        | 4985/24420 [2:04:03<8:01:52,  1.49s/it]
                                                        
{'loss': 0.1399, 'learning_rate': 5e-05, 'epoch': 1.02}

 20%|██        | 4985/24420 [2:04:03<8:01:52,  1.49s/it]
 20%|██        | 4986/24420 [2:04:04<8:01:41,  1.49s/it]
 20%|██        | 4987/24420 [2:04:06<8:02:13,  1.49s/it]
 20%|██        | 4988/24420 [2:04:07<8:01:08,  1.49s/it]
 20%|██        | 4989/24420 [2:04:09<8:00:45,  1.48s/it]
 20%|██        | 4990/24420 [2:04:10<8:00:06,  1.48s/it]
                                                        
{'loss': 0.1417, 'learning_rate': 5e-05, 'epoch': 1.02}

 20%|██        | 4990/24420 [2:04:10<8:00:06,  1.48s/it]
 20%|██        | 4991/24420 [2:04:12<8:00:01,  1.48s/it]
 20%|██        | 4992/24420 [2:04:13<7:59:43,  1.48s/it]
 20%|██        | 4993/24420 [2:04:15<8:00:00,  1.48s/it]
 20%|██        | 4994/24420 [2:04:16<7:59:16,  1.48s/it]
 20%|██        | 4995/24420 [2:04:17<7:59:11,  1.48s/it]
                                                        
{'loss': 0.1533, 'learning_rate': 5e-05, 'epoch': 1.02}

 20%|██        | 4995/24420 [2:04:17<7:59:11,  1.48s/it]
 20%|██        | 4996/24420 [2:04:19<7:59:03,  1.48s/it]
 20%|██        | 4997/24420 [2:04:20<7:59:26,  1.48s/it]
 20%|██        | 4998/24420 [2:04:22<8:00:40,  1.48s/it]
 20%|██        | 4999/24420 [2:04:23<8:00:54,  1.49s/it]
 20%|██        | 5000/24420 [2:04:25<8:00:05,  1.48s/it]
                                                        
{'loss': 0.1333, 'learning_rate': 5e-05, 'epoch': 1.02}

 20%|██        | 5000/24420 [2:04:25<8:00:05,  1.48s/it]
 20%|██        | 5001/24420 [2:04:30<13:19:17,  2.47s/it]
 20%|██        | 5002/24420 [2:04:31<11:43:00,  2.17s/it]
 20%|██        | 5003/24420 [2:04:33<10:36:26,  1.97s/it]
 20%|██        | 5004/24420 [2:04:34<9:49:09,  1.82s/it] 
 20%|██        | 5005/24420 [2:04:36<9:16:25,  1.72s/it]
                                                        
{'loss': 0.1343, 'learning_rate': 5e-05, 'epoch': 1.02}

 20%|██        | 5005/24420 [2:04:36<9:16:25,  1.72s/it]
 20%|██        | 5006/24420 [2:04:37<8:53:51,  1.65s/it]
 21%|██        | 5007/24420 [2:04:39<8:36:51,  1.60s/it]
 21%|██        | 5008/24420 [2:04:40<8:26:00,  1.56s/it]
 21%|██        | 5009/24420 [2:04:42<8:17:51,  1.54s/it]
 21%|██        | 5010/24420 [2:04:43<8:12:04,  1.52s/it]
                                                        
{'loss': 0.1411, 'learning_rate': 5e-05, 'epoch': 1.03}

 21%|██        | 5010/24420 [2:04:43<8:12:04,  1.52s/it]
 21%|██        | 5011/24420 [2:04:44<8:08:10,  1.51s/it]
 21%|██        | 5012/24420 [2:04:46<8:05:05,  1.50s/it]
 21%|██        | 5013/24420 [2:04:47<8:03:50,  1.50s/it]
 21%|██        | 5014/24420 [2:04:49<8:03:05,  1.49s/it]
 21%|██        | 5015/24420 [2:04:50<8:01:41,  1.49s/it]
                                                        
{'loss': 0.1227, 'learning_rate': 5e-05, 'epoch': 1.03}

 21%|██        | 5015/24420 [2:04:50<8:01:41,  1.49s/it]
 21%|██        | 5016/24420 [2:04:52<8:01:58,  1.49s/it]
 21%|██        | 5017/24420 [2:04:53<8:02:22,  1.49s/it]
 21%|██        | 5018/24420 [2:04:55<8:01:27,  1.49s/it]
 21%|██        | 5019/24420 [2:04:56<8:00:38,  1.49s/it]
 21%|██        | 5020/24420 [2:04:58<8:00:30,  1.49s/it]
                                                        
{'loss': 0.133, 'learning_rate': 5e-05, 'epoch': 1.03}

 21%|██        | 5020/24420 [2:04:58<8:00:30,  1.49s/it]
 21%|██        | 5021/24420 [2:04:59<7:59:58,  1.48s/it]
 21%|██        | 5022/24420 [2:05:01<7:59:40,  1.48s/it]
 21%|██        | 5023/24420 [2:05:02<7:59:06,  1.48s/it]
 21%|██        | 5024/24420 [2:05:04<7:59:16,  1.48s/it]
 21%|██        | 5025/24420 [2:05:05<7:59:06,  1.48s/it]
                                                        
{'loss': 0.1301, 'learning_rate': 5e-05, 'epoch': 1.03}

 21%|██        | 5025/24420 [2:05:05<7:59:06,  1.48s/it]
 21%|██        | 5026/24420 [2:05:07<7:59:31,  1.48s/it]
 21%|██        | 5027/24420 [2:05:08<7:59:11,  1.48s/it]
 21%|██        | 5028/24420 [2:05:10<8:00:11,  1.49s/it]
 21%|██        | 5029/24420 [2:05:11<7:59:35,  1.48s/it]
 21%|██        | 5030/24420 [2:05:13<7:59:23,  1.48s/it]
                                                        
{'loss': 0.1444, 'learning_rate': 5e-05, 'epoch': 1.03}

 21%|██        | 5030/24420 [2:05:13<7:59:23,  1.48s/it]
 21%|██        | 5031/24420 [2:05:14<7:59:27,  1.48s/it]
 21%|██        | 5032/24420 [2:05:16<8:00:45,  1.49s/it]
 21%|██        | 5033/24420 [2:05:17<7:59:37,  1.48s/it]
 21%|██        | 5034/24420 [2:05:19<7:59:46,  1.48s/it]
 21%|██        | 5035/24420 [2:05:20<7:59:23,  1.48s/it]
                                                        
{'loss': 0.148, 'learning_rate': 5e-05, 'epoch': 1.03}

 21%|██        | 5035/24420 [2:05:20<7:59:23,  1.48s/it]
 21%|██        | 5036/24420 [2:05:22<7:59:56,  1.49s/it]
 21%|██        | 5037/24420 [2:05:23<8:00:44,  1.49s/it]
 21%|██        | 5038/24420 [2:05:25<8:00:14,  1.49s/it]
 21%|██        | 5039/24420 [2:05:26<7:59:34,  1.48s/it]
 21%|██        | 5040/24420 [2:05:28<8:00:16,  1.49s/it]
                                                        
{'loss': 0.1666, 'learning_rate': 5e-05, 'epoch': 1.03}

 21%|██        | 5040/24420 [2:05:28<8:00:16,  1.49s/it]
 21%|██        | 5041/24420 [2:05:29<7:59:52,  1.49s/it]
 21%|██        | 5042/24420 [2:05:31<8:00:14,  1.49s/it]
 21%|██        | 5043/24420 [2:05:32<7:59:45,  1.49s/it]
 21%|██        | 5044/24420 [2:05:34<8:00:07,  1.49s/it]
 21%|██        | 5045/24420 [2:05:35<8:00:10,  1.49s/it]
                                                        
{'loss': 0.1587, 'learning_rate': 5e-05, 'epoch': 1.03}

 21%|██        | 5045/24420 [2:05:35<8:00:10,  1.49s/it]
 21%|██        | 5046/24420 [2:05:37<8:02:24,  1.49s/it]
 21%|██        | 5047/24420 [2:05:38<8:01:54,  1.49s/it]
 21%|██        | 5048/24420 [2:05:39<8:01:11,  1.49s/it]
 21%|██        | 5049/24420 [2:05:41<8:00:25,  1.49s/it]
 21%|██        | 5050/24420 [2:05:42<7:59:33,  1.49s/it]
                                                        
{'loss': 0.1341, 'learning_rate': 5e-05, 'epoch': 1.03}

 21%|██        | 5050/24420 [2:05:42<7:59:33,  1.49s/it]
 21%|██        | 5051/24420 [2:05:44<8:00:10,  1.49s/it]
 21%|██        | 5052/24420 [2:05:45<8:00:02,  1.49s/it]
 21%|██        | 5053/24420 [2:05:47<8:00:36,  1.49s/it]
 21%|██        | 5054/24420 [2:05:48<7:59:36,  1.49s/it]
 21%|██        | 5055/24420 [2:05:50<7:58:47,  1.48s/it]
                                                        
{'loss': 0.1339, 'learning_rate': 5e-05, 'epoch': 1.04}

 21%|██        | 5055/24420 [2:05:50<7:58:47,  1.48s/it]
 21%|██        | 5056/24420 [2:05:51<7:58:26,  1.48s/it]
 21%|██        | 5057/24420 [2:05:53<7:58:00,  1.48s/it]
 21%|██        | 5058/24420 [2:05:54<7:59:12,  1.48s/it]
 21%|██        | 5059/24420 [2:05:56<8:00:08,  1.49s/it]
 21%|██        | 5060/24420 [2:05:57<8:00:13,  1.49s/it]
                                                        
{'loss': 0.1686, 'learning_rate': 5e-05, 'epoch': 1.04}

 21%|██        | 5060/24420 [2:05:57<8:00:13,  1.49s/it]
 21%|██        | 5061/24420 [2:05:59<8:00:11,  1.49s/it]
 21%|██        | 5062/24420 [2:06:00<7:58:59,  1.48s/it]
 21%|██        | 5063/24420 [2:06:02<7:59:04,  1.48s/it]
 21%|██        | 5064/24420 [2:06:03<7:58:50,  1.48s/it]
 21%|██        | 5065/24420 [2:06:05<8:00:56,  1.49s/it]
                                                        
{'loss': 0.157, 'learning_rate': 5e-05, 'epoch': 1.04}

 21%|██        | 5065/24420 [2:06:05<8:00:56,  1.49s/it]
 21%|██        | 5066/24420 [2:06:06<8:00:56,  1.49s/it]
 21%|██        | 5067/24420 [2:06:08<7:59:57,  1.49s/it]
 21%|██        | 5068/24420 [2:06:09<7:59:44,  1.49s/it]
 21%|██        | 5069/24420 [2:06:11<7:59:15,  1.49s/it]
 21%|██        | 5070/24420 [2:06:12<7:59:10,  1.49s/it]
                                                        
{'loss': 0.1442, 'learning_rate': 5e-05, 'epoch': 1.04}

 21%|██        | 5070/24420 [2:06:12<7:59:10,  1.49s/it]
 21%|██        | 5071/24420 [2:06:14<8:00:22,  1.49s/it]
 21%|██        | 5072/24420 [2:06:15<7:59:41,  1.49s/it]
 21%|██        | 5073/24420 [2:06:17<8:00:03,  1.49s/it]
 21%|██        | 5074/24420 [2:06:18<7:59:49,  1.49s/it]
 21%|██        | 5075/24420 [2:06:20<7:58:57,  1.49s/it]
                                                        
{'loss': 0.1352, 'learning_rate': 5e-05, 'epoch': 1.04}

 21%|██        | 5075/24420 [2:06:20<7:58:57,  1.49s/it]
 21%|██        | 5076/24420 [2:06:21<7:58:34,  1.48s/it]
 21%|██        | 5077/24420 [2:06:23<7:58:44,  1.49s/it]
 21%|██        | 5078/24420 [2:06:24<7:58:08,  1.48s/it]
 21%|██        | 5079/24420 [2:06:26<7:57:47,  1.48s/it]
 21%|██        | 5080/24420 [2:06:27<7:57:56,  1.48s/it]
                                                        
{'loss': 0.1387, 'learning_rate': 5e-05, 'epoch': 1.04}

 21%|██        | 5080/24420 [2:06:27<7:57:56,  1.48s/it]
 21%|██        | 5081/24420 [2:06:29<7:59:21,  1.49s/it]
 21%|██        | 5082/24420 [2:06:30<7:59:11,  1.49s/it]
 21%|██        | 5083/24420 [2:06:31<7:58:37,  1.49s/it]
 21%|██        | 5084/24420 [2:06:33<7:58:18,  1.48s/it]
 21%|██        | 5085/24420 [2:06:34<7:58:11,  1.48s/it]
                                                        
{'loss': 0.1512, 'learning_rate': 5e-05, 'epoch': 1.04}

 21%|██        | 5085/24420 [2:06:34<7:58:11,  1.48s/it]
 21%|██        | 5086/24420 [2:06:36<7:58:35,  1.49s/it]
 21%|██        | 5087/24420 [2:06:37<7:57:24,  1.48s/it]
 21%|██        | 5088/24420 [2:06:39<7:57:46,  1.48s/it]
 21%|██        | 5089/24420 [2:06:40<7:57:29,  1.48s/it]
 21%|██        | 5090/24420 [2:06:42<7:57:36,  1.48s/it]
                                                        
{'loss': 0.1491, 'learning_rate': 5e-05, 'epoch': 1.04}

 21%|██        | 5090/24420 [2:06:42<7:57:36,  1.48s/it]
 21%|██        | 5091/24420 [2:06:43<7:57:51,  1.48s/it]
 21%|██        | 5092/24420 [2:06:45<7:59:07,  1.49s/it]
 21%|██        | 5093/24420 [2:06:46<7:58:53,  1.49s/it]
 21%|██        | 5094/24420 [2:06:48<7:58:16,  1.48s/it]
 21%|██        | 5095/24420 [2:06:49<7:57:53,  1.48s/it]
                                                        
{'loss': 0.1381, 'learning_rate': 5e-05, 'epoch': 1.04}

 21%|██        | 5095/24420 [2:06:49<7:57:53,  1.48s/it]
 21%|██        | 5096/24420 [2:06:51<7:58:03,  1.48s/it]
 21%|██        | 5097/24420 [2:06:52<7:58:07,  1.48s/it]
 21%|██        | 5098/24420 [2:06:54<7:57:25,  1.48s/it]
 21%|██        | 5099/24420 [2:06:55<7:56:38,  1.48s/it]
 21%|██        | 5100/24420 [2:06:57<7:56:07,  1.48s/it]
                                                        
{'loss': 0.1401, 'learning_rate': 5e-05, 'epoch': 1.04}

 21%|██        | 5100/24420 [2:06:57<7:56:07,  1.48s/it]
 21%|██        | 5101/24420 [2:06:58<7:56:21,  1.48s/it]
 21%|██        | 5102/24420 [2:07:00<7:56:14,  1.48s/it]
 21%|██        | 5103/24420 [2:07:01<7:56:13,  1.48s/it]
 21%|██        | 5104/24420 [2:07:03<7:56:45,  1.48s/it]
 21%|██        | 5105/24420 [2:07:04<7:57:37,  1.48s/it]
                                                        
{'loss': 0.1763, 'learning_rate': 5e-05, 'epoch': 1.05}

 21%|██        | 5105/24420 [2:07:04<7:57:37,  1.48s/it]
 21%|██        | 5106/24420 [2:07:06<7:58:22,  1.49s/it]
 21%|██        | 5107/24420 [2:07:07<7:57:13,  1.48s/it]
 21%|██        | 5108/24420 [2:07:09<7:57:07,  1.48s/it]
 21%|██        | 5109/24420 [2:07:10<7:56:57,  1.48s/it]
 21%|██        | 5110/24420 [2:07:12<7:56:25,  1.48s/it]
                                                        
{'loss': 0.1483, 'learning_rate': 5e-05, 'epoch': 1.05}

 21%|██        | 5110/24420 [2:07:12<7:56:25,  1.48s/it]
 21%|██        | 5111/24420 [2:07:13<7:56:45,  1.48s/it]
 21%|██        | 5112/24420 [2:07:14<7:56:50,  1.48s/it]
 21%|██        | 5113/24420 [2:07:16<7:56:06,  1.48s/it]
 21%|██        | 5114/24420 [2:07:17<7:56:54,  1.48s/it]
 21%|██        | 5115/24420 [2:07:19<7:57:19,  1.48s/it]
                                                        
{'loss': 0.1315, 'learning_rate': 5e-05, 'epoch': 1.05}

 21%|██        | 5115/24420 [2:07:19<7:57:19,  1.48s/it]
 21%|██        | 5116/24420 [2:07:20<7:58:04,  1.49s/it]
 21%|██        | 5117/24420 [2:07:22<7:57:37,  1.48s/it]
 21%|██        | 5118/24420 [2:07:23<7:57:19,  1.48s/it]
 21%|██        | 5119/24420 [2:07:25<7:57:41,  1.48s/it]
 21%|██        | 5120/24420 [2:07:26<7:57:31,  1.48s/it]
                                                        
{'loss': 0.1468, 'learning_rate': 5e-05, 'epoch': 1.05}

 21%|██        | 5120/24420 [2:07:26<7:57:31,  1.48s/it]
 21%|██        | 5121/24420 [2:07:28<7:57:51,  1.49s/it]
 21%|██        | 5122/24420 [2:07:29<7:56:53,  1.48s/it]
 21%|██        | 5123/24420 [2:07:31<7:56:42,  1.48s/it]
 21%|██        | 5124/24420 [2:07:32<7:56:15,  1.48s/it]
 21%|██        | 5125/24420 [2:07:34<7:57:21,  1.48s/it]
                                                        
{'loss': 0.1465, 'learning_rate': 5e-05, 'epoch': 1.05}

 21%|██        | 5125/24420 [2:07:34<7:57:21,  1.48s/it]
 21%|██        | 5126/24420 [2:07:35<7:57:13,  1.48s/it]
 21%|██        | 5127/24420 [2:07:37<7:58:08,  1.49s/it]
 21%|██        | 5128/24420 [2:07:38<7:57:40,  1.49s/it]
 21%|██        | 5129/24420 [2:07:40<7:56:51,  1.48s/it]
 21%|██        | 5130/24420 [2:07:41<7:56:17,  1.48s/it]
                                                        
{'loss': 0.1631, 'learning_rate': 5e-05, 'epoch': 1.05}

 21%|██        | 5130/24420 [2:07:41<7:56:17,  1.48s/it]
 21%|██        | 5131/24420 [2:07:43<7:56:05,  1.48s/it]
 21%|██        | 5132/24420 [2:07:44<7:55:49,  1.48s/it]
 21%|██        | 5133/24420 [2:07:46<7:56:25,  1.48s/it]
 21%|██        | 5134/24420 [2:07:47<7:56:50,  1.48s/it]
 21%|██        | 5135/24420 [2:07:49<7:56:27,  1.48s/it]
                                                        
{'loss': 0.1396, 'learning_rate': 5e-05, 'epoch': 1.05}

 21%|██        | 5135/24420 [2:07:49<7:56:27,  1.48s/it]
 21%|██        | 5136/24420 [2:07:50<7:55:54,  1.48s/it]
 21%|██        | 5137/24420 [2:07:52<7:56:37,  1.48s/it]
 21%|██        | 5138/24420 [2:07:53<7:56:13,  1.48s/it]
 21%|██        | 5139/24420 [2:07:55<7:56:26,  1.48s/it]
 21%|██        | 5140/24420 [2:07:56<7:56:19,  1.48s/it]
                                                        
{'loss': 0.1306, 'learning_rate': 5e-05, 'epoch': 1.05}

 21%|██        | 5140/24420 [2:07:56<7:56:19,  1.48s/it]
 21%|██        | 5141/24420 [2:07:57<7:56:15,  1.48s/it]
 21%|██        | 5142/24420 [2:07:59<7:56:21,  1.48s/it]
 21%|██        | 5143/24420 [2:08:00<7:56:08,  1.48s/it]
 21%|██        | 5144/24420 [2:08:02<7:55:28,  1.48s/it]
 21%|██        | 5145/24420 [2:08:03<7:55:36,  1.48s/it]
                                                        
{'loss': 0.137, 'learning_rate': 5e-05, 'epoch': 1.05}

 21%|██        | 5145/24420 [2:08:03<7:55:36,  1.48s/it]
 21%|██        | 5146/24420 [2:08:05<7:56:23,  1.48s/it]
 21%|██        | 5147/24420 [2:08:06<7:56:33,  1.48s/it]
 21%|██        | 5148/24420 [2:08:08<7:56:05,  1.48s/it]
 21%|██        | 5149/24420 [2:08:09<7:56:07,  1.48s/it]
 21%|██        | 5150/24420 [2:08:11<7:56:02,  1.48s/it]
                                                        
{'loss': 0.1419, 'learning_rate': 5e-05, 'epoch': 1.05}

 21%|██        | 5150/24420 [2:08:11<7:56:02,  1.48s/it]
 21%|██        | 5151/24420 [2:08:12<7:56:04,  1.48s/it]
 21%|██        | 5152/24420 [2:08:14<7:56:29,  1.48s/it]
 21%|██        | 5153/24420 [2:08:15<7:56:55,  1.49s/it]
 21%|██        | 5154/24420 [2:08:17<7:56:55,  1.49s/it]
 21%|██        | 5155/24420 [2:08:18<7:56:37,  1.48s/it]
                                                        
{'loss': 0.1318, 'learning_rate': 5e-05, 'epoch': 1.06}

 21%|██        | 5155/24420 [2:08:18<7:56:37,  1.48s/it]
 21%|██        | 5156/24420 [2:08:20<7:56:27,  1.48s/it]
 21%|██        | 5157/24420 [2:08:21<7:56:52,  1.49s/it]
 21%|██        | 5158/24420 [2:08:23<7:56:40,  1.48s/it]
 21%|██        | 5159/24420 [2:08:24<7:57:15,  1.49s/it]
 21%|██        | 5160/24420 [2:08:26<7:56:31,  1.48s/it]
                                                        
{'loss': 0.1477, 'learning_rate': 5e-05, 'epoch': 1.06}

 21%|██        | 5160/24420 [2:08:26<7:56:31,  1.48s/it]
 21%|██        | 5161/24420 [2:08:27<7:56:49,  1.49s/it]
 21%|██        | 5162/24420 [2:08:29<7:56:22,  1.48s/it]
 21%|██        | 5163/24420 [2:08:30<7:56:17,  1.48s/it]
 21%|██        | 5164/24420 [2:08:32<7:56:26,  1.48s/it]
 21%|██        | 5165/24420 [2:08:33<7:56:48,  1.49s/it]
                                                        
{'loss': 0.1282, 'learning_rate': 5e-05, 'epoch': 1.06}

 21%|██        | 5165/24420 [2:08:33<7:56:48,  1.49s/it]
 21%|██        | 5166/24420 [2:08:35<7:57:21,  1.49s/it]
 21%|██        | 5167/24420 [2:08:36<7:57:22,  1.49s/it]
 21%|██        | 5168/24420 [2:08:38<7:56:32,  1.49s/it]
 21%|██        | 5169/24420 [2:08:39<7:57:23,  1.49s/it]
 21%|██        | 5170/24420 [2:08:41<7:56:31,  1.49s/it]
                                                        
{'loss': 0.1336, 'learning_rate': 5e-05, 'epoch': 1.06}

 21%|██        | 5170/24420 [2:08:41<7:56:31,  1.49s/it]
 21%|██        | 5171/24420 [2:08:42<7:55:55,  1.48s/it]
 21%|██        | 5172/24420 [2:08:43<7:56:01,  1.48s/it]
 21%|██        | 5173/24420 [2:08:45<7:55:56,  1.48s/it]
 21%|██        | 5174/24420 [2:08:46<7:55:58,  1.48s/it]
 21%|██        | 5175/24420 [2:08:48<7:56:22,  1.49s/it]
                                                        
{'loss': 0.1504, 'learning_rate': 5e-05, 'epoch': 1.06}

 21%|██        | 5175/24420 [2:08:48<7:56:22,  1.49s/it]
 21%|██        | 5176/24420 [2:08:49<7:56:14,  1.48s/it]
 21%|██        | 5177/24420 [2:08:51<7:55:31,  1.48s/it]
 21%|██        | 5178/24420 [2:08:52<7:55:05,  1.48s/it]
 21%|██        | 5179/24420 [2:08:54<7:54:37,  1.48s/it]
 21%|██        | 5180/24420 [2:08:55<7:55:02,  1.48s/it]
                                                        
{'loss': 0.1568, 'learning_rate': 5e-05, 'epoch': 1.06}

 21%|██        | 5180/24420 [2:08:55<7:55:02,  1.48s/it]
 21%|██        | 5181/24420 [2:08:57<7:56:15,  1.49s/it]
 21%|██        | 5182/24420 [2:08:58<7:55:07,  1.48s/it]
 21%|██        | 5183/24420 [2:09:00<7:55:05,  1.48s/it]
 21%|██        | 5184/24420 [2:09:01<7:55:04,  1.48s/it]
 21%|██        | 5185/24420 [2:09:03<7:54:30,  1.48s/it]
                                                        
{'loss': 0.1414, 'learning_rate': 5e-05, 'epoch': 1.06}

 21%|██        | 5185/24420 [2:09:03<7:54:30,  1.48s/it]
 21%|██        | 5186/24420 [2:09:04<7:54:11,  1.48s/it]
 21%|██        | 5187/24420 [2:09:06<7:53:56,  1.48s/it]
 21%|██        | 5188/24420 [2:09:07<7:53:57,  1.48s/it]
 21%|██        | 5189/24420 [2:09:09<7:55:03,  1.48s/it]
 21%|██▏       | 5190/24420 [2:09:10<7:55:16,  1.48s/it]
                                                        
{'loss': 0.1429, 'learning_rate': 5e-05, 'epoch': 1.06}

 21%|██▏       | 5190/24420 [2:09:10<7:55:16,  1.48s/it]
 21%|██▏       | 5191/24420 [2:09:12<7:55:26,  1.48s/it]
 21%|██▏       | 5192/24420 [2:09:13<7:54:48,  1.48s/it]
 21%|██▏       | 5193/24420 [2:09:15<7:54:53,  1.48s/it]
 21%|██▏       | 5194/24420 [2:09:16<7:55:31,  1.48s/it]
 21%|██▏       | 5195/24420 [2:09:18<7:55:18,  1.48s/it]
                                                        
{'loss': 0.1416, 'learning_rate': 5e-05, 'epoch': 1.06}

 21%|██▏       | 5195/24420 [2:09:18<7:55:18,  1.48s/it]
 21%|██▏       | 5196/24420 [2:09:19<7:54:46,  1.48s/it]
 21%|██▏       | 5197/24420 [2:09:21<7:54:30,  1.48s/it]
 21%|██▏       | 5198/24420 [2:09:22<7:54:00,  1.48s/it]
 21%|██▏       | 5199/24420 [2:09:23<7:54:00,  1.48s/it]
 21%|██▏       | 5200/24420 [2:09:25<7:53:30,  1.48s/it]
                                                        
{'loss': 0.134, 'learning_rate': 5e-05, 'epoch': 1.06}

 21%|██▏       | 5200/24420 [2:09:25<7:53:30,  1.48s/it]
 21%|██▏       | 5201/24420 [2:09:29<12:46:04,  2.39s/it]
 21%|██▏       | 5202/24420 [2:09:31<11:18:05,  2.12s/it]
 21%|██▏       | 5203/24420 [2:09:32<10:16:52,  1.93s/it]
 21%|██▏       | 5204/24420 [2:09:34<9:34:28,  1.79s/it] 
 21%|██▏       | 5205/24420 [2:09:35<9:05:03,  1.70s/it]
                                                        
{'loss': 0.1445, 'learning_rate': 5e-05, 'epoch': 1.07}

 21%|██▏       | 5205/24420 [2:09:35<9:05:03,  1.70s/it]
 21%|██▏       | 5206/24420 [2:09:37<8:44:30,  1.64s/it]
 21%|██▏       | 5207/24420 [2:09:38<8:28:47,  1.59s/it]
 21%|██▏       | 5208/24420 [2:09:40<8:18:58,  1.56s/it]
 21%|██▏       | 5209/24420 [2:09:41<8:11:36,  1.54s/it]
 21%|██▏       | 5210/24420 [2:09:43<8:07:02,  1.52s/it]
                                                        
{'loss': 0.1766, 'learning_rate': 5e-05, 'epoch': 1.07}

 21%|██▏       | 5210/24420 [2:09:43<8:07:02,  1.52s/it]
 21%|██▏       | 5211/24420 [2:09:44<8:03:23,  1.51s/it]
 21%|██▏       | 5212/24420 [2:09:46<8:01:02,  1.50s/it]
 21%|██▏       | 5213/24420 [2:09:47<7:59:20,  1.50s/it]
 21%|██▏       | 5214/24420 [2:09:49<7:58:10,  1.49s/it]
 21%|██▏       | 5215/24420 [2:09:50<7:57:47,  1.49s/it]
                                                        
{'loss': 0.1356, 'learning_rate': 5e-05, 'epoch': 1.07}

 21%|██▏       | 5215/24420 [2:09:50<7:57:47,  1.49s/it]
 21%|██▏       | 5216/24420 [2:09:52<7:57:09,  1.49s/it]
 21%|██▏       | 5217/24420 [2:09:53<7:56:28,  1.49s/it]
 21%|██▏       | 5218/24420 [2:09:55<7:56:27,  1.49s/it]
 21%|██▏       | 5219/24420 [2:09:56<7:55:14,  1.49s/it]
 21%|██▏       | 5220/24420 [2:09:58<7:55:51,  1.49s/it]
                                                        
{'loss': 0.1571, 'learning_rate': 5e-05, 'epoch': 1.07}

 21%|██▏       | 5220/24420 [2:09:58<7:55:51,  1.49s/it]
 21%|██▏       | 5221/24420 [2:09:59<7:55:36,  1.49s/it]
 21%|██▏       | 5222/24420 [2:10:01<7:55:04,  1.48s/it]
 21%|██▏       | 5223/24420 [2:10:02<7:55:23,  1.49s/it]
 21%|██▏       | 5224/24420 [2:10:04<7:55:02,  1.48s/it]
 21%|██▏       | 5225/24420 [2:10:06<8:53:52,  1.67s/it]
                                                        
{'loss': 0.1382, 'learning_rate': 5e-05, 'epoch': 1.07}

 21%|██▏       | 5225/24420 [2:10:06<8:53:52,  1.67s/it]
 21%|██▏       | 5226/24420 [2:10:07<8:36:43,  1.62s/it]
 21%|██▏       | 5227/24420 [2:10:09<8:22:58,  1.57s/it]
 21%|██▏       | 5228/24420 [2:10:10<8:14:10,  1.54s/it]
 21%|██▏       | 5229/24420 [2:10:12<8:08:58,  1.53s/it]
 21%|██▏       | 5230/24420 [2:10:13<8:04:06,  1.51s/it]
                                                        
{'loss': 0.1313, 'learning_rate': 5e-05, 'epoch': 1.07}

 21%|██▏       | 5230/24420 [2:10:13<8:04:06,  1.51s/it]
 21%|██▏       | 5231/24420 [2:10:15<8:00:59,  1.50s/it]
 21%|██▏       | 5232/24420 [2:10:16<7:58:49,  1.50s/it]
 21%|██▏       | 5233/24420 [2:10:18<7:57:45,  1.49s/it]
 21%|██▏       | 5234/24420 [2:10:19<7:55:42,  1.49s/it]
 21%|██▏       | 5235/24420 [2:10:21<7:55:10,  1.49s/it]
                                                        
{'loss': 0.1336, 'learning_rate': 5e-05, 'epoch': 1.07}

 21%|██▏       | 5235/24420 [2:10:21<7:55:10,  1.49s/it]
 21%|██▏       | 5236/24420 [2:10:22<7:55:22,  1.49s/it]
 21%|██▏       | 5237/24420 [2:10:24<7:54:26,  1.48s/it]
 21%|██▏       | 5238/24420 [2:10:25<7:54:00,  1.48s/it]
 21%|██▏       | 5239/24420 [2:10:26<7:53:56,  1.48s/it]
 21%|██▏       | 5240/24420 [2:10:28<7:54:07,  1.48s/it]
                                                        
{'loss': 0.1305, 'learning_rate': 5e-05, 'epoch': 1.07}

 21%|██▏       | 5240/24420 [2:10:28<7:54:07,  1.48s/it]
 21%|██▏       | 5241/24420 [2:10:29<7:53:51,  1.48s/it]
 21%|██▏       | 5242/24420 [2:10:31<7:54:02,  1.48s/it]
 21%|██▏       | 5243/24420 [2:10:32<7:53:57,  1.48s/it]
 21%|██▏       | 5244/24420 [2:10:34<7:55:36,  1.49s/it]
 21%|██▏       | 5245/24420 [2:10:35<7:54:48,  1.49s/it]
                                                        
{'loss': 0.1309, 'learning_rate': 5e-05, 'epoch': 1.07}

 21%|██▏       | 5245/24420 [2:10:35<7:54:48,  1.49s/it]
 21%|██▏       | 5246/24420 [2:10:37<7:54:21,  1.48s/it]
 21%|██▏       | 5247/24420 [2:10:38<7:53:48,  1.48s/it]
 21%|██▏       | 5248/24420 [2:10:40<7:53:53,  1.48s/it]
 21%|██▏       | 5249/24420 [2:10:41<7:54:12,  1.48s/it]
 21%|██▏       | 5250/24420 [2:10:43<7:54:36,  1.49s/it]
                                                        
{'loss': 0.1082, 'learning_rate': 5e-05, 'epoch': 1.07}

 21%|██▏       | 5250/24420 [2:10:43<7:54:36,  1.49s/it]
 22%|██▏       | 5251/24420 [2:10:44<7:54:26,  1.49s/it]
 22%|██▏       | 5252/24420 [2:10:46<7:54:26,  1.49s/it]
 22%|██▏       | 5253/24420 [2:10:47<7:54:46,  1.49s/it]
 22%|██▏       | 5254/24420 [2:10:49<7:54:36,  1.49s/it]
 22%|██▏       | 5255/24420 [2:10:50<7:53:41,  1.48s/it]
                                                        
{'loss': 0.1463, 'learning_rate': 5e-05, 'epoch': 1.08}

 22%|██▏       | 5255/24420 [2:10:50<7:53:41,  1.48s/it]
 22%|██▏       | 5256/24420 [2:10:52<7:53:27,  1.48s/it]
 22%|██▏       | 5257/24420 [2:10:53<7:53:03,  1.48s/it]
 22%|██▏       | 5258/24420 [2:10:55<7:52:45,  1.48s/it]
 22%|██▏       | 5259/24420 [2:10:56<7:52:46,  1.48s/it]
 22%|██▏       | 5260/24420 [2:10:58<7:52:46,  1.48s/it]
                                                        
{'loss': 0.1141, 'learning_rate': 5e-05, 'epoch': 1.08}

 22%|██▏       | 5260/24420 [2:10:58<7:52:46,  1.48s/it]
 22%|██▏       | 5261/24420 [2:10:59<7:52:53,  1.48s/it]
 22%|██▏       | 5262/24420 [2:11:01<7:53:32,  1.48s/it]
 22%|██▏       | 5263/24420 [2:11:02<7:53:08,  1.48s/it]
 22%|██▏       | 5264/24420 [2:11:04<7:53:29,  1.48s/it]
 22%|██▏       | 5265/24420 [2:11:05<7:53:01,  1.48s/it]
                                                        
{'loss': 0.1438, 'learning_rate': 5e-05, 'epoch': 1.08}

 22%|██▏       | 5265/24420 [2:11:05<7:53:01,  1.48s/it]
 22%|██▏       | 5266/24420 [2:11:07<7:53:39,  1.48s/it]
 22%|██▏       | 5267/24420 [2:11:08<7:52:51,  1.48s/it]
 22%|██▏       | 5268/24420 [2:11:09<7:53:10,  1.48s/it]
 22%|██▏       | 5269/24420 [2:11:11<7:53:32,  1.48s/it]
 22%|██▏       | 5270/24420 [2:11:12<7:53:04,  1.48s/it]
                                                        
{'loss': 0.1434, 'learning_rate': 5e-05, 'epoch': 1.08}

 22%|██▏       | 5270/24420 [2:11:12<7:53:04,  1.48s/it]
 22%|██▏       | 5271/24420 [2:11:14<7:53:04,  1.48s/it]
 22%|██▏       | 5272/24420 [2:11:15<7:53:22,  1.48s/it]
 22%|██▏       | 5273/24420 [2:11:17<7:54:20,  1.49s/it]
 22%|██▏       | 5274/24420 [2:11:18<7:54:42,  1.49s/it]
 22%|██▏       | 5275/24420 [2:11:20<7:53:58,  1.49s/it]
                                                        
{'loss': 0.146, 'learning_rate': 5e-05, 'epoch': 1.08}

 22%|██▏       | 5275/24420 [2:11:20<7:53:58,  1.49s/it]
 22%|██▏       | 5276/24420 [2:11:21<7:53:56,  1.49s/it]
 22%|██▏       | 5277/24420 [2:11:23<7:54:31,  1.49s/it]
 22%|██▏       | 5278/24420 [2:11:24<7:54:33,  1.49s/it]
 22%|██▏       | 5279/24420 [2:11:26<7:53:49,  1.49s/it]
 22%|██▏       | 5280/24420 [2:11:27<7:54:55,  1.49s/it]
                                                        
{'loss': 0.1381, 'learning_rate': 5e-05, 'epoch': 1.08}

 22%|██▏       | 5280/24420 [2:11:27<7:54:55,  1.49s/it]
 22%|██▏       | 5281/24420 [2:11:29<7:54:50,  1.49s/it]
 22%|██▏       | 5282/24420 [2:11:30<7:54:10,  1.49s/it]
 22%|██▏       | 5283/24420 [2:11:32<7:56:21,  1.49s/it]
 22%|██▏       | 5284/24420 [2:11:33<7:56:16,  1.49s/it]
 22%|██▏       | 5285/24420 [2:11:35<7:55:41,  1.49s/it]
                                                        
{'loss': 0.1441, 'learning_rate': 5e-05, 'epoch': 1.08}

 22%|██▏       | 5285/24420 [2:11:35<7:55:41,  1.49s/it]
 22%|██▏       | 5286/24420 [2:11:36<7:55:04,  1.49s/it]
 22%|██▏       | 5287/24420 [2:11:38<7:54:30,  1.49s/it]
 22%|██▏       | 5288/24420 [2:11:39<7:55:17,  1.49s/it]
 22%|██▏       | 5289/24420 [2:11:41<7:53:38,  1.49s/it]
 22%|██▏       | 5290/24420 [2:11:42<7:53:20,  1.48s/it]
                                                        
{'loss': 0.1367, 'learning_rate': 5e-05, 'epoch': 1.08}

 22%|██▏       | 5290/24420 [2:11:42<7:53:20,  1.48s/it]
 22%|██▏       | 5291/24420 [2:11:44<7:53:52,  1.49s/it]
 22%|██▏       | 5292/24420 [2:11:45<7:53:57,  1.49s/it]
 22%|██▏       | 5293/24420 [2:11:47<7:54:08,  1.49s/it]
 22%|██▏       | 5294/24420 [2:11:48<7:55:16,  1.49s/it]
 22%|██▏       | 5295/24420 [2:11:50<7:55:36,  1.49s/it]
                                                        
{'loss': 0.1481, 'learning_rate': 5e-05, 'epoch': 1.08}

 22%|██▏       | 5295/24420 [2:11:50<7:55:36,  1.49s/it]
 22%|██▏       | 5296/24420 [2:11:51<7:55:36,  1.49s/it]
 22%|██▏       | 5297/24420 [2:11:53<7:55:19,  1.49s/it]
 22%|██▏       | 5298/24420 [2:11:54<7:54:43,  1.49s/it]
 22%|██▏       | 5299/24420 [2:11:56<7:54:47,  1.49s/it]
 22%|██▏       | 5300/24420 [2:11:57<7:54:59,  1.49s/it]
                                                        
{'loss': 0.1715, 'learning_rate': 5e-05, 'epoch': 1.09}

 22%|██▏       | 5300/24420 [2:11:57<7:54:59,  1.49s/it]
 22%|██▏       | 5301/24420 [2:11:59<7:54:22,  1.49s/it]
 22%|██▏       | 5302/24420 [2:12:00<7:54:09,  1.49s/it]
 22%|██▏       | 5303/24420 [2:12:02<7:53:59,  1.49s/it]
 22%|██▏       | 5304/24420 [2:12:03<7:54:24,  1.49s/it]
 22%|██▏       | 5305/24420 [2:12:05<7:53:49,  1.49s/it]
                                                        
{'loss': 0.1435, 'learning_rate': 5e-05, 'epoch': 1.09}

 22%|██▏       | 5305/24420 [2:12:05<7:53:49,  1.49s/it]
 22%|██▏       | 5306/24420 [2:12:06<7:53:30,  1.49s/it]
 22%|██▏       | 5307/24420 [2:12:08<7:53:23,  1.49s/it]
 22%|██▏       | 5308/24420 [2:12:09<7:52:55,  1.48s/it]
 22%|██▏       | 5309/24420 [2:12:11<7:53:07,  1.49s/it]
 22%|██▏       | 5310/24420 [2:12:12<7:53:46,  1.49s/it]
                                                        
{'loss': 0.1307, 'learning_rate': 5e-05, 'epoch': 1.09}

 22%|██▏       | 5310/24420 [2:12:12<7:53:46,  1.49s/it]
 22%|██▏       | 5311/24420 [2:12:13<7:54:52,  1.49s/it]
 22%|██▏       | 5312/24420 [2:12:15<7:54:54,  1.49s/it]
 22%|██▏       | 5313/24420 [2:12:16<7:54:46,  1.49s/it]
 22%|██▏       | 5314/24420 [2:12:18<7:54:20,  1.49s/it]
 22%|██▏       | 5315/24420 [2:12:19<7:53:22,  1.49s/it]
                                                        
{'loss': 0.1405, 'learning_rate': 5e-05, 'epoch': 1.09}

 22%|██▏       | 5315/24420 [2:12:19<7:53:22,  1.49s/it]
 22%|██▏       | 5316/24420 [2:12:21<7:53:33,  1.49s/it]
 22%|██▏       | 5317/24420 [2:12:22<7:53:25,  1.49s/it]
 22%|██▏       | 5318/24420 [2:12:24<7:53:54,  1.49s/it]
 22%|██▏       | 5319/24420 [2:12:25<7:55:21,  1.49s/it]
 22%|██▏       | 5320/24420 [2:12:27<7:54:03,  1.49s/it]
                                                        
{'loss': 0.1455, 'learning_rate': 5e-05, 'epoch': 1.09}

 22%|██▏       | 5320/24420 [2:12:27<7:54:03,  1.49s/it]
 22%|██▏       | 5321/24420 [2:12:28<7:54:24,  1.49s/it]
 22%|██▏       | 5322/24420 [2:12:30<7:52:52,  1.49s/it]
 22%|██▏       | 5323/24420 [2:12:31<7:52:27,  1.48s/it]
 22%|██▏       | 5324/24420 [2:12:33<7:52:42,  1.49s/it]
 22%|██▏       | 5325/24420 [2:12:34<7:53:30,  1.49s/it]
                                                        
{'loss': 0.1477, 'learning_rate': 5e-05, 'epoch': 1.09}

 22%|██▏       | 5325/24420 [2:12:34<7:53:30,  1.49s/it]
 22%|██▏       | 5326/24420 [2:12:36<7:55:41,  1.49s/it]
 22%|██▏       | 5327/24420 [2:12:37<7:54:09,  1.49s/it]
 22%|██▏       | 5328/24420 [2:12:39<7:53:03,  1.49s/it]
 22%|██▏       | 5329/24420 [2:12:40<7:52:30,  1.49s/it]
 22%|██▏       | 5330/24420 [2:12:42<7:52:16,  1.48s/it]
                                                        
{'loss': 0.1308, 'learning_rate': 5e-05, 'epoch': 1.09}

 22%|██▏       | 5330/24420 [2:12:42<7:52:16,  1.48s/it]
 22%|██▏       | 5331/24420 [2:12:43<7:51:40,  1.48s/it]
 22%|██▏       | 5332/24420 [2:12:45<7:50:57,  1.48s/it]
 22%|██▏       | 5333/24420 [2:12:46<7:51:12,  1.48s/it]
 22%|██▏       | 5334/24420 [2:12:48<7:51:13,  1.48s/it]
 22%|██▏       | 5335/24420 [2:12:49<7:53:06,  1.49s/it]
                                                        
{'loss': 0.1472, 'learning_rate': 5e-05, 'epoch': 1.09}

 22%|██▏       | 5335/24420 [2:12:49<7:53:06,  1.49s/it]
 22%|██▏       | 5336/24420 [2:12:51<7:53:55,  1.49s/it]
 22%|██▏       | 5337/24420 [2:12:52<7:53:48,  1.49s/it]
 22%|██▏       | 5338/24420 [2:12:54<7:53:15,  1.49s/it]
 22%|██▏       | 5339/24420 [2:12:55<7:52:44,  1.49s/it]
 22%|██▏       | 5340/24420 [2:12:57<7:52:06,  1.48s/it]
                                                        
{'loss': 0.119, 'learning_rate': 5e-05, 'epoch': 1.09}

 22%|██▏       | 5340/24420 [2:12:57<7:52:06,  1.48s/it]
 22%|██▏       | 5341/24420 [2:12:58<7:52:07,  1.48s/it]
 22%|██▏       | 5342/24420 [2:13:00<7:51:13,  1.48s/it]
 22%|██▏       | 5343/24420 [2:13:01<7:52:09,  1.49s/it]
 22%|██▏       | 5344/24420 [2:13:03<7:53:02,  1.49s/it]
 22%|██▏       | 5345/24420 [2:13:04<7:53:00,  1.49s/it]
                                                        
{'loss': 0.1299, 'learning_rate': 5e-05, 'epoch': 1.09}

 22%|██▏       | 5345/24420 [2:13:04<7:53:00,  1.49s/it]
 22%|██▏       | 5346/24420 [2:13:06<7:53:02,  1.49s/it]
 22%|██▏       | 5347/24420 [2:13:07<7:53:33,  1.49s/it]
 22%|██▏       | 5348/24420 [2:13:09<7:52:10,  1.49s/it]
 22%|██▏       | 5349/24420 [2:13:10<7:51:38,  1.48s/it]
 22%|██▏       | 5350/24420 [2:13:11<7:51:15,  1.48s/it]
                                                        
{'loss': 0.128, 'learning_rate': 5e-05, 'epoch': 1.1}

 22%|██▏       | 5350/24420 [2:13:11<7:51:15,  1.48s/it]
 22%|██▏       | 5351/24420 [2:13:13<7:51:33,  1.48s/it]
 22%|██▏       | 5352/24420 [2:13:14<7:51:38,  1.48s/it]
 22%|██▏       | 5353/24420 [2:13:16<7:50:06,  1.48s/it]
 22%|██▏       | 5354/24420 [2:13:17<7:50:20,  1.48s/it]
 22%|██▏       | 5355/24420 [2:13:19<7:50:27,  1.48s/it]
                                                        
{'loss': 0.1354, 'learning_rate': 5e-05, 'epoch': 1.1}

 22%|██▏       | 5355/24420 [2:13:19<7:50:27,  1.48s/it]
 22%|██▏       | 5356/24420 [2:13:20<7:50:31,  1.48s/it]
 22%|██▏       | 5357/24420 [2:13:22<7:50:04,  1.48s/it]
 22%|██▏       | 5358/24420 [2:13:23<7:49:45,  1.48s/it]
 22%|██▏       | 5359/24420 [2:13:25<7:49:36,  1.48s/it]
 22%|██▏       | 5360/24420 [2:13:26<7:49:06,  1.48s/it]
                                                        
{'loss': 0.1332, 'learning_rate': 5e-05, 'epoch': 1.1}

 22%|██▏       | 5360/24420 [2:13:26<7:49:06,  1.48s/it]
 22%|██▏       | 5361/24420 [2:13:28<7:49:20,  1.48s/it]
 22%|██▏       | 5362/24420 [2:13:29<7:50:06,  1.48s/it]
 22%|██▏       | 5363/24420 [2:13:31<7:50:41,  1.48s/it]
 22%|██▏       | 5364/24420 [2:13:32<7:50:42,  1.48s/it]
 22%|██▏       | 5365/24420 [2:13:34<7:51:20,  1.48s/it]
                                                        
{'loss': 0.1667, 'learning_rate': 5e-05, 'epoch': 1.1}

 22%|██▏       | 5365/24420 [2:13:34<7:51:20,  1.48s/it]
 22%|██▏       | 5366/24420 [2:13:35<7:51:32,  1.48s/it]
 22%|██▏       | 5367/24420 [2:13:37<7:51:20,  1.48s/it]
 22%|██▏       | 5368/24420 [2:13:38<7:51:22,  1.48s/it]
 22%|██▏       | 5369/24420 [2:13:40<7:50:55,  1.48s/it]
 22%|██▏       | 5370/24420 [2:13:41<7:49:59,  1.48s/it]
                                                        
{'loss': 0.1544, 'learning_rate': 5e-05, 'epoch': 1.1}

 22%|██▏       | 5370/24420 [2:13:41<7:49:59,  1.48s/it]
 22%|██▏       | 5371/24420 [2:13:43<7:50:39,  1.48s/it]
 22%|██▏       | 5372/24420 [2:13:44<7:50:16,  1.48s/it]
 22%|██▏       | 5373/24420 [2:13:46<7:50:30,  1.48s/it]
 22%|██▏       | 5374/24420 [2:13:47<7:50:09,  1.48s/it]
 22%|██▏       | 5375/24420 [2:13:48<7:50:19,  1.48s/it]
                                                        
{'loss': 0.1509, 'learning_rate': 5e-05, 'epoch': 1.1}

 22%|██▏       | 5375/24420 [2:13:48<7:50:19,  1.48s/it]
 22%|██▏       | 5376/24420 [2:13:50<7:50:49,  1.48s/it]
 22%|██▏       | 5377/24420 [2:13:51<7:51:30,  1.49s/it]
 22%|██▏       | 5378/24420 [2:13:53<7:51:49,  1.49s/it]
 22%|██▏       | 5379/24420 [2:13:54<7:51:16,  1.49s/it]
 22%|██▏       | 5380/24420 [2:13:56<7:51:08,  1.48s/it]
                                                        
{'loss': 0.1408, 'learning_rate': 5e-05, 'epoch': 1.1}

 22%|██▏       | 5380/24420 [2:13:56<7:51:08,  1.48s/it]
 22%|██▏       | 5381/24420 [2:13:57<7:52:01,  1.49s/it]
 22%|██▏       | 5382/24420 [2:13:59<7:51:41,  1.49s/it]
 22%|██▏       | 5383/24420 [2:14:00<7:52:00,  1.49s/it]
 22%|██▏       | 5384/24420 [2:14:02<7:51:33,  1.49s/it]
 22%|██▏       | 5385/24420 [2:14:03<7:52:12,  1.49s/it]
                                                        
{'loss': 0.1545, 'learning_rate': 5e-05, 'epoch': 1.1}

 22%|██▏       | 5385/24420 [2:14:03<7:52:12,  1.49s/it]
 22%|██▏       | 5386/24420 [2:14:05<7:52:36,  1.49s/it]
 22%|██▏       | 5387/24420 [2:14:06<7:52:03,  1.49s/it]
 22%|██▏       | 5388/24420 [2:14:08<7:51:50,  1.49s/it]
 22%|██▏       | 5389/24420 [2:14:09<7:51:00,  1.48s/it]
 22%|██▏       | 5390/24420 [2:14:11<7:50:28,  1.48s/it]
                                                        
{'loss': 0.1741, 'learning_rate': 5e-05, 'epoch': 1.1}

 22%|██▏       | 5390/24420 [2:14:11<7:50:28,  1.48s/it]
 22%|██▏       | 5391/24420 [2:14:12<7:51:04,  1.49s/it]
 22%|██▏       | 5392/24420 [2:14:14<7:50:11,  1.48s/it]
 22%|██▏       | 5393/24420 [2:14:15<7:50:08,  1.48s/it]
 22%|██▏       | 5394/24420 [2:14:17<7:50:02,  1.48s/it]
 22%|██▏       | 5395/24420 [2:14:18<7:50:00,  1.48s/it]
                                                        
{'loss': 0.1421, 'learning_rate': 5e-05, 'epoch': 1.1}

 22%|██▏       | 5395/24420 [2:14:18<7:50:00,  1.48s/it]
 22%|██▏       | 5396/24420 [2:14:20<7:50:13,  1.48s/it]
 22%|██▏       | 5397/24420 [2:14:21<7:50:03,  1.48s/it]
 22%|██▏       | 5398/24420 [2:14:23<7:50:35,  1.48s/it]
 22%|██▏       | 5399/24420 [2:14:24<7:49:41,  1.48s/it]
 22%|██▏       | 5400/24420 [2:14:26<7:49:50,  1.48s/it]
                                                        
{'loss': 0.1257, 'learning_rate': 5e-05, 'epoch': 1.11}

 22%|██▏       | 5400/24420 [2:14:26<7:49:50,  1.48s/it]
 22%|██▏       | 5401/24420 [2:14:32<15:18:01,  2.90s/it]
 22%|██▏       | 5402/24420 [2:14:33<13:02:42,  2.47s/it]
 22%|██▏       | 5403/24420 [2:14:35<11:28:09,  2.17s/it]
 22%|██▏       | 5404/24420 [2:14:36<10:22:37,  1.96s/it]
 22%|██▏       | 5405/24420 [2:14:38<9:36:46,  1.82s/it] 
                                                        
{'loss': 0.1235, 'learning_rate': 5e-05, 'epoch': 1.11}

 22%|██▏       | 5405/24420 [2:14:38<9:36:46,  1.82s/it]
 22%|██▏       | 5406/24420 [2:14:39<9:04:40,  1.72s/it]
 22%|██▏       | 5407/24420 [2:14:41<8:41:54,  1.65s/it]
 22%|██▏       | 5408/24420 [2:14:42<8:26:25,  1.60s/it]
 22%|██▏       | 5409/24420 [2:14:44<8:16:15,  1.57s/it]
 22%|██▏       | 5410/24420 [2:14:45<8:07:24,  1.54s/it]
                                                        
{'loss': 0.1362, 'learning_rate': 5e-05, 'epoch': 1.11}

 22%|██▏       | 5410/24420 [2:14:45<8:07:24,  1.54s/it]
 22%|██▏       | 5411/24420 [2:14:47<8:01:38,  1.52s/it]
 22%|██▏       | 5412/24420 [2:14:48<7:57:33,  1.51s/it]
 22%|██▏       | 5413/24420 [2:14:50<7:55:18,  1.50s/it]
 22%|██▏       | 5414/24420 [2:14:51<7:53:05,  1.49s/it]
 22%|██▏       | 5415/24420 [2:14:53<7:51:44,  1.49s/it]
                                                        
{'loss': 0.1453, 'learning_rate': 5e-05, 'epoch': 1.11}

 22%|██▏       | 5415/24420 [2:14:53<7:51:44,  1.49s/it]
 22%|██▏       | 5416/24420 [2:14:54<7:51:13,  1.49s/it]
 22%|██▏       | 5417/24420 [2:14:55<7:49:44,  1.48s/it]
 22%|██▏       | 5418/24420 [2:14:57<7:49:17,  1.48s/it]
 22%|██▏       | 5419/24420 [2:14:58<7:49:21,  1.48s/it]
 22%|██▏       | 5420/24420 [2:15:00<7:49:19,  1.48s/it]
                                                        
{'loss': 0.1412, 'learning_rate': 5e-05, 'epoch': 1.11}

 22%|██▏       | 5420/24420 [2:15:00<7:49:19,  1.48s/it]
 22%|██▏       | 5421/24420 [2:15:01<7:49:35,  1.48s/it]
 22%|██▏       | 5422/24420 [2:15:03<7:49:09,  1.48s/it]
 22%|██▏       | 5423/24420 [2:15:04<7:48:22,  1.48s/it]
 22%|██▏       | 5424/24420 [2:15:06<7:48:22,  1.48s/it]
 22%|██▏       | 5425/24420 [2:15:07<7:48:45,  1.48s/it]
                                                        
{'loss': 0.1403, 'learning_rate': 5e-05, 'epoch': 1.11}

 22%|██▏       | 5425/24420 [2:15:07<7:48:45,  1.48s/it]
 22%|██▏       | 5426/24420 [2:15:09<7:49:10,  1.48s/it]
 22%|██▏       | 5427/24420 [2:15:10<7:49:05,  1.48s/it]
 22%|██▏       | 5428/24420 [2:15:12<7:49:33,  1.48s/it]
 22%|██▏       | 5429/24420 [2:15:13<7:49:38,  1.48s/it]
 22%|██▏       | 5430/24420 [2:15:15<7:49:15,  1.48s/it]
                                                        
{'loss': 0.1566, 'learning_rate': 5e-05, 'epoch': 1.11}

 22%|██▏       | 5430/24420 [2:15:15<7:49:15,  1.48s/it]
 22%|██▏       | 5431/24420 [2:15:16<7:49:29,  1.48s/it]
 22%|██▏       | 5432/24420 [2:15:18<7:49:52,  1.48s/it]
 22%|██▏       | 5433/24420 [2:15:19<7:49:42,  1.48s/it]
 22%|██▏       | 5434/24420 [2:15:21<7:48:48,  1.48s/it]
 22%|██▏       | 5435/24420 [2:15:22<7:48:38,  1.48s/it]
                                                        
{'loss': 0.1513, 'learning_rate': 5e-05, 'epoch': 1.11}

 22%|██▏       | 5435/24420 [2:15:22<7:48:38,  1.48s/it]
 22%|██▏       | 5436/24420 [2:15:24<7:49:21,  1.48s/it]
 22%|██▏       | 5437/24420 [2:15:25<7:48:54,  1.48s/it]
 22%|██▏       | 5438/24420 [2:15:27<8:15:23,  1.57s/it]
 22%|██▏       | 5439/24420 [2:15:28<8:06:48,  1.54s/it]
 22%|██▏       | 5440/24420 [2:15:30<8:01:07,  1.52s/it]
                                                        
{'loss': 0.1726, 'learning_rate': 5e-05, 'epoch': 1.11}

 22%|██▏       | 5440/24420 [2:15:30<8:01:07,  1.52s/it]
 22%|██▏       | 5441/24420 [2:15:31<7:57:24,  1.51s/it]
 22%|██▏       | 5442/24420 [2:15:33<7:54:48,  1.50s/it]
 22%|██▏       | 5443/24420 [2:15:34<7:52:54,  1.50s/it]
 22%|██▏       | 5444/24420 [2:15:36<7:51:47,  1.49s/it]
 22%|██▏       | 5445/24420 [2:15:37<7:50:15,  1.49s/it]
                                                        
{'loss': 0.1428, 'learning_rate': 5e-05, 'epoch': 1.11}

 22%|██▏       | 5445/24420 [2:15:37<7:50:15,  1.49s/it]
 22%|██▏       | 5446/24420 [2:15:39<7:50:06,  1.49s/it]
 22%|██▏       | 5447/24420 [2:15:40<7:50:06,  1.49s/it]
 22%|██▏       | 5448/24420 [2:15:42<7:49:19,  1.48s/it]
 22%|██▏       | 5449/24420 [2:15:43<7:49:01,  1.48s/it]
 22%|██▏       | 5450/24420 [2:15:45<7:48:09,  1.48s/it]
                                                        
{'loss': 0.1377, 'learning_rate': 5e-05, 'epoch': 1.12}

 22%|██▏       | 5450/24420 [2:15:45<7:48:09,  1.48s/it]
 22%|██▏       | 5451/24420 [2:15:46<7:48:15,  1.48s/it]
 22%|██▏       | 5452/24420 [2:15:48<7:47:27,  1.48s/it]
 22%|██▏       | 5453/24420 [2:15:49<7:47:31,  1.48s/it]
 22%|██▏       | 5454/24420 [2:15:51<7:47:44,  1.48s/it]
 22%|██▏       | 5455/24420 [2:15:52<7:47:50,  1.48s/it]
                                                        
{'loss': 0.1378, 'learning_rate': 5e-05, 'epoch': 1.12}

 22%|██▏       | 5455/24420 [2:15:52<7:47:50,  1.48s/it]
 22%|██▏       | 5456/24420 [2:15:54<7:48:06,  1.48s/it]
 22%|██▏       | 5457/24420 [2:15:55<7:47:27,  1.48s/it]
 22%|██▏       | 5458/24420 [2:15:56<7:47:29,  1.48s/it]
 22%|██▏       | 5459/24420 [2:15:58<7:47:22,  1.48s/it]
 22%|██▏       | 5460/24420 [2:15:59<7:47:03,  1.48s/it]
                                                        
{'loss': 0.1299, 'learning_rate': 5e-05, 'epoch': 1.12}

 22%|██▏       | 5460/24420 [2:15:59<7:47:03,  1.48s/it]
 22%|██▏       | 5461/24420 [2:16:01<7:47:23,  1.48s/it]
 22%|██▏       | 5462/24420 [2:16:02<7:47:23,  1.48s/it]
 22%|██▏       | 5463/24420 [2:16:04<7:47:26,  1.48s/it]
 22%|██▏       | 5464/24420 [2:16:05<7:47:28,  1.48s/it]
 22%|██▏       | 5465/24420 [2:16:07<7:46:42,  1.48s/it]
                                                        
{'loss': 0.1132, 'learning_rate': 5e-05, 'epoch': 1.12}

 22%|██▏       | 5465/24420 [2:16:07<7:46:42,  1.48s/it]
 22%|██▏       | 5466/24420 [2:16:08<7:46:39,  1.48s/it]
 22%|██▏       | 5467/24420 [2:16:10<7:48:02,  1.48s/it]
 22%|██▏       | 5468/24420 [2:16:11<7:48:19,  1.48s/it]
 22%|██▏       | 5469/24420 [2:16:13<7:47:16,  1.48s/it]
 22%|██▏       | 5470/24420 [2:16:14<7:46:51,  1.48s/it]
                                                        
{'loss': 0.1313, 'learning_rate': 5e-05, 'epoch': 1.12}

 22%|██▏       | 5470/24420 [2:16:14<7:46:51,  1.48s/it]
 22%|██▏       | 5471/24420 [2:16:16<7:47:27,  1.48s/it]
 22%|██▏       | 5472/24420 [2:16:17<7:47:41,  1.48s/it]
 22%|██▏       | 5473/24420 [2:16:19<7:47:23,  1.48s/it]
 22%|██▏       | 5474/24420 [2:16:20<7:47:20,  1.48s/it]
 22%|██▏       | 5475/24420 [2:16:22<7:47:19,  1.48s/it]
                                                        
{'loss': 0.1613, 'learning_rate': 5e-05, 'epoch': 1.12}

 22%|██▏       | 5475/24420 [2:16:22<7:47:19,  1.48s/it]
 22%|██▏       | 5476/24420 [2:16:23<7:47:19,  1.48s/it]
 22%|██▏       | 5477/24420 [2:16:25<7:47:16,  1.48s/it]
 22%|██▏       | 5478/24420 [2:16:26<7:46:56,  1.48s/it]
 22%|██▏       | 5479/24420 [2:16:28<7:47:19,  1.48s/it]
 22%|██▏       | 5480/24420 [2:16:29<7:47:16,  1.48s/it]
                                                        
{'loss': 0.1128, 'learning_rate': 5e-05, 'epoch': 1.12}

 22%|██▏       | 5480/24420 [2:16:29<7:47:16,  1.48s/it]
 22%|██▏       | 5481/24420 [2:16:31<7:47:19,  1.48s/it]
 22%|██▏       | 5482/24420 [2:16:32<7:46:40,  1.48s/it]
 22%|██▏       | 5483/24420 [2:16:33<7:46:09,  1.48s/it]
 22%|██▏       | 5484/24420 [2:16:35<7:46:05,  1.48s/it]
 22%|██▏       | 5485/24420 [2:16:36<7:45:59,  1.48s/it]
                                                        
{'loss': 0.1321, 'learning_rate': 5e-05, 'epoch': 1.12}

 22%|██▏       | 5485/24420 [2:16:36<7:45:59,  1.48s/it]
 22%|██▏       | 5486/24420 [2:16:38<7:46:29,  1.48s/it]
 22%|██▏       | 5487/24420 [2:16:39<7:47:18,  1.48s/it]
 22%|██▏       | 5488/24420 [2:16:41<7:46:56,  1.48s/it]
 22%|██▏       | 5489/24420 [2:16:42<7:46:39,  1.48s/it]
 22%|██▏       | 5490/24420 [2:16:44<7:46:19,  1.48s/it]
                                                        
{'loss': 0.104, 'learning_rate': 5e-05, 'epoch': 1.12}

 22%|██▏       | 5490/24420 [2:16:44<7:46:19,  1.48s/it]
 22%|██▏       | 5491/24420 [2:16:45<7:46:05,  1.48s/it]
 22%|██▏       | 5492/24420 [2:16:47<7:46:12,  1.48s/it]
 22%|██▏       | 5493/24420 [2:16:48<7:46:48,  1.48s/it]
 22%|██▏       | 5494/24420 [2:16:50<7:45:52,  1.48s/it]
 23%|██▎       | 5495/24420 [2:16:51<7:45:55,  1.48s/it]
                                                        
{'loss': 0.1409, 'learning_rate': 5e-05, 'epoch': 1.13}

 23%|██▎       | 5495/24420 [2:16:51<7:45:55,  1.48s/it]
 23%|██▎       | 5496/24420 [2:16:53<7:47:01,  1.48s/it]
 23%|██▎       | 5497/24420 [2:16:54<7:46:38,  1.48s/it]
 23%|██▎       | 5498/24420 [2:16:56<7:46:55,  1.48s/it]
 23%|██▎       | 5499/24420 [2:16:57<7:46:51,  1.48s/it]
 23%|██▎       | 5500/24420 [2:16:59<7:47:24,  1.48s/it]
                                                        
{'loss': 0.1685, 'learning_rate': 5e-05, 'epoch': 1.13}

 23%|██▎       | 5500/24420 [2:16:59<7:47:24,  1.48s/it]
 23%|██▎       | 5501/24420 [2:17:00<7:47:06,  1.48s/it]
 23%|██▎       | 5502/24420 [2:17:02<7:46:37,  1.48s/it]
 23%|██▎       | 5503/24420 [2:17:03<7:46:12,  1.48s/it]
 23%|██▎       | 5504/24420 [2:17:05<7:45:34,  1.48s/it]
 23%|██▎       | 5505/24420 [2:17:06<7:46:18,  1.48s/it]
                                                        
{'loss': 0.1483, 'learning_rate': 5e-05, 'epoch': 1.13}

 23%|██▎       | 5505/24420 [2:17:06<7:46:18,  1.48s/it]
 23%|██▎       | 5506/24420 [2:17:08<7:46:32,  1.48s/it]
 23%|██▎       | 5507/24420 [2:17:09<7:45:58,  1.48s/it]
 23%|██▎       | 5508/24420 [2:17:10<7:46:06,  1.48s/it]
 23%|██▎       | 5509/24420 [2:17:12<7:46:18,  1.48s/it]
 23%|██▎       | 5510/24420 [2:17:13<7:47:01,  1.48s/it]
                                                        
{'loss': 0.1402, 'learning_rate': 5e-05, 'epoch': 1.13}

 23%|██▎       | 5510/24420 [2:17:13<7:47:01,  1.48s/it]
 23%|██▎       | 5511/24420 [2:17:15<7:47:45,  1.48s/it]
 23%|██▎       | 5512/24420 [2:17:16<7:47:30,  1.48s/it]
 23%|██▎       | 5513/24420 [2:17:18<7:46:53,  1.48s/it]
 23%|██▎       | 5514/24420 [2:17:19<7:46:40,  1.48s/it]
 23%|██▎       | 5515/24420 [2:17:21<7:46:58,  1.48s/it]
                                                        
{'loss': 0.1203, 'learning_rate': 5e-05, 'epoch': 1.13}

 23%|██▎       | 5515/24420 [2:17:21<7:46:58,  1.48s/it]
 23%|██▎       | 5516/24420 [2:17:22<7:47:20,  1.48s/it]
 23%|██▎       | 5517/24420 [2:17:24<7:47:29,  1.48s/it]
 23%|██▎       | 5518/24420 [2:17:25<7:46:13,  1.48s/it]
 23%|██▎       | 5519/24420 [2:17:27<7:46:12,  1.48s/it]
 23%|██▎       | 5520/24420 [2:17:28<7:46:29,  1.48s/it]
                                                        
{'loss': 0.1261, 'learning_rate': 5e-05, 'epoch': 1.13}

 23%|██▎       | 5520/24420 [2:17:28<7:46:29,  1.48s/it]
 23%|██▎       | 5521/24420 [2:17:30<7:46:27,  1.48s/it]
 23%|██▎       | 5522/24420 [2:17:31<7:46:21,  1.48s/it]
 23%|██▎       | 5523/24420 [2:17:33<7:45:53,  1.48s/it]
 23%|██▎       | 5524/24420 [2:17:34<7:45:33,  1.48s/it]
 23%|██▎       | 5525/24420 [2:17:36<7:45:26,  1.48s/it]
                                                        
{'loss': 0.1311, 'learning_rate': 5e-05, 'epoch': 1.13}

 23%|██▎       | 5525/24420 [2:17:36<7:45:26,  1.48s/it]
 23%|██▎       | 5526/24420 [2:17:37<7:45:51,  1.48s/it]
 23%|██▎       | 5527/24420 [2:17:39<7:46:18,  1.48s/it]
 23%|██▎       | 5528/24420 [2:17:40<7:45:12,  1.48s/it]
 23%|██▎       | 5529/24420 [2:17:42<7:44:46,  1.48s/it]
 23%|██▎       | 5530/24420 [2:17:43<7:44:46,  1.48s/it]
                                                        
{'loss': 0.138, 'learning_rate': 5e-05, 'epoch': 1.13}

 23%|██▎       | 5530/24420 [2:17:43<7:44:46,  1.48s/it]
 23%|██▎       | 5531/24420 [2:17:45<7:45:27,  1.48s/it]
 23%|██▎       | 5532/24420 [2:17:46<7:46:37,  1.48s/it]
 23%|██▎       | 5533/24420 [2:17:47<7:46:42,  1.48s/it]
 23%|██▎       | 5534/24420 [2:17:49<7:45:41,  1.48s/it]
 23%|██▎       | 5535/24420 [2:17:50<7:45:14,  1.48s/it]
                                                        
{'loss': 0.1541, 'learning_rate': 5e-05, 'epoch': 1.13}

 23%|██▎       | 5535/24420 [2:17:50<7:45:14,  1.48s/it]
 23%|██▎       | 5536/24420 [2:17:52<7:46:39,  1.48s/it]
 23%|██▎       | 5537/24420 [2:17:53<7:46:29,  1.48s/it]
 23%|██▎       | 5538/24420 [2:17:55<7:46:32,  1.48s/it]
 23%|██▎       | 5539/24420 [2:17:56<7:46:37,  1.48s/it]
 23%|██▎       | 5540/24420 [2:17:58<7:45:30,  1.48s/it]
                                                        
{'loss': 0.1345, 'learning_rate': 5e-05, 'epoch': 1.13}

 23%|██▎       | 5540/24420 [2:17:58<7:45:30,  1.48s/it]
 23%|██▎       | 5541/24420 [2:17:59<7:45:20,  1.48s/it]
 23%|██▎       | 5542/24420 [2:18:01<7:45:37,  1.48s/it]
 23%|██▎       | 5543/24420 [2:18:02<7:46:19,  1.48s/it]
 23%|██▎       | 5544/24420 [2:18:04<7:46:43,  1.48s/it]
 23%|██▎       | 5545/24420 [2:18:05<7:45:53,  1.48s/it]
                                                        
{'loss': 0.1228, 'learning_rate': 5e-05, 'epoch': 1.14}

 23%|██▎       | 5545/24420 [2:18:05<7:45:53,  1.48s/it]
 23%|██▎       | 5546/24420 [2:18:07<7:46:02,  1.48s/it]
 23%|██▎       | 5547/24420 [2:18:08<7:46:08,  1.48s/it]
 23%|██▎       | 5548/24420 [2:18:10<7:46:00,  1.48s/it]
 23%|██▎       | 5549/24420 [2:18:11<7:46:30,  1.48s/it]
 23%|██▎       | 5550/24420 [2:18:13<7:46:40,  1.48s/it]
                                                        
{'loss': 0.1232, 'learning_rate': 5e-05, 'epoch': 1.14}

 23%|██▎       | 5550/24420 [2:18:13<7:46:40,  1.48s/it]
 23%|██▎       | 5551/24420 [2:18:14<7:46:37,  1.48s/it]
 23%|██▎       | 5552/24420 [2:18:16<7:46:24,  1.48s/it]
 23%|██▎       | 5553/24420 [2:18:17<7:46:25,  1.48s/it]
 23%|██▎       | 5554/24420 [2:18:19<7:45:11,  1.48s/it]
 23%|██▎       | 5555/24420 [2:18:20<7:45:13,  1.48s/it]
                                                        
{'loss': 0.1412, 'learning_rate': 5e-05, 'epoch': 1.14}

 23%|██▎       | 5555/24420 [2:18:20<7:45:13,  1.48s/it]
 23%|██▎       | 5556/24420 [2:18:22<7:45:12,  1.48s/it]
 23%|██▎       | 5557/24420 [2:18:23<7:45:02,  1.48s/it]
 23%|██▎       | 5558/24420 [2:18:25<7:45:06,  1.48s/it]
 23%|██▎       | 5559/24420 [2:18:26<7:45:14,  1.48s/it]
 23%|██▎       | 5560/24420 [2:18:27<7:45:10,  1.48s/it]
                                                        
{'loss': 0.1209, 'learning_rate': 5e-05, 'epoch': 1.14}

 23%|██▎       | 5560/24420 [2:18:27<7:45:10,  1.48s/it]
 23%|██▎       | 5561/24420 [2:18:29<7:45:17,  1.48s/it]
 23%|██▎       | 5562/24420 [2:18:30<7:45:20,  1.48s/it]
 23%|██▎       | 5563/24420 [2:18:32<7:44:50,  1.48s/it]
 23%|██▎       | 5564/24420 [2:18:33<7:44:26,  1.48s/it]
 23%|██▎       | 5565/24420 [2:18:35<7:44:45,  1.48s/it]
                                                        
{'loss': 0.1235, 'learning_rate': 5e-05, 'epoch': 1.14}

 23%|██▎       | 5565/24420 [2:18:35<7:44:45,  1.48s/it]
 23%|██▎       | 5566/24420 [2:18:36<7:45:06,  1.48s/it]
 23%|██▎       | 5567/24420 [2:18:38<7:44:36,  1.48s/it]
 23%|██▎       | 5568/24420 [2:18:39<7:46:07,  1.48s/it]
 23%|██▎       | 5569/24420 [2:18:41<7:45:57,  1.48s/it]
 23%|██▎       | 5570/24420 [2:18:42<7:45:58,  1.48s/it]
                                                        
{'loss': 0.1577, 'learning_rate': 5e-05, 'epoch': 1.14}

 23%|██▎       | 5570/24420 [2:18:42<7:45:58,  1.48s/it]
 23%|██▎       | 5571/24420 [2:18:44<7:46:10,  1.48s/it]
 23%|██▎       | 5572/24420 [2:18:45<7:46:35,  1.49s/it]
 23%|██▎       | 5573/24420 [2:18:47<7:46:55,  1.49s/it]
 23%|██▎       | 5574/24420 [2:18:48<7:46:27,  1.49s/it]
 23%|██▎       | 5575/24420 [2:18:50<7:46:16,  1.48s/it]
                                                        
{'loss': 0.1342, 'learning_rate': 5e-05, 'epoch': 1.14}

 23%|██▎       | 5575/24420 [2:18:50<7:46:16,  1.48s/it]
 23%|██▎       | 5576/24420 [2:18:51<7:46:23,  1.49s/it]
 23%|██▎       | 5577/24420 [2:18:53<7:45:28,  1.48s/it]
 23%|██▎       | 5578/24420 [2:18:54<7:44:49,  1.48s/it]
 23%|██▎       | 5579/24420 [2:18:56<7:45:44,  1.48s/it]
 23%|██▎       | 5580/24420 [2:18:57<7:46:20,  1.49s/it]
                                                        
{'loss': 0.1506, 'learning_rate': 5e-05, 'epoch': 1.14}

 23%|██▎       | 5580/24420 [2:18:57<7:46:20,  1.49s/it]
 23%|██▎       | 5581/24420 [2:18:59<7:45:27,  1.48s/it]
 23%|██▎       | 5582/24420 [2:19:00<7:44:52,  1.48s/it]
 23%|██▎       | 5583/24420 [2:19:02<7:44:30,  1.48s/it]
 23%|██▎       | 5584/24420 [2:19:03<7:43:48,  1.48s/it]
 23%|██▎       | 5585/24420 [2:19:05<7:44:45,  1.48s/it]
                                                        
{'loss': 0.1258, 'learning_rate': 5e-05, 'epoch': 1.14}

 23%|██▎       | 5585/24420 [2:19:05<7:44:45,  1.48s/it]
 23%|██▎       | 5586/24420 [2:19:06<7:45:33,  1.48s/it]
 23%|██▎       | 5587/24420 [2:19:07<7:45:03,  1.48s/it]
 23%|██▎       | 5588/24420 [2:19:09<7:44:34,  1.48s/it]
 23%|██▎       | 5589/24420 [2:19:10<7:44:56,  1.48s/it]
 23%|██▎       | 5590/24420 [2:19:12<7:44:15,  1.48s/it]
                                                        
{'loss': 0.1311, 'learning_rate': 5e-05, 'epoch': 1.14}

 23%|██▎       | 5590/24420 [2:19:12<7:44:15,  1.48s/it]
 23%|██▎       | 5591/24420 [2:19:13<7:45:01,  1.48s/it]
 23%|██▎       | 5592/24420 [2:19:15<7:44:02,  1.48s/it]
 23%|██▎       | 5593/24420 [2:19:16<7:44:02,  1.48s/it]
 23%|██▎       | 5594/24420 [2:19:18<7:43:11,  1.48s/it]
 23%|██▎       | 5595/24420 [2:19:19<7:43:09,  1.48s/it]
                                                        
{'loss': 0.1398, 'learning_rate': 5e-05, 'epoch': 1.15}

 23%|██▎       | 5595/24420 [2:19:19<7:43:09,  1.48s/it]
 23%|██▎       | 5596/24420 [2:19:21<7:44:15,  1.48s/it]
 23%|██▎       | 5597/24420 [2:19:22<7:44:19,  1.48s/it]
 23%|██▎       | 5598/24420 [2:19:24<7:44:30,  1.48s/it]
 23%|██▎       | 5599/24420 [2:19:25<7:44:12,  1.48s/it]
 23%|██▎       | 5600/24420 [2:19:27<7:44:28,  1.48s/it]
                                                        
{'loss': 0.1739, 'learning_rate': 5e-05, 'epoch': 1.15}

 23%|██▎       | 5600/24420 [2:19:27<7:44:28,  1.48s/it]
 23%|██▎       | 5601/24420 [2:19:31<12:40:13,  2.42s/it]
 23%|██▎       | 5602/24420 [2:19:33<11:10:41,  2.14s/it]
 23%|██▎       | 5603/24420 [2:19:34<10:08:26,  1.94s/it]
 23%|██▎       | 5604/24420 [2:19:36<9:24:43,  1.80s/it] 
 23%|██▎       | 5605/24420 [2:19:37<8:54:50,  1.71s/it]
                                                        
{'loss': 0.1341, 'learning_rate': 5e-05, 'epoch': 1.15}

 23%|██▎       | 5605/24420 [2:19:37<8:54:50,  1.71s/it]
 23%|██▎       | 5606/24420 [2:19:39<8:34:27,  1.64s/it]
 23%|██▎       | 5607/24420 [2:19:40<8:19:21,  1.59s/it]
 23%|██▎       | 5608/24420 [2:19:42<8:08:52,  1.56s/it]
 23%|██▎       | 5609/24420 [2:19:43<8:01:02,  1.53s/it]
 23%|██▎       | 5610/24420 [2:19:45<7:55:08,  1.52s/it]
                                                        
{'loss': 0.1083, 'learning_rate': 5e-05, 'epoch': 1.15}

 23%|██▎       | 5610/24420 [2:19:45<7:55:08,  1.52s/it]
 23%|██▎       | 5611/24420 [2:19:46<7:52:01,  1.51s/it]
 23%|██▎       | 5612/24420 [2:19:48<7:50:00,  1.50s/it]
 23%|██▎       | 5613/24420 [2:19:49<7:48:38,  1.50s/it]
 23%|██▎       | 5614/24420 [2:19:51<7:47:14,  1.49s/it]
 23%|██▎       | 5615/24420 [2:19:52<7:46:24,  1.49s/it]
                                                        
{'loss': 0.1416, 'learning_rate': 5e-05, 'epoch': 1.15}

 23%|██▎       | 5615/24420 [2:19:52<7:46:24,  1.49s/it]
 23%|██▎       | 5616/24420 [2:19:54<7:46:09,  1.49s/it]
 23%|██▎       | 5617/24420 [2:19:55<7:44:59,  1.48s/it]
 23%|██▎       | 5618/24420 [2:19:57<7:46:43,  1.49s/it]
 23%|██▎       | 5619/24420 [2:19:58<7:45:39,  1.49s/it]
 23%|██▎       | 5620/24420 [2:19:59<7:45:47,  1.49s/it]
                                                        
{'loss': 0.1439, 'learning_rate': 5e-05, 'epoch': 1.15}

 23%|██▎       | 5620/24420 [2:19:59<7:45:47,  1.49s/it]
 23%|██▎       | 5621/24420 [2:20:01<7:46:09,  1.49s/it]
 23%|██▎       | 5622/24420 [2:20:02<7:46:07,  1.49s/it]
 23%|██▎       | 5623/24420 [2:20:04<7:44:59,  1.48s/it]
 23%|██▎       | 5624/24420 [2:20:05<7:43:57,  1.48s/it]
 23%|██▎       | 5625/24420 [2:20:07<7:43:54,  1.48s/it]
                                                        
{'loss': 0.1417, 'learning_rate': 5e-05, 'epoch': 1.15}

 23%|██▎       | 5625/24420 [2:20:07<7:43:54,  1.48s/it]
 23%|██▎       | 5626/24420 [2:20:08<7:44:09,  1.48s/it]
 23%|██▎       | 5627/24420 [2:20:10<7:44:45,  1.48s/it]
 23%|██▎       | 5628/24420 [2:20:11<7:45:14,  1.49s/it]
 23%|██▎       | 5629/24420 [2:20:13<7:45:28,  1.49s/it]
 23%|██▎       | 5630/24420 [2:20:14<7:45:03,  1.49s/it]
                                                        
{'loss': 0.1542, 'learning_rate': 5e-05, 'epoch': 1.15}

 23%|██▎       | 5630/24420 [2:20:14<7:45:03,  1.49s/it]
 23%|██▎       | 5631/24420 [2:20:16<7:44:01,  1.48s/it]
 23%|██▎       | 5632/24420 [2:20:17<7:43:40,  1.48s/it]
 23%|██▎       | 5633/24420 [2:20:19<7:44:38,  1.48s/it]
 23%|██▎       | 5634/24420 [2:20:20<7:43:29,  1.48s/it]
 23%|██▎       | 5635/24420 [2:20:22<7:43:57,  1.48s/it]
                                                        
{'loss': 0.125, 'learning_rate': 5e-05, 'epoch': 1.15}

 23%|██▎       | 5635/24420 [2:20:22<7:43:57,  1.48s/it]
 23%|██▎       | 5636/24420 [2:20:23<7:43:25,  1.48s/it]
 23%|██▎       | 5637/24420 [2:20:25<7:42:57,  1.48s/it]
 23%|██▎       | 5638/24420 [2:20:26<7:43:21,  1.48s/it]
 23%|██▎       | 5639/24420 [2:20:28<7:42:40,  1.48s/it]
 23%|██▎       | 5640/24420 [2:20:29<7:42:31,  1.48s/it]
                                                        
{'loss': 0.1255, 'learning_rate': 5e-05, 'epoch': 1.15}

 23%|██▎       | 5640/24420 [2:20:29<7:42:31,  1.48s/it]
 23%|██▎       | 5641/24420 [2:20:31<7:42:28,  1.48s/it]
 23%|██▎       | 5642/24420 [2:20:32<7:41:57,  1.48s/it]
 23%|██▎       | 5643/24420 [2:20:34<7:42:09,  1.48s/it]
 23%|██▎       | 5644/24420 [2:20:35<7:43:01,  1.48s/it]
 23%|██▎       | 5645/24420 [2:20:37<7:42:39,  1.48s/it]
                                                        
{'loss': 0.1201, 'learning_rate': 5e-05, 'epoch': 1.16}

 23%|██▎       | 5645/24420 [2:20:37<7:42:39,  1.48s/it]
 23%|██▎       | 5646/24420 [2:20:38<7:42:40,  1.48s/it]
 23%|██▎       | 5647/24420 [2:20:39<7:42:06,  1.48s/it]
 23%|██▎       | 5648/24420 [2:20:41<7:42:03,  1.48s/it]
 23%|██▎       | 5649/24420 [2:20:42<7:42:07,  1.48s/it]
 23%|██▎       | 5650/24420 [2:20:44<7:42:18,  1.48s/it]
                                                        
{'loss': 0.1421, 'learning_rate': 5e-05, 'epoch': 1.16}

 23%|██▎       | 5650/24420 [2:20:44<7:42:18,  1.48s/it]
 23%|██▎       | 5651/24420 [2:20:45<7:42:07,  1.48s/it]
 23%|██▎       | 5652/24420 [2:20:47<7:42:42,  1.48s/it]
 23%|██▎       | 5653/24420 [2:20:48<7:43:20,  1.48s/it]
 23%|██▎       | 5654/24420 [2:20:50<7:42:55,  1.48s/it]
 23%|██▎       | 5655/24420 [2:20:51<7:42:48,  1.48s/it]
                                                        
{'loss': 0.1236, 'learning_rate': 5e-05, 'epoch': 1.16}

 23%|██▎       | 5655/24420 [2:20:51<7:42:48,  1.48s/it]
 23%|██▎       | 5656/24420 [2:20:53<7:42:55,  1.48s/it]
 23%|██▎       | 5657/24420 [2:20:54<7:41:55,  1.48s/it]
 23%|██▎       | 5658/24420 [2:20:56<7:41:45,  1.48s/it]
 23%|██▎       | 5659/24420 [2:20:57<7:42:26,  1.48s/it]
 23%|██▎       | 5660/24420 [2:20:59<7:42:35,  1.48s/it]
                                                        
{'loss': 0.1411, 'learning_rate': 5e-05, 'epoch': 1.16}

 23%|██▎       | 5660/24420 [2:20:59<7:42:35,  1.48s/it]
 23%|██▎       | 5661/24420 [2:21:00<7:42:44,  1.48s/it]
 23%|██▎       | 5662/24420 [2:21:02<7:54:50,  1.52s/it]
 23%|██▎       | 5663/24420 [2:21:03<7:51:22,  1.51s/it]
 23%|██▎       | 5664/24420 [2:21:05<7:48:52,  1.50s/it]
 23%|██▎       | 5665/24420 [2:21:06<7:47:25,  1.50s/it]
                                                        
{'loss': 0.1575, 'learning_rate': 5e-05, 'epoch': 1.16}

 23%|██▎       | 5665/24420 [2:21:06<7:47:25,  1.50s/it]
 23%|██▎       | 5666/24420 [2:21:08<7:45:40,  1.49s/it]
 23%|██▎       | 5667/24420 [2:21:09<7:44:55,  1.49s/it]
 23%|██▎       | 5668/24420 [2:21:11<7:45:52,  1.49s/it]
 23%|██▎       | 5669/24420 [2:21:12<7:45:14,  1.49s/it]
 23%|██▎       | 5670/24420 [2:21:14<7:43:49,  1.48s/it]
                                                        
{'loss': 0.1276, 'learning_rate': 5e-05, 'epoch': 1.16}

 23%|██▎       | 5670/24420 [2:21:14<7:43:49,  1.48s/it]
 23%|██▎       | 5671/24420 [2:21:15<7:44:14,  1.49s/it]
 23%|██▎       | 5672/24420 [2:21:17<7:44:00,  1.48s/it]
 23%|██▎       | 5673/24420 [2:21:18<7:43:25,  1.48s/it]
 23%|██▎       | 5674/24420 [2:21:20<7:42:42,  1.48s/it]
 23%|██▎       | 5675/24420 [2:21:21<7:42:38,  1.48s/it]
                                                        
{'loss': 0.1617, 'learning_rate': 5e-05, 'epoch': 1.16}

 23%|██▎       | 5675/24420 [2:21:21<7:42:38,  1.48s/it]
 23%|██▎       | 5676/24420 [2:21:23<7:43:25,  1.48s/it]
 23%|██▎       | 5677/24420 [2:21:24<7:43:13,  1.48s/it]
 23%|██▎       | 5678/24420 [2:21:26<7:44:34,  1.49s/it]
 23%|██▎       | 5679/24420 [2:21:27<7:43:31,  1.48s/it]
 23%|██▎       | 5680/24420 [2:21:28<7:42:16,  1.48s/it]
                                                        
{'loss': 0.1344, 'learning_rate': 5e-05, 'epoch': 1.16}

 23%|██▎       | 5680/24420 [2:21:28<7:42:16,  1.48s/it]
 23%|██▎       | 5681/24420 [2:21:30<7:42:29,  1.48s/it]
 23%|██▎       | 5682/24420 [2:21:31<7:42:07,  1.48s/it]
 23%|██▎       | 5683/24420 [2:21:33<7:41:55,  1.48s/it]
 23%|██▎       | 5684/24420 [2:21:34<7:41:16,  1.48s/it]
 23%|██▎       | 5685/24420 [2:21:36<7:41:12,  1.48s/it]
                                                        
{'loss': 0.1142, 'learning_rate': 5e-05, 'epoch': 1.16}

 23%|██▎       | 5685/24420 [2:21:36<7:41:12,  1.48s/it]
 23%|██▎       | 5686/24420 [2:21:37<7:41:19,  1.48s/it]
 23%|██▎       | 5687/24420 [2:21:39<7:40:47,  1.48s/it]
 23%|██▎       | 5688/24420 [2:21:40<7:42:55,  1.48s/it]
 23%|██▎       | 5689/24420 [2:21:42<7:43:51,  1.49s/it]
 23%|██▎       | 5690/24420 [2:21:43<7:44:04,  1.49s/it]
                                                        
{'loss': 0.1343, 'learning_rate': 5e-05, 'epoch': 1.17}

 23%|██▎       | 5690/24420 [2:21:43<7:44:04,  1.49s/it]
 23%|██▎       | 5691/24420 [2:21:45<7:43:50,  1.49s/it]
 23%|██▎       | 5692/24420 [2:21:46<7:43:58,  1.49s/it]
 23%|██▎       | 5693/24420 [2:21:48<7:44:10,  1.49s/it]
 23%|██▎       | 5694/24420 [2:21:49<7:43:20,  1.48s/it]
 23%|██▎       | 5695/24420 [2:21:51<7:42:54,  1.48s/it]
                                                        
{'loss': 0.1438, 'learning_rate': 5e-05, 'epoch': 1.17}

 23%|██▎       | 5695/24420 [2:21:51<7:42:54,  1.48s/it]
 23%|██▎       | 5696/24420 [2:21:52<7:42:57,  1.48s/it]
 23%|██▎       | 5697/24420 [2:21:54<7:42:46,  1.48s/it]
 23%|██▎       | 5698/24420 [2:21:55<7:42:28,  1.48s/it]
 23%|██▎       | 5699/24420 [2:21:57<7:43:23,  1.49s/it]
 23%|██▎       | 5700/24420 [2:21:58<7:43:28,  1.49s/it]
                                                        
{'loss': 0.1425, 'learning_rate': 5e-05, 'epoch': 1.17}

 23%|██▎       | 5700/24420 [2:21:58<7:43:28,  1.49s/it]
 23%|██▎       | 5701/24420 [2:22:00<7:43:56,  1.49s/it]
 23%|██▎       | 5702/24420 [2:22:01<7:42:36,  1.48s/it]
 23%|██▎       | 5703/24420 [2:22:03<7:42:18,  1.48s/it]
 23%|██▎       | 5704/24420 [2:22:04<7:41:59,  1.48s/it]
 23%|██▎       | 5705/24420 [2:22:06<7:42:28,  1.48s/it]
                                                        
{'loss': 0.115, 'learning_rate': 5e-05, 'epoch': 1.17}

 23%|██▎       | 5705/24420 [2:22:06<7:42:28,  1.48s/it]
 23%|██▎       | 5706/24420 [2:22:07<7:42:18,  1.48s/it]
 23%|██▎       | 5707/24420 [2:22:09<7:42:30,  1.48s/it]
 23%|██▎       | 5708/24420 [2:22:10<7:46:08,  1.49s/it]
 23%|██▎       | 5709/24420 [2:22:12<7:44:21,  1.49s/it]
 23%|██▎       | 5710/24420 [2:22:13<7:43:12,  1.49s/it]
                                                        
{'loss': 0.1382, 'learning_rate': 5e-05, 'epoch': 1.17}

 23%|██▎       | 5710/24420 [2:22:13<7:43:12,  1.49s/it]
 23%|██▎       | 5711/24420 [2:22:14<7:42:23,  1.48s/it]
 23%|██▎       | 5712/24420 [2:22:16<7:42:10,  1.48s/it]
 23%|██▎       | 5713/24420 [2:22:17<7:43:51,  1.49s/it]
 23%|██▎       | 5714/24420 [2:22:19<7:43:21,  1.49s/it]
 23%|██▎       | 5715/24420 [2:22:20<7:42:43,  1.48s/it]
                                                        
{'loss': 0.1496, 'learning_rate': 5e-05, 'epoch': 1.17}

 23%|██▎       | 5715/24420 [2:22:20<7:42:43,  1.48s/it]
 23%|██▎       | 5716/24420 [2:22:22<7:42:29,  1.48s/it]
 23%|██▎       | 5717/24420 [2:22:23<7:42:13,  1.48s/it]
 23%|██▎       | 5718/24420 [2:22:25<7:41:24,  1.48s/it]
 23%|██▎       | 5719/24420 [2:22:26<7:41:15,  1.48s/it]
 23%|██▎       | 5720/24420 [2:22:28<7:41:36,  1.48s/it]
                                                        
{'loss': 0.1334, 'learning_rate': 5e-05, 'epoch': 1.17}

 23%|██▎       | 5720/24420 [2:22:28<7:41:36,  1.48s/it]
 23%|██▎       | 5721/24420 [2:22:29<7:41:56,  1.48s/it]
 23%|██▎       | 5722/24420 [2:22:31<7:41:38,  1.48s/it]
 23%|██▎       | 5723/24420 [2:22:32<7:41:29,  1.48s/it]
 23%|██▎       | 5724/24420 [2:22:34<7:41:21,  1.48s/it]
 23%|██▎       | 5725/24420 [2:22:35<7:41:26,  1.48s/it]
                                                        
{'loss': 0.1505, 'learning_rate': 5e-05, 'epoch': 1.17}

 23%|██▎       | 5725/24420 [2:22:35<7:41:26,  1.48s/it]
 23%|██▎       | 5726/24420 [2:22:37<7:41:55,  1.48s/it]
 23%|██▎       | 5727/24420 [2:22:38<7:42:05,  1.48s/it]
 23%|██▎       | 5728/24420 [2:22:40<7:42:08,  1.48s/it]
 23%|██▎       | 5729/24420 [2:22:41<7:42:35,  1.48s/it]
 23%|██▎       | 5730/24420 [2:22:43<7:42:22,  1.48s/it]
                                                        
{'loss': 0.1359, 'learning_rate': 5e-05, 'epoch': 1.17}

 23%|██▎       | 5730/24420 [2:22:43<7:42:22,  1.48s/it]
 23%|██▎       | 5731/24420 [2:22:44<7:42:08,  1.48s/it]
 23%|██▎       | 5732/24420 [2:22:46<7:42:56,  1.49s/it]
 23%|██▎       | 5733/24420 [2:22:47<7:43:00,  1.49s/it]
 23%|██▎       | 5734/24420 [2:22:49<7:44:16,  1.49s/it]
 23%|██▎       | 5735/24420 [2:22:50<7:43:24,  1.49s/it]
                                                        
{'loss': 0.1301, 'learning_rate': 5e-05, 'epoch': 1.17}

 23%|██▎       | 5735/24420 [2:22:50<7:43:24,  1.49s/it]
 23%|██▎       | 5736/24420 [2:22:52<7:43:02,  1.49s/it]
 23%|██▎       | 5737/24420 [2:22:53<7:42:41,  1.49s/it]
 23%|██▎       | 5738/24420 [2:22:55<7:41:55,  1.48s/it]
 24%|██▎       | 5739/24420 [2:22:56<7:41:58,  1.48s/it]
 24%|██▎       | 5740/24420 [2:22:57<7:41:14,  1.48s/it]
                                                        
{'loss': 0.1472, 'learning_rate': 5e-05, 'epoch': 1.18}

 24%|██▎       | 5740/24420 [2:22:57<7:41:14,  1.48s/it]
 24%|██▎       | 5741/24420 [2:22:59<7:42:19,  1.49s/it]
 24%|██▎       | 5742/24420 [2:23:00<7:41:50,  1.48s/it]
 24%|██▎       | 5743/24420 [2:23:02<7:41:25,  1.48s/it]
 24%|██▎       | 5744/24420 [2:23:03<7:41:54,  1.48s/it]
 24%|██▎       | 5745/24420 [2:23:05<7:41:52,  1.48s/it]
                                                        
{'loss': 0.1298, 'learning_rate': 5e-05, 'epoch': 1.18}

 24%|██▎       | 5745/24420 [2:23:05<7:41:52,  1.48s/it]
 24%|██▎       | 5746/24420 [2:23:06<7:42:28,  1.49s/it]
 24%|██▎       | 5747/24420 [2:23:08<7:42:10,  1.49s/it]
 24%|██▎       | 5748/24420 [2:23:09<7:41:16,  1.48s/it]
 24%|██▎       | 5749/24420 [2:23:11<7:42:20,  1.49s/it]
 24%|██▎       | 5750/24420 [2:23:12<7:42:41,  1.49s/it]
                                                        
{'loss': 0.141, 'learning_rate': 5e-05, 'epoch': 1.18}

 24%|██▎       | 5750/24420 [2:23:12<7:42:41,  1.49s/it]
 24%|██▎       | 5751/24420 [2:23:14<7:42:17,  1.49s/it]
 24%|██▎       | 5752/24420 [2:23:15<7:41:48,  1.48s/it]
 24%|██▎       | 5753/24420 [2:23:17<7:41:48,  1.48s/it]
 24%|██▎       | 5754/24420 [2:23:18<7:41:58,  1.48s/it]
 24%|██▎       | 5755/24420 [2:23:20<7:41:46,  1.48s/it]
                                                        
{'loss': 0.1387, 'learning_rate': 5e-05, 'epoch': 1.18}

 24%|██▎       | 5755/24420 [2:23:20<7:41:46,  1.48s/it]
 24%|██▎       | 5756/24420 [2:23:21<7:42:33,  1.49s/it]
 24%|██▎       | 5757/24420 [2:23:23<7:40:56,  1.48s/it]
 24%|██▎       | 5758/24420 [2:23:24<7:40:43,  1.48s/it]
 24%|██▎       | 5759/24420 [2:23:26<7:40:20,  1.48s/it]
 24%|██▎       | 5760/24420 [2:23:27<7:40:31,  1.48s/it]
                                                        
{'loss': 0.1281, 'learning_rate': 5e-05, 'epoch': 1.18}

 24%|██▎       | 5760/24420 [2:23:27<7:40:31,  1.48s/it]
 24%|██▎       | 5761/24420 [2:23:29<7:41:04,  1.48s/it]
 24%|██▎       | 5762/24420 [2:23:30<7:40:53,  1.48s/it]
 24%|██▎       | 5763/24420 [2:23:32<7:41:05,  1.48s/it]
 24%|██▎       | 5764/24420 [2:23:33<7:41:12,  1.48s/it]
 24%|██▎       | 5765/24420 [2:23:35<7:40:55,  1.48s/it]
                                                        
{'loss': 0.1498, 'learning_rate': 5e-05, 'epoch': 1.18}

 24%|██▎       | 5765/24420 [2:23:35<7:40:55,  1.48s/it]
 24%|██▎       | 5766/24420 [2:23:36<7:40:59,  1.48s/it]
 24%|██▎       | 5767/24420 [2:23:38<7:40:56,  1.48s/it]
 24%|██▎       | 5768/24420 [2:23:39<7:40:13,  1.48s/it]
 24%|██▎       | 5769/24420 [2:23:41<7:40:03,  1.48s/it]
 24%|██▎       | 5770/24420 [2:23:42<7:40:11,  1.48s/it]
                                                        
{'loss': 0.1286, 'learning_rate': 5e-05, 'epoch': 1.18}

 24%|██▎       | 5770/24420 [2:23:42<7:40:11,  1.48s/it]
 24%|██▎       | 5771/24420 [2:23:43<7:41:00,  1.48s/it]
 24%|██▎       | 5772/24420 [2:23:45<7:41:37,  1.49s/it]
 24%|██▎       | 5773/24420 [2:23:46<7:41:35,  1.49s/it]
 24%|██▎       | 5774/24420 [2:23:48<7:41:09,  1.48s/it]
 24%|██▎       | 5775/24420 [2:23:49<7:41:15,  1.48s/it]
                                                        
{'loss': 0.1308, 'learning_rate': 5e-05, 'epoch': 1.18}

 24%|██▎       | 5775/24420 [2:23:49<7:41:15,  1.48s/it]
 24%|██▎       | 5776/24420 [2:23:51<7:41:43,  1.49s/it]
 24%|██▎       | 5777/24420 [2:23:52<7:41:17,  1.48s/it]
 24%|██▎       | 5778/24420 [2:23:54<7:41:23,  1.49s/it]
 24%|██▎       | 5779/24420 [2:23:55<7:42:45,  1.49s/it]
 24%|██▎       | 5780/24420 [2:23:57<7:42:14,  1.49s/it]
                                                        
{'loss': 0.1331, 'learning_rate': 5e-05, 'epoch': 1.18}

 24%|██▎       | 5780/24420 [2:23:57<7:42:14,  1.49s/it]
 24%|██▎       | 5781/24420 [2:23:58<7:41:41,  1.49s/it]
 24%|██▎       | 5782/24420 [2:24:00<7:42:38,  1.49s/it]
 24%|██▎       | 5783/24420 [2:24:01<7:41:17,  1.49s/it]
 24%|██▎       | 5784/24420 [2:24:03<7:40:55,  1.48s/it]
 24%|██▎       | 5785/24420 [2:24:04<7:40:38,  1.48s/it]
                                                        
{'loss': 0.1357, 'learning_rate': 5e-05, 'epoch': 1.18}

 24%|██▎       | 5785/24420 [2:24:04<7:40:38,  1.48s/it]
 24%|██▎       | 5786/24420 [2:24:06<7:42:25,  1.49s/it]
 24%|██▎       | 5787/24420 [2:24:07<7:41:19,  1.49s/it]
 24%|██▎       | 5788/24420 [2:24:09<7:41:01,  1.48s/it]
 24%|██▎       | 5789/24420 [2:24:10<7:40:21,  1.48s/it]
 24%|██▎       | 5790/24420 [2:24:12<7:40:47,  1.48s/it]
                                                        
{'loss': 0.1487, 'learning_rate': 5e-05, 'epoch': 1.19}

 24%|██▎       | 5790/24420 [2:24:12<7:40:47,  1.48s/it]
 24%|██▎       | 5791/24420 [2:24:13<7:42:11,  1.49s/it]
 24%|██▎       | 5792/24420 [2:24:15<7:41:22,  1.49s/it]
 24%|██▎       | 5793/24420 [2:24:16<7:41:08,  1.49s/it]
 24%|██▎       | 5794/24420 [2:24:18<7:39:47,  1.48s/it]
 24%|██▎       | 5795/24420 [2:24:19<7:39:29,  1.48s/it]
                                                        
{'loss': 0.126, 'learning_rate': 5e-05, 'epoch': 1.19}

 24%|██▎       | 5795/24420 [2:24:19<7:39:29,  1.48s/it]
 24%|██▎       | 5796/24420 [2:24:21<7:39:36,  1.48s/it]
 24%|██▎       | 5797/24420 [2:24:22<7:39:05,  1.48s/it]
 24%|██▎       | 5798/24420 [2:24:24<7:40:26,  1.48s/it]
 24%|██▎       | 5799/24420 [2:24:25<7:41:00,  1.49s/it]
 24%|██▍       | 5800/24420 [2:24:27<7:41:02,  1.49s/it]
                                                        
{'loss': 0.1195, 'learning_rate': 5e-05, 'epoch': 1.19}

 24%|██▍       | 5800/24420 [2:24:27<7:41:02,  1.49s/it]
 24%|██▍       | 5801/24420 [2:24:32<14:34:23,  2.82s/it]
 24%|██▍       | 5802/24420 [2:24:34<12:30:25,  2.42s/it]
 24%|██▍       | 5803/24420 [2:24:35<11:03:58,  2.14s/it]
 24%|██▍       | 5804/24420 [2:24:37<10:03:20,  1.94s/it]
 24%|██▍       | 5805/24420 [2:24:38<9:20:29,  1.81s/it] 
                                                        
{'loss': 0.1214, 'learning_rate': 5e-05, 'epoch': 1.19}

 24%|██▍       | 5805/24420 [2:24:38<9:20:29,  1.81s/it]
 24%|██▍       | 5806/24420 [2:24:40<8:50:41,  1.71s/it]
 24%|██▍       | 5807/24420 [2:24:41<8:30:12,  1.64s/it]
 24%|██▍       | 5808/24420 [2:24:43<8:16:00,  1.60s/it]
 24%|██▍       | 5809/24420 [2:24:44<8:05:14,  1.56s/it]
 24%|██▍       | 5810/24420 [2:24:46<7:57:51,  1.54s/it]
                                                        
{'loss': 0.1312, 'learning_rate': 5e-05, 'epoch': 1.19}

 24%|██▍       | 5810/24420 [2:24:46<7:57:51,  1.54s/it]
 24%|██▍       | 5811/24420 [2:24:47<7:52:23,  1.52s/it]
 24%|██▍       | 5812/24420 [2:24:49<7:48:28,  1.51s/it]
 24%|██▍       | 5813/24420 [2:24:50<7:45:30,  1.50s/it]
 24%|██▍       | 5814/24420 [2:24:52<7:43:45,  1.50s/it]
 24%|██▍       | 5815/24420 [2:24:53<7:42:57,  1.49s/it]
                                                        
{'loss': 0.1104, 'learning_rate': 5e-05, 'epoch': 1.19}

 24%|██▍       | 5815/24420 [2:24:53<7:42:57,  1.49s/it]
 24%|██▍       | 5816/24420 [2:24:55<7:41:29,  1.49s/it]
 24%|██▍       | 5817/24420 [2:24:56<7:41:19,  1.49s/it]
 24%|██▍       | 5818/24420 [2:24:58<7:40:28,  1.49s/it]
 24%|██▍       | 5819/24420 [2:24:59<7:39:57,  1.48s/it]
 24%|██▍       | 5820/24420 [2:25:01<7:41:23,  1.49s/it]
                                                        
{'loss': 0.1268, 'learning_rate': 5e-05, 'epoch': 1.19}

 24%|██▍       | 5820/24420 [2:25:01<7:41:23,  1.49s/it]
 24%|██▍       | 5821/24420 [2:25:02<7:40:53,  1.49s/it]
 24%|██▍       | 5822/24420 [2:25:04<7:40:36,  1.49s/it]
 24%|██▍       | 5823/24420 [2:25:05<7:40:27,  1.49s/it]
 24%|██▍       | 5824/24420 [2:25:07<7:40:04,  1.48s/it]
 24%|██▍       | 5825/24420 [2:25:08<7:39:13,  1.48s/it]
                                                        
{'loss': 0.1373, 'learning_rate': 5e-05, 'epoch': 1.19}

 24%|██▍       | 5825/24420 [2:25:08<7:39:13,  1.48s/it]
 24%|██▍       | 5826/24420 [2:25:10<7:38:40,  1.48s/it]
 24%|██▍       | 5827/24420 [2:25:11<7:38:24,  1.48s/it]
 24%|██▍       | 5828/24420 [2:25:13<7:38:29,  1.48s/it]
 24%|██▍       | 5829/24420 [2:25:14<7:38:32,  1.48s/it]
 24%|██▍       | 5830/24420 [2:25:15<7:37:46,  1.48s/it]
                                                        
{'loss': 0.1256, 'learning_rate': 5e-05, 'epoch': 1.19}

 24%|██▍       | 5830/24420 [2:25:15<7:37:46,  1.48s/it]
 24%|██▍       | 5831/24420 [2:25:17<7:37:40,  1.48s/it]
 24%|██▍       | 5832/24420 [2:25:18<7:37:25,  1.48s/it]
 24%|██▍       | 5833/24420 [2:25:20<7:37:58,  1.48s/it]
 24%|██▍       | 5834/24420 [2:25:21<7:39:02,  1.48s/it]
 24%|██▍       | 5835/24420 [2:25:23<7:39:35,  1.48s/it]
                                                        
{'loss': 0.1381, 'learning_rate': 5e-05, 'epoch': 1.19}

 24%|██▍       | 5835/24420 [2:25:23<7:39:35,  1.48s/it]
 24%|██▍       | 5836/24420 [2:25:24<7:39:17,  1.48s/it]
 24%|██▍       | 5837/24420 [2:25:26<7:38:16,  1.48s/it]
 24%|██▍       | 5838/24420 [2:25:27<7:38:07,  1.48s/it]
 24%|██▍       | 5839/24420 [2:25:29<7:37:40,  1.48s/it]
 24%|██▍       | 5840/24420 [2:25:30<7:37:34,  1.48s/it]
                                                        
{'loss': 0.1162, 'learning_rate': 5e-05, 'epoch': 1.2}

 24%|██▍       | 5840/24420 [2:25:30<7:37:34,  1.48s/it]
 24%|██▍       | 5841/24420 [2:25:32<7:37:21,  1.48s/it]
 24%|██▍       | 5842/24420 [2:25:33<7:36:32,  1.47s/it]
 24%|██▍       | 5843/24420 [2:25:35<7:35:45,  1.47s/it]
 24%|██▍       | 5844/24420 [2:25:36<7:37:09,  1.48s/it]
 24%|██▍       | 5845/24420 [2:25:38<7:37:26,  1.48s/it]
                                                        
{'loss': 0.1218, 'learning_rate': 5e-05, 'epoch': 1.2}

 24%|██▍       | 5845/24420 [2:25:38<7:37:26,  1.48s/it]
 24%|██▍       | 5846/24420 [2:25:39<7:37:42,  1.48s/it]
 24%|██▍       | 5847/24420 [2:25:41<7:38:45,  1.48s/it]
 24%|██▍       | 5848/24420 [2:25:42<7:39:24,  1.48s/it]
 24%|██▍       | 5849/24420 [2:25:44<7:38:45,  1.48s/it]
 24%|██▍       | 5850/24420 [2:25:45<7:38:57,  1.48s/it]
                                                        
{'loss': 0.137, 'learning_rate': 5e-05, 'epoch': 1.2}

 24%|██▍       | 5850/24420 [2:25:45<7:38:57,  1.48s/it]
 24%|██▍       | 5851/24420 [2:25:47<7:38:41,  1.48s/it]
 24%|██▍       | 5852/24420 [2:25:48<7:38:19,  1.48s/it]
 24%|██▍       | 5853/24420 [2:25:50<7:38:07,  1.48s/it]
 24%|██▍       | 5854/24420 [2:25:51<7:38:13,  1.48s/it]
 24%|██▍       | 5855/24420 [2:25:52<7:37:53,  1.48s/it]
                                                        
{'loss': 0.1317, 'learning_rate': 5e-05, 'epoch': 1.2}

 24%|██▍       | 5855/24420 [2:25:52<7:37:53,  1.48s/it]
 24%|██▍       | 5856/24420 [2:25:54<7:38:26,  1.48s/it]
 24%|██▍       | 5857/24420 [2:25:55<7:37:58,  1.48s/it]
 24%|██▍       | 5858/24420 [2:25:57<7:37:39,  1.48s/it]
 24%|██▍       | 5859/24420 [2:25:58<7:38:17,  1.48s/it]
 24%|██▍       | 5860/24420 [2:26:00<7:37:55,  1.48s/it]
                                                        
{'loss': 0.1275, 'learning_rate': 5e-05, 'epoch': 1.2}

 24%|██▍       | 5860/24420 [2:26:00<7:37:55,  1.48s/it]
 24%|██▍       | 5861/24420 [2:26:01<7:38:11,  1.48s/it]
 24%|██▍       | 5862/24420 [2:26:03<7:38:52,  1.48s/it]
 24%|██▍       | 5863/24420 [2:26:04<7:39:24,  1.49s/it]
 24%|██▍       | 5864/24420 [2:26:06<7:38:37,  1.48s/it]
 24%|██▍       | 5865/24420 [2:26:07<7:37:40,  1.48s/it]
                                                        
{'loss': 0.1517, 'learning_rate': 5e-05, 'epoch': 1.2}

 24%|██▍       | 5865/24420 [2:26:07<7:37:40,  1.48s/it]
 24%|██▍       | 5866/24420 [2:26:09<7:37:40,  1.48s/it]
 24%|██▍       | 5867/24420 [2:26:10<7:38:44,  1.48s/it]
 24%|██▍       | 5868/24420 [2:26:12<7:38:22,  1.48s/it]
 24%|██▍       | 5869/24420 [2:26:13<7:39:45,  1.49s/it]
 24%|██▍       | 5870/24420 [2:26:15<7:39:19,  1.49s/it]
                                                        
{'loss': 0.1531, 'learning_rate': 5e-05, 'epoch': 1.2}

 24%|██▍       | 5870/24420 [2:26:15<7:39:19,  1.49s/it]
 24%|██▍       | 5871/24420 [2:26:16<7:39:50,  1.49s/it]
 24%|██▍       | 5872/24420 [2:26:18<7:39:40,  1.49s/it]
 24%|██▍       | 5873/24420 [2:26:19<7:39:02,  1.49s/it]
 24%|██▍       | 5874/24420 [2:26:21<7:39:03,  1.49s/it]
 24%|██▍       | 5875/24420 [2:26:22<7:38:00,  1.48s/it]
                                                        
{'loss': 0.1173, 'learning_rate': 5e-05, 'epoch': 1.2}

 24%|██▍       | 5875/24420 [2:26:22<7:38:00,  1.48s/it]
 24%|██▍       | 5876/24420 [2:26:24<7:37:49,  1.48s/it]
 24%|██▍       | 5877/24420 [2:26:25<7:37:43,  1.48s/it]
 24%|██▍       | 5878/24420 [2:26:27<7:37:46,  1.48s/it]
 24%|██▍       | 5879/24420 [2:26:28<7:38:13,  1.48s/it]
 24%|██▍       | 5880/24420 [2:26:30<7:37:32,  1.48s/it]
                                                        
{'loss': 0.1253, 'learning_rate': 5e-05, 'epoch': 1.2}

 24%|██▍       | 5880/24420 [2:26:30<7:37:32,  1.48s/it]
 24%|██▍       | 5881/24420 [2:26:31<7:37:41,  1.48s/it]
 24%|██▍       | 5882/24420 [2:26:33<7:37:43,  1.48s/it]
 24%|██▍       | 5883/24420 [2:26:34<7:37:15,  1.48s/it]
 24%|██▍       | 5884/24420 [2:26:35<7:37:21,  1.48s/it]
 24%|██▍       | 5885/24420 [2:26:37<7:38:29,  1.48s/it]
                                                        
{'loss': 0.1384, 'learning_rate': 5e-05, 'epoch': 1.2}

 24%|██▍       | 5885/24420 [2:26:37<7:38:29,  1.48s/it]
 24%|██▍       | 5886/24420 [2:26:38<7:38:07,  1.48s/it]
 24%|██▍       | 5887/24420 [2:26:40<7:37:06,  1.48s/it]
 24%|██▍       | 5888/24420 [2:26:41<7:36:39,  1.48s/it]
 24%|██▍       | 5889/24420 [2:26:43<7:35:47,  1.48s/it]
 24%|██▍       | 5890/24420 [2:26:44<7:35:40,  1.48s/it]
                                                        
{'loss': 0.1349, 'learning_rate': 5e-05, 'epoch': 1.21}

 24%|██▍       | 5890/24420 [2:26:44<7:35:40,  1.48s/it]
 24%|██▍       | 5891/24420 [2:26:46<7:35:45,  1.48s/it]
 24%|██▍       | 5892/24420 [2:26:47<7:36:28,  1.48s/it]
 24%|██▍       | 5893/24420 [2:26:49<7:36:18,  1.48s/it]
 24%|██▍       | 5894/24420 [2:26:50<7:36:51,  1.48s/it]
 24%|██▍       | 5895/24420 [2:26:52<7:36:55,  1.48s/it]
                                                        
{'loss': 0.1323, 'learning_rate': 5e-05, 'epoch': 1.21}

 24%|██▍       | 5895/24420 [2:26:52<7:36:55,  1.48s/it]
 24%|██▍       | 5896/24420 [2:26:53<7:37:48,  1.48s/it]
 24%|██▍       | 5897/24420 [2:26:55<7:37:12,  1.48s/it]
 24%|██▍       | 5898/24420 [2:26:56<7:36:33,  1.48s/it]
 24%|██▍       | 5899/24420 [2:26:58<7:36:28,  1.48s/it]
 24%|██▍       | 5900/24420 [2:26:59<7:36:51,  1.48s/it]
                                                        
{'loss': 0.1133, 'learning_rate': 5e-05, 'epoch': 1.21}

 24%|██▍       | 5900/24420 [2:26:59<7:36:51,  1.48s/it]
 24%|██▍       | 5901/24420 [2:27:01<7:38:52,  1.49s/it]
 24%|██▍       | 5902/24420 [2:27:02<8:06:13,  1.58s/it]
 24%|██▍       | 5903/24420 [2:27:04<7:57:46,  1.55s/it]
 24%|██▍       | 5904/24420 [2:27:05<7:51:32,  1.53s/it]
 24%|██▍       | 5905/24420 [2:27:07<7:46:58,  1.51s/it]
                                                        
{'loss': 0.1195, 'learning_rate': 5e-05, 'epoch': 1.21}

 24%|██▍       | 5905/24420 [2:27:07<7:46:58,  1.51s/it]
 24%|██▍       | 5906/24420 [2:27:08<7:44:06,  1.50s/it]
 24%|██▍       | 5907/24420 [2:27:10<7:42:10,  1.50s/it]
 24%|██▍       | 5908/24420 [2:27:11<7:40:44,  1.49s/it]
 24%|██▍       | 5909/24420 [2:27:13<7:41:24,  1.50s/it]
 24%|██▍       | 5910/24420 [2:27:14<7:39:35,  1.49s/it]
                                                        
{'loss': 0.1455, 'learning_rate': 5e-05, 'epoch': 1.21}

 24%|██▍       | 5910/24420 [2:27:14<7:39:35,  1.49s/it]
 24%|██▍       | 5911/24420 [2:27:16<7:38:34,  1.49s/it]
 24%|██▍       | 5912/24420 [2:27:17<7:38:05,  1.49s/it]
 24%|██▍       | 5913/24420 [2:27:19<7:37:14,  1.48s/it]
 24%|██▍       | 5914/24420 [2:27:20<7:36:00,  1.48s/it]
 24%|██▍       | 5915/24420 [2:27:22<7:36:05,  1.48s/it]
                                                        
{'loss': 0.1198, 'learning_rate': 5e-05, 'epoch': 1.21}

 24%|██▍       | 5915/24420 [2:27:22<7:36:05,  1.48s/it]
 24%|██▍       | 5916/24420 [2:27:23<7:36:48,  1.48s/it]
 24%|██▍       | 5917/24420 [2:27:25<7:36:39,  1.48s/it]
 24%|██▍       | 5918/24420 [2:27:26<7:36:40,  1.48s/it]
 24%|██▍       | 5919/24420 [2:27:28<7:36:26,  1.48s/it]
 24%|██▍       | 5920/24420 [2:27:29<7:35:55,  1.48s/it]
                                                        
{'loss': 0.1368, 'learning_rate': 5e-05, 'epoch': 1.21}

 24%|██▍       | 5920/24420 [2:27:29<7:35:55,  1.48s/it]
 24%|██▍       | 5921/24420 [2:27:31<7:36:08,  1.48s/it]
 24%|██▍       | 5922/24420 [2:27:32<7:35:49,  1.48s/it]
 24%|██▍       | 5923/24420 [2:27:34<7:37:18,  1.48s/it]
 24%|██▍       | 5924/24420 [2:27:35<7:37:03,  1.48s/it]
 24%|██▍       | 5925/24420 [2:27:37<7:37:23,  1.48s/it]
                                                        
{'loss': 0.1262, 'learning_rate': 5e-05, 'epoch': 1.21}

 24%|██▍       | 5925/24420 [2:27:37<7:37:23,  1.48s/it]
 24%|██▍       | 5926/24420 [2:27:38<7:37:24,  1.48s/it]
 24%|██▍       | 5927/24420 [2:27:39<7:37:29,  1.48s/it]
 24%|██▍       | 5928/24420 [2:27:41<7:36:58,  1.48s/it]
 24%|██▍       | 5929/24420 [2:27:42<7:36:36,  1.48s/it]
 24%|██▍       | 5930/24420 [2:27:44<7:36:31,  1.48s/it]
                                                        
{'loss': 0.1341, 'learning_rate': 5e-05, 'epoch': 1.21}

 24%|██▍       | 5930/24420 [2:27:44<7:36:31,  1.48s/it]
 24%|██▍       | 5931/24420 [2:27:45<7:36:44,  1.48s/it]
 24%|██▍       | 5932/24420 [2:27:47<7:35:37,  1.48s/it]
 24%|██▍       | 5933/24420 [2:27:48<7:35:35,  1.48s/it]
 24%|██▍       | 5934/24420 [2:27:50<7:35:41,  1.48s/it]
 24%|██▍       | 5935/24420 [2:27:51<7:35:55,  1.48s/it]
                                                        
{'loss': 0.1208, 'learning_rate': 5e-05, 'epoch': 1.22}

 24%|██▍       | 5935/24420 [2:27:51<7:35:55,  1.48s/it]
 24%|██▍       | 5936/24420 [2:27:53<7:36:33,  1.48s/it]
 24%|██▍       | 5937/24420 [2:27:54<7:37:18,  1.48s/it]
 24%|██▍       | 5938/24420 [2:27:56<7:36:55,  1.48s/it]
 24%|██▍       | 5939/24420 [2:27:57<7:36:29,  1.48s/it]
 24%|██▍       | 5940/24420 [2:27:59<7:37:10,  1.48s/it]
                                                        
{'loss': 0.1655, 'learning_rate': 5e-05, 'epoch': 1.22}

 24%|██▍       | 5940/24420 [2:27:59<7:37:10,  1.48s/it]
 24%|██▍       | 5941/24420 [2:28:00<7:37:05,  1.48s/it]
 24%|██▍       | 5942/24420 [2:28:02<7:37:15,  1.48s/it]
 24%|██▍       | 5943/24420 [2:28:03<7:38:23,  1.49s/it]
 24%|██▍       | 5944/24420 [2:28:05<7:38:20,  1.49s/it]
 24%|██▍       | 5945/24420 [2:28:06<7:37:55,  1.49s/it]
                                                        
{'loss': 0.1212, 'learning_rate': 5e-05, 'epoch': 1.22}

 24%|██▍       | 5945/24420 [2:28:06<7:37:55,  1.49s/it]
 24%|██▍       | 5946/24420 [2:28:08<7:38:09,  1.49s/it]
 24%|██▍       | 5947/24420 [2:28:09<7:37:41,  1.49s/it]
 24%|██▍       | 5948/24420 [2:28:11<7:37:00,  1.48s/it]
 24%|██▍       | 5949/24420 [2:28:12<7:36:43,  1.48s/it]
 24%|██▍       | 5950/24420 [2:28:14<7:36:34,  1.48s/it]
                                                        
{'loss': 0.1585, 'learning_rate': 5e-05, 'epoch': 1.22}

 24%|██▍       | 5950/24420 [2:28:14<7:36:34,  1.48s/it]
 24%|██▍       | 5951/24420 [2:28:15<7:36:27,  1.48s/it]
 24%|██▍       | 5952/24420 [2:28:17<7:36:59,  1.48s/it]
 24%|██▍       | 5953/24420 [2:28:18<7:36:04,  1.48s/it]
 24%|██▍       | 5954/24420 [2:28:20<7:36:19,  1.48s/it]
 24%|██▍       | 5955/24420 [2:28:21<7:36:28,  1.48s/it]
                                                        
{'loss': 0.1372, 'learning_rate': 5e-05, 'epoch': 1.22}

 24%|██▍       | 5955/24420 [2:28:21<7:36:28,  1.48s/it]
 24%|██▍       | 5956/24420 [2:28:22<7:37:08,  1.49s/it]
 24%|██▍       | 5957/24420 [2:28:24<7:37:06,  1.49s/it]
 24%|██▍       | 5958/24420 [2:28:25<7:37:22,  1.49s/it]
 24%|██▍       | 5959/24420 [2:28:27<7:36:38,  1.48s/it]
 24%|██▍       | 5960/24420 [2:28:28<7:36:07,  1.48s/it]
                                                        
{'loss': 0.1288, 'learning_rate': 5e-05, 'epoch': 1.22}

 24%|██▍       | 5960/24420 [2:28:28<7:36:07,  1.48s/it]
 24%|██▍       | 5961/24420 [2:28:30<7:35:55,  1.48s/it]
 24%|██▍       | 5962/24420 [2:28:31<7:35:20,  1.48s/it]
 24%|██▍       | 5963/24420 [2:28:33<7:35:36,  1.48s/it]
 24%|██▍       | 5964/24420 [2:28:34<7:34:55,  1.48s/it]
 24%|██▍       | 5965/24420 [2:28:36<7:34:45,  1.48s/it]
                                                        
{'loss': 0.145, 'learning_rate': 5e-05, 'epoch': 1.22}

 24%|██▍       | 5965/24420 [2:28:36<7:34:45,  1.48s/it]
 24%|██▍       | 5966/24420 [2:28:37<7:35:45,  1.48s/it]
 24%|██▍       | 5967/24420 [2:28:39<7:35:17,  1.48s/it]
 24%|██▍       | 5968/24420 [2:28:40<7:35:18,  1.48s/it]
 24%|██▍       | 5969/24420 [2:28:42<7:35:12,  1.48s/it]
 24%|██▍       | 5970/24420 [2:28:43<7:35:02,  1.48s/it]
                                                        
{'loss': 0.11, 'learning_rate': 5e-05, 'epoch': 1.22}

 24%|██▍       | 5970/24420 [2:28:43<7:35:02,  1.48s/it]
 24%|██▍       | 5971/24420 [2:28:45<7:34:53,  1.48s/it]
 24%|██▍       | 5972/24420 [2:28:46<7:36:05,  1.48s/it]
 24%|██▍       | 5973/24420 [2:28:48<7:36:01,  1.48s/it]
 24%|██▍       | 5974/24420 [2:28:49<7:35:15,  1.48s/it]
 24%|██▍       | 5975/24420 [2:28:51<7:34:57,  1.48s/it]
                                                        
{'loss': 0.1276, 'learning_rate': 5e-05, 'epoch': 1.22}

 24%|██▍       | 5975/24420 [2:28:51<7:34:57,  1.48s/it]
 24%|██▍       | 5976/24420 [2:28:52<7:34:49,  1.48s/it]
 24%|██▍       | 5977/24420 [2:28:54<7:34:39,  1.48s/it]
 24%|██▍       | 5978/24420 [2:28:55<7:35:25,  1.48s/it]
 24%|██▍       | 5979/24420 [2:28:57<7:34:21,  1.48s/it]
 24%|██▍       | 5980/24420 [2:28:58<7:33:52,  1.48s/it]
                                                        
{'loss': 0.1195, 'learning_rate': 5e-05, 'epoch': 1.22}

 24%|██▍       | 5980/24420 [2:28:58<7:33:52,  1.48s/it]
 24%|██▍       | 5981/24420 [2:29:00<7:34:48,  1.48s/it]
 24%|██▍       | 5982/24420 [2:29:01<7:35:51,  1.48s/it]
 25%|██▍       | 5983/24420 [2:29:02<7:35:17,  1.48s/it]
 25%|██▍       | 5984/24420 [2:29:04<7:34:26,  1.48s/it]
 25%|██▍       | 5985/24420 [2:29:05<7:34:03,  1.48s/it]
                                                        
{'loss': 0.1381, 'learning_rate': 5e-05, 'epoch': 1.23}

 25%|██▍       | 5985/24420 [2:29:05<7:34:03,  1.48s/it]
 25%|██▍       | 5986/24420 [2:29:07<7:34:11,  1.48s/it]
 25%|██▍       | 5987/24420 [2:29:08<7:34:27,  1.48s/it]
 25%|██▍       | 5988/24420 [2:29:10<7:34:44,  1.48s/it]
 25%|██▍       | 5989/24420 [2:29:11<7:35:55,  1.48s/it]
 25%|██▍       | 5990/24420 [2:29:13<7:35:36,  1.48s/it]
                                                        
{'loss': 0.1215, 'learning_rate': 5e-05, 'epoch': 1.23}

 25%|██▍       | 5990/24420 [2:29:13<7:35:36,  1.48s/it]
 25%|██▍       | 5991/24420 [2:29:14<7:35:39,  1.48s/it]
 25%|██▍       | 5992/24420 [2:29:16<7:35:53,  1.48s/it]
 25%|██▍       | 5993/24420 [2:29:17<7:35:31,  1.48s/it]
 25%|██▍       | 5994/24420 [2:29:19<7:35:33,  1.48s/it]
 25%|██▍       | 5995/24420 [2:29:20<7:35:23,  1.48s/it]
                                                        
{'loss': 0.117, 'learning_rate': 5e-05, 'epoch': 1.23}

 25%|██▍       | 5995/24420 [2:29:20<7:35:23,  1.48s/it]
 25%|██▍       | 5996/24420 [2:29:22<7:34:45,  1.48s/it]
 25%|██▍       | 5997/24420 [2:29:23<7:34:51,  1.48s/it]
 25%|██▍       | 5998/24420 [2:29:25<7:34:35,  1.48s/it]
 25%|██▍       | 5999/24420 [2:29:26<7:34:42,  1.48s/it]
 25%|██▍       | 6000/24420 [2:29:28<7:35:15,  1.48s/it]
                                                        
{'loss': 0.1261, 'learning_rate': 5e-05, 'epoch': 1.23}

 25%|██▍       | 6000/24420 [2:29:28<7:35:15,  1.48s/it]
 25%|██▍       | 6001/24420 [2:29:34<14:48:22,  2.89s/it]
 25%|██▍       | 6002/24420 [2:29:35<12:37:21,  2.47s/it]
 25%|██▍       | 6003/24420 [2:29:37<11:05:24,  2.17s/it]
 25%|██▍       | 6004/24420 [2:29:38<10:03:22,  1.97s/it]
 25%|██▍       | 6005/24420 [2:29:40<9:18:24,  1.82s/it] 
                                                        
{'loss': 0.1391, 'learning_rate': 5e-05, 'epoch': 1.23}

 25%|██▍       | 6005/24420 [2:29:40<9:18:24,  1.82s/it]
 25%|██▍       | 6006/24420 [2:29:41<8:46:20,  1.72s/it]
 25%|██▍       | 6007/24420 [2:29:43<8:24:54,  1.65s/it]
 25%|██▍       | 6008/24420 [2:29:44<8:09:47,  1.60s/it]
 25%|██▍       | 6009/24420 [2:29:46<7:58:27,  1.56s/it]
 25%|██▍       | 6010/24420 [2:29:47<7:51:48,  1.54s/it]
                                                        
{'loss': 0.1146, 'learning_rate': 5e-05, 'epoch': 1.23}

 25%|██▍       | 6010/24420 [2:29:47<7:51:48,  1.54s/it]
 25%|██▍       | 6011/24420 [2:29:49<7:47:45,  1.52s/it]
 25%|██▍       | 6012/24420 [2:29:50<7:44:11,  1.51s/it]
 25%|██▍       | 6013/24420 [2:29:52<7:41:13,  1.50s/it]
 25%|██▍       | 6014/24420 [2:29:53<7:39:30,  1.50s/it]
 25%|██▍       | 6015/24420 [2:29:55<7:38:37,  1.50s/it]
                                                        
{'loss': 0.1467, 'learning_rate': 5e-05, 'epoch': 1.23}

 25%|██▍       | 6015/24420 [2:29:55<7:38:37,  1.50s/it]
 25%|██▍       | 6016/24420 [2:29:56<7:37:50,  1.49s/it]
 25%|██▍       | 6017/24420 [2:29:58<7:37:10,  1.49s/it]
 25%|██▍       | 6018/24420 [2:29:59<7:35:49,  1.49s/it]
 25%|██▍       | 6019/24420 [2:30:01<7:34:42,  1.48s/it]
 25%|██▍       | 6020/24420 [2:30:02<7:34:08,  1.48s/it]
                                                        
{'loss': 0.1285, 'learning_rate': 5e-05, 'epoch': 1.23}

 25%|██▍       | 6020/24420 [2:30:02<7:34:08,  1.48s/it]
 25%|██▍       | 6021/24420 [2:30:03<7:34:56,  1.48s/it]
 25%|██▍       | 6022/24420 [2:30:05<7:35:02,  1.48s/it]
 25%|██▍       | 6023/24420 [2:30:06<7:34:38,  1.48s/it]
 25%|██▍       | 6024/24420 [2:30:08<7:33:53,  1.48s/it]
 25%|██▍       | 6025/24420 [2:30:09<7:34:44,  1.48s/it]
                                                        
{'loss': 0.1368, 'learning_rate': 5e-05, 'epoch': 1.23}

 25%|██▍       | 6025/24420 [2:30:09<7:34:44,  1.48s/it]
 25%|██▍       | 6026/24420 [2:30:11<7:34:41,  1.48s/it]
 25%|██▍       | 6027/24420 [2:30:12<7:33:53,  1.48s/it]
 25%|██▍       | 6028/24420 [2:30:14<7:34:23,  1.48s/it]
 25%|██▍       | 6029/24420 [2:30:15<7:33:57,  1.48s/it]
 25%|██▍       | 6030/24420 [2:30:17<7:33:34,  1.48s/it]
                                                        
{'loss': 0.1111, 'learning_rate': 5e-05, 'epoch': 1.23}

 25%|██▍       | 6030/24420 [2:30:17<7:33:34,  1.48s/it]
 25%|██▍       | 6031/24420 [2:30:18<7:34:09,  1.48s/it]
 25%|██▍       | 6032/24420 [2:30:20<7:34:35,  1.48s/it]
 25%|██▍       | 6033/24420 [2:30:21<7:34:26,  1.48s/it]
 25%|██▍       | 6034/24420 [2:30:23<7:34:05,  1.48s/it]
 25%|██▍       | 6035/24420 [2:30:24<7:33:17,  1.48s/it]
                                                        
{'loss': 0.1278, 'learning_rate': 5e-05, 'epoch': 1.24}

 25%|██▍       | 6035/24420 [2:30:24<7:33:17,  1.48s/it]
 25%|██▍       | 6036/24420 [2:30:26<7:33:48,  1.48s/it]
 25%|██▍       | 6037/24420 [2:30:27<7:33:50,  1.48s/it]
 25%|██▍       | 6038/24420 [2:30:29<7:34:08,  1.48s/it]
 25%|██▍       | 6039/24420 [2:30:30<7:34:28,  1.48s/it]
 25%|██▍       | 6040/24420 [2:30:32<7:34:14,  1.48s/it]
                                                        
{'loss': 0.1618, 'learning_rate': 5e-05, 'epoch': 1.24}

 25%|██▍       | 6040/24420 [2:30:32<7:34:14,  1.48s/it]
 25%|██▍       | 6041/24420 [2:30:33<7:35:17,  1.49s/it]
 25%|██▍       | 6042/24420 [2:30:35<7:34:15,  1.48s/it]
 25%|██▍       | 6043/24420 [2:30:36<7:34:45,  1.48s/it]
 25%|██▍       | 6044/24420 [2:30:38<7:33:11,  1.48s/it]
 25%|██▍       | 6045/24420 [2:30:39<7:33:22,  1.48s/it]
                                                        
{'loss': 0.1327, 'learning_rate': 5e-05, 'epoch': 1.24}

 25%|██▍       | 6045/24420 [2:30:39<7:33:22,  1.48s/it]
 25%|██▍       | 6046/24420 [2:30:41<7:33:48,  1.48s/it]
 25%|██▍       | 6047/24420 [2:30:42<7:33:32,  1.48s/it]
 25%|██▍       | 6048/24420 [2:30:44<7:34:14,  1.48s/it]
 25%|██▍       | 6049/24420 [2:30:45<7:33:05,  1.48s/it]
 25%|██▍       | 6050/24420 [2:30:46<7:32:49,  1.48s/it]
                                                        
{'loss': 0.1413, 'learning_rate': 5e-05, 'epoch': 1.24}

 25%|██▍       | 6050/24420 [2:30:46<7:32:49,  1.48s/it]
 25%|██▍       | 6051/24420 [2:30:48<7:32:49,  1.48s/it]
 25%|██▍       | 6052/24420 [2:30:49<7:32:56,  1.48s/it]
 25%|██▍       | 6053/24420 [2:30:51<7:33:46,  1.48s/it]
 25%|██▍       | 6054/24420 [2:30:52<7:33:25,  1.48s/it]
 25%|██▍       | 6055/24420 [2:30:54<7:33:02,  1.48s/it]
                                                        
{'loss': 0.1795, 'learning_rate': 5e-05, 'epoch': 1.24}

 25%|██▍       | 6055/24420 [2:30:54<7:33:02,  1.48s/it]
 25%|██▍       | 6056/24420 [2:30:55<7:33:14,  1.48s/it]
 25%|██▍       | 6057/24420 [2:30:57<7:33:17,  1.48s/it]
 25%|██▍       | 6058/24420 [2:30:58<7:32:59,  1.48s/it]
 25%|██▍       | 6059/24420 [2:31:00<7:33:05,  1.48s/it]
 25%|██▍       | 6060/24420 [2:31:01<7:33:06,  1.48s/it]
                                                        
{'loss': 0.1289, 'learning_rate': 5e-05, 'epoch': 1.24}

 25%|██▍       | 6060/24420 [2:31:01<7:33:06,  1.48s/it]
 25%|██▍       | 6061/24420 [2:31:03<7:34:03,  1.48s/it]
 25%|██▍       | 6062/24420 [2:31:04<7:33:20,  1.48s/it]
 25%|██▍       | 6063/24420 [2:31:06<7:33:58,  1.48s/it]
 25%|██▍       | 6064/24420 [2:31:07<7:33:26,  1.48s/it]
 25%|██▍       | 6065/24420 [2:31:09<7:33:08,  1.48s/it]
                                                        
{'loss': 0.1158, 'learning_rate': 5e-05, 'epoch': 1.24}

 25%|██▍       | 6065/24420 [2:31:09<7:33:08,  1.48s/it]
 25%|██▍       | 6066/24420 [2:31:10<7:32:53,  1.48s/it]
 25%|██▍       | 6067/24420 [2:31:12<7:32:49,  1.48s/it]
 25%|██▍       | 6068/24420 [2:31:13<7:32:27,  1.48s/it]
 25%|██▍       | 6069/24420 [2:31:15<7:31:58,  1.48s/it]
 25%|██▍       | 6070/24420 [2:31:16<7:32:16,  1.48s/it]
                                                        
{'loss': 0.1361, 'learning_rate': 5e-05, 'epoch': 1.24}

 25%|██▍       | 6070/24420 [2:31:16<7:32:16,  1.48s/it]
 25%|██▍       | 6071/24420 [2:31:18<7:32:27,  1.48s/it]
 25%|██▍       | 6072/24420 [2:31:19<7:32:30,  1.48s/it]
 25%|██▍       | 6073/24420 [2:31:21<7:32:30,  1.48s/it]
 25%|██▍       | 6074/24420 [2:31:22<7:32:08,  1.48s/it]
 25%|██▍       | 6075/24420 [2:31:23<7:32:29,  1.48s/it]
                                                        
{'loss': 0.1236, 'learning_rate': 5e-05, 'epoch': 1.24}

 25%|██▍       | 6075/24420 [2:31:23<7:32:29,  1.48s/it]
 25%|██▍       | 6076/24420 [2:31:25<7:32:10,  1.48s/it]
 25%|██▍       | 6077/24420 [2:31:26<7:33:16,  1.48s/it]
 25%|██▍       | 6078/24420 [2:31:28<7:34:28,  1.49s/it]
 25%|██▍       | 6079/24420 [2:31:29<7:33:26,  1.48s/it]
 25%|██▍       | 6080/24420 [2:31:31<7:32:49,  1.48s/it]
                                                        
{'loss': 0.1136, 'learning_rate': 5e-05, 'epoch': 1.24}

 25%|██▍       | 6080/24420 [2:31:31<7:32:49,  1.48s/it]
 25%|██▍       | 6081/24420 [2:31:32<7:33:15,  1.48s/it]
 25%|██▍       | 6082/24420 [2:31:34<7:32:30,  1.48s/it]
 25%|██▍       | 6083/24420 [2:31:35<7:31:53,  1.48s/it]
 25%|██▍       | 6084/24420 [2:31:37<7:31:18,  1.48s/it]
 25%|██▍       | 6085/24420 [2:31:38<7:32:09,  1.48s/it]
                                                        
{'loss': 0.1155, 'learning_rate': 5e-05, 'epoch': 1.25}

 25%|██▍       | 6085/24420 [2:31:38<7:32:09,  1.48s/it]
 25%|██▍       | 6086/24420 [2:31:40<7:32:44,  1.48s/it]
 25%|██▍       | 6087/24420 [2:31:41<7:32:07,  1.48s/it]
 25%|██▍       | 6088/24420 [2:31:43<7:32:05,  1.48s/it]
 25%|██▍       | 6089/24420 [2:31:44<7:31:14,  1.48s/it]
 25%|██▍       | 6090/24420 [2:31:46<7:31:16,  1.48s/it]
                                                        
{'loss': 0.1354, 'learning_rate': 5e-05, 'epoch': 1.25}

 25%|██▍       | 6090/24420 [2:31:46<7:31:16,  1.48s/it]
 25%|██▍       | 6091/24420 [2:31:47<7:31:26,  1.48s/it]
 25%|██▍       | 6092/24420 [2:31:49<7:32:03,  1.48s/it]
 25%|██▍       | 6093/24420 [2:31:50<7:32:32,  1.48s/it]
 25%|██▍       | 6094/24420 [2:31:52<7:32:57,  1.48s/it]
 25%|██▍       | 6095/24420 [2:31:53<7:32:00,  1.48s/it]
                                                        
{'loss': 0.1359, 'learning_rate': 5e-05, 'epoch': 1.25}

 25%|██▍       | 6095/24420 [2:31:53<7:32:00,  1.48s/it]
 25%|██▍       | 6096/24420 [2:31:55<7:31:58,  1.48s/it]
 25%|██▍       | 6097/24420 [2:31:56<7:32:13,  1.48s/it]
 25%|██▍       | 6098/24420 [2:31:58<7:32:07,  1.48s/it]
 25%|██▍       | 6099/24420 [2:31:59<7:32:14,  1.48s/it]
 25%|██▍       | 6100/24420 [2:32:00<7:33:22,  1.48s/it]
                                                        
{'loss': 0.1223, 'learning_rate': 5e-05, 'epoch': 1.25}

 25%|██▍       | 6100/24420 [2:32:00<7:33:22,  1.48s/it]
 25%|██▍       | 6101/24420 [2:32:02<7:33:18,  1.48s/it]
 25%|██▍       | 6102/24420 [2:32:03<7:32:45,  1.48s/it]
 25%|██▍       | 6103/24420 [2:32:05<7:32:45,  1.48s/it]
 25%|██▍       | 6104/24420 [2:32:06<7:32:21,  1.48s/it]
 25%|██▌       | 6105/24420 [2:32:08<7:32:24,  1.48s/it]
                                                        
{'loss': 0.132, 'learning_rate': 5e-05, 'epoch': 1.25}

 25%|██▌       | 6105/24420 [2:32:08<7:32:24,  1.48s/it]
 25%|██▌       | 6106/24420 [2:32:09<7:32:29,  1.48s/it]
 25%|██▌       | 6107/24420 [2:32:11<7:32:19,  1.48s/it]
 25%|██▌       | 6108/24420 [2:32:12<7:32:14,  1.48s/it]
 25%|██▌       | 6109/24420 [2:32:14<7:32:45,  1.48s/it]
 25%|██▌       | 6110/24420 [2:32:15<7:33:24,  1.49s/it]
                                                        
{'loss': 0.1345, 'learning_rate': 5e-05, 'epoch': 1.25}

 25%|██▌       | 6110/24420 [2:32:15<7:33:24,  1.49s/it]
 25%|██▌       | 6111/24420 [2:32:17<7:32:50,  1.48s/it]
 25%|██▌       | 6112/24420 [2:32:18<7:33:46,  1.49s/it]
 25%|██▌       | 6113/24420 [2:32:20<7:32:55,  1.48s/it]
 25%|██▌       | 6114/24420 [2:32:21<7:31:39,  1.48s/it]
 25%|██▌       | 6115/24420 [2:32:23<7:31:39,  1.48s/it]
                                                        
{'loss': 0.1251, 'learning_rate': 5e-05, 'epoch': 1.25}

 25%|██▌       | 6115/24420 [2:32:23<7:31:39,  1.48s/it]
 25%|██▌       | 6116/24420 [2:32:24<7:31:59,  1.48s/it]
 25%|██▌       | 6117/24420 [2:32:26<7:31:29,  1.48s/it]
 25%|██▌       | 6118/24420 [2:32:27<7:30:57,  1.48s/it]
 25%|██▌       | 6119/24420 [2:32:29<7:31:14,  1.48s/it]
 25%|██▌       | 6120/24420 [2:32:30<7:30:51,  1.48s/it]
                                                        
{'loss': 0.1211, 'learning_rate': 5e-05, 'epoch': 1.25}

 25%|██▌       | 6120/24420 [2:32:30<7:30:51,  1.48s/it]
 25%|██▌       | 6121/24420 [2:32:32<7:31:25,  1.48s/it]
 25%|██▌       | 6122/24420 [2:32:33<7:31:15,  1.48s/it]
 25%|██▌       | 6123/24420 [2:32:35<7:30:40,  1.48s/it]
 25%|██▌       | 6124/24420 [2:32:36<7:31:36,  1.48s/it]
 25%|██▌       | 6125/24420 [2:32:38<7:30:52,  1.48s/it]
                                                        
{'loss': 0.1276, 'learning_rate': 5e-05, 'epoch': 1.25}

 25%|██▌       | 6125/24420 [2:32:38<7:30:52,  1.48s/it]
 25%|██▌       | 6126/24420 [2:32:39<7:31:38,  1.48s/it]
 25%|██▌       | 6127/24420 [2:32:40<7:31:32,  1.48s/it]
 25%|██▌       | 6128/24420 [2:32:42<7:30:59,  1.48s/it]
 25%|██▌       | 6129/24420 [2:32:43<7:30:53,  1.48s/it]
 25%|██▌       | 6130/24420 [2:32:45<7:31:39,  1.48s/it]
                                                        
{'loss': 0.1346, 'learning_rate': 5e-05, 'epoch': 1.26}

 25%|██▌       | 6130/24420 [2:32:45<7:31:39,  1.48s/it]
 25%|██▌       | 6131/24420 [2:32:46<7:32:26,  1.48s/it]
 25%|██▌       | 6132/24420 [2:32:48<7:31:57,  1.48s/it]
 25%|██▌       | 6133/24420 [2:32:49<7:30:57,  1.48s/it]
 25%|██▌       | 6134/24420 [2:32:51<7:31:11,  1.48s/it]
 25%|██▌       | 6135/24420 [2:32:52<7:30:52,  1.48s/it]
                                                        
{'loss': 0.1316, 'learning_rate': 5e-05, 'epoch': 1.26}

 25%|██▌       | 6135/24420 [2:32:52<7:30:52,  1.48s/it]
 25%|██▌       | 6136/24420 [2:32:54<7:31:08,  1.48s/it]
 25%|██▌       | 6137/24420 [2:32:55<7:31:59,  1.48s/it]
 25%|██▌       | 6138/24420 [2:32:57<7:31:24,  1.48s/it]
 25%|██▌       | 6139/24420 [2:32:58<7:32:07,  1.48s/it]
 25%|██▌       | 6140/24420 [2:33:00<7:32:00,  1.48s/it]
                                                        
{'loss': 0.1327, 'learning_rate': 5e-05, 'epoch': 1.26}

 25%|██▌       | 6140/24420 [2:33:00<7:32:00,  1.48s/it]
 25%|██▌       | 6141/24420 [2:33:01<7:31:19,  1.48s/it]
 25%|██▌       | 6142/24420 [2:33:03<7:31:23,  1.48s/it]
 25%|██▌       | 6143/24420 [2:33:04<7:31:10,  1.48s/it]
 25%|██▌       | 6144/24420 [2:33:06<7:31:11,  1.48s/it]
 25%|██▌       | 6145/24420 [2:33:07<7:30:27,  1.48s/it]
                                                        
{'loss': 0.1257, 'learning_rate': 5e-05, 'epoch': 1.26}

 25%|██▌       | 6145/24420 [2:33:07<7:30:27,  1.48s/it]
 25%|██▌       | 6146/24420 [2:33:09<7:30:53,  1.48s/it]
 25%|██▌       | 6147/24420 [2:33:10<7:31:52,  1.48s/it]
 25%|██▌       | 6148/24420 [2:33:12<7:30:55,  1.48s/it]
 25%|██▌       | 6149/24420 [2:33:13<7:30:54,  1.48s/it]
 25%|██▌       | 6150/24420 [2:33:15<7:30:42,  1.48s/it]
                                                        
{'loss': 0.1216, 'learning_rate': 5e-05, 'epoch': 1.26}

 25%|██▌       | 6150/24420 [2:33:15<7:30:42,  1.48s/it]
 25%|██▌       | 6151/24420 [2:33:16<7:30:49,  1.48s/it]
 25%|██▌       | 6152/24420 [2:33:18<7:31:21,  1.48s/it]
 25%|██▌       | 6153/24420 [2:33:19<7:30:18,  1.48s/it]
 25%|██▌       | 6154/24420 [2:33:20<7:30:26,  1.48s/it]
 25%|██▌       | 6155/24420 [2:33:22<7:31:11,  1.48s/it]
                                                        
{'loss': 0.1224, 'learning_rate': 5e-05, 'epoch': 1.26}

 25%|██▌       | 6155/24420 [2:33:22<7:31:11,  1.48s/it]
 25%|██▌       | 6156/24420 [2:33:23<7:31:55,  1.48s/it]
 25%|██▌       | 6157/24420 [2:33:25<7:31:05,  1.48s/it]
 25%|██▌       | 6158/24420 [2:33:26<7:32:03,  1.49s/it]
 25%|██▌       | 6159/24420 [2:33:28<7:31:48,  1.48s/it]
 25%|██▌       | 6160/24420 [2:33:29<7:31:23,  1.48s/it]
                                                        
{'loss': 0.1194, 'learning_rate': 5e-05, 'epoch': 1.26}

 25%|██▌       | 6160/24420 [2:33:29<7:31:23,  1.48s/it]
 25%|██▌       | 6161/24420 [2:33:31<7:31:33,  1.48s/it]
 25%|██▌       | 6162/24420 [2:33:32<7:30:56,  1.48s/it]
 25%|██▌       | 6163/24420 [2:33:34<7:31:12,  1.48s/it]
 25%|██▌       | 6164/24420 [2:33:35<7:30:40,  1.48s/it]
 25%|██▌       | 6165/24420 [2:33:37<7:30:51,  1.48s/it]
                                                        
{'loss': 0.1054, 'learning_rate': 5e-05, 'epoch': 1.26}

 25%|██▌       | 6165/24420 [2:33:37<7:30:51,  1.48s/it]
 25%|██▌       | 6166/24420 [2:33:38<7:42:08,  1.52s/it]
 25%|██▌       | 6167/24420 [2:33:40<7:38:33,  1.51s/it]
 25%|██▌       | 6168/24420 [2:33:41<7:36:04,  1.50s/it]
 25%|██▌       | 6169/24420 [2:33:43<7:33:46,  1.49s/it]
 25%|██▌       | 6170/24420 [2:33:44<7:32:38,  1.49s/it]
                                                        
{'loss': 0.134, 'learning_rate': 5e-05, 'epoch': 1.26}

 25%|██▌       | 6170/24420 [2:33:44<7:32:38,  1.49s/it]
 25%|██▌       | 6171/24420 [2:33:46<7:32:35,  1.49s/it]
 25%|██▌       | 6172/24420 [2:33:47<7:32:03,  1.49s/it]
 25%|██▌       | 6173/24420 [2:33:49<7:31:34,  1.48s/it]
 25%|██▌       | 6174/24420 [2:33:50<7:30:56,  1.48s/it]
 25%|██▌       | 6175/24420 [2:33:52<7:30:35,  1.48s/it]
                                                        
{'loss': 0.1371, 'learning_rate': 5e-05, 'epoch': 1.26}

 25%|██▌       | 6175/24420 [2:33:52<7:30:35,  1.48s/it]
 25%|██▌       | 6176/24420 [2:33:53<7:30:48,  1.48s/it]
 25%|██▌       | 6177/24420 [2:33:55<7:31:14,  1.48s/it]
 25%|██▌       | 6178/24420 [2:33:56<7:30:28,  1.48s/it]
 25%|██▌       | 6179/24420 [2:33:58<7:30:26,  1.48s/it]
 25%|██▌       | 6180/24420 [2:33:59<7:30:13,  1.48s/it]
                                                        
{'loss': 0.1222, 'learning_rate': 5e-05, 'epoch': 1.27}

 25%|██▌       | 6180/24420 [2:33:59<7:30:13,  1.48s/it]
 25%|██▌       | 6181/24420 [2:34:01<7:30:26,  1.48s/it]
 25%|██▌       | 6182/24420 [2:34:02<7:31:12,  1.48s/it]
 25%|██▌       | 6183/24420 [2:34:04<7:32:28,  1.49s/it]
 25%|██▌       | 6184/24420 [2:34:05<7:31:28,  1.49s/it]
 25%|██▌       | 6185/24420 [2:34:07<7:31:26,  1.49s/it]
                                                        
{'loss': 0.113, 'learning_rate': 5e-05, 'epoch': 1.27}

 25%|██▌       | 6185/24420 [2:34:07<7:31:26,  1.49s/it]
 25%|██▌       | 6186/24420 [2:34:08<7:32:06,  1.49s/it]
 25%|██▌       | 6187/24420 [2:34:10<7:31:47,  1.49s/it]
 25%|██▌       | 6188/24420 [2:34:11<7:30:54,  1.48s/it]
 25%|██▌       | 6189/24420 [2:34:13<7:30:32,  1.48s/it]
 25%|██▌       | 6190/24420 [2:34:14<7:30:09,  1.48s/it]
                                                        
{'loss': 0.1318, 'learning_rate': 5e-05, 'epoch': 1.27}

 25%|██▌       | 6190/24420 [2:34:14<7:30:09,  1.48s/it]
 25%|██▌       | 6191/24420 [2:34:15<7:30:13,  1.48s/it]
 25%|██▌       | 6192/24420 [2:34:17<7:30:03,  1.48s/it]
 25%|██▌       | 6193/24420 [2:34:18<7:30:11,  1.48s/it]
 25%|██▌       | 6194/24420 [2:34:20<7:30:14,  1.48s/it]
 25%|██▌       | 6195/24420 [2:34:21<7:30:16,  1.48s/it]
                                                        
{'loss': 0.131, 'learning_rate': 5e-05, 'epoch': 1.27}

 25%|██▌       | 6195/24420 [2:34:21<7:30:16,  1.48s/it]
 25%|██▌       | 6196/24420 [2:34:23<7:30:00,  1.48s/it]
 25%|██▌       | 6197/24420 [2:34:24<7:30:03,  1.48s/it]
 25%|██▌       | 6198/24420 [2:34:26<7:29:07,  1.48s/it]
 25%|██▌       | 6199/24420 [2:34:27<7:29:24,  1.48s/it]
 25%|██▌       | 6200/24420 [2:34:29<7:29:04,  1.48s/it]
                                                        
{'loss': 0.1273, 'learning_rate': 5e-05, 'epoch': 1.27}

 25%|██▌       | 6200/24420 [2:34:29<7:29:04,  1.48s/it]
 25%|██▌       | 6201/24420 [2:34:33<12:09:09,  2.40s/it]
 25%|██▌       | 6202/24420 [2:34:35<10:44:38,  2.12s/it]
 25%|██▌       | 6203/24420 [2:34:36<9:45:26,  1.93s/it] 
 25%|██▌       | 6204/24420 [2:34:38<9:04:37,  1.79s/it]
 25%|██▌       | 6205/24420 [2:34:39<8:36:17,  1.70s/it]
                                                        
{'loss': 0.166, 'learning_rate': 5e-05, 'epoch': 1.27}

 25%|██▌       | 6205/24420 [2:34:39<8:36:17,  1.70s/it]
 25%|██▌       | 6206/24420 [2:34:41<8:18:00,  1.64s/it]
 25%|██▌       | 6207/24420 [2:34:42<8:03:53,  1.59s/it]
 25%|██▌       | 6208/24420 [2:34:44<7:53:53,  1.56s/it]
 25%|██▌       | 6209/24420 [2:34:45<7:47:45,  1.54s/it]
 25%|██▌       | 6210/24420 [2:34:47<7:41:21,  1.52s/it]
                                                        
{'loss': 0.1406, 'learning_rate': 5e-05, 'epoch': 1.27}

 25%|██▌       | 6210/24420 [2:34:47<7:41:21,  1.52s/it]
 25%|██▌       | 6211/24420 [2:34:48<7:38:24,  1.51s/it]
 25%|██▌       | 6212/24420 [2:34:50<7:35:39,  1.50s/it]
 25%|██▌       | 6213/24420 [2:34:51<7:33:36,  1.49s/it]
 25%|██▌       | 6214/24420 [2:34:53<7:32:21,  1.49s/it]
 25%|██▌       | 6215/24420 [2:34:54<7:31:09,  1.49s/it]
                                                        
{'loss': 0.1597, 'learning_rate': 5e-05, 'epoch': 1.27}

 25%|██▌       | 6215/24420 [2:34:54<7:31:09,  1.49s/it]
 25%|██▌       | 6216/24420 [2:34:56<7:30:49,  1.49s/it]
 25%|██▌       | 6217/24420 [2:34:57<7:30:08,  1.48s/it]
 25%|██▌       | 6218/24420 [2:34:59<7:31:29,  1.49s/it]
 25%|██▌       | 6219/24420 [2:35:00<7:30:13,  1.48s/it]
 25%|██▌       | 6220/24420 [2:35:02<7:30:31,  1.49s/it]
                                                        
{'loss': 0.1532, 'learning_rate': 5e-05, 'epoch': 1.27}

 25%|██▌       | 6220/24420 [2:35:02<7:30:31,  1.49s/it]
 25%|██▌       | 6221/24420 [2:35:03<7:29:58,  1.48s/it]
 25%|██▌       | 6222/24420 [2:35:04<7:29:04,  1.48s/it]
 25%|██▌       | 6223/24420 [2:35:06<7:28:48,  1.48s/it]
 25%|██▌       | 6224/24420 [2:35:07<7:28:39,  1.48s/it]
 25%|██▌       | 6225/24420 [2:35:09<7:27:56,  1.48s/it]
                                                        
{'loss': 0.1357, 'learning_rate': 5e-05, 'epoch': 1.27}

 25%|██▌       | 6225/24420 [2:35:09<7:27:56,  1.48s/it]
 25%|██▌       | 6226/24420 [2:35:10<7:27:56,  1.48s/it]
 25%|██▌       | 6227/24420 [2:35:12<7:27:53,  1.48s/it]
 26%|██▌       | 6228/24420 [2:35:13<7:28:40,  1.48s/it]
 26%|██▌       | 6229/24420 [2:35:15<7:29:29,  1.48s/it]
 26%|██▌       | 6230/24420 [2:35:16<7:30:18,  1.49s/it]
                                                        
{'loss': 0.1333, 'learning_rate': 5e-05, 'epoch': 1.28}

 26%|██▌       | 6230/24420 [2:35:16<7:30:18,  1.49s/it]
 26%|██▌       | 6231/24420 [2:35:18<7:30:35,  1.49s/it]
 26%|██▌       | 6232/24420 [2:35:19<7:30:14,  1.49s/it]
 26%|██▌       | 6233/24420 [2:35:21<7:30:08,  1.49s/it]
 26%|██▌       | 6234/24420 [2:35:22<7:29:31,  1.48s/it]
 26%|██▌       | 6235/24420 [2:35:24<7:28:40,  1.48s/it]
                                                        
{'loss': 0.1268, 'learning_rate': 5e-05, 'epoch': 1.28}

 26%|██▌       | 6235/24420 [2:35:24<7:28:40,  1.48s/it]
 26%|██▌       | 6236/24420 [2:35:25<7:28:06,  1.48s/it]
 26%|██▌       | 6237/24420 [2:35:27<7:27:40,  1.48s/it]
 26%|██▌       | 6238/24420 [2:35:28<7:27:46,  1.48s/it]
 26%|██▌       | 6239/24420 [2:35:30<7:28:43,  1.48s/it]
 26%|██▌       | 6240/24420 [2:35:31<7:28:41,  1.48s/it]
                                                        
{'loss': 0.1425, 'learning_rate': 5e-05, 'epoch': 1.28}

 26%|██▌       | 6240/24420 [2:35:31<7:28:41,  1.48s/it]
 26%|██▌       | 6241/24420 [2:35:33<7:28:37,  1.48s/it]
 26%|██▌       | 6242/24420 [2:35:34<7:27:56,  1.48s/it]
 26%|██▌       | 6243/24420 [2:35:36<7:27:41,  1.48s/it]
 26%|██▌       | 6244/24420 [2:35:37<7:27:03,  1.48s/it]
 26%|██▌       | 6245/24420 [2:35:39<7:27:04,  1.48s/it]
                                                        
{'loss': 0.1307, 'learning_rate': 5e-05, 'epoch': 1.28}

 26%|██▌       | 6245/24420 [2:35:39<7:27:04,  1.48s/it]
 26%|██▌       | 6246/24420 [2:35:40<7:27:32,  1.48s/it]
 26%|██▌       | 6247/24420 [2:35:41<7:27:45,  1.48s/it]
 26%|██▌       | 6248/24420 [2:35:43<7:27:45,  1.48s/it]
 26%|██▌       | 6249/24420 [2:35:44<7:28:29,  1.48s/it]
 26%|██▌       | 6250/24420 [2:35:46<7:27:57,  1.48s/it]
                                                        
{'loss': 0.1128, 'learning_rate': 5e-05, 'epoch': 1.28}

 26%|██▌       | 6250/24420 [2:35:46<7:27:57,  1.48s/it]
 26%|██▌       | 6251/24420 [2:35:47<7:27:48,  1.48s/it]
 26%|██▌       | 6252/24420 [2:35:49<7:27:50,  1.48s/it]
 26%|██▌       | 6253/24420 [2:35:50<7:28:16,  1.48s/it]
 26%|██▌       | 6254/24420 [2:35:52<7:28:52,  1.48s/it]
 26%|██▌       | 6255/24420 [2:35:53<7:28:54,  1.48s/it]
                                                        
{'loss': 0.1437, 'learning_rate': 5e-05, 'epoch': 1.28}

 26%|██▌       | 6255/24420 [2:35:53<7:28:54,  1.48s/it]
 26%|██▌       | 6256/24420 [2:35:55<7:28:33,  1.48s/it]
 26%|██▌       | 6257/24420 [2:35:56<7:28:00,  1.48s/it]
 26%|██▌       | 6258/24420 [2:35:58<7:28:16,  1.48s/it]
 26%|██▌       | 6259/24420 [2:35:59<7:27:53,  1.48s/it]
 26%|██▌       | 6260/24420 [2:36:01<7:27:54,  1.48s/it]
                                                        
{'loss': 0.1389, 'learning_rate': 5e-05, 'epoch': 1.28}

 26%|██▌       | 6260/24420 [2:36:01<7:27:54,  1.48s/it]
 26%|██▌       | 6261/24420 [2:36:02<7:28:22,  1.48s/it]
 26%|██▌       | 6262/24420 [2:36:04<7:28:34,  1.48s/it]
 26%|██▌       | 6263/24420 [2:36:05<7:28:51,  1.48s/it]
 26%|██▌       | 6264/24420 [2:36:07<7:28:31,  1.48s/it]
 26%|██▌       | 6265/24420 [2:36:08<7:28:29,  1.48s/it]
                                                        
{'loss': 0.1311, 'learning_rate': 5e-05, 'epoch': 1.28}

 26%|██▌       | 6265/24420 [2:36:08<7:28:29,  1.48s/it]
 26%|██▌       | 6266/24420 [2:36:10<7:28:15,  1.48s/it]
 26%|██▌       | 6267/24420 [2:36:11<7:27:59,  1.48s/it]
 26%|██▌       | 6268/24420 [2:36:13<7:27:57,  1.48s/it]
 26%|██▌       | 6269/24420 [2:36:14<7:28:02,  1.48s/it]
 26%|██▌       | 6270/24420 [2:36:16<7:27:36,  1.48s/it]
                                                        
{'loss': 0.134, 'learning_rate': 5e-05, 'epoch': 1.28}

 26%|██▌       | 6270/24420 [2:36:16<7:27:36,  1.48s/it]
 26%|██▌       | 6271/24420 [2:36:17<7:28:11,  1.48s/it]
 26%|██▌       | 6272/24420 [2:36:18<7:27:44,  1.48s/it]
 26%|██▌       | 6273/24420 [2:36:20<7:27:33,  1.48s/it]
 26%|██▌       | 6274/24420 [2:36:21<7:27:59,  1.48s/it]
 26%|██▌       | 6275/24420 [2:36:23<7:28:33,  1.48s/it]
                                                        
{'loss': 0.1316, 'learning_rate': 5e-05, 'epoch': 1.28}

 26%|██▌       | 6275/24420 [2:36:23<7:28:33,  1.48s/it]
 26%|██▌       | 6276/24420 [2:36:24<7:28:23,  1.48s/it]
 26%|██▌       | 6277/24420 [2:36:26<7:28:06,  1.48s/it]
 26%|██▌       | 6278/24420 [2:36:27<7:28:09,  1.48s/it]
 26%|██▌       | 6279/24420 [2:36:29<7:28:18,  1.48s/it]
 26%|██▌       | 6280/24420 [2:36:30<7:27:55,  1.48s/it]
                                                        
{'loss': 0.1224, 'learning_rate': 5e-05, 'epoch': 1.29}

 26%|██▌       | 6280/24420 [2:36:30<7:27:55,  1.48s/it]
 26%|██▌       | 6281/24420 [2:36:32<7:27:56,  1.48s/it]
 26%|██▌       | 6282/24420 [2:36:33<7:27:32,  1.48s/it]
 26%|██▌       | 6283/24420 [2:36:35<7:28:17,  1.48s/it]
 26%|██▌       | 6284/24420 [2:36:36<7:28:07,  1.48s/it]
 26%|██▌       | 6285/24420 [2:36:38<7:29:01,  1.49s/it]
                                                        
{'loss': 0.1527, 'learning_rate': 5e-05, 'epoch': 1.29}

 26%|██▌       | 6285/24420 [2:36:38<7:29:01,  1.49s/it]
 26%|██▌       | 6286/24420 [2:36:39<7:30:04,  1.49s/it]
 26%|██▌       | 6287/24420 [2:36:41<7:29:39,  1.49s/it]
 26%|██▌       | 6288/24420 [2:36:42<7:29:10,  1.49s/it]
 26%|██▌       | 6289/24420 [2:36:44<7:29:17,  1.49s/it]
 26%|██▌       | 6290/24420 [2:36:45<7:29:27,  1.49s/it]
                                                        
{'loss': 0.1338, 'learning_rate': 5e-05, 'epoch': 1.29}

 26%|██▌       | 6290/24420 [2:36:45<7:29:27,  1.49s/it]
 26%|██▌       | 6291/24420 [2:36:47<7:28:37,  1.48s/it]
 26%|██▌       | 6292/24420 [2:36:48<7:28:57,  1.49s/it]
 26%|██▌       | 6293/24420 [2:36:50<7:29:19,  1.49s/it]
 26%|██▌       | 6294/24420 [2:36:51<7:28:50,  1.49s/it]
 26%|██▌       | 6295/24420 [2:36:53<7:28:45,  1.49s/it]
                                                        
{'loss': 0.131, 'learning_rate': 5e-05, 'epoch': 1.29}

 26%|██▌       | 6295/24420 [2:36:53<7:28:45,  1.49s/it]
 26%|██▌       | 6296/24420 [2:36:54<7:29:08,  1.49s/it]
 26%|██▌       | 6297/24420 [2:36:56<7:28:44,  1.49s/it]
 26%|██▌       | 6298/24420 [2:36:57<7:28:14,  1.48s/it]
 26%|██▌       | 6299/24420 [2:36:59<7:28:00,  1.48s/it]
 26%|██▌       | 6300/24420 [2:37:00<7:28:29,  1.49s/it]
                                                        
{'loss': 0.1429, 'learning_rate': 5e-05, 'epoch': 1.29}

 26%|██▌       | 6300/24420 [2:37:00<7:28:29,  1.49s/it]
 26%|██▌       | 6301/24420 [2:37:02<7:28:50,  1.49s/it]
 26%|██▌       | 6302/24420 [2:37:03<7:28:04,  1.48s/it]
 26%|██▌       | 6303/24420 [2:37:05<7:28:10,  1.48s/it]
 26%|██▌       | 6304/24420 [2:37:06<7:27:55,  1.48s/it]
 26%|██▌       | 6305/24420 [2:37:07<7:27:32,  1.48s/it]
                                                        
{'loss': 0.1262, 'learning_rate': 5e-05, 'epoch': 1.29}

 26%|██▌       | 6305/24420 [2:37:07<7:27:32,  1.48s/it]
 26%|██▌       | 6306/24420 [2:37:09<7:27:40,  1.48s/it]
 26%|██▌       | 6307/24420 [2:37:10<7:27:52,  1.48s/it]
 26%|██▌       | 6308/24420 [2:37:12<7:27:30,  1.48s/it]
 26%|██▌       | 6309/24420 [2:37:13<7:29:33,  1.49s/it]
 26%|██▌       | 6310/24420 [2:37:15<7:29:27,  1.49s/it]
                                                        
{'loss': 0.1301, 'learning_rate': 5e-05, 'epoch': 1.29}

 26%|██▌       | 6310/24420 [2:37:15<7:29:27,  1.49s/it]
 26%|██▌       | 6311/24420 [2:37:16<7:28:35,  1.49s/it]
 26%|██▌       | 6312/24420 [2:37:18<7:27:00,  1.48s/it]
 26%|██▌       | 6313/24420 [2:37:19<7:27:21,  1.48s/it]
 26%|██▌       | 6314/24420 [2:37:21<7:27:48,  1.48s/it]
 26%|██▌       | 6315/24420 [2:37:22<7:27:19,  1.48s/it]
                                                        
{'loss': 0.1185, 'learning_rate': 5e-05, 'epoch': 1.29}

 26%|██▌       | 6315/24420 [2:37:22<7:27:19,  1.48s/it]
 26%|██▌       | 6316/24420 [2:37:24<7:27:29,  1.48s/it]
 26%|██▌       | 6317/24420 [2:37:25<7:26:44,  1.48s/it]
 26%|██▌       | 6318/24420 [2:37:27<7:27:02,  1.48s/it]
 26%|██▌       | 6319/24420 [2:37:28<7:26:34,  1.48s/it]
 26%|██▌       | 6320/24420 [2:37:30<7:26:25,  1.48s/it]
                                                        
{'loss': 0.1302, 'learning_rate': 5e-05, 'epoch': 1.29}

 26%|██▌       | 6320/24420 [2:37:30<7:26:25,  1.48s/it]
 26%|██▌       | 6321/24420 [2:37:31<7:26:02,  1.48s/it]
 26%|██▌       | 6322/24420 [2:37:33<7:25:41,  1.48s/it]
 26%|██▌       | 6323/24420 [2:37:34<7:26:18,  1.48s/it]
 26%|██▌       | 6324/24420 [2:37:36<7:26:31,  1.48s/it]
 26%|██▌       | 6325/24420 [2:37:37<7:26:28,  1.48s/it]
                                                        
{'loss': 0.1181, 'learning_rate': 5e-05, 'epoch': 1.3}

 26%|██▌       | 6325/24420 [2:37:37<7:26:28,  1.48s/it]
 26%|██▌       | 6326/24420 [2:37:39<7:27:11,  1.48s/it]
 26%|██▌       | 6327/24420 [2:37:40<7:26:48,  1.48s/it]
 26%|██▌       | 6328/24420 [2:37:42<7:26:25,  1.48s/it]
 26%|██▌       | 6329/24420 [2:37:43<7:25:59,  1.48s/it]
 26%|██▌       | 6330/24420 [2:37:45<7:25:56,  1.48s/it]
                                                        
{'loss': 0.1379, 'learning_rate': 5e-05, 'epoch': 1.3}

 26%|██▌       | 6330/24420 [2:37:45<7:25:56,  1.48s/it]
 26%|██▌       | 6331/24420 [2:37:46<7:25:45,  1.48s/it]
 26%|██▌       | 6332/24420 [2:37:47<7:25:55,  1.48s/it]
 26%|██▌       | 6333/24420 [2:37:49<7:26:36,  1.48s/it]
 26%|██▌       | 6334/24420 [2:37:50<7:26:59,  1.48s/it]
 26%|██▌       | 6335/24420 [2:37:52<7:28:21,  1.49s/it]
                                                        
{'loss': 0.1386, 'learning_rate': 5e-05, 'epoch': 1.3}

 26%|██▌       | 6335/24420 [2:37:52<7:28:21,  1.49s/it]
 26%|██▌       | 6336/24420 [2:37:53<7:28:27,  1.49s/it]
 26%|██▌       | 6337/24420 [2:37:55<7:28:09,  1.49s/it]
 26%|██▌       | 6338/24420 [2:37:56<7:28:29,  1.49s/it]
 26%|██▌       | 6339/24420 [2:37:58<7:28:36,  1.49s/it]
 26%|██▌       | 6340/24420 [2:37:59<7:28:31,  1.49s/it]
                                                        
{'loss': 0.1166, 'learning_rate': 5e-05, 'epoch': 1.3}

 26%|██▌       | 6340/24420 [2:37:59<7:28:31,  1.49s/it]
 26%|██▌       | 6341/24420 [2:38:01<7:27:32,  1.49s/it]
 26%|██▌       | 6342/24420 [2:38:02<7:26:43,  1.48s/it]
 26%|██▌       | 6343/24420 [2:38:04<7:27:28,  1.49s/it]
 26%|██▌       | 6344/24420 [2:38:05<7:27:48,  1.49s/it]
 26%|██▌       | 6345/24420 [2:38:07<7:28:38,  1.49s/it]
                                                        
{'loss': 0.105, 'learning_rate': 5e-05, 'epoch': 1.3}

 26%|██▌       | 6345/24420 [2:38:07<7:28:38,  1.49s/it]
 26%|██▌       | 6346/24420 [2:38:08<7:28:30,  1.49s/it]
 26%|██▌       | 6347/24420 [2:38:10<7:28:50,  1.49s/it]
 26%|██▌       | 6348/24420 [2:38:11<7:28:33,  1.49s/it]
 26%|██▌       | 6349/24420 [2:38:13<7:28:53,  1.49s/it]
 26%|██▌       | 6350/24420 [2:38:14<7:28:19,  1.49s/it]
                                                        
{'loss': 0.1274, 'learning_rate': 5e-05, 'epoch': 1.3}

 26%|██▌       | 6350/24420 [2:38:14<7:28:19,  1.49s/it]
 26%|██▌       | 6351/24420 [2:38:16<7:27:46,  1.49s/it]
 26%|██▌       | 6352/24420 [2:38:17<7:28:23,  1.49s/it]
 26%|██▌       | 6353/24420 [2:38:19<7:27:13,  1.49s/it]
 26%|██▌       | 6354/24420 [2:38:20<7:27:09,  1.49s/it]
 26%|██▌       | 6355/24420 [2:38:22<7:27:44,  1.49s/it]
                                                        
{'loss': 0.1329, 'learning_rate': 5e-05, 'epoch': 1.3}

 26%|██▌       | 6355/24420 [2:38:22<7:27:44,  1.49s/it]
 26%|██▌       | 6356/24420 [2:38:23<7:28:23,  1.49s/it]
 26%|██▌       | 6357/24420 [2:38:25<7:28:24,  1.49s/it]
 26%|██▌       | 6358/24420 [2:38:26<7:29:02,  1.49s/it]
 26%|██▌       | 6359/24420 [2:38:28<7:29:29,  1.49s/it]
 26%|██▌       | 6360/24420 [2:38:29<7:29:13,  1.49s/it]
                                                        
{'loss': 0.1298, 'learning_rate': 5e-05, 'epoch': 1.3}

 26%|██▌       | 6360/24420 [2:38:29<7:29:13,  1.49s/it]
 26%|██▌       | 6361/24420 [2:38:31<7:28:53,  1.49s/it]
 26%|██▌       | 6362/24420 [2:38:32<7:28:36,  1.49s/it]
 26%|██▌       | 6363/24420 [2:38:34<7:29:04,  1.49s/it]
 26%|██▌       | 6364/24420 [2:38:35<7:29:12,  1.49s/it]
 26%|██▌       | 6365/24420 [2:38:37<7:28:59,  1.49s/it]
                                                        
{'loss': 0.13, 'learning_rate': 5e-05, 'epoch': 1.3}

 26%|██▌       | 6365/24420 [2:38:37<7:28:59,  1.49s/it]
 26%|██▌       | 6366/24420 [2:38:38<7:28:49,  1.49s/it]
 26%|██▌       | 6367/24420 [2:38:40<7:28:10,  1.49s/it]
 26%|██▌       | 6368/24420 [2:38:41<7:27:55,  1.49s/it]
 26%|██▌       | 6369/24420 [2:38:43<7:27:13,  1.49s/it]
 26%|██▌       | 6370/24420 [2:38:44<7:27:05,  1.49s/it]
                                                        
{'loss': 0.1122, 'learning_rate': 5e-05, 'epoch': 1.3}

 26%|██▌       | 6370/24420 [2:38:44<7:27:05,  1.49s/it]
 26%|██▌       | 6371/24420 [2:38:46<7:27:55,  1.49s/it]
 26%|██▌       | 6372/24420 [2:38:47<7:27:48,  1.49s/it]
 26%|██▌       | 6373/24420 [2:38:49<7:28:03,  1.49s/it]
 26%|██▌       | 6374/24420 [2:38:50<7:28:32,  1.49s/it]
 26%|██▌       | 6375/24420 [2:38:52<7:28:28,  1.49s/it]
                                                        
{'loss': 0.1391, 'learning_rate': 5e-05, 'epoch': 1.31}

 26%|██▌       | 6375/24420 [2:38:52<7:28:28,  1.49s/it]
 26%|██▌       | 6376/24420 [2:38:53<7:27:50,  1.49s/it]
 26%|██▌       | 6377/24420 [2:38:54<7:28:35,  1.49s/it]
 26%|██▌       | 6378/24420 [2:38:56<7:28:14,  1.49s/it]
 26%|██▌       | 6379/24420 [2:38:57<7:27:54,  1.49s/it]
 26%|██▌       | 6380/24420 [2:38:59<7:27:51,  1.49s/it]
                                                        
{'loss': 0.1528, 'learning_rate': 5e-05, 'epoch': 1.31}

 26%|██▌       | 6380/24420 [2:38:59<7:27:51,  1.49s/it]
 26%|██▌       | 6381/24420 [2:39:00<7:27:56,  1.49s/it]
 26%|██▌       | 6382/24420 [2:39:02<7:27:48,  1.49s/it]
 26%|██▌       | 6383/24420 [2:39:03<7:28:01,  1.49s/it]
 26%|██▌       | 6384/24420 [2:39:05<7:28:04,  1.49s/it]
 26%|██▌       | 6385/24420 [2:39:06<7:27:36,  1.49s/it]
                                                        
{'loss': 0.1239, 'learning_rate': 5e-05, 'epoch': 1.31}

 26%|██▌       | 6385/24420 [2:39:06<7:27:36,  1.49s/it]
 26%|██▌       | 6386/24420 [2:39:08<7:26:59,  1.49s/it]
 26%|██▌       | 6387/24420 [2:39:09<7:26:12,  1.48s/it]
 26%|██▌       | 6388/24420 [2:39:11<7:26:06,  1.48s/it]
 26%|██▌       | 6389/24420 [2:39:12<7:25:09,  1.48s/it]
 26%|██▌       | 6390/24420 [2:39:14<7:24:55,  1.48s/it]
                                                        
{'loss': 0.1257, 'learning_rate': 5e-05, 'epoch': 1.31}

 26%|██▌       | 6390/24420 [2:39:14<7:24:55,  1.48s/it]
 26%|██▌       | 6391/24420 [2:39:15<7:25:09,  1.48s/it]
 26%|██▌       | 6392/24420 [2:39:17<7:26:53,  1.49s/it]
 26%|██▌       | 6393/24420 [2:39:18<7:26:06,  1.48s/it]
 26%|██▌       | 6394/24420 [2:39:20<7:26:45,  1.49s/it]
 26%|██▌       | 6395/24420 [2:39:21<7:27:11,  1.49s/it]
                                                        
{'loss': 0.122, 'learning_rate': 5e-05, 'epoch': 1.31}

 26%|██▌       | 6395/24420 [2:39:21<7:27:11,  1.49s/it]
 26%|██▌       | 6396/24420 [2:39:23<7:29:05,  1.49s/it]
 26%|██▌       | 6397/24420 [2:39:24<7:29:54,  1.50s/it]
 26%|██▌       | 6398/24420 [2:39:26<7:30:12,  1.50s/it]
 26%|██▌       | 6399/24420 [2:39:27<7:30:13,  1.50s/it]
 26%|██▌       | 6400/24420 [2:39:29<7:31:47,  1.50s/it]
                                                        
{'loss': 0.1105, 'learning_rate': 5e-05, 'epoch': 1.31}

 26%|██▌       | 6400/24420 [2:39:29<7:31:47,  1.50s/it]
 26%|██▌       | 6401/24420 [2:39:34<13:12:23,  2.64s/it]
 26%|██▌       | 6402/24420 [2:39:36<11:27:29,  2.29s/it]
 26%|██▌       | 6403/24420 [2:39:37<10:15:21,  2.05s/it]
 26%|██▌       | 6404/24420 [2:39:39<9:23:48,  1.88s/it] 
 26%|██▌       | 6405/24420 [2:39:40<8:49:16,  1.76s/it]
                                                        
{'loss': 0.1034, 'learning_rate': 5e-05, 'epoch': 1.31}

 26%|██▌       | 6405/24420 [2:39:40<8:49:16,  1.76s/it]
 26%|██▌       | 6406/24420 [2:39:42<8:24:50,  1.68s/it]
 26%|██▌       | 6407/24420 [2:39:43<8:07:17,  1.62s/it]
 26%|██▌       | 6408/24420 [2:39:44<7:54:50,  1.58s/it]
 26%|██▌       | 6409/24420 [2:39:46<7:46:39,  1.55s/it]
 26%|██▌       | 6410/24420 [2:39:47<7:41:29,  1.54s/it]
                                                        
{'loss': 0.1206, 'learning_rate': 5e-05, 'epoch': 1.31}

 26%|██▌       | 6410/24420 [2:39:47<7:41:29,  1.54s/it]
 26%|██▋       | 6411/24420 [2:39:49<7:37:05,  1.52s/it]
 26%|██▋       | 6412/24420 [2:39:50<7:34:33,  1.51s/it]
 26%|██▋       | 6413/24420 [2:39:52<7:32:09,  1.51s/it]
 26%|██▋       | 6414/24420 [2:39:53<7:30:45,  1.50s/it]
 26%|██▋       | 6415/24420 [2:39:55<7:30:03,  1.50s/it]
                                                        
{'loss': 0.1001, 'learning_rate': 5e-05, 'epoch': 1.31}

 26%|██▋       | 6415/24420 [2:39:55<7:30:03,  1.50s/it]
 26%|██▋       | 6416/24420 [2:39:56<7:30:30,  1.50s/it]
 26%|██▋       | 6417/24420 [2:39:58<7:28:48,  1.50s/it]
 26%|██▋       | 6418/24420 [2:39:59<7:27:03,  1.49s/it]
 26%|██▋       | 6419/24420 [2:40:01<7:26:27,  1.49s/it]
 26%|██▋       | 6420/24420 [2:40:02<7:25:10,  1.48s/it]
                                                        
{'loss': 0.1337, 'learning_rate': 5e-05, 'epoch': 1.31}

 26%|██▋       | 6420/24420 [2:40:02<7:25:10,  1.48s/it]
 26%|██▋       | 6421/24420 [2:40:04<7:25:11,  1.48s/it]
 26%|██▋       | 6422/24420 [2:40:05<7:25:00,  1.48s/it]
 26%|██▋       | 6423/24420 [2:40:07<7:27:06,  1.49s/it]
 26%|██▋       | 6424/24420 [2:40:08<7:28:14,  1.49s/it]
 26%|██▋       | 6425/24420 [2:40:10<7:26:54,  1.49s/it]
                                                        
{'loss': 0.1347, 'learning_rate': 5e-05, 'epoch': 1.32}

 26%|██▋       | 6425/24420 [2:40:10<7:26:54,  1.49s/it]
 26%|██▋       | 6426/24420 [2:40:11<7:29:09,  1.50s/it]
 26%|██▋       | 6427/24420 [2:40:13<7:28:55,  1.50s/it]
 26%|██▋       | 6428/24420 [2:40:14<7:29:12,  1.50s/it]
 26%|██▋       | 6429/24420 [2:40:16<7:29:25,  1.50s/it]
 26%|██▋       | 6430/24420 [2:40:17<7:31:15,  1.51s/it]
                                                        
{'loss': 0.1415, 'learning_rate': 5e-05, 'epoch': 1.32}

 26%|██▋       | 6430/24420 [2:40:17<7:31:15,  1.51s/it]
 26%|██▋       | 6431/24420 [2:40:19<7:29:08,  1.50s/it]
 26%|██▋       | 6432/24420 [2:40:20<7:29:19,  1.50s/it]
 26%|██▋       | 6433/24420 [2:40:22<7:28:45,  1.50s/it]
 26%|██▋       | 6434/24420 [2:40:23<7:28:09,  1.50s/it]
 26%|██▋       | 6435/24420 [2:40:25<7:27:58,  1.49s/it]
                                                        
{'loss': 0.1395, 'learning_rate': 5e-05, 'epoch': 1.32}

 26%|██▋       | 6435/24420 [2:40:25<7:27:58,  1.49s/it]
 26%|██▋       | 6436/24420 [2:40:26<7:27:50,  1.49s/it]
 26%|██▋       | 6437/24420 [2:40:28<7:27:33,  1.49s/it]
 26%|██▋       | 6438/24420 [2:40:29<7:27:34,  1.49s/it]
 26%|██▋       | 6439/24420 [2:40:31<7:27:30,  1.49s/it]
 26%|██▋       | 6440/24420 [2:40:32<7:26:37,  1.49s/it]
                                                        
{'loss': 0.1133, 'learning_rate': 5e-05, 'epoch': 1.32}

 26%|██▋       | 6440/24420 [2:40:32<7:26:37,  1.49s/it]
 26%|██▋       | 6441/24420 [2:40:34<7:26:40,  1.49s/it]
 26%|██▋       | 6442/24420 [2:40:35<7:27:02,  1.49s/it]
 26%|██▋       | 6443/24420 [2:40:37<7:26:37,  1.49s/it]
 26%|██▋       | 6444/24420 [2:40:38<7:25:34,  1.49s/it]
 26%|██▋       | 6445/24420 [2:40:40<7:24:30,  1.48s/it]
                                                        
{'loss': 0.1345, 'learning_rate': 5e-05, 'epoch': 1.32}

 26%|██▋       | 6445/24420 [2:40:40<7:24:30,  1.48s/it]
 26%|██▋       | 6446/24420 [2:40:41<7:25:02,  1.49s/it]
 26%|██▋       | 6447/24420 [2:40:43<7:24:09,  1.48s/it]
 26%|██▋       | 6448/24420 [2:40:44<7:24:14,  1.48s/it]
 26%|██▋       | 6449/24420 [2:40:46<7:24:20,  1.48s/it]
 26%|██▋       | 6450/24420 [2:40:47<7:24:31,  1.48s/it]
                                                        
{'loss': 0.1284, 'learning_rate': 5e-05, 'epoch': 1.32}

 26%|██▋       | 6450/24420 [2:40:47<7:24:31,  1.48s/it]
 26%|██▋       | 6451/24420 [2:40:49<7:24:42,  1.48s/it]
 26%|██▋       | 6452/24420 [2:40:50<7:24:44,  1.49s/it]
 26%|██▋       | 6453/24420 [2:40:52<7:25:28,  1.49s/it]
 26%|██▋       | 6454/24420 [2:40:53<7:24:23,  1.48s/it]
 26%|██▋       | 6455/24420 [2:40:55<7:24:35,  1.48s/it]
                                                        
{'loss': 0.117, 'learning_rate': 5e-05, 'epoch': 1.32}

 26%|██▋       | 6455/24420 [2:40:55<7:24:35,  1.48s/it]
 26%|██▋       | 6456/24420 [2:40:56<7:24:47,  1.49s/it]
 26%|██▋       | 6457/24420 [2:40:57<7:24:52,  1.49s/it]
 26%|██▋       | 6458/24420 [2:40:59<7:24:29,  1.48s/it]
 26%|██▋       | 6459/24420 [2:41:00<7:24:16,  1.48s/it]
 26%|██▋       | 6460/24420 [2:41:02<7:24:16,  1.48s/it]
                                                        
{'loss': 0.1199, 'learning_rate': 5e-05, 'epoch': 1.32}

 26%|██▋       | 6460/24420 [2:41:02<7:24:16,  1.48s/it]
 26%|██▋       | 6461/24420 [2:41:03<7:25:07,  1.49s/it]
 26%|██▋       | 6462/24420 [2:41:05<7:24:14,  1.48s/it]
 26%|██▋       | 6463/24420 [2:41:06<7:23:53,  1.48s/it]
 26%|██▋       | 6464/24420 [2:41:08<7:24:26,  1.49s/it]
 26%|██▋       | 6465/24420 [2:41:09<7:24:09,  1.48s/it]
                                                        
{'loss': 0.1235, 'learning_rate': 5e-05, 'epoch': 1.32}

 26%|██▋       | 6465/24420 [2:41:09<7:24:09,  1.48s/it]
 26%|██▋       | 6466/24420 [2:41:11<7:23:49,  1.48s/it]
 26%|██▋       | 6467/24420 [2:41:12<7:22:54,  1.48s/it]
 26%|██▋       | 6468/24420 [2:41:14<7:22:45,  1.48s/it]
 26%|██▋       | 6469/24420 [2:41:15<7:22:13,  1.48s/it]
 26%|██▋       | 6470/24420 [2:41:17<7:21:55,  1.48s/it]
                                                        
{'loss': 0.1227, 'learning_rate': 5e-05, 'epoch': 1.32}

 26%|██▋       | 6470/24420 [2:41:17<7:21:55,  1.48s/it]
 26%|██▋       | 6471/24420 [2:41:18<7:22:05,  1.48s/it]
 27%|██▋       | 6472/24420 [2:41:20<7:21:49,  1.48s/it]
 27%|██▋       | 6473/24420 [2:41:21<7:21:10,  1.47s/it]
 27%|██▋       | 6474/24420 [2:41:23<7:22:07,  1.48s/it]
 27%|██▋       | 6475/24420 [2:41:24<7:22:07,  1.48s/it]
                                                        
{'loss': 0.111, 'learning_rate': 5e-05, 'epoch': 1.33}

 27%|██▋       | 6475/24420 [2:41:24<7:22:07,  1.48s/it]
 27%|██▋       | 6476/24420 [2:41:26<7:22:52,  1.48s/it]
 27%|██▋       | 6477/24420 [2:41:27<7:23:15,  1.48s/it]
 27%|██▋       | 6478/24420 [2:41:29<7:23:55,  1.48s/it]
 27%|██▋       | 6479/24420 [2:41:30<7:23:04,  1.48s/it]
 27%|██▋       | 6480/24420 [2:41:32<7:22:38,  1.48s/it]
                                                        
{'loss': 0.1092, 'learning_rate': 5e-05, 'epoch': 1.33}

 27%|██▋       | 6480/24420 [2:41:32<7:22:38,  1.48s/it]
 27%|██▋       | 6481/24420 [2:41:33<7:23:13,  1.48s/it]
 27%|██▋       | 6482/24420 [2:41:35<7:23:33,  1.48s/it]
 27%|██▋       | 6483/24420 [2:41:36<7:23:42,  1.48s/it]
 27%|██▋       | 6484/24420 [2:41:37<7:22:35,  1.48s/it]
 27%|██▋       | 6485/24420 [2:41:39<7:23:53,  1.48s/it]
                                                        
{'loss': 0.1405, 'learning_rate': 5e-05, 'epoch': 1.33}

 27%|██▋       | 6485/24420 [2:41:39<7:23:53,  1.48s/it]
 27%|██▋       | 6486/24420 [2:41:40<7:24:04,  1.49s/it]
 27%|██▋       | 6487/24420 [2:41:42<7:23:04,  1.48s/it]
 27%|██▋       | 6488/24420 [2:41:43<7:23:58,  1.49s/it]
 27%|██▋       | 6489/24420 [2:41:45<7:23:32,  1.48s/it]
 27%|██▋       | 6490/24420 [2:41:46<7:23:12,  1.48s/it]
                                                        
{'loss': 0.1335, 'learning_rate': 5e-05, 'epoch': 1.33}

 27%|██▋       | 6490/24420 [2:41:46<7:23:12,  1.48s/it]
 27%|██▋       | 6491/24420 [2:41:48<7:22:51,  1.48s/it]
 27%|██▋       | 6492/24420 [2:41:49<7:22:37,  1.48s/it]
 27%|██▋       | 6493/24420 [2:41:51<7:22:29,  1.48s/it]
 27%|██▋       | 6494/24420 [2:41:52<7:21:23,  1.48s/it]
 27%|██▋       | 6495/24420 [2:41:54<7:21:08,  1.48s/it]
                                                        
{'loss': 0.1169, 'learning_rate': 5e-05, 'epoch': 1.33}

 27%|██▋       | 6495/24420 [2:41:54<7:21:08,  1.48s/it]
 27%|██▋       | 6496/24420 [2:41:55<7:21:12,  1.48s/it]
 27%|██▋       | 6497/24420 [2:41:57<7:21:01,  1.48s/it]
 27%|██▋       | 6498/24420 [2:41:58<7:21:09,  1.48s/it]
 27%|██▋       | 6499/24420 [2:42:00<7:21:14,  1.48s/it]
 27%|██▋       | 6500/24420 [2:42:01<7:21:08,  1.48s/it]
                                                        
{'loss': 0.1233, 'learning_rate': 5e-05, 'epoch': 1.33}

 27%|██▋       | 6500/24420 [2:42:01<7:21:08,  1.48s/it]
 27%|██▋       | 6501/24420 [2:42:03<7:21:30,  1.48s/it]
 27%|██▋       | 6502/24420 [2:42:04<7:20:58,  1.48s/it]
 27%|██▋       | 6503/24420 [2:42:06<7:22:40,  1.48s/it]
 27%|██▋       | 6504/24420 [2:42:07<7:22:55,  1.48s/it]
 27%|██▋       | 6505/24420 [2:42:09<7:21:57,  1.48s/it]
                                                        
{'loss': 0.1217, 'learning_rate': 5e-05, 'epoch': 1.33}

 27%|██▋       | 6505/24420 [2:42:09<7:21:57,  1.48s/it]
 27%|██▋       | 6506/24420 [2:42:10<7:25:22,  1.49s/it]
 27%|██▋       | 6507/24420 [2:42:12<7:25:09,  1.49s/it]
 27%|██▋       | 6508/24420 [2:42:13<7:24:20,  1.49s/it]
 27%|██▋       | 6509/24420 [2:42:15<7:23:40,  1.49s/it]
 27%|██▋       | 6510/24420 [2:42:16<7:23:45,  1.49s/it]
                                                        
{'loss': 0.1274, 'learning_rate': 5e-05, 'epoch': 1.33}

 27%|██▋       | 6510/24420 [2:42:16<7:23:45,  1.49s/it]
 27%|██▋       | 6511/24420 [2:42:18<7:23:44,  1.49s/it]
 27%|██▋       | 6512/24420 [2:42:19<7:23:48,  1.49s/it]
 27%|██▋       | 6513/24420 [2:42:20<7:23:18,  1.49s/it]
 27%|██▋       | 6514/24420 [2:42:22<7:22:31,  1.48s/it]
 27%|██▋       | 6515/24420 [2:42:23<7:21:52,  1.48s/it]
                                                        
{'loss': 0.1357, 'learning_rate': 5e-05, 'epoch': 1.33}

 27%|██▋       | 6515/24420 [2:42:23<7:21:52,  1.48s/it]
 27%|██▋       | 6516/24420 [2:42:25<7:22:12,  1.48s/it]
 27%|██▋       | 6517/24420 [2:42:26<7:21:13,  1.48s/it]
 27%|██▋       | 6518/24420 [2:42:28<7:20:51,  1.48s/it]
 27%|██▋       | 6519/24420 [2:42:29<7:20:41,  1.48s/it]
 27%|██▋       | 6520/24420 [2:42:31<7:20:27,  1.48s/it]
                                                        
{'loss': 0.1214, 'learning_rate': 5e-05, 'epoch': 1.33}

 27%|██▋       | 6520/24420 [2:42:31<7:20:27,  1.48s/it]
 27%|██▋       | 6521/24420 [2:42:32<7:21:16,  1.48s/it]
 27%|██▋       | 6522/24420 [2:42:34<7:21:50,  1.48s/it]
 27%|██▋       | 6523/24420 [2:42:35<7:22:10,  1.48s/it]
 27%|██▋       | 6524/24420 [2:42:37<7:21:57,  1.48s/it]
 27%|██▋       | 6525/24420 [2:42:38<7:21:03,  1.48s/it]
                                                        
{'loss': 0.1336, 'learning_rate': 5e-05, 'epoch': 1.34}

 27%|██▋       | 6525/24420 [2:42:38<7:21:03,  1.48s/it]
 27%|██▋       | 6526/24420 [2:42:40<7:21:35,  1.48s/it]
 27%|██▋       | 6527/24420 [2:42:41<7:20:53,  1.48s/it]
 27%|██▋       | 6528/24420 [2:42:43<7:20:46,  1.48s/it]
 27%|██▋       | 6529/24420 [2:42:44<7:20:45,  1.48s/it]
 27%|██▋       | 6530/24420 [2:42:46<7:21:09,  1.48s/it]
                                                        
{'loss': 0.1197, 'learning_rate': 5e-05, 'epoch': 1.34}

 27%|██▋       | 6530/24420 [2:42:46<7:21:09,  1.48s/it]
 27%|██▋       | 6531/24420 [2:42:47<7:21:34,  1.48s/it]
 27%|██▋       | 6532/24420 [2:42:49<7:21:00,  1.48s/it]
 27%|██▋       | 6533/24420 [2:42:50<7:21:09,  1.48s/it]
 27%|██▋       | 6534/24420 [2:42:52<7:20:47,  1.48s/it]
 27%|██▋       | 6535/24420 [2:42:53<7:21:38,  1.48s/it]
                                                        
{'loss': 0.1589, 'learning_rate': 5e-05, 'epoch': 1.34}

 27%|██▋       | 6535/24420 [2:42:53<7:21:38,  1.48s/it]
 27%|██▋       | 6536/24420 [2:42:55<7:21:28,  1.48s/it]
 27%|██▋       | 6537/24420 [2:42:56<7:20:57,  1.48s/it]
 27%|██▋       | 6538/24420 [2:42:57<7:21:00,  1.48s/it]
 27%|██▋       | 6539/24420 [2:42:59<7:21:41,  1.48s/it]
 27%|██▋       | 6540/24420 [2:43:00<7:21:56,  1.48s/it]
                                                        
{'loss': 0.1256, 'learning_rate': 5e-05, 'epoch': 1.34}

 27%|██▋       | 6540/24420 [2:43:00<7:21:56,  1.48s/it]
 27%|██▋       | 6541/24420 [2:43:02<7:21:39,  1.48s/it]
 27%|██▋       | 6542/24420 [2:43:03<7:21:30,  1.48s/it]
 27%|██▋       | 6543/24420 [2:43:05<7:20:55,  1.48s/it]
 27%|██▋       | 6544/24420 [2:43:06<7:22:19,  1.48s/it]
 27%|██▋       | 6545/24420 [2:43:08<7:21:49,  1.48s/it]
                                                        
{'loss': 0.1282, 'learning_rate': 5e-05, 'epoch': 1.34}

 27%|██▋       | 6545/24420 [2:43:08<7:21:49,  1.48s/it]
 27%|██▋       | 6546/24420 [2:43:09<7:22:07,  1.48s/it]
 27%|██▋       | 6547/24420 [2:43:11<7:21:25,  1.48s/it]
 27%|██▋       | 6548/24420 [2:43:12<7:21:23,  1.48s/it]
 27%|██▋       | 6549/24420 [2:43:14<7:21:20,  1.48s/it]
 27%|██▋       | 6550/24420 [2:43:15<7:21:14,  1.48s/it]
                                                        
{'loss': 0.1426, 'learning_rate': 5e-05, 'epoch': 1.34}

 27%|██▋       | 6550/24420 [2:43:15<7:21:14,  1.48s/it]
 27%|██▋       | 6551/24420 [2:43:17<7:21:37,  1.48s/it]
 27%|██▋       | 6552/24420 [2:43:18<7:20:45,  1.48s/it]
 27%|██▋       | 6553/24420 [2:43:20<7:21:21,  1.48s/it]
 27%|██▋       | 6554/24420 [2:43:21<7:20:57,  1.48s/it]
 27%|██▋       | 6555/24420 [2:43:23<7:21:11,  1.48s/it]
                                                        
{'loss': 0.1107, 'learning_rate': 5e-05, 'epoch': 1.34}

 27%|██▋       | 6555/24420 [2:43:23<7:21:11,  1.48s/it]
 27%|██▋       | 6556/24420 [2:43:24<7:20:48,  1.48s/it]
 27%|██▋       | 6557/24420 [2:43:26<7:20:34,  1.48s/it]
 27%|██▋       | 6558/24420 [2:43:27<7:20:36,  1.48s/it]
 27%|██▋       | 6559/24420 [2:43:29<7:20:32,  1.48s/it]
 27%|██▋       | 6560/24420 [2:43:30<7:21:07,  1.48s/it]
                                                        
{'loss': 0.1308, 'learning_rate': 5e-05, 'epoch': 1.34}

 27%|██▋       | 6560/24420 [2:43:30<7:21:07,  1.48s/it]
 27%|██▋       | 6561/24420 [2:43:32<7:20:19,  1.48s/it]
 27%|██▋       | 6562/24420 [2:43:33<7:19:49,  1.48s/it]
 27%|██▋       | 6563/24420 [2:43:35<7:25:04,  1.50s/it]
 27%|██▋       | 6564/24420 [2:43:36<7:23:11,  1.49s/it]
 27%|██▋       | 6565/24420 [2:43:38<7:22:22,  1.49s/it]
                                                        
{'loss': 0.138, 'learning_rate': 5e-05, 'epoch': 1.34}

 27%|██▋       | 6565/24420 [2:43:38<7:22:22,  1.49s/it]
 27%|██▋       | 6566/24420 [2:43:39<7:22:26,  1.49s/it]
 27%|██▋       | 6567/24420 [2:43:40<7:21:43,  1.48s/it]
 27%|██▋       | 6568/24420 [2:43:42<7:21:22,  1.48s/it]
 27%|██▋       | 6569/24420 [2:43:43<7:21:22,  1.48s/it]
 27%|██▋       | 6570/24420 [2:43:45<7:21:36,  1.48s/it]
                                                        
{'loss': 0.1212, 'learning_rate': 5e-05, 'epoch': 1.35}

 27%|██▋       | 6570/24420 [2:43:45<7:21:36,  1.48s/it]
 27%|██▋       | 6571/24420 [2:43:46<7:21:45,  1.48s/it]
 27%|██▋       | 6572/24420 [2:43:48<7:21:42,  1.48s/it]
 27%|██▋       | 6573/24420 [2:43:49<7:21:18,  1.48s/it]
 27%|██▋       | 6574/24420 [2:43:51<7:21:08,  1.48s/it]
 27%|██▋       | 6575/24420 [2:43:52<7:21:19,  1.48s/it]
                                                        
{'loss': 0.1509, 'learning_rate': 5e-05, 'epoch': 1.35}

 27%|██▋       | 6575/24420 [2:43:52<7:21:19,  1.48s/it]
 27%|██▋       | 6576/24420 [2:43:54<7:21:05,  1.48s/it]
 27%|██▋       | 6577/24420 [2:43:55<7:21:42,  1.49s/it]
 27%|██▋       | 6578/24420 [2:43:57<7:20:52,  1.48s/it]
 27%|██▋       | 6579/24420 [2:43:58<7:20:39,  1.48s/it]
 27%|██▋       | 6580/24420 [2:44:00<7:20:34,  1.48s/it]
                                                        
{'loss': 0.1496, 'learning_rate': 5e-05, 'epoch': 1.35}

 27%|██▋       | 6580/24420 [2:44:00<7:20:34,  1.48s/it]
 27%|██▋       | 6581/24420 [2:44:01<7:19:58,  1.48s/it]
 27%|██▋       | 6582/24420 [2:44:03<7:20:18,  1.48s/it]
 27%|██▋       | 6583/24420 [2:44:04<7:19:20,  1.48s/it]
 27%|██▋       | 6584/24420 [2:44:06<7:19:07,  1.48s/it]
 27%|██▋       | 6585/24420 [2:44:07<7:19:05,  1.48s/it]
                                                        
{'loss': 0.1238, 'learning_rate': 5e-05, 'epoch': 1.35}

 27%|██▋       | 6585/24420 [2:44:07<7:19:05,  1.48s/it]
 27%|██▋       | 6586/24420 [2:44:09<7:19:57,  1.48s/it]
 27%|██▋       | 6587/24420 [2:44:10<7:19:08,  1.48s/it]
 27%|██▋       | 6588/24420 [2:44:12<7:20:07,  1.48s/it]
 27%|██▋       | 6589/24420 [2:44:13<7:20:03,  1.48s/it]
 27%|██▋       | 6590/24420 [2:44:15<7:19:35,  1.48s/it]
                                                        
{'loss': 0.1368, 'learning_rate': 5e-05, 'epoch': 1.35}

 27%|██▋       | 6590/24420 [2:44:15<7:19:35,  1.48s/it]
 27%|██▋       | 6591/24420 [2:44:16<7:19:41,  1.48s/it]
 27%|██▋       | 6592/24420 [2:44:18<7:19:39,  1.48s/it]
 27%|██▋       | 6593/24420 [2:44:19<7:20:22,  1.48s/it]
 27%|██▋       | 6594/24420 [2:44:20<7:20:28,  1.48s/it]
 27%|██▋       | 6595/24420 [2:44:22<7:20:29,  1.48s/it]
                                                        
{'loss': 0.1258, 'learning_rate': 5e-05, 'epoch': 1.35}

 27%|██▋       | 6595/24420 [2:44:22<7:20:29,  1.48s/it]
 27%|██▋       | 6596/24420 [2:44:23<7:20:26,  1.48s/it]
 27%|██▋       | 6597/24420 [2:44:25<7:19:53,  1.48s/it]
 27%|██▋       | 6598/24420 [2:44:26<7:19:18,  1.48s/it]
 27%|██▋       | 6599/24420 [2:44:28<7:20:00,  1.48s/it]
 27%|██▋       | 6600/24420 [2:44:29<7:19:51,  1.48s/it]
                                                        
{'loss': 0.1238, 'learning_rate': 5e-05, 'epoch': 1.35}

 27%|██▋       | 6600/24420 [2:44:29<7:19:51,  1.48s/it]
 27%|██▋       | 6601/24420 [2:44:34<12:00:56,  2.43s/it]
 27%|██▋       | 6602/24420 [2:44:35<10:36:14,  2.14s/it]
 27%|██▋       | 6603/24420 [2:44:37<9:37:47,  1.95s/it] 
 27%|██▋       | 6604/24420 [2:44:38<8:56:05,  1.81s/it]
 27%|██▋       | 6605/24420 [2:44:40<8:27:26,  1.71s/it]
                                                        
{'loss': 0.132, 'learning_rate': 5e-05, 'epoch': 1.35}

 27%|██▋       | 6605/24420 [2:44:40<8:27:26,  1.71s/it]
 27%|██▋       | 6606/24420 [2:44:41<8:07:34,  1.64s/it]
 27%|██▋       | 6607/24420 [2:44:43<7:53:41,  1.60s/it]
 27%|██▋       | 6608/24420 [2:44:44<7:44:34,  1.56s/it]
 27%|██▋       | 6609/24420 [2:44:46<7:36:51,  1.54s/it]
 27%|██▋       | 6610/24420 [2:44:47<7:32:22,  1.52s/it]
                                                        
{'loss': 0.1517, 'learning_rate': 5e-05, 'epoch': 1.35}

 27%|██▋       | 6610/24420 [2:44:47<7:32:22,  1.52s/it]
 27%|██▋       | 6611/24420 [2:44:49<7:28:35,  1.51s/it]
 27%|██▋       | 6612/24420 [2:44:50<7:25:46,  1.50s/it]
 27%|██▋       | 6613/24420 [2:44:52<7:24:15,  1.50s/it]
 27%|██▋       | 6614/24420 [2:44:53<7:22:36,  1.49s/it]
 27%|██▋       | 6615/24420 [2:44:55<7:21:53,  1.49s/it]
                                                        
{'loss': 0.127, 'learning_rate': 5e-05, 'epoch': 1.35}

 27%|██▋       | 6615/24420 [2:44:55<7:21:53,  1.49s/it]
 27%|██▋       | 6616/24420 [2:44:56<7:21:08,  1.49s/it]
 27%|██▋       | 6617/24420 [2:44:58<7:20:10,  1.48s/it]
 27%|██▋       | 6618/24420 [2:44:59<7:20:05,  1.48s/it]
 27%|██▋       | 6619/24420 [2:45:01<7:19:35,  1.48s/it]
 27%|██▋       | 6620/24420 [2:45:02<7:19:23,  1.48s/it]
                                                        
{'loss': 0.1208, 'learning_rate': 5e-05, 'epoch': 1.36}

 27%|██▋       | 6620/24420 [2:45:02<7:19:23,  1.48s/it]
 27%|██▋       | 6621/24420 [2:45:04<7:19:10,  1.48s/it]
 27%|██▋       | 6622/24420 [2:45:05<7:26:49,  1.51s/it]
 27%|██▋       | 6623/24420 [2:45:07<7:24:26,  1.50s/it]
 27%|██▋       | 6624/24420 [2:45:08<7:22:50,  1.49s/it]
 27%|██▋       | 6625/24420 [2:45:10<7:21:25,  1.49s/it]
                                                        
{'loss': 0.1053, 'learning_rate': 5e-05, 'epoch': 1.36}

 27%|██▋       | 6625/24420 [2:45:10<7:21:25,  1.49s/it]
 27%|██▋       | 6626/24420 [2:45:11<7:20:40,  1.49s/it]
 27%|██▋       | 6627/24420 [2:45:13<7:20:34,  1.49s/it]
 27%|██▋       | 6628/24420 [2:45:14<7:19:50,  1.48s/it]
 27%|██▋       | 6629/24420 [2:45:16<7:21:19,  1.49s/it]
 27%|██▋       | 6630/24420 [2:45:17<7:21:02,  1.49s/it]
                                                        
{'loss': 0.1084, 'learning_rate': 5e-05, 'epoch': 1.36}

 27%|██▋       | 6630/24420 [2:45:17<7:21:02,  1.49s/it]
 27%|██▋       | 6631/24420 [2:45:19<7:21:22,  1.49s/it]
 27%|██▋       | 6632/24420 [2:45:20<7:19:50,  1.48s/it]
 27%|██▋       | 6633/24420 [2:45:22<7:19:40,  1.48s/it]
 27%|██▋       | 6634/24420 [2:45:23<7:19:53,  1.48s/it]
 27%|██▋       | 6635/24420 [2:45:24<7:20:00,  1.48s/it]
                                                        
{'loss': 0.1106, 'learning_rate': 5e-05, 'epoch': 1.36}

 27%|██▋       | 6635/24420 [2:45:24<7:20:00,  1.48s/it]
 27%|██▋       | 6636/24420 [2:45:26<7:20:20,  1.49s/it]
 27%|██▋       | 6637/24420 [2:45:27<7:20:07,  1.49s/it]
 27%|██▋       | 6638/24420 [2:45:29<7:19:01,  1.48s/it]
 27%|██▋       | 6639/24420 [2:45:30<7:18:34,  1.48s/it]
 27%|██▋       | 6640/24420 [2:45:32<7:18:31,  1.48s/it]
                                                        
{'loss': 0.1147, 'learning_rate': 5e-05, 'epoch': 1.36}

 27%|██▋       | 6640/24420 [2:45:32<7:18:31,  1.48s/it]
 27%|██▋       | 6641/24420 [2:45:33<7:19:22,  1.48s/it]
 27%|██▋       | 6642/24420 [2:45:35<7:18:45,  1.48s/it]
 27%|██▋       | 6643/24420 [2:45:36<7:19:16,  1.48s/it]
 27%|██▋       | 6644/24420 [2:45:38<7:18:56,  1.48s/it]
 27%|██▋       | 6645/24420 [2:45:39<7:19:18,  1.48s/it]
                                                        
{'loss': 0.1283, 'learning_rate': 5e-05, 'epoch': 1.36}

 27%|██▋       | 6645/24420 [2:45:39<7:19:18,  1.48s/it]
 27%|██▋       | 6646/24420 [2:45:41<7:18:59,  1.48s/it]
 27%|██▋       | 6647/24420 [2:45:42<7:18:36,  1.48s/it]
 27%|██▋       | 6648/24420 [2:45:44<7:18:46,  1.48s/it]
 27%|██▋       | 6649/24420 [2:45:45<7:18:14,  1.48s/it]
 27%|██▋       | 6650/24420 [2:45:47<7:18:25,  1.48s/it]
                                                        
{'loss': 0.121, 'learning_rate': 5e-05, 'epoch': 1.36}

 27%|██▋       | 6650/24420 [2:45:47<7:18:25,  1.48s/it]
 27%|██▋       | 6651/24420 [2:45:48<7:19:11,  1.48s/it]
 27%|██▋       | 6652/24420 [2:45:50<7:18:30,  1.48s/it]
 27%|██▋       | 6653/24420 [2:45:51<7:18:53,  1.48s/it]
 27%|██▋       | 6654/24420 [2:45:53<7:18:17,  1.48s/it]
 27%|██▋       | 6655/24420 [2:45:54<7:18:59,  1.48s/it]
                                                        
{'loss': 0.1224, 'learning_rate': 5e-05, 'epoch': 1.36}

 27%|██▋       | 6655/24420 [2:45:54<7:18:59,  1.48s/it]
 27%|██▋       | 6656/24420 [2:45:56<7:18:39,  1.48s/it]
 27%|██▋       | 6657/24420 [2:45:57<7:18:30,  1.48s/it]
 27%|██▋       | 6658/24420 [2:45:59<7:18:21,  1.48s/it]
 27%|██▋       | 6659/24420 [2:46:00<7:18:06,  1.48s/it]
 27%|██▋       | 6660/24420 [2:46:02<7:18:07,  1.48s/it]
                                                        
{'loss': 0.1078, 'learning_rate': 5e-05, 'epoch': 1.36}

 27%|██▋       | 6660/24420 [2:46:02<7:18:07,  1.48s/it]
 27%|██▋       | 6661/24420 [2:46:03<7:17:58,  1.48s/it]
 27%|██▋       | 6662/24420 [2:46:04<7:17:41,  1.48s/it]
 27%|██▋       | 6663/24420 [2:46:06<7:17:27,  1.48s/it]
 27%|██▋       | 6664/24420 [2:46:07<7:17:58,  1.48s/it]
 27%|██▋       | 6665/24420 [2:46:09<7:17:53,  1.48s/it]
                                                        
{'loss': 0.1271, 'learning_rate': 5e-05, 'epoch': 1.36}

 27%|██▋       | 6665/24420 [2:46:09<7:17:53,  1.48s/it]
 27%|██▋       | 6666/24420 [2:46:10<7:18:02,  1.48s/it]
 27%|██▋       | 6667/24420 [2:46:12<7:20:19,  1.49s/it]
 27%|██▋       | 6668/24420 [2:46:13<7:19:38,  1.49s/it]
 27%|██▋       | 6669/24420 [2:46:15<7:19:48,  1.49s/it]
 27%|██▋       | 6670/24420 [2:46:16<7:19:44,  1.49s/it]
                                                        
{'loss': 0.1486, 'learning_rate': 5e-05, 'epoch': 1.37}

 27%|██▋       | 6670/24420 [2:46:16<7:19:44,  1.49s/it]
 27%|██▋       | 6671/24420 [2:46:18<7:19:09,  1.48s/it]
 27%|██▋       | 6672/24420 [2:46:19<7:18:52,  1.48s/it]
 27%|██▋       | 6673/24420 [2:46:21<7:18:19,  1.48s/it]
 27%|██▋       | 6674/24420 [2:46:22<7:19:21,  1.49s/it]
 27%|██▋       | 6675/24420 [2:46:24<7:19:13,  1.49s/it]
                                                        
{'loss': 0.1202, 'learning_rate': 5e-05, 'epoch': 1.37}

 27%|██▋       | 6675/24420 [2:46:24<7:19:13,  1.49s/it]
 27%|██▋       | 6676/24420 [2:46:25<7:18:53,  1.48s/it]
 27%|██▋       | 6677/24420 [2:46:27<7:18:57,  1.48s/it]
 27%|██▋       | 6678/24420 [2:46:28<7:18:22,  1.48s/it]
 27%|██▋       | 6679/24420 [2:46:30<7:18:15,  1.48s/it]
 27%|██▋       | 6680/24420 [2:46:31<7:19:10,  1.49s/it]
                                                        
{'loss': 0.117, 'learning_rate': 5e-05, 'epoch': 1.37}

 27%|██▋       | 6680/24420 [2:46:31<7:19:10,  1.49s/it]
 27%|██▋       | 6681/24420 [2:46:33<7:18:29,  1.48s/it]
 27%|██▋       | 6682/24420 [2:46:34<7:18:25,  1.48s/it]
 27%|██▋       | 6683/24420 [2:46:36<7:18:01,  1.48s/it]
 27%|██▋       | 6684/24420 [2:46:37<7:17:48,  1.48s/it]
 27%|██▋       | 6685/24420 [2:46:39<7:18:48,  1.48s/it]
                                                        
{'loss': 0.106, 'learning_rate': 5e-05, 'epoch': 1.37}

 27%|██▋       | 6685/24420 [2:46:39<7:18:48,  1.48s/it]
 27%|██▋       | 6686/24420 [2:46:40<7:19:00,  1.49s/it]
 27%|██▋       | 6687/24420 [2:46:42<7:18:30,  1.48s/it]
 27%|██▋       | 6688/24420 [2:46:43<7:18:38,  1.48s/it]
 27%|██▋       | 6689/24420 [2:46:45<7:18:31,  1.48s/it]
 27%|██▋       | 6690/24420 [2:46:46<7:17:41,  1.48s/it]
                                                        
{'loss': 0.1252, 'learning_rate': 5e-05, 'epoch': 1.37}

 27%|██▋       | 6690/24420 [2:46:46<7:17:41,  1.48s/it]
 27%|██▋       | 6691/24420 [2:46:48<7:17:42,  1.48s/it]
 27%|██▋       | 6692/24420 [2:46:49<7:17:15,  1.48s/it]
 27%|██▋       | 6693/24420 [2:46:50<7:17:12,  1.48s/it]
 27%|██▋       | 6694/24420 [2:46:52<7:18:30,  1.48s/it]
 27%|██▋       | 6695/24420 [2:46:53<7:18:26,  1.48s/it]
                                                        
{'loss': 0.1286, 'learning_rate': 5e-05, 'epoch': 1.37}

 27%|██▋       | 6695/24420 [2:46:53<7:18:26,  1.48s/it]
 27%|██▋       | 6696/24420 [2:46:55<7:18:58,  1.49s/it]
 27%|██▋       | 6697/24420 [2:46:56<7:20:01,  1.49s/it]
 27%|██▋       | 6698/24420 [2:46:58<7:19:31,  1.49s/it]
 27%|██▋       | 6699/24420 [2:46:59<7:19:20,  1.49s/it]
 27%|██▋       | 6700/24420 [2:47:01<7:18:23,  1.48s/it]
                                                        
{'loss': 0.1408, 'learning_rate': 5e-05, 'epoch': 1.37}

 27%|██▋       | 6700/24420 [2:47:01<7:18:23,  1.48s/it]
 27%|██▋       | 6701/24420 [2:47:02<7:18:29,  1.48s/it]
 27%|██▋       | 6702/24420 [2:47:04<7:18:13,  1.48s/it]
 27%|██▋       | 6703/24420 [2:47:05<7:18:25,  1.48s/it]
 27%|██▋       | 6704/24420 [2:47:07<7:18:52,  1.49s/it]
 27%|██▋       | 6705/24420 [2:47:08<7:18:57,  1.49s/it]
                                                        
{'loss': 0.1232, 'learning_rate': 5e-05, 'epoch': 1.37}

 27%|██▋       | 6705/24420 [2:47:08<7:18:57,  1.49s/it]
 27%|██▋       | 6706/24420 [2:47:10<7:18:22,  1.48s/it]
 27%|██▋       | 6707/24420 [2:47:11<7:18:08,  1.48s/it]
 27%|██▋       | 6708/24420 [2:47:13<7:17:51,  1.48s/it]
 27%|██▋       | 6709/24420 [2:47:14<7:18:08,  1.48s/it]
 27%|██▋       | 6710/24420 [2:47:16<7:18:58,  1.49s/it]
                                                        
{'loss': 0.133, 'learning_rate': 5e-05, 'epoch': 1.37}

 27%|██▋       | 6710/24420 [2:47:16<7:18:58,  1.49s/it]
 27%|██▋       | 6711/24420 [2:47:17<7:18:21,  1.49s/it]
 27%|██▋       | 6712/24420 [2:47:19<7:17:55,  1.48s/it]
 27%|██▋       | 6713/24420 [2:47:20<7:17:45,  1.48s/it]
 27%|██▋       | 6714/24420 [2:47:22<7:17:51,  1.48s/it]
 27%|██▋       | 6715/24420 [2:47:23<7:17:39,  1.48s/it]
                                                        
{'loss': 0.1266, 'learning_rate': 5e-05, 'epoch': 1.37}

 27%|██▋       | 6715/24420 [2:47:23<7:17:39,  1.48s/it]
 28%|██▊       | 6716/24420 [2:47:25<7:17:24,  1.48s/it]
 28%|██▊       | 6717/24420 [2:47:26<7:17:47,  1.48s/it]
 28%|██▊       | 6718/24420 [2:47:28<7:17:19,  1.48s/it]
 28%|██▊       | 6719/24420 [2:47:29<7:18:00,  1.48s/it]
 28%|██▊       | 6720/24420 [2:47:31<7:17:53,  1.48s/it]
                                                        
{'loss': 0.1335, 'learning_rate': 5e-05, 'epoch': 1.38}

 28%|██▊       | 6720/24420 [2:47:31<7:17:53,  1.48s/it]
 28%|██▊       | 6721/24420 [2:47:32<7:17:15,  1.48s/it]
 28%|██▊       | 6722/24420 [2:47:34<7:17:22,  1.48s/it]
 28%|██▊       | 6723/24420 [2:47:35<7:16:46,  1.48s/it]
 28%|██▊       | 6724/24420 [2:47:37<7:19:09,  1.49s/it]
 28%|██▊       | 6725/24420 [2:47:38<7:18:33,  1.49s/it]
                                                        
{'loss': 0.1365, 'learning_rate': 5e-05, 'epoch': 1.38}

 28%|██▊       | 6725/24420 [2:47:38<7:18:33,  1.49s/it]
 28%|██▊       | 6726/24420 [2:47:39<7:19:03,  1.49s/it]
 28%|██▊       | 6727/24420 [2:47:41<7:18:31,  1.49s/it]
 28%|██▊       | 6728/24420 [2:47:42<7:18:18,  1.49s/it]
 28%|██▊       | 6729/24420 [2:47:44<7:17:48,  1.48s/it]
 28%|██▊       | 6730/24420 [2:47:45<7:17:18,  1.48s/it]
                                                        
{'loss': 0.1111, 'learning_rate': 5e-05, 'epoch': 1.38}

 28%|██▊       | 6730/24420 [2:47:45<7:17:18,  1.48s/it]
 28%|██▊       | 6731/24420 [2:47:47<7:17:35,  1.48s/it]
 28%|██▊       | 6732/24420 [2:47:48<7:18:11,  1.49s/it]
 28%|██▊       | 6733/24420 [2:47:50<7:17:28,  1.48s/it]
 28%|██▊       | 6734/24420 [2:47:51<7:16:54,  1.48s/it]
 28%|██▊       | 6735/24420 [2:47:53<7:17:11,  1.48s/it]
                                                        
{'loss': 0.1088, 'learning_rate': 5e-05, 'epoch': 1.38}

 28%|██▊       | 6735/24420 [2:47:53<7:17:11,  1.48s/it]
 28%|██▊       | 6736/24420 [2:47:54<7:17:17,  1.48s/it]
 28%|██▊       | 6737/24420 [2:47:56<7:16:56,  1.48s/it]
 28%|██▊       | 6738/24420 [2:47:57<7:16:57,  1.48s/it]
 28%|██▊       | 6739/24420 [2:47:59<7:16:53,  1.48s/it]
 28%|██▊       | 6740/24420 [2:48:00<7:16:45,  1.48s/it]
                                                        
{'loss': 0.1388, 'learning_rate': 5e-05, 'epoch': 1.38}

 28%|██▊       | 6740/24420 [2:48:00<7:16:45,  1.48s/it]
 28%|██▊       | 6741/24420 [2:48:02<7:17:35,  1.49s/it]
 28%|██▊       | 6742/24420 [2:48:03<7:17:07,  1.48s/it]
 28%|██▊       | 6743/24420 [2:48:05<7:16:43,  1.48s/it]
 28%|██▊       | 6744/24420 [2:48:06<7:16:51,  1.48s/it]
 28%|██▊       | 6745/24420 [2:48:08<7:16:34,  1.48s/it]
                                                        
{'loss': 0.1194, 'learning_rate': 5e-05, 'epoch': 1.38}

 28%|██▊       | 6745/24420 [2:48:08<7:16:34,  1.48s/it]
 28%|██▊       | 6746/24420 [2:48:09<7:16:18,  1.48s/it]
 28%|██▊       | 6747/24420 [2:48:11<7:16:18,  1.48s/it]
 28%|██▊       | 6748/24420 [2:48:12<7:17:13,  1.48s/it]
 28%|██▊       | 6749/24420 [2:48:14<7:16:55,  1.48s/it]
 28%|██▊       | 6750/24420 [2:48:15<7:16:19,  1.48s/it]
                                                        
{'loss': 0.1422, 'learning_rate': 5e-05, 'epoch': 1.38}

 28%|██▊       | 6750/24420 [2:48:15<7:16:19,  1.48s/it]
 28%|██▊       | 6751/24420 [2:48:17<7:16:55,  1.48s/it]
 28%|██▊       | 6752/24420 [2:48:18<7:16:31,  1.48s/it]
 28%|██▊       | 6753/24420 [2:48:20<7:16:45,  1.48s/it]
 28%|██▊       | 6754/24420 [2:48:21<7:16:04,  1.48s/it]
 28%|██▊       | 6755/24420 [2:48:22<7:16:06,  1.48s/it]
                                                        
{'loss': 0.135, 'learning_rate': 5e-05, 'epoch': 1.38}

 28%|██▊       | 6755/24420 [2:48:22<7:16:06,  1.48s/it]
 28%|██▊       | 6756/24420 [2:48:24<7:16:05,  1.48s/it]
 28%|██▊       | 6757/24420 [2:48:25<7:16:36,  1.48s/it]
 28%|██▊       | 6758/24420 [2:48:27<7:16:29,  1.48s/it]
 28%|██▊       | 6759/24420 [2:48:28<7:16:43,  1.48s/it]
 28%|██▊       | 6760/24420 [2:48:30<7:16:31,  1.48s/it]
                                                        
{'loss': 0.1179, 'learning_rate': 5e-05, 'epoch': 1.38}

 28%|██▊       | 6760/24420 [2:48:30<7:16:31,  1.48s/it]
 28%|██▊       | 6761/24420 [2:48:31<7:16:50,  1.48s/it]
 28%|██▊       | 6762/24420 [2:48:33<7:16:41,  1.48s/it]
 28%|██▊       | 6763/24420 [2:48:34<7:16:29,  1.48s/it]
 28%|██▊       | 6764/24420 [2:48:36<7:16:41,  1.48s/it]
 28%|██▊       | 6765/24420 [2:48:37<7:18:43,  1.49s/it]
                                                        
{'loss': 0.1822, 'learning_rate': 5e-05, 'epoch': 1.39}

 28%|██▊       | 6765/24420 [2:48:37<7:18:43,  1.49s/it]
 28%|██▊       | 6766/24420 [2:48:39<7:18:13,  1.49s/it]
 28%|██▊       | 6767/24420 [2:48:40<7:17:43,  1.49s/it]
 28%|██▊       | 6768/24420 [2:48:42<7:17:39,  1.49s/it]
 28%|██▊       | 6769/24420 [2:48:43<7:16:28,  1.48s/it]
 28%|██▊       | 6770/24420 [2:48:45<7:15:19,  1.48s/it]
                                                        
{'loss': 0.1247, 'learning_rate': 5e-05, 'epoch': 1.39}

 28%|██▊       | 6770/24420 [2:48:45<7:15:19,  1.48s/it]
 28%|██▊       | 6771/24420 [2:48:46<7:15:26,  1.48s/it]
 28%|██▊       | 6772/24420 [2:48:48<7:15:57,  1.48s/it]
 28%|██▊       | 6773/24420 [2:48:49<7:15:19,  1.48s/it]
 28%|██▊       | 6774/24420 [2:48:51<7:14:44,  1.48s/it]
 28%|██▊       | 6775/24420 [2:48:52<7:14:54,  1.48s/it]
                                                        
{'loss': 0.1218, 'learning_rate': 5e-05, 'epoch': 1.39}

 28%|██▊       | 6775/24420 [2:48:52<7:14:54,  1.48s/it]
 28%|██▊       | 6776/24420 [2:48:54<7:15:42,  1.48s/it]
 28%|██▊       | 6777/24420 [2:48:55<7:15:57,  1.48s/it]
 28%|██▊       | 6778/24420 [2:48:57<7:14:54,  1.48s/it]
 28%|██▊       | 6779/24420 [2:48:58<7:15:08,  1.48s/it]
 28%|██▊       | 6780/24420 [2:49:00<7:15:03,  1.48s/it]
                                                        
{'loss': 0.1004, 'learning_rate': 5e-05, 'epoch': 1.39}

 28%|██▊       | 6780/24420 [2:49:00<7:15:03,  1.48s/it]
 28%|██▊       | 6781/24420 [2:49:01<7:15:45,  1.48s/it]
 28%|██▊       | 6782/24420 [2:49:03<7:15:46,  1.48s/it]
 28%|██▊       | 6783/24420 [2:49:04<7:16:19,  1.48s/it]
 28%|██▊       | 6784/24420 [2:49:05<7:15:42,  1.48s/it]
 28%|██▊       | 6785/24420 [2:49:07<7:15:21,  1.48s/it]
                                                        
{'loss': 0.1353, 'learning_rate': 5e-05, 'epoch': 1.39}

 28%|██▊       | 6785/24420 [2:49:07<7:15:21,  1.48s/it]
 28%|██▊       | 6786/24420 [2:49:08<7:14:16,  1.48s/it]
 28%|██▊       | 6787/24420 [2:49:10<7:14:42,  1.48s/it]
 28%|██▊       | 6788/24420 [2:49:11<7:14:52,  1.48s/it]
 28%|██▊       | 6789/24420 [2:49:13<7:14:35,  1.48s/it]
 28%|██▊       | 6790/24420 [2:49:14<7:14:20,  1.48s/it]
                                                        
{'loss': 0.1256, 'learning_rate': 5e-05, 'epoch': 1.39}

 28%|██▊       | 6790/24420 [2:49:14<7:14:20,  1.48s/it]
 28%|██▊       | 6791/24420 [2:49:16<7:14:40,  1.48s/it]
 28%|██▊       | 6792/24420 [2:49:17<7:15:25,  1.48s/it]
 28%|██▊       | 6793/24420 [2:49:19<7:15:41,  1.48s/it]
 28%|██▊       | 6794/24420 [2:49:20<7:15:59,  1.48s/it]
 28%|██▊       | 6795/24420 [2:49:22<7:15:36,  1.48s/it]
                                                        
{'loss': 0.125, 'learning_rate': 5e-05, 'epoch': 1.39}

 28%|██▊       | 6795/24420 [2:49:22<7:15:36,  1.48s/it]
 28%|██▊       | 6796/24420 [2:49:23<7:15:16,  1.48s/it]
 28%|██▊       | 6797/24420 [2:49:25<7:14:36,  1.48s/it]
 28%|██▊       | 6798/24420 [2:49:26<7:14:08,  1.48s/it]
 28%|██▊       | 6799/24420 [2:49:28<7:15:08,  1.48s/it]
 28%|██▊       | 6800/24420 [2:49:29<7:14:57,  1.48s/it]
                                                        
{'loss': 0.128, 'learning_rate': 5e-05, 'epoch': 1.39}

 28%|██▊       | 6800/24420 [2:49:29<7:14:57,  1.48s/it]
 28%|██▊       | 6801/24420 [2:49:34<11:51:26,  2.42s/it]
 28%|██▊       | 6802/24420 [2:49:35<10:28:27,  2.14s/it]
 28%|██▊       | 6803/24420 [2:49:37<9:30:54,  1.94s/it] 
 28%|██▊       | 6804/24420 [2:49:38<8:49:55,  1.80s/it]
 28%|██▊       | 6805/24420 [2:49:40<8:20:40,  1.71s/it]
                                                        
{'loss': 0.1015, 'learning_rate': 5e-05, 'epoch': 1.39}

 28%|██▊       | 6805/24420 [2:49:40<8:20:40,  1.71s/it]
 28%|██▊       | 6806/24420 [2:49:41<8:00:56,  1.64s/it]
 28%|██▊       | 6807/24420 [2:49:43<7:48:29,  1.60s/it]
 28%|██▊       | 6808/24420 [2:49:44<7:38:14,  1.56s/it]
 28%|██▊       | 6809/24420 [2:49:46<7:30:54,  1.54s/it]
 28%|██▊       | 6810/24420 [2:49:47<7:26:11,  1.52s/it]
                                                        
{'loss': 0.1282, 'learning_rate': 5e-05, 'epoch': 1.39}

 28%|██▊       | 6810/24420 [2:49:47<7:26:11,  1.52s/it]
 28%|██▊       | 6811/24420 [2:49:49<7:22:31,  1.51s/it]
 28%|██▊       | 6812/24420 [2:49:50<7:20:26,  1.50s/it]
 28%|██▊       | 6813/24420 [2:49:52<7:18:32,  1.49s/it]
 28%|██▊       | 6814/24420 [2:49:53<7:17:33,  1.49s/it]
 28%|██▊       | 6815/24420 [2:49:55<7:17:45,  1.49s/it]
                                                        
{'loss': 0.1302, 'learning_rate': 5e-05, 'epoch': 1.4}

 28%|██▊       | 6815/24420 [2:49:55<7:17:45,  1.49s/it]
 28%|██▊       | 6816/24420 [2:49:56<7:16:58,  1.49s/it]
 28%|██▊       | 6817/24420 [2:49:58<7:16:50,  1.49s/it]
 28%|██▊       | 6818/24420 [2:49:59<7:17:16,  1.49s/it]
 28%|██▊       | 6819/24420 [2:50:01<7:16:38,  1.49s/it]
 28%|██▊       | 6820/24420 [2:50:02<7:16:58,  1.49s/it]
                                                        
{'loss': 0.1512, 'learning_rate': 5e-05, 'epoch': 1.4}

 28%|██▊       | 6820/24420 [2:50:02<7:16:58,  1.49s/it]
 28%|██▊       | 6821/24420 [2:50:03<7:16:29,  1.49s/it]
 28%|██▊       | 6822/24420 [2:50:05<7:15:58,  1.49s/it]
 28%|██▊       | 6823/24420 [2:50:06<7:15:28,  1.48s/it]
 28%|██▊       | 6824/24420 [2:50:08<7:14:56,  1.48s/it]
 28%|██▊       | 6825/24420 [2:50:09<7:15:15,  1.48s/it]
                                                        
{'loss': 0.1444, 'learning_rate': 5e-05, 'epoch': 1.4}

 28%|██▊       | 6825/24420 [2:50:09<7:15:15,  1.48s/it]
 28%|██▊       | 6826/24420 [2:50:11<7:14:35,  1.48s/it]
 28%|██▊       | 6827/24420 [2:50:12<7:15:54,  1.49s/it]
 28%|██▊       | 6828/24420 [2:50:14<7:15:04,  1.48s/it]
 28%|██▊       | 6829/24420 [2:50:15<7:14:12,  1.48s/it]
 28%|██▊       | 6830/24420 [2:50:17<7:13:51,  1.48s/it]
                                                        
{'loss': 0.0845, 'learning_rate': 5e-05, 'epoch': 1.4}

 28%|██▊       | 6830/24420 [2:50:17<7:13:51,  1.48s/it]
 28%|██▊       | 6831/24420 [2:50:18<7:15:11,  1.48s/it]
 28%|██▊       | 6832/24420 [2:50:20<7:14:23,  1.48s/it]
 28%|██▊       | 6833/24420 [2:50:21<7:14:56,  1.48s/it]
 28%|██▊       | 6834/24420 [2:50:23<7:14:52,  1.48s/it]
 28%|██▊       | 6835/24420 [2:50:24<7:14:41,  1.48s/it]
                                                        
{'loss': 0.1108, 'learning_rate': 5e-05, 'epoch': 1.4}

 28%|██▊       | 6835/24420 [2:50:24<7:14:41,  1.48s/it]
 28%|██▊       | 6836/24420 [2:50:26<7:13:58,  1.48s/it]
 28%|██▊       | 6837/24420 [2:50:27<7:14:25,  1.48s/it]
 28%|██▊       | 6838/24420 [2:50:29<7:14:27,  1.48s/it]
 28%|██▊       | 6839/24420 [2:50:30<7:14:02,  1.48s/it]
 28%|██▊       | 6840/24420 [2:50:32<7:14:05,  1.48s/it]
                                                        
{'loss': 0.145, 'learning_rate': 5e-05, 'epoch': 1.4}

 28%|██▊       | 6840/24420 [2:50:32<7:14:05,  1.48s/it]
 28%|██▊       | 6841/24420 [2:50:33<7:13:48,  1.48s/it]
 28%|██▊       | 6842/24420 [2:50:35<7:13:47,  1.48s/it]
 28%|██▊       | 6843/24420 [2:50:36<7:13:29,  1.48s/it]
 28%|██▊       | 6844/24420 [2:50:38<7:13:01,  1.48s/it]
 28%|██▊       | 6845/24420 [2:50:39<7:13:03,  1.48s/it]
                                                        
{'loss': 0.1401, 'learning_rate': 5e-05, 'epoch': 1.4}

 28%|██▊       | 6845/24420 [2:50:39<7:13:03,  1.48s/it]
 28%|██▊       | 6846/24420 [2:50:41<7:12:13,  1.48s/it]
 28%|██▊       | 6847/24420 [2:50:42<7:12:01,  1.48s/it]
 28%|██▊       | 6848/24420 [2:50:43<7:12:09,  1.48s/it]
 28%|██▊       | 6849/24420 [2:50:45<7:13:28,  1.48s/it]
 28%|██▊       | 6850/24420 [2:50:46<7:13:29,  1.48s/it]
                                                        
{'loss': 0.1291, 'learning_rate': 5e-05, 'epoch': 1.4}

 28%|██▊       | 6850/24420 [2:50:46<7:13:29,  1.48s/it]
 28%|██▊       | 6851/24420 [2:50:48<7:13:03,  1.48s/it]
 28%|██▊       | 6852/24420 [2:50:49<7:14:41,  1.48s/it]
 28%|██▊       | 6853/24420 [2:50:51<7:14:40,  1.48s/it]
 28%|██▊       | 6854/24420 [2:50:52<7:14:26,  1.48s/it]
 28%|██▊       | 6855/24420 [2:50:54<7:13:50,  1.48s/it]
                                                        
{'loss': 0.122, 'learning_rate': 5e-05, 'epoch': 1.4}

 28%|██▊       | 6855/24420 [2:50:54<7:13:50,  1.48s/it]
 28%|██▊       | 6856/24420 [2:50:55<7:13:28,  1.48s/it]
 28%|██▊       | 6857/24420 [2:50:57<7:13:26,  1.48s/it]
 28%|██▊       | 6858/24420 [2:50:58<7:14:57,  1.49s/it]
 28%|██▊       | 6859/24420 [2:51:00<7:13:56,  1.48s/it]
 28%|██▊       | 6860/24420 [2:51:01<7:13:52,  1.48s/it]
                                                        
{'loss': 0.1306, 'learning_rate': 5e-05, 'epoch': 1.4}

 28%|██▊       | 6860/24420 [2:51:01<7:13:52,  1.48s/it]
 28%|██▊       | 6861/24420 [2:51:03<7:13:13,  1.48s/it]
 28%|██▊       | 6862/24420 [2:51:04<7:13:06,  1.48s/it]
 28%|██▊       | 6863/24420 [2:51:06<7:13:32,  1.48s/it]
 28%|██▊       | 6864/24420 [2:51:07<7:14:05,  1.48s/it]
 28%|██▊       | 6865/24420 [2:51:09<7:14:24,  1.48s/it]
                                                        
{'loss': 0.123, 'learning_rate': 5e-05, 'epoch': 1.41}

 28%|██▊       | 6865/24420 [2:51:09<7:14:24,  1.48s/it]
 28%|██▊       | 6866/24420 [2:51:10<7:14:48,  1.49s/it]
 28%|██▊       | 6867/24420 [2:51:12<7:13:58,  1.48s/it]
 28%|██▊       | 6868/24420 [2:51:13<7:14:09,  1.48s/it]
 28%|██▊       | 6869/24420 [2:51:15<7:13:18,  1.48s/it]
 28%|██▊       | 6870/24420 [2:51:16<7:13:18,  1.48s/it]
                                                        
{'loss': 0.1307, 'learning_rate': 5e-05, 'epoch': 1.41}

 28%|██▊       | 6870/24420 [2:51:16<7:13:18,  1.48s/it]
 28%|██▊       | 6871/24420 [2:51:18<7:12:50,  1.48s/it]
 28%|██▊       | 6872/24420 [2:51:19<7:12:35,  1.48s/it]
 28%|██▊       | 6873/24420 [2:51:21<7:11:52,  1.48s/it]
 28%|██▊       | 6874/24420 [2:51:22<7:11:30,  1.48s/it]
 28%|██▊       | 6875/24420 [2:51:23<7:12:50,  1.48s/it]
                                                        
{'loss': 0.1141, 'learning_rate': 5e-05, 'epoch': 1.41}

 28%|██▊       | 6875/24420 [2:51:23<7:12:50,  1.48s/it]
 28%|██▊       | 6876/24420 [2:51:25<7:13:30,  1.48s/it]
 28%|██▊       | 6877/24420 [2:51:26<7:13:28,  1.48s/it]
 28%|██▊       | 6878/24420 [2:51:28<7:14:28,  1.49s/it]
 28%|██▊       | 6879/24420 [2:51:29<7:13:33,  1.48s/it]
 28%|██▊       | 6880/24420 [2:51:31<7:13:48,  1.48s/it]
                                                        
{'loss': 0.1409, 'learning_rate': 5e-05, 'epoch': 1.41}

 28%|██▊       | 6880/24420 [2:51:31<7:13:48,  1.48s/it]
 28%|██▊       | 6881/24420 [2:51:32<7:14:06,  1.49s/it]
 28%|██▊       | 6882/24420 [2:51:34<7:13:49,  1.48s/it]
 28%|██▊       | 6883/24420 [2:51:35<7:13:09,  1.48s/it]
 28%|██▊       | 6884/24420 [2:51:37<7:13:34,  1.48s/it]
 28%|██▊       | 6885/24420 [2:51:38<7:14:05,  1.49s/it]
                                                        
{'loss': 0.106, 'learning_rate': 5e-05, 'epoch': 1.41}

 28%|██▊       | 6885/24420 [2:51:38<7:14:05,  1.49s/it]
 28%|██▊       | 6886/24420 [2:51:40<7:13:47,  1.48s/it]
 28%|██▊       | 6887/24420 [2:51:41<7:12:56,  1.48s/it]
 28%|██▊       | 6888/24420 [2:51:43<7:12:40,  1.48s/it]
 28%|██▊       | 6889/24420 [2:51:44<7:12:35,  1.48s/it]
 28%|██▊       | 6890/24420 [2:51:46<7:12:12,  1.48s/it]
                                                        
{'loss': 0.1256, 'learning_rate': 5e-05, 'epoch': 1.41}

 28%|██▊       | 6890/24420 [2:51:46<7:12:12,  1.48s/it]
 28%|██▊       | 6891/24420 [2:51:47<7:12:01,  1.48s/it]
 28%|██▊       | 6892/24420 [2:51:49<7:11:45,  1.48s/it]
 28%|██▊       | 6893/24420 [2:51:50<7:12:31,  1.48s/it]
 28%|██▊       | 6894/24420 [2:51:52<7:12:50,  1.48s/it]
 28%|██▊       | 6895/24420 [2:51:53<7:12:49,  1.48s/it]
                                                        
{'loss': 0.126, 'learning_rate': 5e-05, 'epoch': 1.41}

 28%|██▊       | 6895/24420 [2:51:53<7:12:49,  1.48s/it]
 28%|██▊       | 6896/24420 [2:51:55<7:12:22,  1.48s/it]
 28%|██▊       | 6897/24420 [2:51:56<7:12:11,  1.48s/it]
 28%|██▊       | 6898/24420 [2:51:58<7:12:20,  1.48s/it]
 28%|██▊       | 6899/24420 [2:51:59<7:12:46,  1.48s/it]
 28%|██▊       | 6900/24420 [2:52:01<7:13:41,  1.49s/it]
                                                        
{'loss': 0.1265, 'learning_rate': 5e-05, 'epoch': 1.41}

 28%|██▊       | 6900/24420 [2:52:01<7:13:41,  1.49s/it]
 28%|██▊       | 6901/24420 [2:52:02<7:13:01,  1.48s/it]
 28%|██▊       | 6902/24420 [2:52:03<7:12:32,  1.48s/it]
 28%|██▊       | 6903/24420 [2:52:05<7:12:20,  1.48s/it]
 28%|██▊       | 6904/24420 [2:52:06<7:12:19,  1.48s/it]
 28%|██▊       | 6905/24420 [2:52:08<7:12:15,  1.48s/it]
                                                        
{'loss': 0.1183, 'learning_rate': 5e-05, 'epoch': 1.41}

 28%|██▊       | 6905/24420 [2:52:08<7:12:15,  1.48s/it]
 28%|██▊       | 6906/24420 [2:52:09<7:12:04,  1.48s/it]
 28%|██▊       | 6907/24420 [2:52:11<7:12:06,  1.48s/it]
 28%|██▊       | 6908/24420 [2:52:12<7:12:48,  1.48s/it]
 28%|██▊       | 6909/24420 [2:52:14<7:11:41,  1.48s/it]
 28%|██▊       | 6910/24420 [2:52:15<7:11:52,  1.48s/it]
                                                        
{'loss': 0.1496, 'learning_rate': 5e-05, 'epoch': 1.41}

 28%|██▊       | 6910/24420 [2:52:15<7:11:52,  1.48s/it]
 28%|██▊       | 6911/24420 [2:52:17<7:11:52,  1.48s/it]
 28%|██▊       | 6912/24420 [2:52:18<7:12:34,  1.48s/it]
 28%|██▊       | 6913/24420 [2:52:20<7:12:59,  1.48s/it]
 28%|██▊       | 6914/24420 [2:52:21<7:13:36,  1.49s/it]
 28%|██▊       | 6915/24420 [2:52:23<7:12:43,  1.48s/it]
                                                        
{'loss': 0.12, 'learning_rate': 5e-05, 'epoch': 1.42}

 28%|██▊       | 6915/24420 [2:52:23<7:12:43,  1.48s/it]
 28%|██▊       | 6916/24420 [2:52:24<7:13:08,  1.48s/it]
 28%|██▊       | 6917/24420 [2:52:26<7:12:41,  1.48s/it]
 28%|██▊       | 6918/24420 [2:52:27<7:12:29,  1.48s/it]
 28%|██▊       | 6919/24420 [2:52:29<7:11:19,  1.48s/it]
 28%|██▊       | 6920/24420 [2:52:30<7:11:53,  1.48s/it]
                                                        
{'loss': 0.1249, 'learning_rate': 5e-05, 'epoch': 1.42}

 28%|██▊       | 6920/24420 [2:52:30<7:11:53,  1.48s/it]
 28%|██▊       | 6921/24420 [2:52:32<7:11:44,  1.48s/it]
 28%|██▊       | 6922/24420 [2:52:33<7:11:50,  1.48s/it]
 28%|██▊       | 6923/24420 [2:52:35<7:12:02,  1.48s/it]
 28%|██▊       | 6924/24420 [2:52:36<7:12:50,  1.48s/it]
 28%|██▊       | 6925/24420 [2:52:38<7:12:03,  1.48s/it]
                                                        
{'loss': 0.1214, 'learning_rate': 5e-05, 'epoch': 1.42}

 28%|██▊       | 6925/24420 [2:52:38<7:12:03,  1.48s/it]
 28%|██▊       | 6926/24420 [2:52:39<7:11:55,  1.48s/it]
 28%|██▊       | 6927/24420 [2:52:41<7:11:53,  1.48s/it]
 28%|██▊       | 6928/24420 [2:52:42<7:12:15,  1.48s/it]
 28%|██▊       | 6929/24420 [2:52:43<7:11:50,  1.48s/it]
 28%|██▊       | 6930/24420 [2:52:45<7:11:39,  1.48s/it]
                                                        
{'loss': 0.1141, 'learning_rate': 5e-05, 'epoch': 1.42}

 28%|██▊       | 6930/24420 [2:52:45<7:11:39,  1.48s/it]
 28%|██▊       | 6931/24420 [2:52:46<7:11:26,  1.48s/it]
 28%|██▊       | 6932/24420 [2:52:48<7:11:28,  1.48s/it]
 28%|██▊       | 6933/24420 [2:52:49<7:11:18,  1.48s/it]
 28%|██▊       | 6934/24420 [2:52:51<7:11:02,  1.48s/it]
 28%|██▊       | 6935/24420 [2:52:52<7:10:51,  1.48s/it]
                                                        
{'loss': 0.1198, 'learning_rate': 5e-05, 'epoch': 1.42}

 28%|██▊       | 6935/24420 [2:52:52<7:10:51,  1.48s/it]
 28%|██▊       | 6936/24420 [2:52:54<7:10:51,  1.48s/it]
 28%|██▊       | 6937/24420 [2:52:55<7:10:54,  1.48s/it]
 28%|██▊       | 6938/24420 [2:52:57<7:12:07,  1.48s/it]
 28%|██▊       | 6939/24420 [2:52:58<7:12:04,  1.48s/it]
 28%|██▊       | 6940/24420 [2:53:00<7:12:07,  1.48s/it]
                                                        
{'loss': 0.1231, 'learning_rate': 5e-05, 'epoch': 1.42}

 28%|██▊       | 6940/24420 [2:53:00<7:12:07,  1.48s/it]
 28%|██▊       | 6941/24420 [2:53:01<7:11:36,  1.48s/it]
 28%|██▊       | 6942/24420 [2:53:03<7:11:19,  1.48s/it]
 28%|██▊       | 6943/24420 [2:53:04<7:11:30,  1.48s/it]
 28%|██▊       | 6944/24420 [2:53:06<7:12:19,  1.48s/it]
 28%|██▊       | 6945/24420 [2:53:07<7:13:19,  1.49s/it]
                                                        
{'loss': 0.1258, 'learning_rate': 5e-05, 'epoch': 1.42}

 28%|██▊       | 6945/24420 [2:53:07<7:13:19,  1.49s/it]
 28%|██▊       | 6946/24420 [2:53:09<7:12:43,  1.49s/it]
 28%|██▊       | 6947/24420 [2:53:10<7:12:56,  1.49s/it]
 28%|██▊       | 6948/24420 [2:53:12<7:12:33,  1.49s/it]
 28%|██▊       | 6949/24420 [2:53:13<7:12:31,  1.49s/it]
 28%|██▊       | 6950/24420 [2:53:15<7:12:11,  1.48s/it]
                                                        
{'loss': 0.1523, 'learning_rate': 5e-05, 'epoch': 1.42}

 28%|██▊       | 6950/24420 [2:53:15<7:12:11,  1.48s/it]
 28%|██▊       | 6951/24420 [2:53:16<7:11:34,  1.48s/it]
 28%|██▊       | 6952/24420 [2:53:18<7:11:37,  1.48s/it]
 28%|██▊       | 6953/24420 [2:53:19<7:11:10,  1.48s/it]
 28%|██▊       | 6954/24420 [2:53:21<7:11:16,  1.48s/it]
 28%|██▊       | 6955/24420 [2:53:22<7:10:54,  1.48s/it]
                                                        
{'loss': 0.1357, 'learning_rate': 5e-05, 'epoch': 1.42}

 28%|██▊       | 6955/24420 [2:53:22<7:10:54,  1.48s/it]
 28%|██▊       | 6956/24420 [2:53:24<7:11:13,  1.48s/it]
 28%|██▊       | 6957/24420 [2:53:25<7:10:44,  1.48s/it]
 28%|██▊       | 6958/24420 [2:53:26<7:10:21,  1.48s/it]
 28%|██▊       | 6959/24420 [2:53:28<7:10:23,  1.48s/it]
 29%|██▊       | 6960/24420 [2:53:29<7:10:58,  1.48s/it]
                                                        
{'loss': 0.1223, 'learning_rate': 5e-05, 'epoch': 1.43}

 29%|██▊       | 6960/24420 [2:53:29<7:10:58,  1.48s/it]
 29%|██▊       | 6961/24420 [2:53:31<7:11:25,  1.48s/it]
 29%|██▊       | 6962/24420 [2:53:32<7:11:53,  1.48s/it]
 29%|██▊       | 6963/24420 [2:53:34<7:11:55,  1.48s/it]
 29%|██▊       | 6964/24420 [2:53:35<7:11:27,  1.48s/it]
 29%|██▊       | 6965/24420 [2:53:37<7:11:54,  1.48s/it]
                                                        
{'loss': 0.1198, 'learning_rate': 5e-05, 'epoch': 1.43}

 29%|██▊       | 6965/24420 [2:53:37<7:11:54,  1.48s/it]
 29%|██▊       | 6966/24420 [2:53:38<7:12:17,  1.49s/it]
 29%|██▊       | 6967/24420 [2:53:40<7:11:52,  1.48s/it]
 29%|██▊       | 6968/24420 [2:53:41<7:12:12,  1.49s/it]
 29%|██▊       | 6969/24420 [2:53:43<7:12:16,  1.49s/it]
 29%|██▊       | 6970/24420 [2:53:44<7:12:11,  1.49s/it]
                                                        
{'loss': 0.1132, 'learning_rate': 5e-05, 'epoch': 1.43}

 29%|██▊       | 6970/24420 [2:53:44<7:12:11,  1.49s/it]
 29%|██▊       | 6971/24420 [2:53:46<7:12:47,  1.49s/it]
 29%|██▊       | 6972/24420 [2:53:47<7:13:46,  1.49s/it]
 29%|██▊       | 6973/24420 [2:53:49<7:13:21,  1.49s/it]
 29%|██▊       | 6974/24420 [2:53:50<7:12:27,  1.49s/it]
 29%|██▊       | 6975/24420 [2:53:52<7:12:33,  1.49s/it]
                                                        
{'loss': 0.1368, 'learning_rate': 5e-05, 'epoch': 1.43}

 29%|██▊       | 6975/24420 [2:53:52<7:12:33,  1.49s/it]
 29%|██▊       | 6976/24420 [2:53:53<7:12:29,  1.49s/it]
 29%|██▊       | 6977/24420 [2:53:55<7:11:43,  1.49s/it]
 29%|██▊       | 6978/24420 [2:53:56<7:11:55,  1.49s/it]
 29%|██▊       | 6979/24420 [2:53:58<7:12:21,  1.49s/it]
 29%|██▊       | 6980/24420 [2:53:59<7:11:49,  1.49s/it]
                                                        
{'loss': 0.1271, 'learning_rate': 5e-05, 'epoch': 1.43}

 29%|██▊       | 6980/24420 [2:53:59<7:11:49,  1.49s/it]
 29%|██▊       | 6981/24420 [2:54:01<7:11:50,  1.49s/it]
 29%|██▊       | 6982/24420 [2:54:02<7:11:47,  1.49s/it]
 29%|██▊       | 6983/24420 [2:54:04<7:11:26,  1.48s/it]
 29%|██▊       | 6984/24420 [2:54:05<7:11:14,  1.48s/it]
 29%|██▊       | 6985/24420 [2:54:07<7:11:04,  1.48s/it]
                                                        
{'loss': 0.0999, 'learning_rate': 5e-05, 'epoch': 1.43}

 29%|██▊       | 6985/24420 [2:54:07<7:11:04,  1.48s/it]
 29%|██▊       | 6986/24420 [2:54:08<7:11:50,  1.49s/it]
 29%|██▊       | 6987/24420 [2:54:10<7:10:55,  1.48s/it]
 29%|██▊       | 6988/24420 [2:54:11<7:10:54,  1.48s/it]
 29%|██▊       | 6989/24420 [2:54:13<7:10:52,  1.48s/it]
 29%|██▊       | 6990/24420 [2:54:14<7:10:20,  1.48s/it]
                                                        
{'loss': 0.1383, 'learning_rate': 5e-05, 'epoch': 1.43}

 29%|██▊       | 6990/24420 [2:54:14<7:10:20,  1.48s/it]
 29%|██▊       | 6991/24420 [2:54:15<7:10:31,  1.48s/it]
 29%|██▊       | 6992/24420 [2:54:17<7:10:56,  1.48s/it]
 29%|██▊       | 6993/24420 [2:54:18<7:11:02,  1.48s/it]
 29%|██▊       | 6994/24420 [2:54:20<7:11:00,  1.48s/it]
 29%|██▊       | 6995/24420 [2:54:21<7:10:54,  1.48s/it]
                                                        
{'loss': 0.1245, 'learning_rate': 5e-05, 'epoch': 1.43}

 29%|██▊       | 6995/24420 [2:54:21<7:10:54,  1.48s/it]
 29%|██▊       | 6996/24420 [2:54:23<7:10:41,  1.48s/it]
 29%|██▊       | 6997/24420 [2:54:24<7:11:23,  1.49s/it]
 29%|██▊       | 6998/24420 [2:54:26<7:11:14,  1.49s/it]
 29%|██▊       | 6999/24420 [2:54:27<7:10:46,  1.48s/it]
 29%|██▊       | 7000/24420 [2:54:29<7:10:54,  1.48s/it]
                                                        
{'loss': 0.1105, 'learning_rate': 5e-05, 'epoch': 1.43}

 29%|██▊       | 7000/24420 [2:54:29<7:10:54,  1.48s/it]
 29%|██▊       | 7001/24420 [2:54:34<11:51:45,  2.45s/it]
 29%|██▊       | 7002/24420 [2:54:35<10:27:13,  2.16s/it]
 29%|██▊       | 7003/24420 [2:54:37<9:27:44,  1.96s/it] 
 29%|██▊       | 7004/24420 [2:54:38<8:46:56,  1.82s/it]
 29%|██▊       | 7005/24420 [2:54:39<8:17:32,  1.71s/it]
                                                        
{'loss': 0.1145, 'learning_rate': 5e-05, 'epoch': 1.43}

 29%|██▊       | 7005/24420 [2:54:39<8:17:32,  1.71s/it]
 29%|██▊       | 7006/24420 [2:54:41<7:57:29,  1.65s/it]
 29%|██▊       | 7007/24420 [2:54:42<7:42:40,  1.59s/it]
 29%|██▊       | 7008/24420 [2:54:44<7:32:46,  1.56s/it]
 29%|██▊       | 7009/24420 [2:54:45<7:27:19,  1.54s/it]
 29%|██▊       | 7010/24420 [2:54:47<7:22:51,  1.53s/it]
                                                        
{'loss': 0.1365, 'learning_rate': 5e-05, 'epoch': 1.44}

 29%|██▊       | 7010/24420 [2:54:47<7:22:51,  1.53s/it]
 29%|██▊       | 7011/24420 [2:54:48<7:19:42,  1.52s/it]
 29%|██▊       | 7012/24420 [2:54:50<7:16:33,  1.50s/it]
 29%|██▊       | 7013/24420 [2:54:51<7:14:25,  1.50s/it]
 29%|██▊       | 7014/24420 [2:54:53<7:12:14,  1.49s/it]
 29%|██▊       | 7015/24420 [2:54:54<7:10:55,  1.49s/it]
                                                        
{'loss': 0.1167, 'learning_rate': 5e-05, 'epoch': 1.44}

 29%|██▊       | 7015/24420 [2:54:54<7:10:55,  1.49s/it]
 29%|██▊       | 7016/24420 [2:54:56<7:10:15,  1.48s/it]
 29%|██▊       | 7017/24420 [2:54:57<7:09:56,  1.48s/it]
 29%|██▊       | 7018/24420 [2:54:59<7:09:20,  1.48s/it]
 29%|██▊       | 7019/24420 [2:55:00<7:09:14,  1.48s/it]
 29%|██▊       | 7020/24420 [2:55:02<7:11:24,  1.49s/it]
                                                        
{'loss': 0.1398, 'learning_rate': 5e-05, 'epoch': 1.44}

 29%|██▊       | 7020/24420 [2:55:02<7:11:24,  1.49s/it]
 29%|██▉       | 7021/24420 [2:55:03<7:11:29,  1.49s/it]
 29%|██▉       | 7022/24420 [2:55:05<7:10:16,  1.48s/it]
 29%|██▉       | 7023/24420 [2:55:06<7:09:19,  1.48s/it]
 29%|██▉       | 7024/24420 [2:55:08<7:08:48,  1.48s/it]
 29%|██▉       | 7025/24420 [2:55:09<7:10:58,  1.49s/it]
                                                        
{'loss': 0.1123, 'learning_rate': 5e-05, 'epoch': 1.44}

 29%|██▉       | 7025/24420 [2:55:09<7:10:58,  1.49s/it]
 29%|██▉       | 7026/24420 [2:55:11<7:10:18,  1.48s/it]
 29%|██▉       | 7027/24420 [2:55:12<7:10:15,  1.48s/it]
 29%|██▉       | 7028/24420 [2:55:14<7:09:48,  1.48s/it]
 29%|██▉       | 7029/24420 [2:55:15<7:08:53,  1.48s/it]
 29%|██▉       | 7030/24420 [2:55:17<7:08:31,  1.48s/it]
                                                        
{'loss': 0.1075, 'learning_rate': 5e-05, 'epoch': 1.44}

 29%|██▉       | 7030/24420 [2:55:17<7:08:31,  1.48s/it]
 29%|██▉       | 7031/24420 [2:55:18<7:09:07,  1.48s/it]
 29%|██▉       | 7032/24420 [2:55:19<7:09:06,  1.48s/it]
 29%|██▉       | 7033/24420 [2:55:21<7:09:28,  1.48s/it]
 29%|██▉       | 7034/24420 [2:55:22<7:09:42,  1.48s/it]
 29%|██▉       | 7035/24420 [2:55:24<7:08:52,  1.48s/it]
                                                        
{'loss': 0.12, 'learning_rate': 5e-05, 'epoch': 1.44}

 29%|██▉       | 7035/24420 [2:55:24<7:08:52,  1.48s/it]
 29%|██▉       | 7036/24420 [2:55:25<7:08:32,  1.48s/it]
 29%|██▉       | 7037/24420 [2:55:27<7:07:46,  1.48s/it]
 29%|██▉       | 7038/24420 [2:55:28<7:07:58,  1.48s/it]
 29%|██▉       | 7039/24420 [2:55:30<7:08:39,  1.48s/it]
 29%|██▉       | 7040/24420 [2:55:31<7:08:16,  1.48s/it]
                                                        
{'loss': 0.1228, 'learning_rate': 5e-05, 'epoch': 1.44}

 29%|██▉       | 7040/24420 [2:55:31<7:08:16,  1.48s/it]
 29%|██▉       | 7041/24420 [2:55:33<7:08:13,  1.48s/it]
 29%|██▉       | 7042/24420 [2:55:34<7:08:15,  1.48s/it]
 29%|██▉       | 7043/24420 [2:55:36<7:08:21,  1.48s/it]
 29%|██▉       | 7044/24420 [2:55:37<7:07:56,  1.48s/it]
 29%|██▉       | 7045/24420 [2:55:39<7:08:49,  1.48s/it]
                                                        
{'loss': 0.1272, 'learning_rate': 5e-05, 'epoch': 1.44}

 29%|██▉       | 7045/24420 [2:55:39<7:08:49,  1.48s/it]
 29%|██▉       | 7046/24420 [2:55:40<7:08:18,  1.48s/it]
 29%|██▉       | 7047/24420 [2:55:42<7:07:36,  1.48s/it]
 29%|██▉       | 7048/24420 [2:55:43<7:07:51,  1.48s/it]
 29%|██▉       | 7049/24420 [2:55:45<7:08:39,  1.48s/it]
 29%|██▉       | 7050/24420 [2:55:46<7:08:49,  1.48s/it]
                                                        
{'loss': 0.1165, 'learning_rate': 5e-05, 'epoch': 1.44}

 29%|██▉       | 7050/24420 [2:55:46<7:08:49,  1.48s/it]
 29%|██▉       | 7051/24420 [2:55:48<7:09:30,  1.48s/it]
 29%|██▉       | 7052/24420 [2:55:49<7:09:39,  1.48s/it]
 29%|██▉       | 7053/24420 [2:55:51<7:10:29,  1.49s/it]
 29%|██▉       | 7054/24420 [2:55:52<7:09:43,  1.48s/it]
 29%|██▉       | 7055/24420 [2:55:54<7:08:52,  1.48s/it]
                                                        
{'loss': 0.1347, 'learning_rate': 5e-05, 'epoch': 1.44}

 29%|██▉       | 7055/24420 [2:55:54<7:08:52,  1.48s/it]
 29%|██▉       | 7056/24420 [2:55:55<7:08:27,  1.48s/it]
 29%|██▉       | 7057/24420 [2:55:56<7:07:29,  1.48s/it]
 29%|██▉       | 7058/24420 [2:55:58<7:08:00,  1.48s/it]
 29%|██▉       | 7059/24420 [2:55:59<7:08:33,  1.48s/it]
 29%|██▉       | 7060/24420 [2:56:01<7:09:07,  1.48s/it]
                                                        
{'loss': 0.1055, 'learning_rate': 5e-05, 'epoch': 1.45}

 29%|██▉       | 7060/24420 [2:56:01<7:09:07,  1.48s/it]
 29%|██▉       | 7061/24420 [2:56:02<7:08:54,  1.48s/it]
 29%|██▉       | 7062/24420 [2:56:04<7:08:57,  1.48s/it]
 29%|██▉       | 7063/24420 [2:56:05<7:09:01,  1.48s/it]
 29%|██▉       | 7064/24420 [2:56:07<7:09:06,  1.48s/it]
 29%|██▉       | 7065/24420 [2:56:08<7:09:48,  1.49s/it]
                                                        
{'loss': 0.1044, 'learning_rate': 5e-05, 'epoch': 1.45}

 29%|██▉       | 7065/24420 [2:56:08<7:09:48,  1.49s/it]
 29%|██▉       | 7066/24420 [2:56:10<7:10:11,  1.49s/it]
 29%|██▉       | 7067/24420 [2:56:11<7:09:15,  1.48s/it]
 29%|██▉       | 7068/24420 [2:56:13<7:08:26,  1.48s/it]
 29%|██▉       | 7069/24420 [2:56:14<7:08:18,  1.48s/it]
 29%|██▉       | 7070/24420 [2:56:16<7:08:08,  1.48s/it]
                                                        
{'loss': 0.1264, 'learning_rate': 5e-05, 'epoch': 1.45}

 29%|██▉       | 7070/24420 [2:56:16<7:08:08,  1.48s/it]
 29%|██▉       | 7071/24420 [2:56:17<7:08:17,  1.48s/it]
 29%|██▉       | 7072/24420 [2:56:19<7:07:09,  1.48s/it]
 29%|██▉       | 7073/24420 [2:56:20<7:07:07,  1.48s/it]
 29%|██▉       | 7074/24420 [2:56:22<7:06:49,  1.48s/it]
 29%|██▉       | 7075/24420 [2:56:23<7:06:40,  1.48s/it]
                                                        
{'loss': 0.1402, 'learning_rate': 5e-05, 'epoch': 1.45}

 29%|██▉       | 7075/24420 [2:56:23<7:06:40,  1.48s/it]
 29%|██▉       | 7076/24420 [2:56:25<7:06:37,  1.48s/it]
 29%|██▉       | 7077/24420 [2:56:26<7:07:06,  1.48s/it]
 29%|██▉       | 7078/24420 [2:56:28<7:07:10,  1.48s/it]
 29%|██▉       | 7079/24420 [2:56:29<7:07:31,  1.48s/it]
 29%|██▉       | 7080/24420 [2:56:31<7:07:12,  1.48s/it]
                                                        
{'loss': 0.1091, 'learning_rate': 5e-05, 'epoch': 1.45}

 29%|██▉       | 7080/24420 [2:56:31<7:07:12,  1.48s/it]
 29%|██▉       | 7081/24420 [2:56:32<7:29:43,  1.56s/it]
 29%|██▉       | 7082/24420 [2:56:34<7:22:42,  1.53s/it]
 29%|██▉       | 7083/24420 [2:56:35<7:17:37,  1.51s/it]
 29%|██▉       | 7084/24420 [2:56:37<7:14:11,  1.50s/it]
 29%|██▉       | 7085/24420 [2:56:38<7:12:28,  1.50s/it]
                                                        
{'loss': 0.1183, 'learning_rate': 5e-05, 'epoch': 1.45}

 29%|██▉       | 7085/24420 [2:56:38<7:12:28,  1.50s/it]
 29%|██▉       | 7086/24420 [2:56:40<7:11:48,  1.49s/it]
 29%|██▉       | 7087/24420 [2:56:41<7:10:15,  1.49s/it]
 29%|██▉       | 7088/24420 [2:56:43<7:08:48,  1.48s/it]
 29%|██▉       | 7089/24420 [2:56:44<7:09:03,  1.49s/it]
 29%|██▉       | 7090/24420 [2:56:46<7:09:05,  1.49s/it]
                                                        
{'loss': 0.1089, 'learning_rate': 5e-05, 'epoch': 1.45}

 29%|██▉       | 7090/24420 [2:56:46<7:09:05,  1.49s/it]
 29%|██▉       | 7091/24420 [2:56:47<7:08:45,  1.48s/it]
 29%|██▉       | 7092/24420 [2:56:49<7:08:22,  1.48s/it]
 29%|██▉       | 7093/24420 [2:56:50<7:07:50,  1.48s/it]
 29%|██▉       | 7094/24420 [2:56:52<7:07:34,  1.48s/it]
 29%|██▉       | 7095/24420 [2:56:53<7:07:21,  1.48s/it]
                                                        
{'loss': 0.1047, 'learning_rate': 5e-05, 'epoch': 1.45}

 29%|██▉       | 7095/24420 [2:56:53<7:07:21,  1.48s/it]
 29%|██▉       | 7096/24420 [2:56:54<7:08:04,  1.48s/it]
 29%|██▉       | 7097/24420 [2:56:56<7:07:59,  1.48s/it]
 29%|██▉       | 7098/24420 [2:56:57<7:08:26,  1.48s/it]
 29%|██▉       | 7099/24420 [2:56:59<7:07:53,  1.48s/it]
 29%|██▉       | 7100/24420 [2:57:00<7:08:08,  1.48s/it]
                                                        
{'loss': 0.102, 'learning_rate': 5e-05, 'epoch': 1.45}

 29%|██▉       | 7100/24420 [2:57:00<7:08:08,  1.48s/it]
 29%|██▉       | 7101/24420 [2:57:02<7:07:50,  1.48s/it]
 29%|██▉       | 7102/24420 [2:57:03<7:07:45,  1.48s/it]
 29%|██▉       | 7103/24420 [2:57:05<7:07:39,  1.48s/it]
 29%|██▉       | 7104/24420 [2:57:06<7:07:58,  1.48s/it]
 29%|██▉       | 7105/24420 [2:57:08<7:07:41,  1.48s/it]
                                                        
{'loss': 0.1288, 'learning_rate': 5e-05, 'epoch': 1.45}

 29%|██▉       | 7105/24420 [2:57:08<7:07:41,  1.48s/it]
 29%|██▉       | 7106/24420 [2:57:09<7:08:03,  1.48s/it]
 29%|██▉       | 7107/24420 [2:57:11<7:07:43,  1.48s/it]
 29%|██▉       | 7108/24420 [2:57:12<7:08:23,  1.48s/it]
 29%|██▉       | 7109/24420 [2:57:14<7:08:07,  1.48s/it]
 29%|██▉       | 7110/24420 [2:57:15<7:07:46,  1.48s/it]
                                                        
{'loss': 0.0905, 'learning_rate': 5e-05, 'epoch': 1.46}

 29%|██▉       | 7110/24420 [2:57:15<7:07:46,  1.48s/it]
 29%|██▉       | 7111/24420 [2:57:17<7:07:41,  1.48s/it]
 29%|██▉       | 7112/24420 [2:57:18<7:07:25,  1.48s/it]
 29%|██▉       | 7113/24420 [2:57:20<7:06:37,  1.48s/it]
 29%|██▉       | 7114/24420 [2:57:21<7:06:26,  1.48s/it]
 29%|██▉       | 7115/24420 [2:57:23<7:06:17,  1.48s/it]
                                                        
{'loss': 0.1085, 'learning_rate': 5e-05, 'epoch': 1.46}

 29%|██▉       | 7115/24420 [2:57:23<7:06:17,  1.48s/it]
 29%|██▉       | 7116/24420 [2:57:24<7:06:12,  1.48s/it]
 29%|██▉       | 7117/24420 [2:57:26<7:06:21,  1.48s/it]
 29%|██▉       | 7118/24420 [2:57:27<7:06:27,  1.48s/it]
 29%|██▉       | 7119/24420 [2:57:29<7:06:54,  1.48s/it]
 29%|██▉       | 7120/24420 [2:57:30<7:06:53,  1.48s/it]
                                                        
{'loss': 0.1116, 'learning_rate': 5e-05, 'epoch': 1.46}

 29%|██▉       | 7120/24420 [2:57:30<7:06:53,  1.48s/it]
 29%|██▉       | 7121/24420 [2:57:32<7:07:01,  1.48s/it]
 29%|██▉       | 7122/24420 [2:57:33<7:07:05,  1.48s/it]
 29%|██▉       | 7123/24420 [2:57:34<7:06:57,  1.48s/it]
 29%|██▉       | 7124/24420 [2:57:36<7:07:36,  1.48s/it]
 29%|██▉       | 7125/24420 [2:57:37<7:07:29,  1.48s/it]
                                                        
{'loss': 0.1199, 'learning_rate': 5e-05, 'epoch': 1.46}

 29%|██▉       | 7125/24420 [2:57:37<7:07:29,  1.48s/it]
 29%|██▉       | 7126/24420 [2:57:39<7:07:05,  1.48s/it]
 29%|██▉       | 7127/24420 [2:57:40<7:07:10,  1.48s/it]
 29%|██▉       | 7128/24420 [2:57:42<7:08:49,  1.49s/it]
 29%|██▉       | 7129/24420 [2:57:43<7:08:38,  1.49s/it]
 29%|██▉       | 7130/24420 [2:57:45<7:07:16,  1.48s/it]
                                                        
{'loss': 0.1258, 'learning_rate': 5e-05, 'epoch': 1.46}

 29%|██▉       | 7130/24420 [2:57:45<7:07:16,  1.48s/it]
 29%|██▉       | 7131/24420 [2:57:46<7:08:18,  1.49s/it]
 29%|██▉       | 7132/24420 [2:57:48<7:07:50,  1.48s/it]
 29%|██▉       | 7133/24420 [2:57:49<7:07:21,  1.48s/it]
 29%|██▉       | 7134/24420 [2:57:51<7:06:32,  1.48s/it]
 29%|██▉       | 7135/24420 [2:57:52<7:06:05,  1.48s/it]
                                                        
{'loss': 0.1407, 'learning_rate': 5e-05, 'epoch': 1.46}

 29%|██▉       | 7135/24420 [2:57:52<7:06:05,  1.48s/it]
 29%|██▉       | 7136/24420 [2:57:54<7:07:07,  1.48s/it]
 29%|██▉       | 7137/24420 [2:57:55<7:06:44,  1.48s/it]
 29%|██▉       | 7138/24420 [2:57:57<7:06:54,  1.48s/it]
 29%|██▉       | 7139/24420 [2:57:58<7:06:57,  1.48s/it]
 29%|██▉       | 7140/24420 [2:58:00<7:06:37,  1.48s/it]
                                                        
{'loss': 0.1348, 'learning_rate': 5e-05, 'epoch': 1.46}

 29%|██▉       | 7140/24420 [2:58:00<7:06:37,  1.48s/it]
 29%|██▉       | 7141/24420 [2:58:01<7:06:49,  1.48s/it]
 29%|██▉       | 7142/24420 [2:58:03<7:06:26,  1.48s/it]
 29%|██▉       | 7143/24420 [2:58:04<7:07:00,  1.48s/it]
 29%|██▉       | 7144/24420 [2:58:06<7:07:03,  1.48s/it]
 29%|██▉       | 7145/24420 [2:58:07<7:07:27,  1.48s/it]
                                                        
{'loss': 0.1123, 'learning_rate': 5e-05, 'epoch': 1.46}

 29%|██▉       | 7145/24420 [2:58:07<7:07:27,  1.48s/it]
 29%|██▉       | 7146/24420 [2:58:09<7:07:49,  1.49s/it]
 29%|██▉       | 7147/24420 [2:58:10<7:06:57,  1.48s/it]
 29%|██▉       | 7148/24420 [2:58:12<7:07:39,  1.49s/it]
 29%|██▉       | 7149/24420 [2:58:13<7:07:21,  1.48s/it]
 29%|██▉       | 7150/24420 [2:58:15<7:07:05,  1.48s/it]
                                                        
{'loss': 0.1114, 'learning_rate': 5e-05, 'epoch': 1.46}

 29%|██▉       | 7150/24420 [2:58:15<7:07:05,  1.48s/it]
 29%|██▉       | 7151/24420 [2:58:16<7:06:39,  1.48s/it]
 29%|██▉       | 7152/24420 [2:58:18<7:05:59,  1.48s/it]
 29%|██▉       | 7153/24420 [2:58:19<7:05:15,  1.48s/it]
 29%|██▉       | 7154/24420 [2:58:20<7:05:39,  1.48s/it]
 29%|██▉       | 7155/24420 [2:58:22<7:06:15,  1.48s/it]
                                                        
{'loss': 0.1168, 'learning_rate': 5e-05, 'epoch': 1.46}

 29%|██▉       | 7155/24420 [2:58:22<7:06:15,  1.48s/it]
 29%|██▉       | 7156/24420 [2:58:23<7:06:26,  1.48s/it]
 29%|██▉       | 7157/24420 [2:58:25<7:07:16,  1.49s/it]
 29%|██▉       | 7158/24420 [2:58:26<7:06:38,  1.48s/it]
 29%|██▉       | 7159/24420 [2:58:28<7:05:50,  1.48s/it]
 29%|██▉       | 7160/24420 [2:58:29<7:05:28,  1.48s/it]
                                                        
{'loss': 0.1129, 'learning_rate': 5e-05, 'epoch': 1.47}

 29%|██▉       | 7160/24420 [2:58:29<7:05:28,  1.48s/it]
 29%|██▉       | 7161/24420 [2:58:31<7:05:45,  1.48s/it]
 29%|██▉       | 7162/24420 [2:58:32<7:06:10,  1.48s/it]
 29%|██▉       | 7163/24420 [2:58:34<7:05:41,  1.48s/it]
 29%|██▉       | 7164/24420 [2:58:35<7:05:46,  1.48s/it]
 29%|██▉       | 7165/24420 [2:58:37<7:06:18,  1.48s/it]
                                                        
{'loss': 0.1043, 'learning_rate': 5e-05, 'epoch': 1.47}

 29%|██▉       | 7165/24420 [2:58:37<7:06:18,  1.48s/it]
 29%|██▉       | 7166/24420 [2:58:38<7:07:19,  1.49s/it]
 29%|██▉       | 7167/24420 [2:58:40<7:07:02,  1.49s/it]
 29%|██▉       | 7168/24420 [2:58:41<7:06:40,  1.48s/it]
 29%|██▉       | 7169/24420 [2:58:43<7:06:12,  1.48s/it]
 29%|██▉       | 7170/24420 [2:58:44<7:06:44,  1.48s/it]
                                                        
{'loss': 0.1236, 'learning_rate': 5e-05, 'epoch': 1.47}

 29%|██▉       | 7170/24420 [2:58:44<7:06:44,  1.48s/it]
 29%|██▉       | 7171/24420 [2:58:46<7:06:20,  1.48s/it]
 29%|██▉       | 7172/24420 [2:58:47<7:05:50,  1.48s/it]
 29%|██▉       | 7173/24420 [2:58:49<7:06:54,  1.49s/it]
 29%|██▉       | 7174/24420 [2:58:50<7:06:28,  1.48s/it]
 29%|██▉       | 7175/24420 [2:58:52<7:05:51,  1.48s/it]
                                                        
{'loss': 0.0949, 'learning_rate': 5e-05, 'epoch': 1.47}

 29%|██▉       | 7175/24420 [2:58:52<7:05:51,  1.48s/it]
 29%|██▉       | 7176/24420 [2:58:53<7:05:46,  1.48s/it]
 29%|██▉       | 7177/24420 [2:58:55<7:05:51,  1.48s/it]
 29%|██▉       | 7178/24420 [2:58:56<7:05:51,  1.48s/it]
 29%|██▉       | 7179/24420 [2:58:58<7:04:51,  1.48s/it]
 29%|██▉       | 7180/24420 [2:58:59<7:05:30,  1.48s/it]
                                                        
{'loss': 0.1236, 'learning_rate': 5e-05, 'epoch': 1.47}

 29%|██▉       | 7180/24420 [2:58:59<7:05:30,  1.48s/it]
 29%|██▉       | 7181/24420 [2:59:00<7:05:33,  1.48s/it]
 29%|██▉       | 7182/24420 [2:59:02<7:05:56,  1.48s/it]
 29%|██▉       | 7183/24420 [2:59:03<7:06:06,  1.48s/it]
 29%|██▉       | 7184/24420 [2:59:05<7:06:43,  1.49s/it]
 29%|██▉       | 7185/24420 [2:59:06<7:07:08,  1.49s/it]
                                                        
{'loss': 0.1135, 'learning_rate': 5e-05, 'epoch': 1.47}

 29%|██▉       | 7185/24420 [2:59:06<7:07:08,  1.49s/it]
 29%|██▉       | 7186/24420 [2:59:08<7:08:21,  1.49s/it]
 29%|██▉       | 7187/24420 [2:59:09<7:07:09,  1.49s/it]
 29%|██▉       | 7188/24420 [2:59:11<7:06:17,  1.48s/it]
 29%|██▉       | 7189/24420 [2:59:12<7:05:43,  1.48s/it]
 29%|██▉       | 7190/24420 [2:59:14<7:06:49,  1.49s/it]
                                                        
{'loss': 0.1183, 'learning_rate': 5e-05, 'epoch': 1.47}

 29%|██▉       | 7190/24420 [2:59:14<7:06:49,  1.49s/it]
 29%|██▉       | 7191/24420 [2:59:15<7:06:41,  1.49s/it]
 29%|██▉       | 7192/24420 [2:59:17<7:07:08,  1.49s/it]
 29%|██▉       | 7193/24420 [2:59:18<7:07:07,  1.49s/it]
 29%|██▉       | 7194/24420 [2:59:20<7:07:13,  1.49s/it]
 29%|██▉       | 7195/24420 [2:59:21<7:06:51,  1.49s/it]
                                                        
{'loss': 0.1158, 'learning_rate': 5e-05, 'epoch': 1.47}

 29%|██▉       | 7195/24420 [2:59:21<7:06:51,  1.49s/it]
 29%|██▉       | 7196/24420 [2:59:23<7:06:07,  1.48s/it]
 29%|██▉       | 7197/24420 [2:59:24<7:06:28,  1.49s/it]
 29%|██▉       | 7198/24420 [2:59:26<7:05:48,  1.48s/it]
 29%|██▉       | 7199/24420 [2:59:27<7:06:15,  1.49s/it]
 29%|██▉       | 7200/24420 [2:59:29<7:06:22,  1.49s/it]
                                                        
{'loss': 0.1254, 'learning_rate': 5e-05, 'epoch': 1.47}

 29%|██▉       | 7200/24420 [2:59:29<7:06:22,  1.49s/it]
 29%|██▉       | 7201/24420 [2:59:33<11:30:40,  2.41s/it]
 29%|██▉       | 7202/24420 [2:59:35<10:10:44,  2.13s/it]
 29%|██▉       | 7203/24420 [2:59:36<9:14:06,  1.93s/it] 
 30%|██▉       | 7204/24420 [2:59:38<8:34:49,  1.79s/it]
 30%|██▉       | 7205/24420 [2:59:39<8:08:57,  1.70s/it]
                                                        
{'loss': 0.1236, 'learning_rate': 5e-05, 'epoch': 1.48}

 30%|██▉       | 7205/24420 [2:59:39<8:08:57,  1.70s/it]
 30%|██▉       | 7206/24420 [2:59:41<7:49:18,  1.64s/it]
 30%|██▉       | 7207/24420 [2:59:42<7:35:31,  1.59s/it]
 30%|██▉       | 7208/24420 [2:59:44<7:26:19,  1.56s/it]
 30%|██▉       | 7209/24420 [2:59:45<7:18:58,  1.53s/it]
 30%|██▉       | 7210/24420 [2:59:47<7:14:42,  1.52s/it]
                                                        
{'loss': 0.1132, 'learning_rate': 5e-05, 'epoch': 1.48}

 30%|██▉       | 7210/24420 [2:59:47<7:14:42,  1.52s/it]
 30%|██▉       | 7211/24420 [2:59:48<7:12:32,  1.51s/it]
 30%|██▉       | 7212/24420 [2:59:50<7:10:56,  1.50s/it]
 30%|██▉       | 7213/24420 [2:59:51<7:08:20,  1.49s/it]
 30%|██▉       | 7214/24420 [2:59:53<7:07:31,  1.49s/it]
 30%|██▉       | 7215/24420 [2:59:54<7:06:42,  1.49s/it]
                                                        
{'loss': 0.1293, 'learning_rate': 5e-05, 'epoch': 1.48}

 30%|██▉       | 7215/24420 [2:59:54<7:06:42,  1.49s/it]
 30%|██▉       | 7216/24420 [2:59:55<7:06:21,  1.49s/it]
 30%|██▉       | 7217/24420 [2:59:57<7:05:28,  1.48s/it]
 30%|██▉       | 7218/24420 [2:59:58<7:05:14,  1.48s/it]
 30%|██▉       | 7219/24420 [3:00:00<7:05:09,  1.48s/it]
 30%|██▉       | 7220/24420 [3:00:01<7:04:53,  1.48s/it]
                                                        
{'loss': 0.094, 'learning_rate': 5e-05, 'epoch': 1.48}

 30%|██▉       | 7220/24420 [3:00:01<7:04:53,  1.48s/it]
 30%|██▉       | 7221/24420 [3:00:03<7:04:57,  1.48s/it]
 30%|██▉       | 7222/24420 [3:00:04<7:04:46,  1.48s/it]
 30%|██▉       | 7223/24420 [3:00:06<7:05:00,  1.48s/it]
 30%|██▉       | 7224/24420 [3:00:07<7:04:41,  1.48s/it]
 30%|██▉       | 7225/24420 [3:00:09<7:04:35,  1.48s/it]
                                                        
{'loss': 0.1135, 'learning_rate': 5e-05, 'epoch': 1.48}

 30%|██▉       | 7225/24420 [3:00:09<7:04:35,  1.48s/it]
 30%|██▉       | 7226/24420 [3:00:10<7:04:25,  1.48s/it]
 30%|██▉       | 7227/24420 [3:00:12<7:10:21,  1.50s/it]
 30%|██▉       | 7228/24420 [3:00:13<7:08:28,  1.50s/it]
 30%|██▉       | 7229/24420 [3:00:15<7:07:20,  1.49s/it]
 30%|██▉       | 7230/24420 [3:00:16<7:06:19,  1.49s/it]
                                                        
{'loss': 0.1279, 'learning_rate': 5e-05, 'epoch': 1.48}

 30%|██▉       | 7230/24420 [3:00:16<7:06:19,  1.49s/it]
 30%|██▉       | 7231/24420 [3:00:18<7:05:45,  1.49s/it]
 30%|██▉       | 7232/24420 [3:00:19<7:06:24,  1.49s/it]
 30%|██▉       | 7233/24420 [3:00:21<7:05:27,  1.49s/it]
 30%|██▉       | 7234/24420 [3:00:22<7:05:32,  1.49s/it]
 30%|██▉       | 7235/24420 [3:00:24<7:05:00,  1.48s/it]
                                                        
{'loss': 0.1155, 'learning_rate': 5e-05, 'epoch': 1.48}

 30%|██▉       | 7235/24420 [3:00:24<7:05:00,  1.48s/it]
 30%|██▉       | 7236/24420 [3:00:25<7:05:30,  1.49s/it]
 30%|██▉       | 7237/24420 [3:00:27<7:05:25,  1.49s/it]
 30%|██▉       | 7238/24420 [3:00:28<7:04:44,  1.48s/it]
 30%|██▉       | 7239/24420 [3:00:30<7:04:04,  1.48s/it]
 30%|██▉       | 7240/24420 [3:00:31<7:03:11,  1.48s/it]
                                                        
{'loss': 0.1338, 'learning_rate': 5e-05, 'epoch': 1.48}

 30%|██▉       | 7240/24420 [3:00:31<7:03:11,  1.48s/it]
 30%|██▉       | 7241/24420 [3:00:33<7:03:04,  1.48s/it]
 30%|██▉       | 7242/24420 [3:00:34<7:03:17,  1.48s/it]
 30%|██▉       | 7243/24420 [3:00:36<7:03:12,  1.48s/it]
 30%|██▉       | 7244/24420 [3:00:37<7:03:38,  1.48s/it]
 30%|██▉       | 7245/24420 [3:00:39<7:03:29,  1.48s/it]
                                                        
{'loss': 0.1116, 'learning_rate': 5e-05, 'epoch': 1.48}

 30%|██▉       | 7245/24420 [3:00:39<7:03:29,  1.48s/it]
 30%|██▉       | 7246/24420 [3:00:40<7:03:21,  1.48s/it]
 30%|██▉       | 7247/24420 [3:00:41<7:03:49,  1.48s/it]
 30%|██▉       | 7248/24420 [3:00:43<7:03:23,  1.48s/it]
 30%|██▉       | 7249/24420 [3:00:44<7:03:13,  1.48s/it]
 30%|██▉       | 7250/24420 [3:00:46<7:04:00,  1.48s/it]
                                                        
{'loss': 0.1182, 'learning_rate': 5e-05, 'epoch': 1.48}

 30%|██▉       | 7250/24420 [3:00:46<7:04:00,  1.48s/it]
 30%|██▉       | 7251/24420 [3:00:47<7:04:05,  1.48s/it]
 30%|██▉       | 7252/24420 [3:00:49<7:03:31,  1.48s/it]
 30%|██▉       | 7253/24420 [3:00:50<7:03:01,  1.48s/it]
 30%|██▉       | 7254/24420 [3:00:52<7:02:32,  1.48s/it]
 30%|██▉       | 7255/24420 [3:00:53<7:03:26,  1.48s/it]
                                                        
{'loss': 0.1079, 'learning_rate': 5e-05, 'epoch': 1.49}

 30%|██▉       | 7255/24420 [3:00:53<7:03:26,  1.48s/it]
 30%|██▉       | 7256/24420 [3:00:55<7:03:01,  1.48s/it]
 30%|██▉       | 7257/24420 [3:00:56<7:03:20,  1.48s/it]
 30%|██▉       | 7258/24420 [3:00:58<7:03:40,  1.48s/it]
 30%|██▉       | 7259/24420 [3:00:59<7:04:02,  1.48s/it]
 30%|██▉       | 7260/24420 [3:01:01<7:03:19,  1.48s/it]
                                                        
{'loss': 0.1313, 'learning_rate': 5e-05, 'epoch': 1.49}

 30%|██▉       | 7260/24420 [3:01:01<7:03:19,  1.48s/it]
 30%|██▉       | 7261/24420 [3:01:02<7:03:30,  1.48s/it]
 30%|██▉       | 7262/24420 [3:01:04<7:03:40,  1.48s/it]
 30%|██▉       | 7263/24420 [3:01:05<7:03:20,  1.48s/it]
 30%|██▉       | 7264/24420 [3:01:07<7:03:28,  1.48s/it]
 30%|██▉       | 7265/24420 [3:01:08<7:03:52,  1.48s/it]
                                                        
{'loss': 0.1266, 'learning_rate': 5e-05, 'epoch': 1.49}

 30%|██▉       | 7265/24420 [3:01:08<7:03:52,  1.48s/it]
 30%|██▉       | 7266/24420 [3:01:10<7:04:57,  1.49s/it]
 30%|██▉       | 7267/24420 [3:01:11<7:04:08,  1.48s/it]
 30%|██▉       | 7268/24420 [3:01:13<7:04:28,  1.48s/it]
 30%|██▉       | 7269/24420 [3:01:14<7:03:19,  1.48s/it]
 30%|██▉       | 7270/24420 [3:01:16<7:03:01,  1.48s/it]
                                                        
{'loss': 0.1108, 'learning_rate': 5e-05, 'epoch': 1.49}

 30%|██▉       | 7270/24420 [3:01:16<7:03:01,  1.48s/it]
 30%|██▉       | 7271/24420 [3:01:17<7:03:14,  1.48s/it]
 30%|██▉       | 7272/24420 [3:01:18<7:02:44,  1.48s/it]
 30%|██▉       | 7273/24420 [3:01:20<7:02:59,  1.48s/it]
 30%|██▉       | 7274/24420 [3:01:21<7:02:26,  1.48s/it]
 30%|██▉       | 7275/24420 [3:01:23<7:03:10,  1.48s/it]
                                                        
{'loss': 0.1106, 'learning_rate': 5e-05, 'epoch': 1.49}

 30%|██▉       | 7275/24420 [3:01:23<7:03:10,  1.48s/it]
 30%|██▉       | 7276/24420 [3:01:24<7:03:32,  1.48s/it]
 30%|██▉       | 7277/24420 [3:01:26<7:02:40,  1.48s/it]
 30%|██▉       | 7278/24420 [3:01:27<7:02:25,  1.48s/it]
 30%|██▉       | 7279/24420 [3:01:29<7:02:26,  1.48s/it]
 30%|██▉       | 7280/24420 [3:01:30<7:02:33,  1.48s/it]
                                                        
{'loss': 0.115, 'learning_rate': 5e-05, 'epoch': 1.49}

 30%|██▉       | 7280/24420 [3:01:30<7:02:33,  1.48s/it]
 30%|██▉       | 7281/24420 [3:01:32<7:03:17,  1.48s/it]
 30%|██▉       | 7282/24420 [3:01:33<7:02:32,  1.48s/it]
 30%|██▉       | 7283/24420 [3:01:35<7:02:17,  1.48s/it]
 30%|██▉       | 7284/24420 [3:01:36<7:03:16,  1.48s/it]
 30%|██▉       | 7285/24420 [3:01:38<7:03:23,  1.48s/it]
                                                        
{'loss': 0.1182, 'learning_rate': 5e-05, 'epoch': 1.49}

 30%|██▉       | 7285/24420 [3:01:38<7:03:23,  1.48s/it]
 30%|██▉       | 7286/24420 [3:01:39<7:03:20,  1.48s/it]
 30%|██▉       | 7287/24420 [3:01:41<7:03:23,  1.48s/it]
 30%|██▉       | 7288/24420 [3:01:42<7:03:24,  1.48s/it]
 30%|██▉       | 7289/24420 [3:01:44<7:03:06,  1.48s/it]
 30%|██▉       | 7290/24420 [3:01:45<7:03:00,  1.48s/it]
                                                        
{'loss': 0.135, 'learning_rate': 5e-05, 'epoch': 1.49}

 30%|██▉       | 7290/24420 [3:01:45<7:03:00,  1.48s/it]
 30%|██▉       | 7291/24420 [3:01:47<7:02:51,  1.48s/it]
 30%|██▉       | 7292/24420 [3:01:48<7:04:11,  1.49s/it]
 30%|██▉       | 7293/24420 [3:01:50<7:03:12,  1.48s/it]
 30%|██▉       | 7294/24420 [3:01:51<7:03:37,  1.48s/it]
 30%|██▉       | 7295/24420 [3:01:53<7:04:17,  1.49s/it]
                                                        
{'loss': 0.1073, 'learning_rate': 5e-05, 'epoch': 1.49}

 30%|██▉       | 7295/24420 [3:01:53<7:04:17,  1.49s/it]
 30%|██▉       | 7296/24420 [3:01:54<7:04:08,  1.49s/it]
 30%|██▉       | 7297/24420 [3:01:56<7:03:29,  1.48s/it]
 30%|██▉       | 7298/24420 [3:01:57<7:03:57,  1.49s/it]
 30%|██▉       | 7299/24420 [3:01:59<7:03:55,  1.49s/it]
 30%|██▉       | 7300/24420 [3:02:00<7:03:32,  1.48s/it]
                                                        
{'loss': 0.1441, 'learning_rate': 5e-05, 'epoch': 1.49}

 30%|██▉       | 7300/24420 [3:02:00<7:03:32,  1.48s/it]
 30%|██▉       | 7301/24420 [3:02:01<7:04:05,  1.49s/it]
 30%|██▉       | 7302/24420 [3:02:03<7:03:21,  1.48s/it]
 30%|██▉       | 7303/24420 [3:02:04<7:03:12,  1.48s/it]
 30%|██▉       | 7304/24420 [3:02:06<7:02:50,  1.48s/it]
 30%|██▉       | 7305/24420 [3:02:07<7:02:15,  1.48s/it]
                                                        
{'loss': 0.1225, 'learning_rate': 5e-05, 'epoch': 1.5}

 30%|██▉       | 7305/24420 [3:02:07<7:02:15,  1.48s/it]
 30%|██▉       | 7306/24420 [3:02:09<7:03:19,  1.48s/it]
 30%|██▉       | 7307/24420 [3:02:10<7:03:55,  1.49s/it]
 30%|██▉       | 7308/24420 [3:02:12<7:03:27,  1.48s/it]
 30%|██▉       | 7309/24420 [3:02:13<7:02:43,  1.48s/it]
 30%|██▉       | 7310/24420 [3:02:15<7:03:00,  1.48s/it]
                                                        
{'loss': 0.1232, 'learning_rate': 5e-05, 'epoch': 1.5}

 30%|██▉       | 7310/24420 [3:02:15<7:03:00,  1.48s/it]
 30%|██▉       | 7311/24420 [3:02:16<7:04:01,  1.49s/it]
 30%|██▉       | 7312/24420 [3:02:18<7:02:46,  1.48s/it]
 30%|██▉       | 7313/24420 [3:02:19<7:02:28,  1.48s/it]
 30%|██▉       | 7314/24420 [3:02:21<7:02:47,  1.48s/it]
 30%|██▉       | 7315/24420 [3:02:22<7:02:23,  1.48s/it]
                                                        
{'loss': 0.1052, 'learning_rate': 5e-05, 'epoch': 1.5}

 30%|██▉       | 7315/24420 [3:02:22<7:02:23,  1.48s/it]
 30%|██▉       | 7316/24420 [3:02:24<7:02:39,  1.48s/it]
 30%|██▉       | 7317/24420 [3:02:25<7:02:19,  1.48s/it]
 30%|██▉       | 7318/24420 [3:02:27<7:02:24,  1.48s/it]
 30%|██▉       | 7319/24420 [3:02:28<7:02:44,  1.48s/it]
 30%|██▉       | 7320/24420 [3:02:30<7:03:38,  1.49s/it]
                                                        
{'loss': 0.1019, 'learning_rate': 5e-05, 'epoch': 1.5}

 30%|██▉       | 7320/24420 [3:02:30<7:03:38,  1.49s/it]
 30%|██▉       | 7321/24420 [3:02:31<7:03:52,  1.49s/it]
 30%|██▉       | 7322/24420 [3:02:33<7:03:33,  1.49s/it]
 30%|██▉       | 7323/24420 [3:02:34<7:02:21,  1.48s/it]
 30%|██▉       | 7324/24420 [3:02:36<7:01:49,  1.48s/it]
 30%|██▉       | 7325/24420 [3:02:37<7:01:12,  1.48s/it]
                                                        
{'loss': 0.107, 'learning_rate': 5e-05, 'epoch': 1.5}

 30%|██▉       | 7325/24420 [3:02:37<7:01:12,  1.48s/it]
 30%|███       | 7326/24420 [3:02:39<7:01:33,  1.48s/it]
 30%|███       | 7327/24420 [3:02:40<7:01:36,  1.48s/it]
 30%|███       | 7328/24420 [3:02:42<7:02:15,  1.48s/it]
 30%|███       | 7329/24420 [3:02:43<7:01:48,  1.48s/it]
 30%|███       | 7330/24420 [3:02:44<7:01:44,  1.48s/it]
                                                        
{'loss': 0.1262, 'learning_rate': 5e-05, 'epoch': 1.5}

 30%|███       | 7330/24420 [3:02:44<7:01:44,  1.48s/it]
 30%|███       | 7331/24420 [3:02:46<7:02:18,  1.48s/it]
 30%|███       | 7332/24420 [3:02:47<7:01:59,  1.48s/it]
 30%|███       | 7333/24420 [3:02:49<7:01:54,  1.48s/it]
 30%|███       | 7334/24420 [3:02:50<7:01:37,  1.48s/it]
 30%|███       | 7335/24420 [3:02:52<7:01:37,  1.48s/it]
                                                        
{'loss': 0.1131, 'learning_rate': 5e-05, 'epoch': 1.5}

 30%|███       | 7335/24420 [3:02:52<7:01:37,  1.48s/it]
 30%|███       | 7336/24420 [3:02:53<7:01:38,  1.48s/it]
 30%|███       | 7337/24420 [3:02:55<7:01:59,  1.48s/it]
 30%|███       | 7338/24420 [3:02:56<7:02:00,  1.48s/it]
 30%|███       | 7339/24420 [3:02:58<7:02:04,  1.48s/it]
 30%|███       | 7340/24420 [3:02:59<7:01:44,  1.48s/it]
                                                        
{'loss': 0.1142, 'learning_rate': 5e-05, 'epoch': 1.5}

 30%|███       | 7340/24420 [3:02:59<7:01:44,  1.48s/it]
 30%|███       | 7341/24420 [3:03:01<7:01:43,  1.48s/it]
 30%|███       | 7342/24420 [3:03:02<7:02:33,  1.48s/it]
 30%|███       | 7343/24420 [3:03:04<7:02:31,  1.48s/it]
 30%|███       | 7344/24420 [3:03:05<7:02:08,  1.48s/it]
 30%|███       | 7345/24420 [3:03:07<7:01:20,  1.48s/it]
                                                        
{'loss': 0.1204, 'learning_rate': 5e-05, 'epoch': 1.5}

 30%|███       | 7345/24420 [3:03:07<7:01:20,  1.48s/it]
 30%|███       | 7346/24420 [3:03:08<7:01:47,  1.48s/it]
 30%|███       | 7347/24420 [3:03:10<7:01:31,  1.48s/it]
 30%|███       | 7348/24420 [3:03:11<7:01:42,  1.48s/it]
 30%|███       | 7349/24420 [3:03:13<7:02:00,  1.48s/it]
 30%|███       | 7350/24420 [3:03:14<7:01:55,  1.48s/it]
                                                        
{'loss': 0.1265, 'learning_rate': 5e-05, 'epoch': 1.5}

 30%|███       | 7350/24420 [3:03:14<7:01:55,  1.48s/it]
 30%|███       | 7351/24420 [3:03:16<7:02:19,  1.48s/it]
 30%|███       | 7352/24420 [3:03:17<7:01:14,  1.48s/it]
 30%|███       | 7353/24420 [3:03:19<7:01:43,  1.48s/it]
 30%|███       | 7354/24420 [3:03:20<7:01:59,  1.48s/it]
 30%|███       | 7355/24420 [3:03:22<7:01:04,  1.48s/it]
                                                        
{'loss': 0.099, 'learning_rate': 5e-05, 'epoch': 1.51}

 30%|███       | 7355/24420 [3:03:22<7:01:04,  1.48s/it]
 30%|███       | 7356/24420 [3:03:23<7:01:13,  1.48s/it]
 30%|███       | 7357/24420 [3:03:24<7:01:09,  1.48s/it]
 30%|███       | 7358/24420 [3:03:26<7:01:12,  1.48s/it]
 30%|███       | 7359/24420 [3:03:27<7:01:09,  1.48s/it]
 30%|███       | 7360/24420 [3:03:29<7:01:19,  1.48s/it]
                                                        
{'loss': 0.1275, 'learning_rate': 5e-05, 'epoch': 1.51}

 30%|███       | 7360/24420 [3:03:29<7:01:19,  1.48s/it]
 30%|███       | 7361/24420 [3:03:30<7:01:41,  1.48s/it]
 30%|███       | 7362/24420 [3:03:32<7:00:37,  1.48s/it]
 30%|███       | 7363/24420 [3:03:33<7:01:22,  1.48s/it]
 30%|███       | 7364/24420 [3:03:35<7:01:15,  1.48s/it]
 30%|███       | 7365/24420 [3:03:36<7:01:47,  1.48s/it]
                                                        
{'loss': 0.1199, 'learning_rate': 5e-05, 'epoch': 1.51}

 30%|███       | 7365/24420 [3:03:36<7:01:47,  1.48s/it]
 30%|███       | 7366/24420 [3:03:38<7:01:47,  1.48s/it]
 30%|███       | 7367/24420 [3:03:39<7:01:33,  1.48s/it]
 30%|███       | 7368/24420 [3:03:41<7:01:40,  1.48s/it]
 30%|███       | 7369/24420 [3:03:42<7:01:10,  1.48s/it]
 30%|███       | 7370/24420 [3:03:44<7:00:47,  1.48s/it]
                                                        
{'loss': 0.1365, 'learning_rate': 5e-05, 'epoch': 1.51}

 30%|███       | 7370/24420 [3:03:44<7:00:47,  1.48s/it]
 30%|███       | 7371/24420 [3:03:45<7:02:07,  1.49s/it]
 30%|███       | 7372/24420 [3:03:47<7:01:28,  1.48s/it]
 30%|███       | 7373/24420 [3:03:48<7:01:30,  1.48s/it]
 30%|███       | 7374/24420 [3:03:50<7:01:29,  1.48s/it]
 30%|███       | 7375/24420 [3:03:51<7:01:38,  1.48s/it]
                                                        
{'loss': 0.1154, 'learning_rate': 5e-05, 'epoch': 1.51}

 30%|███       | 7375/24420 [3:03:51<7:01:38,  1.48s/it]
 30%|███       | 7376/24420 [3:03:53<7:01:18,  1.48s/it]
 30%|███       | 7377/24420 [3:03:54<7:02:03,  1.49s/it]
 30%|███       | 7378/24420 [3:03:56<7:01:17,  1.48s/it]
 30%|███       | 7379/24420 [3:03:57<7:00:33,  1.48s/it]
 30%|███       | 7380/24420 [3:03:59<6:59:53,  1.48s/it]
                                                        
{'loss': 0.125, 'learning_rate': 5e-05, 'epoch': 1.51}

 30%|███       | 7380/24420 [3:03:59<6:59:53,  1.48s/it]
 30%|███       | 7381/24420 [3:04:00<7:00:08,  1.48s/it]
 30%|███       | 7382/24420 [3:04:02<7:00:29,  1.48s/it]
 30%|███       | 7383/24420 [3:04:03<7:00:40,  1.48s/it]
 30%|███       | 7384/24420 [3:04:05<7:00:44,  1.48s/it]
 30%|███       | 7385/24420 [3:04:06<7:00:12,  1.48s/it]
                                                        
{'loss': 0.0993, 'learning_rate': 5e-05, 'epoch': 1.51}

 30%|███       | 7385/24420 [3:04:06<7:00:12,  1.48s/it]
 30%|███       | 7386/24420 [3:04:07<7:00:12,  1.48s/it]
 30%|███       | 7387/24420 [3:04:09<7:00:24,  1.48s/it]
 30%|███       | 7388/24420 [3:04:10<7:00:12,  1.48s/it]
 30%|███       | 7389/24420 [3:04:12<7:00:07,  1.48s/it]
 30%|███       | 7390/24420 [3:04:13<7:00:12,  1.48s/it]
                                                        
{'loss': 0.1093, 'learning_rate': 5e-05, 'epoch': 1.51}

 30%|███       | 7390/24420 [3:04:13<7:00:12,  1.48s/it]
 30%|███       | 7391/24420 [3:04:15<7:00:14,  1.48s/it]
 30%|███       | 7392/24420 [3:04:16<7:00:40,  1.48s/it]
 30%|███       | 7393/24420 [3:04:18<7:00:32,  1.48s/it]
 30%|███       | 7394/24420 [3:04:19<7:00:00,  1.48s/it]
 30%|███       | 7395/24420 [3:04:21<6:59:45,  1.48s/it]
                                                        
{'loss': 0.1022, 'learning_rate': 5e-05, 'epoch': 1.51}

 30%|███       | 7395/24420 [3:04:21<6:59:45,  1.48s/it]
 30%|███       | 7396/24420 [3:04:22<6:59:33,  1.48s/it]
 30%|███       | 7397/24420 [3:04:24<6:59:37,  1.48s/it]
 30%|███       | 7398/24420 [3:04:25<7:00:53,  1.48s/it]
 30%|███       | 7399/24420 [3:04:27<7:00:17,  1.48s/it]
 30%|███       | 7400/24420 [3:04:28<6:59:32,  1.48s/it]
                                                        
{'loss': 0.1017, 'learning_rate': 5e-05, 'epoch': 1.52}

 30%|███       | 7400/24420 [3:04:28<6:59:32,  1.48s/it]
 30%|███       | 7401/24420 [3:04:33<11:22:14,  2.41s/it]
 30%|███       | 7402/24420 [3:04:34<10:03:30,  2.13s/it]
 30%|███       | 7403/24420 [3:04:36<9:09:06,  1.94s/it] 
 30%|███       | 7404/24420 [3:04:37<8:30:19,  1.80s/it]
 30%|███       | 7405/24420 [3:04:39<8:02:55,  1.70s/it]
                                                        
{'loss': 0.1336, 'learning_rate': 5e-05, 'epoch': 1.52}

 30%|███       | 7405/24420 [3:04:39<8:02:55,  1.70s/it]
 30%|███       | 7406/24420 [3:04:40<7:44:36,  1.64s/it]
 30%|███       | 7407/24420 [3:04:42<7:31:13,  1.59s/it]
 30%|███       | 7408/24420 [3:04:43<7:21:44,  1.56s/it]
 30%|███       | 7409/24420 [3:04:45<7:14:54,  1.53s/it]
 30%|███       | 7410/24420 [3:04:46<7:10:43,  1.52s/it]
                                                        
{'loss': 0.1348, 'learning_rate': 5e-05, 'epoch': 1.52}

 30%|███       | 7410/24420 [3:04:46<7:10:43,  1.52s/it]
 30%|███       | 7411/24420 [3:04:48<7:07:44,  1.51s/it]
 30%|███       | 7412/24420 [3:04:49<7:05:59,  1.50s/it]
 30%|███       | 7413/24420 [3:04:51<7:04:13,  1.50s/it]
 30%|███       | 7414/24420 [3:04:52<7:03:03,  1.49s/it]
 30%|███       | 7415/24420 [3:04:54<7:02:07,  1.49s/it]
                                                        
{'loss': 0.1298, 'learning_rate': 5e-05, 'epoch': 1.52}

 30%|███       | 7415/24420 [3:04:54<7:02:07,  1.49s/it]
 30%|███       | 7416/24420 [3:04:55<7:01:23,  1.49s/it]
 30%|███       | 7417/24420 [3:04:56<7:00:55,  1.49s/it]
 30%|███       | 7418/24420 [3:04:58<7:01:07,  1.49s/it]
 30%|███       | 7419/24420 [3:04:59<7:00:41,  1.48s/it]
 30%|███       | 7420/24420 [3:05:01<7:00:40,  1.48s/it]
                                                        
{'loss': 0.1031, 'learning_rate': 5e-05, 'epoch': 1.52}

 30%|███       | 7420/24420 [3:05:01<7:00:40,  1.48s/it]
 30%|███       | 7421/24420 [3:05:02<7:00:23,  1.48s/it]
 30%|███       | 7422/24420 [3:05:04<7:01:23,  1.49s/it]
 30%|███       | 7423/24420 [3:05:05<7:01:00,  1.49s/it]
 30%|███       | 7424/24420 [3:05:07<7:00:39,  1.49s/it]
 30%|███       | 7425/24420 [3:05:08<7:00:38,  1.49s/it]
                                                        
{'loss': 0.1647, 'learning_rate': 5e-05, 'epoch': 1.52}

 30%|███       | 7425/24420 [3:05:08<7:00:38,  1.49s/it]
 30%|███       | 7426/24420 [3:05:10<7:00:39,  1.49s/it]
 30%|███       | 7427/24420 [3:05:11<7:00:27,  1.48s/it]
 30%|███       | 7428/24420 [3:05:13<6:59:49,  1.48s/it]
 30%|███       | 7429/24420 [3:05:14<6:59:03,  1.48s/it]
 30%|███       | 7430/24420 [3:05:16<6:58:56,  1.48s/it]
                                                        
{'loss': 0.0953, 'learning_rate': 5e-05, 'epoch': 1.52}

 30%|███       | 7430/24420 [3:05:16<6:58:56,  1.48s/it]
 30%|███       | 7431/24420 [3:05:17<6:58:53,  1.48s/it]
 30%|███       | 7432/24420 [3:05:19<6:58:46,  1.48s/it]
 30%|███       | 7433/24420 [3:05:20<6:58:10,  1.48s/it]
 30%|███       | 7434/24420 [3:05:22<6:58:27,  1.48s/it]
 30%|███       | 7435/24420 [3:05:23<6:58:35,  1.48s/it]
                                                        
{'loss': 0.1137, 'learning_rate': 5e-05, 'epoch': 1.52}

 30%|███       | 7435/24420 [3:05:23<6:58:35,  1.48s/it]
 30%|███       | 7436/24420 [3:05:25<6:58:45,  1.48s/it]
 30%|███       | 7437/24420 [3:05:26<6:58:38,  1.48s/it]
 30%|███       | 7438/24420 [3:05:28<6:58:38,  1.48s/it]
 30%|███       | 7439/24420 [3:05:29<6:59:23,  1.48s/it]
 30%|███       | 7440/24420 [3:05:31<6:59:03,  1.48s/it]
                                                        
{'loss': 0.1174, 'learning_rate': 5e-05, 'epoch': 1.52}

 30%|███       | 7440/24420 [3:05:31<6:59:03,  1.48s/it]
 30%|███       | 7441/24420 [3:05:32<6:59:17,  1.48s/it]
 30%|███       | 7442/24420 [3:05:34<6:59:13,  1.48s/it]
 30%|███       | 7443/24420 [3:05:35<6:59:14,  1.48s/it]
 30%|███       | 7444/24420 [3:05:36<6:58:39,  1.48s/it]
 30%|███       | 7445/24420 [3:05:38<6:58:22,  1.48s/it]
                                                        
{'loss': 0.1157, 'learning_rate': 5e-05, 'epoch': 1.52}

 30%|███       | 7445/24420 [3:05:38<6:58:22,  1.48s/it]
 30%|███       | 7446/24420 [3:05:39<6:59:08,  1.48s/it]
 30%|███       | 7447/24420 [3:05:41<6:58:55,  1.48s/it]
 30%|███       | 7448/24420 [3:05:42<6:58:54,  1.48s/it]
 31%|███       | 7449/24420 [3:05:44<6:59:03,  1.48s/it]
 31%|███       | 7450/24420 [3:05:45<6:59:16,  1.48s/it]
                                                        
{'loss': 0.1128, 'learning_rate': 5e-05, 'epoch': 1.53}

 31%|███       | 7450/24420 [3:05:45<6:59:16,  1.48s/it]
 31%|███       | 7451/24420 [3:05:47<6:59:31,  1.48s/it]
 31%|███       | 7452/24420 [3:05:48<6:58:43,  1.48s/it]
 31%|███       | 7453/24420 [3:05:50<6:58:31,  1.48s/it]
 31%|███       | 7454/24420 [3:05:51<6:58:52,  1.48s/it]
 31%|███       | 7455/24420 [3:05:53<6:59:37,  1.48s/it]
                                                        
{'loss': 0.1059, 'learning_rate': 5e-05, 'epoch': 1.53}

 31%|███       | 7455/24420 [3:05:53<6:59:37,  1.48s/it]
 31%|███       | 7456/24420 [3:05:54<6:59:25,  1.48s/it]
 31%|███       | 7457/24420 [3:05:56<7:00:14,  1.49s/it]
 31%|███       | 7458/24420 [3:05:57<6:59:35,  1.48s/it]
 31%|███       | 7459/24420 [3:05:59<6:59:14,  1.48s/it]
 31%|███       | 7460/24420 [3:06:00<6:58:27,  1.48s/it]
                                                        
{'loss': 0.1214, 'learning_rate': 5e-05, 'epoch': 1.53}

 31%|███       | 7460/24420 [3:06:00<6:58:27,  1.48s/it]
 31%|███       | 7461/24420 [3:06:02<6:58:51,  1.48s/it]
 31%|███       | 7462/24420 [3:06:03<6:58:55,  1.48s/it]
 31%|███       | 7463/24420 [3:06:05<6:58:13,  1.48s/it]
 31%|███       | 7464/24420 [3:06:06<6:58:18,  1.48s/it]
 31%|███       | 7465/24420 [3:06:08<6:59:02,  1.48s/it]
                                                        
{'loss': 0.1214, 'learning_rate': 5e-05, 'epoch': 1.53}

 31%|███       | 7465/24420 [3:06:08<6:59:02,  1.48s/it]
 31%|███       | 7466/24420 [3:06:09<6:59:20,  1.48s/it]
 31%|███       | 7467/24420 [3:06:11<6:58:57,  1.48s/it]
 31%|███       | 7468/24420 [3:06:12<6:58:44,  1.48s/it]
 31%|███       | 7469/24420 [3:06:14<6:58:20,  1.48s/it]
 31%|███       | 7470/24420 [3:06:15<6:58:13,  1.48s/it]
                                                        
{'loss': 0.1105, 'learning_rate': 5e-05, 'epoch': 1.53}

 31%|███       | 7470/24420 [3:06:15<6:58:13,  1.48s/it]
 31%|███       | 7471/24420 [3:06:17<6:58:08,  1.48s/it]
 31%|███       | 7472/24420 [3:06:18<6:58:47,  1.48s/it]
 31%|███       | 7473/24420 [3:06:19<6:58:56,  1.48s/it]
 31%|███       | 7474/24420 [3:06:21<6:59:20,  1.48s/it]
 31%|███       | 7475/24420 [3:06:22<6:58:52,  1.48s/it]
                                                        
{'loss': 0.1308, 'learning_rate': 5e-05, 'epoch': 1.53}

 31%|███       | 7475/24420 [3:06:22<6:58:52,  1.48s/it]
 31%|███       | 7476/24420 [3:06:24<6:58:31,  1.48s/it]
 31%|███       | 7477/24420 [3:06:25<6:58:13,  1.48s/it]
 31%|███       | 7478/24420 [3:06:27<6:57:25,  1.48s/it]
 31%|███       | 7479/24420 [3:06:28<6:57:06,  1.48s/it]
 31%|███       | 7480/24420 [3:06:30<6:58:06,  1.48s/it]
                                                        
{'loss': 0.125, 'learning_rate': 5e-05, 'epoch': 1.53}

 31%|███       | 7480/24420 [3:06:30<6:58:06,  1.48s/it]
 31%|███       | 7481/24420 [3:06:31<6:58:44,  1.48s/it]
 31%|███       | 7482/24420 [3:06:33<6:58:50,  1.48s/it]
 31%|███       | 7483/24420 [3:06:34<6:58:19,  1.48s/it]
 31%|███       | 7484/24420 [3:06:36<6:59:50,  1.49s/it]
 31%|███       | 7485/24420 [3:06:37<6:59:44,  1.49s/it]
                                                        
{'loss': 0.134, 'learning_rate': 5e-05, 'epoch': 1.53}

 31%|███       | 7485/24420 [3:06:37<6:59:44,  1.49s/it]
 31%|███       | 7486/24420 [3:06:39<6:59:12,  1.49s/it]
 31%|███       | 7487/24420 [3:06:40<6:59:11,  1.49s/it]
 31%|███       | 7488/24420 [3:06:42<6:57:42,  1.48s/it]
 31%|███       | 7489/24420 [3:06:43<6:57:29,  1.48s/it]
 31%|███       | 7490/24420 [3:06:45<6:58:15,  1.48s/it]
                                                        
{'loss': 0.0948, 'learning_rate': 5e-05, 'epoch': 1.53}

 31%|███       | 7490/24420 [3:06:45<6:58:15,  1.48s/it]
 31%|███       | 7491/24420 [3:06:46<6:58:18,  1.48s/it]
 31%|███       | 7492/24420 [3:06:48<6:58:16,  1.48s/it]
 31%|███       | 7493/24420 [3:06:49<6:58:00,  1.48s/it]
 31%|███       | 7494/24420 [3:06:51<6:57:47,  1.48s/it]
 31%|███       | 7495/24420 [3:06:52<6:57:14,  1.48s/it]
                                                        
{'loss': 0.1139, 'learning_rate': 5e-05, 'epoch': 1.53}

 31%|███       | 7495/24420 [3:06:52<6:57:14,  1.48s/it]
 31%|███       | 7496/24420 [3:06:54<6:57:46,  1.48s/it]
 31%|███       | 7497/24420 [3:06:55<6:57:39,  1.48s/it]
 31%|███       | 7498/24420 [3:06:57<6:57:38,  1.48s/it]
 31%|███       | 7499/24420 [3:06:58<6:57:21,  1.48s/it]
 31%|███       | 7500/24420 [3:06:59<6:57:30,  1.48s/it]
                                                        
{'loss': 0.1159, 'learning_rate': 5e-05, 'epoch': 1.54}

 31%|███       | 7500/24420 [3:06:59<6:57:30,  1.48s/it]
 31%|███       | 7501/24420 [3:07:01<6:58:40,  1.48s/it]
 31%|███       | 7502/24420 [3:07:02<6:58:14,  1.48s/it]
 31%|███       | 7503/24420 [3:07:04<6:57:26,  1.48s/it]
 31%|███       | 7504/24420 [3:07:05<6:57:07,  1.48s/it]
 31%|███       | 7505/24420 [3:07:07<6:56:35,  1.48s/it]
                                                        
{'loss': 0.1333, 'learning_rate': 5e-05, 'epoch': 1.54}

 31%|███       | 7505/24420 [3:07:07<6:56:35,  1.48s/it]
 31%|███       | 7506/24420 [3:07:08<6:57:57,  1.48s/it]
 31%|███       | 7507/24420 [3:07:10<6:57:38,  1.48s/it]
 31%|███       | 7508/24420 [3:07:11<6:57:18,  1.48s/it]
 31%|███       | 7509/24420 [3:07:13<6:57:40,  1.48s/it]
 31%|███       | 7510/24420 [3:07:14<6:57:08,  1.48s/it]
                                                        
{'loss': 0.1184, 'learning_rate': 5e-05, 'epoch': 1.54}

 31%|███       | 7510/24420 [3:07:14<6:57:08,  1.48s/it]
 31%|███       | 7511/24420 [3:07:16<6:57:07,  1.48s/it]
 31%|███       | 7512/24420 [3:07:17<6:57:29,  1.48s/it]
 31%|███       | 7513/24420 [3:07:19<6:58:38,  1.49s/it]
 31%|███       | 7514/24420 [3:07:20<6:58:12,  1.48s/it]
 31%|███       | 7515/24420 [3:07:22<6:57:51,  1.48s/it]
                                                        
{'loss': 0.0958, 'learning_rate': 5e-05, 'epoch': 1.54}

 31%|███       | 7515/24420 [3:07:22<6:57:51,  1.48s/it]
 31%|███       | 7516/24420 [3:07:23<6:58:20,  1.48s/it]
 31%|███       | 7517/24420 [3:07:25<6:57:46,  1.48s/it]
 31%|███       | 7518/24420 [3:07:26<6:59:11,  1.49s/it]
 31%|███       | 7519/24420 [3:07:28<7:00:08,  1.49s/it]
 31%|███       | 7520/24420 [3:07:29<6:59:37,  1.49s/it]
                                                        
{'loss': 0.1107, 'learning_rate': 5e-05, 'epoch': 1.54}

 31%|███       | 7520/24420 [3:07:29<6:59:37,  1.49s/it]
 31%|███       | 7521/24420 [3:07:31<6:59:21,  1.49s/it]
 31%|███       | 7522/24420 [3:07:32<6:58:32,  1.49s/it]
 31%|███       | 7523/24420 [3:07:34<6:57:48,  1.48s/it]
 31%|███       | 7524/24420 [3:07:35<6:57:27,  1.48s/it]
 31%|███       | 7525/24420 [3:07:37<6:56:59,  1.48s/it]
                                                        
{'loss': 0.1183, 'learning_rate': 5e-05, 'epoch': 1.54}

 31%|███       | 7525/24420 [3:07:37<6:56:59,  1.48s/it]
 31%|███       | 7526/24420 [3:07:38<6:57:22,  1.48s/it]
 31%|███       | 7527/24420 [3:07:40<6:58:35,  1.49s/it]
 31%|███       | 7528/24420 [3:07:41<6:57:53,  1.48s/it]
 31%|███       | 7529/24420 [3:07:43<6:58:37,  1.49s/it]
 31%|███       | 7530/24420 [3:07:44<6:57:25,  1.48s/it]
                                                        
{'loss': 0.1222, 'learning_rate': 5e-05, 'epoch': 1.54}

 31%|███       | 7530/24420 [3:07:44<6:57:25,  1.48s/it]
 31%|███       | 7531/24420 [3:07:45<6:56:58,  1.48s/it]
 31%|███       | 7532/24420 [3:07:47<6:57:29,  1.48s/it]
 31%|███       | 7533/24420 [3:07:48<6:57:05,  1.48s/it]
 31%|███       | 7534/24420 [3:07:50<6:57:34,  1.48s/it]
 31%|███       | 7535/24420 [3:07:51<6:57:06,  1.48s/it]
                                                        
{'loss': 0.1247, 'learning_rate': 5e-05, 'epoch': 1.54}

 31%|███       | 7535/24420 [3:07:51<6:57:06,  1.48s/it]
 31%|███       | 7536/24420 [3:07:53<6:57:26,  1.48s/it]
 31%|███       | 7537/24420 [3:07:54<6:57:29,  1.48s/it]
 31%|███       | 7538/24420 [3:07:56<6:57:32,  1.48s/it]
 31%|███       | 7539/24420 [3:07:57<6:58:05,  1.49s/it]
 31%|███       | 7540/24420 [3:07:59<6:57:09,  1.48s/it]
                                                        
{'loss': 0.1407, 'learning_rate': 5e-05, 'epoch': 1.54}

 31%|███       | 7540/24420 [3:07:59<6:57:09,  1.48s/it]
 31%|███       | 7541/24420 [3:08:00<6:57:00,  1.48s/it]
 31%|███       | 7542/24420 [3:08:02<6:57:54,  1.49s/it]
 31%|███       | 7543/24420 [3:08:03<6:57:26,  1.48s/it]
 31%|███       | 7544/24420 [3:08:05<6:58:39,  1.49s/it]
 31%|███       | 7545/24420 [3:08:06<6:57:56,  1.49s/it]
                                                        
{'loss': 0.1172, 'learning_rate': 5e-05, 'epoch': 1.54}

 31%|███       | 7545/24420 [3:08:06<6:57:56,  1.49s/it]
 31%|███       | 7546/24420 [3:08:08<6:57:29,  1.48s/it]
 31%|███       | 7547/24420 [3:08:09<6:57:06,  1.48s/it]
 31%|███       | 7548/24420 [3:08:11<6:56:59,  1.48s/it]
 31%|███       | 7549/24420 [3:08:12<6:56:59,  1.48s/it]
 31%|███       | 7550/24420 [3:08:14<6:56:18,  1.48s/it]
                                                        
{'loss': 0.1106, 'learning_rate': 5e-05, 'epoch': 1.55}

 31%|███       | 7550/24420 [3:08:14<6:56:18,  1.48s/it]
 31%|███       | 7551/24420 [3:08:15<6:56:12,  1.48s/it]
 31%|███       | 7552/24420 [3:08:17<6:56:14,  1.48s/it]
 31%|███       | 7553/24420 [3:08:18<6:56:24,  1.48s/it]
 31%|███       | 7554/24420 [3:08:20<6:56:18,  1.48s/it]
 31%|███       | 7555/24420 [3:08:21<6:56:41,  1.48s/it]
                                                        
{'loss': 0.1234, 'learning_rate': 5e-05, 'epoch': 1.55}

 31%|███       | 7555/24420 [3:08:21<6:56:41,  1.48s/it]
 31%|███       | 7556/24420 [3:08:23<6:57:36,  1.49s/it]
 31%|███       | 7557/24420 [3:08:24<6:57:14,  1.48s/it]
 31%|███       | 7558/24420 [3:08:26<6:57:12,  1.48s/it]
 31%|███       | 7559/24420 [3:08:27<6:57:05,  1.48s/it]
 31%|███       | 7560/24420 [3:08:28<6:56:45,  1.48s/it]
                                                        
{'loss': 0.1306, 'learning_rate': 5e-05, 'epoch': 1.55}

 31%|███       | 7560/24420 [3:08:28<6:56:45,  1.48s/it]
 31%|███       | 7561/24420 [3:08:30<6:57:09,  1.48s/it]
 31%|███       | 7562/24420 [3:08:31<6:56:53,  1.48s/it]
 31%|███       | 7563/24420 [3:08:33<6:57:05,  1.48s/it]
 31%|███       | 7564/24420 [3:08:34<6:57:00,  1.48s/it]
 31%|███       | 7565/24420 [3:08:36<6:56:19,  1.48s/it]
                                                        
{'loss': 0.114, 'learning_rate': 5e-05, 'epoch': 1.55}

 31%|███       | 7565/24420 [3:08:36<6:56:19,  1.48s/it]
 31%|███       | 7566/24420 [3:08:37<6:56:12,  1.48s/it]
 31%|███       | 7567/24420 [3:08:39<6:55:44,  1.48s/it]
 31%|███       | 7568/24420 [3:08:40<6:56:00,  1.48s/it]
 31%|███       | 7569/24420 [3:08:42<6:56:11,  1.48s/it]
 31%|███       | 7570/24420 [3:08:43<6:55:18,  1.48s/it]
                                                        
{'loss': 0.1126, 'learning_rate': 5e-05, 'epoch': 1.55}

 31%|███       | 7570/24420 [3:08:43<6:55:18,  1.48s/it]
 31%|███       | 7571/24420 [3:08:45<6:56:12,  1.48s/it]
 31%|███       | 7572/24420 [3:08:46<6:56:08,  1.48s/it]
 31%|███       | 7573/24420 [3:08:48<6:56:25,  1.48s/it]
 31%|███       | 7574/24420 [3:08:49<6:56:43,  1.48s/it]
 31%|███       | 7575/24420 [3:08:51<6:56:34,  1.48s/it]
                                                        
{'loss': 0.1186, 'learning_rate': 5e-05, 'epoch': 1.55}

 31%|███       | 7575/24420 [3:08:51<6:56:34,  1.48s/it]
 31%|███       | 7576/24420 [3:08:52<6:56:25,  1.48s/it]
 31%|███       | 7577/24420 [3:08:54<6:56:39,  1.48s/it]
 31%|███       | 7578/24420 [3:08:55<6:56:05,  1.48s/it]
 31%|███       | 7579/24420 [3:08:57<6:55:37,  1.48s/it]
 31%|███       | 7580/24420 [3:08:58<6:55:27,  1.48s/it]
                                                        
{'loss': 0.1238, 'learning_rate': 5e-05, 'epoch': 1.55}

 31%|███       | 7580/24420 [3:08:58<6:55:27,  1.48s/it]
 31%|███       | 7581/24420 [3:09:00<6:57:37,  1.49s/it]
 31%|███       | 7582/24420 [3:09:01<6:57:28,  1.49s/it]
 31%|███       | 7583/24420 [3:09:03<6:56:32,  1.48s/it]
 31%|███       | 7584/24420 [3:09:04<6:55:42,  1.48s/it]
 31%|███       | 7585/24420 [3:09:06<6:56:04,  1.48s/it]
                                                        
{'loss': 0.1313, 'learning_rate': 5e-05, 'epoch': 1.55}

 31%|███       | 7585/24420 [3:09:06<6:56:04,  1.48s/it]
 31%|███       | 7586/24420 [3:09:07<6:55:56,  1.48s/it]
 31%|███       | 7587/24420 [3:09:09<6:56:09,  1.48s/it]
 31%|███       | 7588/24420 [3:09:10<6:55:37,  1.48s/it]
 31%|███       | 7589/24420 [3:09:12<6:55:54,  1.48s/it]
 31%|███       | 7590/24420 [3:09:13<6:55:23,  1.48s/it]
                                                        
{'loss': 0.1188, 'learning_rate': 5e-05, 'epoch': 1.55}

 31%|███       | 7590/24420 [3:09:13<6:55:23,  1.48s/it]
 31%|███       | 7591/24420 [3:09:14<6:55:28,  1.48s/it]
 31%|███       | 7592/24420 [3:09:16<6:55:32,  1.48s/it]
 31%|███       | 7593/24420 [3:09:17<6:56:02,  1.48s/it]
 31%|███       | 7594/24420 [3:09:19<6:55:41,  1.48s/it]
 31%|███       | 7595/24420 [3:09:20<6:55:37,  1.48s/it]
                                                        
{'loss': 0.1135, 'learning_rate': 5e-05, 'epoch': 1.56}

 31%|███       | 7595/24420 [3:09:20<6:55:37,  1.48s/it]
 31%|███       | 7596/24420 [3:09:22<6:56:23,  1.48s/it]
 31%|███       | 7597/24420 [3:09:23<6:56:27,  1.49s/it]
 31%|███       | 7598/24420 [3:09:25<6:55:53,  1.48s/it]
 31%|███       | 7599/24420 [3:09:26<6:56:41,  1.49s/it]
 31%|███       | 7600/24420 [3:09:28<6:56:01,  1.48s/it]
                                                        
{'loss': 0.1311, 'learning_rate': 5e-05, 'epoch': 1.56}

 31%|███       | 7600/24420 [3:09:28<6:56:01,  1.48s/it]
 31%|███       | 7601/24420 [3:09:32<11:12:38,  2.40s/it]
 31%|███       | 7602/24420 [3:09:34<9:55:25,  2.12s/it] 
 31%|███       | 7603/24420 [3:09:35<9:01:37,  1.93s/it]
 31%|███       | 7604/24420 [3:09:37<8:23:14,  1.80s/it]
 31%|███       | 7605/24420 [3:09:38<7:57:20,  1.70s/it]
                                                        
{'loss': 0.1247, 'learning_rate': 5e-05, 'epoch': 1.56}

 31%|███       | 7605/24420 [3:09:38<7:57:20,  1.70s/it]
 31%|███       | 7606/24420 [3:09:40<7:38:20,  1.64s/it]
 31%|███       | 7607/24420 [3:09:41<7:24:49,  1.59s/it]
 31%|███       | 7608/24420 [3:09:43<7:16:49,  1.56s/it]
 31%|███       | 7609/24420 [3:09:44<7:11:24,  1.54s/it]
 31%|███       | 7610/24420 [3:09:46<7:06:21,  1.52s/it]
                                                        
{'loss': 0.1126, 'learning_rate': 5e-05, 'epoch': 1.56}

 31%|███       | 7610/24420 [3:09:46<7:06:21,  1.52s/it]
 31%|███       | 7611/24420 [3:09:47<7:03:00,  1.51s/it]
 31%|███       | 7612/24420 [3:09:49<6:59:45,  1.50s/it]
 31%|███       | 7613/24420 [3:09:50<6:57:42,  1.49s/it]
 31%|███       | 7614/24420 [3:09:52<6:56:26,  1.49s/it]
 31%|███       | 7615/24420 [3:09:53<6:56:03,  1.49s/it]
                                                        
{'loss': 0.1111, 'learning_rate': 5e-05, 'epoch': 1.56}

 31%|███       | 7615/24420 [3:09:53<6:56:03,  1.49s/it]
 31%|███       | 7616/24420 [3:09:55<6:56:18,  1.49s/it]
 31%|███       | 7617/24420 [3:09:56<6:55:40,  1.48s/it]
 31%|███       | 7618/24420 [3:09:58<6:55:18,  1.48s/it]
 31%|███       | 7619/24420 [3:09:59<6:54:16,  1.48s/it]
 31%|███       | 7620/24420 [3:10:00<6:54:05,  1.48s/it]
                                                        
{'loss': 0.108, 'learning_rate': 5e-05, 'epoch': 1.56}

 31%|███       | 7620/24420 [3:10:00<6:54:05,  1.48s/it]
 31%|███       | 7621/24420 [3:10:02<6:54:25,  1.48s/it]
 31%|███       | 7622/24420 [3:10:03<6:54:21,  1.48s/it]
 31%|███       | 7623/24420 [3:10:05<6:54:02,  1.48s/it]
 31%|███       | 7624/24420 [3:10:06<6:54:11,  1.48s/it]
 31%|███       | 7625/24420 [3:10:08<6:54:36,  1.48s/it]
                                                        
{'loss': 0.1115, 'learning_rate': 5e-05, 'epoch': 1.56}

 31%|███       | 7625/24420 [3:10:08<6:54:36,  1.48s/it]
 31%|███       | 7626/24420 [3:10:09<6:54:38,  1.48s/it]
 31%|███       | 7627/24420 [3:10:11<6:54:17,  1.48s/it]
 31%|███       | 7628/24420 [3:10:12<6:53:50,  1.48s/it]
 31%|███       | 7629/24420 [3:10:14<6:54:54,  1.48s/it]
 31%|███       | 7630/24420 [3:10:15<6:55:39,  1.49s/it]
                                                        
{'loss': 0.092, 'learning_rate': 5e-05, 'epoch': 1.56}

 31%|███       | 7630/24420 [3:10:15<6:55:39,  1.49s/it]
 31%|███       | 7631/24420 [3:10:17<6:55:18,  1.48s/it]
 31%|███▏      | 7632/24420 [3:10:18<6:56:03,  1.49s/it]
 31%|███▏      | 7633/24420 [3:10:20<6:55:04,  1.48s/it]
 31%|███▏      | 7634/24420 [3:10:21<6:54:24,  1.48s/it]
 31%|███▏      | 7635/24420 [3:10:23<6:54:47,  1.48s/it]
                                                        
{'loss': 0.1031, 'learning_rate': 5e-05, 'epoch': 1.56}

 31%|███▏      | 7635/24420 [3:10:23<6:54:47,  1.48s/it]
 31%|███▏      | 7636/24420 [3:10:24<6:54:52,  1.48s/it]
 31%|███▏      | 7637/24420 [3:10:26<6:54:29,  1.48s/it]
 31%|███▏      | 7638/24420 [3:10:27<6:54:02,  1.48s/it]
 31%|███▏      | 7639/24420 [3:10:29<6:53:53,  1.48s/it]
 31%|███▏      | 7640/24420 [3:10:30<6:54:03,  1.48s/it]
                                                        
{'loss': 0.1066, 'learning_rate': 5e-05, 'epoch': 1.56}

 31%|███▏      | 7640/24420 [3:10:30<6:54:03,  1.48s/it]
 31%|███▏      | 7641/24420 [3:10:32<6:54:07,  1.48s/it]
 31%|███▏      | 7642/24420 [3:10:33<6:53:41,  1.48s/it]
 31%|███▏      | 7643/24420 [3:10:35<6:55:07,  1.48s/it]
 31%|███▏      | 7644/24420 [3:10:36<6:54:52,  1.48s/it]
 31%|███▏      | 7645/24420 [3:10:38<6:54:20,  1.48s/it]
                                                        
{'loss': 0.1057, 'learning_rate': 5e-05, 'epoch': 1.57}

 31%|███▏      | 7645/24420 [3:10:38<6:54:20,  1.48s/it]
 31%|███▏      | 7646/24420 [3:10:39<6:54:17,  1.48s/it]
 31%|███▏      | 7647/24420 [3:10:41<6:54:22,  1.48s/it]
 31%|███▏      | 7648/24420 [3:10:42<6:54:03,  1.48s/it]
 31%|███▏      | 7649/24420 [3:10:43<6:53:49,  1.48s/it]
 31%|███▏      | 7650/24420 [3:10:45<6:54:48,  1.48s/it]
                                                        
{'loss': 0.1074, 'learning_rate': 5e-05, 'epoch': 1.57}

 31%|███▏      | 7650/24420 [3:10:45<6:54:48,  1.48s/it]
 31%|███▏      | 7651/24420 [3:10:46<6:54:51,  1.48s/it]
 31%|███▏      | 7652/24420 [3:10:48<6:54:39,  1.48s/it]
 31%|███▏      | 7653/24420 [3:10:49<6:54:50,  1.48s/it]
 31%|███▏      | 7654/24420 [3:10:51<6:54:58,  1.49s/it]
 31%|███▏      | 7655/24420 [3:10:52<6:54:48,  1.48s/it]
                                                        
{'loss': 0.1161, 'learning_rate': 5e-05, 'epoch': 1.57}

 31%|███▏      | 7655/24420 [3:10:52<6:54:48,  1.48s/it]
 31%|███▏      | 7656/24420 [3:10:54<6:55:12,  1.49s/it]
 31%|███▏      | 7657/24420 [3:10:55<6:54:44,  1.48s/it]
 31%|███▏      | 7658/24420 [3:10:57<6:53:47,  1.48s/it]
 31%|███▏      | 7659/24420 [3:10:58<6:53:52,  1.48s/it]
 31%|███▏      | 7660/24420 [3:11:00<6:55:02,  1.49s/it]
                                                        
{'loss': 0.1147, 'learning_rate': 5e-05, 'epoch': 1.57}

 31%|███▏      | 7660/24420 [3:11:00<6:55:02,  1.49s/it]
 31%|███▏      | 7661/24420 [3:11:01<6:54:13,  1.48s/it]
 31%|███▏      | 7662/24420 [3:11:03<6:53:48,  1.48s/it]
 31%|███▏      | 7663/24420 [3:11:04<6:53:21,  1.48s/it]
 31%|███▏      | 7664/24420 [3:11:06<6:53:31,  1.48s/it]
 31%|███▏      | 7665/24420 [3:11:07<6:53:56,  1.48s/it]
                                                        
{'loss': 0.1142, 'learning_rate': 5e-05, 'epoch': 1.57}

 31%|███▏      | 7665/24420 [3:11:07<6:53:56,  1.48s/it]
 31%|███▏      | 7666/24420 [3:11:09<6:53:57,  1.48s/it]
 31%|███▏      | 7667/24420 [3:11:10<6:53:25,  1.48s/it]
 31%|███▏      | 7668/24420 [3:11:12<6:53:17,  1.48s/it]
 31%|███▏      | 7669/24420 [3:11:13<6:54:01,  1.48s/it]
 31%|███▏      | 7670/24420 [3:11:15<6:53:54,  1.48s/it]
                                                        
{'loss': 0.1139, 'learning_rate': 5e-05, 'epoch': 1.57}

 31%|███▏      | 7670/24420 [3:11:15<6:53:54,  1.48s/it]
 31%|███▏      | 7671/24420 [3:11:16<6:54:28,  1.48s/it]
 31%|███▏      | 7672/24420 [3:11:18<6:54:02,  1.48s/it]
 31%|███▏      | 7673/24420 [3:11:19<6:53:57,  1.48s/it]
 31%|███▏      | 7674/24420 [3:11:21<6:53:47,  1.48s/it]
 31%|███▏      | 7675/24420 [3:11:22<6:53:32,  1.48s/it]
                                                        
{'loss': 0.107, 'learning_rate': 5e-05, 'epoch': 1.57}

 31%|███▏      | 7675/24420 [3:11:22<6:53:32,  1.48s/it]
 31%|███▏      | 7676/24420 [3:11:24<6:53:34,  1.48s/it]
 31%|███▏      | 7677/24420 [3:11:25<6:53:19,  1.48s/it]
 31%|███▏      | 7678/24420 [3:11:26<6:53:03,  1.48s/it]
 31%|███▏      | 7679/24420 [3:11:28<6:53:22,  1.48s/it]
 31%|███▏      | 7680/24420 [3:11:29<6:53:50,  1.48s/it]
                                                        
{'loss': 0.1297, 'learning_rate': 5e-05, 'epoch': 1.57}

 31%|███▏      | 7680/24420 [3:11:29<6:53:50,  1.48s/it]
 31%|███▏      | 7681/24420 [3:11:31<6:53:26,  1.48s/it]
 31%|███▏      | 7682/24420 [3:11:32<6:53:18,  1.48s/it]
 31%|███▏      | 7683/24420 [3:11:34<6:56:00,  1.49s/it]
 31%|███▏      | 7684/24420 [3:11:35<6:54:58,  1.49s/it]
 31%|███▏      | 7685/24420 [3:11:37<6:55:01,  1.49s/it]
                                                        
{'loss': 0.1256, 'learning_rate': 5e-05, 'epoch': 1.57}

 31%|███▏      | 7685/24420 [3:11:37<6:55:01,  1.49s/it]
 31%|███▏      | 7686/24420 [3:11:38<6:54:17,  1.49s/it]
 31%|███▏      | 7687/24420 [3:11:40<6:53:56,  1.48s/it]
 31%|███▏      | 7688/24420 [3:11:41<6:54:21,  1.49s/it]
 31%|███▏      | 7689/24420 [3:11:43<6:54:00,  1.48s/it]
 31%|███▏      | 7690/24420 [3:11:44<6:53:53,  1.48s/it]
                                                        
{'loss': 0.1216, 'learning_rate': 5e-05, 'epoch': 1.57}

 31%|███▏      | 7690/24420 [3:11:44<6:53:53,  1.48s/it]
 31%|███▏      | 7691/24420 [3:11:46<6:54:09,  1.49s/it]
 31%|███▏      | 7692/24420 [3:11:47<6:53:06,  1.48s/it]
 32%|███▏      | 7693/24420 [3:11:49<6:52:41,  1.48s/it]
 32%|███▏      | 7694/24420 [3:11:50<6:53:20,  1.48s/it]
 32%|███▏      | 7695/24420 [3:11:52<6:53:56,  1.48s/it]
                                                        
{'loss': 0.1294, 'learning_rate': 5e-05, 'epoch': 1.58}

 32%|███▏      | 7695/24420 [3:11:52<6:53:56,  1.48s/it]
 32%|███▏      | 7696/24420 [3:11:53<6:53:51,  1.48s/it]
 32%|███▏      | 7697/24420 [3:11:55<6:53:29,  1.48s/it]
 32%|███▏      | 7698/24420 [3:11:56<6:53:22,  1.48s/it]
 32%|███▏      | 7699/24420 [3:11:58<6:53:53,  1.49s/it]
 32%|███▏      | 7700/24420 [3:11:59<6:54:09,  1.49s/it]
                                                        
{'loss': 0.1122, 'learning_rate': 5e-05, 'epoch': 1.58}

 32%|███▏      | 7700/24420 [3:11:59<6:54:09,  1.49s/it]
 32%|███▏      | 7701/24420 [3:12:01<6:53:56,  1.49s/it]
 32%|███▏      | 7702/24420 [3:12:02<6:53:53,  1.49s/it]
 32%|███▏      | 7703/24420 [3:12:04<6:53:21,  1.48s/it]
 32%|███▏      | 7704/24420 [3:12:05<6:52:41,  1.48s/it]
 32%|███▏      | 7705/24420 [3:12:07<6:52:52,  1.48s/it]
                                                        
{'loss': 0.1087, 'learning_rate': 5e-05, 'epoch': 1.58}

 32%|███▏      | 7705/24420 [3:12:07<6:52:52,  1.48s/it]
 32%|███▏      | 7706/24420 [3:12:08<6:52:32,  1.48s/it]
 32%|███▏      | 7707/24420 [3:12:10<6:53:15,  1.48s/it]
 32%|███▏      | 7708/24420 [3:12:11<6:52:29,  1.48s/it]
 32%|███▏      | 7709/24420 [3:12:12<6:53:17,  1.48s/it]
 32%|███▏      | 7710/24420 [3:12:14<6:53:22,  1.48s/it]
                                                        
{'loss': 0.1054, 'learning_rate': 5e-05, 'epoch': 1.58}

 32%|███▏      | 7710/24420 [3:12:14<6:53:22,  1.48s/it]
 32%|███▏      | 7711/24420 [3:12:15<6:52:44,  1.48s/it]
 32%|███▏      | 7712/24420 [3:12:17<6:52:56,  1.48s/it]
 32%|███▏      | 7713/24420 [3:12:18<6:52:56,  1.48s/it]
 32%|███▏      | 7714/24420 [3:12:20<6:52:29,  1.48s/it]
 32%|███▏      | 7715/24420 [3:12:21<6:52:37,  1.48s/it]
                                                        
{'loss': 0.1222, 'learning_rate': 5e-05, 'epoch': 1.58}

 32%|███▏      | 7715/24420 [3:12:21<6:52:37,  1.48s/it]
 32%|███▏      | 7716/24420 [3:12:23<6:52:36,  1.48s/it]
 32%|███▏      | 7717/24420 [3:12:24<6:52:40,  1.48s/it]
 32%|███▏      | 7718/24420 [3:12:26<6:53:00,  1.48s/it]
 32%|███▏      | 7719/24420 [3:12:27<6:52:56,  1.48s/it]
 32%|███▏      | 7720/24420 [3:12:29<6:53:05,  1.48s/it]
                                                        
{'loss': 0.1454, 'learning_rate': 5e-05, 'epoch': 1.58}

 32%|███▏      | 7720/24420 [3:12:29<6:53:05,  1.48s/it]
 32%|███▏      | 7721/24420 [3:12:30<6:52:43,  1.48s/it]
 32%|███▏      | 7722/24420 [3:12:32<6:52:45,  1.48s/it]
 32%|███▏      | 7723/24420 [3:12:33<6:52:42,  1.48s/it]
 32%|███▏      | 7724/24420 [3:12:35<6:52:56,  1.48s/it]
 32%|███▏      | 7725/24420 [3:12:36<6:52:48,  1.48s/it]
                                                        
{'loss': 0.1457, 'learning_rate': 5e-05, 'epoch': 1.58}

 32%|███▏      | 7725/24420 [3:12:36<6:52:48,  1.48s/it]
 32%|███▏      | 7726/24420 [3:12:38<6:52:29,  1.48s/it]
 32%|███▏      | 7727/24420 [3:12:39<6:52:47,  1.48s/it]
 32%|███▏      | 7728/24420 [3:12:41<6:52:02,  1.48s/it]
 32%|███▏      | 7729/24420 [3:12:42<6:51:37,  1.48s/it]
 32%|███▏      | 7730/24420 [3:12:44<6:50:59,  1.48s/it]
                                                        
{'loss': 0.1244, 'learning_rate': 5e-05, 'epoch': 1.58}

 32%|███▏      | 7730/24420 [3:12:44<6:50:59,  1.48s/it]
 32%|███▏      | 7731/24420 [3:12:45<6:51:13,  1.48s/it]
 32%|███▏      | 7732/24420 [3:12:47<6:51:41,  1.48s/it]
 32%|███▏      | 7733/24420 [3:12:48<6:51:38,  1.48s/it]
 32%|███▏      | 7734/24420 [3:12:50<6:51:43,  1.48s/it]
 32%|███▏      | 7735/24420 [3:12:51<6:51:35,  1.48s/it]
                                                        
{'loss': 0.176, 'learning_rate': 5e-05, 'epoch': 1.58}

 32%|███▏      | 7735/24420 [3:12:51<6:51:35,  1.48s/it]
 32%|███▏      | 7736/24420 [3:12:52<6:51:53,  1.48s/it]
 32%|███▏      | 7737/24420 [3:12:54<6:52:14,  1.48s/it]
 32%|███▏      | 7738/24420 [3:12:55<6:52:17,  1.48s/it]
 32%|███▏      | 7739/24420 [3:12:57<6:51:59,  1.48s/it]
 32%|███▏      | 7740/24420 [3:12:58<6:52:26,  1.48s/it]
                                                        
{'loss': 0.1421, 'learning_rate': 5e-05, 'epoch': 1.58}

 32%|███▏      | 7740/24420 [3:12:58<6:52:26,  1.48s/it]
 32%|███▏      | 7741/24420 [3:13:00<6:52:21,  1.48s/it]
 32%|███▏      | 7742/24420 [3:13:01<6:51:57,  1.48s/it]
 32%|███▏      | 7743/24420 [3:13:03<6:52:08,  1.48s/it]
 32%|███▏      | 7744/24420 [3:13:04<6:52:02,  1.48s/it]
 32%|███▏      | 7745/24420 [3:13:06<6:52:45,  1.49s/it]
                                                        
{'loss': 0.1176, 'learning_rate': 5e-05, 'epoch': 1.59}

 32%|███▏      | 7745/24420 [3:13:06<6:52:45,  1.49s/it]
 32%|███▏      | 7746/24420 [3:13:07<6:52:41,  1.49s/it]
 32%|███▏      | 7747/24420 [3:13:09<6:52:22,  1.48s/it]
 32%|███▏      | 7748/24420 [3:13:10<6:51:53,  1.48s/it]
 32%|███▏      | 7749/24420 [3:13:12<6:51:45,  1.48s/it]
 32%|███▏      | 7750/24420 [3:13:13<6:52:03,  1.48s/it]
                                                        
{'loss': 0.1069, 'learning_rate': 5e-05, 'epoch': 1.59}

 32%|███▏      | 7750/24420 [3:13:13<6:52:03,  1.48s/it]
 32%|███▏      | 7751/24420 [3:13:15<6:52:52,  1.49s/it]
 32%|███▏      | 7752/24420 [3:13:16<6:53:09,  1.49s/it]
 32%|███▏      | 7753/24420 [3:13:18<6:52:30,  1.49s/it]
 32%|███▏      | 7754/24420 [3:13:19<6:52:21,  1.48s/it]
 32%|███▏      | 7755/24420 [3:13:21<6:51:40,  1.48s/it]
                                                        
{'loss': 0.1183, 'learning_rate': 5e-05, 'epoch': 1.59}

 32%|███▏      | 7755/24420 [3:13:21<6:51:40,  1.48s/it]
 32%|███▏      | 7756/24420 [3:13:22<6:51:26,  1.48s/it]
 32%|███▏      | 7757/24420 [3:13:24<6:53:09,  1.49s/it]
 32%|███▏      | 7758/24420 [3:13:25<6:52:22,  1.48s/it]
 32%|███▏      | 7759/24420 [3:13:27<6:51:50,  1.48s/it]
 32%|███▏      | 7760/24420 [3:13:28<6:51:07,  1.48s/it]
                                                        
{'loss': 0.1225, 'learning_rate': 5e-05, 'epoch': 1.59}

 32%|███▏      | 7760/24420 [3:13:28<6:51:07,  1.48s/it]
 32%|███▏      | 7761/24420 [3:13:30<6:51:34,  1.48s/it]
 32%|███▏      | 7762/24420 [3:13:31<6:52:02,  1.48s/it]
 32%|███▏      | 7763/24420 [3:13:33<6:52:44,  1.49s/it]
 32%|███▏      | 7764/24420 [3:13:34<6:52:36,  1.49s/it]
 32%|███▏      | 7765/24420 [3:13:36<6:52:26,  1.49s/it]
                                                        
{'loss': 0.098, 'learning_rate': 5e-05, 'epoch': 1.59}

 32%|███▏      | 7765/24420 [3:13:36<6:52:26,  1.49s/it]
 32%|███▏      | 7766/24420 [3:13:37<6:51:52,  1.48s/it]
 32%|███▏      | 7767/24420 [3:13:38<6:51:53,  1.48s/it]
 32%|███▏      | 7768/24420 [3:13:40<6:51:56,  1.48s/it]
 32%|███▏      | 7769/24420 [3:13:41<6:52:30,  1.49s/it]
 32%|███▏      | 7770/24420 [3:13:43<6:51:39,  1.48s/it]
                                                        
{'loss': 0.1054, 'learning_rate': 5e-05, 'epoch': 1.59}

 32%|███▏      | 7770/24420 [3:13:43<6:51:39,  1.48s/it]
 32%|███▏      | 7771/24420 [3:13:44<6:51:34,  1.48s/it]
 32%|███▏      | 7772/24420 [3:13:46<6:51:47,  1.48s/it]
 32%|███▏      | 7773/24420 [3:13:47<6:51:52,  1.48s/it]
 32%|███▏      | 7774/24420 [3:13:49<6:51:08,  1.48s/it]
 32%|███▏      | 7775/24420 [3:13:50<6:52:09,  1.49s/it]
                                                        
{'loss': 0.1227, 'learning_rate': 5e-05, 'epoch': 1.59}

 32%|███▏      | 7775/24420 [3:13:50<6:52:09,  1.49s/it]
 32%|███▏      | 7776/24420 [3:13:52<6:51:18,  1.48s/it]
 32%|███▏      | 7777/24420 [3:13:53<6:51:19,  1.48s/it]
 32%|███▏      | 7778/24420 [3:13:55<6:51:33,  1.48s/it]
 32%|███▏      | 7779/24420 [3:13:56<6:50:47,  1.48s/it]
 32%|███▏      | 7780/24420 [3:13:58<6:51:07,  1.48s/it]
                                                        
{'loss': 0.121, 'learning_rate': 5e-05, 'epoch': 1.59}

 32%|███▏      | 7780/24420 [3:13:58<6:51:07,  1.48s/it]
 32%|███▏      | 7781/24420 [3:13:59<6:51:34,  1.48s/it]
 32%|███▏      | 7782/24420 [3:14:01<6:51:14,  1.48s/it]
 32%|███▏      | 7783/24420 [3:14:02<6:51:43,  1.48s/it]
 32%|███▏      | 7784/24420 [3:14:04<6:51:49,  1.49s/it]
 32%|███▏      | 7785/24420 [3:14:05<6:51:22,  1.48s/it]
                                                        
{'loss': 0.1005, 'learning_rate': 5e-05, 'epoch': 1.59}

 32%|███▏      | 7785/24420 [3:14:05<6:51:22,  1.48s/it]
 32%|███▏      | 7786/24420 [3:14:07<6:52:29,  1.49s/it]
 32%|███▏      | 7787/24420 [3:14:08<6:51:38,  1.48s/it]
 32%|███▏      | 7788/24420 [3:14:10<6:51:17,  1.48s/it]
 32%|███▏      | 7789/24420 [3:14:11<6:51:04,  1.48s/it]
 32%|███▏      | 7790/24420 [3:14:13<6:50:36,  1.48s/it]
                                                        
{'loss': 0.1077, 'learning_rate': 5e-05, 'epoch': 1.6}

 32%|███▏      | 7790/24420 [3:14:13<6:50:36,  1.48s/it]
 32%|███▏      | 7791/24420 [3:14:14<6:50:18,  1.48s/it]
 32%|███▏      | 7792/24420 [3:14:16<6:49:46,  1.48s/it]
 32%|███▏      | 7793/24420 [3:14:17<6:49:45,  1.48s/it]
 32%|███▏      | 7794/24420 [3:14:19<6:50:56,  1.48s/it]
 32%|███▏      | 7795/24420 [3:14:20<6:50:28,  1.48s/it]
                                                        
{'loss': 0.1277, 'learning_rate': 5e-05, 'epoch': 1.6}

 32%|███▏      | 7795/24420 [3:14:20<6:50:28,  1.48s/it]
 32%|███▏      | 7796/24420 [3:14:21<6:50:18,  1.48s/it]
 32%|███▏      | 7797/24420 [3:14:23<6:50:34,  1.48s/it]
 32%|███▏      | 7798/24420 [3:14:24<6:50:49,  1.48s/it]
 32%|███▏      | 7799/24420 [3:14:26<6:50:36,  1.48s/it]
 32%|███▏      | 7800/24420 [3:14:27<6:50:16,  1.48s/it]
                                                        
{'loss': 0.1044, 'learning_rate': 5e-05, 'epoch': 1.6}

 32%|███▏      | 7800/24420 [3:14:27<6:50:16,  1.48s/it]
 32%|███▏      | 7801/24420 [3:14:32<11:05:37,  2.40s/it]
 32%|███▏      | 7802/24420 [3:14:33<9:50:07,  2.13s/it] 
 32%|███▏      | 7803/24420 [3:14:35<8:55:22,  1.93s/it]
 32%|███▏      | 7804/24420 [3:14:36<8:17:32,  1.80s/it]
 32%|███▏      | 7805/24420 [3:14:38<7:51:38,  1.70s/it]
                                                        
{'loss': 0.1172, 'learning_rate': 5e-05, 'epoch': 1.6}

 32%|███▏      | 7805/24420 [3:14:38<7:51:38,  1.70s/it]
 32%|███▏      | 7806/24420 [3:14:39<7:34:21,  1.64s/it]
 32%|███▏      | 7807/24420 [3:14:41<7:19:58,  1.59s/it]
 32%|███▏      | 7808/24420 [3:14:42<7:10:22,  1.55s/it]
 32%|███▏      | 7809/24420 [3:14:44<7:04:57,  1.53s/it]
 32%|███▏      | 7810/24420 [3:14:45<7:00:30,  1.52s/it]
                                                        
{'loss': 0.0963, 'learning_rate': 5e-05, 'epoch': 1.6}

 32%|███▏      | 7810/24420 [3:14:45<7:00:30,  1.52s/it]
 32%|███▏      | 7811/24420 [3:14:47<6:57:38,  1.51s/it]
 32%|███▏      | 7812/24420 [3:14:48<6:55:32,  1.50s/it]
 32%|███▏      | 7813/24420 [3:14:50<6:53:35,  1.49s/it]
 32%|███▏      | 7814/24420 [3:14:51<6:53:23,  1.49s/it]
 32%|███▏      | 7815/24420 [3:14:53<6:52:52,  1.49s/it]
                                                        
{'loss': 0.1056, 'learning_rate': 5e-05, 'epoch': 1.6}

 32%|███▏      | 7815/24420 [3:14:53<6:52:52,  1.49s/it]
 32%|███▏      | 7816/24420 [3:14:54<6:51:54,  1.49s/it]
 32%|███▏      | 7817/24420 [3:14:56<6:52:05,  1.49s/it]
 32%|███▏      | 7818/24420 [3:14:57<6:51:58,  1.49s/it]
 32%|███▏      | 7819/24420 [3:14:59<6:51:31,  1.49s/it]
 32%|███▏      | 7820/24420 [3:15:00<6:50:44,  1.48s/it]
                                                        
{'loss': 0.1727, 'learning_rate': 5e-05, 'epoch': 1.6}

 32%|███▏      | 7820/24420 [3:15:00<6:50:44,  1.48s/it]
 32%|███▏      | 7821/24420 [3:15:02<6:51:45,  1.49s/it]
 32%|███▏      | 7822/24420 [3:15:03<6:51:28,  1.49s/it]
 32%|███▏      | 7823/24420 [3:15:05<6:51:08,  1.49s/it]
 32%|███▏      | 7824/24420 [3:15:06<6:51:15,  1.49s/it]
 32%|███▏      | 7825/24420 [3:15:08<6:50:46,  1.49s/it]
                                                        
{'loss': 0.1124, 'learning_rate': 5e-05, 'epoch': 1.6}

 32%|███▏      | 7825/24420 [3:15:08<6:50:46,  1.49s/it]
 32%|███▏      | 7826/24420 [3:15:09<6:50:13,  1.48s/it]
 32%|███▏      | 7827/24420 [3:15:11<6:49:41,  1.48s/it]
 32%|███▏      | 7828/24420 [3:15:12<6:49:33,  1.48s/it]
 32%|███▏      | 7829/24420 [3:15:14<6:49:19,  1.48s/it]
 32%|███▏      | 7830/24420 [3:15:15<6:49:00,  1.48s/it]
                                                        
{'loss': 0.1333, 'learning_rate': 5e-05, 'epoch': 1.6}

 32%|███▏      | 7830/24420 [3:15:15<6:49:00,  1.48s/it]
 32%|███▏      | 7831/24420 [3:15:16<6:49:55,  1.48s/it]
 32%|███▏      | 7832/24420 [3:15:18<6:50:09,  1.48s/it]
 32%|███▏      | 7833/24420 [3:15:19<6:50:28,  1.48s/it]
 32%|███▏      | 7834/24420 [3:15:21<6:49:58,  1.48s/it]
 32%|███▏      | 7835/24420 [3:15:22<6:49:39,  1.48s/it]
                                                        
{'loss': 0.1177, 'learning_rate': 5e-05, 'epoch': 1.6}

 32%|███▏      | 7835/24420 [3:15:22<6:49:39,  1.48s/it]
 32%|███▏      | 7836/24420 [3:15:24<6:50:12,  1.48s/it]
 32%|███▏      | 7837/24420 [3:15:25<6:50:05,  1.48s/it]
 32%|███▏      | 7838/24420 [3:15:27<6:49:58,  1.48s/it]
 32%|███▏      | 7839/24420 [3:15:28<6:49:16,  1.48s/it]
 32%|███▏      | 7840/24420 [3:15:30<6:49:30,  1.48s/it]
                                                        
{'loss': 0.116, 'learning_rate': 5e-05, 'epoch': 1.61}

 32%|███▏      | 7840/24420 [3:15:30<6:49:30,  1.48s/it]
 32%|███▏      | 7841/24420 [3:15:31<6:49:48,  1.48s/it]
 32%|███▏      | 7842/24420 [3:15:33<6:49:23,  1.48s/it]
 32%|███▏      | 7843/24420 [3:15:34<6:48:56,  1.48s/it]
 32%|███▏      | 7844/24420 [3:15:36<6:48:53,  1.48s/it]
 32%|███▏      | 7845/24420 [3:15:37<6:48:43,  1.48s/it]
                                                        
{'loss': 0.1206, 'learning_rate': 5e-05, 'epoch': 1.61}

 32%|███▏      | 7845/24420 [3:15:37<6:48:43,  1.48s/it]
 32%|███▏      | 7846/24420 [3:15:39<6:49:11,  1.48s/it]
 32%|███▏      | 7847/24420 [3:15:40<6:49:24,  1.48s/it]
 32%|███▏      | 7848/24420 [3:15:42<6:48:52,  1.48s/it]
 32%|███▏      | 7849/24420 [3:15:43<6:48:20,  1.48s/it]
 32%|███▏      | 7850/24420 [3:15:45<6:48:22,  1.48s/it]
                                                        
{'loss': 0.0962, 'learning_rate': 5e-05, 'epoch': 1.61}

 32%|███▏      | 7850/24420 [3:15:45<6:48:22,  1.48s/it]
 32%|███▏      | 7851/24420 [3:15:46<6:48:59,  1.48s/it]
 32%|███▏      | 7852/24420 [3:15:48<6:48:52,  1.48s/it]
 32%|███▏      | 7853/24420 [3:15:49<6:49:04,  1.48s/it]
 32%|███▏      | 7854/24420 [3:15:51<6:48:48,  1.48s/it]
 32%|███▏      | 7855/24420 [3:15:52<6:49:09,  1.48s/it]
                                                        
{'loss': 0.1168, 'learning_rate': 5e-05, 'epoch': 1.61}

 32%|███▏      | 7855/24420 [3:15:52<6:49:09,  1.48s/it]
 32%|███▏      | 7856/24420 [3:15:54<6:49:13,  1.48s/it]
 32%|███▏      | 7857/24420 [3:15:55<6:48:36,  1.48s/it]
 32%|███▏      | 7858/24420 [3:15:56<6:48:20,  1.48s/it]
 32%|███▏      | 7859/24420 [3:15:58<6:48:37,  1.48s/it]
 32%|███▏      | 7860/24420 [3:15:59<6:48:47,  1.48s/it]
                                                        
{'loss': 0.0982, 'learning_rate': 5e-05, 'epoch': 1.61}

 32%|███▏      | 7860/24420 [3:15:59<6:48:47,  1.48s/it]
 32%|███▏      | 7861/24420 [3:16:01<6:48:53,  1.48s/it]
 32%|███▏      | 7862/24420 [3:16:02<6:48:48,  1.48s/it]
 32%|███▏      | 7863/24420 [3:16:04<6:48:58,  1.48s/it]
 32%|███▏      | 7864/24420 [3:16:05<6:48:30,  1.48s/it]
 32%|███▏      | 7865/24420 [3:16:07<6:48:14,  1.48s/it]
                                                        
{'loss': 0.1162, 'learning_rate': 5e-05, 'epoch': 1.61}

 32%|███▏      | 7865/24420 [3:16:07<6:48:14,  1.48s/it]
 32%|███▏      | 7866/24420 [3:16:08<6:48:06,  1.48s/it]
 32%|███▏      | 7867/24420 [3:16:10<6:48:54,  1.48s/it]
 32%|███▏      | 7868/24420 [3:16:11<6:48:13,  1.48s/it]
 32%|███▏      | 7869/24420 [3:16:13<6:48:56,  1.48s/it]
 32%|███▏      | 7870/24420 [3:16:14<6:48:55,  1.48s/it]
                                                        
{'loss': 0.1092, 'learning_rate': 5e-05, 'epoch': 1.61}

 32%|███▏      | 7870/24420 [3:16:14<6:48:55,  1.48s/it]
 32%|███▏      | 7871/24420 [3:16:16<6:48:57,  1.48s/it]
 32%|███▏      | 7872/24420 [3:16:17<6:48:52,  1.48s/it]
 32%|███▏      | 7873/24420 [3:16:19<6:48:27,  1.48s/it]
 32%|███▏      | 7874/24420 [3:16:20<6:49:15,  1.48s/it]
 32%|███▏      | 7875/24420 [3:16:22<6:49:55,  1.49s/it]
                                                        
{'loss': 0.1117, 'learning_rate': 5e-05, 'epoch': 1.61}

 32%|███▏      | 7875/24420 [3:16:22<6:49:55,  1.49s/it]
 32%|███▏      | 7876/24420 [3:16:23<6:49:52,  1.49s/it]
 32%|███▏      | 7877/24420 [3:16:25<6:50:32,  1.49s/it]
 32%|███▏      | 7878/24420 [3:16:26<6:50:00,  1.49s/it]
 32%|███▏      | 7879/24420 [3:16:28<6:49:26,  1.49s/it]
 32%|███▏      | 7880/24420 [3:16:29<6:49:21,  1.48s/it]
                                                        
{'loss': 0.0906, 'learning_rate': 5e-05, 'epoch': 1.61}

 32%|███▏      | 7880/24420 [3:16:29<6:49:21,  1.48s/it]
 32%|███▏      | 7881/24420 [3:16:31<6:49:36,  1.49s/it]
 32%|███▏      | 7882/24420 [3:16:32<6:49:29,  1.49s/it]
 32%|███▏      | 7883/24420 [3:16:34<6:49:28,  1.49s/it]
 32%|███▏      | 7884/24420 [3:16:35<6:49:13,  1.48s/it]
 32%|███▏      | 7885/24420 [3:16:37<6:49:15,  1.49s/it]
                                                        
{'loss': 0.1274, 'learning_rate': 5e-05, 'epoch': 1.61}

 32%|███▏      | 7885/24420 [3:16:37<6:49:15,  1.49s/it]
 32%|███▏      | 7886/24420 [3:16:38<6:49:38,  1.49s/it]
 32%|███▏      | 7887/24420 [3:16:40<6:49:55,  1.49s/it]
 32%|███▏      | 7888/24420 [3:16:41<6:49:39,  1.49s/it]
 32%|███▏      | 7889/24420 [3:16:42<6:49:16,  1.49s/it]
 32%|███▏      | 7890/24420 [3:16:44<6:48:48,  1.48s/it]
                                                        
{'loss': 0.1053, 'learning_rate': 5e-05, 'epoch': 1.62}

 32%|███▏      | 7890/24420 [3:16:44<6:48:48,  1.48s/it]
 32%|███▏      | 7891/24420 [3:16:45<6:48:51,  1.48s/it]
 32%|███▏      | 7892/24420 [3:16:47<6:48:45,  1.48s/it]
 32%|███▏      | 7893/24420 [3:16:48<6:48:47,  1.48s/it]
 32%|███▏      | 7894/24420 [3:16:50<6:49:33,  1.49s/it]
 32%|███▏      | 7895/24420 [3:16:51<6:49:49,  1.49s/it]
                                                        
{'loss': 0.1342, 'learning_rate': 5e-05, 'epoch': 1.62}

 32%|███▏      | 7895/24420 [3:16:51<6:49:49,  1.49s/it]
 32%|███▏      | 7896/24420 [3:16:53<6:49:52,  1.49s/it]
 32%|███▏      | 7897/24420 [3:16:54<6:49:17,  1.49s/it]
 32%|███▏      | 7898/24420 [3:16:56<6:48:49,  1.48s/it]
 32%|███▏      | 7899/24420 [3:16:57<6:48:38,  1.48s/it]
 32%|███▏      | 7900/24420 [3:16:59<6:48:21,  1.48s/it]
                                                        
{'loss': 0.1166, 'learning_rate': 5e-05, 'epoch': 1.62}

 32%|███▏      | 7900/24420 [3:16:59<6:48:21,  1.48s/it]
 32%|███▏      | 7901/24420 [3:17:00<6:49:11,  1.49s/it]
 32%|███▏      | 7902/24420 [3:17:02<6:48:12,  1.48s/it]
 32%|███▏      | 7903/24420 [3:17:03<6:48:03,  1.48s/it]
 32%|███▏      | 7904/24420 [3:17:05<6:48:02,  1.48s/it]
 32%|███▏      | 7905/24420 [3:17:06<6:48:02,  1.48s/it]
                                                        
{'loss': 0.127, 'learning_rate': 5e-05, 'epoch': 1.62}

 32%|███▏      | 7905/24420 [3:17:06<6:48:02,  1.48s/it]
 32%|███▏      | 7906/24420 [3:17:08<6:48:59,  1.49s/it]
 32%|███▏      | 7907/24420 [3:17:09<6:49:28,  1.49s/it]
 32%|███▏      | 7908/24420 [3:17:11<6:49:03,  1.49s/it]
 32%|███▏      | 7909/24420 [3:17:12<6:48:05,  1.48s/it]
 32%|███▏      | 7910/24420 [3:17:14<6:48:06,  1.48s/it]
                                                        
{'loss': 0.1077, 'learning_rate': 5e-05, 'epoch': 1.62}

 32%|███▏      | 7910/24420 [3:17:14<6:48:06,  1.48s/it]
 32%|███▏      | 7911/24420 [3:17:15<6:47:23,  1.48s/it]
 32%|███▏      | 7912/24420 [3:17:17<6:46:51,  1.48s/it]
 32%|███▏      | 7913/24420 [3:17:18<6:47:54,  1.48s/it]
 32%|███▏      | 7914/24420 [3:17:20<6:48:01,  1.48s/it]
 32%|███▏      | 7915/24420 [3:17:21<6:47:37,  1.48s/it]
                                                        
{'loss': 0.1178, 'learning_rate': 5e-05, 'epoch': 1.62}

 32%|███▏      | 7915/24420 [3:17:21<6:47:37,  1.48s/it]
 32%|███▏      | 7916/24420 [3:17:23<6:47:58,  1.48s/it]
 32%|███▏      | 7917/24420 [3:17:24<6:48:23,  1.48s/it]
 32%|███▏      | 7918/24420 [3:17:26<6:47:54,  1.48s/it]
 32%|███▏      | 7919/24420 [3:17:27<6:48:18,  1.48s/it]
 32%|███▏      | 7920/24420 [3:17:28<6:47:37,  1.48s/it]
                                                        
{'loss': 0.1105, 'learning_rate': 5e-05, 'epoch': 1.62}

 32%|███▏      | 7920/24420 [3:17:28<6:47:37,  1.48s/it]
 32%|███▏      | 7921/24420 [3:17:31<7:35:09,  1.66s/it]
 32%|███▏      | 7922/24420 [3:17:32<7:19:47,  1.60s/it]
 32%|███▏      | 7923/24420 [3:17:33<7:09:30,  1.56s/it]
 32%|███▏      | 7924/24420 [3:17:35<7:03:40,  1.54s/it]
 32%|███▏      | 7925/24420 [3:17:36<6:58:57,  1.52s/it]
                                                        
{'loss': 0.1101, 'learning_rate': 5e-05, 'epoch': 1.62}

 32%|███▏      | 7925/24420 [3:17:36<6:58:57,  1.52s/it]
 32%|███▏      | 7926/24420 [3:17:38<6:55:44,  1.51s/it]
 32%|███▏      | 7927/24420 [3:17:39<6:53:00,  1.50s/it]
 32%|███▏      | 7928/24420 [3:17:41<6:51:20,  1.50s/it]
 32%|███▏      | 7929/24420 [3:17:42<6:50:24,  1.49s/it]
 32%|███▏      | 7930/24420 [3:17:44<6:49:18,  1.49s/it]
                                                        
{'loss': 0.1282, 'learning_rate': 5e-05, 'epoch': 1.62}

 32%|███▏      | 7930/24420 [3:17:44<6:49:18,  1.49s/it]
 32%|███▏      | 7931/24420 [3:17:45<6:48:39,  1.49s/it]
 32%|███▏      | 7932/24420 [3:17:47<6:48:19,  1.49s/it]
 32%|███▏      | 7933/24420 [3:17:48<6:47:52,  1.48s/it]
 32%|███▏      | 7934/24420 [3:17:50<6:47:52,  1.48s/it]
 32%|███▏      | 7935/24420 [3:17:51<6:47:37,  1.48s/it]
                                                        
{'loss': 0.1057, 'learning_rate': 5e-05, 'epoch': 1.62}

 32%|███▏      | 7935/24420 [3:17:51<6:47:37,  1.48s/it]
 32%|███▏      | 7936/24420 [3:17:53<6:47:15,  1.48s/it]
 33%|███▎      | 7937/24420 [3:17:54<6:47:37,  1.48s/it]
 33%|███▎      | 7938/24420 [3:17:56<6:48:49,  1.49s/it]
 33%|███▎      | 7939/24420 [3:17:57<6:48:25,  1.49s/it]
 33%|███▎      | 7940/24420 [3:17:59<6:49:42,  1.49s/it]
                                                        
{'loss': 0.1205, 'learning_rate': 5e-05, 'epoch': 1.63}

 33%|███▎      | 7940/24420 [3:17:59<6:49:42,  1.49s/it]
 33%|███▎      | 7941/24420 [3:18:00<6:49:00,  1.49s/it]
 33%|███▎      | 7942/24420 [3:18:02<6:48:53,  1.49s/it]
 33%|███▎      | 7943/24420 [3:18:03<6:48:09,  1.49s/it]
 33%|███▎      | 7944/24420 [3:18:05<6:47:54,  1.49s/it]
 33%|███▎      | 7945/24420 [3:18:06<6:47:45,  1.49s/it]
                                                        
{'loss': 0.1006, 'learning_rate': 5e-05, 'epoch': 1.63}

 33%|███▎      | 7945/24420 [3:18:06<6:47:45,  1.49s/it]
 33%|███▎      | 7946/24420 [3:18:08<6:47:20,  1.48s/it]
 33%|███▎      | 7947/24420 [3:18:09<6:47:08,  1.48s/it]
 33%|███▎      | 7948/24420 [3:18:11<6:46:37,  1.48s/it]
 33%|███▎      | 7949/24420 [3:18:12<6:45:57,  1.48s/it]
 33%|███▎      | 7950/24420 [3:18:14<6:46:10,  1.48s/it]
                                                        
{'loss': 0.0991, 'learning_rate': 5e-05, 'epoch': 1.63}

 33%|███▎      | 7950/24420 [3:18:14<6:46:10,  1.48s/it]
 33%|███▎      | 7951/24420 [3:18:15<6:46:43,  1.48s/it]
 33%|███▎      | 7952/24420 [3:18:17<6:46:14,  1.48s/it]
 33%|███▎      | 7953/24420 [3:18:18<6:46:06,  1.48s/it]
 33%|███▎      | 7954/24420 [3:18:19<6:45:40,  1.48s/it]
 33%|███▎      | 7955/24420 [3:18:21<6:45:38,  1.48s/it]
                                                        
{'loss': 0.1111, 'learning_rate': 5e-05, 'epoch': 1.63}

 33%|███▎      | 7955/24420 [3:18:21<6:45:38,  1.48s/it]
 33%|███▎      | 7956/24420 [3:18:22<6:46:34,  1.48s/it]
 33%|███▎      | 7957/24420 [3:18:24<6:46:55,  1.48s/it]
 33%|███▎      | 7958/24420 [3:18:25<6:48:16,  1.49s/it]
 33%|███▎      | 7959/24420 [3:18:27<6:47:27,  1.49s/it]
 33%|███▎      | 7960/24420 [3:18:28<6:48:19,  1.49s/it]
                                                        
{'loss': 0.1045, 'learning_rate': 5e-05, 'epoch': 1.63}

 33%|███▎      | 7960/24420 [3:18:28<6:48:19,  1.49s/it]
 33%|███▎      | 7961/24420 [3:18:30<6:48:01,  1.49s/it]
 33%|███▎      | 7962/24420 [3:18:31<6:47:29,  1.49s/it]
 33%|███▎      | 7963/24420 [3:18:33<6:47:07,  1.48s/it]
 33%|███▎      | 7964/24420 [3:18:34<6:47:35,  1.49s/it]
 33%|███▎      | 7965/24420 [3:18:36<6:47:07,  1.48s/it]
                                                        
{'loss': 0.117, 'learning_rate': 5e-05, 'epoch': 1.63}

 33%|███▎      | 7965/24420 [3:18:36<6:47:07,  1.48s/it]
 33%|███▎      | 7966/24420 [3:18:37<6:47:01,  1.48s/it]
 33%|███▎      | 7967/24420 [3:18:39<6:46:53,  1.48s/it]
 33%|███▎      | 7968/24420 [3:18:40<6:46:18,  1.48s/it]
 33%|███▎      | 7969/24420 [3:18:42<6:46:40,  1.48s/it]
 33%|███▎      | 7970/24420 [3:18:43<6:47:07,  1.48s/it]
                                                        
{'loss': 0.1045, 'learning_rate': 5e-05, 'epoch': 1.63}

 33%|███▎      | 7970/24420 [3:18:43<6:47:07,  1.48s/it]
 33%|███▎      | 7971/24420 [3:18:45<6:47:12,  1.49s/it]
 33%|███▎      | 7972/24420 [3:18:46<6:46:43,  1.48s/it]
 33%|███▎      | 7973/24420 [3:18:48<6:46:16,  1.48s/it]
 33%|███▎      | 7974/24420 [3:18:49<6:46:29,  1.48s/it]
 33%|███▎      | 7975/24420 [3:18:51<6:45:59,  1.48s/it]
                                                        
{'loss': 0.1215, 'learning_rate': 5e-05, 'epoch': 1.63}

 33%|███▎      | 7975/24420 [3:18:51<6:45:59,  1.48s/it]
 33%|███▎      | 7976/24420 [3:18:52<6:46:00,  1.48s/it]
 33%|███▎      | 7977/24420 [3:18:54<6:46:23,  1.48s/it]
 33%|███▎      | 7978/24420 [3:18:55<6:46:15,  1.48s/it]
 33%|███▎      | 7979/24420 [3:18:57<6:46:06,  1.48s/it]
 33%|███▎      | 7980/24420 [3:18:58<6:46:07,  1.48s/it]
                                                        
{'loss': 0.0901, 'learning_rate': 5e-05, 'epoch': 1.63}

 33%|███▎      | 7980/24420 [3:18:58<6:46:07,  1.48s/it]
 33%|███▎      | 7981/24420 [3:19:00<6:46:18,  1.48s/it]
 33%|███▎      | 7982/24420 [3:19:01<6:46:32,  1.48s/it]
 33%|███▎      | 7983/24420 [3:19:03<6:45:54,  1.48s/it]
 33%|███▎      | 7984/24420 [3:19:04<6:46:11,  1.48s/it]
 33%|███▎      | 7985/24420 [3:19:05<6:46:57,  1.49s/it]
                                                        
{'loss': 0.1413, 'learning_rate': 5e-05, 'epoch': 1.63}

 33%|███▎      | 7985/24420 [3:19:05<6:46:57,  1.49s/it]
 33%|███▎      | 7986/24420 [3:19:07<6:47:37,  1.49s/it]
 33%|███▎      | 7987/24420 [3:19:08<6:47:00,  1.49s/it]
 33%|███▎      | 7988/24420 [3:19:10<6:47:36,  1.49s/it]
 33%|███▎      | 7989/24420 [3:19:11<6:46:42,  1.49s/it]
 33%|███▎      | 7990/24420 [3:19:13<6:47:14,  1.49s/it]
                                                        
{'loss': 0.0975, 'learning_rate': 5e-05, 'epoch': 1.64}

 33%|███▎      | 7990/24420 [3:19:13<6:47:14,  1.49s/it]
 33%|███▎      | 7991/24420 [3:19:14<6:46:41,  1.49s/it]
 33%|███▎      | 7992/24420 [3:19:16<6:46:38,  1.49s/it]
 33%|███▎      | 7993/24420 [3:19:17<6:45:56,  1.48s/it]
 33%|███▎      | 7994/24420 [3:19:19<6:45:49,  1.48s/it]
 33%|███▎      | 7995/24420 [3:19:20<6:45:59,  1.48s/it]
                                                        
{'loss': 0.1224, 'learning_rate': 5e-05, 'epoch': 1.64}

 33%|███▎      | 7995/24420 [3:19:20<6:45:59,  1.48s/it]
 33%|███▎      | 7996/24420 [3:19:22<6:45:39,  1.48s/it]
 33%|███▎      | 7997/24420 [3:19:23<6:45:12,  1.48s/it]
 33%|███▎      | 7998/24420 [3:19:25<6:45:05,  1.48s/it]
 33%|███▎      | 7999/24420 [3:19:26<6:45:43,  1.48s/it]
 33%|███▎      | 8000/24420 [3:19:28<6:46:06,  1.48s/it]
                                                        
{'loss': 0.1152, 'learning_rate': 5e-05, 'epoch': 1.64}

 33%|███▎      | 8000/24420 [3:19:28<6:46:06,  1.48s/it]
 33%|███▎      | 8001/24420 [3:19:32<11:07:51,  2.44s/it]
 33%|███▎      | 8002/24420 [3:19:34<9:48:38,  2.15s/it] 
 33%|███▎      | 8003/24420 [3:19:35<8:54:09,  1.95s/it]
 33%|███▎      | 8004/24420 [3:19:37<8:15:51,  1.81s/it]
 33%|███▎      | 8005/24420 [3:19:38<7:49:10,  1.71s/it]
                                                        
{'loss': 0.1131, 'learning_rate': 5e-05, 'epoch': 1.64}

 33%|███▎      | 8005/24420 [3:19:38<7:49:10,  1.71s/it]
 33%|███▎      | 8006/24420 [3:19:40<7:29:46,  1.64s/it]
 33%|███▎      | 8007/24420 [3:19:41<7:16:40,  1.60s/it]
 33%|███▎      | 8008/24420 [3:19:43<7:07:50,  1.56s/it]
 33%|███▎      | 8009/24420 [3:19:44<7:02:04,  1.54s/it]
 33%|███▎      | 8010/24420 [3:19:46<6:58:37,  1.53s/it]
                                                        
{'loss': 0.1201, 'learning_rate': 5e-05, 'epoch': 1.64}

 33%|███▎      | 8010/24420 [3:19:46<6:58:37,  1.53s/it]
 33%|███▎      | 8011/24420 [3:19:47<6:54:23,  1.52s/it]
 33%|███▎      | 8012/24420 [3:19:49<6:51:32,  1.50s/it]
 33%|███▎      | 8013/24420 [3:19:50<6:49:23,  1.50s/it]
 33%|███▎      | 8014/24420 [3:19:52<6:48:29,  1.49s/it]
 33%|███▎      | 8015/24420 [3:19:53<6:47:45,  1.49s/it]
                                                        
{'loss': 0.089, 'learning_rate': 5e-05, 'epoch': 1.64}

 33%|███▎      | 8015/24420 [3:19:53<6:47:45,  1.49s/it]
 33%|███▎      | 8016/24420 [3:19:55<6:46:37,  1.49s/it]
 33%|███▎      | 8017/24420 [3:19:56<6:46:02,  1.49s/it]
 33%|███▎      | 8018/24420 [3:19:58<6:45:48,  1.48s/it]
 33%|███▎      | 8019/24420 [3:19:59<6:47:00,  1.49s/it]
 33%|███▎      | 8020/24420 [3:20:01<6:46:02,  1.49s/it]
                                                        
{'loss': 0.0921, 'learning_rate': 5e-05, 'epoch': 1.64}

 33%|███▎      | 8020/24420 [3:20:01<6:46:02,  1.49s/it]
 33%|███▎      | 8021/24420 [3:20:02<6:45:52,  1.48s/it]
 33%|███▎      | 8022/24420 [3:20:04<6:45:37,  1.48s/it]
 33%|███▎      | 8023/24420 [3:20:05<6:45:37,  1.48s/it]
 33%|███▎      | 8024/24420 [3:20:07<6:44:52,  1.48s/it]
 33%|███▎      | 8025/24420 [3:20:08<6:44:27,  1.48s/it]
                                                        
{'loss': 0.1224, 'learning_rate': 5e-05, 'epoch': 1.64}

 33%|███▎      | 8025/24420 [3:20:08<6:44:27,  1.48s/it]
 33%|███▎      | 8026/24420 [3:20:10<6:44:37,  1.48s/it]
 33%|███▎      | 8027/24420 [3:20:11<6:44:12,  1.48s/it]
 33%|███▎      | 8028/24420 [3:20:12<6:43:59,  1.48s/it]
 33%|███▎      | 8029/24420 [3:20:14<6:43:51,  1.48s/it]
 33%|███▎      | 8030/24420 [3:20:15<6:43:46,  1.48s/it]
                                                        
{'loss': 0.1055, 'learning_rate': 5e-05, 'epoch': 1.64}

 33%|███▎      | 8030/24420 [3:20:15<6:43:46,  1.48s/it]
 33%|███▎      | 8031/24420 [3:20:17<6:43:01,  1.48s/it]
 33%|███▎      | 8032/24420 [3:20:18<6:43:15,  1.48s/it]
 33%|███▎      | 8033/24420 [3:20:20<6:43:34,  1.48s/it]
 33%|███▎      | 8034/24420 [3:20:21<6:43:44,  1.48s/it]
 33%|███▎      | 8035/24420 [3:20:23<6:44:01,  1.48s/it]
                                                        
{'loss': 0.108, 'learning_rate': 5e-05, 'epoch': 1.65}

 33%|███▎      | 8035/24420 [3:20:23<6:44:01,  1.48s/it]
 33%|███▎      | 8036/24420 [3:20:24<6:44:29,  1.48s/it]
 33%|███▎      | 8037/24420 [3:20:26<6:43:53,  1.48s/it]
 33%|███▎      | 8038/24420 [3:20:27<6:44:05,  1.48s/it]
 33%|███▎      | 8039/24420 [3:20:29<6:43:45,  1.48s/it]
 33%|███▎      | 8040/24420 [3:20:30<6:44:14,  1.48s/it]
                                                        
{'loss': 0.1177, 'learning_rate': 5e-05, 'epoch': 1.65}

 33%|███▎      | 8040/24420 [3:20:30<6:44:14,  1.48s/it]
 33%|███▎      | 8041/24420 [3:20:32<6:44:08,  1.48s/it]
 33%|███▎      | 8042/24420 [3:20:33<6:43:37,  1.48s/it]
 33%|███▎      | 8043/24420 [3:20:35<6:43:56,  1.48s/it]
 33%|███▎      | 8044/24420 [3:20:36<6:44:21,  1.48s/it]
 33%|███▎      | 8045/24420 [3:20:38<6:44:33,  1.48s/it]
                                                        
{'loss': 0.1097, 'learning_rate': 5e-05, 'epoch': 1.65}

 33%|███▎      | 8045/24420 [3:20:38<6:44:33,  1.48s/it]
 33%|███▎      | 8046/24420 [3:20:39<6:44:58,  1.48s/it]
 33%|███▎      | 8047/24420 [3:20:41<6:44:43,  1.48s/it]
 33%|███▎      | 8048/24420 [3:20:42<6:44:02,  1.48s/it]
 33%|███▎      | 8049/24420 [3:20:44<6:43:58,  1.48s/it]
 33%|███▎      | 8050/24420 [3:20:45<6:43:56,  1.48s/it]
                                                        
{'loss': 0.1184, 'learning_rate': 5e-05, 'epoch': 1.65}

 33%|███▎      | 8050/24420 [3:20:45<6:43:56,  1.48s/it]
 33%|███▎      | 8051/24420 [3:20:47<6:44:49,  1.48s/it]
 33%|███▎      | 8052/24420 [3:20:48<6:44:29,  1.48s/it]
 33%|███▎      | 8053/24420 [3:20:49<6:44:23,  1.48s/it]
 33%|███▎      | 8054/24420 [3:20:51<6:43:44,  1.48s/it]
 33%|███▎      | 8055/24420 [3:20:52<6:43:38,  1.48s/it]
                                                        
{'loss': 0.1247, 'learning_rate': 5e-05, 'epoch': 1.65}

 33%|███▎      | 8055/24420 [3:20:52<6:43:38,  1.48s/it]
 33%|███▎      | 8056/24420 [3:20:54<6:43:19,  1.48s/it]
 33%|███▎      | 8057/24420 [3:20:55<6:43:05,  1.48s/it]
 33%|███▎      | 8058/24420 [3:20:57<6:42:42,  1.48s/it]
 33%|███▎      | 8059/24420 [3:20:58<6:43:06,  1.48s/it]
 33%|███▎      | 8060/24420 [3:21:00<6:43:51,  1.48s/it]
                                                        
{'loss': 0.1139, 'learning_rate': 5e-05, 'epoch': 1.65}

 33%|███▎      | 8060/24420 [3:21:00<6:43:51,  1.48s/it]
 33%|███▎      | 8061/24420 [3:21:01<6:44:08,  1.48s/it]
 33%|███▎      | 8062/24420 [3:21:03<6:44:47,  1.48s/it]
 33%|███▎      | 8063/24420 [3:21:04<6:44:09,  1.48s/it]
 33%|███▎      | 8064/24420 [3:21:06<6:44:14,  1.48s/it]
 33%|███▎      | 8065/24420 [3:21:07<6:43:21,  1.48s/it]
                                                        
{'loss': 0.1054, 'learning_rate': 5e-05, 'epoch': 1.65}

 33%|███▎      | 8065/24420 [3:21:07<6:43:21,  1.48s/it]
 33%|███▎      | 8066/24420 [3:21:09<6:43:26,  1.48s/it]
 33%|███▎      | 8067/24420 [3:21:10<6:45:04,  1.49s/it]
 33%|███▎      | 8068/24420 [3:21:12<6:46:03,  1.49s/it]
 33%|███▎      | 8069/24420 [3:21:13<6:45:50,  1.49s/it]
 33%|███▎      | 8070/24420 [3:21:15<6:45:50,  1.49s/it]
                                                        
{'loss': 0.108, 'learning_rate': 5e-05, 'epoch': 1.65}

 33%|███▎      | 8070/24420 [3:21:15<6:45:50,  1.49s/it]
 33%|███▎      | 8071/24420 [3:21:16<6:45:19,  1.49s/it]
 33%|███▎      | 8072/24420 [3:21:18<6:45:01,  1.49s/it]
 33%|███▎      | 8073/24420 [3:21:19<6:44:35,  1.49s/it]
 33%|███▎      | 8074/24420 [3:21:21<6:44:12,  1.48s/it]
 33%|███▎      | 8075/24420 [3:21:22<6:44:27,  1.48s/it]
                                                        
{'loss': 0.1122, 'learning_rate': 5e-05, 'epoch': 1.65}

 33%|███▎      | 8075/24420 [3:21:22<6:44:27,  1.48s/it]
 33%|███▎      | 8076/24420 [3:21:24<6:44:29,  1.48s/it]
 33%|███▎      | 8077/24420 [3:21:25<6:44:51,  1.49s/it]
 33%|███▎      | 8078/24420 [3:21:27<6:44:50,  1.49s/it]
 33%|███▎      | 8079/24420 [3:21:28<6:43:30,  1.48s/it]
 33%|███▎      | 8080/24420 [3:21:30<6:44:16,  1.48s/it]
                                                        
{'loss': 0.1035, 'learning_rate': 5e-05, 'epoch': 1.65}

 33%|███▎      | 8080/24420 [3:21:30<6:44:16,  1.48s/it]
 33%|███▎      | 8081/24420 [3:21:31<6:43:44,  1.48s/it]
 33%|███▎      | 8082/24420 [3:21:32<6:43:19,  1.48s/it]
 33%|███▎      | 8083/24420 [3:21:34<6:42:55,  1.48s/it]
 33%|███▎      | 8084/24420 [3:21:35<6:43:25,  1.48s/it]
 33%|███▎      | 8085/24420 [3:21:37<6:44:14,  1.48s/it]
                                                        
{'loss': 0.1081, 'learning_rate': 5e-05, 'epoch': 1.66}

 33%|███▎      | 8085/24420 [3:21:37<6:44:14,  1.48s/it]
 33%|███▎      | 8086/24420 [3:21:38<6:44:32,  1.49s/it]
 33%|███▎      | 8087/24420 [3:21:40<6:43:47,  1.48s/it]
 33%|███▎      | 8088/24420 [3:21:41<6:44:21,  1.49s/it]
 33%|███▎      | 8089/24420 [3:21:43<6:44:13,  1.49s/it]
 33%|███▎      | 8090/24420 [3:21:44<6:44:03,  1.48s/it]
                                                        
{'loss': 0.1159, 'learning_rate': 5e-05, 'epoch': 1.66}

 33%|███▎      | 8090/24420 [3:21:44<6:44:03,  1.48s/it]
 33%|███▎      | 8091/24420 [3:21:46<6:44:11,  1.49s/it]
 33%|███▎      | 8092/24420 [3:21:47<6:44:42,  1.49s/it]
 33%|███▎      | 8093/24420 [3:21:49<6:44:54,  1.49s/it]
 33%|███▎      | 8094/24420 [3:21:50<6:45:05,  1.49s/it]
 33%|███▎      | 8095/24420 [3:21:52<6:45:08,  1.49s/it]
                                                        
{'loss': 0.1147, 'learning_rate': 5e-05, 'epoch': 1.66}

 33%|███▎      | 8095/24420 [3:21:52<6:45:08,  1.49s/it]
 33%|███▎      | 8096/24420 [3:21:53<6:44:35,  1.49s/it]
 33%|███▎      | 8097/24420 [3:21:55<6:44:09,  1.49s/it]
 33%|███▎      | 8098/24420 [3:21:56<6:44:42,  1.49s/it]
 33%|███▎      | 8099/24420 [3:21:58<6:44:20,  1.49s/it]
 33%|███▎      | 8100/24420 [3:21:59<6:43:56,  1.49s/it]
                                                        
{'loss': 0.1211, 'learning_rate': 5e-05, 'epoch': 1.66}

 33%|███▎      | 8100/24420 [3:21:59<6:43:56,  1.49s/it]
 33%|███▎      | 8101/24420 [3:22:01<6:44:13,  1.49s/it]
 33%|███▎      | 8102/24420 [3:22:02<6:44:11,  1.49s/it]
 33%|███▎      | 8103/24420 [3:22:04<6:44:18,  1.49s/it]
 33%|███▎      | 8104/24420 [3:22:05<6:44:36,  1.49s/it]
 33%|███▎      | 8105/24420 [3:22:07<6:43:51,  1.49s/it]
                                                        
{'loss': 0.0965, 'learning_rate': 5e-05, 'epoch': 1.66}

 33%|███▎      | 8105/24420 [3:22:07<6:43:51,  1.49s/it]
 33%|███▎      | 8106/24420 [3:22:08<6:44:13,  1.49s/it]
 33%|███▎      | 8107/24420 [3:22:10<6:44:06,  1.49s/it]
 33%|███▎      | 8108/24420 [3:22:11<6:45:42,  1.49s/it]
 33%|███▎      | 8109/24420 [3:22:13<6:45:22,  1.49s/it]
 33%|███▎      | 8110/24420 [3:22:14<6:44:28,  1.49s/it]
                                                        
{'loss': 0.1194, 'learning_rate': 5e-05, 'epoch': 1.66}

 33%|███▎      | 8110/24420 [3:22:14<6:44:28,  1.49s/it]
 33%|███▎      | 8111/24420 [3:22:16<6:45:41,  1.49s/it]
 33%|███▎      | 8112/24420 [3:22:17<6:44:42,  1.49s/it]
 33%|███▎      | 8113/24420 [3:22:19<6:44:01,  1.49s/it]
 33%|███▎      | 8114/24420 [3:22:20<6:44:23,  1.49s/it]
 33%|███▎      | 8115/24420 [3:22:22<6:44:09,  1.49s/it]
                                                        
{'loss': 0.0978, 'learning_rate': 5e-05, 'epoch': 1.66}

 33%|███▎      | 8115/24420 [3:22:22<6:44:09,  1.49s/it]
 33%|███▎      | 8116/24420 [3:22:23<6:44:58,  1.49s/it]
 33%|███▎      | 8117/24420 [3:22:25<6:45:04,  1.49s/it]
 33%|███▎      | 8118/24420 [3:22:26<6:44:30,  1.49s/it]
 33%|███▎      | 8119/24420 [3:22:28<6:44:09,  1.49s/it]
 33%|███▎      | 8120/24420 [3:22:29<6:44:17,  1.49s/it]
                                                        
{'loss': 0.1168, 'learning_rate': 5e-05, 'epoch': 1.66}

 33%|███▎      | 8120/24420 [3:22:29<6:44:17,  1.49s/it]
 33%|███▎      | 8121/24420 [3:22:31<6:43:53,  1.49s/it]
 33%|███▎      | 8122/24420 [3:22:32<6:44:13,  1.49s/it]
 33%|███▎      | 8123/24420 [3:22:33<6:43:28,  1.49s/it]
 33%|███▎      | 8124/24420 [3:22:35<6:44:46,  1.49s/it]
 33%|███▎      | 8125/24420 [3:22:36<6:44:18,  1.49s/it]
                                                        
{'loss': 0.0996, 'learning_rate': 5e-05, 'epoch': 1.66}

 33%|███▎      | 8125/24420 [3:22:36<6:44:18,  1.49s/it]
 33%|███▎      | 8126/24420 [3:22:38<6:44:39,  1.49s/it]
 33%|███▎      | 8127/24420 [3:22:39<6:44:19,  1.49s/it]
 33%|███▎      | 8128/24420 [3:22:41<6:44:20,  1.49s/it]
 33%|███▎      | 8129/24420 [3:22:42<6:44:29,  1.49s/it]
 33%|███▎      | 8130/24420 [3:22:44<6:44:00,  1.49s/it]
                                                        
{'loss': 0.1112, 'learning_rate': 5e-05, 'epoch': 1.66}

 33%|███▎      | 8130/24420 [3:22:44<6:44:00,  1.49s/it]
 33%|███▎      | 8131/24420 [3:22:45<6:43:32,  1.49s/it]
 33%|███▎      | 8132/24420 [3:22:47<6:43:35,  1.49s/it]
 33%|███▎      | 8133/24420 [3:22:48<6:43:32,  1.49s/it]
 33%|███▎      | 8134/24420 [3:22:50<6:42:46,  1.48s/it]
 33%|███▎      | 8135/24420 [3:22:51<6:42:19,  1.48s/it]
                                                        
{'loss': 0.1131, 'learning_rate': 5e-05, 'epoch': 1.67}

 33%|███▎      | 8135/24420 [3:22:51<6:42:19,  1.48s/it]
 33%|███▎      | 8136/24420 [3:22:53<6:42:14,  1.48s/it]
 33%|███▎      | 8137/24420 [3:22:54<6:43:21,  1.49s/it]
 33%|███▎      | 8138/24420 [3:22:56<6:43:17,  1.49s/it]
 33%|███▎      | 8139/24420 [3:22:57<6:42:50,  1.48s/it]
 33%|███▎      | 8140/24420 [3:22:59<6:43:18,  1.49s/it]
                                                        
{'loss': 0.107, 'learning_rate': 5e-05, 'epoch': 1.67}

 33%|███▎      | 8140/24420 [3:22:59<6:43:18,  1.49s/it]
 33%|███▎      | 8141/24420 [3:23:00<6:43:58,  1.49s/it]
 33%|███▎      | 8142/24420 [3:23:02<6:43:12,  1.49s/it]
 33%|███▎      | 8143/24420 [3:23:03<6:43:14,  1.49s/it]
 33%|███▎      | 8144/24420 [3:23:05<6:43:32,  1.49s/it]
 33%|███▎      | 8145/24420 [3:23:06<6:44:02,  1.49s/it]
                                                        
{'loss': 0.0951, 'learning_rate': 5e-05, 'epoch': 1.67}

 33%|███▎      | 8145/24420 [3:23:06<6:44:02,  1.49s/it]
 33%|███▎      | 8146/24420 [3:23:08<6:44:19,  1.49s/it]
 33%|███▎      | 8147/24420 [3:23:09<6:43:54,  1.49s/it]
 33%|███▎      | 8148/24420 [3:23:11<6:43:16,  1.49s/it]
 33%|███▎      | 8149/24420 [3:23:12<6:42:44,  1.49s/it]
 33%|███▎      | 8150/24420 [3:23:14<6:42:03,  1.48s/it]
                                                        
{'loss': 0.1335, 'learning_rate': 5e-05, 'epoch': 1.67}

 33%|███▎      | 8150/24420 [3:23:14<6:42:03,  1.48s/it]
 33%|███▎      | 8151/24420 [3:23:15<6:41:46,  1.48s/it]
 33%|███▎      | 8152/24420 [3:23:17<6:41:43,  1.48s/it]
 33%|███▎      | 8153/24420 [3:23:18<6:41:48,  1.48s/it]
 33%|███▎      | 8154/24420 [3:23:20<6:41:54,  1.48s/it]
 33%|███▎      | 8155/24420 [3:23:21<6:42:11,  1.48s/it]
                                                        
{'loss': 0.1073, 'learning_rate': 5e-05, 'epoch': 1.67}

 33%|███▎      | 8155/24420 [3:23:21<6:42:11,  1.48s/it]
 33%|███▎      | 8156/24420 [3:23:23<6:42:32,  1.49s/it]
 33%|███▎      | 8157/24420 [3:23:24<6:43:56,  1.49s/it]
 33%|███▎      | 8158/24420 [3:23:25<6:42:58,  1.49s/it]
 33%|███▎      | 8159/24420 [3:23:27<6:43:08,  1.49s/it]
 33%|███▎      | 8160/24420 [3:23:28<6:45:12,  1.50s/it]
                                                        
{'loss': 0.1186, 'learning_rate': 5e-05, 'epoch': 1.67}

 33%|███▎      | 8160/24420 [3:23:29<6:45:12,  1.50s/it]
 33%|███▎      | 8161/24420 [3:23:30<6:44:28,  1.49s/it]
 33%|███▎      | 8162/24420 [3:23:31<6:43:12,  1.49s/it]
 33%|███▎      | 8163/24420 [3:23:33<6:42:47,  1.49s/it]
 33%|███▎      | 8164/24420 [3:23:34<6:42:22,  1.49s/it]
 33%|███▎      | 8165/24420 [3:23:36<6:42:53,  1.49s/it]
                                                        
{'loss': 0.111, 'learning_rate': 5e-05, 'epoch': 1.67}

 33%|███▎      | 8165/24420 [3:23:36<6:42:53,  1.49s/it]
 33%|███▎      | 8166/24420 [3:23:37<6:42:43,  1.49s/it]
 33%|███▎      | 8167/24420 [3:23:39<6:43:09,  1.49s/it]
 33%|███▎      | 8168/24420 [3:23:40<6:43:51,  1.49s/it]
 33%|███▎      | 8169/24420 [3:23:42<6:42:56,  1.49s/it]
 33%|███▎      | 8170/24420 [3:23:43<6:43:03,  1.49s/it]
                                                        
{'loss': 0.1163, 'learning_rate': 5e-05, 'epoch': 1.67}

 33%|███▎      | 8170/24420 [3:23:43<6:43:03,  1.49s/it]
 33%|███▎      | 8171/24420 [3:23:45<6:43:43,  1.49s/it]
 33%|███▎      | 8172/24420 [3:23:46<6:42:19,  1.49s/it]
 33%|███▎      | 8173/24420 [3:23:48<6:42:56,  1.49s/it]
 33%|███▎      | 8174/24420 [3:23:49<6:42:25,  1.49s/it]
 33%|███▎      | 8175/24420 [3:23:51<6:42:27,  1.49s/it]
                                                        
{'loss': 0.1172, 'learning_rate': 5e-05, 'epoch': 1.67}

 33%|███▎      | 8175/24420 [3:23:51<6:42:27,  1.49s/it]
 33%|███▎      | 8176/24420 [3:23:52<6:43:32,  1.49s/it]
 33%|███▎      | 8177/24420 [3:23:54<6:43:31,  1.49s/it]
 33%|███▎      | 8178/24420 [3:23:55<6:43:00,  1.49s/it]
 33%|███▎      | 8179/24420 [3:23:57<6:42:45,  1.49s/it]
 33%|███▎      | 8180/24420 [3:23:58<6:41:46,  1.48s/it]
                                                        
{'loss': 0.1163, 'learning_rate': 5e-05, 'epoch': 1.67}

 33%|███▎      | 8180/24420 [3:23:58<6:41:46,  1.48s/it]
 34%|███▎      | 8181/24420 [3:24:00<6:41:28,  1.48s/it]
 34%|███▎      | 8182/24420 [3:24:01<6:41:10,  1.48s/it]
 34%|███▎      | 8183/24420 [3:24:03<6:41:09,  1.48s/it]
 34%|███▎      | 8184/24420 [3:24:04<6:41:34,  1.48s/it]
 34%|███▎      | 8185/24420 [3:24:06<6:42:21,  1.49s/it]
                                                        
{'loss': 0.1135, 'learning_rate': 5e-05, 'epoch': 1.68}

 34%|███▎      | 8185/24420 [3:24:06<6:42:21,  1.49s/it]
 34%|███▎      | 8186/24420 [3:24:07<6:42:18,  1.49s/it]
 34%|███▎      | 8187/24420 [3:24:09<6:42:00,  1.49s/it]
 34%|███▎      | 8188/24420 [3:24:10<6:41:35,  1.48s/it]
 34%|███▎      | 8189/24420 [3:24:12<6:41:05,  1.48s/it]
 34%|███▎      | 8190/24420 [3:24:13<6:40:27,  1.48s/it]
                                                        
{'loss': 0.1103, 'learning_rate': 5e-05, 'epoch': 1.68}

 34%|███▎      | 8190/24420 [3:24:13<6:40:27,  1.48s/it]
 34%|███▎      | 8191/24420 [3:24:15<6:41:05,  1.48s/it]
 34%|███▎      | 8192/24420 [3:24:16<6:41:44,  1.49s/it]
 34%|███▎      | 8193/24420 [3:24:18<6:42:45,  1.49s/it]
 34%|███▎      | 8194/24420 [3:24:19<6:42:41,  1.49s/it]
 34%|███▎      | 8195/24420 [3:24:21<6:45:15,  1.50s/it]
                                                        
{'loss': 0.131, 'learning_rate': 5e-05, 'epoch': 1.68}

 34%|███▎      | 8195/24420 [3:24:21<6:45:15,  1.50s/it]
 34%|███▎      | 8196/24420 [3:24:22<6:44:16,  1.50s/it]
 34%|███▎      | 8197/24420 [3:24:24<6:43:42,  1.49s/it]
 34%|███▎      | 8198/24420 [3:24:25<6:43:36,  1.49s/it]
 34%|███▎      | 8199/24420 [3:24:26<6:42:17,  1.49s/it]
 34%|███▎      | 8200/24420 [3:24:28<6:41:53,  1.49s/it]
                                                        
{'loss': 0.1295, 'learning_rate': 5e-05, 'epoch': 1.68}

 34%|███▎      | 8200/24420 [3:24:28<6:41:53,  1.49s/it]
 34%|███▎      | 8201/24420 [3:24:33<10:57:07,  2.43s/it]
 34%|███▎      | 8202/24420 [3:24:34<9:40:26,  2.15s/it] 
 34%|███▎      | 8203/24420 [3:24:36<8:46:14,  1.95s/it]
 34%|███▎      | 8204/24420 [3:24:37<8:09:01,  1.81s/it]
 34%|███▎      | 8205/24420 [3:24:39<7:42:41,  1.71s/it]
                                                        
{'loss': 0.1018, 'learning_rate': 5e-05, 'epoch': 1.68}

 34%|███▎      | 8205/24420 [3:24:39<7:42:41,  1.71s/it]
 34%|███▎      | 8206/24420 [3:24:40<7:24:36,  1.65s/it]
 34%|███▎      | 8207/24420 [3:24:42<7:11:26,  1.60s/it]
 34%|███▎      | 8208/24420 [3:24:43<7:01:54,  1.56s/it]
 34%|███▎      | 8209/24420 [3:24:44<6:55:46,  1.54s/it]
 34%|███▎      | 8210/24420 [3:24:46<6:51:39,  1.52s/it]
                                                        
{'loss': 0.104, 'learning_rate': 5e-05, 'epoch': 1.68}

 34%|███▎      | 8210/24420 [3:24:46<6:51:39,  1.52s/it]
 34%|███▎      | 8211/24420 [3:24:47<6:48:09,  1.51s/it]
 34%|███▎      | 8212/24420 [3:24:49<6:46:50,  1.51s/it]
 34%|███▎      | 8213/24420 [3:24:50<6:44:55,  1.50s/it]
 34%|███▎      | 8214/24420 [3:24:52<6:43:25,  1.49s/it]
 34%|███▎      | 8215/24420 [3:24:53<6:42:33,  1.49s/it]
                                                        
{'loss': 0.1205, 'learning_rate': 5e-05, 'epoch': 1.68}

 34%|███▎      | 8215/24420 [3:24:53<6:42:33,  1.49s/it]
 34%|███▎      | 8216/24420 [3:24:55<6:42:37,  1.49s/it]
 34%|███▎      | 8217/24420 [3:24:56<6:42:06,  1.49s/it]
 34%|███▎      | 8218/24420 [3:24:58<6:42:00,  1.49s/it]
 34%|███▎      | 8219/24420 [3:24:59<6:43:08,  1.49s/it]
 34%|███▎      | 8220/24420 [3:25:01<6:42:32,  1.49s/it]
                                                        
{'loss': 0.1217, 'learning_rate': 5e-05, 'epoch': 1.68}

 34%|███▎      | 8220/24420 [3:25:01<6:42:32,  1.49s/it]
 34%|███▎      | 8221/24420 [3:25:02<6:41:28,  1.49s/it]
 34%|███▎      | 8222/24420 [3:25:04<6:40:55,  1.49s/it]
 34%|███▎      | 8223/24420 [3:25:05<6:40:15,  1.48s/it]
 34%|███▎      | 8224/24420 [3:25:07<6:40:32,  1.48s/it]
 34%|███▎      | 8225/24420 [3:25:08<6:40:56,  1.49s/it]
                                                        
{'loss': 0.1128, 'learning_rate': 5e-05, 'epoch': 1.68}

 34%|███▎      | 8225/24420 [3:25:08<6:40:56,  1.49s/it]
 34%|███▎      | 8226/24420 [3:25:10<6:40:47,  1.48s/it]
 34%|███▎      | 8227/24420 [3:25:11<6:40:48,  1.49s/it]
 34%|███▎      | 8228/24420 [3:25:13<6:41:37,  1.49s/it]
 34%|███▎      | 8229/24420 [3:25:14<6:41:54,  1.49s/it]
 34%|███▎      | 8230/24420 [3:25:16<6:41:50,  1.49s/it]
                                                        
{'loss': 0.1483, 'learning_rate': 5e-05, 'epoch': 1.69}

 34%|███▎      | 8230/24420 [3:25:16<6:41:50,  1.49s/it]
 34%|███▎      | 8231/24420 [3:25:17<6:42:03,  1.49s/it]
 34%|███▎      | 8232/24420 [3:25:19<6:41:27,  1.49s/it]
 34%|███▎      | 8233/24420 [3:25:20<6:40:46,  1.49s/it]
 34%|███▎      | 8234/24420 [3:25:22<6:40:06,  1.48s/it]
 34%|███▎      | 8235/24420 [3:25:23<6:40:16,  1.48s/it]
                                                        
{'loss': 0.109, 'learning_rate': 5e-05, 'epoch': 1.69}

 34%|███▎      | 8235/24420 [3:25:23<6:40:16,  1.48s/it]
 34%|███▎      | 8236/24420 [3:25:25<6:40:21,  1.48s/it]
 34%|███▎      | 8237/24420 [3:25:26<6:40:16,  1.48s/it]
 34%|███▎      | 8238/24420 [3:25:28<6:40:18,  1.48s/it]
 34%|███▎      | 8239/24420 [3:25:29<6:39:49,  1.48s/it]
 34%|███▎      | 8240/24420 [3:25:31<6:39:30,  1.48s/it]
                                                        
{'loss': 0.1082, 'learning_rate': 5e-05, 'epoch': 1.69}

 34%|███▎      | 8240/24420 [3:25:31<6:39:30,  1.48s/it]
 34%|███▎      | 8241/24420 [3:25:32<6:40:04,  1.48s/it]
 34%|███▍      | 8242/24420 [3:25:34<6:39:44,  1.48s/it]
 34%|███▍      | 8243/24420 [3:25:35<6:39:42,  1.48s/it]
 34%|███▍      | 8244/24420 [3:25:36<6:40:03,  1.48s/it]
 34%|███▍      | 8245/24420 [3:25:38<6:40:20,  1.49s/it]
                                                        
{'loss': 0.1039, 'learning_rate': 5e-05, 'epoch': 1.69}

 34%|███▍      | 8245/24420 [3:25:38<6:40:20,  1.49s/it]
 34%|███▍      | 8246/24420 [3:25:39<6:39:53,  1.48s/it]
 34%|███▍      | 8247/24420 [3:25:41<6:40:08,  1.48s/it]
 34%|███▍      | 8248/24420 [3:25:42<6:39:58,  1.48s/it]
 34%|███▍      | 8249/24420 [3:25:44<6:40:34,  1.49s/it]
 34%|███▍      | 8250/24420 [3:25:45<6:41:00,  1.49s/it]
                                                        
{'loss': 0.1066, 'learning_rate': 5e-05, 'epoch': 1.69}

 34%|███▍      | 8250/24420 [3:25:45<6:41:00,  1.49s/it]
 34%|███▍      | 8251/24420 [3:25:47<6:41:18,  1.49s/it]
 34%|███▍      | 8252/24420 [3:25:48<6:40:40,  1.49s/it]
 34%|███▍      | 8253/24420 [3:25:50<6:40:36,  1.49s/it]
 34%|███▍      | 8254/24420 [3:25:51<6:41:02,  1.49s/it]
 34%|███▍      | 8255/24420 [3:25:53<6:40:13,  1.49s/it]
                                                        
{'loss': 0.1047, 'learning_rate': 5e-05, 'epoch': 1.69}

 34%|███▍      | 8255/24420 [3:25:53<6:40:13,  1.49s/it]
 34%|███▍      | 8256/24420 [3:25:54<6:40:15,  1.49s/it]
 34%|███▍      | 8257/24420 [3:25:56<6:40:49,  1.49s/it]
 34%|███▍      | 8258/24420 [3:25:57<6:40:29,  1.49s/it]
 34%|███▍      | 8259/24420 [3:25:59<6:41:29,  1.49s/it]
 34%|███▍      | 8260/24420 [3:26:00<6:41:27,  1.49s/it]
                                                        
{'loss': 0.1069, 'learning_rate': 5e-05, 'epoch': 1.69}

 34%|███▍      | 8260/24420 [3:26:00<6:41:27,  1.49s/it]
 34%|███▍      | 8261/24420 [3:26:02<6:40:41,  1.49s/it]
 34%|███▍      | 8262/24420 [3:26:03<6:40:06,  1.49s/it]
 34%|███▍      | 8263/24420 [3:26:05<6:40:32,  1.49s/it]
 34%|███▍      | 8264/24420 [3:26:06<6:40:59,  1.49s/it]
 34%|███▍      | 8265/24420 [3:26:08<6:40:30,  1.49s/it]
                                                        
{'loss': 0.1183, 'learning_rate': 5e-05, 'epoch': 1.69}

 34%|███▍      | 8265/24420 [3:26:08<6:40:30,  1.49s/it]
 34%|███▍      | 8266/24420 [3:26:09<6:40:44,  1.49s/it]
 34%|███▍      | 8267/24420 [3:26:11<6:40:28,  1.49s/it]
 34%|███▍      | 8268/24420 [3:26:12<6:40:35,  1.49s/it]
 34%|███▍      | 8269/24420 [3:26:14<6:39:48,  1.49s/it]
 34%|███▍      | 8270/24420 [3:26:15<6:39:56,  1.49s/it]
                                                        
{'loss': 0.1034, 'learning_rate': 5e-05, 'epoch': 1.69}

 34%|███▍      | 8270/24420 [3:26:15<6:39:56,  1.49s/it]
 34%|███▍      | 8271/24420 [3:26:17<6:40:15,  1.49s/it]
 34%|███▍      | 8272/24420 [3:26:18<6:42:35,  1.50s/it]
 34%|███▍      | 8273/24420 [3:26:20<6:42:03,  1.49s/it]
 34%|███▍      | 8274/24420 [3:26:21<6:42:00,  1.49s/it]
 34%|███▍      | 8275/24420 [3:26:23<6:41:03,  1.49s/it]
                                                        
{'loss': 0.1141, 'learning_rate': 5e-05, 'epoch': 1.69}

 34%|███▍      | 8275/24420 [3:26:23<6:41:03,  1.49s/it]
 34%|███▍      | 8276/24420 [3:26:24<6:40:49,  1.49s/it]
 34%|███▍      | 8277/24420 [3:26:26<6:41:15,  1.49s/it]
 34%|███▍      | 8278/24420 [3:26:27<6:40:01,  1.49s/it]
 34%|███▍      | 8279/24420 [3:26:29<6:40:04,  1.49s/it]
 34%|███▍      | 8280/24420 [3:26:30<6:39:32,  1.49s/it]
                                                        
{'loss': 0.1169, 'learning_rate': 5e-05, 'epoch': 1.7}

 34%|███▍      | 8280/24420 [3:26:30<6:39:32,  1.49s/it]
 34%|███▍      | 8281/24420 [3:26:32<6:39:05,  1.48s/it]
 34%|███▍      | 8282/24420 [3:26:33<6:38:52,  1.48s/it]
 34%|███▍      | 8283/24420 [3:26:34<6:39:22,  1.48s/it]
 34%|███▍      | 8284/24420 [3:26:36<6:40:18,  1.49s/it]
 34%|███▍      | 8285/24420 [3:26:37<6:39:16,  1.48s/it]
                                                        
{'loss': 0.1094, 'learning_rate': 5e-05, 'epoch': 1.7}

 34%|███▍      | 8285/24420 [3:26:37<6:39:16,  1.48s/it]
 34%|███▍      | 8286/24420 [3:26:39<6:39:03,  1.48s/it]
 34%|███▍      | 8287/24420 [3:26:40<6:38:56,  1.48s/it]
 34%|███▍      | 8288/24420 [3:26:42<6:39:17,  1.49s/it]
 34%|███▍      | 8289/24420 [3:26:43<6:39:08,  1.48s/it]
 34%|███▍      | 8290/24420 [3:26:45<6:39:17,  1.49s/it]
                                                        
{'loss': 0.1203, 'learning_rate': 5e-05, 'epoch': 1.7}

 34%|███▍      | 8290/24420 [3:26:45<6:39:17,  1.49s/it]
 34%|███▍      | 8291/24420 [3:26:46<6:39:01,  1.48s/it]
 34%|███▍      | 8292/24420 [3:26:48<6:38:48,  1.48s/it]
 34%|███▍      | 8293/24420 [3:26:49<6:38:46,  1.48s/it]
 34%|███▍      | 8294/24420 [3:26:51<6:38:34,  1.48s/it]
 34%|███▍      | 8295/24420 [3:26:52<6:38:38,  1.48s/it]
                                                        
{'loss': 0.1085, 'learning_rate': 5e-05, 'epoch': 1.7}

 34%|███▍      | 8295/24420 [3:26:52<6:38:38,  1.48s/it]
 34%|███▍      | 8296/24420 [3:26:54<6:38:51,  1.48s/it]
 34%|███▍      | 8297/24420 [3:26:55<6:38:30,  1.48s/it]
 34%|███▍      | 8298/24420 [3:26:57<6:38:32,  1.48s/it]
 34%|███▍      | 8299/24420 [3:26:58<6:38:21,  1.48s/it]
 34%|███▍      | 8300/24420 [3:27:00<6:38:11,  1.48s/it]
                                                        
{'loss': 0.1166, 'learning_rate': 5e-05, 'epoch': 1.7}

 34%|███▍      | 8300/24420 [3:27:00<6:38:11,  1.48s/it]
 34%|███▍      | 8301/24420 [3:27:01<6:37:40,  1.48s/it]
 34%|███▍      | 8302/24420 [3:27:03<6:37:25,  1.48s/it]
 34%|███▍      | 8303/24420 [3:27:04<6:37:08,  1.48s/it]
 34%|███▍      | 8304/24420 [3:27:06<6:37:10,  1.48s/it]
 34%|███▍      | 8305/24420 [3:27:07<6:37:32,  1.48s/it]
                                                        
{'loss': 0.0906, 'learning_rate': 5e-05, 'epoch': 1.7}

 34%|███▍      | 8305/24420 [3:27:07<6:37:32,  1.48s/it]
 34%|███▍      | 8306/24420 [3:27:09<6:38:23,  1.48s/it]
 34%|███▍      | 8307/24420 [3:27:10<6:38:15,  1.48s/it]
 34%|███▍      | 8308/24420 [3:27:12<6:38:31,  1.48s/it]
 34%|███▍      | 8309/24420 [3:27:13<6:37:53,  1.48s/it]
 34%|███▍      | 8310/24420 [3:27:15<6:37:36,  1.48s/it]
                                                        
{'loss': 0.1091, 'learning_rate': 5e-05, 'epoch': 1.7}

 34%|███▍      | 8310/24420 [3:27:15<6:37:36,  1.48s/it]
 34%|███▍      | 8311/24420 [3:27:16<6:38:05,  1.48s/it]
 34%|███▍      | 8312/24420 [3:27:18<6:39:04,  1.49s/it]
 34%|███▍      | 8313/24420 [3:27:19<6:39:00,  1.49s/it]
 34%|███▍      | 8314/24420 [3:27:20<6:38:48,  1.49s/it]
 34%|███▍      | 8315/24420 [3:27:22<6:38:34,  1.48s/it]
                                                        
{'loss': 0.1084, 'learning_rate': 5e-05, 'epoch': 1.7}

 34%|███▍      | 8315/24420 [3:27:22<6:38:34,  1.48s/it]
 34%|███▍      | 8316/24420 [3:27:23<6:38:47,  1.49s/it]
 34%|███▍      | 8317/24420 [3:27:25<6:38:32,  1.48s/it]
 34%|███▍      | 8318/24420 [3:27:26<6:38:09,  1.48s/it]
 34%|███▍      | 8319/24420 [3:27:28<6:37:44,  1.48s/it]
 34%|███▍      | 8320/24420 [3:27:29<6:37:26,  1.48s/it]
                                                        
{'loss': 0.102, 'learning_rate': 5e-05, 'epoch': 1.7}

 34%|███▍      | 8320/24420 [3:27:29<6:37:26,  1.48s/it]
 34%|███▍      | 8321/24420 [3:27:31<6:40:10,  1.49s/it]
 34%|███▍      | 8322/24420 [3:27:32<6:38:50,  1.49s/it]
 34%|███▍      | 8323/24420 [3:27:34<6:38:34,  1.49s/it]
 34%|███▍      | 8324/24420 [3:27:35<6:38:22,  1.49s/it]
 34%|███▍      | 8325/24420 [3:27:37<6:38:36,  1.49s/it]
                                                        
{'loss': 0.1276, 'learning_rate': 5e-05, 'epoch': 1.7}

 34%|███▍      | 8325/24420 [3:27:37<6:38:36,  1.49s/it]
 34%|███▍      | 8326/24420 [3:27:38<6:38:52,  1.49s/it]
 34%|███▍      | 8327/24420 [3:27:40<6:38:41,  1.49s/it]
 34%|███▍      | 8328/24420 [3:27:41<6:38:26,  1.49s/it]
 34%|███▍      | 8329/24420 [3:27:43<6:37:27,  1.48s/it]
 34%|███▍      | 8330/24420 [3:27:44<6:37:33,  1.48s/it]
                                                        
{'loss': 0.1033, 'learning_rate': 5e-05, 'epoch': 1.71}

 34%|███▍      | 8330/24420 [3:27:44<6:37:33,  1.48s/it]
 34%|███▍      | 8331/24420 [3:27:46<6:37:21,  1.48s/it]
 34%|███▍      | 8332/24420 [3:27:47<6:37:03,  1.48s/it]
 34%|███▍      | 8333/24420 [3:27:49<6:36:45,  1.48s/it]
 34%|███▍      | 8334/24420 [3:27:50<6:37:13,  1.48s/it]
 34%|███▍      | 8335/24420 [3:27:52<6:38:01,  1.48s/it]
                                                        
{'loss': 0.1097, 'learning_rate': 5e-05, 'epoch': 1.71}

 34%|███▍      | 8335/24420 [3:27:52<6:38:01,  1.48s/it]
 34%|███▍      | 8336/24420 [3:27:53<6:37:25,  1.48s/it]
 34%|███▍      | 8337/24420 [3:27:55<6:37:05,  1.48s/it]
 34%|███▍      | 8338/24420 [3:27:56<6:37:16,  1.48s/it]
 34%|███▍      | 8339/24420 [3:27:58<6:37:53,  1.48s/it]
 34%|███▍      | 8340/24420 [3:27:59<6:37:38,  1.48s/it]
                                                        
{'loss': 0.0995, 'learning_rate': 5e-05, 'epoch': 1.71}

 34%|███▍      | 8340/24420 [3:27:59<6:37:38,  1.48s/it]
 34%|███▍      | 8341/24420 [3:28:01<6:37:21,  1.48s/it]
 34%|███▍      | 8342/24420 [3:28:02<6:38:08,  1.49s/it]
 34%|███▍      | 8343/24420 [3:28:04<6:37:42,  1.48s/it]
 34%|███▍      | 8344/24420 [3:28:05<6:37:23,  1.48s/it]
 34%|███▍      | 8345/24420 [3:28:06<6:37:08,  1.48s/it]
                                                        
{'loss': 0.1187, 'learning_rate': 5e-05, 'epoch': 1.71}

 34%|███▍      | 8345/24420 [3:28:06<6:37:08,  1.48s/it]
 34%|███▍      | 8346/24420 [3:28:08<6:37:14,  1.48s/it]
 34%|███▍      | 8347/24420 [3:28:09<6:37:24,  1.48s/it]
 34%|███▍      | 8348/24420 [3:28:11<6:37:25,  1.48s/it]
 34%|███▍      | 8349/24420 [3:28:12<6:37:28,  1.48s/it]
 34%|███▍      | 8350/24420 [3:28:14<6:37:31,  1.48s/it]
                                                        
{'loss': 0.1076, 'learning_rate': 5e-05, 'epoch': 1.71}

 34%|███▍      | 8350/24420 [3:28:14<6:37:31,  1.48s/it]
 34%|███▍      | 8351/24420 [3:28:15<6:37:40,  1.48s/it]
 34%|███▍      | 8352/24420 [3:28:17<6:37:21,  1.48s/it]
 34%|███▍      | 8353/24420 [3:28:18<6:38:00,  1.49s/it]
 34%|███▍      | 8354/24420 [3:28:20<6:38:22,  1.49s/it]
 34%|███▍      | 8355/24420 [3:28:21<6:38:00,  1.49s/it]
                                                        
{'loss': 0.1163, 'learning_rate': 5e-05, 'epoch': 1.71}

 34%|███▍      | 8355/24420 [3:28:21<6:38:00,  1.49s/it]
 34%|███▍      | 8356/24420 [3:28:23<6:38:15,  1.49s/it]
 34%|███▍      | 8357/24420 [3:28:24<6:37:24,  1.48s/it]
 34%|███▍      | 8358/24420 [3:28:26<6:36:38,  1.48s/it]
 34%|███▍      | 8359/24420 [3:28:27<6:36:29,  1.48s/it]
 34%|███▍      | 8360/24420 [3:28:29<6:35:38,  1.48s/it]
                                                        
{'loss': 0.105, 'learning_rate': 5e-05, 'epoch': 1.71}

 34%|███▍      | 8360/24420 [3:28:29<6:35:38,  1.48s/it]
 34%|███▍      | 8361/24420 [3:28:30<6:35:15,  1.48s/it]
 34%|███▍      | 8362/24420 [3:28:32<6:35:26,  1.48s/it]
 34%|███▍      | 8363/24420 [3:28:33<6:36:01,  1.48s/it]
 34%|███▍      | 8364/24420 [3:28:35<6:36:08,  1.48s/it]
 34%|███▍      | 8365/24420 [3:28:36<6:36:38,  1.48s/it]
                                                        
{'loss': 0.1201, 'learning_rate': 5e-05, 'epoch': 1.71}

 34%|███▍      | 8365/24420 [3:28:36<6:36:38,  1.48s/it]
 34%|███▍      | 8366/24420 [3:28:38<6:36:22,  1.48s/it]
 34%|███▍      | 8367/24420 [3:28:39<6:37:00,  1.48s/it]
 34%|███▍      | 8368/24420 [3:28:41<6:36:17,  1.48s/it]
 34%|███▍      | 8369/24420 [3:28:42<6:36:18,  1.48s/it]
 34%|███▍      | 8370/24420 [3:28:44<6:35:56,  1.48s/it]
                                                        
{'loss': 0.0985, 'learning_rate': 5e-05, 'epoch': 1.71}

 34%|███▍      | 8370/24420 [3:28:44<6:35:56,  1.48s/it]
 34%|███▍      | 8371/24420 [3:28:45<6:36:46,  1.48s/it]
 34%|███▍      | 8372/24420 [3:28:47<6:37:25,  1.49s/it]
 34%|███▍      | 8373/24420 [3:28:48<6:36:50,  1.48s/it]
 34%|███▍      | 8374/24420 [3:28:49<6:36:21,  1.48s/it]
 34%|███▍      | 8375/24420 [3:28:51<6:36:16,  1.48s/it]
                                                        
{'loss': 0.1173, 'learning_rate': 5e-05, 'epoch': 1.71}

 34%|███▍      | 8375/24420 [3:28:51<6:36:16,  1.48s/it]
 34%|███▍      | 8376/24420 [3:28:52<6:36:01,  1.48s/it]
 34%|███▍      | 8377/24420 [3:28:54<6:35:43,  1.48s/it]
 34%|███▍      | 8378/24420 [3:28:55<6:38:06,  1.49s/it]
 34%|███▍      | 8379/24420 [3:28:57<6:37:59,  1.49s/it]
 34%|███▍      | 8380/24420 [3:28:58<6:39:14,  1.49s/it]
                                                        
{'loss': 0.1119, 'learning_rate': 5e-05, 'epoch': 1.72}

 34%|███▍      | 8380/24420 [3:28:58<6:39:14,  1.49s/it]
 34%|███▍      | 8381/24420 [3:29:00<6:38:30,  1.49s/it]
 34%|███▍      | 8382/24420 [3:29:01<6:38:16,  1.49s/it]
 34%|███▍      | 8383/24420 [3:29:03<6:37:46,  1.49s/it]
 34%|███▍      | 8384/24420 [3:29:04<6:37:42,  1.49s/it]
 34%|███▍      | 8385/24420 [3:29:06<6:37:31,  1.49s/it]
                                                        
{'loss': 0.1068, 'learning_rate': 5e-05, 'epoch': 1.72}

 34%|███▍      | 8385/24420 [3:29:06<6:37:31,  1.49s/it]
 34%|███▍      | 8386/24420 [3:29:07<6:38:13,  1.49s/it]
 34%|███▍      | 8387/24420 [3:29:09<6:37:16,  1.49s/it]
 34%|███▍      | 8388/24420 [3:29:10<6:36:56,  1.49s/it]
 34%|███▍      | 8389/24420 [3:29:12<6:36:38,  1.48s/it]
 34%|███▍      | 8390/24420 [3:29:13<6:37:26,  1.49s/it]
                                                        
{'loss': 0.1027, 'learning_rate': 5e-05, 'epoch': 1.72}

 34%|███▍      | 8390/24420 [3:29:13<6:37:26,  1.49s/it]
 34%|███▍      | 8391/24420 [3:29:15<6:37:19,  1.49s/it]
 34%|███▍      | 8392/24420 [3:29:16<6:36:09,  1.48s/it]
 34%|███▍      | 8393/24420 [3:29:18<6:36:40,  1.49s/it]
 34%|███▍      | 8394/24420 [3:29:19<6:37:15,  1.49s/it]
 34%|███▍      | 8395/24420 [3:29:21<6:36:50,  1.49s/it]
                                                        
{'loss': 0.1247, 'learning_rate': 5e-05, 'epoch': 1.72}

 34%|███▍      | 8395/24420 [3:29:21<6:36:50,  1.49s/it]
 34%|███▍      | 8396/24420 [3:29:22<6:37:01,  1.49s/it]
 34%|███▍      | 8397/24420 [3:29:24<6:36:55,  1.49s/it]
 34%|███▍      | 8398/24420 [3:29:25<6:36:19,  1.48s/it]
 34%|███▍      | 8399/24420 [3:29:27<6:36:31,  1.49s/it]
 34%|███▍      | 8400/24420 [3:29:28<6:36:59,  1.49s/it]
                                                        
{'loss': 0.0979, 'learning_rate': 5e-05, 'epoch': 1.72}

 34%|███▍      | 8400/24420 [3:29:28<6:36:59,  1.49s/it]
 34%|███▍      | 8401/24420 [3:29:33<10:44:38,  2.41s/it]
 34%|███▍      | 8402/24420 [3:29:34<9:29:39,  2.13s/it] 
 34%|███▍      | 8403/24420 [3:29:36<8:37:44,  1.94s/it]
 34%|███▍      | 8404/24420 [3:29:37<8:00:49,  1.80s/it]
 34%|███▍      | 8405/24420 [3:29:39<7:35:41,  1.71s/it]
                                                        
{'loss': 0.1024, 'learning_rate': 5e-05, 'epoch': 1.72}

 34%|███▍      | 8405/24420 [3:29:39<7:35:41,  1.71s/it]
 34%|███▍      | 8406/24420 [3:29:40<7:17:53,  1.64s/it]
 34%|███▍      | 8407/24420 [3:29:42<7:05:07,  1.59s/it]
 34%|███▍      | 8408/24420 [3:29:43<6:56:04,  1.56s/it]
 34%|███▍      | 8409/24420 [3:29:45<6:49:53,  1.54s/it]
 34%|███▍      | 8410/24420 [3:29:46<6:45:20,  1.52s/it]
                                                        
{'loss': 0.0977, 'learning_rate': 5e-05, 'epoch': 1.72}

 34%|███▍      | 8410/24420 [3:29:46<6:45:20,  1.52s/it]
 34%|███▍      | 8411/24420 [3:29:48<6:42:48,  1.51s/it]
 34%|███▍      | 8412/24420 [3:29:49<6:40:30,  1.50s/it]
 34%|███▍      | 8413/24420 [3:29:51<6:40:19,  1.50s/it]
 34%|███▍      | 8414/24420 [3:29:52<6:38:16,  1.49s/it]
 34%|███▍      | 8415/24420 [3:29:53<6:37:16,  1.49s/it]
                                                        
{'loss': 0.1111, 'learning_rate': 5e-05, 'epoch': 1.72}

 34%|███▍      | 8415/24420 [3:29:53<6:37:16,  1.49s/it]
 34%|███▍      | 8416/24420 [3:29:55<6:36:34,  1.49s/it]
 34%|███▍      | 8417/24420 [3:29:56<6:36:16,  1.49s/it]
 34%|███▍      | 8418/24420 [3:29:58<6:36:13,  1.49s/it]
 34%|███▍      | 8419/24420 [3:29:59<6:35:26,  1.48s/it]
 34%|███▍      | 8420/24420 [3:30:01<6:35:27,  1.48s/it]
                                                        
{'loss': 0.123, 'learning_rate': 5e-05, 'epoch': 1.72}

 34%|███▍      | 8420/24420 [3:30:01<6:35:27,  1.48s/it]
 34%|███▍      | 8421/24420 [3:30:02<6:35:33,  1.48s/it]
 34%|███▍      | 8422/24420 [3:30:04<6:35:48,  1.48s/it]
 34%|███▍      | 8423/24420 [3:30:05<6:36:01,  1.49s/it]
 34%|███▍      | 8424/24420 [3:30:07<6:35:57,  1.49s/it]
 35%|███▍      | 8425/24420 [3:30:08<6:36:49,  1.49s/it]
                                                        
{'loss': 0.1307, 'learning_rate': 5e-05, 'epoch': 1.73}

 35%|███▍      | 8425/24420 [3:30:08<6:36:49,  1.49s/it]
 35%|███▍      | 8426/24420 [3:30:10<6:37:01,  1.49s/it]
 35%|███▍      | 8427/24420 [3:30:11<6:36:49,  1.49s/it]
 35%|███▍      | 8428/24420 [3:30:13<6:36:47,  1.49s/it]
 35%|███▍      | 8429/24420 [3:30:14<6:36:28,  1.49s/it]
 35%|███▍      | 8430/24420 [3:30:16<6:36:12,  1.49s/it]
                                                        
{'loss': 0.1166, 'learning_rate': 5e-05, 'epoch': 1.73}

 35%|███▍      | 8430/24420 [3:30:16<6:36:12,  1.49s/it]
 35%|███▍      | 8431/24420 [3:30:17<6:36:43,  1.49s/it]
 35%|███▍      | 8432/24420 [3:30:19<6:36:34,  1.49s/it]
 35%|███▍      | 8433/24420 [3:30:20<6:36:02,  1.49s/it]
 35%|███▍      | 8434/24420 [3:30:22<6:35:54,  1.49s/it]
 35%|███▍      | 8435/24420 [3:30:23<6:35:27,  1.48s/it]
                                                        
{'loss': 0.1059, 'learning_rate': 5e-05, 'epoch': 1.73}

 35%|███▍      | 8435/24420 [3:30:23<6:35:27,  1.48s/it]
 35%|███▍      | 8436/24420 [3:30:25<6:35:23,  1.48s/it]
 35%|███▍      | 8437/24420 [3:30:26<6:35:00,  1.48s/it]
 35%|███▍      | 8438/24420 [3:30:28<6:35:19,  1.48s/it]
 35%|███▍      | 8439/24420 [3:30:29<6:34:27,  1.48s/it]
 35%|███▍      | 8440/24420 [3:30:31<6:34:18,  1.48s/it]
                                                        
{'loss': 0.093, 'learning_rate': 5e-05, 'epoch': 1.73}

 35%|███▍      | 8440/24420 [3:30:31<6:34:18,  1.48s/it]
 35%|███▍      | 8441/24420 [3:30:32<6:33:59,  1.48s/it]
 35%|███▍      | 8442/24420 [3:30:34<6:33:53,  1.48s/it]
 35%|███▍      | 8443/24420 [3:30:35<6:33:59,  1.48s/it]
 35%|███▍      | 8444/24420 [3:30:37<6:33:16,  1.48s/it]
 35%|███▍      | 8445/24420 [3:30:38<6:33:49,  1.48s/it]
                                                        
{'loss': 0.1119, 'learning_rate': 5e-05, 'epoch': 1.73}

 35%|███▍      | 8445/24420 [3:30:38<6:33:49,  1.48s/it]
 35%|███▍      | 8446/24420 [3:30:39<6:34:42,  1.48s/it]
 35%|███▍      | 8447/24420 [3:30:41<6:34:17,  1.48s/it]
 35%|███▍      | 8448/24420 [3:30:42<6:34:09,  1.48s/it]
 35%|███▍      | 8449/24420 [3:30:44<6:34:26,  1.48s/it]
 35%|███▍      | 8450/24420 [3:30:45<6:38:01,  1.50s/it]
                                                        
{'loss': 0.1094, 'learning_rate': 5e-05, 'epoch': 1.73}

 35%|███▍      | 8450/24420 [3:30:45<6:38:01,  1.50s/it]
 35%|███▍      | 8451/24420 [3:30:47<6:37:30,  1.49s/it]
 35%|███▍      | 8452/24420 [3:30:48<6:36:38,  1.49s/it]
 35%|███▍      | 8453/24420 [3:30:50<6:35:51,  1.49s/it]
 35%|███▍      | 8454/24420 [3:30:51<6:35:06,  1.48s/it]
 35%|███▍      | 8455/24420 [3:30:53<6:36:15,  1.49s/it]
                                                        
{'loss': 0.1263, 'learning_rate': 5e-05, 'epoch': 1.73}

 35%|███▍      | 8455/24420 [3:30:53<6:36:15,  1.49s/it]
 35%|███▍      | 8456/24420 [3:30:54<6:35:13,  1.49s/it]
 35%|███▍      | 8457/24420 [3:30:56<6:34:36,  1.48s/it]
 35%|███▍      | 8458/24420 [3:30:57<6:35:03,  1.49s/it]
 35%|███▍      | 8459/24420 [3:30:59<6:34:44,  1.48s/it]
 35%|███▍      | 8460/24420 [3:31:00<6:34:29,  1.48s/it]
                                                        
{'loss': 0.1264, 'learning_rate': 5e-05, 'epoch': 1.73}

 35%|███▍      | 8460/24420 [3:31:00<6:34:29,  1.48s/it]
 35%|███▍      | 8461/24420 [3:31:02<6:34:59,  1.49s/it]
 35%|███▍      | 8462/24420 [3:31:03<6:34:32,  1.48s/it]
 35%|███▍      | 8463/24420 [3:31:05<6:33:59,  1.48s/it]
 35%|███▍      | 8464/24420 [3:31:06<6:34:22,  1.48s/it]
 35%|███▍      | 8465/24420 [3:31:08<6:33:41,  1.48s/it]
                                                        
{'loss': 0.0925, 'learning_rate': 5e-05, 'epoch': 1.73}

 35%|███▍      | 8465/24420 [3:31:08<6:33:41,  1.48s/it]
 35%|███▍      | 8466/24420 [3:31:09<6:33:49,  1.48s/it]
 35%|███▍      | 8467/24420 [3:31:11<6:33:52,  1.48s/it]
 35%|███▍      | 8468/24420 [3:31:12<6:33:59,  1.48s/it]
 35%|███▍      | 8469/24420 [3:31:14<6:34:39,  1.48s/it]
 35%|███▍      | 8470/24420 [3:31:15<6:34:55,  1.49s/it]
                                                        
{'loss': 0.1107, 'learning_rate': 5e-05, 'epoch': 1.73}

 35%|███▍      | 8470/24420 [3:31:15<6:34:55,  1.49s/it]
 35%|███▍      | 8471/24420 [3:31:17<6:34:35,  1.48s/it]
 35%|███▍      | 8472/24420 [3:31:18<6:34:59,  1.49s/it]
 35%|███▍      | 8473/24420 [3:31:20<6:34:23,  1.48s/it]
 35%|███▍      | 8474/24420 [3:31:21<6:34:38,  1.48s/it]
 35%|███▍      | 8475/24420 [3:31:23<6:34:51,  1.49s/it]
                                                        
{'loss': 0.1299, 'learning_rate': 5e-05, 'epoch': 1.74}

 35%|███▍      | 8475/24420 [3:31:23<6:34:51,  1.49s/it]
 35%|███▍      | 8476/24420 [3:31:24<6:35:30,  1.49s/it]
 35%|███▍      | 8477/24420 [3:31:26<6:34:44,  1.49s/it]
 35%|███▍      | 8478/24420 [3:31:27<6:34:37,  1.49s/it]
 35%|███▍      | 8479/24420 [3:31:28<6:35:07,  1.49s/it]
 35%|███▍      | 8480/24420 [3:31:30<6:34:36,  1.49s/it]
                                                        
{'loss': 0.1066, 'learning_rate': 5e-05, 'epoch': 1.74}

 35%|███▍      | 8480/24420 [3:31:30<6:34:36,  1.49s/it]
 35%|███▍      | 8481/24420 [3:31:31<6:33:54,  1.48s/it]
 35%|███▍      | 8482/24420 [3:31:33<6:33:32,  1.48s/it]
 35%|███▍      | 8483/24420 [3:31:34<6:33:39,  1.48s/it]
 35%|███▍      | 8484/24420 [3:31:36<6:33:55,  1.48s/it]
 35%|███▍      | 8485/24420 [3:31:37<6:33:50,  1.48s/it]
                                                        
{'loss': 0.0902, 'learning_rate': 5e-05, 'epoch': 1.74}

 35%|███▍      | 8485/24420 [3:31:37<6:33:50,  1.48s/it]
 35%|███▍      | 8486/24420 [3:31:39<6:34:00,  1.48s/it]
 35%|███▍      | 8487/24420 [3:31:40<6:34:08,  1.48s/it]
 35%|███▍      | 8488/24420 [3:31:42<6:33:54,  1.48s/it]
 35%|███▍      | 8489/24420 [3:31:43<6:33:53,  1.48s/it]
 35%|███▍      | 8490/24420 [3:31:45<6:33:31,  1.48s/it]
                                                        
{'loss': 0.1143, 'learning_rate': 5e-05, 'epoch': 1.74}

 35%|███▍      | 8490/24420 [3:31:45<6:33:31,  1.48s/it]
 35%|███▍      | 8491/24420 [3:31:46<6:33:38,  1.48s/it]
 35%|███▍      | 8492/24420 [3:31:48<6:33:32,  1.48s/it]
 35%|███▍      | 8493/24420 [3:31:49<6:33:55,  1.48s/it]
 35%|███▍      | 8494/24420 [3:31:51<6:33:54,  1.48s/it]
 35%|███▍      | 8495/24420 [3:31:52<6:33:46,  1.48s/it]
                                                        
{'loss': 0.0989, 'learning_rate': 5e-05, 'epoch': 1.74}

 35%|███▍      | 8495/24420 [3:31:52<6:33:46,  1.48s/it]
 35%|███▍      | 8496/24420 [3:31:54<6:33:55,  1.48s/it]
 35%|███▍      | 8497/24420 [3:31:55<6:33:40,  1.48s/it]
 35%|███▍      | 8498/24420 [3:31:57<6:33:35,  1.48s/it]
 35%|███▍      | 8499/24420 [3:31:58<6:33:39,  1.48s/it]
 35%|███▍      | 8500/24420 [3:32:00<6:33:27,  1.48s/it]
                                                        
{'loss': 0.1065, 'learning_rate': 5e-05, 'epoch': 1.74}

 35%|███▍      | 8500/24420 [3:32:00<6:33:27,  1.48s/it]
 35%|███▍      | 8501/24420 [3:32:01<6:33:46,  1.48s/it]
 35%|███▍      | 8502/24420 [3:32:03<6:33:20,  1.48s/it]
 35%|███▍      | 8503/24420 [3:32:04<6:34:10,  1.49s/it]
 35%|███▍      | 8504/24420 [3:32:06<6:33:34,  1.48s/it]
 35%|███▍      | 8505/24420 [3:32:07<6:34:01,  1.49s/it]
                                                        
{'loss': 0.1208, 'learning_rate': 5e-05, 'epoch': 1.74}

 35%|███▍      | 8505/24420 [3:32:07<6:34:01,  1.49s/it]
 35%|███▍      | 8506/24420 [3:32:09<6:33:36,  1.48s/it]
 35%|███▍      | 8507/24420 [3:32:10<6:33:29,  1.48s/it]
 35%|███▍      | 8508/24420 [3:32:12<6:33:30,  1.48s/it]
 35%|███▍      | 8509/24420 [3:32:13<6:33:40,  1.48s/it]
 35%|███▍      | 8510/24420 [3:32:14<6:33:26,  1.48s/it]
                                                        
{'loss': 0.123, 'learning_rate': 5e-05, 'epoch': 1.74}

 35%|███▍      | 8510/24420 [3:32:14<6:33:26,  1.48s/it]
 35%|███▍      | 8511/24420 [3:32:16<6:33:23,  1.48s/it]
 35%|███▍      | 8512/24420 [3:32:17<6:34:13,  1.49s/it]
 35%|███▍      | 8513/24420 [3:32:19<6:34:20,  1.49s/it]
 35%|███▍      | 8514/24420 [3:32:20<6:34:13,  1.49s/it]
 35%|███▍      | 8515/24420 [3:32:22<6:33:34,  1.48s/it]
                                                        
{'loss': 0.1024, 'learning_rate': 5e-05, 'epoch': 1.74}

 35%|███▍      | 8515/24420 [3:32:22<6:33:34,  1.48s/it]
 35%|███▍      | 8516/24420 [3:32:23<6:34:00,  1.49s/it]
 35%|███▍      | 8517/24420 [3:32:25<6:33:56,  1.49s/it]
 35%|███▍      | 8518/24420 [3:32:26<6:34:13,  1.49s/it]
 35%|███▍      | 8519/24420 [3:32:28<6:34:27,  1.49s/it]
 35%|███▍      | 8520/24420 [3:32:29<6:34:13,  1.49s/it]
                                                        
{'loss': 0.1052, 'learning_rate': 5e-05, 'epoch': 1.74}

 35%|███▍      | 8520/24420 [3:32:29<6:34:13,  1.49s/it]
 35%|███▍      | 8521/24420 [3:32:31<6:33:58,  1.49s/it]
 35%|███▍      | 8522/24420 [3:32:32<6:33:33,  1.49s/it]
 35%|███▍      | 8523/24420 [3:32:34<6:34:00,  1.49s/it]
 35%|███▍      | 8524/24420 [3:32:35<6:34:15,  1.49s/it]
 35%|███▍      | 8525/24420 [3:32:37<6:34:34,  1.49s/it]
                                                        
{'loss': 0.1152, 'learning_rate': 5e-05, 'epoch': 1.75}

 35%|███▍      | 8525/24420 [3:32:37<6:34:34,  1.49s/it]
 35%|███▍      | 8526/24420 [3:32:38<6:34:25,  1.49s/it]
 35%|███▍      | 8527/24420 [3:32:40<6:33:21,  1.49s/it]
 35%|███▍      | 8528/24420 [3:32:41<6:32:40,  1.48s/it]
 35%|███▍      | 8529/24420 [3:32:43<6:32:47,  1.48s/it]
 35%|███▍      | 8530/24420 [3:32:44<6:33:00,  1.48s/it]
                                                        
{'loss': 0.106, 'learning_rate': 5e-05, 'epoch': 1.75}

 35%|███▍      | 8530/24420 [3:32:44<6:33:00,  1.48s/it]
 35%|███▍      | 8531/24420 [3:32:46<6:33:01,  1.48s/it]
 35%|███▍      | 8532/24420 [3:32:47<6:32:18,  1.48s/it]
 35%|███▍      | 8533/24420 [3:32:49<6:32:12,  1.48s/it]
 35%|███▍      | 8534/24420 [3:32:50<6:32:01,  1.48s/it]
 35%|███▍      | 8535/24420 [3:32:52<6:31:46,  1.48s/it]
                                                        
{'loss': 0.1356, 'learning_rate': 5e-05, 'epoch': 1.75}

 35%|███▍      | 8535/24420 [3:32:52<6:31:46,  1.48s/it]
 35%|███▍      | 8536/24420 [3:32:53<6:45:48,  1.53s/it]
 35%|███▍      | 8537/24420 [3:32:55<6:41:28,  1.52s/it]
 35%|███▍      | 8538/24420 [3:32:56<6:38:09,  1.50s/it]
 35%|███▍      | 8539/24420 [3:32:58<6:35:36,  1.49s/it]
 35%|███▍      | 8540/24420 [3:32:59<6:34:07,  1.49s/it]
                                                        
{'loss': 0.1268, 'learning_rate': 5e-05, 'epoch': 1.75}

 35%|███▍      | 8540/24420 [3:32:59<6:34:07,  1.49s/it]
 35%|███▍      | 8541/24420 [3:33:01<6:33:36,  1.49s/it]
 35%|███▍      | 8542/24420 [3:33:02<6:34:04,  1.49s/it]
 35%|███▍      | 8543/24420 [3:33:04<6:33:58,  1.49s/it]
 35%|███▍      | 8544/24420 [3:33:05<6:34:35,  1.49s/it]
 35%|███▍      | 8545/24420 [3:33:07<6:33:38,  1.49s/it]
                                                        
{'loss': 0.1315, 'learning_rate': 5e-05, 'epoch': 1.75}

 35%|███▍      | 8545/24420 [3:33:07<6:33:38,  1.49s/it]
 35%|███▍      | 8546/24420 [3:33:08<6:32:51,  1.48s/it]
 35%|███▌      | 8547/24420 [3:33:10<6:32:28,  1.48s/it]
 35%|███▌      | 8548/24420 [3:33:11<6:32:52,  1.49s/it]
 35%|███▌      | 8549/24420 [3:33:13<6:32:53,  1.49s/it]
 35%|███▌      | 8550/24420 [3:33:14<6:32:42,  1.48s/it]
                                                        
{'loss': 0.0977, 'learning_rate': 5e-05, 'epoch': 1.75}

 35%|███▌      | 8550/24420 [3:33:14<6:32:42,  1.48s/it]
 35%|███▌      | 8551/24420 [3:33:16<6:32:43,  1.48s/it]
 35%|███▌      | 8552/24420 [3:33:17<6:32:27,  1.48s/it]
 35%|███▌      | 8553/24420 [3:33:18<6:32:47,  1.49s/it]
 35%|███▌      | 8554/24420 [3:33:20<6:32:14,  1.48s/it]
 35%|███▌      | 8555/24420 [3:33:21<6:32:07,  1.48s/it]
                                                        
{'loss': 0.1267, 'learning_rate': 5e-05, 'epoch': 1.75}

 35%|███▌      | 8555/24420 [3:33:21<6:32:07,  1.48s/it]
 35%|███▌      | 8556/24420 [3:33:23<6:32:45,  1.49s/it]
 35%|███▌      | 8557/24420 [3:33:24<6:32:41,  1.49s/it]
 35%|███▌      | 8558/24420 [3:33:26<6:32:59,  1.49s/it]
 35%|███▌      | 8559/24420 [3:33:27<6:33:05,  1.49s/it]
 35%|███▌      | 8560/24420 [3:33:29<6:33:08,  1.49s/it]
                                                        
{'loss': 0.1096, 'learning_rate': 5e-05, 'epoch': 1.75}

 35%|███▌      | 8560/24420 [3:33:29<6:33:08,  1.49s/it]
 35%|███▌      | 8561/24420 [3:33:30<6:32:43,  1.49s/it]
 35%|███▌      | 8562/24420 [3:33:32<6:31:56,  1.48s/it]
 35%|███▌      | 8563/24420 [3:33:33<6:31:33,  1.48s/it]
 35%|███▌      | 8564/24420 [3:33:35<6:31:33,  1.48s/it]
 35%|███▌      | 8565/24420 [3:33:36<6:32:09,  1.48s/it]
                                                        
{'loss': 0.0983, 'learning_rate': 5e-05, 'epoch': 1.75}

 35%|███▌      | 8565/24420 [3:33:36<6:32:09,  1.48s/it]
 35%|███▌      | 8566/24420 [3:33:38<6:31:28,  1.48s/it]
 35%|███▌      | 8567/24420 [3:33:39<6:31:18,  1.48s/it]
 35%|███▌      | 8568/24420 [3:33:41<6:31:47,  1.48s/it]
 35%|███▌      | 8569/24420 [3:33:42<6:31:41,  1.48s/it]
 35%|███▌      | 8570/24420 [3:33:44<6:31:07,  1.48s/it]
                                                        
{'loss': 0.1148, 'learning_rate': 5e-05, 'epoch': 1.75}

 35%|███▌      | 8570/24420 [3:33:44<6:31:07,  1.48s/it]
 35%|███▌      | 8571/24420 [3:33:45<6:31:09,  1.48s/it]
 35%|███▌      | 8572/24420 [3:33:47<6:31:30,  1.48s/it]
 35%|███▌      | 8573/24420 [3:33:48<6:31:27,  1.48s/it]
 35%|███▌      | 8574/24420 [3:33:50<6:31:23,  1.48s/it]
 35%|███▌      | 8575/24420 [3:33:51<6:31:38,  1.48s/it]
                                                        
{'loss': 0.1056, 'learning_rate': 5e-05, 'epoch': 1.76}

 35%|███▌      | 8575/24420 [3:33:51<6:31:38,  1.48s/it]
 35%|███▌      | 8576/24420 [3:33:53<6:31:29,  1.48s/it]
 35%|███▌      | 8577/24420 [3:33:54<6:31:04,  1.48s/it]
 35%|███▌      | 8578/24420 [3:33:56<6:31:01,  1.48s/it]
 35%|███▌      | 8579/24420 [3:33:57<6:31:17,  1.48s/it]
 35%|███▌      | 8580/24420 [3:33:59<6:30:59,  1.48s/it]
                                                        
{'loss': 0.0973, 'learning_rate': 5e-05, 'epoch': 1.76}

 35%|███▌      | 8580/24420 [3:33:59<6:30:59,  1.48s/it]
 35%|███▌      | 8581/24420 [3:34:00<6:30:51,  1.48s/it]
 35%|███▌      | 8582/24420 [3:34:01<6:32:14,  1.49s/it]
 35%|███▌      | 8583/24420 [3:34:03<6:31:57,  1.48s/it]
 35%|███▌      | 8584/24420 [3:34:04<6:31:10,  1.48s/it]
 35%|███▌      | 8585/24420 [3:34:06<6:32:33,  1.49s/it]
                                                        
{'loss': 0.1168, 'learning_rate': 5e-05, 'epoch': 1.76}

 35%|███▌      | 8585/24420 [3:34:06<6:32:33,  1.49s/it]
 35%|███▌      | 8586/24420 [3:34:07<6:32:33,  1.49s/it]
 35%|███▌      | 8587/24420 [3:34:09<6:32:35,  1.49s/it]
 35%|███▌      | 8588/24420 [3:34:10<6:32:23,  1.49s/it]
 35%|███▌      | 8589/24420 [3:34:12<6:32:13,  1.49s/it]
 35%|███▌      | 8590/24420 [3:34:13<6:32:34,  1.49s/it]
                                                        
{'loss': 0.1323, 'learning_rate': 5e-05, 'epoch': 1.76}

 35%|███▌      | 8590/24420 [3:34:13<6:32:34,  1.49s/it]
 35%|███▌      | 8591/24420 [3:34:15<6:32:54,  1.49s/it]
 35%|███▌      | 8592/24420 [3:34:16<6:31:45,  1.49s/it]
 35%|███▌      | 8593/24420 [3:34:18<6:31:47,  1.49s/it]
 35%|███▌      | 8594/24420 [3:34:19<6:31:40,  1.48s/it]
 35%|███▌      | 8595/24420 [3:34:21<6:31:27,  1.48s/it]
                                                        
{'loss': 0.1039, 'learning_rate': 5e-05, 'epoch': 1.76}

 35%|███▌      | 8595/24420 [3:34:21<6:31:27,  1.48s/it]
 35%|███▌      | 8596/24420 [3:34:22<6:31:19,  1.48s/it]
 35%|███▌      | 8597/24420 [3:34:24<6:31:28,  1.48s/it]
 35%|███▌      | 8598/24420 [3:34:25<6:31:21,  1.48s/it]
 35%|███▌      | 8599/24420 [3:34:27<6:31:22,  1.48s/it]
 35%|███▌      | 8600/24420 [3:34:28<6:30:44,  1.48s/it]
                                                        
{'loss': 0.0978, 'learning_rate': 5e-05, 'epoch': 1.76}

 35%|███▌      | 8600/24420 [3:34:28<6:30:44,  1.48s/it]
 35%|███▌      | 8601/24420 [3:34:33<10:40:12,  2.43s/it]
 35%|███▌      | 8602/24420 [3:34:34<9:25:19,  2.14s/it] 
 35%|███▌      | 8603/24420 [3:34:36<8:32:45,  1.95s/it]
 35%|███▌      | 8604/24420 [3:34:37<7:56:37,  1.81s/it]
 35%|███▌      | 8605/24420 [3:34:39<7:30:44,  1.71s/it]
                                                        
{'loss': 0.1305, 'learning_rate': 5e-05, 'epoch': 1.76}

 35%|███▌      | 8605/24420 [3:34:39<7:30:44,  1.71s/it]
 35%|███▌      | 8606/24420 [3:34:40<7:13:17,  1.64s/it]
 35%|███▌      | 8607/24420 [3:34:42<7:00:24,  1.60s/it]
 35%|███▌      | 8608/24420 [3:34:43<6:51:15,  1.56s/it]
 35%|███▌      | 8609/24420 [3:34:45<6:45:33,  1.54s/it]
 35%|███▌      | 8610/24420 [3:34:46<6:41:25,  1.52s/it]
                                                        
{'loss': 0.1058, 'learning_rate': 5e-05, 'epoch': 1.76}

 35%|███▌      | 8610/24420 [3:34:46<6:41:25,  1.52s/it]
 35%|███▌      | 8611/24420 [3:34:48<6:38:12,  1.51s/it]
 35%|███▌      | 8612/24420 [3:34:49<6:36:12,  1.50s/it]
 35%|███▌      | 8613/24420 [3:34:51<6:34:45,  1.50s/it]
 35%|███▌      | 8614/24420 [3:34:52<6:34:12,  1.50s/it]
 35%|███▌      | 8615/24420 [3:34:54<6:33:10,  1.49s/it]
                                                        
{'loss': 0.1102, 'learning_rate': 5e-05, 'epoch': 1.76}

 35%|███▌      | 8615/24420 [3:34:54<6:33:10,  1.49s/it]
 35%|███▌      | 8616/24420 [3:34:55<6:32:40,  1.49s/it]
 35%|███▌      | 8617/24420 [3:34:57<6:31:58,  1.49s/it]
 35%|███▌      | 8618/24420 [3:34:58<6:31:49,  1.49s/it]
 35%|███▌      | 8619/24420 [3:35:00<6:31:45,  1.49s/it]
 35%|███▌      | 8620/24420 [3:35:01<6:31:25,  1.49s/it]
                                                        
{'loss': 0.135, 'learning_rate': 5e-05, 'epoch': 1.76}

 35%|███▌      | 8620/24420 [3:35:01<6:31:25,  1.49s/it]
 35%|███▌      | 8621/24420 [3:35:03<6:31:50,  1.49s/it]
 35%|███▌      | 8622/24420 [3:35:04<6:31:01,  1.49s/it]
 35%|███▌      | 8623/24420 [3:35:06<6:30:39,  1.48s/it]
 35%|███▌      | 8624/24420 [3:35:07<6:30:16,  1.48s/it]
 35%|███▌      | 8625/24420 [3:35:08<6:29:46,  1.48s/it]
                                                        
{'loss': 0.0939, 'learning_rate': 5e-05, 'epoch': 1.77}

 35%|███▌      | 8625/24420 [3:35:08<6:29:46,  1.48s/it]
 35%|███▌      | 8626/24420 [3:35:10<6:30:27,  1.48s/it]
 35%|███▌      | 8627/24420 [3:35:11<6:30:03,  1.48s/it]
 35%|███▌      | 8628/24420 [3:35:13<6:29:47,  1.48s/it]
 35%|███▌      | 8629/24420 [3:35:14<6:29:47,  1.48s/it]
 35%|███▌      | 8630/24420 [3:35:16<6:29:24,  1.48s/it]
                                                        
{'loss': 0.0998, 'learning_rate': 5e-05, 'epoch': 1.77}

 35%|███▌      | 8630/24420 [3:35:16<6:29:24,  1.48s/it]
 35%|███▌      | 8631/24420 [3:35:17<6:29:15,  1.48s/it]
 35%|███▌      | 8632/24420 [3:35:19<6:29:53,  1.48s/it]
 35%|███▌      | 8633/24420 [3:35:20<6:29:24,  1.48s/it]
 35%|███▌      | 8634/24420 [3:35:22<6:29:17,  1.48s/it]
 35%|███▌      | 8635/24420 [3:35:23<6:29:01,  1.48s/it]
                                                        
{'loss': 0.1128, 'learning_rate': 5e-05, 'epoch': 1.77}

 35%|███▌      | 8635/24420 [3:35:23<6:29:01,  1.48s/it]
 35%|███▌      | 8636/24420 [3:35:25<6:29:39,  1.48s/it]
 35%|███▌      | 8637/24420 [3:35:26<6:29:27,  1.48s/it]
 35%|███▌      | 8638/24420 [3:35:28<6:28:47,  1.48s/it]
 35%|███▌      | 8639/24420 [3:35:29<6:30:10,  1.48s/it]
 35%|███▌      | 8640/24420 [3:35:31<6:29:15,  1.48s/it]
                                                        
{'loss': 0.1065, 'learning_rate': 5e-05, 'epoch': 1.77}

 35%|███▌      | 8640/24420 [3:35:31<6:29:15,  1.48s/it]
 35%|███▌      | 8641/24420 [3:35:32<6:29:12,  1.48s/it]
 35%|███▌      | 8642/24420 [3:35:34<6:29:04,  1.48s/it]
 35%|███▌      | 8643/24420 [3:35:35<6:29:27,  1.48s/it]
 35%|███▌      | 8644/24420 [3:35:37<6:29:28,  1.48s/it]
 35%|███▌      | 8645/24420 [3:35:38<6:29:31,  1.48s/it]
                                                        
{'loss': 0.1279, 'learning_rate': 5e-05, 'epoch': 1.77}

 35%|███▌      | 8645/24420 [3:35:38<6:29:31,  1.48s/it]
 35%|███▌      | 8646/24420 [3:35:40<6:29:01,  1.48s/it]
 35%|███▌      | 8647/24420 [3:35:41<6:28:45,  1.48s/it]
 35%|███▌      | 8648/24420 [3:35:43<6:29:36,  1.48s/it]
 35%|███▌      | 8649/24420 [3:35:44<6:30:48,  1.49s/it]
 35%|███▌      | 8650/24420 [3:35:46<6:29:34,  1.48s/it]
                                                        
{'loss': 0.0896, 'learning_rate': 5e-05, 'epoch': 1.77}

 35%|███▌      | 8650/24420 [3:35:46<6:29:34,  1.48s/it]
 35%|███▌      | 8651/24420 [3:35:47<6:29:16,  1.48s/it]
 35%|███▌      | 8652/24420 [3:35:48<6:29:56,  1.48s/it]
 35%|███▌      | 8653/24420 [3:35:50<6:29:35,  1.48s/it]
 35%|███▌      | 8654/24420 [3:35:51<6:29:06,  1.48s/it]
 35%|███▌      | 8655/24420 [3:35:53<6:28:52,  1.48s/it]
                                                        
{'loss': 0.1064, 'learning_rate': 5e-05, 'epoch': 1.77}

 35%|███▌      | 8655/24420 [3:35:53<6:28:52,  1.48s/it]
 35%|███▌      | 8656/24420 [3:35:54<6:29:20,  1.48s/it]
 35%|███▌      | 8657/24420 [3:35:56<6:29:07,  1.48s/it]
 35%|███▌      | 8658/24420 [3:35:57<6:29:00,  1.48s/it]
 35%|███▌      | 8659/24420 [3:35:59<6:29:00,  1.48s/it]
 35%|███▌      | 8660/24420 [3:36:00<6:28:41,  1.48s/it]
                                                        
{'loss': 0.1027, 'learning_rate': 5e-05, 'epoch': 1.77}

 35%|███▌      | 8660/24420 [3:36:00<6:28:41,  1.48s/it]
 35%|███▌      | 8661/24420 [3:36:02<6:28:49,  1.48s/it]
 35%|███▌      | 8662/24420 [3:36:03<6:28:59,  1.48s/it]
 35%|███▌      | 8663/24420 [3:36:05<6:29:16,  1.48s/it]
 35%|███▌      | 8664/24420 [3:36:06<6:28:47,  1.48s/it]
 35%|███▌      | 8665/24420 [3:36:08<6:28:55,  1.48s/it]
                                                        
{'loss': 0.0929, 'learning_rate': 5e-05, 'epoch': 1.77}

 35%|███▌      | 8665/24420 [3:36:08<6:28:55,  1.48s/it]
 35%|███▌      | 8666/24420 [3:36:09<6:29:37,  1.48s/it]
 35%|███▌      | 8667/24420 [3:36:11<6:28:54,  1.48s/it]
 35%|███▌      | 8668/24420 [3:36:12<6:29:21,  1.48s/it]
 35%|███▌      | 8669/24420 [3:36:14<6:28:46,  1.48s/it]
 36%|███▌      | 8670/24420 [3:36:15<6:29:21,  1.48s/it]
                                                        
{'loss': 0.124, 'learning_rate': 5e-05, 'epoch': 1.78}

 36%|███▌      | 8670/24420 [3:36:15<6:29:21,  1.48s/it]
 36%|███▌      | 8671/24420 [3:36:17<6:29:16,  1.48s/it]
 36%|███▌      | 8672/24420 [3:36:18<6:28:37,  1.48s/it]
 36%|███▌      | 8673/24420 [3:36:20<6:28:33,  1.48s/it]
 36%|███▌      | 8674/24420 [3:36:21<6:28:29,  1.48s/it]
 36%|███▌      | 8675/24420 [3:36:23<6:28:27,  1.48s/it]
                                                        
{'loss': 0.0978, 'learning_rate': 5e-05, 'epoch': 1.78}

 36%|███▌      | 8675/24420 [3:36:23<6:28:27,  1.48s/it]
 36%|███▌      | 8676/24420 [3:36:24<6:28:18,  1.48s/it]
 36%|███▌      | 8677/24420 [3:36:25<6:28:18,  1.48s/it]
 36%|███▌      | 8678/24420 [3:36:27<6:28:05,  1.48s/it]
 36%|███▌      | 8679/24420 [3:36:28<6:28:08,  1.48s/it]
 36%|███▌      | 8680/24420 [3:36:30<6:28:18,  1.48s/it]
                                                        
{'loss': 0.1037, 'learning_rate': 5e-05, 'epoch': 1.78}

 36%|███▌      | 8680/24420 [3:36:30<6:28:18,  1.48s/it]
 36%|███▌      | 8681/24420 [3:36:31<6:29:33,  1.49s/it]
 36%|███▌      | 8682/24420 [3:36:33<6:28:33,  1.48s/it]
 36%|███▌      | 8683/24420 [3:36:34<6:28:01,  1.48s/it]
 36%|███▌      | 8684/24420 [3:36:36<6:28:18,  1.48s/it]
 36%|███▌      | 8685/24420 [3:36:37<6:28:06,  1.48s/it]
                                                        
{'loss': 0.0989, 'learning_rate': 5e-05, 'epoch': 1.78}

 36%|███▌      | 8685/24420 [3:36:37<6:28:06,  1.48s/it]
 36%|███▌      | 8686/24420 [3:36:39<6:28:39,  1.48s/it]
 36%|███▌      | 8687/24420 [3:36:40<6:28:17,  1.48s/it]
 36%|███▌      | 8688/24420 [3:36:42<6:28:40,  1.48s/it]
 36%|███▌      | 8689/24420 [3:36:43<6:29:58,  1.49s/it]
 36%|███▌      | 8690/24420 [3:36:45<6:30:33,  1.49s/it]
                                                        
{'loss': 0.0953, 'learning_rate': 5e-05, 'epoch': 1.78}

 36%|███▌      | 8690/24420 [3:36:45<6:30:33,  1.49s/it]
 36%|███▌      | 8691/24420 [3:36:46<6:30:23,  1.49s/it]
 36%|███▌      | 8692/24420 [3:36:48<6:30:07,  1.49s/it]
 36%|███▌      | 8693/24420 [3:36:49<6:30:04,  1.49s/it]
 36%|███▌      | 8694/24420 [3:36:51<6:29:48,  1.49s/it]
 36%|███▌      | 8695/24420 [3:36:52<6:30:16,  1.49s/it]
                                                        
{'loss': 0.1263, 'learning_rate': 5e-05, 'epoch': 1.78}

 36%|███▌      | 8695/24420 [3:36:52<6:30:16,  1.49s/it]
 36%|███▌      | 8696/24420 [3:36:54<6:29:53,  1.49s/it]
 36%|███▌      | 8697/24420 [3:36:55<6:29:07,  1.48s/it]
 36%|███▌      | 8698/24420 [3:36:57<6:29:07,  1.49s/it]
 36%|███▌      | 8699/24420 [3:36:58<6:29:13,  1.49s/it]
 36%|███▌      | 8700/24420 [3:37:00<6:28:43,  1.48s/it]
                                                        
{'loss': 0.1116, 'learning_rate': 5e-05, 'epoch': 1.78}

 36%|███▌      | 8700/24420 [3:37:00<6:28:43,  1.48s/it]
 36%|███▌      | 8701/24420 [3:37:01<6:29:18,  1.49s/it]
 36%|███▌      | 8702/24420 [3:37:03<6:29:35,  1.49s/it]
 36%|███▌      | 8703/24420 [3:37:04<6:29:42,  1.49s/it]
 36%|███▌      | 8704/24420 [3:37:06<6:29:23,  1.49s/it]
 36%|███▌      | 8705/24420 [3:37:07<6:28:29,  1.48s/it]
                                                        
{'loss': 0.1125, 'learning_rate': 5e-05, 'epoch': 1.78}

 36%|███▌      | 8705/24420 [3:37:07<6:28:29,  1.48s/it]
 36%|███▌      | 8706/24420 [3:37:09<6:28:57,  1.49s/it]
 36%|███▌      | 8707/24420 [3:37:10<6:28:48,  1.48s/it]
 36%|███▌      | 8708/24420 [3:37:12<6:28:21,  1.48s/it]
 36%|███▌      | 8709/24420 [3:37:13<6:28:19,  1.48s/it]
 36%|███▌      | 8710/24420 [3:37:14<6:28:00,  1.48s/it]
                                                        
{'loss': 0.1073, 'learning_rate': 5e-05, 'epoch': 1.78}

 36%|███▌      | 8710/24420 [3:37:14<6:28:00,  1.48s/it]
 36%|███▌      | 8711/24420 [3:37:16<6:27:46,  1.48s/it]
 36%|███▌      | 8712/24420 [3:37:17<6:27:37,  1.48s/it]
 36%|███▌      | 8713/24420 [3:37:19<6:27:56,  1.48s/it]
 36%|███▌      | 8714/24420 [3:37:20<6:27:40,  1.48s/it]
 36%|███▌      | 8715/24420 [3:37:22<6:27:54,  1.48s/it]
                                                        
{'loss': 0.1028, 'learning_rate': 5e-05, 'epoch': 1.78}

 36%|███▌      | 8715/24420 [3:37:22<6:27:54,  1.48s/it]
 36%|███▌      | 8716/24420 [3:37:23<6:27:45,  1.48s/it]
 36%|███▌      | 8717/24420 [3:37:25<6:27:45,  1.48s/it]
 36%|███▌      | 8718/24420 [3:37:26<6:28:02,  1.48s/it]
 36%|███▌      | 8719/24420 [3:37:28<6:28:06,  1.48s/it]
 36%|███▌      | 8720/24420 [3:37:29<6:28:05,  1.48s/it]
                                                        
{'loss': 0.1181, 'learning_rate': 5e-05, 'epoch': 1.79}

 36%|███▌      | 8720/24420 [3:37:29<6:28:05,  1.48s/it]
 36%|███▌      | 8721/24420 [3:37:31<6:29:12,  1.49s/it]
 36%|███▌      | 8722/24420 [3:37:32<6:30:01,  1.49s/it]
 36%|███▌      | 8723/24420 [3:37:34<6:29:48,  1.49s/it]
 36%|███▌      | 8724/24420 [3:37:35<6:28:57,  1.49s/it]
 36%|███▌      | 8725/24420 [3:37:37<6:28:48,  1.49s/it]
                                                        
{'loss': 0.107, 'learning_rate': 5e-05, 'epoch': 1.79}

 36%|███▌      | 8725/24420 [3:37:37<6:28:48,  1.49s/it]
 36%|███▌      | 8726/24420 [3:37:38<6:28:46,  1.49s/it]
 36%|███▌      | 8727/24420 [3:37:40<6:27:45,  1.48s/it]
 36%|███▌      | 8728/24420 [3:37:41<6:27:52,  1.48s/it]
 36%|███▌      | 8729/24420 [3:37:43<6:28:05,  1.48s/it]
 36%|███▌      | 8730/24420 [3:37:44<6:27:56,  1.48s/it]
                                                        
{'loss': 0.1204, 'learning_rate': 5e-05, 'epoch': 1.79}

 36%|███▌      | 8730/24420 [3:37:44<6:27:56,  1.48s/it]
 36%|███▌      | 8731/24420 [3:37:46<6:27:38,  1.48s/it]
 36%|███▌      | 8732/24420 [3:37:47<6:29:16,  1.49s/it]
 36%|███▌      | 8733/24420 [3:37:49<6:28:42,  1.49s/it]
 36%|███▌      | 8734/24420 [3:37:50<6:28:17,  1.49s/it]
 36%|███▌      | 8735/24420 [3:37:52<6:28:05,  1.48s/it]
                                                        
{'loss': 0.1034, 'learning_rate': 5e-05, 'epoch': 1.79}

 36%|███▌      | 8735/24420 [3:37:52<6:28:05,  1.48s/it]
 36%|███▌      | 8736/24420 [3:37:53<6:28:00,  1.48s/it]
 36%|███▌      | 8737/24420 [3:37:55<6:27:24,  1.48s/it]
 36%|███▌      | 8738/24420 [3:37:56<6:28:08,  1.49s/it]
 36%|███▌      | 8739/24420 [3:37:58<6:28:06,  1.49s/it]
 36%|███▌      | 8740/24420 [3:37:59<6:27:33,  1.48s/it]
                                                        
{'loss': 0.1179, 'learning_rate': 5e-05, 'epoch': 1.79}

 36%|███▌      | 8740/24420 [3:37:59<6:27:33,  1.48s/it]
 36%|███▌      | 8741/24420 [3:38:00<6:27:21,  1.48s/it]
 36%|███▌      | 8742/24420 [3:38:02<6:27:26,  1.48s/it]
 36%|███▌      | 8743/24420 [3:38:03<6:27:37,  1.48s/it]
 36%|███▌      | 8744/24420 [3:38:05<6:27:16,  1.48s/it]
 36%|███▌      | 8745/24420 [3:38:06<6:27:09,  1.48s/it]
                                                        
{'loss': 0.1101, 'learning_rate': 5e-05, 'epoch': 1.79}

 36%|███▌      | 8745/24420 [3:38:06<6:27:09,  1.48s/it]
 36%|███▌      | 8746/24420 [3:38:08<6:26:53,  1.48s/it]
 36%|███▌      | 8747/24420 [3:38:09<6:27:21,  1.48s/it]
 36%|███▌      | 8748/24420 [3:38:11<6:27:32,  1.48s/it]
 36%|███▌      | 8749/24420 [3:38:12<6:27:43,  1.48s/it]
 36%|███▌      | 8750/24420 [3:38:14<6:28:11,  1.49s/it]
                                                        
{'loss': 0.0978, 'learning_rate': 5e-05, 'epoch': 1.79}

 36%|███▌      | 8750/24420 [3:38:14<6:28:11,  1.49s/it]
 36%|███▌      | 8751/24420 [3:38:15<6:28:28,  1.49s/it]
 36%|███▌      | 8752/24420 [3:38:17<6:28:13,  1.49s/it]
 36%|███▌      | 8753/24420 [3:38:18<6:28:43,  1.49s/it]
 36%|███▌      | 8754/24420 [3:38:20<6:27:56,  1.49s/it]
 36%|███▌      | 8755/24420 [3:38:21<6:27:34,  1.48s/it]
                                                        
{'loss': 0.1145, 'learning_rate': 5e-05, 'epoch': 1.79}

 36%|███▌      | 8755/24420 [3:38:21<6:27:34,  1.48s/it]
 36%|███▌      | 8756/24420 [3:38:23<6:27:44,  1.49s/it]
 36%|███▌      | 8757/24420 [3:38:24<6:27:40,  1.49s/it]
 36%|███▌      | 8758/24420 [3:38:26<6:27:32,  1.48s/it]
 36%|███▌      | 8759/24420 [3:38:27<6:27:12,  1.48s/it]
 36%|███▌      | 8760/24420 [3:38:29<6:27:36,  1.49s/it]
                                                        
{'loss': 0.1129, 'learning_rate': 5e-05, 'epoch': 1.79}

 36%|███▌      | 8760/24420 [3:38:29<6:27:36,  1.49s/it]
 36%|███▌      | 8761/24420 [3:38:30<6:35:51,  1.52s/it]
 36%|███▌      | 8762/24420 [3:38:32<6:33:00,  1.51s/it]
 36%|███▌      | 8763/24420 [3:38:33<6:31:23,  1.50s/it]
 36%|███▌      | 8764/24420 [3:38:35<6:31:30,  1.50s/it]
 36%|███▌      | 8765/24420 [3:38:36<6:29:15,  1.49s/it]
                                                        
{'loss': 0.1106, 'learning_rate': 5e-05, 'epoch': 1.79}

 36%|███▌      | 8765/24420 [3:38:36<6:29:15,  1.49s/it]
 36%|███▌      | 8766/24420 [3:38:38<6:28:05,  1.49s/it]
 36%|███▌      | 8767/24420 [3:38:39<6:27:39,  1.49s/it]
 36%|███▌      | 8768/24420 [3:38:41<6:26:53,  1.48s/it]
 36%|███▌      | 8769/24420 [3:38:42<6:26:49,  1.48s/it]
 36%|███▌      | 8770/24420 [3:38:44<6:26:46,  1.48s/it]
                                                        
{'loss': 0.097, 'learning_rate': 5e-05, 'epoch': 1.8}

 36%|███▌      | 8770/24420 [3:38:44<6:26:46,  1.48s/it]
 36%|███▌      | 8771/24420 [3:38:45<6:27:09,  1.48s/it]
 36%|███▌      | 8772/24420 [3:38:47<6:26:54,  1.48s/it]
 36%|███▌      | 8773/24420 [3:38:48<6:26:46,  1.48s/it]
 36%|███▌      | 8774/24420 [3:38:50<6:27:11,  1.48s/it]
 36%|███▌      | 8775/24420 [3:38:51<6:27:42,  1.49s/it]
                                                        
{'loss': 0.1184, 'learning_rate': 5e-05, 'epoch': 1.8}

 36%|███▌      | 8775/24420 [3:38:51<6:27:42,  1.49s/it]
 36%|███▌      | 8776/24420 [3:38:53<6:26:59,  1.48s/it]
 36%|███▌      | 8777/24420 [3:38:54<6:27:25,  1.49s/it]
 36%|███▌      | 8778/24420 [3:38:56<6:27:23,  1.49s/it]
 36%|███▌      | 8779/24420 [3:38:57<6:27:06,  1.48s/it]
 36%|███▌      | 8780/24420 [3:38:58<6:27:18,  1.49s/it]
                                                        
{'loss': 0.1264, 'learning_rate': 5e-05, 'epoch': 1.8}

 36%|███▌      | 8780/24420 [3:38:58<6:27:18,  1.49s/it]
 36%|███▌      | 8781/24420 [3:39:00<6:26:32,  1.48s/it]
 36%|███▌      | 8782/24420 [3:39:01<6:26:12,  1.48s/it]
 36%|███▌      | 8783/24420 [3:39:03<6:26:20,  1.48s/it]
 36%|███▌      | 8784/24420 [3:39:04<6:26:20,  1.48s/it]
 36%|███▌      | 8785/24420 [3:39:06<6:26:30,  1.48s/it]
                                                        
{'loss': 0.1122, 'learning_rate': 5e-05, 'epoch': 1.8}

 36%|███▌      | 8785/24420 [3:39:06<6:26:30,  1.48s/it]
 36%|███▌      | 8786/24420 [3:39:07<6:26:13,  1.48s/it]
 36%|███▌      | 8787/24420 [3:39:09<6:25:56,  1.48s/it]
 36%|███▌      | 8788/24420 [3:39:10<6:26:14,  1.48s/it]
 36%|███▌      | 8789/24420 [3:39:12<6:25:51,  1.48s/it]
 36%|███▌      | 8790/24420 [3:39:13<6:25:58,  1.48s/it]
                                                        
{'loss': 0.1033, 'learning_rate': 5e-05, 'epoch': 1.8}

 36%|███▌      | 8790/24420 [3:39:13<6:25:58,  1.48s/it]
 36%|███▌      | 8791/24420 [3:39:15<6:26:51,  1.49s/it]
 36%|███▌      | 8792/24420 [3:39:16<6:26:59,  1.49s/it]
 36%|███▌      | 8793/24420 [3:39:18<6:26:12,  1.48s/it]
 36%|███▌      | 8794/24420 [3:39:19<6:26:13,  1.48s/it]
 36%|███▌      | 8795/24420 [3:39:21<6:26:07,  1.48s/it]
                                                        
{'loss': 0.103, 'learning_rate': 5e-05, 'epoch': 1.8}

 36%|███▌      | 8795/24420 [3:39:21<6:26:07,  1.48s/it]
 36%|███▌      | 8796/24420 [3:39:22<6:25:50,  1.48s/it]
 36%|███▌      | 8797/24420 [3:39:24<6:26:15,  1.48s/it]
 36%|███▌      | 8798/24420 [3:39:25<6:26:23,  1.48s/it]
 36%|███▌      | 8799/24420 [3:39:27<6:26:12,  1.48s/it]
 36%|███▌      | 8800/24420 [3:39:28<6:25:46,  1.48s/it]
                                                        
{'loss': 0.1201, 'learning_rate': 5e-05, 'epoch': 1.8}

 36%|███▌      | 8800/24420 [3:39:28<6:25:46,  1.48s/it]
 36%|███▌      | 8801/24420 [3:39:33<10:48:59,  2.49s/it]
 36%|███▌      | 8802/24420 [3:39:34<9:30:03,  2.19s/it] 
 36%|███▌      | 8803/24420 [3:39:36<8:34:46,  1.98s/it]
 36%|███▌      | 8804/24420 [3:39:37<7:55:43,  1.83s/it]
 36%|███▌      | 8805/24420 [3:39:39<7:29:31,  1.73s/it]
                                                        
{'loss': 0.1039, 'learning_rate': 5e-05, 'epoch': 1.8}

 36%|███▌      | 8805/24420 [3:39:39<7:29:31,  1.73s/it]
 36%|███▌      | 8806/24420 [3:39:40<7:10:02,  1.65s/it]
 36%|███▌      | 8807/24420 [3:39:42<6:56:33,  1.60s/it]
 36%|███▌      | 8808/24420 [3:39:43<6:47:06,  1.56s/it]
 36%|███▌      | 8809/24420 [3:39:45<6:41:19,  1.54s/it]
 36%|███▌      | 8810/24420 [3:39:46<6:36:16,  1.52s/it]
                                                        
{'loss': 0.1179, 'learning_rate': 5e-05, 'epoch': 1.8}

 36%|███▌      | 8810/24420 [3:39:46<6:36:16,  1.52s/it]
 36%|███▌      | 8811/24420 [3:39:48<6:33:26,  1.51s/it]
 36%|███▌      | 8812/24420 [3:39:49<6:31:18,  1.50s/it]
 36%|███▌      | 8813/24420 [3:39:51<6:29:33,  1.50s/it]
 36%|███▌      | 8814/24420 [3:39:52<6:28:17,  1.49s/it]
 36%|███▌      | 8815/24420 [3:39:54<6:27:17,  1.49s/it]
                                                        
{'loss': 0.1305, 'learning_rate': 5e-05, 'epoch': 1.8}

 36%|███▌      | 8815/24420 [3:39:54<6:27:17,  1.49s/it]
 36%|███▌      | 8816/24420 [3:39:55<6:26:32,  1.49s/it]
 36%|███▌      | 8817/24420 [3:39:57<6:25:59,  1.48s/it]
 36%|███▌      | 8818/24420 [3:39:58<6:25:45,  1.48s/it]
 36%|███▌      | 8819/24420 [3:40:00<6:26:03,  1.48s/it]
 36%|███▌      | 8820/24420 [3:40:01<6:26:25,  1.49s/it]
                                                        
{'loss': 0.1136, 'learning_rate': 5e-05, 'epoch': 1.81}

 36%|███▌      | 8820/24420 [3:40:01<6:26:25,  1.49s/it]
 36%|███▌      | 8821/24420 [3:40:03<6:26:02,  1.48s/it]
 36%|███▌      | 8822/24420 [3:40:04<6:25:56,  1.48s/it]
 36%|███▌      | 8823/24420 [3:40:06<6:25:37,  1.48s/it]
 36%|███▌      | 8824/24420 [3:40:07<6:25:41,  1.48s/it]
 36%|███▌      | 8825/24420 [3:40:09<6:25:36,  1.48s/it]
                                                        
{'loss': 0.1091, 'learning_rate': 5e-05, 'epoch': 1.81}

 36%|███▌      | 8825/24420 [3:40:09<6:25:36,  1.48s/it]
 36%|███▌      | 8826/24420 [3:40:10<6:24:50,  1.48s/it]
 36%|███▌      | 8827/24420 [3:40:12<6:24:56,  1.48s/it]
 36%|███▌      | 8828/24420 [3:40:13<6:25:03,  1.48s/it]
 36%|███▌      | 8829/24420 [3:40:15<6:24:53,  1.48s/it]
 36%|███▌      | 8830/24420 [3:40:16<6:25:15,  1.48s/it]
                                                        
{'loss': 0.132, 'learning_rate': 5e-05, 'epoch': 1.81}

 36%|███▌      | 8830/24420 [3:40:16<6:25:15,  1.48s/it]
 36%|███▌      | 8831/24420 [3:40:17<6:25:04,  1.48s/it]
 36%|███▌      | 8832/24420 [3:40:19<6:25:40,  1.48s/it]
 36%|███▌      | 8833/24420 [3:40:20<6:25:45,  1.48s/it]
 36%|███▌      | 8834/24420 [3:40:22<6:25:28,  1.48s/it]
 36%|███▌      | 8835/24420 [3:40:23<6:25:06,  1.48s/it]
                                                        
{'loss': 0.1035, 'learning_rate': 5e-05, 'epoch': 1.81}

 36%|███▌      | 8835/24420 [3:40:23<6:25:06,  1.48s/it]
 36%|███▌      | 8836/24420 [3:40:25<6:25:07,  1.48s/it]
 36%|███▌      | 8837/24420 [3:40:26<6:24:34,  1.48s/it]
 36%|███▌      | 8838/24420 [3:40:28<6:25:23,  1.48s/it]
 36%|███▌      | 8839/24420 [3:40:29<6:24:50,  1.48s/it]
 36%|███▌      | 8840/24420 [3:40:31<6:24:45,  1.48s/it]
                                                        
{'loss': 0.102, 'learning_rate': 5e-05, 'epoch': 1.81}

 36%|███▌      | 8840/24420 [3:40:31<6:24:45,  1.48s/it]
 36%|███▌      | 8841/24420 [3:40:32<6:25:17,  1.48s/it]
 36%|███▌      | 8842/24420 [3:40:34<6:25:45,  1.49s/it]
 36%|███▌      | 8843/24420 [3:40:35<6:25:00,  1.48s/it]
 36%|███▌      | 8844/24420 [3:40:37<6:24:53,  1.48s/it]
 36%|███▌      | 8845/24420 [3:40:38<6:24:34,  1.48s/it]
                                                        
{'loss': 0.1024, 'learning_rate': 5e-05, 'epoch': 1.81}

 36%|███▌      | 8845/24420 [3:40:38<6:24:34,  1.48s/it]
 36%|███▌      | 8846/24420 [3:40:40<6:25:01,  1.48s/it]
 36%|███▌      | 8847/24420 [3:40:41<6:24:20,  1.48s/it]
 36%|███▌      | 8848/24420 [3:40:43<6:24:06,  1.48s/it]
 36%|███▌      | 8849/24420 [3:40:44<6:24:36,  1.48s/it]
 36%|███▌      | 8850/24420 [3:40:46<6:23:57,  1.48s/it]
                                                        
{'loss': 0.0997, 'learning_rate': 5e-05, 'epoch': 1.81}

 36%|███▌      | 8850/24420 [3:40:46<6:23:57,  1.48s/it]
 36%|███▌      | 8851/24420 [3:40:47<6:24:03,  1.48s/it]
 36%|███▌      | 8852/24420 [3:40:49<6:24:14,  1.48s/it]
 36%|███▋      | 8853/24420 [3:40:50<6:24:38,  1.48s/it]
 36%|███▋      | 8854/24420 [3:40:52<6:24:46,  1.48s/it]
 36%|███▋      | 8855/24420 [3:40:53<6:24:38,  1.48s/it]
                                                        
{'loss': 0.1228, 'learning_rate': 5e-05, 'epoch': 1.81}

 36%|███▋      | 8855/24420 [3:40:53<6:24:38,  1.48s/it]
 36%|███▋      | 8856/24420 [3:40:55<6:24:11,  1.48s/it]
 36%|███▋      | 8857/24420 [3:40:56<6:23:58,  1.48s/it]
 36%|███▋      | 8858/24420 [3:40:57<6:24:27,  1.48s/it]
 36%|███▋      | 8859/24420 [3:40:59<6:24:47,  1.48s/it]
 36%|███▋      | 8860/24420 [3:41:00<6:25:47,  1.49s/it]
                                                        
{'loss': 0.1051, 'learning_rate': 5e-05, 'epoch': 1.81}

 36%|███▋      | 8860/24420 [3:41:00<6:25:47,  1.49s/it]
 36%|███▋      | 8861/24420 [3:41:02<6:25:32,  1.49s/it]
 36%|███▋      | 8862/24420 [3:41:03<6:24:36,  1.48s/it]
 36%|███▋      | 8863/24420 [3:41:05<6:24:31,  1.48s/it]
 36%|███▋      | 8864/24420 [3:41:06<6:25:34,  1.49s/it]
 36%|███▋      | 8865/24420 [3:41:08<6:25:15,  1.49s/it]
                                                        
{'loss': 0.1113, 'learning_rate': 5e-05, 'epoch': 1.82}

 36%|███▋      | 8865/24420 [3:41:08<6:25:15,  1.49s/it]
 36%|███▋      | 8866/24420 [3:41:09<6:25:18,  1.49s/it]
 36%|███▋      | 8867/24420 [3:41:11<6:25:28,  1.49s/it]
 36%|███▋      | 8868/24420 [3:41:12<6:25:47,  1.49s/it]
 36%|███▋      | 8869/24420 [3:41:14<6:25:05,  1.49s/it]
 36%|███▋      | 8870/24420 [3:41:15<6:24:31,  1.48s/it]
                                                        
{'loss': 0.1221, 'learning_rate': 5e-05, 'epoch': 1.82}

 36%|███▋      | 8870/24420 [3:41:15<6:24:31,  1.48s/it]
 36%|███▋      | 8871/24420 [3:41:17<6:24:32,  1.48s/it]
 36%|███▋      | 8872/24420 [3:41:18<6:24:58,  1.49s/it]
 36%|███▋      | 8873/24420 [3:41:20<6:24:57,  1.49s/it]
 36%|███▋      | 8874/24420 [3:41:21<6:24:54,  1.49s/it]
 36%|███▋      | 8875/24420 [3:41:23<6:24:50,  1.49s/it]
                                                        
{'loss': 0.1271, 'learning_rate': 5e-05, 'epoch': 1.82}

 36%|███▋      | 8875/24420 [3:41:23<6:24:50,  1.49s/it]
 36%|███▋      | 8876/24420 [3:41:24<6:24:40,  1.48s/it]
 36%|███▋      | 8877/24420 [3:41:26<6:25:01,  1.49s/it]
 36%|███▋      | 8878/24420 [3:41:27<6:24:19,  1.48s/it]
 36%|███▋      | 8879/24420 [3:41:29<6:23:52,  1.48s/it]
 36%|███▋      | 8880/24420 [3:41:30<6:23:52,  1.48s/it]
                                                        
{'loss': 0.1164, 'learning_rate': 5e-05, 'epoch': 1.82}

 36%|███▋      | 8880/24420 [3:41:30<6:23:52,  1.48s/it]
 36%|███▋      | 8881/24420 [3:41:32<6:24:45,  1.49s/it]
 36%|███▋      | 8882/24420 [3:41:33<6:24:03,  1.48s/it]
 36%|███▋      | 8883/24420 [3:41:35<6:23:47,  1.48s/it]
 36%|███▋      | 8884/24420 [3:41:36<6:23:28,  1.48s/it]
 36%|███▋      | 8885/24420 [3:41:38<6:24:03,  1.48s/it]
                                                        
{'loss': 0.1126, 'learning_rate': 5e-05, 'epoch': 1.82}

 36%|███▋      | 8885/24420 [3:41:38<6:24:03,  1.48s/it]
 36%|███▋      | 8886/24420 [3:41:39<6:24:02,  1.48s/it]
 36%|███▋      | 8887/24420 [3:41:41<6:23:32,  1.48s/it]
 36%|███▋      | 8888/24420 [3:41:42<6:24:16,  1.48s/it]
 36%|███▋      | 8889/24420 [3:41:44<6:24:31,  1.49s/it]
 36%|███▋      | 8890/24420 [3:41:45<6:24:36,  1.49s/it]
                                                        
{'loss': 0.1131, 'learning_rate': 5e-05, 'epoch': 1.82}

 36%|███▋      | 8890/24420 [3:41:45<6:24:36,  1.49s/it]
 36%|███▋      | 8891/24420 [3:41:47<6:25:18,  1.49s/it]
 36%|███▋      | 8892/24420 [3:41:48<6:25:24,  1.49s/it]
 36%|███▋      | 8893/24420 [3:41:49<6:24:50,  1.49s/it]
 36%|███▋      | 8894/24420 [3:41:51<6:24:43,  1.49s/it]
 36%|███▋      | 8895/24420 [3:41:52<6:24:34,  1.49s/it]
                                                        
{'loss': 0.113, 'learning_rate': 5e-05, 'epoch': 1.82}

 36%|███▋      | 8895/24420 [3:41:52<6:24:34,  1.49s/it]
 36%|███▋      | 8896/24420 [3:41:54<6:24:11,  1.48s/it]
 36%|███▋      | 8897/24420 [3:41:55<6:23:59,  1.48s/it]
 36%|███▋      | 8898/24420 [3:41:57<6:23:29,  1.48s/it]
 36%|███▋      | 8899/24420 [3:41:58<6:23:35,  1.48s/it]
 36%|███▋      | 8900/24420 [3:42:00<6:24:23,  1.49s/it]
                                                        
{'loss': 0.1002, 'learning_rate': 5e-05, 'epoch': 1.82}

 36%|███▋      | 8900/24420 [3:42:00<6:24:23,  1.49s/it]
 36%|███▋      | 8901/24420 [3:42:01<6:24:48,  1.49s/it]
 36%|███▋      | 8902/24420 [3:42:03<6:24:44,  1.49s/it]
 36%|███▋      | 8903/24420 [3:42:04<6:24:46,  1.49s/it]
 36%|███▋      | 8904/24420 [3:42:06<6:24:29,  1.49s/it]
 36%|███▋      | 8905/24420 [3:42:07<6:24:40,  1.49s/it]
                                                        
{'loss': 0.1238, 'learning_rate': 5e-05, 'epoch': 1.82}

 36%|███▋      | 8905/24420 [3:42:07<6:24:40,  1.49s/it]
 36%|███▋      | 8906/24420 [3:42:09<6:25:09,  1.49s/it]
 36%|███▋      | 8907/24420 [3:42:10<6:25:17,  1.49s/it]
 36%|███▋      | 8908/24420 [3:42:12<6:27:28,  1.50s/it]
 36%|███▋      | 8909/24420 [3:42:13<6:26:36,  1.50s/it]
 36%|███▋      | 8910/24420 [3:42:15<6:25:54,  1.49s/it]
                                                        
{'loss': 0.0992, 'learning_rate': 5e-05, 'epoch': 1.82}

 36%|███▋      | 8910/24420 [3:42:15<6:25:54,  1.49s/it]
 36%|███▋      | 8911/24420 [3:42:16<6:26:10,  1.49s/it]
 36%|███▋      | 8912/24420 [3:42:18<6:25:47,  1.49s/it]
 36%|███▋      | 8913/24420 [3:42:19<6:24:50,  1.49s/it]
 37%|███▋      | 8914/24420 [3:42:21<6:24:24,  1.49s/it]
 37%|███▋      | 8915/24420 [3:42:22<6:24:11,  1.49s/it]
                                                        
{'loss': 0.1228, 'learning_rate': 5e-05, 'epoch': 1.83}

 37%|███▋      | 8915/24420 [3:42:22<6:24:11,  1.49s/it]
 37%|███▋      | 8916/24420 [3:42:24<6:24:25,  1.49s/it]
 37%|███▋      | 8917/24420 [3:42:25<6:23:56,  1.49s/it]
 37%|███▋      | 8918/24420 [3:42:27<6:24:45,  1.49s/it]
 37%|███▋      | 8919/24420 [3:42:28<6:24:47,  1.49s/it]
 37%|███▋      | 8920/24420 [3:42:30<6:23:41,  1.49s/it]
                                                        
{'loss': 0.1032, 'learning_rate': 5e-05, 'epoch': 1.83}

 37%|███▋      | 8920/24420 [3:42:30<6:23:41,  1.49s/it]
 37%|███▋      | 8921/24420 [3:42:31<6:23:26,  1.48s/it]
 37%|███▋      | 8922/24420 [3:42:33<6:23:22,  1.48s/it]
 37%|███▋      | 8923/24420 [3:42:34<6:23:39,  1.49s/it]
 37%|███▋      | 8924/24420 [3:42:36<6:23:55,  1.49s/it]
 37%|███▋      | 8925/24420 [3:42:37<6:25:09,  1.49s/it]
                                                        
{'loss': 0.115, 'learning_rate': 5e-05, 'epoch': 1.83}

 37%|███▋      | 8925/24420 [3:42:37<6:25:09,  1.49s/it]
 37%|███▋      | 8926/24420 [3:42:39<6:25:05,  1.49s/it]
 37%|███▋      | 8927/24420 [3:42:40<6:24:19,  1.49s/it]
 37%|███▋      | 8928/24420 [3:42:42<6:24:22,  1.49s/it]
 37%|███▋      | 8929/24420 [3:42:43<6:24:26,  1.49s/it]
 37%|███▋      | 8930/24420 [3:42:45<6:24:50,  1.49s/it]
                                                        
{'loss': 0.1056, 'learning_rate': 5e-05, 'epoch': 1.83}

 37%|███▋      | 8930/24420 [3:42:45<6:24:50,  1.49s/it]
 37%|███▋      | 8931/24420 [3:42:46<6:24:41,  1.49s/it]
 37%|███▋      | 8932/24420 [3:42:48<6:24:32,  1.49s/it]
 37%|███▋      | 8933/24420 [3:42:49<6:24:31,  1.49s/it]
 37%|███▋      | 8934/24420 [3:42:51<6:24:43,  1.49s/it]
 37%|███▋      | 8935/24420 [3:42:52<6:24:47,  1.49s/it]
                                                        
{'loss': 0.1028, 'learning_rate': 5e-05, 'epoch': 1.83}

 37%|███▋      | 8935/24420 [3:42:52<6:24:47,  1.49s/it]
 37%|███▋      | 8936/24420 [3:42:53<6:24:19,  1.49s/it]
 37%|███▋      | 8937/24420 [3:42:55<6:23:18,  1.49s/it]
 37%|███▋      | 8938/24420 [3:42:56<6:23:46,  1.49s/it]
 37%|███▋      | 8939/24420 [3:42:58<6:23:41,  1.49s/it]
 37%|███▋      | 8940/24420 [3:42:59<6:23:09,  1.49s/it]
                                                        
{'loss': 0.1054, 'learning_rate': 5e-05, 'epoch': 1.83}

 37%|███▋      | 8940/24420 [3:42:59<6:23:09,  1.49s/it]
 37%|███▋      | 8941/24420 [3:43:01<6:22:47,  1.48s/it]
 37%|███▋      | 8942/24420 [3:43:02<6:22:52,  1.48s/it]
 37%|███▋      | 8943/24420 [3:43:04<6:23:09,  1.49s/it]
 37%|███▋      | 8944/24420 [3:43:05<6:22:36,  1.48s/it]
 37%|███▋      | 8945/24420 [3:43:07<6:22:40,  1.48s/it]
                                                        
{'loss': 0.1134, 'learning_rate': 5e-05, 'epoch': 1.83}

 37%|███▋      | 8945/24420 [3:43:07<6:22:40,  1.48s/it]
 37%|███▋      | 8946/24420 [3:43:08<6:22:47,  1.48s/it]
 37%|███▋      | 8947/24420 [3:43:10<6:22:28,  1.48s/it]
 37%|███▋      | 8948/24420 [3:43:11<6:22:47,  1.48s/it]
 37%|███▋      | 8949/24420 [3:43:13<6:22:36,  1.48s/it]
 37%|███▋      | 8950/24420 [3:43:14<6:22:10,  1.48s/it]
                                                        
{'loss': 0.1176, 'learning_rate': 5e-05, 'epoch': 1.83}

 37%|███▋      | 8950/24420 [3:43:14<6:22:10,  1.48s/it]
 37%|███▋      | 8951/24420 [3:43:16<6:22:35,  1.48s/it]
 37%|███▋      | 8952/24420 [3:43:17<6:23:03,  1.49s/it]
 37%|███▋      | 8953/24420 [3:43:19<6:23:01,  1.49s/it]
 37%|███▋      | 8954/24420 [3:43:20<6:23:06,  1.49s/it]
 37%|███▋      | 8955/24420 [3:43:22<6:23:33,  1.49s/it]
                                                        
{'loss': 0.1046, 'learning_rate': 5e-05, 'epoch': 1.83}

 37%|███▋      | 8955/24420 [3:43:22<6:23:33,  1.49s/it]
 37%|███▋      | 8956/24420 [3:43:23<6:24:15,  1.49s/it]
 37%|███▋      | 8957/24420 [3:43:25<6:23:34,  1.49s/it]
 37%|███▋      | 8958/24420 [3:43:26<6:25:23,  1.50s/it]
 37%|███▋      | 8959/24420 [3:43:28<6:25:06,  1.49s/it]
 37%|███▋      | 8960/24420 [3:43:29<6:25:15,  1.50s/it]
                                                        
{'loss': 0.0864, 'learning_rate': 5e-05, 'epoch': 1.83}

 37%|███▋      | 8960/24420 [3:43:29<6:25:15,  1.50s/it]
 37%|███▋      | 8961/24420 [3:43:31<6:23:59,  1.49s/it]
 37%|███▋      | 8962/24420 [3:43:32<6:24:05,  1.49s/it]
 37%|███▋      | 8963/24420 [3:43:34<6:24:12,  1.49s/it]
 37%|███▋      | 8964/24420 [3:43:35<6:23:15,  1.49s/it]
 37%|███▋      | 8965/24420 [3:43:37<6:22:25,  1.48s/it]
                                                        
{'loss': 0.0971, 'learning_rate': 5e-05, 'epoch': 1.84}

 37%|███▋      | 8965/24420 [3:43:37<6:22:25,  1.48s/it]
 37%|███▋      | 8966/24420 [3:43:38<6:23:05,  1.49s/it]
 37%|███▋      | 8967/24420 [3:43:40<6:22:25,  1.48s/it]
 37%|███▋      | 8968/24420 [3:43:41<6:22:51,  1.49s/it]
 37%|███▋      | 8969/24420 [3:43:43<6:22:09,  1.48s/it]
 37%|███▋      | 8970/24420 [3:43:44<6:22:02,  1.48s/it]
                                                        
{'loss': 0.0952, 'learning_rate': 5e-05, 'epoch': 1.84}

 37%|███▋      | 8970/24420 [3:43:44<6:22:02,  1.48s/it]
 37%|███▋      | 8971/24420 [3:43:46<6:21:48,  1.48s/it]
 37%|███▋      | 8972/24420 [3:43:47<6:22:32,  1.49s/it]
 37%|███▋      | 8973/24420 [3:43:48<6:22:13,  1.48s/it]
 37%|███▋      | 8974/24420 [3:43:50<6:22:46,  1.49s/it]
 37%|███▋      | 8975/24420 [3:43:51<6:22:45,  1.49s/it]
                                                        
{'loss': 0.1198, 'learning_rate': 5e-05, 'epoch': 1.84}

 37%|███▋      | 8975/24420 [3:43:51<6:22:45,  1.49s/it]
 37%|███▋      | 8976/24420 [3:43:53<6:22:22,  1.49s/it]
 37%|███▋      | 8977/24420 [3:43:54<6:22:09,  1.48s/it]
 37%|███▋      | 8978/24420 [3:43:56<6:21:40,  1.48s/it]
 37%|███▋      | 8979/24420 [3:43:57<6:21:26,  1.48s/it]
 37%|███▋      | 8980/24420 [3:43:59<6:21:27,  1.48s/it]
                                                        
{'loss': 0.1116, 'learning_rate': 5e-05, 'epoch': 1.84}

 37%|███▋      | 8980/24420 [3:43:59<6:21:27,  1.48s/it]
 37%|███▋      | 8981/24420 [3:44:00<6:21:33,  1.48s/it]
 37%|███▋      | 8982/24420 [3:44:02<6:21:46,  1.48s/it]
 37%|███▋      | 8983/24420 [3:44:03<6:21:50,  1.48s/it]
 37%|███▋      | 8984/24420 [3:44:05<6:21:55,  1.48s/it]
 37%|███▋      | 8985/24420 [3:44:06<6:21:53,  1.48s/it]
                                                        
{'loss': 0.1086, 'learning_rate': 5e-05, 'epoch': 1.84}

 37%|███▋      | 8985/24420 [3:44:06<6:21:53,  1.48s/it]
 37%|███▋      | 8986/24420 [3:44:08<6:22:02,  1.49s/it]
 37%|███▋      | 8987/24420 [3:44:09<6:22:06,  1.49s/it]
 37%|███▋      | 8988/24420 [3:44:11<6:22:02,  1.49s/it]
 37%|███▋      | 8989/24420 [3:44:12<6:22:33,  1.49s/it]
 37%|███▋      | 8990/24420 [3:44:14<6:22:11,  1.49s/it]
                                                        
{'loss': 0.0913, 'learning_rate': 5e-05, 'epoch': 1.84}

 37%|███▋      | 8990/24420 [3:44:14<6:22:11,  1.49s/it]
 37%|███▋      | 8991/24420 [3:44:15<6:21:54,  1.49s/it]
 37%|███▋      | 8992/24420 [3:44:17<6:22:18,  1.49s/it]
 37%|███▋      | 8993/24420 [3:44:18<6:22:31,  1.49s/it]
 37%|███▋      | 8994/24420 [3:44:20<6:22:11,  1.49s/it]
 37%|███▋      | 8995/24420 [3:44:21<6:22:28,  1.49s/it]
                                                        
{'loss': 0.1039, 'learning_rate': 5e-05, 'epoch': 1.84}

 37%|███▋      | 8995/24420 [3:44:21<6:22:28,  1.49s/it]
 37%|███▋      | 8996/24420 [3:44:23<6:22:54,  1.49s/it]
 37%|███▋      | 8997/24420 [3:44:24<6:22:12,  1.49s/it]
 37%|███▋      | 8998/24420 [3:44:26<6:21:59,  1.49s/it]
 37%|███▋      | 8999/24420 [3:44:27<6:22:05,  1.49s/it]
 37%|███▋      | 9000/24420 [3:44:29<6:21:48,  1.49s/it]
                                                        
{'loss': 0.1151, 'learning_rate': 5e-05, 'epoch': 1.84}

 37%|███▋      | 9000/24420 [3:44:29<6:21:48,  1.49s/it]
 37%|███▋      | 9001/24420 [3:44:33<10:18:18,  2.41s/it]
 37%|███▋      | 9002/24420 [3:44:35<9:07:05,  2.13s/it] 
 37%|███▋      | 9003/24420 [3:44:36<8:17:46,  1.94s/it]
 37%|███▋      | 9004/24420 [3:44:38<7:42:47,  1.80s/it]
 37%|███▋      | 9005/24420 [3:44:39<7:19:33,  1.71s/it]
                                                        
{'loss': 0.1108, 'learning_rate': 5e-05, 'epoch': 1.84}

 37%|███▋      | 9005/24420 [3:44:39<7:19:33,  1.71s/it]
 37%|███▋      | 9006/24420 [3:44:41<7:02:34,  1.64s/it]
 37%|███▋      | 9007/24420 [3:44:42<6:50:03,  1.60s/it]
 37%|███▋      | 9008/24420 [3:44:44<6:42:10,  1.57s/it]
 37%|███▋      | 9009/24420 [3:44:45<6:35:57,  1.54s/it]
 37%|███▋      | 9010/24420 [3:44:47<6:31:21,  1.52s/it]
                                                        
{'loss': 0.0897, 'learning_rate': 5e-05, 'epoch': 1.84}

 37%|███▋      | 9010/24420 [3:44:47<6:31:21,  1.52s/it]
 37%|███▋      | 9011/24420 [3:44:48<6:28:20,  1.51s/it]
 37%|███▋      | 9012/24420 [3:44:50<6:25:52,  1.50s/it]
 37%|███▋      | 9013/24420 [3:44:51<6:24:27,  1.50s/it]
 37%|███▋      | 9014/24420 [3:44:52<6:23:12,  1.49s/it]
 37%|███▋      | 9015/24420 [3:44:54<6:22:16,  1.49s/it]
                                                        
{'loss': 0.0918, 'learning_rate': 5e-05, 'epoch': 1.85}

 37%|███▋      | 9015/24420 [3:44:54<6:22:16,  1.49s/it]
 37%|███▋      | 9016/24420 [3:44:55<6:21:33,  1.49s/it]
 37%|███▋      | 9017/24420 [3:44:57<6:21:12,  1.48s/it]
 37%|███▋      | 9018/24420 [3:44:58<6:20:40,  1.48s/it]
 37%|███▋      | 9019/24420 [3:45:00<6:21:17,  1.49s/it]
 37%|███▋      | 9020/24420 [3:45:01<6:20:56,  1.48s/it]
                                                        
{'loss': 0.093, 'learning_rate': 5e-05, 'epoch': 1.85}

 37%|███▋      | 9020/24420 [3:45:01<6:20:56,  1.48s/it]
 37%|███▋      | 9021/24420 [3:45:03<6:20:40,  1.48s/it]
 37%|███▋      | 9022/24420 [3:45:04<6:20:26,  1.48s/it]
 37%|███▋      | 9023/24420 [3:45:06<6:20:07,  1.48s/it]
 37%|███▋      | 9024/24420 [3:45:07<6:20:09,  1.48s/it]
 37%|███▋      | 9025/24420 [3:45:09<6:20:06,  1.48s/it]
                                                        
{'loss': 0.0881, 'learning_rate': 5e-05, 'epoch': 1.85}

 37%|███▋      | 9025/24420 [3:45:09<6:20:06,  1.48s/it]
 37%|███▋      | 9026/24420 [3:45:10<6:20:01,  1.48s/it]
 37%|███▋      | 9027/24420 [3:45:12<6:19:57,  1.48s/it]
 37%|███▋      | 9028/24420 [3:45:13<6:19:50,  1.48s/it]
 37%|███▋      | 9029/24420 [3:45:15<6:19:59,  1.48s/it]
 37%|███▋      | 9030/24420 [3:45:16<6:20:04,  1.48s/it]
                                                        
{'loss': 0.1268, 'learning_rate': 5e-05, 'epoch': 1.85}

 37%|███▋      | 9030/24420 [3:45:16<6:20:04,  1.48s/it]
 37%|███▋      | 9031/24420 [3:45:18<6:20:39,  1.48s/it]
 37%|███▋      | 9032/24420 [3:45:19<6:20:34,  1.48s/it]
 37%|███▋      | 9033/24420 [3:45:21<6:20:02,  1.48s/it]
 37%|███▋      | 9034/24420 [3:45:22<6:20:10,  1.48s/it]
 37%|███▋      | 9035/24420 [3:45:24<6:20:13,  1.48s/it]
                                                        
{'loss': 0.1002, 'learning_rate': 5e-05, 'epoch': 1.85}

 37%|███▋      | 9035/24420 [3:45:24<6:20:13,  1.48s/it]
 37%|███▋      | 9036/24420 [3:45:25<6:20:39,  1.48s/it]
 37%|███▋      | 9037/24420 [3:45:27<6:20:53,  1.49s/it]
 37%|███▋      | 9038/24420 [3:45:28<6:21:35,  1.49s/it]
 37%|███▋      | 9039/24420 [3:45:30<6:22:10,  1.49s/it]
 37%|███▋      | 9040/24420 [3:45:31<6:22:41,  1.49s/it]
                                                        
{'loss': 0.1341, 'learning_rate': 5e-05, 'epoch': 1.85}

 37%|███▋      | 9040/24420 [3:45:31<6:22:41,  1.49s/it]
 37%|███▋      | 9041/24420 [3:45:33<6:21:43,  1.49s/it]
 37%|███▋      | 9042/24420 [3:45:34<6:21:33,  1.49s/it]
 37%|███▋      | 9043/24420 [3:45:36<6:21:35,  1.49s/it]
 37%|███▋      | 9044/24420 [3:45:37<6:21:15,  1.49s/it]
 37%|███▋      | 9045/24420 [3:45:38<6:21:18,  1.49s/it]
                                                        
{'loss': 0.0968, 'learning_rate': 5e-05, 'epoch': 1.85}

 37%|███▋      | 9045/24420 [3:45:38<6:21:18,  1.49s/it]
 37%|███▋      | 9046/24420 [3:45:40<6:20:38,  1.49s/it]
 37%|███▋      | 9047/24420 [3:45:41<6:20:45,  1.49s/it]
 37%|███▋      | 9048/24420 [3:45:43<6:19:58,  1.48s/it]
 37%|███▋      | 9049/24420 [3:45:44<6:19:32,  1.48s/it]
 37%|███▋      | 9050/24420 [3:45:46<6:19:45,  1.48s/it]
                                                        
{'loss': 0.0962, 'learning_rate': 5e-05, 'epoch': 1.85}

 37%|███▋      | 9050/24420 [3:45:46<6:19:45,  1.48s/it]
 37%|███▋      | 9051/24420 [3:45:47<6:19:28,  1.48s/it]
 37%|███▋      | 9052/24420 [3:45:49<6:19:47,  1.48s/it]
 37%|███▋      | 9053/24420 [3:45:50<6:19:57,  1.48s/it]
 37%|███▋      | 9054/24420 [3:45:52<6:19:48,  1.48s/it]
 37%|███▋      | 9055/24420 [3:45:53<6:19:35,  1.48s/it]
                                                        
{'loss': 0.1101, 'learning_rate': 5e-05, 'epoch': 1.85}

 37%|███▋      | 9055/24420 [3:45:53<6:19:35,  1.48s/it]
 37%|███▋      | 9056/24420 [3:45:55<6:19:28,  1.48s/it]
 37%|███▋      | 9057/24420 [3:45:56<6:19:54,  1.48s/it]
 37%|███▋      | 9058/24420 [3:45:58<6:19:07,  1.48s/it]
 37%|███▋      | 9059/24420 [3:45:59<6:19:25,  1.48s/it]
 37%|███▋      | 9060/24420 [3:46:01<6:19:45,  1.48s/it]
                                                        
{'loss': 0.1161, 'learning_rate': 5e-05, 'epoch': 1.86}

 37%|███▋      | 9060/24420 [3:46:01<6:19:45,  1.48s/it]
 37%|███▋      | 9061/24420 [3:46:02<6:19:51,  1.48s/it]
 37%|███▋      | 9062/24420 [3:46:04<6:19:23,  1.48s/it]
 37%|███▋      | 9063/24420 [3:46:05<6:19:39,  1.48s/it]
 37%|███▋      | 9064/24420 [3:46:07<6:19:26,  1.48s/it]
 37%|███▋      | 9065/24420 [3:46:08<6:19:58,  1.48s/it]
                                                        
{'loss': 0.1024, 'learning_rate': 5e-05, 'epoch': 1.86}

 37%|███▋      | 9065/24420 [3:46:08<6:19:58,  1.48s/it]
 37%|███▋      | 9066/24420 [3:46:10<6:20:32,  1.49s/it]
 37%|███▋      | 9067/24420 [3:46:11<6:21:05,  1.49s/it]
 37%|███▋      | 9068/24420 [3:46:13<6:20:38,  1.49s/it]
 37%|███▋      | 9069/24420 [3:46:14<6:20:50,  1.49s/it]
 37%|███▋      | 9070/24420 [3:46:16<6:21:05,  1.49s/it]
                                                        
{'loss': 0.1051, 'learning_rate': 5e-05, 'epoch': 1.86}

 37%|███▋      | 9070/24420 [3:46:16<6:21:05,  1.49s/it]
 37%|███▋      | 9071/24420 [3:46:17<6:20:53,  1.49s/it]
 37%|███▋      | 9072/24420 [3:46:19<6:20:34,  1.49s/it]
 37%|███▋      | 9073/24420 [3:46:20<6:20:31,  1.49s/it]
 37%|███▋      | 9074/24420 [3:46:22<6:20:27,  1.49s/it]
 37%|███▋      | 9075/24420 [3:46:23<6:20:25,  1.49s/it]
                                                        
{'loss': 0.0939, 'learning_rate': 5e-05, 'epoch': 1.86}

 37%|███▋      | 9075/24420 [3:46:23<6:20:25,  1.49s/it]
 37%|███▋      | 9076/24420 [3:46:25<6:20:27,  1.49s/it]
 37%|███▋      | 9077/24420 [3:46:26<6:20:08,  1.49s/it]
 37%|███▋      | 9078/24420 [3:46:28<6:24:51,  1.51s/it]
 37%|███▋      | 9079/24420 [3:46:29<6:22:43,  1.50s/it]
 37%|███▋      | 9080/24420 [3:46:31<6:22:07,  1.49s/it]
                                                        
{'loss': 0.1108, 'learning_rate': 5e-05, 'epoch': 1.86}

 37%|███▋      | 9080/24420 [3:46:31<6:22:07,  1.49s/it]
 37%|███▋      | 9081/24420 [3:46:32<6:21:41,  1.49s/it]
 37%|███▋      | 9082/24420 [3:46:33<6:21:12,  1.49s/it]
 37%|███▋      | 9083/24420 [3:46:35<6:20:17,  1.49s/it]
 37%|███▋      | 9084/24420 [3:46:36<6:20:23,  1.49s/it]
 37%|███▋      | 9085/24420 [3:46:38<6:19:36,  1.49s/it]
                                                        
{'loss': 0.1136, 'learning_rate': 5e-05, 'epoch': 1.86}

 37%|███▋      | 9085/24420 [3:46:38<6:19:36,  1.49s/it]
 37%|███▋      | 9086/24420 [3:46:39<6:19:35,  1.49s/it]
 37%|███▋      | 9087/24420 [3:46:41<6:19:07,  1.48s/it]
 37%|███▋      | 9088/24420 [3:46:42<6:19:13,  1.48s/it]
 37%|███▋      | 9089/24420 [3:46:44<6:20:17,  1.49s/it]
 37%|███▋      | 9090/24420 [3:46:45<6:21:01,  1.49s/it]
                                                        
{'loss': 0.1105, 'learning_rate': 5e-05, 'epoch': 1.86}

 37%|███▋      | 9090/24420 [3:46:45<6:21:01,  1.49s/it]
 37%|███▋      | 9091/24420 [3:46:47<6:21:35,  1.49s/it]
 37%|███▋      | 9092/24420 [3:46:48<6:20:21,  1.49s/it]
 37%|███▋      | 9093/24420 [3:46:50<6:19:12,  1.48s/it]
 37%|███▋      | 9094/24420 [3:46:51<6:19:48,  1.49s/it]
 37%|███▋      | 9095/24420 [3:46:53<6:19:33,  1.49s/it]
                                                        
{'loss': 0.1099, 'learning_rate': 5e-05, 'epoch': 1.86}

 37%|███▋      | 9095/24420 [3:46:53<6:19:33,  1.49s/it]
 37%|███▋      | 9096/24420 [3:46:54<6:19:17,  1.49s/it]
 37%|███▋      | 9097/24420 [3:46:56<6:19:35,  1.49s/it]
 37%|███▋      | 9098/24420 [3:46:57<6:19:03,  1.48s/it]
 37%|███▋      | 9099/24420 [3:46:59<6:18:23,  1.48s/it]
 37%|███▋      | 9100/24420 [3:47:00<6:18:39,  1.48s/it]
                                                        
{'loss': 0.1436, 'learning_rate': 5e-05, 'epoch': 1.86}

 37%|███▋      | 9100/24420 [3:47:00<6:18:39,  1.48s/it]
 37%|███▋      | 9101/24420 [3:47:02<6:18:00,  1.48s/it]
 37%|███▋      | 9102/24420 [3:47:03<6:18:33,  1.48s/it]
 37%|███▋      | 9103/24420 [3:47:05<6:18:09,  1.48s/it]
 37%|███▋      | 9104/24420 [3:47:06<6:18:32,  1.48s/it]
 37%|███▋      | 9105/24420 [3:47:08<6:18:02,  1.48s/it]
                                                        
{'loss': 0.0972, 'learning_rate': 5e-05, 'epoch': 1.86}

 37%|███▋      | 9105/24420 [3:47:08<6:18:02,  1.48s/it]
 37%|███▋      | 9106/24420 [3:47:09<6:17:55,  1.48s/it]
 37%|███▋      | 9107/24420 [3:47:11<6:17:25,  1.48s/it]
 37%|███▋      | 9108/24420 [3:47:12<6:17:33,  1.48s/it]
 37%|███▋      | 9109/24420 [3:47:14<6:18:08,  1.48s/it]
 37%|███▋      | 9110/24420 [3:47:15<6:17:54,  1.48s/it]
                                                        
{'loss': 0.0932, 'learning_rate': 5e-05, 'epoch': 1.87}

 37%|███▋      | 9110/24420 [3:47:15<6:17:54,  1.48s/it]
 37%|███▋      | 9111/24420 [3:47:17<6:18:06,  1.48s/it]
 37%|███▋      | 9112/24420 [3:47:18<6:17:51,  1.48s/it]
 37%|███▋      | 9113/24420 [3:47:19<6:18:09,  1.48s/it]
 37%|███▋      | 9114/24420 [3:47:21<6:18:21,  1.48s/it]
 37%|███▋      | 9115/24420 [3:47:22<6:18:25,  1.48s/it]
                                                        
{'loss': 0.0914, 'learning_rate': 5e-05, 'epoch': 1.87}

 37%|███▋      | 9115/24420 [3:47:22<6:18:25,  1.48s/it]
 37%|███▋      | 9116/24420 [3:47:24<6:18:06,  1.48s/it]
 37%|███▋      | 9117/24420 [3:47:25<6:17:50,  1.48s/it]
 37%|███▋      | 9118/24420 [3:47:27<6:17:41,  1.48s/it]
 37%|███▋      | 9119/24420 [3:47:28<6:17:44,  1.48s/it]
 37%|███▋      | 9120/24420 [3:47:30<6:17:44,  1.48s/it]
                                                        
{'loss': 0.0973, 'learning_rate': 5e-05, 'epoch': 1.87}

 37%|███▋      | 9120/24420 [3:47:30<6:17:44,  1.48s/it]
 37%|███▋      | 9121/24420 [3:47:31<6:17:50,  1.48s/it]
 37%|███▋      | 9122/24420 [3:47:33<6:17:54,  1.48s/it]
 37%|███▋      | 9123/24420 [3:47:34<6:17:55,  1.48s/it]
 37%|███▋      | 9124/24420 [3:47:36<6:17:39,  1.48s/it]
 37%|███▋      | 9125/24420 [3:47:37<6:17:43,  1.48s/it]
                                                        
{'loss': 0.109, 'learning_rate': 5e-05, 'epoch': 1.87}

 37%|███▋      | 9125/24420 [3:47:37<6:17:43,  1.48s/it]
 37%|███▋      | 9126/24420 [3:47:39<6:17:29,  1.48s/it]
 37%|███▋      | 9127/24420 [3:47:40<6:18:02,  1.48s/it]
 37%|███▋      | 9128/24420 [3:47:42<6:18:08,  1.48s/it]
 37%|███▋      | 9129/24420 [3:47:43<6:17:52,  1.48s/it]
 37%|███▋      | 9130/24420 [3:47:45<6:17:31,  1.48s/it]
                                                        
{'loss': 0.0938, 'learning_rate': 5e-05, 'epoch': 1.87}

 37%|███▋      | 9130/24420 [3:47:45<6:17:31,  1.48s/it]
 37%|███▋      | 9131/24420 [3:47:46<6:17:17,  1.48s/it]
 37%|███▋      | 9132/24420 [3:47:48<6:17:44,  1.48s/it]
 37%|███▋      | 9133/24420 [3:47:49<6:18:10,  1.48s/it]
 37%|███▋      | 9134/24420 [3:47:51<6:18:33,  1.49s/it]
 37%|███▋      | 9135/24420 [3:47:52<6:17:59,  1.48s/it]
                                                        
{'loss': 0.0997, 'learning_rate': 5e-05, 'epoch': 1.87}

 37%|███▋      | 9135/24420 [3:47:52<6:17:59,  1.48s/it]
 37%|███▋      | 9136/24420 [3:47:54<6:17:24,  1.48s/it]
 37%|███▋      | 9137/24420 [3:47:55<6:17:33,  1.48s/it]
 37%|███▋      | 9138/24420 [3:47:57<6:16:54,  1.48s/it]
 37%|███▋      | 9139/24420 [3:47:58<6:17:50,  1.48s/it]
 37%|███▋      | 9140/24420 [3:48:00<6:17:21,  1.48s/it]
                                                        
{'loss': 0.1138, 'learning_rate': 5e-05, 'epoch': 1.87}

 37%|███▋      | 9140/24420 [3:48:00<6:17:21,  1.48s/it]
 37%|███▋      | 9141/24420 [3:48:01<6:16:51,  1.48s/it]
 37%|███▋      | 9142/24420 [3:48:02<6:17:06,  1.48s/it]
 37%|███▋      | 9143/24420 [3:48:04<6:17:29,  1.48s/it]
 37%|███▋      | 9144/24420 [3:48:05<6:17:24,  1.48s/it]
 37%|███▋      | 9145/24420 [3:48:07<6:17:45,  1.48s/it]
                                                        
{'loss': 0.1142, 'learning_rate': 5e-05, 'epoch': 1.87}

 37%|███▋      | 9145/24420 [3:48:07<6:17:45,  1.48s/it]
 37%|███▋      | 9146/24420 [3:48:08<6:17:32,  1.48s/it]
 37%|███▋      | 9147/24420 [3:48:10<6:17:21,  1.48s/it]
 37%|███▋      | 9148/24420 [3:48:11<6:16:54,  1.48s/it]
 37%|███▋      | 9149/24420 [3:48:13<6:17:05,  1.48s/it]
 37%|███▋      | 9150/24420 [3:48:14<6:17:25,  1.48s/it]
                                                        
{'loss': 0.0862, 'learning_rate': 5e-05, 'epoch': 1.87}

 37%|███▋      | 9150/24420 [3:48:14<6:17:25,  1.48s/it]
 37%|███▋      | 9151/24420 [3:48:16<6:17:29,  1.48s/it]
 37%|███▋      | 9152/24420 [3:48:17<6:17:34,  1.48s/it]
 37%|███▋      | 9153/24420 [3:48:19<6:17:41,  1.48s/it]
 37%|███▋      | 9154/24420 [3:48:20<6:17:47,  1.48s/it]
 37%|███▋      | 9155/24420 [3:48:22<6:17:32,  1.48s/it]
                                                        
{'loss': 0.1275, 'learning_rate': 5e-05, 'epoch': 1.87}

 37%|███▋      | 9155/24420 [3:48:22<6:17:32,  1.48s/it]
 37%|███▋      | 9156/24420 [3:48:23<6:18:05,  1.49s/it]
 37%|███▋      | 9157/24420 [3:48:25<6:17:43,  1.48s/it]
 38%|███▊      | 9158/24420 [3:48:26<6:17:18,  1.48s/it]
 38%|███▊      | 9159/24420 [3:48:28<6:17:20,  1.48s/it]
 38%|███▊      | 9160/24420 [3:48:29<6:16:56,  1.48s/it]
                                                        
{'loss': 0.0904, 'learning_rate': 5e-05, 'epoch': 1.88}

 38%|███▊      | 9160/24420 [3:48:29<6:16:56,  1.48s/it]
 38%|███▊      | 9161/24420 [3:48:31<6:16:50,  1.48s/it]
 38%|███▊      | 9162/24420 [3:48:32<6:17:05,  1.48s/it]
 38%|███▊      | 9163/24420 [3:48:34<6:17:09,  1.48s/it]
 38%|███▊      | 9164/24420 [3:48:35<6:17:38,  1.49s/it]
 38%|███▊      | 9165/24420 [3:48:37<6:17:30,  1.48s/it]
                                                        
{'loss': 0.1088, 'learning_rate': 5e-05, 'epoch': 1.88}

 38%|███▊      | 9165/24420 [3:48:37<6:17:30,  1.48s/it]
 38%|███▊      | 9166/24420 [3:48:38<6:17:54,  1.49s/it]
 38%|███▊      | 9167/24420 [3:48:40<6:17:14,  1.48s/it]
 38%|███▊      | 9168/24420 [3:48:41<6:17:26,  1.48s/it]
 38%|███▊      | 9169/24420 [3:48:43<6:17:41,  1.49s/it]
 38%|███▊      | 9170/24420 [3:48:44<6:17:34,  1.49s/it]
                                                        
{'loss': 0.1122, 'learning_rate': 5e-05, 'epoch': 1.88}

 38%|███▊      | 9170/24420 [3:48:44<6:17:34,  1.49s/it]
 38%|███▊      | 9171/24420 [3:48:46<6:17:28,  1.49s/it]
 38%|███▊      | 9172/24420 [3:48:47<6:18:30,  1.49s/it]
 38%|███▊      | 9173/24420 [3:48:49<6:19:40,  1.49s/it]
 38%|███▊      | 9174/24420 [3:48:50<6:19:22,  1.49s/it]
 38%|███▊      | 9175/24420 [3:48:51<6:18:32,  1.49s/it]
                                                        
{'loss': 0.1031, 'learning_rate': 5e-05, 'epoch': 1.88}

 38%|███▊      | 9175/24420 [3:48:51<6:18:32,  1.49s/it]
 38%|███▊      | 9176/24420 [3:48:53<6:18:21,  1.49s/it]
 38%|███▊      | 9177/24420 [3:48:54<6:18:10,  1.49s/it]
 38%|███▊      | 9178/24420 [3:48:56<6:18:37,  1.49s/it]
 38%|███▊      | 9179/24420 [3:48:57<6:18:05,  1.49s/it]
 38%|███▊      | 9180/24420 [3:48:59<6:18:40,  1.49s/it]
                                                        
{'loss': 0.1519, 'learning_rate': 5e-05, 'epoch': 1.88}

 38%|███▊      | 9180/24420 [3:48:59<6:18:40,  1.49s/it]
 38%|███▊      | 9181/24420 [3:49:00<6:19:38,  1.49s/it]
 38%|███▊      | 9182/24420 [3:49:02<6:19:50,  1.50s/it]
 38%|███▊      | 9183/24420 [3:49:03<6:19:01,  1.49s/it]
 38%|███▊      | 9184/24420 [3:49:05<6:18:16,  1.49s/it]
 38%|███▊      | 9185/24420 [3:49:06<6:18:20,  1.49s/it]
                                                        
{'loss': 0.1011, 'learning_rate': 5e-05, 'epoch': 1.88}

 38%|███▊      | 9185/24420 [3:49:06<6:18:20,  1.49s/it]
 38%|███▊      | 9186/24420 [3:49:08<6:17:58,  1.49s/it]
 38%|███▊      | 9187/24420 [3:49:09<6:17:15,  1.49s/it]
 38%|███▊      | 9188/24420 [3:49:11<6:18:11,  1.49s/it]
 38%|███▊      | 9189/24420 [3:49:12<6:17:54,  1.49s/it]
 38%|███▊      | 9190/24420 [3:49:14<6:18:16,  1.49s/it]
                                                        
{'loss': 0.1179, 'learning_rate': 5e-05, 'epoch': 1.88}

 38%|███▊      | 9190/24420 [3:49:14<6:18:16,  1.49s/it]
 38%|███▊      | 9191/24420 [3:49:15<6:17:58,  1.49s/it]
 38%|███▊      | 9192/24420 [3:49:17<6:18:40,  1.49s/it]
 38%|███▊      | 9193/24420 [3:49:18<6:18:17,  1.49s/it]
 38%|███▊      | 9194/24420 [3:49:20<6:17:35,  1.49s/it]
 38%|███▊      | 9195/24420 [3:49:21<6:19:32,  1.50s/it]
                                                        
{'loss': 0.1061, 'learning_rate': 5e-05, 'epoch': 1.88}

 38%|███▊      | 9195/24420 [3:49:21<6:19:32,  1.50s/it]
 38%|███▊      | 9196/24420 [3:49:23<6:19:02,  1.49s/it]
 38%|███▊      | 9197/24420 [3:49:24<6:19:13,  1.49s/it]
 38%|███▊      | 9198/24420 [3:49:26<6:19:10,  1.49s/it]
 38%|███▊      | 9199/24420 [3:49:27<6:18:43,  1.49s/it]
 38%|███▊      | 9200/24420 [3:49:29<6:17:38,  1.49s/it]
                                                        
{'loss': 0.1205, 'learning_rate': 5e-05, 'epoch': 1.88}

 38%|███▊      | 9200/24420 [3:49:29<6:17:38,  1.49s/it]
 38%|███▊      | 9201/24420 [3:49:33<10:23:50,  2.46s/it]
 38%|███▊      | 9202/24420 [3:49:35<9:09:21,  2.17s/it] 
 38%|███▊      | 9203/24420 [3:49:36<8:17:17,  1.96s/it]
 38%|███▊      | 9204/24420 [3:49:38<7:40:54,  1.82s/it]
 38%|███▊      | 9205/24420 [3:49:39<7:16:02,  1.72s/it]
                                                        
{'loss': 0.0964, 'learning_rate': 5e-05, 'epoch': 1.88}

 38%|███▊      | 9205/24420 [3:49:39<7:16:02,  1.72s/it]
 38%|███▊      | 9206/24420 [3:49:41<6:58:42,  1.65s/it]
 38%|███▊      | 9207/24420 [3:49:42<6:45:39,  1.60s/it]
 38%|███▊      | 9208/24420 [3:49:44<6:36:59,  1.57s/it]
 38%|███▊      | 9209/24420 [3:49:45<6:30:46,  1.54s/it]
 38%|███▊      | 9210/24420 [3:49:47<6:26:20,  1.52s/it]
                                                        
{'loss': 0.1013, 'learning_rate': 5e-05, 'epoch': 1.89}

 38%|███▊      | 9210/24420 [3:49:47<6:26:20,  1.52s/it]
 38%|███▊      | 9211/24420 [3:49:48<6:23:22,  1.51s/it]
 38%|███▊      | 9212/24420 [3:49:50<6:20:51,  1.50s/it]
 38%|███▊      | 9213/24420 [3:49:51<6:20:41,  1.50s/it]
 38%|███▊      | 9214/24420 [3:49:53<6:19:16,  1.50s/it]
 38%|███▊      | 9215/24420 [3:49:54<6:18:24,  1.49s/it]
                                                        
{'loss': 0.1204, 'learning_rate': 5e-05, 'epoch': 1.89}

 38%|███▊      | 9215/24420 [3:49:54<6:18:24,  1.49s/it]
 38%|███▊      | 9216/24420 [3:49:56<6:17:56,  1.49s/it]
 38%|███▊      | 9217/24420 [3:49:57<6:16:43,  1.49s/it]
 38%|███▊      | 9218/24420 [3:49:59<6:16:56,  1.49s/it]
 38%|███▊      | 9219/24420 [3:50:00<6:17:04,  1.49s/it]
 38%|███▊      | 9220/24420 [3:50:02<6:16:46,  1.49s/it]
                                                        
{'loss': 0.1, 'learning_rate': 5e-05, 'epoch': 1.89}

 38%|███▊      | 9220/24420 [3:50:02<6:16:46,  1.49s/it]
 38%|███▊      | 9221/24420 [3:50:03<6:17:23,  1.49s/it]
 38%|███▊      | 9222/24420 [3:50:05<6:17:33,  1.49s/it]
 38%|███▊      | 9223/24420 [3:50:06<6:17:14,  1.49s/it]
 38%|███▊      | 9224/24420 [3:50:08<6:17:05,  1.49s/it]
 38%|███▊      | 9225/24420 [3:50:09<6:17:13,  1.49s/it]
                                                        
{'loss': 0.1045, 'learning_rate': 5e-05, 'epoch': 1.89}

 38%|███▊      | 9225/24420 [3:50:09<6:17:13,  1.49s/it]
 38%|███▊      | 9226/24420 [3:50:11<6:17:05,  1.49s/it]
 38%|███▊      | 9227/24420 [3:50:12<6:16:40,  1.49s/it]
 38%|███▊      | 9228/24420 [3:50:14<6:16:57,  1.49s/it]
 38%|███▊      | 9229/24420 [3:50:15<6:16:43,  1.49s/it]
 38%|███▊      | 9230/24420 [3:50:17<6:16:38,  1.49s/it]
                                                        
{'loss': 0.1115, 'learning_rate': 5e-05, 'epoch': 1.89}

 38%|███▊      | 9230/24420 [3:50:17<6:16:38,  1.49s/it]
 38%|███▊      | 9231/24420 [3:50:18<6:17:05,  1.49s/it]
 38%|███▊      | 9232/24420 [3:50:20<6:16:40,  1.49s/it]
 38%|███▊      | 9233/24420 [3:50:21<6:16:18,  1.49s/it]
 38%|███▊      | 9234/24420 [3:50:23<6:16:10,  1.49s/it]
 38%|███▊      | 9235/24420 [3:50:24<6:15:51,  1.49s/it]
                                                        
{'loss': 0.1, 'learning_rate': 5e-05, 'epoch': 1.89}

 38%|███▊      | 9235/24420 [3:50:24<6:15:51,  1.49s/it]
 38%|███▊      | 9236/24420 [3:50:26<6:15:32,  1.48s/it]
 38%|███▊      | 9237/24420 [3:50:27<6:15:58,  1.49s/it]
 38%|███▊      | 9238/24420 [3:50:29<6:17:40,  1.49s/it]
 38%|███▊      | 9239/24420 [3:50:30<6:17:23,  1.49s/it]
 38%|███▊      | 9240/24420 [3:50:31<6:17:29,  1.49s/it]
                                                        
{'loss': 0.1188, 'learning_rate': 5e-05, 'epoch': 1.89}

 38%|███▊      | 9240/24420 [3:50:31<6:17:29,  1.49s/it]
 38%|███▊      | 9241/24420 [3:50:33<6:17:09,  1.49s/it]
 38%|███▊      | 9242/24420 [3:50:34<6:16:37,  1.49s/it]
 38%|███▊      | 9243/24420 [3:50:36<6:15:54,  1.49s/it]
 38%|███▊      | 9244/24420 [3:50:37<6:16:25,  1.49s/it]
 38%|███▊      | 9245/24420 [3:50:39<6:15:44,  1.49s/it]
                                                        
{'loss': 0.0896, 'learning_rate': 5e-05, 'epoch': 1.89}

 38%|███▊      | 9245/24420 [3:50:39<6:15:44,  1.49s/it]
 38%|███▊      | 9246/24420 [3:50:40<6:15:59,  1.49s/it]
 38%|███▊      | 9247/24420 [3:50:42<6:16:08,  1.49s/it]
 38%|███▊      | 9248/24420 [3:50:43<6:16:35,  1.49s/it]
 38%|███▊      | 9249/24420 [3:50:45<6:16:44,  1.49s/it]
 38%|███▊      | 9250/24420 [3:50:46<6:16:31,  1.49s/it]
                                                        
{'loss': 0.0965, 'learning_rate': 5e-05, 'epoch': 1.89}

 38%|███▊      | 9250/24420 [3:50:46<6:16:31,  1.49s/it]
 38%|███▊      | 9251/24420 [3:50:48<6:16:35,  1.49s/it]
 38%|███▊      | 9252/24420 [3:50:49<6:16:23,  1.49s/it]
 38%|███▊      | 9253/24420 [3:50:51<6:16:17,  1.49s/it]
 38%|███▊      | 9254/24420 [3:50:52<6:16:07,  1.49s/it]
 38%|███▊      | 9255/24420 [3:50:54<6:15:33,  1.49s/it]
                                                        
{'loss': 0.1059, 'learning_rate': 5e-05, 'epoch': 1.89}

 38%|███▊      | 9255/24420 [3:50:54<6:15:33,  1.49s/it]
 38%|███▊      | 9256/24420 [3:50:55<6:15:50,  1.49s/it]
 38%|███▊      | 9257/24420 [3:50:57<6:15:49,  1.49s/it]
 38%|███▊      | 9258/24420 [3:50:58<6:16:23,  1.49s/it]
 38%|███▊      | 9259/24420 [3:51:00<6:15:50,  1.49s/it]
 38%|███▊      | 9260/24420 [3:51:01<6:15:20,  1.49s/it]
                                                        
{'loss': 0.1077, 'learning_rate': 5e-05, 'epoch': 1.9}

 38%|███▊      | 9260/24420 [3:51:01<6:15:20,  1.49s/it]
 38%|███▊      | 9261/24420 [3:51:03<6:16:03,  1.49s/it]
 38%|███▊      | 9262/24420 [3:51:04<6:15:59,  1.49s/it]
 38%|███▊      | 9263/24420 [3:51:06<6:15:26,  1.49s/it]
 38%|███▊      | 9264/24420 [3:51:07<6:15:27,  1.49s/it]
 38%|███▊      | 9265/24420 [3:51:09<6:15:31,  1.49s/it]
                                                        
{'loss': 0.0926, 'learning_rate': 5e-05, 'epoch': 1.9}

 38%|███▊      | 9265/24420 [3:51:09<6:15:31,  1.49s/it]
 38%|███▊      | 9266/24420 [3:51:10<6:16:13,  1.49s/it]
 38%|███▊      | 9267/24420 [3:51:12<6:15:57,  1.49s/it]
 38%|███▊      | 9268/24420 [3:51:13<6:16:31,  1.49s/it]
 38%|███▊      | 9269/24420 [3:51:15<6:16:24,  1.49s/it]
 38%|███▊      | 9270/24420 [3:51:16<6:15:21,  1.49s/it]
                                                        
{'loss': 0.1075, 'learning_rate': 5e-05, 'epoch': 1.9}

 38%|███▊      | 9270/24420 [3:51:16<6:15:21,  1.49s/it]
 38%|███▊      | 9271/24420 [3:51:18<6:15:36,  1.49s/it]
 38%|███▊      | 9272/24420 [3:51:19<6:15:38,  1.49s/it]
 38%|███▊      | 9273/24420 [3:51:21<6:14:49,  1.48s/it]
 38%|███▊      | 9274/24420 [3:51:22<6:14:59,  1.49s/it]
 38%|███▊      | 9275/24420 [3:51:24<6:14:26,  1.48s/it]
                                                        
{'loss': 0.1211, 'learning_rate': 5e-05, 'epoch': 1.9}

 38%|███▊      | 9275/24420 [3:51:24<6:14:26,  1.48s/it]
 38%|███▊      | 9276/24420 [3:51:25<6:14:47,  1.48s/it]
 38%|███▊      | 9277/24420 [3:51:27<6:14:35,  1.48s/it]
 38%|███▊      | 9278/24420 [3:51:28<6:14:12,  1.48s/it]
 38%|███▊      | 9279/24420 [3:51:29<6:13:53,  1.48s/it]
 38%|███▊      | 9280/24420 [3:51:31<6:13:59,  1.48s/it]
                                                        
{'loss': 0.097, 'learning_rate': 5e-05, 'epoch': 1.9}

 38%|███▊      | 9280/24420 [3:51:31<6:13:59,  1.48s/it]
 38%|███▊      | 9281/24420 [3:51:32<6:13:56,  1.48s/it]
 38%|███▊      | 9282/24420 [3:51:34<6:13:50,  1.48s/it]
 38%|███▊      | 9283/24420 [3:51:35<6:14:28,  1.48s/it]
 38%|███▊      | 9284/24420 [3:51:37<6:14:23,  1.48s/it]
 38%|███▊      | 9285/24420 [3:51:38<6:14:41,  1.49s/it]
                                                        
{'loss': 0.1115, 'learning_rate': 5e-05, 'epoch': 1.9}

 38%|███▊      | 9285/24420 [3:51:38<6:14:41,  1.49s/it]
 38%|███▊      | 9286/24420 [3:51:40<6:15:01,  1.49s/it]
 38%|███▊      | 9287/24420 [3:51:41<6:15:03,  1.49s/it]
 38%|███▊      | 9288/24420 [3:51:43<6:15:27,  1.49s/it]
 38%|███▊      | 9289/24420 [3:51:44<6:15:17,  1.49s/it]
 38%|███▊      | 9290/24420 [3:51:46<6:14:51,  1.49s/it]
                                                        
{'loss': 0.1114, 'learning_rate': 5e-05, 'epoch': 1.9}

 38%|███▊      | 9290/24420 [3:51:46<6:14:51,  1.49s/it]
 38%|███▊      | 9291/24420 [3:51:47<6:15:02,  1.49s/it]
 38%|███▊      | 9292/24420 [3:51:49<6:15:41,  1.49s/it]
 38%|███▊      | 9293/24420 [3:51:50<6:16:10,  1.49s/it]
 38%|███▊      | 9294/24420 [3:51:52<6:15:34,  1.49s/it]
 38%|███▊      | 9295/24420 [3:51:53<6:14:59,  1.49s/it]
                                                        
{'loss': 0.113, 'learning_rate': 5e-05, 'epoch': 1.9}

 38%|███▊      | 9295/24420 [3:51:53<6:14:59,  1.49s/it]
 38%|███▊      | 9296/24420 [3:51:55<6:15:01,  1.49s/it]
 38%|███▊      | 9297/24420 [3:51:56<6:15:06,  1.49s/it]
 38%|███▊      | 9298/24420 [3:51:58<6:14:51,  1.49s/it]
 38%|███▊      | 9299/24420 [3:51:59<6:15:21,  1.49s/it]
 38%|███▊      | 9300/24420 [3:52:01<6:15:26,  1.49s/it]
                                                        
{'loss': 0.099, 'learning_rate': 5e-05, 'epoch': 1.9}

 38%|███▊      | 9300/24420 [3:52:01<6:15:26,  1.49s/it]
 38%|███▊      | 9301/24420 [3:52:02<6:14:58,  1.49s/it]
 38%|███▊      | 9302/24420 [3:52:04<6:15:27,  1.49s/it]
 38%|███▊      | 9303/24420 [3:52:05<6:14:45,  1.49s/it]
 38%|███▊      | 9304/24420 [3:52:07<6:14:42,  1.49s/it]
 38%|███▊      | 9305/24420 [3:52:08<6:13:53,  1.48s/it]
                                                        
{'loss': 0.0928, 'learning_rate': 5e-05, 'epoch': 1.91}

 38%|███▊      | 9305/24420 [3:52:08<6:13:53,  1.48s/it]
 38%|███▊      | 9306/24420 [3:52:10<6:14:13,  1.49s/it]
 38%|███▊      | 9307/24420 [3:52:11<6:13:53,  1.48s/it]
 38%|███▊      | 9308/24420 [3:52:13<6:13:46,  1.48s/it]
 38%|███▊      | 9309/24420 [3:52:14<6:13:56,  1.48s/it]
 38%|███▊      | 9310/24420 [3:52:16<6:13:22,  1.48s/it]
                                                        
{'loss': 0.0987, 'learning_rate': 5e-05, 'epoch': 1.91}

 38%|███▊      | 9310/24420 [3:52:16<6:13:22,  1.48s/it]
 38%|███▊      | 9311/24420 [3:52:17<6:13:28,  1.48s/it]
 38%|███▊      | 9312/24420 [3:52:19<6:13:47,  1.48s/it]
 38%|███▊      | 9313/24420 [3:52:20<6:14:16,  1.49s/it]
 38%|███▊      | 9314/24420 [3:52:21<6:13:52,  1.48s/it]
 38%|███▊      | 9315/24420 [3:52:23<6:13:53,  1.49s/it]
                                                        
{'loss': 0.123, 'learning_rate': 5e-05, 'epoch': 1.91}

 38%|███▊      | 9315/24420 [3:52:23<6:13:53,  1.49s/it]
 38%|███▊      | 9316/24420 [3:52:24<6:14:39,  1.49s/it]
 38%|███▊      | 9317/24420 [3:52:26<6:15:45,  1.49s/it]
 38%|███▊      | 9318/24420 [3:52:27<6:15:27,  1.49s/it]
 38%|███▊      | 9319/24420 [3:52:29<6:16:15,  1.49s/it]
 38%|███▊      | 9320/24420 [3:52:30<6:16:06,  1.49s/it]
                                                        
{'loss': 0.0981, 'learning_rate': 5e-05, 'epoch': 1.91}

 38%|███▊      | 9320/24420 [3:52:30<6:16:06,  1.49s/it]
 38%|███▊      | 9321/24420 [3:52:32<6:15:36,  1.49s/it]
 38%|███▊      | 9322/24420 [3:52:33<6:15:09,  1.49s/it]
 38%|███▊      | 9323/24420 [3:52:35<6:14:07,  1.49s/it]
 38%|███▊      | 9324/24420 [3:52:36<6:13:55,  1.49s/it]
 38%|███▊      | 9325/24420 [3:52:38<6:13:26,  1.48s/it]
                                                        
{'loss': 0.0871, 'learning_rate': 5e-05, 'epoch': 1.91}

 38%|███▊      | 9325/24420 [3:52:38<6:13:26,  1.48s/it]
 38%|███▊      | 9326/24420 [3:52:39<6:13:44,  1.49s/it]
 38%|███▊      | 9327/24420 [3:52:41<6:13:34,  1.49s/it]
 38%|███▊      | 9328/24420 [3:52:42<6:13:13,  1.48s/it]
 38%|███▊      | 9329/24420 [3:52:44<6:13:09,  1.48s/it]
 38%|███▊      | 9330/24420 [3:52:45<6:12:55,  1.48s/it]
                                                        
{'loss': 0.0975, 'learning_rate': 5e-05, 'epoch': 1.91}

 38%|███▊      | 9330/24420 [3:52:45<6:12:55,  1.48s/it]
 38%|███▊      | 9331/24420 [3:52:47<6:12:56,  1.48s/it]
 38%|███▊      | 9332/24420 [3:52:48<6:14:03,  1.49s/it]
 38%|███▊      | 9333/24420 [3:52:50<6:14:44,  1.49s/it]
 38%|███▊      | 9334/24420 [3:52:51<6:14:52,  1.49s/it]
 38%|███▊      | 9335/24420 [3:52:53<6:14:55,  1.49s/it]
                                                        
{'loss': 0.1005, 'learning_rate': 5e-05, 'epoch': 1.91}

 38%|███▊      | 9335/24420 [3:52:53<6:14:55,  1.49s/it]
 38%|███▊      | 9336/24420 [3:52:54<6:14:32,  1.49s/it]
 38%|███▊      | 9337/24420 [3:52:56<6:15:21,  1.49s/it]
 38%|███▊      | 9338/24420 [3:52:57<6:14:41,  1.49s/it]
 38%|███▊      | 9339/24420 [3:52:59<6:13:22,  1.49s/it]
 38%|███▊      | 9340/24420 [3:53:00<6:13:25,  1.49s/it]
                                                        
{'loss': 0.1192, 'learning_rate': 5e-05, 'epoch': 1.91}

 38%|███▊      | 9340/24420 [3:53:00<6:13:25,  1.49s/it]
 38%|███▊      | 9341/24420 [3:53:02<6:13:01,  1.48s/it]
 38%|███▊      | 9342/24420 [3:53:03<6:13:46,  1.49s/it]
 38%|███▊      | 9343/24420 [3:53:05<6:14:34,  1.49s/it]
 38%|███▊      | 9344/24420 [3:53:06<6:14:09,  1.49s/it]
 38%|███▊      | 9345/24420 [3:53:08<6:13:49,  1.49s/it]
                                                        
{'loss': 0.0967, 'learning_rate': 5e-05, 'epoch': 1.91}

 38%|███▊      | 9345/24420 [3:53:08<6:13:49,  1.49s/it]
 38%|███▊      | 9346/24420 [3:53:09<6:13:42,  1.49s/it]
 38%|███▊      | 9347/24420 [3:53:11<6:13:25,  1.49s/it]
 38%|███▊      | 9348/24420 [3:53:12<6:13:22,  1.49s/it]
 38%|███▊      | 9349/24420 [3:53:14<6:13:25,  1.49s/it]
 38%|███▊      | 9350/24420 [3:53:15<6:12:41,  1.48s/it]
                                                        
{'loss': 0.1015, 'learning_rate': 5e-05, 'epoch': 1.91}

 38%|███▊      | 9350/24420 [3:53:15<6:12:41,  1.48s/it]
 38%|███▊      | 9351/24420 [3:53:17<6:12:34,  1.48s/it]
 38%|███▊      | 9352/24420 [3:53:18<6:11:45,  1.48s/it]
 38%|███▊      | 9353/24420 [3:53:20<6:12:24,  1.48s/it]
 38%|███▊      | 9354/24420 [3:53:21<6:12:50,  1.48s/it]
 38%|███▊      | 9355/24420 [3:53:22<6:12:45,  1.48s/it]
                                                        
{'loss': 0.0901, 'learning_rate': 5e-05, 'epoch': 1.92}

 38%|███▊      | 9355/24420 [3:53:22<6:12:45,  1.48s/it]
 38%|███▊      | 9356/24420 [3:53:24<6:12:42,  1.48s/it]
 38%|███▊      | 9357/24420 [3:53:25<6:13:00,  1.49s/it]
 38%|███▊      | 9358/24420 [3:53:27<6:13:00,  1.49s/it]
 38%|███▊      | 9359/24420 [3:53:28<6:12:32,  1.48s/it]
 38%|███▊      | 9360/24420 [3:53:30<6:12:26,  1.48s/it]
                                                        
{'loss': 0.1007, 'learning_rate': 5e-05, 'epoch': 1.92}

 38%|███▊      | 9360/24420 [3:53:30<6:12:26,  1.48s/it]
 38%|███▊      | 9361/24420 [3:53:31<6:13:25,  1.49s/it]
 38%|███▊      | 9362/24420 [3:53:33<6:13:29,  1.49s/it]
 38%|███▊      | 9363/24420 [3:53:34<6:13:26,  1.49s/it]
 38%|███▊      | 9364/24420 [3:53:36<6:12:54,  1.49s/it]
 38%|███▊      | 9365/24420 [3:53:37<6:12:28,  1.48s/it]
                                                        
{'loss': 0.099, 'learning_rate': 5e-05, 'epoch': 1.92}

 38%|███▊      | 9365/24420 [3:53:37<6:12:28,  1.48s/it]
 38%|███▊      | 9366/24420 [3:53:39<6:12:54,  1.49s/it]
 38%|███▊      | 9367/24420 [3:53:40<6:12:42,  1.49s/it]
 38%|███▊      | 9368/24420 [3:53:42<6:12:52,  1.49s/it]
 38%|███▊      | 9369/24420 [3:53:43<6:14:13,  1.49s/it]
 38%|███▊      | 9370/24420 [3:53:45<6:14:27,  1.49s/it]
                                                        
{'loss': 0.1079, 'learning_rate': 5e-05, 'epoch': 1.92}

 38%|███▊      | 9370/24420 [3:53:45<6:14:27,  1.49s/it]
 38%|███▊      | 9371/24420 [3:53:46<6:13:59,  1.49s/it]
 38%|███▊      | 9372/24420 [3:53:48<6:15:10,  1.50s/it]
 38%|███▊      | 9373/24420 [3:53:49<6:13:43,  1.49s/it]
 38%|███▊      | 9374/24420 [3:53:51<6:13:41,  1.49s/it]
 38%|███▊      | 9375/24420 [3:53:52<6:13:38,  1.49s/it]
                                                        
{'loss': 0.1081, 'learning_rate': 5e-05, 'epoch': 1.92}

 38%|███▊      | 9375/24420 [3:53:52<6:13:38,  1.49s/it]
 38%|███▊      | 9376/24420 [3:53:54<6:13:45,  1.49s/it]
 38%|███▊      | 9377/24420 [3:53:55<6:13:54,  1.49s/it]
 38%|███▊      | 9378/24420 [3:53:57<6:14:33,  1.49s/it]
 38%|███▊      | 9379/24420 [3:53:58<6:14:32,  1.49s/it]
 38%|███▊      | 9380/24420 [3:54:00<6:14:20,  1.49s/it]
                                                        
{'loss': 0.0931, 'learning_rate': 5e-05, 'epoch': 1.92}

 38%|███▊      | 9380/24420 [3:54:00<6:14:20,  1.49s/it]
 38%|███▊      | 9381/24420 [3:54:01<6:14:17,  1.49s/it]
 38%|███▊      | 9382/24420 [3:54:03<6:14:03,  1.49s/it]
 38%|███▊      | 9383/24420 [3:54:04<6:13:35,  1.49s/it]
 38%|███▊      | 9384/24420 [3:54:06<6:13:47,  1.49s/it]
 38%|███▊      | 9385/24420 [3:54:07<6:13:22,  1.49s/it]
                                                        
{'loss': 0.1238, 'learning_rate': 5e-05, 'epoch': 1.92}

 38%|███▊      | 9385/24420 [3:54:07<6:13:22,  1.49s/it]
 38%|███▊      | 9386/24420 [3:54:09<6:13:26,  1.49s/it]
 38%|███▊      | 9387/24420 [3:54:10<6:12:57,  1.49s/it]
 38%|███▊      | 9388/24420 [3:54:12<6:13:08,  1.49s/it]
 38%|███▊      | 9389/24420 [3:54:13<6:12:40,  1.49s/it]
 38%|███▊      | 9390/24420 [3:54:15<6:12:02,  1.49s/it]
                                                        
{'loss': 0.1117, 'learning_rate': 5e-05, 'epoch': 1.92}

 38%|███▊      | 9390/24420 [3:54:15<6:12:02,  1.49s/it]
 38%|███▊      | 9391/24420 [3:54:16<6:12:13,  1.49s/it]
 38%|███▊      | 9392/24420 [3:54:18<6:12:18,  1.49s/it]
 38%|███▊      | 9393/24420 [3:54:19<6:12:25,  1.49s/it]
 38%|███▊      | 9394/24420 [3:54:21<6:11:57,  1.49s/it]
 38%|███▊      | 9395/24420 [3:54:22<6:12:29,  1.49s/it]
                                                        
{'loss': 0.1163, 'learning_rate': 5e-05, 'epoch': 1.92}

 38%|███▊      | 9395/24420 [3:54:22<6:12:29,  1.49s/it]
 38%|███▊      | 9396/24420 [3:54:24<6:12:56,  1.49s/it]
 38%|███▊      | 9397/24420 [3:54:25<6:12:24,  1.49s/it]
 38%|███▊      | 9398/24420 [3:54:27<6:12:49,  1.49s/it]
 38%|███▊      | 9399/24420 [3:54:28<6:12:46,  1.49s/it]
 38%|███▊      | 9400/24420 [3:54:30<6:14:31,  1.50s/it]
                                                        
{'loss': 0.0997, 'learning_rate': 5e-05, 'epoch': 1.92}

 38%|███▊      | 9400/24420 [3:54:30<6:14:31,  1.50s/it]
 38%|███▊      | 9401/24420 [3:54:34<10:21:54,  2.48s/it]
 39%|███▊      | 9402/24420 [3:54:36<9:07:59,  2.19s/it] 
 39%|███▊      | 9403/24420 [3:54:37<8:15:31,  1.98s/it]
 39%|███▊      | 9404/24420 [3:54:39<7:38:10,  1.83s/it]
 39%|███▊      | 9405/24420 [3:54:40<7:11:30,  1.72s/it]
                                                        
{'loss': 0.1135, 'learning_rate': 5e-05, 'epoch': 1.93}

 39%|███▊      | 9405/24420 [3:54:40<7:11:30,  1.72s/it]
 39%|███▊      | 9406/24420 [3:54:42<6:53:32,  1.65s/it]
 39%|███▊      | 9407/24420 [3:54:43<6:40:43,  1.60s/it]
 39%|███▊      | 9408/24420 [3:54:45<6:32:11,  1.57s/it]
 39%|███▊      | 9409/24420 [3:54:46<6:26:52,  1.55s/it]
 39%|███▊      | 9410/24420 [3:54:48<6:21:30,  1.53s/it]
                                                        
{'loss': 0.1084, 'learning_rate': 5e-05, 'epoch': 1.93}

 39%|███▊      | 9410/24420 [3:54:48<6:21:30,  1.53s/it]
 39%|███▊      | 9411/24420 [3:54:49<6:18:43,  1.51s/it]
 39%|███▊      | 9412/24420 [3:54:51<6:16:31,  1.51s/it]
 39%|███▊      | 9413/24420 [3:54:52<6:14:30,  1.50s/it]
 39%|███▊      | 9414/24420 [3:54:54<6:13:12,  1.49s/it]
 39%|███▊      | 9415/24420 [3:54:55<6:12:31,  1.49s/it]
                                                        
{'loss': 0.0961, 'learning_rate': 5e-05, 'epoch': 1.93}

 39%|███▊      | 9415/24420 [3:54:55<6:12:31,  1.49s/it]
 39%|███▊      | 9416/24420 [3:54:57<6:12:11,  1.49s/it]
 39%|███▊      | 9417/24420 [3:54:59<7:16:03,  1.74s/it]
 39%|███▊      | 9418/24420 [3:55:00<6:56:18,  1.67s/it]
 39%|███▊      | 9419/24420 [3:55:02<6:43:02,  1.61s/it]
 39%|███▊      | 9420/24420 [3:55:03<6:33:27,  1.57s/it]
                                                        
{'loss': 0.109, 'learning_rate': 5e-05, 'epoch': 1.93}

 39%|███▊      | 9420/24420 [3:55:03<6:33:27,  1.57s/it]
 39%|███▊      | 9421/24420 [3:55:05<6:26:44,  1.55s/it]
 39%|███▊      | 9422/24420 [3:55:06<6:21:19,  1.53s/it]
 39%|███▊      | 9423/24420 [3:55:08<6:17:44,  1.51s/it]
 39%|███▊      | 9424/24420 [3:55:09<6:15:52,  1.50s/it]
 39%|███▊      | 9425/24420 [3:55:11<6:14:26,  1.50s/it]
                                                        
{'loss': 0.0959, 'learning_rate': 5e-05, 'epoch': 1.93}

 39%|███▊      | 9425/24420 [3:55:11<6:14:26,  1.50s/it]
 39%|███▊      | 9426/24420 [3:55:12<6:14:17,  1.50s/it]
 39%|███▊      | 9427/24420 [3:55:14<6:14:00,  1.50s/it]
 39%|███▊      | 9428/24420 [3:55:15<6:12:59,  1.49s/it]
 39%|███▊      | 9429/24420 [3:55:17<6:12:53,  1.49s/it]
 39%|███▊      | 9430/24420 [3:55:18<6:12:13,  1.49s/it]
                                                        
{'loss': 0.1238, 'learning_rate': 5e-05, 'epoch': 1.93}

 39%|███▊      | 9430/24420 [3:55:18<6:12:13,  1.49s/it]
 39%|███▊      | 9431/24420 [3:55:20<6:12:20,  1.49s/it]
 39%|███▊      | 9432/24420 [3:55:21<6:11:43,  1.49s/it]
 39%|███▊      | 9433/24420 [3:55:23<6:11:18,  1.49s/it]
 39%|███▊      | 9434/24420 [3:55:24<6:10:56,  1.49s/it]
 39%|███▊      | 9435/24420 [3:55:26<6:11:52,  1.49s/it]
                                                        
{'loss': 0.1115, 'learning_rate': 5e-05, 'epoch': 1.93}

 39%|███▊      | 9435/24420 [3:55:26<6:11:52,  1.49s/it]
 39%|███▊      | 9436/24420 [3:55:27<6:12:19,  1.49s/it]
 39%|███▊      | 9437/24420 [3:55:29<6:12:02,  1.49s/it]
 39%|███▊      | 9438/24420 [3:55:30<6:11:45,  1.49s/it]
 39%|███▊      | 9439/24420 [3:55:32<6:10:50,  1.49s/it]
 39%|███▊      | 9440/24420 [3:55:33<6:12:13,  1.49s/it]
                                                        
{'loss': 0.1031, 'learning_rate': 5e-05, 'epoch': 1.93}

 39%|███▊      | 9440/24420 [3:55:33<6:12:13,  1.49s/it]
 39%|███▊      | 9441/24420 [3:55:35<6:12:08,  1.49s/it]
 39%|███▊      | 9442/24420 [3:55:36<6:11:28,  1.49s/it]
 39%|███▊      | 9443/24420 [3:55:38<6:11:21,  1.49s/it]
 39%|███▊      | 9444/24420 [3:55:39<6:11:01,  1.49s/it]
 39%|███▊      | 9445/24420 [3:55:41<6:11:17,  1.49s/it]
                                                        
{'loss': 0.1128, 'learning_rate': 5e-05, 'epoch': 1.93}

 39%|███▊      | 9445/24420 [3:55:41<6:11:17,  1.49s/it]
 39%|███▊      | 9446/24420 [3:55:42<6:11:34,  1.49s/it]
 39%|███▊      | 9447/24420 [3:55:44<6:10:49,  1.49s/it]
 39%|███▊      | 9448/24420 [3:55:45<6:10:46,  1.49s/it]
 39%|███▊      | 9449/24420 [3:55:47<6:10:49,  1.49s/it]
 39%|███▊      | 9450/24420 [3:55:48<6:11:01,  1.49s/it]
                                                        
{'loss': 0.0899, 'learning_rate': 5e-05, 'epoch': 1.93}

 39%|███▊      | 9450/24420 [3:55:48<6:11:01,  1.49s/it]
 39%|███▊      | 9451/24420 [3:55:49<6:10:53,  1.49s/it]
 39%|███▊      | 9452/24420 [3:55:51<6:10:32,  1.49s/it]
 39%|███▊      | 9453/24420 [3:55:52<6:10:02,  1.48s/it]
 39%|███▊      | 9454/24420 [3:55:54<6:10:24,  1.48s/it]
 39%|███▊      | 9455/24420 [3:55:55<6:10:19,  1.48s/it]
                                                        
{'loss': 0.115, 'learning_rate': 5e-05, 'epoch': 1.94}

 39%|███▊      | 9455/24420 [3:55:55<6:10:19,  1.48s/it]
 39%|███▊      | 9456/24420 [3:55:57<6:09:48,  1.48s/it]
 39%|███▊      | 9457/24420 [3:55:58<6:09:24,  1.48s/it]
 39%|███▊      | 9458/24420 [3:56:00<6:09:23,  1.48s/it]
 39%|███▊      | 9459/24420 [3:56:01<6:09:39,  1.48s/it]
 39%|███▊      | 9460/24420 [3:56:03<6:09:49,  1.48s/it]
                                                        
{'loss': 0.0969, 'learning_rate': 5e-05, 'epoch': 1.94}

 39%|███▊      | 9460/24420 [3:56:03<6:09:49,  1.48s/it]
 39%|███▊      | 9461/24420 [3:56:04<6:10:54,  1.49s/it]
 39%|███▊      | 9462/24420 [3:56:06<6:10:54,  1.49s/it]
 39%|███▉      | 9463/24420 [3:56:07<6:10:34,  1.49s/it]
 39%|███▉      | 9464/24420 [3:56:09<6:10:31,  1.49s/it]
 39%|███▉      | 9465/24420 [3:56:10<6:10:21,  1.49s/it]
                                                        
{'loss': 0.1101, 'learning_rate': 5e-05, 'epoch': 1.94}

 39%|███▉      | 9465/24420 [3:56:10<6:10:21,  1.49s/it]
 39%|███▉      | 9466/24420 [3:56:12<6:10:54,  1.49s/it]
 39%|███▉      | 9467/24420 [3:56:13<6:10:42,  1.49s/it]
 39%|███▉      | 9468/24420 [3:56:15<6:10:24,  1.49s/it]
 39%|███▉      | 9469/24420 [3:56:16<6:10:17,  1.49s/it]
 39%|███▉      | 9470/24420 [3:56:18<6:10:48,  1.49s/it]
                                                        
{'loss': 0.1066, 'learning_rate': 5e-05, 'epoch': 1.94}

 39%|███▉      | 9470/24420 [3:56:18<6:10:48,  1.49s/it]
 39%|███▉      | 9471/24420 [3:56:19<6:11:39,  1.49s/it]
 39%|███▉      | 9472/24420 [3:56:21<6:11:18,  1.49s/it]
 39%|███▉      | 9473/24420 [3:56:22<6:11:43,  1.49s/it]
 39%|███▉      | 9474/24420 [3:56:24<6:10:50,  1.49s/it]
 39%|███▉      | 9475/24420 [3:56:25<6:10:38,  1.49s/it]
                                                        
{'loss': 0.1044, 'learning_rate': 5e-05, 'epoch': 1.94}

 39%|███▉      | 9475/24420 [3:56:25<6:10:38,  1.49s/it]
 39%|███▉      | 9476/24420 [3:56:27<6:10:41,  1.49s/it]
 39%|███▉      | 9477/24420 [3:56:28<6:09:44,  1.48s/it]
 39%|███▉      | 9478/24420 [3:56:30<6:09:35,  1.48s/it]
 39%|███▉      | 9479/24420 [3:56:31<6:10:04,  1.49s/it]
 39%|███▉      | 9480/24420 [3:56:33<6:09:58,  1.49s/it]
                                                        
{'loss': 0.1023, 'learning_rate': 5e-05, 'epoch': 1.94}

 39%|███▉      | 9480/24420 [3:56:33<6:09:58,  1.49s/it]
 39%|███▉      | 9481/24420 [3:56:34<6:09:31,  1.48s/it]
 39%|███▉      | 9482/24420 [3:56:36<6:09:56,  1.49s/it]
 39%|███▉      | 9483/24420 [3:56:37<6:09:58,  1.49s/it]
 39%|███▉      | 9484/24420 [3:56:39<6:10:21,  1.49s/it]
 39%|███▉      | 9485/24420 [3:56:40<6:10:34,  1.49s/it]
                                                        
{'loss': 0.1174, 'learning_rate': 5e-05, 'epoch': 1.94}

 39%|███▉      | 9485/24420 [3:56:40<6:10:34,  1.49s/it]
 39%|███▉      | 9486/24420 [3:56:42<6:11:03,  1.49s/it]
 39%|███▉      | 9487/24420 [3:56:43<6:10:23,  1.49s/it]
 39%|███▉      | 9488/24420 [3:56:44<6:10:43,  1.49s/it]
 39%|███▉      | 9489/24420 [3:56:46<6:11:10,  1.49s/it]
 39%|███▉      | 9490/24420 [3:56:47<6:10:24,  1.49s/it]
                                                        
{'loss': 0.1134, 'learning_rate': 5e-05, 'epoch': 1.94}

 39%|███▉      | 9490/24420 [3:56:47<6:10:24,  1.49s/it]
 39%|███▉      | 9491/24420 [3:56:49<6:10:20,  1.49s/it]
 39%|███▉      | 9492/24420 [3:56:50<6:10:08,  1.49s/it]
 39%|███▉      | 9493/24420 [3:56:52<6:10:05,  1.49s/it]
 39%|███▉      | 9494/24420 [3:56:53<6:09:13,  1.48s/it]
 39%|███▉      | 9495/24420 [3:56:55<6:08:33,  1.48s/it]
                                                        
{'loss': 0.0977, 'learning_rate': 5e-05, 'epoch': 1.94}

 39%|███▉      | 9495/24420 [3:56:55<6:08:33,  1.48s/it]
 39%|███▉      | 9496/24420 [3:56:56<6:08:52,  1.48s/it]
 39%|███▉      | 9497/24420 [3:56:58<6:09:01,  1.48s/it]
 39%|███▉      | 9498/24420 [3:56:59<6:09:17,  1.48s/it]
 39%|███▉      | 9499/24420 [3:57:01<6:09:23,  1.49s/it]
 39%|███▉      | 9500/24420 [3:57:02<6:09:22,  1.49s/it]
                                                        
{'loss': 0.1131, 'learning_rate': 5e-05, 'epoch': 1.95}

 39%|███▉      | 9500/24420 [3:57:02<6:09:22,  1.49s/it]
 39%|███▉      | 9501/24420 [3:57:04<6:09:47,  1.49s/it]
 39%|███▉      | 9502/24420 [3:57:05<6:10:04,  1.49s/it]
 39%|███▉      | 9503/24420 [3:57:07<6:10:02,  1.49s/it]
 39%|███▉      | 9504/24420 [3:57:08<6:10:44,  1.49s/it]
 39%|███▉      | 9505/24420 [3:57:10<6:10:32,  1.49s/it]
                                                        
{'loss': 0.1048, 'learning_rate': 5e-05, 'epoch': 1.95}

 39%|███▉      | 9505/24420 [3:57:10<6:10:32,  1.49s/it]
 39%|███▉      | 9506/24420 [3:57:11<6:10:03,  1.49s/it]
 39%|███▉      | 9507/24420 [3:57:13<6:10:01,  1.49s/it]
 39%|███▉      | 9508/24420 [3:57:14<6:09:52,  1.49s/it]
 39%|███▉      | 9509/24420 [3:57:16<6:09:56,  1.49s/it]
 39%|███▉      | 9510/24420 [3:57:17<6:09:40,  1.49s/it]
                                                        
{'loss': 0.1005, 'learning_rate': 5e-05, 'epoch': 1.95}

 39%|███▉      | 9510/24420 [3:57:17<6:09:40,  1.49s/it]
 39%|███▉      | 9511/24420 [3:57:19<6:10:27,  1.49s/it]
 39%|███▉      | 9512/24420 [3:57:20<6:10:13,  1.49s/it]
 39%|███▉      | 9513/24420 [3:57:22<6:10:37,  1.49s/it]
 39%|███▉      | 9514/24420 [3:57:23<6:10:20,  1.49s/it]
 39%|███▉      | 9515/24420 [3:57:25<6:10:42,  1.49s/it]
                                                        
{'loss': 0.1019, 'learning_rate': 5e-05, 'epoch': 1.95}

 39%|███▉      | 9515/24420 [3:57:25<6:10:42,  1.49s/it]
 39%|███▉      | 9516/24420 [3:57:26<6:10:36,  1.49s/it]
 39%|███▉      | 9517/24420 [3:57:28<6:10:26,  1.49s/it]
 39%|███▉      | 9518/24420 [3:57:29<6:10:23,  1.49s/it]
 39%|███▉      | 9519/24420 [3:57:31<6:09:56,  1.49s/it]
 39%|███▉      | 9520/24420 [3:57:32<6:09:31,  1.49s/it]
                                                        
{'loss': 0.1093, 'learning_rate': 5e-05, 'epoch': 1.95}

 39%|███▉      | 9520/24420 [3:57:32<6:09:31,  1.49s/it]
 39%|███▉      | 9521/24420 [3:57:34<6:09:38,  1.49s/it]
 39%|███▉      | 9522/24420 [3:57:35<6:09:26,  1.49s/it]
 39%|███▉      | 9523/24420 [3:57:37<6:08:57,  1.49s/it]
 39%|███▉      | 9524/24420 [3:57:38<6:09:01,  1.49s/it]
 39%|███▉      | 9525/24420 [3:57:40<6:09:05,  1.49s/it]
                                                        
{'loss': 0.1047, 'learning_rate': 5e-05, 'epoch': 1.95}

 39%|███▉      | 9525/24420 [3:57:40<6:09:05,  1.49s/it]
 39%|███▉      | 9526/24420 [3:57:41<6:09:35,  1.49s/it]
 39%|███▉      | 9527/24420 [3:57:43<6:08:56,  1.49s/it]
 39%|███▉      | 9528/24420 [3:57:44<6:09:27,  1.49s/it]
 39%|███▉      | 9529/24420 [3:57:45<6:09:06,  1.49s/it]
 39%|███▉      | 9530/24420 [3:57:47<6:09:05,  1.49s/it]
                                                        
{'loss': 0.1034, 'learning_rate': 5e-05, 'epoch': 1.95}

 39%|███▉      | 9530/24420 [3:57:47<6:09:05,  1.49s/it]
 39%|███▉      | 9531/24420 [3:57:48<6:09:26,  1.49s/it]
 39%|███▉      | 9532/24420 [3:57:50<6:09:14,  1.49s/it]
 39%|███▉      | 9533/24420 [3:57:51<6:09:46,  1.49s/it]
 39%|███▉      | 9534/24420 [3:57:53<6:09:20,  1.49s/it]
 39%|███▉      | 9535/24420 [3:57:54<6:08:40,  1.49s/it]
                                                        
{'loss': 0.1293, 'learning_rate': 5e-05, 'epoch': 1.95}

 39%|███▉      | 9535/24420 [3:57:54<6:08:40,  1.49s/it]
 39%|███▉      | 9536/24420 [3:57:56<6:09:00,  1.49s/it]
 39%|███▉      | 9537/24420 [3:57:57<6:09:17,  1.49s/it]
 39%|███▉      | 9538/24420 [3:57:59<6:08:09,  1.48s/it]
 39%|███▉      | 9539/24420 [3:58:00<6:08:59,  1.49s/it]
 39%|███▉      | 9540/24420 [3:58:02<6:09:08,  1.49s/it]
                                                        
{'loss': 0.1129, 'learning_rate': 5e-05, 'epoch': 1.95}

 39%|███▉      | 9540/24420 [3:58:02<6:09:08,  1.49s/it]
 39%|███▉      | 9541/24420 [3:58:03<6:09:08,  1.49s/it]
 39%|███▉      | 9542/24420 [3:58:05<6:08:51,  1.49s/it]
 39%|███▉      | 9543/24420 [3:58:06<6:08:49,  1.49s/it]
 39%|███▉      | 9544/24420 [3:58:08<6:08:46,  1.49s/it]
 39%|███▉      | 9545/24420 [3:58:09<6:08:25,  1.49s/it]
                                                        
{'loss': 0.1054, 'learning_rate': 5e-05, 'epoch': 1.95}

 39%|███▉      | 9545/24420 [3:58:09<6:08:25,  1.49s/it]
 39%|███▉      | 9546/24420 [3:58:11<6:09:54,  1.49s/it]
 39%|███▉      | 9547/24420 [3:58:12<6:09:12,  1.49s/it]
 39%|███▉      | 9548/24420 [3:58:14<6:09:20,  1.49s/it]
 39%|███▉      | 9549/24420 [3:58:15<6:08:55,  1.49s/it]
 39%|███▉      | 9550/24420 [3:58:17<6:08:40,  1.49s/it]
                                                        
{'loss': 0.0903, 'learning_rate': 5e-05, 'epoch': 1.96}

 39%|███▉      | 9550/24420 [3:58:17<6:08:40,  1.49s/it]
 39%|███▉      | 9551/24420 [3:58:18<6:09:00,  1.49s/it]
 39%|███▉      | 9552/24420 [3:58:20<6:09:33,  1.49s/it]
 39%|███▉      | 9553/24420 [3:58:21<6:09:07,  1.49s/it]
 39%|███▉      | 9554/24420 [3:58:23<6:09:07,  1.49s/it]
 39%|███▉      | 9555/24420 [3:58:24<6:08:32,  1.49s/it]
                                                        
{'loss': 0.11, 'learning_rate': 5e-05, 'epoch': 1.96}

 39%|███▉      | 9555/24420 [3:58:24<6:08:32,  1.49s/it]
 39%|███▉      | 9556/24420 [3:58:26<6:10:20,  1.49s/it]
 39%|███▉      | 9557/24420 [3:58:27<6:09:30,  1.49s/it]
 39%|███▉      | 9558/24420 [3:58:29<6:08:55,  1.49s/it]
 39%|███▉      | 9559/24420 [3:58:30<6:09:19,  1.49s/it]
 39%|███▉      | 9560/24420 [3:58:32<6:08:38,  1.49s/it]
                                                        
{'loss': 0.1043, 'learning_rate': 5e-05, 'epoch': 1.96}

 39%|███▉      | 9560/24420 [3:58:32<6:08:38,  1.49s/it]
 39%|███▉      | 9561/24420 [3:58:33<6:08:44,  1.49s/it]
 39%|███▉      | 9562/24420 [3:58:35<6:08:25,  1.49s/it]
 39%|███▉      | 9563/24420 [3:58:36<6:08:02,  1.49s/it]
 39%|███▉      | 9564/24420 [3:58:38<6:07:30,  1.48s/it]
 39%|███▉      | 9565/24420 [3:58:39<6:07:53,  1.49s/it]
                                                        
{'loss': 0.0926, 'learning_rate': 5e-05, 'epoch': 1.96}

 39%|███▉      | 9565/24420 [3:58:39<6:07:53,  1.49s/it]
 39%|███▉      | 9566/24420 [3:58:41<6:07:33,  1.48s/it]
 39%|███▉      | 9567/24420 [3:58:42<6:07:26,  1.48s/it]
 39%|███▉      | 9568/24420 [3:58:44<6:07:18,  1.48s/it]
 39%|███▉      | 9569/24420 [3:58:45<6:07:08,  1.48s/it]
 39%|███▉      | 9570/24420 [3:58:46<6:07:33,  1.49s/it]
                                                        
{'loss': 0.101, 'learning_rate': 5e-05, 'epoch': 1.96}

 39%|███▉      | 9570/24420 [3:58:47<6:07:33,  1.49s/it]
 39%|███▉      | 9571/24420 [3:58:48<6:07:01,  1.48s/it]
 39%|███▉      | 9572/24420 [3:58:49<6:07:17,  1.48s/it]
 39%|███▉      | 9573/24420 [3:58:51<6:07:30,  1.49s/it]
 39%|███▉      | 9574/24420 [3:58:52<6:07:16,  1.48s/it]
 39%|███▉      | 9575/24420 [3:58:54<6:07:40,  1.49s/it]
                                                        
{'loss': 0.0949, 'learning_rate': 5e-05, 'epoch': 1.96}

 39%|███▉      | 9575/24420 [3:58:54<6:07:40,  1.49s/it]
 39%|███▉      | 9576/24420 [3:58:55<6:07:59,  1.49s/it]
 39%|███▉      | 9577/24420 [3:58:57<6:07:50,  1.49s/it]
 39%|███▉      | 9578/24420 [3:58:58<6:07:26,  1.49s/it]
 39%|███▉      | 9579/24420 [3:59:00<6:07:51,  1.49s/it]
 39%|███▉      | 9580/24420 [3:59:01<6:08:04,  1.49s/it]
                                                        
{'loss': 0.0997, 'learning_rate': 5e-05, 'epoch': 1.96}

 39%|███▉      | 9580/24420 [3:59:01<6:08:04,  1.49s/it]
 39%|███▉      | 9581/24420 [3:59:03<6:08:08,  1.49s/it]
 39%|███▉      | 9582/24420 [3:59:04<6:08:33,  1.49s/it]
 39%|███▉      | 9583/24420 [3:59:06<6:08:10,  1.49s/it]
 39%|███▉      | 9584/24420 [3:59:07<6:07:08,  1.48s/it]
 39%|███▉      | 9585/24420 [3:59:09<6:06:45,  1.48s/it]
                                                        
{'loss': 0.1033, 'learning_rate': 5e-05, 'epoch': 1.96}

 39%|███▉      | 9585/24420 [3:59:09<6:06:45,  1.48s/it]
 39%|███▉      | 9586/24420 [3:59:10<6:06:57,  1.48s/it]
 39%|███▉      | 9587/24420 [3:59:12<6:06:52,  1.48s/it]
 39%|███▉      | 9588/24420 [3:59:13<6:06:18,  1.48s/it]
 39%|███▉      | 9589/24420 [3:59:15<6:05:50,  1.48s/it]
 39%|███▉      | 9590/24420 [3:59:16<6:05:51,  1.48s/it]
                                                        
{'loss': 0.0928, 'learning_rate': 5e-05, 'epoch': 1.96}

 39%|███▉      | 9590/24420 [3:59:16<6:05:51,  1.48s/it]
 39%|███▉      | 9591/24420 [3:59:18<6:06:17,  1.48s/it]
 39%|███▉      | 9592/24420 [3:59:19<6:07:22,  1.49s/it]
 39%|███▉      | 9593/24420 [3:59:21<6:07:08,  1.49s/it]
 39%|███▉      | 9594/24420 [3:59:22<6:07:20,  1.49s/it]
 39%|███▉      | 9595/24420 [3:59:24<6:07:24,  1.49s/it]
                                                        
{'loss': 0.101, 'learning_rate': 5e-05, 'epoch': 1.96}

 39%|███▉      | 9595/24420 [3:59:24<6:07:24,  1.49s/it]
 39%|███▉      | 9596/24420 [3:59:25<6:07:34,  1.49s/it]
 39%|███▉      | 9597/24420 [3:59:27<6:06:58,  1.49s/it]
 39%|███▉      | 9598/24420 [3:59:28<6:06:53,  1.49s/it]
 39%|███▉      | 9599/24420 [3:59:30<6:07:13,  1.49s/it]
 39%|███▉      | 9600/24420 [3:59:31<6:06:25,  1.48s/it]
                                                        
{'loss': 0.1018, 'learning_rate': 5e-05, 'epoch': 1.97}

 39%|███▉      | 9600/24420 [3:59:31<6:06:25,  1.48s/it]
 39%|███▉      | 9601/24420 [3:59:36<9:53:50,  2.40s/it]
 39%|███▉      | 9602/24420 [3:59:37<8:46:01,  2.13s/it]
 39%|███▉      | 9603/24420 [3:59:39<7:58:18,  1.94s/it]
 39%|███▉      | 9604/24420 [3:59:40<7:24:09,  1.80s/it]
 39%|███▉      | 9605/24420 [3:59:42<7:01:02,  1.71s/it]
                                                        
{'loss': 0.1011, 'learning_rate': 5e-05, 'epoch': 1.97}

 39%|███▉      | 9605/24420 [3:59:42<7:01:02,  1.71s/it]
 39%|███▉      | 9606/24420 [3:59:43<6:45:35,  1.64s/it]
 39%|███▉      | 9607/24420 [3:59:45<6:34:04,  1.60s/it]
 39%|███▉      | 9608/24420 [3:59:46<6:25:55,  1.56s/it]
 39%|███▉      | 9609/24420 [3:59:48<6:20:00,  1.54s/it]
 39%|███▉      | 9610/24420 [3:59:49<6:15:35,  1.52s/it]
                                                        
{'loss': 0.125, 'learning_rate': 5e-05, 'epoch': 1.97}

 39%|███▉      | 9610/24420 [3:59:49<6:15:35,  1.52s/it]
 39%|███▉      | 9611/24420 [3:59:50<6:13:52,  1.51s/it]
 39%|███▉      | 9612/24420 [3:59:52<6:11:52,  1.51s/it]
 39%|███▉      | 9613/24420 [3:59:53<6:09:44,  1.50s/it]
 39%|███▉      | 9614/24420 [3:59:55<6:08:49,  1.49s/it]
 39%|███▉      | 9615/24420 [3:59:56<6:07:41,  1.49s/it]
                                                        
{'loss': 0.1161, 'learning_rate': 5e-05, 'epoch': 1.97}

 39%|███▉      | 9615/24420 [3:59:56<6:07:41,  1.49s/it]
 39%|███▉      | 9616/24420 [3:59:58<6:07:11,  1.49s/it]
 39%|███▉      | 9617/24420 [3:59:59<6:07:08,  1.49s/it]
 39%|███▉      | 9618/24420 [4:00:01<6:07:08,  1.49s/it]
 39%|███▉      | 9619/24420 [4:00:02<6:06:30,  1.49s/it]
 39%|███▉      | 9620/24420 [4:00:04<6:06:18,  1.49s/it]
                                                        
{'loss': 0.1178, 'learning_rate': 5e-05, 'epoch': 1.97}

 39%|███▉      | 9620/24420 [4:00:04<6:06:18,  1.49s/it]
 39%|███▉      | 9621/24420 [4:00:05<6:06:27,  1.49s/it]
 39%|███▉      | 9622/24420 [4:00:07<6:06:25,  1.49s/it]
 39%|███▉      | 9623/24420 [4:00:08<6:07:10,  1.49s/it]
 39%|███▉      | 9624/24420 [4:00:10<6:07:08,  1.49s/it]
 39%|███▉      | 9625/24420 [4:00:11<6:06:59,  1.49s/it]
                                                        
{'loss': 0.1208, 'learning_rate': 5e-05, 'epoch': 1.97}

 39%|███▉      | 9625/24420 [4:00:11<6:06:59,  1.49s/it]
 39%|███▉      | 9626/24420 [4:00:13<6:06:20,  1.49s/it]
 39%|███▉      | 9627/24420 [4:00:14<6:05:57,  1.48s/it]
 39%|███▉      | 9628/24420 [4:00:16<6:05:41,  1.48s/it]
 39%|███▉      | 9629/24420 [4:00:17<6:05:36,  1.48s/it]
 39%|███▉      | 9630/24420 [4:00:19<6:05:43,  1.48s/it]
                                                        
{'loss': 0.1028, 'learning_rate': 5e-05, 'epoch': 1.97}

 39%|███▉      | 9630/24420 [4:00:19<6:05:43,  1.48s/it]
 39%|███▉      | 9631/24420 [4:00:20<6:06:22,  1.49s/it]
 39%|███▉      | 9632/24420 [4:00:22<6:05:53,  1.48s/it]
 39%|███▉      | 9633/24420 [4:00:23<6:05:15,  1.48s/it]
 39%|███▉      | 9634/24420 [4:00:25<6:05:08,  1.48s/it]
 39%|███▉      | 9635/24420 [4:00:26<6:05:39,  1.48s/it]
                                                        
{'loss': 0.0909, 'learning_rate': 5e-05, 'epoch': 1.97}

 39%|███▉      | 9635/24420 [4:00:26<6:05:39,  1.48s/it]
 39%|███▉      | 9636/24420 [4:00:28<6:05:47,  1.48s/it]
 39%|███▉      | 9637/24420 [4:00:29<6:05:37,  1.48s/it]
 39%|███▉      | 9638/24420 [4:00:31<6:05:33,  1.48s/it]
 39%|███▉      | 9639/24420 [4:00:32<6:04:57,  1.48s/it]
 39%|███▉      | 9640/24420 [4:00:34<6:04:58,  1.48s/it]
                                                        
{'loss': 0.1258, 'learning_rate': 5e-05, 'epoch': 1.97}

 39%|███▉      | 9640/24420 [4:00:34<6:04:58,  1.48s/it]
 39%|███▉      | 9641/24420 [4:00:35<6:04:58,  1.48s/it]
 39%|███▉      | 9642/24420 [4:00:36<6:04:53,  1.48s/it]
 39%|███▉      | 9643/24420 [4:00:38<6:04:57,  1.48s/it]
 39%|███▉      | 9644/24420 [4:00:39<6:04:54,  1.48s/it]
 39%|███▉      | 9645/24420 [4:00:41<6:04:42,  1.48s/it]
                                                        
{'loss': 0.0922, 'learning_rate': 5e-05, 'epoch': 1.97}

 39%|███▉      | 9645/24420 [4:00:41<6:04:42,  1.48s/it]
 40%|███▉      | 9646/24420 [4:00:42<6:04:56,  1.48s/it]
 40%|███▉      | 9647/24420 [4:00:44<6:05:33,  1.48s/it]
 40%|███▉      | 9648/24420 [4:00:45<6:05:54,  1.49s/it]
 40%|███▉      | 9649/24420 [4:00:47<6:06:33,  1.49s/it]
 40%|███▉      | 9650/24420 [4:00:48<6:06:40,  1.49s/it]
                                                        
{'loss': 0.1085, 'learning_rate': 5e-05, 'epoch': 1.98}

 40%|███▉      | 9650/24420 [4:00:48<6:06:40,  1.49s/it]
 40%|███▉      | 9651/24420 [4:00:50<6:07:11,  1.49s/it]
 40%|███▉      | 9652/24420 [4:00:51<6:06:28,  1.49s/it]
 40%|███▉      | 9653/24420 [4:00:53<6:06:22,  1.49s/it]
 40%|███▉      | 9654/24420 [4:00:54<6:06:17,  1.49s/it]
 40%|███▉      | 9655/24420 [4:00:56<6:05:51,  1.49s/it]
                                                        
{'loss': 0.1164, 'learning_rate': 5e-05, 'epoch': 1.98}

 40%|███▉      | 9655/24420 [4:00:56<6:05:51,  1.49s/it]
 40%|███▉      | 9656/24420 [4:00:57<6:05:33,  1.49s/it]
 40%|███▉      | 9657/24420 [4:00:59<6:05:57,  1.49s/it]
 40%|███▉      | 9658/24420 [4:01:00<6:05:32,  1.49s/it]
 40%|███▉      | 9659/24420 [4:01:02<6:04:55,  1.48s/it]
 40%|███▉      | 9660/24420 [4:01:03<6:04:29,  1.48s/it]
                                                        
{'loss': 0.0965, 'learning_rate': 5e-05, 'epoch': 1.98}

 40%|███▉      | 9660/24420 [4:01:03<6:04:29,  1.48s/it]
 40%|███▉      | 9661/24420 [4:01:05<6:05:18,  1.49s/it]
 40%|███▉      | 9662/24420 [4:01:06<6:05:34,  1.49s/it]
 40%|███▉      | 9663/24420 [4:01:08<6:05:33,  1.49s/it]
 40%|███▉      | 9664/24420 [4:01:09<6:05:32,  1.49s/it]
 40%|███▉      | 9665/24420 [4:01:11<6:05:40,  1.49s/it]
                                                        
{'loss': 0.115, 'learning_rate': 5e-05, 'epoch': 1.98}

 40%|███▉      | 9665/24420 [4:01:11<6:05:40,  1.49s/it]
 40%|███▉      | 9666/24420 [4:01:12<6:05:04,  1.48s/it]
 40%|███▉      | 9667/24420 [4:01:14<6:04:34,  1.48s/it]
 40%|███▉      | 9668/24420 [4:01:15<6:05:07,  1.49s/it]
 40%|███▉      | 9669/24420 [4:01:17<6:05:27,  1.49s/it]
 40%|███▉      | 9670/24420 [4:01:18<6:06:39,  1.49s/it]
                                                        
{'loss': 0.0802, 'learning_rate': 5e-05, 'epoch': 1.98}

 40%|███▉      | 9670/24420 [4:01:18<6:06:39,  1.49s/it]
 40%|███▉      | 9671/24420 [4:01:20<6:06:11,  1.49s/it]
 40%|███▉      | 9672/24420 [4:01:21<6:06:23,  1.49s/it]
 40%|███▉      | 9673/24420 [4:01:23<6:06:47,  1.49s/it]
 40%|███▉      | 9674/24420 [4:01:24<6:06:31,  1.49s/it]
 40%|███▉      | 9675/24420 [4:01:26<6:06:11,  1.49s/it]
                                                        
{'loss': 0.1221, 'learning_rate': 5e-05, 'epoch': 1.98}

 40%|███▉      | 9675/24420 [4:01:26<6:06:11,  1.49s/it]
 40%|███▉      | 9676/24420 [4:01:27<6:05:45,  1.49s/it]
 40%|███▉      | 9677/24420 [4:01:29<6:05:18,  1.49s/it]
 40%|███▉      | 9678/24420 [4:01:30<6:04:40,  1.48s/it]
 40%|███▉      | 9679/24420 [4:01:32<6:05:12,  1.49s/it]
 40%|███▉      | 9680/24420 [4:01:33<6:05:04,  1.49s/it]
                                                        
{'loss': 0.101, 'learning_rate': 5e-05, 'epoch': 1.98}

 40%|███▉      | 9680/24420 [4:01:33<6:05:04,  1.49s/it]
 40%|███▉      | 9681/24420 [4:01:34<6:04:52,  1.49s/it]
 40%|███▉      | 9682/24420 [4:01:36<6:04:26,  1.48s/it]
 40%|███▉      | 9683/24420 [4:01:37<6:04:00,  1.48s/it]
 40%|███▉      | 9684/24420 [4:01:39<6:04:23,  1.48s/it]
 40%|███▉      | 9685/24420 [4:01:40<6:04:16,  1.48s/it]
                                                        
{'loss': 0.1071, 'learning_rate': 5e-05, 'epoch': 1.98}

 40%|███▉      | 9685/24420 [4:01:40<6:04:16,  1.48s/it]
 40%|███▉      | 9686/24420 [4:01:42<6:04:40,  1.49s/it]
 40%|███▉      | 9687/24420 [4:01:43<6:04:47,  1.49s/it]
 40%|███▉      | 9688/24420 [4:01:45<6:04:29,  1.48s/it]
 40%|███▉      | 9689/24420 [4:01:46<6:04:00,  1.48s/it]
 40%|███▉      | 9690/24420 [4:01:48<6:03:41,  1.48s/it]
                                                        
{'loss': 0.1088, 'learning_rate': 5e-05, 'epoch': 1.98}

 40%|███▉      | 9690/24420 [4:01:48<6:03:41,  1.48s/it]
 40%|███▉      | 9691/24420 [4:01:49<6:04:03,  1.48s/it]
 40%|███▉      | 9692/24420 [4:01:51<6:03:20,  1.48s/it]
 40%|███▉      | 9693/24420 [4:01:52<6:03:13,  1.48s/it]
 40%|███▉      | 9694/24420 [4:01:54<6:03:46,  1.48s/it]
 40%|███▉      | 9695/24420 [4:01:55<6:03:44,  1.48s/it]
                                                        
{'loss': 0.0979, 'learning_rate': 5e-05, 'epoch': 1.99}

 40%|███▉      | 9695/24420 [4:01:55<6:03:44,  1.48s/it]
 40%|███▉      | 9696/24420 [4:01:57<6:03:32,  1.48s/it]
 40%|███▉      | 9697/24420 [4:01:58<6:04:11,  1.48s/it]
 40%|███▉      | 9698/24420 [4:02:00<6:04:16,  1.48s/it]
 40%|███▉      | 9699/24420 [4:02:01<6:04:11,  1.48s/it]
 40%|███▉      | 9700/24420 [4:02:03<6:04:04,  1.48s/it]
                                                        
{'loss': 0.1038, 'learning_rate': 5e-05, 'epoch': 1.99}

 40%|███▉      | 9700/24420 [4:02:03<6:04:04,  1.48s/it]
 40%|███▉      | 9701/24420 [4:02:04<6:04:28,  1.49s/it]
 40%|███▉      | 9702/24420 [4:02:06<6:04:41,  1.49s/it]
 40%|███▉      | 9703/24420 [4:02:07<6:04:38,  1.49s/it]
 40%|███▉      | 9704/24420 [4:02:09<6:04:14,  1.49s/it]
 40%|███▉      | 9705/24420 [4:02:10<6:03:28,  1.48s/it]
                                                        
{'loss': 0.1108, 'learning_rate': 5e-05, 'epoch': 1.99}

 40%|███▉      | 9705/24420 [4:02:10<6:03:28,  1.48s/it]
 40%|███▉      | 9706/24420 [4:02:12<6:03:22,  1.48s/it]
 40%|███▉      | 9707/24420 [4:02:13<6:03:52,  1.48s/it]
 40%|███▉      | 9708/24420 [4:02:15<6:03:59,  1.48s/it]
 40%|███▉      | 9709/24420 [4:02:16<6:03:21,  1.48s/it]
 40%|███▉      | 9710/24420 [4:02:17<6:03:20,  1.48s/it]
                                                        
{'loss': 0.1133, 'learning_rate': 5e-05, 'epoch': 1.99}

 40%|███▉      | 9710/24420 [4:02:17<6:03:20,  1.48s/it]
 40%|███▉      | 9711/24420 [4:02:19<6:03:55,  1.48s/it]
 40%|███▉      | 9712/24420 [4:02:20<6:03:44,  1.48s/it]
 40%|███▉      | 9713/24420 [4:02:22<6:04:35,  1.49s/it]
 40%|███▉      | 9714/24420 [4:02:23<6:04:04,  1.49s/it]
 40%|███▉      | 9715/24420 [4:02:25<6:04:06,  1.49s/it]
                                                        
{'loss': 0.1104, 'learning_rate': 5e-05, 'epoch': 1.99}

 40%|███▉      | 9715/24420 [4:02:25<6:04:06,  1.49s/it]
 40%|███▉      | 9716/24420 [4:02:26<6:03:53,  1.48s/it]
 40%|███▉      | 9717/24420 [4:02:28<6:03:53,  1.48s/it]
 40%|███▉      | 9718/24420 [4:02:29<6:03:30,  1.48s/it]
 40%|███▉      | 9719/24420 [4:02:31<6:03:31,  1.48s/it]
 40%|███▉      | 9720/24420 [4:02:32<6:03:16,  1.48s/it]
                                                        
{'loss': 0.1061, 'learning_rate': 5e-05, 'epoch': 1.99}

 40%|███▉      | 9720/24420 [4:02:32<6:03:16,  1.48s/it]
 40%|███▉      | 9721/24420 [4:02:34<6:02:53,  1.48s/it]
 40%|███▉      | 9722/24420 [4:02:35<6:02:10,  1.48s/it]
 40%|███▉      | 9723/24420 [4:02:37<6:02:42,  1.48s/it]
 40%|███▉      | 9724/24420 [4:02:38<6:02:41,  1.48s/it]
 40%|███▉      | 9725/24420 [4:02:40<6:02:48,  1.48s/it]
                                                        
{'loss': 0.0883, 'learning_rate': 5e-05, 'epoch': 1.99}

 40%|███▉      | 9725/24420 [4:02:40<6:02:48,  1.48s/it]
 40%|███▉      | 9726/24420 [4:02:41<6:03:05,  1.48s/it]
 40%|███▉      | 9727/24420 [4:02:43<6:03:08,  1.48s/it]
 40%|███▉      | 9728/24420 [4:02:44<6:03:13,  1.48s/it]
 40%|███▉      | 9729/24420 [4:02:46<6:02:56,  1.48s/it]
 40%|███▉      | 9730/24420 [4:02:47<6:03:17,  1.48s/it]
                                                        
{'loss': 0.1247, 'learning_rate': 5e-05, 'epoch': 1.99}

 40%|███▉      | 9730/24420 [4:02:47<6:03:17,  1.48s/it]
 40%|███▉      | 9731/24420 [4:02:49<6:03:27,  1.48s/it]
 40%|███▉      | 9732/24420 [4:02:50<6:03:29,  1.48s/it]
 40%|███▉      | 9733/24420 [4:02:52<6:03:26,  1.48s/it]
 40%|███▉      | 9734/24420 [4:02:53<6:03:21,  1.48s/it]
 40%|███▉      | 9735/24420 [4:02:55<6:03:04,  1.48s/it]
                                                        
{'loss': 0.1293, 'learning_rate': 5e-05, 'epoch': 1.99}

 40%|███▉      | 9735/24420 [4:02:55<6:03:04,  1.48s/it]
 40%|███▉      | 9736/24420 [4:02:56<6:04:04,  1.49s/it]
 40%|███▉      | 9737/24420 [4:02:58<6:03:52,  1.49s/it]
 40%|███▉      | 9738/24420 [4:02:59<6:04:10,  1.49s/it]
 40%|███▉      | 9739/24420 [4:03:01<6:04:17,  1.49s/it]
 40%|███▉      | 9740/24420 [4:03:02<6:03:38,  1.49s/it]
                                                        
{'loss': 0.0996, 'learning_rate': 5e-05, 'epoch': 1.99}

 40%|███▉      | 9740/24420 [4:03:02<6:03:38,  1.49s/it]
 40%|███▉      | 9741/24420 [4:03:04<6:03:47,  1.49s/it]
 40%|███▉      | 9742/24420 [4:03:05<6:03:11,  1.48s/it]
 40%|███▉      | 9743/24420 [4:03:06<6:04:00,  1.49s/it]
 40%|███▉      | 9744/24420 [4:03:08<6:03:36,  1.49s/it]
 40%|███▉      | 9745/24420 [4:03:09<6:03:49,  1.49s/it]
                                                        
{'loss': 0.0975, 'learning_rate': 5e-05, 'epoch': 2.0}

 40%|███▉      | 9745/24420 [4:03:09<6:03:49,  1.49s/it]
 40%|███▉      | 9746/24420 [4:03:11<6:03:44,  1.49s/it]
 40%|███▉      | 9747/24420 [4:03:12<6:03:05,  1.48s/it]
 40%|███▉      | 9748/24420 [4:03:14<6:01:57,  1.48s/it]
 40%|███▉      | 9749/24420 [4:03:15<6:01:42,  1.48s/it]
 40%|███▉      | 9750/24420 [4:03:17<6:02:29,  1.48s/it]
                                                        
{'loss': 0.1109, 'learning_rate': 5e-05, 'epoch': 2.0}

 40%|███▉      | 9750/24420 [4:03:17<6:02:29,  1.48s/it]
 40%|███▉      | 9751/24420 [4:03:18<6:02:49,  1.48s/it]
 40%|███▉      | 9752/24420 [4:03:20<6:03:11,  1.49s/it]
 40%|███▉      | 9753/24420 [4:03:21<6:03:29,  1.49s/it]
 40%|███▉      | 9754/24420 [4:03:23<6:03:40,  1.49s/it]
 40%|███▉      | 9755/24420 [4:03:24<6:02:54,  1.48s/it]
                                                        
{'loss': 0.1063, 'learning_rate': 5e-05, 'epoch': 2.0}

 40%|███▉      | 9755/24420 [4:03:24<6:02:54,  1.48s/it]
 40%|███▉      | 9756/24420 [4:03:26<6:03:25,  1.49s/it]
 40%|███▉      | 9757/24420 [4:03:27<6:02:09,  1.48s/it]
 40%|███▉      | 9758/24420 [4:03:29<6:04:11,  1.49s/it]
 40%|███▉      | 9759/24420 [4:03:30<6:03:52,  1.49s/it]
 40%|███▉      | 9760/24420 [4:03:32<6:03:52,  1.49s/it]
                                                        
{'loss': 0.1216, 'learning_rate': 5e-05, 'epoch': 2.0}

 40%|███▉      | 9760/24420 [4:03:32<6:03:52,  1.49s/it]
 40%|███▉      | 9761/24420 [4:03:33<6:03:31,  1.49s/it]
 40%|███▉      | 9762/24420 [4:03:35<6:03:40,  1.49s/it]
 40%|███▉      | 9763/24420 [4:03:36<6:03:36,  1.49s/it]
 40%|███▉      | 9764/24420 [4:03:38<6:03:44,  1.49s/it]
 40%|███▉      | 9765/24420 [4:03:39<6:03:05,  1.49s/it]
                                                        
{'loss': 0.1129, 'learning_rate': 5e-05, 'epoch': 2.0}

 40%|███▉      | 9765/24420 [4:03:39<6:03:05,  1.49s/it]
 40%|███▉      | 9766/24420 [4:03:41<6:02:26,  1.48s/it]
 40%|███▉      | 9767/24420 [4:03:42<6:02:03,  1.48s/it]
 40%|████      | 9768/24420 [4:03:44<6:02:18,  1.48s/it]
 40%|████      | 9769/24420 [4:03:45<6:05:47,  1.50s/it]
 40%|████      | 9770/24420 [4:03:47<6:05:00,  1.49s/it]
                                                        
{'loss': 0.087, 'learning_rate': 5e-05, 'epoch': 2.0}

 40%|████      | 9770/24420 [4:03:47<6:05:00,  1.49s/it]
 40%|████      | 9771/24420 [4:03:48<6:04:29,  1.49s/it]
 40%|████      | 9772/24420 [4:03:50<6:04:43,  1.49s/it]
 40%|████      | 9773/24420 [4:03:51<6:04:14,  1.49s/it]
 40%|████      | 9774/24420 [4:03:53<6:03:24,  1.49s/it]
 40%|████      | 9775/24420 [4:03:54<6:03:09,  1.49s/it]
                                                        
{'loss': 0.1247, 'learning_rate': 5e-05, 'epoch': 2.0}

 40%|████      | 9775/24420 [4:03:54<6:03:09,  1.49s/it]
 40%|████      | 9776/24420 [4:03:56<6:02:37,  1.49s/it]
 40%|████      | 9777/24420 [4:03:57<6:02:41,  1.49s/it]
 40%|████      | 9778/24420 [4:03:59<6:01:54,  1.48s/it]
 40%|████      | 9779/24420 [4:04:00<6:02:35,  1.49s/it]
 40%|████      | 9780/24420 [4:04:01<6:01:44,  1.48s/it]
                                                        
{'loss': 0.0955, 'learning_rate': 5e-05, 'epoch': 2.0}

 40%|████      | 9780/24420 [4:04:01<6:01:44,  1.48s/it]
 40%|████      | 9781/24420 [4:04:03<6:01:47,  1.48s/it]
 40%|████      | 9782/24420 [4:04:04<6:01:21,  1.48s/it]
 40%|████      | 9783/24420 [4:04:06<6:02:29,  1.49s/it]
 40%|████      | 9784/24420 [4:04:07<6:01:53,  1.48s/it]
 40%|████      | 9785/24420 [4:04:09<6:02:11,  1.48s/it]
                                                        
{'loss': 0.0857, 'learning_rate': 5e-05, 'epoch': 2.0}

 40%|████      | 9785/24420 [4:04:09<6:02:11,  1.48s/it]
 40%|████      | 9786/24420 [4:04:10<6:01:48,  1.48s/it]
 40%|████      | 9787/24420 [4:04:12<6:01:36,  1.48s/it]
 40%|████      | 9788/24420 [4:04:13<6:01:40,  1.48s/it]
 40%|████      | 9789/24420 [4:04:15<6:01:36,  1.48s/it]
 40%|████      | 9790/24420 [4:04:16<6:01:20,  1.48s/it]
                                                        
{'loss': 0.091, 'learning_rate': 5e-05, 'epoch': 2.0}

 40%|████      | 9790/24420 [4:04:16<6:01:20,  1.48s/it]
 40%|████      | 9791/24420 [4:04:18<6:01:26,  1.48s/it]
 40%|████      | 9792/24420 [4:04:19<6:01:33,  1.48s/it]
 40%|████      | 9793/24420 [4:04:21<6:01:46,  1.48s/it]
 40%|████      | 9794/24420 [4:04:22<6:01:32,  1.48s/it]
 40%|████      | 9795/24420 [4:04:24<6:01:25,  1.48s/it]
                                                        
{'loss': 0.1066, 'learning_rate': 5e-05, 'epoch': 2.01}

 40%|████      | 9795/24420 [4:04:24<6:01:25,  1.48s/it]
 40%|████      | 9796/24420 [4:04:25<6:01:08,  1.48s/it]
 40%|████      | 9797/24420 [4:04:27<6:01:20,  1.48s/it]
 40%|████      | 9798/24420 [4:04:28<6:01:46,  1.48s/it]
 40%|████      | 9799/24420 [4:04:30<6:01:20,  1.48s/it]
 40%|████      | 9800/24420 [4:04:31<6:01:58,  1.49s/it]
                                                        
{'loss': 0.0858, 'learning_rate': 5e-05, 'epoch': 2.01}

 40%|████      | 9800/24420 [4:04:31<6:01:58,  1.49s/it]
 40%|████      | 9801/24420 [4:04:37<11:08:52,  2.75s/it]
 40%|████      | 9802/24420 [4:04:38<9:36:09,  2.36s/it] 
 40%|████      | 9803/24420 [4:04:40<8:31:35,  2.10s/it]
 40%|████      | 9804/24420 [4:04:41<7:46:53,  1.92s/it]
 40%|████      | 9805/24420 [4:04:43<7:14:56,  1.79s/it]
                                                        
{'loss': 0.0943, 'learning_rate': 5e-05, 'epoch': 2.01}

 40%|████      | 9805/24420 [4:04:43<7:14:56,  1.79s/it]
 40%|████      | 9806/24420 [4:04:44<6:52:56,  1.70s/it]
 40%|████      | 9807/24420 [4:04:46<6:37:19,  1.63s/it]
 40%|████      | 9808/24420 [4:04:47<6:26:26,  1.59s/it]
 40%|████      | 9809/24420 [4:04:49<6:18:47,  1.56s/it]
 40%|████      | 9810/24420 [4:04:50<6:13:43,  1.53s/it]
                                                        
{'loss': 0.088, 'learning_rate': 5e-05, 'epoch': 2.01}

 40%|████      | 9810/24420 [4:04:50<6:13:43,  1.53s/it]
 40%|████      | 9811/24420 [4:04:52<6:09:53,  1.52s/it]
 40%|████      | 9812/24420 [4:04:53<6:06:50,  1.51s/it]
 40%|████      | 9813/24420 [4:04:55<6:05:08,  1.50s/it]
 40%|████      | 9814/24420 [4:04:56<6:04:30,  1.50s/it]
 40%|████      | 9815/24420 [4:04:58<6:03:19,  1.49s/it]
                                                        
{'loss': 0.1003, 'learning_rate': 5e-05, 'epoch': 2.01}

 40%|████      | 9815/24420 [4:04:58<6:03:19,  1.49s/it]
 40%|████      | 9816/24420 [4:04:59<6:03:10,  1.49s/it]
 40%|████      | 9817/24420 [4:05:01<6:02:35,  1.49s/it]
 40%|████      | 9818/24420 [4:05:02<6:02:21,  1.49s/it]
 40%|████      | 9819/24420 [4:05:04<6:01:57,  1.49s/it]
 40%|████      | 9820/24420 [4:05:05<6:01:07,  1.48s/it]
                                                        
{'loss': 0.1136, 'learning_rate': 5e-05, 'epoch': 2.01}

 40%|████      | 9820/24420 [4:05:05<6:01:07,  1.48s/it]
 40%|████      | 9821/24420 [4:05:07<6:01:00,  1.48s/it]
 40%|████      | 9822/24420 [4:05:08<6:00:27,  1.48s/it]
 40%|████      | 9823/24420 [4:05:09<5:59:52,  1.48s/it]
 40%|████      | 9824/24420 [4:05:11<5:59:58,  1.48s/it]
 40%|████      | 9825/24420 [4:05:12<6:00:12,  1.48s/it]
                                                        
{'loss': 0.0877, 'learning_rate': 5e-05, 'epoch': 2.01}

 40%|████      | 9825/24420 [4:05:12<6:00:12,  1.48s/it]
 40%|████      | 9826/24420 [4:05:14<5:59:51,  1.48s/it]
 40%|████      | 9827/24420 [4:05:15<5:59:54,  1.48s/it]
 40%|████      | 9828/24420 [4:05:17<5:59:46,  1.48s/it]
 40%|████      | 9829/24420 [4:05:18<5:59:41,  1.48s/it]
 40%|████      | 9830/24420 [4:05:20<5:59:31,  1.48s/it]
                                                        
{'loss': 0.0877, 'learning_rate': 5e-05, 'epoch': 2.01}

 40%|████      | 9830/24420 [4:05:20<5:59:31,  1.48s/it]
 40%|████      | 9831/24420 [4:05:21<6:01:40,  1.49s/it]
 40%|████      | 9832/24420 [4:05:23<6:39:56,  1.64s/it]
 40%|████      | 9833/24420 [4:05:25<6:28:02,  1.60s/it]
 40%|████      | 9834/24420 [4:05:26<6:19:46,  1.56s/it]
 40%|████      | 9835/24420 [4:05:28<6:14:23,  1.54s/it]
                                                        
{'loss': 0.1086, 'learning_rate': 5e-05, 'epoch': 2.01}

 40%|████      | 9835/24420 [4:05:28<6:14:23,  1.54s/it]
 40%|████      | 9836/24420 [4:05:29<6:11:03,  1.53s/it]
 40%|████      | 9837/24420 [4:05:31<6:07:26,  1.51s/it]
 40%|████      | 9838/24420 [4:05:32<6:04:56,  1.50s/it]
 40%|████      | 9839/24420 [4:05:34<6:03:40,  1.50s/it]
 40%|████      | 9840/24420 [4:05:35<6:03:02,  1.49s/it]
                                                        
{'loss': 0.084, 'learning_rate': 5e-05, 'epoch': 2.01}

 40%|████      | 9840/24420 [4:05:35<6:03:02,  1.49s/it]
 40%|████      | 9841/24420 [4:05:37<6:01:50,  1.49s/it]
 40%|████      | 9842/24420 [4:05:38<6:01:40,  1.49s/it]
 40%|████      | 9843/24420 [4:05:40<6:01:20,  1.49s/it]
 40%|████      | 9844/24420 [4:05:41<6:00:58,  1.49s/it]
 40%|████      | 9845/24420 [4:05:43<6:00:45,  1.49s/it]
                                                        
{'loss': 0.106, 'learning_rate': 5e-05, 'epoch': 2.02}

 40%|████      | 9845/24420 [4:05:43<6:00:45,  1.49s/it]
 40%|████      | 9846/24420 [4:05:44<6:00:35,  1.48s/it]
 40%|████      | 9847/24420 [4:05:46<6:01:16,  1.49s/it]
 40%|████      | 9848/24420 [4:05:47<6:00:29,  1.48s/it]
 40%|████      | 9849/24420 [4:05:49<6:00:13,  1.48s/it]
 40%|████      | 9850/24420 [4:05:50<6:00:21,  1.48s/it]
                                                        
{'loss': 0.1018, 'learning_rate': 5e-05, 'epoch': 2.02}

 40%|████      | 9850/24420 [4:05:50<6:00:21,  1.48s/it]
 40%|████      | 9851/24420 [4:05:52<6:00:08,  1.48s/it]
 40%|████      | 9852/24420 [4:05:53<5:59:55,  1.48s/it]
 40%|████      | 9853/24420 [4:05:54<5:59:50,  1.48s/it]
 40%|████      | 9854/24420 [4:05:56<5:59:58,  1.48s/it]
 40%|████      | 9855/24420 [4:05:57<6:00:20,  1.48s/it]
                                                        
{'loss': 0.1017, 'learning_rate': 5e-05, 'epoch': 2.02}

 40%|████      | 9855/24420 [4:05:57<6:00:20,  1.48s/it]
 40%|████      | 9856/24420 [4:05:59<6:00:29,  1.49s/it]
 40%|████      | 9857/24420 [4:06:00<6:00:18,  1.48s/it]
 40%|████      | 9858/24420 [4:06:02<6:00:38,  1.49s/it]
 40%|████      | 9859/24420 [4:06:03<6:00:48,  1.49s/it]
 40%|████      | 9860/24420 [4:06:05<5:59:51,  1.48s/it]
                                                        
{'loss': 0.0951, 'learning_rate': 5e-05, 'epoch': 2.02}

 40%|████      | 9860/24420 [4:06:05<5:59:51,  1.48s/it]
 40%|████      | 9861/24420 [4:06:06<5:59:54,  1.48s/it]
 40%|████      | 9862/24420 [4:06:08<5:59:54,  1.48s/it]
 40%|████      | 9863/24420 [4:06:09<5:59:18,  1.48s/it]
 40%|████      | 9864/24420 [4:06:11<5:59:18,  1.48s/it]
 40%|████      | 9865/24420 [4:06:12<5:59:31,  1.48s/it]
                                                        
{'loss': 0.0931, 'learning_rate': 5e-05, 'epoch': 2.02}

 40%|████      | 9865/24420 [4:06:12<5:59:31,  1.48s/it]
 40%|████      | 9866/24420 [4:06:14<6:00:22,  1.49s/it]
 40%|████      | 9867/24420 [4:06:15<5:59:34,  1.48s/it]
 40%|████      | 9868/24420 [4:06:17<5:59:01,  1.48s/it]
 40%|████      | 9869/24420 [4:06:18<5:59:07,  1.48s/it]
 40%|████      | 9870/24420 [4:06:20<6:00:09,  1.49s/it]
                                                        
{'loss': 0.0918, 'learning_rate': 5e-05, 'epoch': 2.02}

 40%|████      | 9870/24420 [4:06:20<6:00:09,  1.49s/it]
 40%|████      | 9871/24420 [4:06:21<6:00:25,  1.49s/it]
 40%|████      | 9872/24420 [4:06:23<6:00:38,  1.49s/it]
 40%|████      | 9873/24420 [4:06:24<6:00:22,  1.49s/it]
 40%|████      | 9874/24420 [4:06:26<5:59:45,  1.48s/it]
 40%|████      | 9875/24420 [4:06:27<5:59:57,  1.48s/it]
                                                        
{'loss': 0.1028, 'learning_rate': 5e-05, 'epoch': 2.02}

 40%|████      | 9875/24420 [4:06:27<5:59:57,  1.48s/it]
 40%|████      | 9876/24420 [4:06:29<5:59:31,  1.48s/it]
 40%|████      | 9877/24420 [4:06:30<5:59:17,  1.48s/it]
 40%|████      | 9878/24420 [4:06:32<5:59:43,  1.48s/it]
 40%|████      | 9879/24420 [4:06:33<5:59:14,  1.48s/it]
 40%|████      | 9880/24420 [4:06:35<5:59:42,  1.48s/it]
                                                        
{'loss': 0.0905, 'learning_rate': 5e-05, 'epoch': 2.02}

 40%|████      | 9880/24420 [4:06:35<5:59:42,  1.48s/it]
 40%|████      | 9881/24420 [4:06:36<5:59:58,  1.49s/it]
 40%|████      | 9882/24420 [4:06:38<5:59:58,  1.49s/it]
 40%|████      | 9883/24420 [4:06:39<6:00:00,  1.49s/it]
 40%|████      | 9884/24420 [4:06:41<5:59:46,  1.49s/it]
 40%|████      | 9885/24420 [4:06:42<5:59:49,  1.49s/it]
                                                        
{'loss': 0.1043, 'learning_rate': 5e-05, 'epoch': 2.02}

 40%|████      | 9885/24420 [4:06:42<5:59:49,  1.49s/it]
 40%|████      | 9886/24420 [4:06:43<5:59:50,  1.49s/it]
 40%|████      | 9887/24420 [4:06:45<5:59:28,  1.48s/it]
 40%|████      | 9888/24420 [4:06:46<5:59:32,  1.48s/it]
 40%|████      | 9889/24420 [4:06:48<5:58:49,  1.48s/it]
 40%|████      | 9890/24420 [4:06:49<5:58:37,  1.48s/it]
                                                        
{'loss': 0.0916, 'learning_rate': 5e-05, 'epoch': 2.02}

 40%|████      | 9890/24420 [4:06:49<5:58:37,  1.48s/it]
 41%|████      | 9891/24420 [4:06:51<5:59:28,  1.48s/it]
 41%|████      | 9892/24420 [4:06:52<5:59:13,  1.48s/it]
 41%|████      | 9893/24420 [4:06:54<5:58:57,  1.48s/it]
 41%|████      | 9894/24420 [4:06:55<5:58:39,  1.48s/it]
 41%|████      | 9895/24420 [4:06:57<5:58:28,  1.48s/it]
                                                        
{'loss': 0.1017, 'learning_rate': 5e-05, 'epoch': 2.03}

 41%|████      | 9895/24420 [4:06:57<5:58:28,  1.48s/it]
 41%|████      | 9896/24420 [4:06:58<5:58:43,  1.48s/it]
 41%|████      | 9897/24420 [4:07:00<5:58:06,  1.48s/it]
 41%|████      | 9898/24420 [4:07:01<5:58:30,  1.48s/it]
 41%|████      | 9899/24420 [4:07:03<5:59:33,  1.49s/it]
 41%|████      | 9900/24420 [4:07:04<5:59:42,  1.49s/it]
                                                        
{'loss': 0.1001, 'learning_rate': 5e-05, 'epoch': 2.03}

 41%|████      | 9900/24420 [4:07:04<5:59:42,  1.49s/it]
 41%|████      | 9901/24420 [4:07:06<5:59:05,  1.48s/it]
 41%|████      | 9902/24420 [4:07:07<5:58:27,  1.48s/it]
 41%|████      | 9903/24420 [4:07:09<5:58:37,  1.48s/it]
 41%|████      | 9904/24420 [4:07:10<5:58:45,  1.48s/it]
 41%|████      | 9905/24420 [4:07:12<5:58:05,  1.48s/it]
                                                        
{'loss': 0.0785, 'learning_rate': 5e-05, 'epoch': 2.03}

 41%|████      | 9905/24420 [4:07:12<5:58:05,  1.48s/it]
 41%|████      | 9906/24420 [4:07:13<5:58:38,  1.48s/it]
 41%|████      | 9907/24420 [4:07:15<5:58:47,  1.48s/it]
 41%|████      | 9908/24420 [4:07:16<5:59:12,  1.49s/it]
 41%|████      | 9909/24420 [4:07:18<5:59:04,  1.48s/it]
 41%|████      | 9910/24420 [4:07:19<5:58:35,  1.48s/it]
                                                        
{'loss': 0.0924, 'learning_rate': 5e-05, 'epoch': 2.03}

 41%|████      | 9910/24420 [4:07:19<5:58:35,  1.48s/it]
 41%|████      | 9911/24420 [4:07:21<5:58:47,  1.48s/it]
 41%|████      | 9912/24420 [4:07:22<5:58:31,  1.48s/it]
 41%|████      | 9913/24420 [4:07:24<5:58:08,  1.48s/it]
 41%|████      | 9914/24420 [4:07:25<5:57:26,  1.48s/it]
 41%|████      | 9915/24420 [4:07:26<5:58:12,  1.48s/it]
                                                        
{'loss': 0.0921, 'learning_rate': 5e-05, 'epoch': 2.03}

 41%|████      | 9915/24420 [4:07:26<5:58:12,  1.48s/it]
 41%|████      | 9916/24420 [4:07:28<5:58:14,  1.48s/it]
 41%|████      | 9917/24420 [4:07:29<5:58:31,  1.48s/it]
 41%|████      | 9918/24420 [4:07:31<5:58:07,  1.48s/it]
 41%|████      | 9919/24420 [4:07:32<5:58:16,  1.48s/it]
 41%|████      | 9920/24420 [4:07:34<5:57:59,  1.48s/it]
                                                        
{'loss': 0.108, 'learning_rate': 5e-05, 'epoch': 2.03}

 41%|████      | 9920/24420 [4:07:34<5:57:59,  1.48s/it]
 41%|████      | 9921/24420 [4:07:35<5:58:17,  1.48s/it]
 41%|████      | 9922/24420 [4:07:37<5:57:59,  1.48s/it]
 41%|████      | 9923/24420 [4:07:38<5:57:49,  1.48s/it]
 41%|████      | 9924/24420 [4:07:40<5:58:03,  1.48s/it]
 41%|████      | 9925/24420 [4:07:41<5:57:52,  1.48s/it]
                                                        
{'loss': 0.096, 'learning_rate': 5e-05, 'epoch': 2.03}

 41%|████      | 9925/24420 [4:07:41<5:57:52,  1.48s/it]
 41%|████      | 9926/24420 [4:07:43<5:58:03,  1.48s/it]
 41%|████      | 9927/24420 [4:07:44<5:57:11,  1.48s/it]
 41%|████      | 9928/24420 [4:07:46<5:57:18,  1.48s/it]
 41%|████      | 9929/24420 [4:07:47<5:58:32,  1.48s/it]
 41%|████      | 9930/24420 [4:07:49<5:58:12,  1.48s/it]
                                                        
{'loss': 0.084, 'learning_rate': 5e-05, 'epoch': 2.03}

 41%|████      | 9930/24420 [4:07:49<5:58:12,  1.48s/it]
 41%|████      | 9931/24420 [4:07:50<5:59:03,  1.49s/it]
 41%|████      | 9932/24420 [4:07:52<5:59:46,  1.49s/it]
 41%|████      | 9933/24420 [4:07:53<5:59:21,  1.49s/it]
 41%|████      | 9934/24420 [4:07:55<5:58:58,  1.49s/it]
 41%|████      | 9935/24420 [4:07:56<5:59:04,  1.49s/it]
                                                        
{'loss': 0.1061, 'learning_rate': 5e-05, 'epoch': 2.03}

 41%|████      | 9935/24420 [4:07:56<5:59:04,  1.49s/it]
 41%|████      | 9936/24420 [4:07:58<5:59:22,  1.49s/it]
 41%|████      | 9937/24420 [4:07:59<5:58:48,  1.49s/it]
 41%|████      | 9938/24420 [4:08:01<5:58:09,  1.48s/it]
 41%|████      | 9939/24420 [4:08:02<5:58:04,  1.48s/it]
 41%|████      | 9940/24420 [4:08:04<5:58:19,  1.48s/it]
                                                        
{'loss': 0.0977, 'learning_rate': 5e-05, 'epoch': 2.04}

 41%|████      | 9940/24420 [4:08:04<5:58:19,  1.48s/it]
 41%|████      | 9941/24420 [4:08:05<5:58:02,  1.48s/it]
 41%|████      | 9942/24420 [4:08:07<5:58:02,  1.48s/it]
 41%|████      | 9943/24420 [4:08:08<5:57:37,  1.48s/it]
 41%|████      | 9944/24420 [4:08:10<5:58:17,  1.49s/it]
 41%|████      | 9945/24420 [4:08:11<5:57:58,  1.48s/it]
                                                        
{'loss': 0.1127, 'learning_rate': 5e-05, 'epoch': 2.04}

 41%|████      | 9945/24420 [4:08:11<5:57:58,  1.48s/it]
 41%|████      | 9946/24420 [4:08:12<5:57:50,  1.48s/it]
 41%|████      | 9947/24420 [4:08:14<5:58:10,  1.48s/it]
 41%|████      | 9948/24420 [4:08:15<5:57:53,  1.48s/it]
 41%|████      | 9949/24420 [4:08:17<5:59:07,  1.49s/it]
 41%|████      | 9950/24420 [4:08:18<6:00:36,  1.50s/it]
                                                        
{'loss': 0.0889, 'learning_rate': 5e-05, 'epoch': 2.04}

 41%|████      | 9950/24420 [4:08:18<6:00:36,  1.50s/it]
 41%|████      | 9951/24420 [4:08:20<6:00:25,  1.49s/it]
 41%|████      | 9952/24420 [4:08:21<5:59:52,  1.49s/it]
 41%|████      | 9953/24420 [4:08:23<5:59:57,  1.49s/it]
 41%|████      | 9954/24420 [4:08:24<5:59:40,  1.49s/it]
 41%|████      | 9955/24420 [4:08:26<5:59:38,  1.49s/it]
                                                        
{'loss': 0.1108, 'learning_rate': 5e-05, 'epoch': 2.04}

 41%|████      | 9955/24420 [4:08:26<5:59:38,  1.49s/it]
 41%|████      | 9956/24420 [4:08:27<5:59:36,  1.49s/it]
 41%|████      | 9957/24420 [4:08:29<5:59:08,  1.49s/it]
 41%|████      | 9958/24420 [4:08:30<5:59:00,  1.49s/it]
 41%|████      | 9959/24420 [4:08:32<5:58:50,  1.49s/it]
 41%|████      | 9960/24420 [4:08:33<5:58:22,  1.49s/it]
                                                        
{'loss': 0.0958, 'learning_rate': 5e-05, 'epoch': 2.04}

 41%|████      | 9960/24420 [4:08:33<5:58:22,  1.49s/it]
 41%|████      | 9961/24420 [4:08:35<5:58:39,  1.49s/it]
 41%|████      | 9962/24420 [4:08:36<5:57:53,  1.49s/it]
 41%|████      | 9963/24420 [4:08:38<5:58:07,  1.49s/it]
 41%|████      | 9964/24420 [4:08:39<5:57:58,  1.49s/it]
 41%|████      | 9965/24420 [4:08:41<5:57:48,  1.49s/it]
                                                        
{'loss': 0.1, 'learning_rate': 5e-05, 'epoch': 2.04}

 41%|████      | 9965/24420 [4:08:41<5:57:48,  1.49s/it]
 41%|████      | 9966/24420 [4:08:42<5:58:57,  1.49s/it]
 41%|████      | 9967/24420 [4:08:44<5:59:40,  1.49s/it]
 41%|████      | 9968/24420 [4:08:45<5:58:44,  1.49s/it]
 41%|████      | 9969/24420 [4:08:47<5:58:20,  1.49s/it]
 41%|████      | 9970/24420 [4:08:48<5:58:32,  1.49s/it]
                                                        
{'loss': 0.1142, 'learning_rate': 5e-05, 'epoch': 2.04}

 41%|████      | 9970/24420 [4:08:48<5:58:32,  1.49s/it]
 41%|████      | 9971/24420 [4:08:50<5:58:30,  1.49s/it]
 41%|████      | 9972/24420 [4:08:51<5:58:28,  1.49s/it]
 41%|████      | 9973/24420 [4:08:53<5:57:49,  1.49s/it]
 41%|████      | 9974/24420 [4:08:54<5:56:54,  1.48s/it]
 41%|████      | 9975/24420 [4:08:56<5:57:36,  1.49s/it]
                                                        
{'loss': 0.0947, 'learning_rate': 5e-05, 'epoch': 2.04}

 41%|████      | 9975/24420 [4:08:56<5:57:36,  1.49s/it]
 41%|████      | 9976/24420 [4:08:57<5:57:54,  1.49s/it]
 41%|████      | 9977/24420 [4:08:59<5:57:59,  1.49s/it]
 41%|████      | 9978/24420 [4:09:00<5:57:53,  1.49s/it]
 41%|████      | 9979/24420 [4:09:02<5:58:19,  1.49s/it]
 41%|████      | 9980/24420 [4:09:03<5:57:35,  1.49s/it]
                                                        
{'loss': 0.1115, 'learning_rate': 5e-05, 'epoch': 2.04}

 41%|████      | 9980/24420 [4:09:03<5:57:35,  1.49s/it]
 41%|████      | 9981/24420 [4:09:05<5:57:13,  1.48s/it]
 41%|████      | 9982/24420 [4:09:06<5:57:02,  1.48s/it]
 41%|████      | 9983/24420 [4:09:08<5:57:07,  1.48s/it]
 41%|████      | 9984/24420 [4:09:09<5:57:03,  1.48s/it]
 41%|████      | 9985/24420 [4:09:10<5:57:13,  1.48s/it]
                                                        
{'loss': 0.1022, 'learning_rate': 5e-05, 'epoch': 2.04}

 41%|████      | 9985/24420 [4:09:11<5:57:13,  1.48s/it]
 41%|████      | 9986/24420 [4:09:12<5:56:54,  1.48s/it]
 41%|████      | 9987/24420 [4:09:13<5:56:29,  1.48s/it]
 41%|████      | 9988/24420 [4:09:15<5:56:18,  1.48s/it]
 41%|████      | 9989/24420 [4:09:16<5:56:23,  1.48s/it]
 41%|████      | 9990/24420 [4:09:18<5:56:41,  1.48s/it]
                                                        
{'loss': 0.0823, 'learning_rate': 5e-05, 'epoch': 2.05}

 41%|████      | 9990/24420 [4:09:18<5:56:41,  1.48s/it]
 41%|████      | 9991/24420 [4:09:19<5:57:12,  1.49s/it]
 41%|████      | 9992/24420 [4:09:21<5:57:28,  1.49s/it]
 41%|████      | 9993/24420 [4:09:22<5:57:03,  1.48s/it]
 41%|████      | 9994/24420 [4:09:24<5:56:53,  1.48s/it]
 41%|████      | 9995/24420 [4:09:25<5:56:50,  1.48s/it]
                                                        
{'loss': 0.1026, 'learning_rate': 5e-05, 'epoch': 2.05}

 41%|████      | 9995/24420 [4:09:25<5:56:50,  1.48s/it]
 41%|████      | 9996/24420 [4:09:27<5:56:42,  1.48s/it]
 41%|████      | 9997/24420 [4:09:28<5:57:48,  1.49s/it]
 41%|████      | 9998/24420 [4:09:30<5:57:56,  1.49s/it]
 41%|████      | 9999/24420 [4:09:31<5:57:20,  1.49s/it]
 41%|████      | 10000/24420 [4:09:33<5:56:41,  1.48s/it]
                                                         
{'loss': 0.0859, 'learning_rate': 5e-05, 'epoch': 2.05}

 41%|████      | 10000/24420 [4:09:33<5:56:41,  1.48s/it]
 41%|████      | 10001/24420 [4:09:37<9:36:15,  2.40s/it]
 41%|████      | 10002/24420 [4:09:39<8:30:11,  2.12s/it]
 41%|████      | 10003/24420 [4:09:40<7:43:18,  1.93s/it]
 41%|████      | 10004/24420 [4:09:42<7:11:27,  1.80s/it]
 41%|████      | 10005/24420 [4:09:43<6:48:37,  1.70s/it]
                                                         
{'loss': 0.0795, 'learning_rate': 5e-05, 'epoch': 2.05}

 41%|████      | 10005/24420 [4:09:43<6:48:37,  1.70s/it]
 41%|████      | 10006/24420 [4:09:45<6:33:33,  1.64s/it]
 41%|████      | 10007/24420 [4:09:46<6:21:58,  1.59s/it]
 41%|████      | 10008/24420 [4:09:48<6:14:36,  1.56s/it]
 41%|████      | 10009/24420 [4:09:49<6:08:34,  1.53s/it]
 41%|████      | 10010/24420 [4:09:51<6:04:46,  1.52s/it]
                                                         
{'loss': 0.0917, 'learning_rate': 5e-05, 'epoch': 2.05}

 41%|████      | 10010/24420 [4:09:51<6:04:46,  1.52s/it]
 41%|████      | 10011/24420 [4:09:52<6:02:33,  1.51s/it]
 41%|████      | 10012/24420 [4:09:54<6:00:24,  1.50s/it]
 41%|████      | 10013/24420 [4:09:55<5:59:32,  1.50s/it]
 41%|████      | 10014/24420 [4:09:57<5:59:29,  1.50s/it]
 41%|████      | 10015/24420 [4:09:58<5:58:45,  1.49s/it]
                                                         
{'loss': 0.0893, 'learning_rate': 5e-05, 'epoch': 2.05}

 41%|████      | 10015/24420 [4:09:58<5:58:45,  1.49s/it]
 41%|████      | 10016/24420 [4:10:00<5:58:33,  1.49s/it]
 41%|████      | 10017/24420 [4:10:01<5:57:50,  1.49s/it]
 41%|████      | 10018/24420 [4:10:03<5:57:09,  1.49s/it]
 41%|████      | 10019/24420 [4:10:04<5:57:56,  1.49s/it]
 41%|████      | 10020/24420 [4:10:06<5:57:17,  1.49s/it]
                                                         
{'loss': 0.1122, 'learning_rate': 5e-05, 'epoch': 2.05}

 41%|████      | 10020/24420 [4:10:06<5:57:17,  1.49s/it]
 41%|████      | 10021/24420 [4:10:07<5:57:01,  1.49s/it]
 41%|████      | 10022/24420 [4:10:08<5:56:55,  1.49s/it]
 41%|████      | 10023/24420 [4:10:10<5:56:29,  1.49s/it]
 41%|████      | 10024/24420 [4:10:11<5:56:18,  1.49s/it]
 41%|████      | 10025/24420 [4:10:13<5:56:01,  1.48s/it]
                                                         
{'loss': 0.0837, 'learning_rate': 5e-05, 'epoch': 2.05}

 41%|████      | 10025/24420 [4:10:13<5:56:01,  1.48s/it]
 41%|████      | 10026/24420 [4:10:14<5:56:21,  1.49s/it]
 41%|████      | 10027/24420 [4:10:16<5:56:20,  1.49s/it]
 41%|████      | 10028/24420 [4:10:17<5:56:31,  1.49s/it]
 41%|████      | 10029/24420 [4:10:19<5:56:59,  1.49s/it]
 41%|████      | 10030/24420 [4:10:20<5:56:09,  1.49s/it]
                                                         
{'loss': 0.0834, 'learning_rate': 5e-05, 'epoch': 2.05}

 41%|████      | 10030/24420 [4:10:20<5:56:09,  1.49s/it]
 41%|████      | 10031/24420 [4:10:22<5:56:31,  1.49s/it]
 41%|████      | 10032/24420 [4:10:23<5:56:39,  1.49s/it]
 41%|████      | 10033/24420 [4:10:25<5:56:23,  1.49s/it]
 41%|████      | 10034/24420 [4:10:26<5:56:38,  1.49s/it]
 41%|████      | 10035/24420 [4:10:28<5:56:21,  1.49s/it]
                                                         
{'loss': 0.0865, 'learning_rate': 5e-05, 'epoch': 2.05}

 41%|████      | 10035/24420 [4:10:28<5:56:21,  1.49s/it]
 41%|████      | 10036/24420 [4:10:29<5:56:43,  1.49s/it]
 41%|████      | 10037/24420 [4:10:31<5:58:00,  1.49s/it]
 41%|████      | 10038/24420 [4:10:32<5:57:47,  1.49s/it]
 41%|████      | 10039/24420 [4:10:34<5:57:15,  1.49s/it]
 41%|████      | 10040/24420 [4:10:35<5:56:47,  1.49s/it]
                                                         
{'loss': 0.0995, 'learning_rate': 5e-05, 'epoch': 2.06}

 41%|████      | 10040/24420 [4:10:35<5:56:47,  1.49s/it]
 41%|████      | 10041/24420 [4:10:37<5:56:20,  1.49s/it]
 41%|████      | 10042/24420 [4:10:38<5:55:59,  1.49s/it]
 41%|████      | 10043/24420 [4:10:40<5:56:06,  1.49s/it]
 41%|████      | 10044/24420 [4:10:41<5:55:49,  1.49s/it]
 41%|████      | 10045/24420 [4:10:43<5:56:26,  1.49s/it]
                                                         
{'loss': 0.0959, 'learning_rate': 5e-05, 'epoch': 2.06}

 41%|████      | 10045/24420 [4:10:43<5:56:26,  1.49s/it]
 41%|████      | 10046/24420 [4:10:44<5:56:32,  1.49s/it]
 41%|████      | 10047/24420 [4:10:46<5:56:50,  1.49s/it]
 41%|████      | 10048/24420 [4:10:47<5:55:55,  1.49s/it]
 41%|████      | 10049/24420 [4:10:49<5:55:53,  1.49s/it]
 41%|████      | 10050/24420 [4:10:50<5:55:54,  1.49s/it]
                                                         
{'loss': 0.0898, 'learning_rate': 5e-05, 'epoch': 2.06}

 41%|████      | 10050/24420 [4:10:50<5:55:54,  1.49s/it]
 41%|████      | 10051/24420 [4:10:52<5:55:47,  1.49s/it]
 41%|████      | 10052/24420 [4:10:53<5:55:02,  1.48s/it]
 41%|████      | 10053/24420 [4:10:55<5:56:00,  1.49s/it]
 41%|████      | 10054/24420 [4:10:56<5:55:16,  1.48s/it]
 41%|████      | 10055/24420 [4:10:58<5:55:56,  1.49s/it]
                                                         
{'loss': 0.0931, 'learning_rate': 5e-05, 'epoch': 2.06}

 41%|████      | 10055/24420 [4:10:58<5:55:56,  1.49s/it]
 41%|████      | 10056/24420 [4:10:59<5:55:40,  1.49s/it]
 41%|████      | 10057/24420 [4:11:01<5:55:24,  1.48s/it]
 41%|████      | 10058/24420 [4:11:02<5:55:41,  1.49s/it]
 41%|████      | 10059/24420 [4:11:04<5:56:21,  1.49s/it]
 41%|████      | 10060/24420 [4:11:05<5:55:46,  1.49s/it]
                                                         
{'loss': 0.0898, 'learning_rate': 5e-05, 'epoch': 2.06}

 41%|████      | 10060/24420 [4:11:05<5:55:46,  1.49s/it]
 41%|████      | 10061/24420 [4:11:06<5:56:03,  1.49s/it]
 41%|████      | 10062/24420 [4:11:08<5:55:46,  1.49s/it]
 41%|████      | 10063/24420 [4:11:09<5:55:42,  1.49s/it]
 41%|████      | 10064/24420 [4:11:11<5:56:14,  1.49s/it]
 41%|████      | 10065/24420 [4:11:12<5:56:25,  1.49s/it]
                                                         
{'loss': 0.0945, 'learning_rate': 5e-05, 'epoch': 2.06}

 41%|████      | 10065/24420 [4:11:12<5:56:25,  1.49s/it]
 41%|████      | 10066/24420 [4:11:14<5:56:15,  1.49s/it]
 41%|████      | 10067/24420 [4:11:15<5:55:53,  1.49s/it]
 41%|████      | 10068/24420 [4:11:17<5:56:08,  1.49s/it]
 41%|████      | 10069/24420 [4:11:18<5:56:54,  1.49s/it]
 41%|████      | 10070/24420 [4:11:20<5:56:22,  1.49s/it]
                                                         
{'loss': 0.0892, 'learning_rate': 5e-05, 'epoch': 2.06}

 41%|████      | 10070/24420 [4:11:20<5:56:22,  1.49s/it]
 41%|████      | 10071/24420 [4:11:21<5:56:29,  1.49s/it]
 41%|████      | 10072/24420 [4:11:23<5:55:38,  1.49s/it]
 41%|████      | 10073/24420 [4:11:24<5:55:27,  1.49s/it]
 41%|████▏     | 10074/24420 [4:11:26<5:55:00,  1.48s/it]
 41%|████▏     | 10075/24420 [4:11:27<5:55:10,  1.49s/it]
                                                         
{'loss': 0.106, 'learning_rate': 5e-05, 'epoch': 2.06}

 41%|████▏     | 10075/24420 [4:11:27<5:55:10,  1.49s/it]
 41%|████▏     | 10076/24420 [4:11:29<5:55:10,  1.49s/it]
 41%|████▏     | 10077/24420 [4:11:30<5:56:06,  1.49s/it]
 41%|████▏     | 10078/24420 [4:11:32<5:55:51,  1.49s/it]
 41%|████▏     | 10079/24420 [4:11:33<5:55:21,  1.49s/it]
 41%|████▏     | 10080/24420 [4:11:35<5:54:59,  1.49s/it]
                                                         
{'loss': 0.0874, 'learning_rate': 5e-05, 'epoch': 2.06}

 41%|████▏     | 10080/24420 [4:11:35<5:54:59,  1.49s/it]
 41%|████▏     | 10081/24420 [4:11:36<5:54:54,  1.49s/it]
 41%|████▏     | 10082/24420 [4:11:38<5:56:20,  1.49s/it]
 41%|████▏     | 10083/24420 [4:11:39<5:55:38,  1.49s/it]
 41%|████▏     | 10084/24420 [4:11:41<5:55:30,  1.49s/it]
 41%|████▏     | 10085/24420 [4:11:42<5:55:28,  1.49s/it]
                                                         
{'loss': 0.0879, 'learning_rate': 5e-05, 'epoch': 2.06}

 41%|████▏     | 10085/24420 [4:11:42<5:55:28,  1.49s/it]
 41%|████▏     | 10086/24420 [4:11:44<5:55:39,  1.49s/it]
 41%|████▏     | 10087/24420 [4:11:45<5:55:46,  1.49s/it]
 41%|████▏     | 10088/24420 [4:11:47<5:55:18,  1.49s/it]
 41%|████▏     | 10089/24420 [4:11:48<5:55:17,  1.49s/it]
 41%|████▏     | 10090/24420 [4:11:50<5:55:08,  1.49s/it]
                                                         
{'loss': 0.0987, 'learning_rate': 5e-05, 'epoch': 2.07}

 41%|████▏     | 10090/24420 [4:11:50<5:55:08,  1.49s/it]
 41%|████▏     | 10091/24420 [4:11:51<5:55:44,  1.49s/it]
 41%|████▏     | 10092/24420 [4:11:53<5:55:41,  1.49s/it]
 41%|████▏     | 10093/24420 [4:11:54<5:55:04,  1.49s/it]
 41%|████▏     | 10094/24420 [4:11:56<5:54:36,  1.49s/it]
 41%|████▏     | 10095/24420 [4:11:57<5:55:14,  1.49s/it]
                                                         
{'loss': 0.1017, 'learning_rate': 5e-05, 'epoch': 2.07}

 41%|████▏     | 10095/24420 [4:11:57<5:55:14,  1.49s/it]
 41%|████▏     | 10096/24420 [4:11:59<5:55:08,  1.49s/it]
 41%|████▏     | 10097/24420 [4:12:00<5:54:33,  1.49s/it]
 41%|████▏     | 10098/24420 [4:12:02<5:54:41,  1.49s/it]
 41%|████▏     | 10099/24420 [4:12:03<5:54:39,  1.49s/it]
 41%|████▏     | 10100/24420 [4:12:04<5:54:21,  1.48s/it]
                                                         
{'loss': 0.0951, 'learning_rate': 5e-05, 'epoch': 2.07}

 41%|████▏     | 10100/24420 [4:12:04<5:54:21,  1.48s/it]
 41%|████▏     | 10101/24420 [4:12:06<5:54:29,  1.49s/it]
 41%|████▏     | 10102/24420 [4:12:07<5:55:25,  1.49s/it]
 41%|████▏     | 10103/24420 [4:12:09<5:54:37,  1.49s/it]
 41%|████▏     | 10104/24420 [4:12:10<5:54:52,  1.49s/it]
 41%|████▏     | 10105/24420 [4:12:12<5:54:53,  1.49s/it]
                                                         
{'loss': 0.0829, 'learning_rate': 5e-05, 'epoch': 2.07}

 41%|████▏     | 10105/24420 [4:12:12<5:54:53,  1.49s/it]
 41%|████▏     | 10106/24420 [4:12:13<5:55:14,  1.49s/it]
 41%|████▏     | 10107/24420 [4:12:15<5:55:57,  1.49s/it]
 41%|████▏     | 10108/24420 [4:12:16<5:55:11,  1.49s/it]
 41%|████▏     | 10109/24420 [4:12:18<5:54:58,  1.49s/it]
 41%|████▏     | 10110/24420 [4:12:19<5:56:47,  1.50s/it]
                                                         
{'loss': 0.1006, 'learning_rate': 5e-05, 'epoch': 2.07}

 41%|████▏     | 10110/24420 [4:12:19<5:56:47,  1.50s/it]
 41%|████▏     | 10111/24420 [4:12:21<5:56:13,  1.49s/it]
 41%|████▏     | 10112/24420 [4:12:22<5:55:57,  1.49s/it]
 41%|████▏     | 10113/24420 [4:12:24<5:55:04,  1.49s/it]
 41%|████▏     | 10114/24420 [4:12:25<5:54:43,  1.49s/it]
 41%|████▏     | 10115/24420 [4:12:27<5:54:16,  1.49s/it]
                                                         
{'loss': 0.1299, 'learning_rate': 5e-05, 'epoch': 2.07}

 41%|████▏     | 10115/24420 [4:12:27<5:54:16,  1.49s/it]
 41%|████▏     | 10116/24420 [4:12:28<5:54:00,  1.48s/it]
 41%|████▏     | 10117/24420 [4:12:30<5:53:59,  1.48s/it]
 41%|████▏     | 10118/24420 [4:12:31<5:54:12,  1.49s/it]
 41%|████▏     | 10119/24420 [4:12:33<5:53:44,  1.48s/it]
 41%|████▏     | 10120/24420 [4:12:34<5:54:08,  1.49s/it]
                                                         
{'loss': 0.0897, 'learning_rate': 5e-05, 'epoch': 2.07}

 41%|████▏     | 10120/24420 [4:12:34<5:54:08,  1.49s/it]
 41%|████▏     | 10121/24420 [4:12:36<5:53:55,  1.49s/it]
 41%|████▏     | 10122/24420 [4:12:37<5:54:06,  1.49s/it]
 41%|████▏     | 10123/24420 [4:12:39<5:53:45,  1.48s/it]
 41%|████▏     | 10124/24420 [4:12:40<5:54:19,  1.49s/it]
 41%|████▏     | 10125/24420 [4:12:42<5:54:11,  1.49s/it]
                                                         
{'loss': 0.0857, 'learning_rate': 5e-05, 'epoch': 2.07}

 41%|████▏     | 10125/24420 [4:12:42<5:54:11,  1.49s/it]
 41%|████▏     | 10126/24420 [4:12:43<5:54:08,  1.49s/it]
 41%|████▏     | 10127/24420 [4:12:45<5:54:13,  1.49s/it]
 41%|████▏     | 10128/24420 [4:12:46<5:54:14,  1.49s/it]
 41%|████▏     | 10129/24420 [4:12:48<5:54:52,  1.49s/it]
 41%|████▏     | 10130/24420 [4:12:49<5:55:49,  1.49s/it]
                                                         
{'loss': 0.1026, 'learning_rate': 5e-05, 'epoch': 2.07}

 41%|████▏     | 10130/24420 [4:12:49<5:55:49,  1.49s/it]
 41%|████▏     | 10131/24420 [4:12:51<5:55:40,  1.49s/it]
 41%|████▏     | 10132/24420 [4:12:52<5:54:46,  1.49s/it]
 41%|████▏     | 10133/24420 [4:12:54<5:54:54,  1.49s/it]
 41%|████▏     | 10134/24420 [4:12:55<5:56:02,  1.50s/it]
 42%|████▏     | 10135/24420 [4:12:57<5:56:55,  1.50s/it]
                                                         
{'loss': 0.1125, 'learning_rate': 5e-05, 'epoch': 2.08}

 42%|████▏     | 10135/24420 [4:12:57<5:56:55,  1.50s/it]
 42%|████▏     | 10136/24420 [4:12:58<5:55:50,  1.49s/it]
 42%|████▏     | 10137/24420 [4:13:00<5:55:03,  1.49s/it]
 42%|████▏     | 10138/24420 [4:13:01<5:54:32,  1.49s/it]
 42%|████▏     | 10139/24420 [4:13:03<5:54:55,  1.49s/it]
 42%|████▏     | 10140/24420 [4:13:04<5:54:58,  1.49s/it]
                                                         
{'loss': 0.0934, 'learning_rate': 5e-05, 'epoch': 2.08}

 42%|████▏     | 10140/24420 [4:13:04<5:54:58,  1.49s/it]
 42%|████▏     | 10141/24420 [4:13:06<5:54:48,  1.49s/it]
 42%|████▏     | 10142/24420 [4:13:07<5:55:06,  1.49s/it]
 42%|████▏     | 10143/24420 [4:13:09<5:54:51,  1.49s/it]
 42%|████▏     | 10144/24420 [4:13:10<5:54:44,  1.49s/it]
 42%|████▏     | 10145/24420 [4:13:12<5:54:37,  1.49s/it]
                                                         
{'loss': 0.104, 'learning_rate': 5e-05, 'epoch': 2.08}

 42%|████▏     | 10145/24420 [4:13:12<5:54:37,  1.49s/it]
 42%|████▏     | 10146/24420 [4:13:13<5:54:23,  1.49s/it]
 42%|████▏     | 10147/24420 [4:13:14<5:53:54,  1.49s/it]
 42%|████▏     | 10148/24420 [4:13:16<5:53:45,  1.49s/it]
 42%|████▏     | 10149/24420 [4:13:17<5:53:57,  1.49s/it]
 42%|████▏     | 10150/24420 [4:13:19<5:54:09,  1.49s/it]
                                                         
{'loss': 0.0977, 'learning_rate': 5e-05, 'epoch': 2.08}

 42%|████▏     | 10150/24420 [4:13:19<5:54:09,  1.49s/it]
 42%|████▏     | 10151/24420 [4:13:20<5:54:28,  1.49s/it]
 42%|████▏     | 10152/24420 [4:13:22<5:54:51,  1.49s/it]
 42%|████▏     | 10153/24420 [4:13:23<5:54:30,  1.49s/it]
 42%|████▏     | 10154/24420 [4:13:25<5:54:55,  1.49s/it]
 42%|████▏     | 10155/24420 [4:13:26<5:54:48,  1.49s/it]
                                                         
{'loss': 0.107, 'learning_rate': 5e-05, 'epoch': 2.08}

 42%|████▏     | 10155/24420 [4:13:26<5:54:48,  1.49s/it]
 42%|████▏     | 10156/24420 [4:13:28<5:55:00,  1.49s/it]
 42%|████▏     | 10157/24420 [4:13:29<5:54:00,  1.49s/it]
 42%|████▏     | 10158/24420 [4:13:31<5:53:53,  1.49s/it]
 42%|████▏     | 10159/24420 [4:13:32<5:54:14,  1.49s/it]
 42%|████▏     | 10160/24420 [4:13:34<5:54:12,  1.49s/it]
                                                         
{'loss': 0.0821, 'learning_rate': 5e-05, 'epoch': 2.08}

 42%|████▏     | 10160/24420 [4:13:34<5:54:12,  1.49s/it]
 42%|████▏     | 10161/24420 [4:13:35<5:54:23,  1.49s/it]
 42%|████▏     | 10162/24420 [4:13:37<5:53:57,  1.49s/it]
 42%|████▏     | 10163/24420 [4:13:38<5:53:45,  1.49s/it]
 42%|████▏     | 10164/24420 [4:13:40<5:54:13,  1.49s/it]
 42%|████▏     | 10165/24420 [4:13:41<5:53:40,  1.49s/it]
                                                         
{'loss': 0.1042, 'learning_rate': 5e-05, 'epoch': 2.08}

 42%|████▏     | 10165/24420 [4:13:41<5:53:40,  1.49s/it]
 42%|████▏     | 10166/24420 [4:13:43<5:53:52,  1.49s/it]
 42%|████▏     | 10167/24420 [4:13:44<5:53:22,  1.49s/it]
 42%|████▏     | 10168/24420 [4:13:46<5:52:58,  1.49s/it]
 42%|████▏     | 10169/24420 [4:13:47<5:53:09,  1.49s/it]
 42%|████▏     | 10170/24420 [4:13:49<5:54:15,  1.49s/it]
                                                         
{'loss': 0.0872, 'learning_rate': 5e-05, 'epoch': 2.08}

 42%|████▏     | 10170/24420 [4:13:49<5:54:15,  1.49s/it]
 42%|████▏     | 10171/24420 [4:13:50<5:53:44,  1.49s/it]
 42%|████▏     | 10172/24420 [4:13:52<5:53:13,  1.49s/it]
 42%|████▏     | 10173/24420 [4:13:53<5:53:38,  1.49s/it]
 42%|████▏     | 10174/24420 [4:13:55<5:53:25,  1.49s/it]
 42%|████▏     | 10175/24420 [4:13:56<5:52:43,  1.49s/it]
                                                         
{'loss': 0.0963, 'learning_rate': 5e-05, 'epoch': 2.08}

 42%|████▏     | 10175/24420 [4:13:56<5:52:43,  1.49s/it]
 42%|████▏     | 10176/24420 [4:13:58<5:53:21,  1.49s/it]
 42%|████▏     | 10177/24420 [4:13:59<5:53:06,  1.49s/it]
 42%|████▏     | 10178/24420 [4:14:01<5:52:46,  1.49s/it]
 42%|████▏     | 10179/24420 [4:14:02<5:52:55,  1.49s/it]
 42%|████▏     | 10180/24420 [4:14:04<5:52:46,  1.49s/it]
                                                         
{'loss': 0.1003, 'learning_rate': 5e-05, 'epoch': 2.08}

 42%|████▏     | 10180/24420 [4:14:04<5:52:46,  1.49s/it]
 42%|████▏     | 10181/24420 [4:14:05<5:53:14,  1.49s/it]
 42%|████▏     | 10182/24420 [4:14:07<5:53:27,  1.49s/it]
 42%|████▏     | 10183/24420 [4:14:08<5:53:19,  1.49s/it]
 42%|████▏     | 10184/24420 [4:14:10<5:53:12,  1.49s/it]
 42%|████▏     | 10185/24420 [4:14:11<5:53:39,  1.49s/it]
                                                         
{'loss': 0.1091, 'learning_rate': 5e-05, 'epoch': 2.09}

 42%|████▏     | 10185/24420 [4:14:11<5:53:39,  1.49s/it]
 42%|████▏     | 10186/24420 [4:14:13<5:52:47,  1.49s/it]
 42%|████▏     | 10187/24420 [4:14:14<5:53:11,  1.49s/it]
 42%|████▏     | 10188/24420 [4:14:16<5:52:15,  1.49s/it]
 42%|████▏     | 10189/24420 [4:14:17<5:53:17,  1.49s/it]
 42%|████▏     | 10190/24420 [4:14:19<5:53:56,  1.49s/it]
                                                         
{'loss': 0.0916, 'learning_rate': 5e-05, 'epoch': 2.09}

 42%|████▏     | 10190/24420 [4:14:19<5:53:56,  1.49s/it]
 42%|████▏     | 10191/24420 [4:14:20<5:53:18,  1.49s/it]
 42%|████▏     | 10192/24420 [4:14:22<5:56:41,  1.50s/it]
 42%|████▏     | 10193/24420 [4:14:23<5:56:21,  1.50s/it]
 42%|████▏     | 10194/24420 [4:14:25<5:55:50,  1.50s/it]
 42%|████▏     | 10195/24420 [4:14:26<5:54:53,  1.50s/it]
                                                         
{'loss': 0.1123, 'learning_rate': 5e-05, 'epoch': 2.09}

 42%|████▏     | 10195/24420 [4:14:26<5:54:53,  1.50s/it]
 42%|████▏     | 10196/24420 [4:14:28<5:54:12,  1.49s/it]
 42%|████▏     | 10197/24420 [4:14:29<5:53:44,  1.49s/it]
 42%|████▏     | 10198/24420 [4:14:31<5:53:15,  1.49s/it]
 42%|████▏     | 10199/24420 [4:14:32<5:52:43,  1.49s/it]
 42%|████▏     | 10200/24420 [4:14:33<5:52:17,  1.49s/it]
                                                         
{'loss': 0.1124, 'learning_rate': 5e-05, 'epoch': 2.09}

 42%|████▏     | 10200/24420 [4:14:33<5:52:17,  1.49s/it]
 42%|████▏     | 10201/24420 [4:14:38<9:32:03,  2.41s/it]
 42%|████▏     | 10202/24420 [4:14:40<8:26:04,  2.14s/it]
 42%|████▏     | 10203/24420 [4:14:41<7:40:05,  1.94s/it]
 42%|████▏     | 10204/24420 [4:14:43<7:08:49,  1.81s/it]
 42%|████▏     | 10205/24420 [4:14:44<6:45:31,  1.71s/it]
                                                         
{'loss': 0.0907, 'learning_rate': 5e-05, 'epoch': 2.09}

 42%|████▏     | 10205/24420 [4:14:44<6:45:31,  1.71s/it]
 42%|████▏     | 10206/24420 [4:14:45<6:29:40,  1.64s/it]
 42%|████▏     | 10207/24420 [4:14:47<6:17:58,  1.60s/it]
 42%|████▏     | 10208/24420 [4:14:48<6:10:10,  1.56s/it]
 42%|████▏     | 10209/24420 [4:14:50<6:04:20,  1.54s/it]
 42%|████▏     | 10210/24420 [4:14:51<6:00:51,  1.52s/it]
                                                         
{'loss': 0.0972, 'learning_rate': 5e-05, 'epoch': 2.09}

 42%|████▏     | 10210/24420 [4:14:51<6:00:51,  1.52s/it]
 42%|████▏     | 10211/24420 [4:14:53<5:58:20,  1.51s/it]
 42%|████▏     | 10212/24420 [4:14:54<5:56:28,  1.51s/it]
 42%|████▏     | 10213/24420 [4:14:56<5:54:50,  1.50s/it]
 42%|████▏     | 10214/24420 [4:14:57<5:54:21,  1.50s/it]
 42%|████▏     | 10215/24420 [4:14:59<5:53:19,  1.49s/it]
                                                         
{'loss': 0.1066, 'learning_rate': 5e-05, 'epoch': 2.09}

 42%|████▏     | 10215/24420 [4:14:59<5:53:19,  1.49s/it]
 42%|████▏     | 10216/24420 [4:15:00<5:52:30,  1.49s/it]
 42%|████▏     | 10217/24420 [4:15:02<5:52:03,  1.49s/it]
 42%|████▏     | 10218/24420 [4:15:03<5:51:47,  1.49s/it]
 42%|████▏     | 10219/24420 [4:15:05<5:51:55,  1.49s/it]
 42%|████▏     | 10220/24420 [4:15:06<5:52:28,  1.49s/it]
                                                         
{'loss': 0.0847, 'learning_rate': 5e-05, 'epoch': 2.09}

 42%|████▏     | 10220/24420 [4:15:06<5:52:28,  1.49s/it]
 42%|████▏     | 10221/24420 [4:15:08<5:52:31,  1.49s/it]
 42%|████▏     | 10222/24420 [4:15:09<5:52:04,  1.49s/it]
 42%|████▏     | 10223/24420 [4:15:11<5:51:28,  1.49s/it]
 42%|████▏     | 10224/24420 [4:15:12<5:52:42,  1.49s/it]
 42%|████▏     | 10225/24420 [4:15:14<5:52:12,  1.49s/it]
                                                         
{'loss': 0.0968, 'learning_rate': 5e-05, 'epoch': 2.09}

 42%|████▏     | 10225/24420 [4:15:14<5:52:12,  1.49s/it]
 42%|████▏     | 10226/24420 [4:15:15<5:52:29,  1.49s/it]
 42%|████▏     | 10227/24420 [4:15:17<5:52:15,  1.49s/it]
 42%|████▏     | 10228/24420 [4:15:18<5:51:37,  1.49s/it]
 42%|████▏     | 10229/24420 [4:15:20<5:51:21,  1.49s/it]
 42%|████▏     | 10230/24420 [4:15:21<5:51:25,  1.49s/it]
                                                         
{'loss': 0.092, 'learning_rate': 5e-05, 'epoch': 2.09}

 42%|████▏     | 10230/24420 [4:15:21<5:51:25,  1.49s/it]
 42%|████▏     | 10231/24420 [4:15:23<5:51:03,  1.48s/it]
 42%|████▏     | 10232/24420 [4:15:24<5:51:28,  1.49s/it]
 42%|████▏     | 10233/24420 [4:15:26<5:51:43,  1.49s/it]
 42%|████▏     | 10234/24420 [4:15:27<5:52:04,  1.49s/it]
 42%|████▏     | 10235/24420 [4:15:29<5:51:06,  1.49s/it]
                                                         
{'loss': 0.0931, 'learning_rate': 5e-05, 'epoch': 2.1}

 42%|████▏     | 10235/24420 [4:15:29<5:51:06,  1.49s/it]
 42%|████▏     | 10236/24420 [4:15:30<5:51:08,  1.49s/it]
 42%|████▏     | 10237/24420 [4:15:32<5:51:01,  1.49s/it]
 42%|████▏     | 10238/24420 [4:15:33<5:51:42,  1.49s/it]
 42%|████▏     | 10239/24420 [4:15:35<5:51:49,  1.49s/it]
 42%|████▏     | 10240/24420 [4:15:36<5:51:28,  1.49s/it]
                                                         
{'loss': 0.1061, 'learning_rate': 5e-05, 'epoch': 2.1}

 42%|████▏     | 10240/24420 [4:15:36<5:51:28,  1.49s/it]
 42%|████▏     | 10241/24420 [4:15:38<5:52:46,  1.49s/it]
 42%|████▏     | 10242/24420 [4:15:39<5:51:58,  1.49s/it]
 42%|████▏     | 10243/24420 [4:15:41<5:51:27,  1.49s/it]
 42%|████▏     | 10244/24420 [4:15:42<5:51:38,  1.49s/it]
 42%|████▏     | 10245/24420 [4:15:43<5:51:40,  1.49s/it]
                                                         
{'loss': 0.1003, 'learning_rate': 5e-05, 'epoch': 2.1}

 42%|████▏     | 10245/24420 [4:15:43<5:51:40,  1.49s/it]
 42%|████▏     | 10246/24420 [4:15:45<5:51:31,  1.49s/it]
 42%|████▏     | 10247/24420 [4:15:46<5:51:38,  1.49s/it]
 42%|████▏     | 10248/24420 [4:15:48<5:50:56,  1.49s/it]
 42%|████▏     | 10249/24420 [4:15:49<5:50:46,  1.49s/it]
 42%|████▏     | 10250/24420 [4:15:51<5:50:50,  1.49s/it]
                                                         
{'loss': 0.1042, 'learning_rate': 5e-05, 'epoch': 2.1}

 42%|████▏     | 10250/24420 [4:15:51<5:50:50,  1.49s/it]
 42%|████▏     | 10251/24420 [4:15:52<5:50:52,  1.49s/it]
 42%|████▏     | 10252/24420 [4:15:54<5:50:55,  1.49s/it]
 42%|████▏     | 10253/24420 [4:15:55<5:51:15,  1.49s/it]
 42%|████▏     | 10254/24420 [4:15:57<5:51:03,  1.49s/it]
 42%|████▏     | 10255/24420 [4:15:58<5:50:40,  1.49s/it]
                                                         
{'loss': 0.0999, 'learning_rate': 5e-05, 'epoch': 2.1}

 42%|████▏     | 10255/24420 [4:15:58<5:50:40,  1.49s/it]
 42%|████▏     | 10256/24420 [4:16:00<5:51:00,  1.49s/it]
 42%|████▏     | 10257/24420 [4:16:01<5:50:58,  1.49s/it]
 42%|████▏     | 10258/24420 [4:16:03<5:50:38,  1.49s/it]
 42%|████▏     | 10259/24420 [4:16:04<5:50:32,  1.49s/it]
 42%|████▏     | 10260/24420 [4:16:06<5:50:27,  1.48s/it]
                                                         
{'loss': 0.1023, 'learning_rate': 5e-05, 'epoch': 2.1}

 42%|████▏     | 10260/24420 [4:16:06<5:50:27,  1.48s/it]
 42%|████▏     | 10261/24420 [4:16:07<5:51:11,  1.49s/it]
 42%|████▏     | 10262/24420 [4:16:09<5:50:39,  1.49s/it]
 42%|████▏     | 10263/24420 [4:16:10<5:50:25,  1.49s/it]
 42%|████▏     | 10264/24420 [4:16:12<5:50:48,  1.49s/it]
 42%|████▏     | 10265/24420 [4:16:13<5:50:45,  1.49s/it]
                                                         
{'loss': 0.0957, 'learning_rate': 5e-05, 'epoch': 2.1}

 42%|████▏     | 10265/24420 [4:16:13<5:50:45,  1.49s/it]
 42%|████▏     | 10266/24420 [4:16:15<5:50:59,  1.49s/it]
 42%|████▏     | 10267/24420 [4:16:16<5:51:19,  1.49s/it]
 42%|████▏     | 10268/24420 [4:16:18<5:51:45,  1.49s/it]
 42%|████▏     | 10269/24420 [4:16:19<5:50:55,  1.49s/it]
 42%|████▏     | 10270/24420 [4:16:21<5:50:34,  1.49s/it]
                                                         
{'loss': 0.0917, 'learning_rate': 5e-05, 'epoch': 2.1}

 42%|████▏     | 10270/24420 [4:16:21<5:50:34,  1.49s/it]
 42%|████▏     | 10271/24420 [4:16:22<5:50:41,  1.49s/it]
 42%|████▏     | 10272/24420 [4:16:24<5:50:29,  1.49s/it]
 42%|████▏     | 10273/24420 [4:16:25<5:50:13,  1.49s/it]
 42%|████▏     | 10274/24420 [4:16:27<5:49:50,  1.48s/it]
 42%|████▏     | 10275/24420 [4:16:28<5:50:18,  1.49s/it]
                                                         
{'loss': 0.0913, 'learning_rate': 5e-05, 'epoch': 2.1}

 42%|████▏     | 10275/24420 [4:16:28<5:50:18,  1.49s/it]
 42%|████▏     | 10276/24420 [4:16:30<5:50:10,  1.49s/it]
 42%|████▏     | 10277/24420 [4:16:31<5:49:46,  1.48s/it]
 42%|████▏     | 10278/24420 [4:16:33<5:50:18,  1.49s/it]
 42%|████▏     | 10279/24420 [4:16:34<5:50:00,  1.49s/it]
 42%|████▏     | 10280/24420 [4:16:36<5:49:58,  1.49s/it]
                                                         
{'loss': 0.0929, 'learning_rate': 5e-05, 'epoch': 2.1}

 42%|████▏     | 10280/24420 [4:16:36<5:49:58,  1.49s/it]
 42%|████▏     | 10281/24420 [4:16:37<5:51:09,  1.49s/it]
 42%|████▏     | 10282/24420 [4:16:38<5:50:36,  1.49s/it]
 42%|████▏     | 10283/24420 [4:16:40<5:51:07,  1.49s/it]
 42%|████▏     | 10284/24420 [4:16:41<5:50:21,  1.49s/it]
 42%|████▏     | 10285/24420 [4:16:43<5:50:46,  1.49s/it]
                                                         
{'loss': 0.0976, 'learning_rate': 5e-05, 'epoch': 2.11}

 42%|████▏     | 10285/24420 [4:16:43<5:50:46,  1.49s/it]
 42%|████▏     | 10286/24420 [4:16:44<5:50:42,  1.49s/it]
 42%|████▏     | 10287/24420 [4:16:46<5:50:55,  1.49s/it]
 42%|████▏     | 10288/24420 [4:16:47<5:50:24,  1.49s/it]
 42%|████▏     | 10289/24420 [4:16:49<5:50:20,  1.49s/it]
 42%|████▏     | 10290/24420 [4:16:50<5:50:54,  1.49s/it]
                                                         
{'loss': 0.0904, 'learning_rate': 5e-05, 'epoch': 2.11}

 42%|████▏     | 10290/24420 [4:16:50<5:50:54,  1.49s/it]
 42%|████▏     | 10291/24420 [4:16:52<5:51:22,  1.49s/it]
 42%|████▏     | 10292/24420 [4:16:53<5:50:55,  1.49s/it]
 42%|████▏     | 10293/24420 [4:16:55<5:51:06,  1.49s/it]
 42%|████▏     | 10294/24420 [4:16:56<5:50:29,  1.49s/it]
 42%|████▏     | 10295/24420 [4:16:58<5:50:38,  1.49s/it]
                                                         
{'loss': 0.0996, 'learning_rate': 5e-05, 'epoch': 2.11}

 42%|████▏     | 10295/24420 [4:16:58<5:50:38,  1.49s/it]
 42%|████▏     | 10296/24420 [4:16:59<5:49:58,  1.49s/it]
 42%|████▏     | 10297/24420 [4:17:01<5:49:38,  1.49s/it]
 42%|████▏     | 10298/24420 [4:17:02<5:48:59,  1.48s/it]
 42%|████▏     | 10299/24420 [4:17:04<5:49:21,  1.48s/it]
 42%|████▏     | 10300/24420 [4:17:05<5:49:57,  1.49s/it]
                                                         
{'loss': 0.0891, 'learning_rate': 5e-05, 'epoch': 2.11}

 42%|████▏     | 10300/24420 [4:17:05<5:49:57,  1.49s/it]
 42%|████▏     | 10301/24420 [4:17:07<5:50:10,  1.49s/it]
 42%|████▏     | 10302/24420 [4:17:08<5:50:18,  1.49s/it]
 42%|████▏     | 10303/24420 [4:17:10<5:50:13,  1.49s/it]
 42%|████▏     | 10304/24420 [4:17:11<5:50:07,  1.49s/it]
 42%|████▏     | 10305/24420 [4:17:13<5:50:31,  1.49s/it]
                                                         
{'loss': 0.0968, 'learning_rate': 5e-05, 'epoch': 2.11}

 42%|████▏     | 10305/24420 [4:17:13<5:50:31,  1.49s/it]
 42%|████▏     | 10306/24420 [4:17:14<5:51:09,  1.49s/it]
 42%|████▏     | 10307/24420 [4:17:16<5:51:11,  1.49s/it]
 42%|████▏     | 10308/24420 [4:17:17<5:50:25,  1.49s/it]
 42%|████▏     | 10309/24420 [4:17:19<5:50:31,  1.49s/it]
 42%|████▏     | 10310/24420 [4:17:20<5:50:37,  1.49s/it]
                                                         
{'loss': 0.091, 'learning_rate': 5e-05, 'epoch': 2.11}

 42%|████▏     | 10310/24420 [4:17:20<5:50:37,  1.49s/it]
 42%|████▏     | 10311/24420 [4:17:22<5:50:37,  1.49s/it]
 42%|████▏     | 10312/24420 [4:17:23<5:49:37,  1.49s/it]
 42%|████▏     | 10313/24420 [4:17:25<5:49:08,  1.48s/it]
 42%|████▏     | 10314/24420 [4:17:26<5:49:41,  1.49s/it]
 42%|████▏     | 10315/24420 [4:17:28<5:49:41,  1.49s/it]
                                                         
{'loss': 0.1149, 'learning_rate': 5e-05, 'epoch': 2.11}

 42%|████▏     | 10315/24420 [4:17:28<5:49:41,  1.49s/it]
 42%|████▏     | 10316/24420 [4:17:29<5:49:43,  1.49s/it]
 42%|████▏     | 10317/24420 [4:17:31<5:49:08,  1.49s/it]
 42%|████▏     | 10318/24420 [4:17:32<5:49:12,  1.49s/it]
 42%|████▏     | 10319/24420 [4:17:34<5:49:20,  1.49s/it]
 42%|████▏     | 10320/24420 [4:17:35<5:49:46,  1.49s/it]
                                                         
{'loss': 0.0792, 'learning_rate': 5e-05, 'epoch': 2.11}

 42%|████▏     | 10320/24420 [4:17:35<5:49:46,  1.49s/it]
 42%|████▏     | 10321/24420 [4:17:37<5:50:27,  1.49s/it]
 42%|████▏     | 10322/24420 [4:17:38<5:50:28,  1.49s/it]
 42%|████▏     | 10323/24420 [4:17:40<5:50:19,  1.49s/it]
 42%|████▏     | 10324/24420 [4:17:41<5:50:46,  1.49s/it]
 42%|████▏     | 10325/24420 [4:17:43<5:50:06,  1.49s/it]
                                                         
{'loss': 0.0855, 'learning_rate': 5e-05, 'epoch': 2.11}

 42%|████▏     | 10325/24420 [4:17:43<5:50:06,  1.49s/it]
 42%|████▏     | 10326/24420 [4:17:44<5:49:13,  1.49s/it]
 42%|████▏     | 10327/24420 [4:17:45<5:49:23,  1.49s/it]
 42%|████▏     | 10328/24420 [4:17:47<5:50:10,  1.49s/it]
 42%|████▏     | 10329/24420 [4:17:48<5:50:29,  1.49s/it]
 42%|████▏     | 10330/24420 [4:17:50<5:50:39,  1.49s/it]
                                                         
{'loss': 0.0985, 'learning_rate': 5e-05, 'epoch': 2.12}

 42%|████▏     | 10330/24420 [4:17:50<5:50:39,  1.49s/it]
 42%|████▏     | 10331/24420 [4:17:51<5:51:01,  1.49s/it]
 42%|████▏     | 10332/24420 [4:17:53<5:52:10,  1.50s/it]
 42%|████▏     | 10333/24420 [4:17:54<5:51:20,  1.50s/it]
 42%|████▏     | 10334/24420 [4:17:56<5:51:05,  1.50s/it]
 42%|████▏     | 10335/24420 [4:17:57<5:50:32,  1.49s/it]
                                                         
{'loss': 0.1085, 'learning_rate': 5e-05, 'epoch': 2.12}

 42%|████▏     | 10335/24420 [4:17:57<5:50:32,  1.49s/it]
 42%|████▏     | 10336/24420 [4:17:59<5:50:08,  1.49s/it]
 42%|████▏     | 10337/24420 [4:18:00<5:49:50,  1.49s/it]
 42%|████▏     | 10338/24420 [4:18:02<5:49:55,  1.49s/it]
 42%|████▏     | 10339/24420 [4:18:03<5:50:08,  1.49s/it]
 42%|████▏     | 10340/24420 [4:18:05<5:49:49,  1.49s/it]
                                                         
{'loss': 0.0996, 'learning_rate': 5e-05, 'epoch': 2.12}

 42%|████▏     | 10340/24420 [4:18:05<5:49:49,  1.49s/it]
 42%|████▏     | 10341/24420 [4:18:06<5:49:41,  1.49s/it]
 42%|████▏     | 10342/24420 [4:18:08<5:49:23,  1.49s/it]
 42%|████▏     | 10343/24420 [4:18:09<5:49:06,  1.49s/it]
 42%|████▏     | 10344/24420 [4:18:11<5:48:40,  1.49s/it]
 42%|████▏     | 10345/24420 [4:18:12<5:49:00,  1.49s/it]
                                                         
{'loss': 0.0847, 'learning_rate': 5e-05, 'epoch': 2.12}

 42%|████▏     | 10345/24420 [4:18:12<5:49:00,  1.49s/it]
 42%|████▏     | 10346/24420 [4:18:14<5:49:06,  1.49s/it]
 42%|████▏     | 10347/24420 [4:18:15<5:48:51,  1.49s/it]
 42%|████▏     | 10348/24420 [4:18:17<5:48:37,  1.49s/it]
 42%|████▏     | 10349/24420 [4:18:18<5:48:44,  1.49s/it]
 42%|████▏     | 10350/24420 [4:18:20<5:48:50,  1.49s/it]
                                                         
{'loss': 0.1001, 'learning_rate': 5e-05, 'epoch': 2.12}

 42%|████▏     | 10350/24420 [4:18:20<5:48:50,  1.49s/it]
 42%|████▏     | 10351/24420 [4:18:21<5:49:34,  1.49s/it]
 42%|████▏     | 10352/24420 [4:18:23<5:49:14,  1.49s/it]
 42%|████▏     | 10353/24420 [4:18:24<5:49:07,  1.49s/it]
 42%|████▏     | 10354/24420 [4:18:26<5:48:49,  1.49s/it]
 42%|████▏     | 10355/24420 [4:18:27<5:48:57,  1.49s/it]
                                                         
{'loss': 0.0948, 'learning_rate': 5e-05, 'epoch': 2.12}

 42%|████▏     | 10355/24420 [4:18:27<5:48:57,  1.49s/it]
 42%|████▏     | 10356/24420 [4:18:29<5:49:08,  1.49s/it]
 42%|████▏     | 10357/24420 [4:18:30<5:49:27,  1.49s/it]
 42%|████▏     | 10358/24420 [4:18:32<5:49:22,  1.49s/it]
 42%|████▏     | 10359/24420 [4:18:33<5:49:15,  1.49s/it]
 42%|████▏     | 10360/24420 [4:18:35<5:49:07,  1.49s/it]
                                                         
{'loss': 0.089, 'learning_rate': 5e-05, 'epoch': 2.12}

 42%|████▏     | 10360/24420 [4:18:35<5:49:07,  1.49s/it]
 42%|████▏     | 10361/24420 [4:18:36<5:49:16,  1.49s/it]
 42%|████▏     | 10362/24420 [4:18:38<5:49:29,  1.49s/it]
 42%|████▏     | 10363/24420 [4:18:39<5:49:33,  1.49s/it]
 42%|████▏     | 10364/24420 [4:18:41<5:49:06,  1.49s/it]
 42%|████▏     | 10365/24420 [4:18:42<5:49:33,  1.49s/it]
                                                         
{'loss': 0.0779, 'learning_rate': 5e-05, 'epoch': 2.12}

 42%|████▏     | 10365/24420 [4:18:42<5:49:33,  1.49s/it]
 42%|████▏     | 10366/24420 [4:18:44<5:49:34,  1.49s/it]
 42%|████▏     | 10367/24420 [4:18:45<5:49:40,  1.49s/it]
 42%|████▏     | 10368/24420 [4:18:47<5:49:07,  1.49s/it]
 42%|████▏     | 10369/24420 [4:18:48<5:49:22,  1.49s/it]
 42%|████▏     | 10370/24420 [4:18:50<5:48:48,  1.49s/it]
                                                         
{'loss': 0.0766, 'learning_rate': 5e-05, 'epoch': 2.12}

 42%|████▏     | 10370/24420 [4:18:50<5:48:48,  1.49s/it]
 42%|████▏     | 10371/24420 [4:18:51<5:48:21,  1.49s/it]
 42%|████▏     | 10372/24420 [4:18:53<5:48:04,  1.49s/it]
 42%|████▏     | 10373/24420 [4:18:54<5:47:44,  1.49s/it]
 42%|████▏     | 10374/24420 [4:18:56<5:48:30,  1.49s/it]
 42%|████▏     | 10375/24420 [4:18:57<5:48:15,  1.49s/it]
                                                         
{'loss': 0.0974, 'learning_rate': 5e-05, 'epoch': 2.12}

 42%|████▏     | 10375/24420 [4:18:57<5:48:15,  1.49s/it]
 42%|████▏     | 10376/24420 [4:18:59<5:48:29,  1.49s/it]
 42%|████▏     | 10377/24420 [4:19:00<5:48:19,  1.49s/it]
 42%|████▏     | 10378/24420 [4:19:01<5:48:21,  1.49s/it]
 43%|████▎     | 10379/24420 [4:19:03<5:48:46,  1.49s/it]
 43%|████▎     | 10380/24420 [4:19:04<5:48:12,  1.49s/it]
                                                         
{'loss': 0.0883, 'learning_rate': 5e-05, 'epoch': 2.13}

 43%|████▎     | 10380/24420 [4:19:04<5:48:12,  1.49s/it]
 43%|████▎     | 10381/24420 [4:19:06<5:48:19,  1.49s/it]
 43%|████▎     | 10382/24420 [4:19:07<5:49:18,  1.49s/it]
 43%|████▎     | 10383/24420 [4:19:09<5:48:26,  1.49s/it]
 43%|████▎     | 10384/24420 [4:19:10<5:48:24,  1.49s/it]
 43%|████▎     | 10385/24420 [4:19:12<5:48:23,  1.49s/it]
                                                         
{'loss': 0.0841, 'learning_rate': 5e-05, 'epoch': 2.13}

 43%|████▎     | 10385/24420 [4:19:12<5:48:23,  1.49s/it]
 43%|████▎     | 10386/24420 [4:19:13<5:49:18,  1.49s/it]
 43%|████▎     | 10387/24420 [4:19:15<5:49:30,  1.49s/it]
 43%|████▎     | 10388/24420 [4:19:16<5:48:43,  1.49s/it]
 43%|████▎     | 10389/24420 [4:19:18<5:48:43,  1.49s/it]
 43%|████▎     | 10390/24420 [4:19:19<5:48:07,  1.49s/it]
                                                         
{'loss': 0.0971, 'learning_rate': 5e-05, 'epoch': 2.13}

 43%|████▎     | 10390/24420 [4:19:19<5:48:07,  1.49s/it]
 43%|████▎     | 10391/24420 [4:19:21<5:47:32,  1.49s/it]
 43%|████▎     | 10392/24420 [4:19:22<5:47:37,  1.49s/it]
 43%|████▎     | 10393/24420 [4:19:24<5:47:46,  1.49s/it]
 43%|████▎     | 10394/24420 [4:19:25<5:48:52,  1.49s/it]
 43%|████▎     | 10395/24420 [4:19:27<5:48:09,  1.49s/it]
                                                         
{'loss': 0.0916, 'learning_rate': 5e-05, 'epoch': 2.13}

 43%|████▎     | 10395/24420 [4:19:27<5:48:09,  1.49s/it]
 43%|████▎     | 10396/24420 [4:19:28<5:48:33,  1.49s/it]
 43%|████▎     | 10397/24420 [4:19:30<5:48:26,  1.49s/it]
 43%|████▎     | 10398/24420 [4:19:31<5:48:41,  1.49s/it]
 43%|████▎     | 10399/24420 [4:19:33<5:49:17,  1.49s/it]
 43%|████▎     | 10400/24420 [4:19:34<5:48:48,  1.49s/it]
                                                         
{'loss': 0.1002, 'learning_rate': 5e-05, 'epoch': 2.13}

 43%|████▎     | 10400/24420 [4:19:34<5:48:48,  1.49s/it]
 43%|████▎     | 10401/24420 [4:19:40<11:05:21,  2.85s/it]
 43%|████▎     | 10402/24420 [4:19:42<9:29:37,  2.44s/it] 
 43%|████▎     | 10403/24420 [4:19:43<8:24:02,  2.16s/it]
 43%|████▎     | 10404/24420 [4:19:45<7:36:46,  1.96s/it]
 43%|████▎     | 10405/24420 [4:19:46<7:03:40,  1.81s/it]
                                                         
{'loss': 0.0871, 'learning_rate': 5e-05, 'epoch': 2.13}

 43%|████▎     | 10405/24420 [4:19:46<7:03:40,  1.81s/it]
 43%|████▎     | 10406/24420 [4:19:48<6:40:52,  1.72s/it]
 43%|████▎     | 10407/24420 [4:19:49<6:24:33,  1.65s/it]
 43%|████▎     | 10408/24420 [4:19:51<6:13:32,  1.60s/it]
 43%|████▎     | 10409/24420 [4:19:52<6:05:38,  1.57s/it]
 43%|████▎     | 10410/24420 [4:19:54<6:00:13,  1.54s/it]
                                                         
{'loss': 0.0997, 'learning_rate': 5e-05, 'epoch': 2.13}

 43%|████▎     | 10410/24420 [4:19:54<6:00:13,  1.54s/it]
 43%|████▎     | 10411/24420 [4:19:55<5:56:23,  1.53s/it]
 43%|████▎     | 10412/24420 [4:19:57<5:53:36,  1.51s/it]
 43%|████▎     | 10413/24420 [4:19:58<5:51:15,  1.50s/it]
 43%|████▎     | 10414/24420 [4:20:00<5:50:02,  1.50s/it]
 43%|████▎     | 10415/24420 [4:20:01<5:48:56,  1.49s/it]
                                                         
{'loss': 0.087, 'learning_rate': 5e-05, 'epoch': 2.13}

 43%|████▎     | 10415/24420 [4:20:01<5:48:56,  1.49s/it]
 43%|████▎     | 10416/24420 [4:20:03<5:48:18,  1.49s/it]
 43%|████▎     | 10417/24420 [4:20:04<5:51:26,  1.51s/it]
 43%|████▎     | 10418/24420 [4:20:06<5:49:32,  1.50s/it]
 43%|████▎     | 10419/24420 [4:20:07<5:50:06,  1.50s/it]
 43%|████▎     | 10420/24420 [4:20:09<5:49:38,  1.50s/it]
                                                         
{'loss': 0.1369, 'learning_rate': 5e-05, 'epoch': 2.13}

 43%|████▎     | 10420/24420 [4:20:09<5:49:38,  1.50s/it]
 43%|████▎     | 10421/24420 [4:20:10<5:49:32,  1.50s/it]
 43%|████▎     | 10422/24420 [4:20:12<5:48:49,  1.50s/it]
 43%|████▎     | 10423/24420 [4:20:13<5:48:00,  1.49s/it]
 43%|████▎     | 10424/24420 [4:20:15<5:48:22,  1.49s/it]
 43%|████▎     | 10425/24420 [4:20:16<5:48:13,  1.49s/it]
                                                         
{'loss': 0.0885, 'learning_rate': 5e-05, 'epoch': 2.13}

 43%|████▎     | 10425/24420 [4:20:16<5:48:13,  1.49s/it]
 43%|████▎     | 10426/24420 [4:20:18<5:49:17,  1.50s/it]
 43%|████▎     | 10427/24420 [4:20:19<5:47:31,  1.49s/it]
 43%|████▎     | 10428/24420 [4:20:21<5:47:10,  1.49s/it]
 43%|████▎     | 10429/24420 [4:20:22<5:46:38,  1.49s/it]
 43%|████▎     | 10430/24420 [4:20:24<5:46:34,  1.49s/it]
                                                         
{'loss': 0.0969, 'learning_rate': 5e-05, 'epoch': 2.14}

 43%|████▎     | 10430/24420 [4:20:24<5:46:34,  1.49s/it]
 43%|████▎     | 10431/24420 [4:20:25<5:47:43,  1.49s/it]
 43%|████▎     | 10432/24420 [4:20:27<5:48:29,  1.49s/it]
 43%|████▎     | 10433/24420 [4:20:28<5:48:47,  1.50s/it]
 43%|████▎     | 10434/24420 [4:20:30<5:48:50,  1.50s/it]
 43%|████▎     | 10435/24420 [4:20:31<5:48:29,  1.50s/it]
                                                         
{'loss': 0.0844, 'learning_rate': 5e-05, 'epoch': 2.14}

 43%|████▎     | 10435/24420 [4:20:31<5:48:29,  1.50s/it]
 43%|████▎     | 10436/24420 [4:20:33<5:48:29,  1.50s/it]
 43%|████▎     | 10437/24420 [4:20:34<5:48:19,  1.49s/it]
 43%|████▎     | 10438/24420 [4:20:35<5:48:36,  1.50s/it]
 43%|████▎     | 10439/24420 [4:20:37<5:48:24,  1.50s/it]
 43%|████▎     | 10440/24420 [4:20:38<5:48:43,  1.50s/it]
                                                         
{'loss': 0.1071, 'learning_rate': 5e-05, 'epoch': 2.14}

 43%|████▎     | 10440/24420 [4:20:38<5:48:43,  1.50s/it]
 43%|████▎     | 10441/24420 [4:20:40<5:49:25,  1.50s/it]
 43%|████▎     | 10442/24420 [4:20:42<5:49:34,  1.50s/it]
 43%|████▎     | 10443/24420 [4:20:43<5:49:06,  1.50s/it]
 43%|████▎     | 10444/24420 [4:20:44<5:48:25,  1.50s/it]
 43%|████▎     | 10445/24420 [4:20:46<5:47:50,  1.49s/it]
                                                         
{'loss': 0.1059, 'learning_rate': 5e-05, 'epoch': 2.14}

 43%|████▎     | 10445/24420 [4:20:46<5:47:50,  1.49s/it]
 43%|████▎     | 10446/24420 [4:20:47<5:48:19,  1.50s/it]
 43%|████▎     | 10447/24420 [4:20:49<5:47:18,  1.49s/it]
 43%|████▎     | 10448/24420 [4:20:50<5:47:45,  1.49s/it]
 43%|████▎     | 10449/24420 [4:20:52<5:46:47,  1.49s/it]
 43%|████▎     | 10450/24420 [4:20:53<5:46:30,  1.49s/it]
                                                         
{'loss': 0.0967, 'learning_rate': 5e-05, 'epoch': 2.14}

 43%|████▎     | 10450/24420 [4:20:53<5:46:30,  1.49s/it]
 43%|████▎     | 10451/24420 [4:20:55<5:46:28,  1.49s/it]
 43%|████▎     | 10452/24420 [4:20:56<5:46:33,  1.49s/it]
 43%|████▎     | 10453/24420 [4:20:58<5:46:52,  1.49s/it]
 43%|████▎     | 10454/24420 [4:20:59<5:46:09,  1.49s/it]
 43%|████▎     | 10455/24420 [4:21:01<5:45:59,  1.49s/it]
                                                         
{'loss': 0.0889, 'learning_rate': 5e-05, 'epoch': 2.14}

 43%|████▎     | 10455/24420 [4:21:01<5:45:59,  1.49s/it]
 43%|████▎     | 10456/24420 [4:21:02<5:45:43,  1.49s/it]
 43%|████▎     | 10457/24420 [4:21:04<5:45:59,  1.49s/it]
 43%|████▎     | 10458/24420 [4:21:05<5:46:07,  1.49s/it]
 43%|████▎     | 10459/24420 [4:21:07<5:46:17,  1.49s/it]
 43%|████▎     | 10460/24420 [4:21:08<5:46:32,  1.49s/it]
                                                         
{'loss': 0.0859, 'learning_rate': 5e-05, 'epoch': 2.14}

 43%|████▎     | 10460/24420 [4:21:08<5:46:32,  1.49s/it]
 43%|████▎     | 10461/24420 [4:21:10<5:47:25,  1.49s/it]
 43%|████▎     | 10462/24420 [4:21:11<5:47:15,  1.49s/it]
 43%|████▎     | 10463/24420 [4:21:13<5:46:29,  1.49s/it]
 43%|████▎     | 10464/24420 [4:21:14<5:46:28,  1.49s/it]
 43%|████▎     | 10465/24420 [4:21:16<5:46:37,  1.49s/it]
                                                         
{'loss': 0.0996, 'learning_rate': 5e-05, 'epoch': 2.14}

 43%|████▎     | 10465/24420 [4:21:16<5:46:37,  1.49s/it]
 43%|████▎     | 10466/24420 [4:21:17<5:46:36,  1.49s/it]
 43%|████▎     | 10467/24420 [4:21:19<5:46:13,  1.49s/it]
 43%|████▎     | 10468/24420 [4:21:20<5:46:04,  1.49s/it]
 43%|████▎     | 10469/24420 [4:21:22<5:45:47,  1.49s/it]
 43%|████▎     | 10470/24420 [4:21:23<5:45:31,  1.49s/it]
                                                         
{'loss': 0.1086, 'learning_rate': 5e-05, 'epoch': 2.14}

 43%|████▎     | 10470/24420 [4:21:23<5:45:31,  1.49s/it]
 43%|████▎     | 10471/24420 [4:21:25<5:45:30,  1.49s/it]
 43%|████▎     | 10472/24420 [4:21:26<5:45:45,  1.49s/it]
 43%|████▎     | 10473/24420 [4:21:28<5:45:21,  1.49s/it]
 43%|████▎     | 10474/24420 [4:21:29<5:45:13,  1.49s/it]
 43%|████▎     | 10475/24420 [4:21:31<5:45:13,  1.49s/it]
                                                         
{'loss': 0.0806, 'learning_rate': 5e-05, 'epoch': 2.14}

 43%|████▎     | 10475/24420 [4:21:31<5:45:13,  1.49s/it]
 43%|████▎     | 10476/24420 [4:21:32<5:46:24,  1.49s/it]
 43%|████▎     | 10477/24420 [4:21:34<5:46:19,  1.49s/it]
 43%|████▎     | 10478/24420 [4:21:35<5:45:44,  1.49s/it]
 43%|████▎     | 10479/24420 [4:21:37<5:45:27,  1.49s/it]
 43%|████▎     | 10480/24420 [4:21:38<5:46:28,  1.49s/it]
                                                         
{'loss': 0.0818, 'learning_rate': 5e-05, 'epoch': 2.15}

 43%|████▎     | 10480/24420 [4:21:38<5:46:28,  1.49s/it]
 43%|████▎     | 10481/24420 [4:21:40<5:46:34,  1.49s/it]
 43%|████▎     | 10482/24420 [4:21:41<5:46:40,  1.49s/it]
 43%|████▎     | 10483/24420 [4:21:43<5:46:06,  1.49s/it]
 43%|████▎     | 10484/24420 [4:21:44<5:45:39,  1.49s/it]
 43%|████▎     | 10485/24420 [4:21:46<5:45:33,  1.49s/it]
                                                         
{'loss': 0.0991, 'learning_rate': 5e-05, 'epoch': 2.15}

 43%|████▎     | 10485/24420 [4:21:46<5:45:33,  1.49s/it]
 43%|████▎     | 10486/24420 [4:21:47<5:45:37,  1.49s/it]
 43%|████▎     | 10487/24420 [4:21:48<5:45:39,  1.49s/it]
 43%|████▎     | 10488/24420 [4:21:50<5:45:28,  1.49s/it]
 43%|████▎     | 10489/24420 [4:21:51<5:45:33,  1.49s/it]
 43%|████▎     | 10490/24420 [4:21:53<5:46:02,  1.49s/it]
                                                         
{'loss': 0.0906, 'learning_rate': 5e-05, 'epoch': 2.15}

 43%|████▎     | 10490/24420 [4:21:53<5:46:02,  1.49s/it]
 43%|████▎     | 10491/24420 [4:21:54<5:46:06,  1.49s/it]
 43%|████▎     | 10492/24420 [4:21:56<5:46:13,  1.49s/it]
 43%|████▎     | 10493/24420 [4:21:57<5:45:47,  1.49s/it]
 43%|████▎     | 10494/24420 [4:21:59<5:45:38,  1.49s/it]
 43%|████▎     | 10495/24420 [4:22:00<5:45:11,  1.49s/it]
                                                         
{'loss': 0.0767, 'learning_rate': 5e-05, 'epoch': 2.15}

 43%|████▎     | 10495/24420 [4:22:00<5:45:11,  1.49s/it]
 43%|████▎     | 10496/24420 [4:22:02<5:45:24,  1.49s/it]
 43%|████▎     | 10497/24420 [4:22:03<5:45:37,  1.49s/it]
 43%|████▎     | 10498/24420 [4:22:05<5:45:12,  1.49s/it]
 43%|████▎     | 10499/24420 [4:22:06<5:45:06,  1.49s/it]
 43%|████▎     | 10500/24420 [4:22:08<5:45:25,  1.49s/it]
                                                         
{'loss': 0.0998, 'learning_rate': 5e-05, 'epoch': 2.15}

 43%|████▎     | 10500/24420 [4:22:08<5:45:25,  1.49s/it]
 43%|████▎     | 10501/24420 [4:22:09<5:45:25,  1.49s/it]
 43%|████▎     | 10502/24420 [4:22:11<5:44:46,  1.49s/it]
 43%|████▎     | 10503/24420 [4:22:12<5:44:20,  1.48s/it]
 43%|████▎     | 10504/24420 [4:22:14<5:44:18,  1.48s/it]
 43%|████▎     | 10505/24420 [4:22:15<5:44:51,  1.49s/it]
                                                         
{'loss': 0.078, 'learning_rate': 5e-05, 'epoch': 2.15}

 43%|████▎     | 10505/24420 [4:22:15<5:44:51,  1.49s/it]
 43%|████▎     | 10506/24420 [4:22:17<5:44:55,  1.49s/it]
 43%|████▎     | 10507/24420 [4:22:18<5:44:33,  1.49s/it]
 43%|████▎     | 10508/24420 [4:22:20<5:53:29,  1.52s/it]
 43%|████▎     | 10509/24420 [4:22:21<5:51:02,  1.51s/it]
 43%|████▎     | 10510/24420 [4:22:23<5:50:02,  1.51s/it]
                                                         
{'loss': 0.104, 'learning_rate': 5e-05, 'epoch': 2.15}

 43%|████▎     | 10510/24420 [4:22:23<5:50:02,  1.51s/it]
 43%|████▎     | 10511/24420 [4:22:24<5:48:14,  1.50s/it]
 43%|████▎     | 10512/24420 [4:22:26<5:46:52,  1.50s/it]
 43%|████▎     | 10513/24420 [4:22:27<5:46:55,  1.50s/it]
 43%|████▎     | 10514/24420 [4:22:29<5:46:42,  1.50s/it]
 43%|████▎     | 10515/24420 [4:22:30<5:46:12,  1.49s/it]
                                                         
{'loss': 0.088, 'learning_rate': 5e-05, 'epoch': 2.15}

 43%|████▎     | 10515/24420 [4:22:30<5:46:12,  1.49s/it]
 43%|████▎     | 10516/24420 [4:22:32<5:45:58,  1.49s/it]
 43%|████▎     | 10517/24420 [4:22:33<5:45:58,  1.49s/it]
 43%|████▎     | 10518/24420 [4:22:35<5:45:50,  1.49s/it]
 43%|████▎     | 10519/24420 [4:22:36<5:45:24,  1.49s/it]
 43%|████▎     | 10520/24420 [4:22:38<5:45:10,  1.49s/it]
                                                         
{'loss': 0.0844, 'learning_rate': 5e-05, 'epoch': 2.15}

 43%|████▎     | 10520/24420 [4:22:38<5:45:10,  1.49s/it]
 43%|████▎     | 10521/24420 [4:22:39<5:44:51,  1.49s/it]
 43%|████▎     | 10522/24420 [4:22:41<5:45:31,  1.49s/it]
 43%|████▎     | 10523/24420 [4:22:42<5:45:26,  1.49s/it]
 43%|████▎     | 10524/24420 [4:22:44<5:45:12,  1.49s/it]
 43%|████▎     | 10525/24420 [4:22:45<5:45:13,  1.49s/it]
                                                         
{'loss': 0.0901, 'learning_rate': 5e-05, 'epoch': 2.15}

 43%|████▎     | 10525/24420 [4:22:45<5:45:13,  1.49s/it]
 43%|████▎     | 10526/24420 [4:22:47<5:44:39,  1.49s/it]
 43%|████▎     | 10527/24420 [4:22:48<5:45:09,  1.49s/it]
 43%|████▎     | 10528/24420 [4:22:50<5:44:38,  1.49s/it]
 43%|████▎     | 10529/24420 [4:22:51<5:43:45,  1.48s/it]
 43%|████▎     | 10530/24420 [4:22:53<5:44:13,  1.49s/it]
                                                         
{'loss': 0.0974, 'learning_rate': 5e-05, 'epoch': 2.16}

 43%|████▎     | 10530/24420 [4:22:53<5:44:13,  1.49s/it]
 43%|████▎     | 10531/24420 [4:22:54<5:45:04,  1.49s/it]
 43%|████▎     | 10532/24420 [4:22:56<5:44:47,  1.49s/it]
 43%|████▎     | 10533/24420 [4:22:57<5:44:57,  1.49s/it]
 43%|████▎     | 10534/24420 [4:22:59<5:44:47,  1.49s/it]
 43%|████▎     | 10535/24420 [4:23:00<5:45:02,  1.49s/it]
                                                         
{'loss': 0.0913, 'learning_rate': 5e-05, 'epoch': 2.16}

 43%|████▎     | 10535/24420 [4:23:00<5:45:02,  1.49s/it]
 43%|████▎     | 10536/24420 [4:23:02<5:45:46,  1.49s/it]
 43%|████▎     | 10537/24420 [4:23:03<5:45:33,  1.49s/it]
 43%|████▎     | 10538/24420 [4:23:05<5:45:04,  1.49s/it]
 43%|████▎     | 10539/24420 [4:23:06<5:45:55,  1.50s/it]
 43%|████▎     | 10540/24420 [4:23:08<5:45:11,  1.49s/it]
                                                         
{'loss': 0.0899, 'learning_rate': 5e-05, 'epoch': 2.16}

 43%|████▎     | 10540/24420 [4:23:08<5:45:11,  1.49s/it]
 43%|████▎     | 10541/24420 [4:23:09<5:44:54,  1.49s/it]
 43%|████▎     | 10542/24420 [4:23:11<5:44:49,  1.49s/it]
 43%|████▎     | 10543/24420 [4:23:12<5:45:46,  1.49s/it]
 43%|████▎     | 10544/24420 [4:23:14<5:45:27,  1.49s/it]
 43%|████▎     | 10545/24420 [4:23:15<5:45:22,  1.49s/it]
                                                         
{'loss': 0.0872, 'learning_rate': 5e-05, 'epoch': 2.16}

 43%|████▎     | 10545/24420 [4:23:15<5:45:22,  1.49s/it]
 43%|████▎     | 10546/24420 [4:23:17<5:45:14,  1.49s/it]
 43%|████▎     | 10547/24420 [4:23:18<5:44:52,  1.49s/it]
 43%|████▎     | 10548/24420 [4:23:20<5:44:49,  1.49s/it]
 43%|████▎     | 10549/24420 [4:23:21<5:44:24,  1.49s/it]
 43%|████▎     | 10550/24420 [4:23:22<5:44:18,  1.49s/it]
                                                         
{'loss': 0.0943, 'learning_rate': 5e-05, 'epoch': 2.16}

 43%|████▎     | 10550/24420 [4:23:22<5:44:18,  1.49s/it]
 43%|████▎     | 10551/24420 [4:23:24<5:44:17,  1.49s/it]
 43%|████▎     | 10552/24420 [4:23:25<5:43:57,  1.49s/it]
 43%|████▎     | 10553/24420 [4:23:27<5:44:02,  1.49s/it]
 43%|████▎     | 10554/24420 [4:23:28<5:43:37,  1.49s/it]
 43%|████▎     | 10555/24420 [4:23:30<5:44:07,  1.49s/it]
                                                         
{'loss': 0.0809, 'learning_rate': 5e-05, 'epoch': 2.16}

 43%|████▎     | 10555/24420 [4:23:30<5:44:07,  1.49s/it]
 43%|████▎     | 10556/24420 [4:23:31<5:44:39,  1.49s/it]
 43%|████▎     | 10557/24420 [4:23:33<5:44:23,  1.49s/it]
 43%|████▎     | 10558/24420 [4:23:34<5:43:56,  1.49s/it]
 43%|████▎     | 10559/24420 [4:23:36<5:43:13,  1.49s/it]
 43%|████▎     | 10560/24420 [4:23:37<5:43:09,  1.49s/it]
                                                         
{'loss': 0.0968, 'learning_rate': 5e-05, 'epoch': 2.16}

 43%|████▎     | 10560/24420 [4:23:37<5:43:09,  1.49s/it]
 43%|████▎     | 10561/24420 [4:23:39<5:44:13,  1.49s/it]
 43%|████▎     | 10562/24420 [4:23:40<5:44:11,  1.49s/it]
 43%|████▎     | 10563/24420 [4:23:42<5:43:48,  1.49s/it]
 43%|████▎     | 10564/24420 [4:23:43<5:44:05,  1.49s/it]
 43%|████▎     | 10565/24420 [4:23:45<5:44:02,  1.49s/it]
                                                         
{'loss': 0.0841, 'learning_rate': 5e-05, 'epoch': 2.16}

 43%|████▎     | 10565/24420 [4:23:45<5:44:02,  1.49s/it]
 43%|████▎     | 10566/24420 [4:23:46<5:45:13,  1.50s/it]
 43%|████▎     | 10567/24420 [4:23:48<5:44:43,  1.49s/it]
 43%|████▎     | 10568/24420 [4:23:49<5:45:01,  1.49s/it]
 43%|████▎     | 10569/24420 [4:23:51<5:44:47,  1.49s/it]
 43%|████▎     | 10570/24420 [4:23:52<5:44:14,  1.49s/it]
                                                         
{'loss': 0.0934, 'learning_rate': 5e-05, 'epoch': 2.16}

 43%|████▎     | 10570/24420 [4:23:52<5:44:14,  1.49s/it]
 43%|████▎     | 10571/24420 [4:23:54<5:43:31,  1.49s/it]
 43%|████▎     | 10572/24420 [4:23:55<5:43:47,  1.49s/it]
 43%|████▎     | 10573/24420 [4:23:57<5:45:15,  1.50s/it]
 43%|████▎     | 10574/24420 [4:23:58<5:44:40,  1.49s/it]
 43%|████▎     | 10575/24420 [4:24:00<5:43:59,  1.49s/it]
                                                         
{'loss': 0.0968, 'learning_rate': 5e-05, 'epoch': 2.17}

 43%|████▎     | 10575/24420 [4:24:00<5:43:59,  1.49s/it]
 43%|████▎     | 10576/24420 [4:24:01<5:43:32,  1.49s/it]
 43%|████▎     | 10577/24420 [4:24:03<5:44:23,  1.49s/it]
 43%|████▎     | 10578/24420 [4:24:04<5:44:49,  1.49s/it]
 43%|████▎     | 10579/24420 [4:24:06<5:44:12,  1.49s/it]
 43%|████▎     | 10580/24420 [4:24:07<5:45:04,  1.50s/it]
                                                         
{'loss': 0.1094, 'learning_rate': 5e-05, 'epoch': 2.17}

 43%|████▎     | 10580/24420 [4:24:07<5:45:04,  1.50s/it]
 43%|████▎     | 10581/24420 [4:24:09<5:44:55,  1.50s/it]
 43%|████▎     | 10582/24420 [4:24:10<5:44:07,  1.49s/it]
 43%|████▎     | 10583/24420 [4:24:12<5:43:57,  1.49s/it]
 43%|████▎     | 10584/24420 [4:24:13<5:43:19,  1.49s/it]
 43%|████▎     | 10585/24420 [4:24:15<5:44:35,  1.49s/it]
                                                         
{'loss': 0.0875, 'learning_rate': 5e-05, 'epoch': 2.17}

 43%|████▎     | 10585/24420 [4:24:15<5:44:35,  1.49s/it]
 43%|████▎     | 10586/24420 [4:24:16<5:44:44,  1.50s/it]
 43%|████▎     | 10587/24420 [4:24:18<5:44:16,  1.49s/it]
 43%|████▎     | 10588/24420 [4:24:19<5:44:30,  1.49s/it]
 43%|████▎     | 10589/24420 [4:24:21<5:43:41,  1.49s/it]
 43%|████▎     | 10590/24420 [4:24:22<5:44:23,  1.49s/it]
                                                         
{'loss': 0.0997, 'learning_rate': 5e-05, 'epoch': 2.17}

 43%|████▎     | 10590/24420 [4:24:22<5:44:23,  1.49s/it]
 43%|████▎     | 10591/24420 [4:24:24<5:43:46,  1.49s/it]
 43%|████▎     | 10592/24420 [4:24:25<5:43:26,  1.49s/it]
 43%|████▎     | 10593/24420 [4:24:27<5:42:50,  1.49s/it]
 43%|████▎     | 10594/24420 [4:24:28<5:42:28,  1.49s/it]
 43%|████▎     | 10595/24420 [4:24:30<5:43:08,  1.49s/it]
                                                         
{'loss': 0.0888, 'learning_rate': 5e-05, 'epoch': 2.17}

 43%|████▎     | 10595/24420 [4:24:30<5:43:08,  1.49s/it]
 43%|████▎     | 10596/24420 [4:24:31<5:43:08,  1.49s/it]
 43%|████▎     | 10597/24420 [4:24:33<5:42:41,  1.49s/it]
 43%|████▎     | 10598/24420 [4:24:34<5:42:19,  1.49s/it]
 43%|████▎     | 10599/24420 [4:24:36<5:42:40,  1.49s/it]
 43%|████▎     | 10600/24420 [4:24:37<5:42:29,  1.49s/it]
                                                         
{'loss': 0.0968, 'learning_rate': 5e-05, 'epoch': 2.17}

 43%|████▎     | 10600/24420 [4:24:37<5:42:29,  1.49s/it]
 43%|████▎     | 10601/24420 [4:24:42<9:34:17,  2.49s/it]
 43%|████▎     | 10602/24420 [4:24:43<8:24:53,  2.19s/it]
 43%|████▎     | 10603/24420 [4:24:45<7:36:30,  1.98s/it]
 43%|████▎     | 10604/24420 [4:24:46<7:03:02,  1.84s/it]
 43%|████▎     | 10605/24420 [4:24:48<6:39:26,  1.73s/it]
                                                         
{'loss': 0.0956, 'learning_rate': 5e-05, 'epoch': 2.17}

 43%|████▎     | 10605/24420 [4:24:48<6:39:26,  1.73s/it]
 43%|████▎     | 10606/24420 [4:24:49<6:23:23,  1.67s/it]
 43%|████▎     | 10607/24420 [4:24:51<6:11:17,  1.61s/it]
 43%|████▎     | 10608/24420 [4:24:52<6:02:24,  1.57s/it]
 43%|████▎     | 10609/24420 [4:24:54<5:56:23,  1.55s/it]
 43%|████▎     | 10610/24420 [4:24:55<5:52:00,  1.53s/it]
                                                         
{'loss': 0.0853, 'learning_rate': 5e-05, 'epoch': 2.17}

 43%|████▎     | 10610/24420 [4:24:55<5:52:00,  1.53s/it]
 43%|████▎     | 10611/24420 [4:24:57<5:49:08,  1.52s/it]
 43%|████▎     | 10612/24420 [4:24:58<5:46:28,  1.51s/it]
 43%|████▎     | 10613/24420 [4:25:00<5:45:19,  1.50s/it]
 43%|████▎     | 10614/24420 [4:25:01<5:44:35,  1.50s/it]
 43%|████▎     | 10615/24420 [4:25:03<5:44:01,  1.50s/it]
                                                         
{'loss': 0.0878, 'learning_rate': 5e-05, 'epoch': 2.17}

 43%|████▎     | 10615/24420 [4:25:03<5:44:01,  1.50s/it]
 43%|████▎     | 10616/24420 [4:25:04<5:43:20,  1.49s/it]
 43%|████▎     | 10617/24420 [4:25:06<5:43:22,  1.49s/it]
 43%|████▎     | 10618/24420 [4:25:07<5:42:56,  1.49s/it]
 43%|████▎     | 10619/24420 [4:25:09<5:42:38,  1.49s/it]
 43%|████▎     | 10620/24420 [4:25:10<5:44:20,  1.50s/it]
                                                         
{'loss': 0.1021, 'learning_rate': 5e-05, 'epoch': 2.17}

 43%|████▎     | 10620/24420 [4:25:10<5:44:20,  1.50s/it]
 43%|████▎     | 10621/24420 [4:25:12<5:43:28,  1.49s/it]
 43%|████▎     | 10622/24420 [4:25:13<5:42:44,  1.49s/it]
 44%|████▎     | 10623/24420 [4:25:15<5:42:34,  1.49s/it]
 44%|████▎     | 10624/24420 [4:25:16<5:42:32,  1.49s/it]
 44%|████▎     | 10625/24420 [4:25:18<5:42:25,  1.49s/it]
                                                         
{'loss': 0.0891, 'learning_rate': 5e-05, 'epoch': 2.18}

 44%|████▎     | 10625/24420 [4:25:18<5:42:25,  1.49s/it]
 44%|████▎     | 10626/24420 [4:25:19<5:42:10,  1.49s/it]
 44%|████▎     | 10627/24420 [4:25:21<5:42:21,  1.49s/it]
 44%|████▎     | 10628/24420 [4:25:22<5:42:40,  1.49s/it]
 44%|████▎     | 10629/24420 [4:25:24<5:43:02,  1.49s/it]
 44%|████▎     | 10630/24420 [4:25:25<5:43:19,  1.49s/it]
                                                         
{'loss': 0.1043, 'learning_rate': 5e-05, 'epoch': 2.18}

 44%|████▎     | 10630/24420 [4:25:25<5:43:19,  1.49s/it]
 44%|████▎     | 10631/24420 [4:25:27<5:42:41,  1.49s/it]
 44%|████▎     | 10632/24420 [4:25:28<5:42:36,  1.49s/it]
 44%|████▎     | 10633/24420 [4:25:30<5:42:17,  1.49s/it]
 44%|████▎     | 10634/24420 [4:25:31<5:42:18,  1.49s/it]
 44%|████▎     | 10635/24420 [4:25:33<5:42:03,  1.49s/it]
                                                         
{'loss': 0.0869, 'learning_rate': 5e-05, 'epoch': 2.18}

 44%|████▎     | 10635/24420 [4:25:33<5:42:03,  1.49s/it]
 44%|████▎     | 10636/24420 [4:25:34<5:42:27,  1.49s/it]
 44%|████▎     | 10637/24420 [4:25:36<5:42:20,  1.49s/it]
 44%|████▎     | 10638/24420 [4:25:37<5:42:01,  1.49s/it]
 44%|████▎     | 10639/24420 [4:25:38<5:41:37,  1.49s/it]
 44%|████▎     | 10640/24420 [4:25:40<5:41:43,  1.49s/it]
                                                         
{'loss': 0.1097, 'learning_rate': 5e-05, 'epoch': 2.18}

 44%|████▎     | 10640/24420 [4:25:40<5:41:43,  1.49s/it]
 44%|████▎     | 10641/24420 [4:25:41<5:41:51,  1.49s/it]
 44%|████▎     | 10642/24420 [4:25:43<5:41:56,  1.49s/it]
 44%|████▎     | 10643/24420 [4:25:44<5:41:54,  1.49s/it]
 44%|████▎     | 10644/24420 [4:25:46<5:41:19,  1.49s/it]
 44%|████▎     | 10645/24420 [4:25:47<5:41:07,  1.49s/it]
                                                         
{'loss': 0.0958, 'learning_rate': 5e-05, 'epoch': 2.18}

 44%|████▎     | 10645/24420 [4:25:47<5:41:07,  1.49s/it]
 44%|████▎     | 10646/24420 [4:25:49<5:41:17,  1.49s/it]
 44%|████▎     | 10647/24420 [4:25:50<5:41:03,  1.49s/it]
 44%|████▎     | 10648/24420 [4:25:52<5:40:50,  1.48s/it]
 44%|████▎     | 10649/24420 [4:25:53<5:41:02,  1.49s/it]
 44%|████▎     | 10650/24420 [4:25:55<5:41:05,  1.49s/it]
                                                         
{'loss': 0.0987, 'learning_rate': 5e-05, 'epoch': 2.18}

 44%|████▎     | 10650/24420 [4:25:55<5:41:05,  1.49s/it]
 44%|████▎     | 10651/24420 [4:25:56<5:41:16,  1.49s/it]
 44%|████▎     | 10652/24420 [4:25:58<5:41:20,  1.49s/it]
 44%|████▎     | 10653/24420 [4:25:59<5:41:21,  1.49s/it]
 44%|████▎     | 10654/24420 [4:26:01<5:41:07,  1.49s/it]
 44%|████▎     | 10655/24420 [4:26:02<5:41:24,  1.49s/it]
                                                         
{'loss': 0.0993, 'learning_rate': 5e-05, 'epoch': 2.18}

 44%|████▎     | 10655/24420 [4:26:02<5:41:24,  1.49s/it]
 44%|████▎     | 10656/24420 [4:26:04<5:42:04,  1.49s/it]
 44%|████▎     | 10657/24420 [4:26:05<5:41:39,  1.49s/it]
 44%|████▎     | 10658/24420 [4:26:07<5:41:20,  1.49s/it]
 44%|████▎     | 10659/24420 [4:26:08<5:40:57,  1.49s/it]
 44%|████▎     | 10660/24420 [4:26:10<5:40:32,  1.48s/it]
                                                         
{'loss': 0.1052, 'learning_rate': 5e-05, 'epoch': 2.18}

 44%|████▎     | 10660/24420 [4:26:10<5:40:32,  1.48s/it]
 44%|████▎     | 10661/24420 [4:26:11<5:41:24,  1.49s/it]
 44%|████▎     | 10662/24420 [4:26:13<5:40:55,  1.49s/it]
 44%|████▎     | 10663/24420 [4:26:14<5:40:34,  1.49s/it]
 44%|████▎     | 10664/24420 [4:26:16<5:41:17,  1.49s/it]
 44%|████▎     | 10665/24420 [4:26:17<5:41:23,  1.49s/it]
                                                         
{'loss': 0.0962, 'learning_rate': 5e-05, 'epoch': 2.18}

 44%|████▎     | 10665/24420 [4:26:17<5:41:23,  1.49s/it]
 44%|████▎     | 10666/24420 [4:26:19<5:40:26,  1.49s/it]
 44%|████▎     | 10667/24420 [4:26:20<5:40:34,  1.49s/it]
 44%|████▎     | 10668/24420 [4:26:22<5:39:58,  1.48s/it]
 44%|████▎     | 10669/24420 [4:26:23<5:41:43,  1.49s/it]
 44%|████▎     | 10670/24420 [4:26:25<5:42:53,  1.50s/it]
                                                         
{'loss': 0.1024, 'learning_rate': 5e-05, 'epoch': 2.18}

 44%|████▎     | 10670/24420 [4:26:25<5:42:53,  1.50s/it]
 44%|████▎     | 10671/24420 [4:26:26<5:43:24,  1.50s/it]
 44%|████▎     | 10672/24420 [4:26:28<5:43:12,  1.50s/it]
 44%|████▎     | 10673/24420 [4:26:29<5:43:20,  1.50s/it]
 44%|████▎     | 10674/24420 [4:26:31<5:42:33,  1.50s/it]
 44%|████▎     | 10675/24420 [4:26:32<5:42:43,  1.50s/it]
                                                         
{'loss': 0.0975, 'learning_rate': 5e-05, 'epoch': 2.19}

 44%|████▎     | 10675/24420 [4:26:32<5:42:43,  1.50s/it]
 44%|████▎     | 10676/24420 [4:26:34<5:42:13,  1.49s/it]
 44%|████▎     | 10677/24420 [4:26:35<5:41:45,  1.49s/it]
 44%|████▎     | 10678/24420 [4:26:37<5:41:38,  1.49s/it]
 44%|████▎     | 10679/24420 [4:26:38<5:41:25,  1.49s/it]
 44%|████▎     | 10680/24420 [4:26:40<5:40:53,  1.49s/it]
                                                         
{'loss': 0.0833, 'learning_rate': 5e-05, 'epoch': 2.19}

 44%|████▎     | 10680/24420 [4:26:40<5:40:53,  1.49s/it]
 44%|████▎     | 10681/24420 [4:26:41<5:41:35,  1.49s/it]
 44%|████▎     | 10682/24420 [4:26:43<5:41:40,  1.49s/it]
 44%|████▎     | 10683/24420 [4:26:44<5:41:31,  1.49s/it]
 44%|████▍     | 10684/24420 [4:26:46<5:41:48,  1.49s/it]
 44%|████▍     | 10685/24420 [4:26:47<5:41:35,  1.49s/it]
                                                         
{'loss': 0.0942, 'learning_rate': 5e-05, 'epoch': 2.19}

 44%|████▍     | 10685/24420 [4:26:47<5:41:35,  1.49s/it]
 44%|████▍     | 10686/24420 [4:26:49<5:41:54,  1.49s/it]
 44%|████▍     | 10687/24420 [4:26:50<5:41:20,  1.49s/it]
 44%|████▍     | 10688/24420 [4:26:51<5:41:16,  1.49s/it]
 44%|████▍     | 10689/24420 [4:26:53<5:40:29,  1.49s/it]
 44%|████▍     | 10690/24420 [4:26:54<5:39:55,  1.49s/it]
                                                         
{'loss': 0.0929, 'learning_rate': 5e-05, 'epoch': 2.19}

 44%|████▍     | 10690/24420 [4:26:54<5:39:55,  1.49s/it]
 44%|████▍     | 10691/24420 [4:26:56<5:40:50,  1.49s/it]
 44%|████▍     | 10692/24420 [4:26:57<5:41:03,  1.49s/it]
 44%|████▍     | 10693/24420 [4:26:59<5:41:14,  1.49s/it]
 44%|████▍     | 10694/24420 [4:27:00<5:41:21,  1.49s/it]
 44%|████▍     | 10695/24420 [4:27:02<5:59:29,  1.57s/it]
                                                         
{'loss': 0.1286, 'learning_rate': 5e-05, 'epoch': 2.19}

 44%|████▍     | 10695/24420 [4:27:02<5:59:29,  1.57s/it]
 44%|████▍     | 10696/24420 [4:27:04<5:54:03,  1.55s/it]
 44%|████▍     | 10697/24420 [4:27:05<5:50:00,  1.53s/it]
 44%|████▍     | 10698/24420 [4:27:07<5:47:19,  1.52s/it]
 44%|████▍     | 10699/24420 [4:27:08<5:45:24,  1.51s/it]
 44%|████▍     | 10700/24420 [4:27:10<5:44:25,  1.51s/it]
                                                         
{'loss': 0.1061, 'learning_rate': 5e-05, 'epoch': 2.19}

 44%|████▍     | 10700/24420 [4:27:10<5:44:25,  1.51s/it]
 44%|████▍     | 10701/24420 [4:27:11<5:43:06,  1.50s/it]
 44%|████▍     | 10702/24420 [4:27:13<5:41:54,  1.50s/it]
 44%|████▍     | 10703/24420 [4:27:14<5:42:04,  1.50s/it]
 44%|████▍     | 10704/24420 [4:27:16<5:41:29,  1.49s/it]
 44%|████▍     | 10705/24420 [4:27:17<5:41:57,  1.50s/it]
                                                         
{'loss': 0.1057, 'learning_rate': 5e-05, 'epoch': 2.19}

 44%|████▍     | 10705/24420 [4:27:17<5:41:57,  1.50s/it]
 44%|████▍     | 10706/24420 [4:27:19<5:42:19,  1.50s/it]
 44%|████▍     | 10707/24420 [4:27:20<5:41:49,  1.50s/it]
 44%|████▍     | 10708/24420 [4:27:22<5:40:43,  1.49s/it]
 44%|████▍     | 10709/24420 [4:27:23<5:40:45,  1.49s/it]
 44%|████▍     | 10710/24420 [4:27:25<5:40:47,  1.49s/it]
                                                         
{'loss': 0.077, 'learning_rate': 5e-05, 'epoch': 2.19}

 44%|████▍     | 10710/24420 [4:27:25<5:40:47,  1.49s/it]
 44%|████▍     | 10711/24420 [4:27:26<5:41:06,  1.49s/it]
 44%|████▍     | 10712/24420 [4:27:28<5:40:57,  1.49s/it]
 44%|████▍     | 10713/24420 [4:27:29<5:41:12,  1.49s/it]
 44%|████▍     | 10714/24420 [4:27:31<5:41:21,  1.49s/it]
 44%|████▍     | 10715/24420 [4:27:32<5:41:05,  1.49s/it]
                                                         
{'loss': 0.0943, 'learning_rate': 5e-05, 'epoch': 2.19}

 44%|████▍     | 10715/24420 [4:27:32<5:41:05,  1.49s/it]
 44%|████▍     | 10716/24420 [4:27:34<5:40:50,  1.49s/it]
 44%|████▍     | 10717/24420 [4:27:35<5:40:24,  1.49s/it]
 44%|████▍     | 10718/24420 [4:27:37<5:41:10,  1.49s/it]
 44%|████▍     | 10719/24420 [4:27:38<5:41:51,  1.50s/it]
 44%|████▍     | 10720/24420 [4:27:39<5:41:10,  1.49s/it]
                                                         
{'loss': 0.1008, 'learning_rate': 5e-05, 'epoch': 2.19}

 44%|████▍     | 10720/24420 [4:27:39<5:41:10,  1.49s/it]
 44%|████▍     | 10721/24420 [4:27:41<5:40:47,  1.49s/it]
 44%|████▍     | 10722/24420 [4:27:42<5:40:34,  1.49s/it]
 44%|████▍     | 10723/24420 [4:27:44<5:40:49,  1.49s/it]
 44%|████▍     | 10724/24420 [4:27:45<5:40:50,  1.49s/it]
 44%|████▍     | 10725/24420 [4:27:47<5:40:10,  1.49s/it]
                                                         
{'loss': 0.0978, 'learning_rate': 5e-05, 'epoch': 2.2}

 44%|████▍     | 10725/24420 [4:27:47<5:40:10,  1.49s/it]
 44%|████▍     | 10726/24420 [4:27:48<5:40:22,  1.49s/it]
 44%|████▍     | 10727/24420 [4:27:50<5:39:59,  1.49s/it]
 44%|████▍     | 10728/24420 [4:27:51<5:39:42,  1.49s/it]
 44%|████▍     | 10729/24420 [4:27:53<5:39:14,  1.49s/it]
 44%|████▍     | 10730/24420 [4:27:54<5:39:47,  1.49s/it]
                                                         
{'loss': 0.095, 'learning_rate': 5e-05, 'epoch': 2.2}

 44%|████▍     | 10730/24420 [4:27:54<5:39:47,  1.49s/it]
 44%|████▍     | 10731/24420 [4:27:56<5:41:33,  1.50s/it]
 44%|████▍     | 10732/24420 [4:27:57<5:40:40,  1.49s/it]
 44%|████▍     | 10733/24420 [4:27:59<5:40:19,  1.49s/it]
 44%|████▍     | 10734/24420 [4:28:00<5:39:44,  1.49s/it]
 44%|████▍     | 10735/24420 [4:28:02<5:42:44,  1.50s/it]
                                                         
{'loss': 0.0914, 'learning_rate': 5e-05, 'epoch': 2.2}

 44%|████▍     | 10735/24420 [4:28:02<5:42:44,  1.50s/it]
 44%|████▍     | 10736/24420 [4:28:03<5:42:13,  1.50s/it]
 44%|████▍     | 10737/24420 [4:28:05<5:43:34,  1.51s/it]
 44%|████▍     | 10738/24420 [4:28:06<5:45:18,  1.51s/it]
 44%|████▍     | 10739/24420 [4:28:08<5:45:38,  1.52s/it]
 44%|████▍     | 10740/24420 [4:28:09<5:43:40,  1.51s/it]
                                                         
{'loss': 0.1034, 'learning_rate': 5e-05, 'epoch': 2.2}

 44%|████▍     | 10740/24420 [4:28:09<5:43:40,  1.51s/it]
 44%|████▍     | 10741/24420 [4:28:11<5:41:44,  1.50s/it]
 44%|████▍     | 10742/24420 [4:28:12<5:41:26,  1.50s/it]
 44%|████▍     | 10743/24420 [4:28:14<5:40:48,  1.50s/it]
 44%|████▍     | 10744/24420 [4:28:15<5:40:14,  1.49s/it]
 44%|████▍     | 10745/24420 [4:28:17<5:40:34,  1.49s/it]
                                                         
{'loss': 0.0972, 'learning_rate': 5e-05, 'epoch': 2.2}

 44%|████▍     | 10745/24420 [4:28:17<5:40:34,  1.49s/it]
 44%|████▍     | 10746/24420 [4:28:18<5:40:42,  1.49s/it]
 44%|████▍     | 10747/24420 [4:28:20<5:40:35,  1.49s/it]
 44%|████▍     | 10748/24420 [4:28:21<5:39:54,  1.49s/it]
 44%|████▍     | 10749/24420 [4:28:23<5:40:30,  1.49s/it]
 44%|████▍     | 10750/24420 [4:28:24<5:39:56,  1.49s/it]
                                                         
{'loss': 0.078, 'learning_rate': 5e-05, 'epoch': 2.2}

 44%|████▍     | 10750/24420 [4:28:24<5:39:56,  1.49s/it]
 44%|████▍     | 10751/24420 [4:28:26<5:40:17,  1.49s/it]
 44%|████▍     | 10752/24420 [4:28:27<5:39:56,  1.49s/it]
 44%|████▍     | 10753/24420 [4:28:29<5:39:00,  1.49s/it]
 44%|████▍     | 10754/24420 [4:28:30<5:39:51,  1.49s/it]
 44%|████▍     | 10755/24420 [4:28:32<5:39:26,  1.49s/it]
                                                         
{'loss': 0.0887, 'learning_rate': 5e-05, 'epoch': 2.2}

 44%|████▍     | 10755/24420 [4:28:32<5:39:26,  1.49s/it]
 44%|████▍     | 10756/24420 [4:28:33<5:39:05,  1.49s/it]
 44%|████▍     | 10757/24420 [4:28:35<5:38:45,  1.49s/it]
 44%|████▍     | 10758/24420 [4:28:36<5:38:03,  1.48s/it]
 44%|████▍     | 10759/24420 [4:28:38<5:38:35,  1.49s/it]
 44%|████▍     | 10760/24420 [4:28:39<5:38:36,  1.49s/it]
                                                         
{'loss': 0.0957, 'learning_rate': 5e-05, 'epoch': 2.2}

 44%|████▍     | 10760/24420 [4:28:39<5:38:36,  1.49s/it]
 44%|████▍     | 10761/24420 [4:28:41<5:39:47,  1.49s/it]
 44%|████▍     | 10762/24420 [4:28:42<5:40:12,  1.49s/it]
 44%|████▍     | 10763/24420 [4:28:44<5:39:27,  1.49s/it]
 44%|████▍     | 10764/24420 [4:28:45<5:39:21,  1.49s/it]
 44%|████▍     | 10765/24420 [4:28:47<5:39:56,  1.49s/it]
                                                         
{'loss': 0.0868, 'learning_rate': 5e-05, 'epoch': 2.2}

 44%|████▍     | 10765/24420 [4:28:47<5:39:56,  1.49s/it]
 44%|████▍     | 10766/24420 [4:28:48<5:39:51,  1.49s/it]
 44%|████▍     | 10767/24420 [4:28:50<5:39:41,  1.49s/it]
 44%|████▍     | 10768/24420 [4:28:51<5:39:21,  1.49s/it]
 44%|████▍     | 10769/24420 [4:28:53<5:39:28,  1.49s/it]
 44%|████▍     | 10770/24420 [4:28:54<5:38:53,  1.49s/it]
                                                         
{'loss': 0.0939, 'learning_rate': 5e-05, 'epoch': 2.21}

 44%|████▍     | 10770/24420 [4:28:54<5:38:53,  1.49s/it]
 44%|████▍     | 10771/24420 [4:28:56<5:38:40,  1.49s/it]
 44%|████▍     | 10772/24420 [4:28:57<5:38:11,  1.49s/it]
 44%|████▍     | 10773/24420 [4:28:59<5:37:53,  1.49s/it]
 44%|████▍     | 10774/24420 [4:29:00<5:37:37,  1.48s/it]
 44%|████▍     | 10775/24420 [4:29:02<5:38:11,  1.49s/it]
                                                         
{'loss': 0.0794, 'learning_rate': 5e-05, 'epoch': 2.21}

 44%|████▍     | 10775/24420 [4:29:02<5:38:11,  1.49s/it]
 44%|████▍     | 10776/24420 [4:29:03<5:38:02,  1.49s/it]
 44%|████▍     | 10777/24420 [4:29:05<5:37:58,  1.49s/it]
 44%|████▍     | 10778/24420 [4:29:06<5:37:47,  1.49s/it]
 44%|████▍     | 10779/24420 [4:29:08<5:37:44,  1.49s/it]
 44%|████▍     | 10780/24420 [4:29:09<5:37:14,  1.48s/it]
                                                         
{'loss': 0.084, 'learning_rate': 5e-05, 'epoch': 2.21}

 44%|████▍     | 10780/24420 [4:29:09<5:37:14,  1.48s/it]
 44%|████▍     | 10781/24420 [4:29:11<5:37:12,  1.48s/it]
 44%|████▍     | 10782/24420 [4:29:12<5:37:13,  1.48s/it]
 44%|████▍     | 10783/24420 [4:29:13<5:37:43,  1.49s/it]
 44%|████▍     | 10784/24420 [4:29:15<5:37:45,  1.49s/it]
 44%|████▍     | 10785/24420 [4:29:16<5:38:24,  1.49s/it]
                                                         
{'loss': 0.0809, 'learning_rate': 5e-05, 'epoch': 2.21}

 44%|████▍     | 10785/24420 [4:29:16<5:38:24,  1.49s/it]
 44%|████▍     | 10786/24420 [4:29:18<5:39:08,  1.49s/it]
 44%|████▍     | 10787/24420 [4:29:19<5:39:55,  1.50s/it]
 44%|████▍     | 10788/24420 [4:29:21<5:39:25,  1.49s/it]
 44%|████▍     | 10789/24420 [4:29:22<5:38:54,  1.49s/it]
 44%|████▍     | 10790/24420 [4:29:24<5:38:10,  1.49s/it]
                                                         
{'loss': 0.0971, 'learning_rate': 5e-05, 'epoch': 2.21}

 44%|████▍     | 10790/24420 [4:29:24<5:38:10,  1.49s/it]
 44%|████▍     | 10791/24420 [4:29:25<5:38:02,  1.49s/it]
 44%|████▍     | 10792/24420 [4:29:27<5:37:40,  1.49s/it]
 44%|████▍     | 10793/24420 [4:29:28<5:37:41,  1.49s/it]
 44%|████▍     | 10794/24420 [4:29:30<5:37:00,  1.48s/it]
 44%|████▍     | 10795/24420 [4:29:31<5:36:59,  1.48s/it]
                                                         
{'loss': 0.0857, 'learning_rate': 5e-05, 'epoch': 2.21}

 44%|████▍     | 10795/24420 [4:29:31<5:36:59,  1.48s/it]
 44%|████▍     | 10796/24420 [4:29:33<5:37:17,  1.49s/it]
 44%|████▍     | 10797/24420 [4:29:34<5:37:14,  1.49s/it]
 44%|████▍     | 10798/24420 [4:29:36<5:37:18,  1.49s/it]
 44%|████▍     | 10799/24420 [4:29:37<5:37:49,  1.49s/it]
 44%|████▍     | 10800/24420 [4:29:39<5:37:48,  1.49s/it]
                                                         
{'loss': 0.0978, 'learning_rate': 5e-05, 'epoch': 2.21}

 44%|████▍     | 10800/24420 [4:29:39<5:37:48,  1.49s/it]
 44%|████▍     | 10801/24420 [4:29:43<9:06:20,  2.41s/it]
 44%|████▍     | 10802/24420 [4:29:45<8:03:07,  2.13s/it]
 44%|████▍     | 10803/24420 [4:29:46<7:19:30,  1.94s/it]
 44%|████▍     | 10804/24420 [4:29:48<6:48:50,  1.80s/it]
 44%|████▍     | 10805/24420 [4:29:49<6:26:46,  1.70s/it]
                                                         
{'loss': 0.0886, 'learning_rate': 5e-05, 'epoch': 2.21}

 44%|████▍     | 10805/24420 [4:29:49<6:26:46,  1.70s/it]
 44%|████▍     | 10806/24420 [4:29:51<6:13:07,  1.64s/it]
 44%|████▍     | 10807/24420 [4:29:52<6:02:08,  1.60s/it]
 44%|████▍     | 10808/24420 [4:29:54<5:54:42,  1.56s/it]
 44%|████▍     | 10809/24420 [4:29:55<5:49:21,  1.54s/it]
 44%|████▍     | 10810/24420 [4:29:57<5:46:20,  1.53s/it]
                                                         
{'loss': 0.1128, 'learning_rate': 5e-05, 'epoch': 2.21}

 44%|████▍     | 10810/24420 [4:29:57<5:46:20,  1.53s/it]
 44%|████▍     | 10811/24420 [4:29:58<5:43:47,  1.52s/it]
 44%|████▍     | 10812/24420 [4:30:00<5:42:03,  1.51s/it]
 44%|████▍     | 10813/24420 [4:30:01<5:40:57,  1.50s/it]
 44%|████▍     | 10814/24420 [4:30:03<5:39:37,  1.50s/it]
 44%|████▍     | 10815/24420 [4:30:04<5:38:58,  1.49s/it]
                                                         
{'loss': 0.1035, 'learning_rate': 5e-05, 'epoch': 2.21}

 44%|████▍     | 10815/24420 [4:30:04<5:38:58,  1.49s/it]
 44%|████▍     | 10816/24420 [4:30:06<5:38:19,  1.49s/it]
 44%|████▍     | 10817/24420 [4:30:07<5:37:45,  1.49s/it]
 44%|████▍     | 10818/24420 [4:30:09<5:38:09,  1.49s/it]
 44%|████▍     | 10819/24420 [4:30:10<5:38:06,  1.49s/it]
 44%|████▍     | 10820/24420 [4:30:12<5:37:45,  1.49s/it]
                                                         
{'loss': 0.0845, 'learning_rate': 5e-05, 'epoch': 2.22}

 44%|████▍     | 10820/24420 [4:30:12<5:37:45,  1.49s/it]
 44%|████▍     | 10821/24420 [4:30:13<5:37:19,  1.49s/it]
 44%|████▍     | 10822/24420 [4:30:15<5:37:29,  1.49s/it]
 44%|████▍     | 10823/24420 [4:30:16<5:37:54,  1.49s/it]
 44%|████▍     | 10824/24420 [4:30:18<5:37:40,  1.49s/it]
 44%|████▍     | 10825/24420 [4:30:19<5:36:42,  1.49s/it]
                                                         
{'loss': 0.1177, 'learning_rate': 5e-05, 'epoch': 2.22}

 44%|████▍     | 10825/24420 [4:30:19<5:36:42,  1.49s/it]
 44%|████▍     | 10826/24420 [4:30:21<5:36:42,  1.49s/it]
 44%|████▍     | 10827/24420 [4:30:22<5:37:14,  1.49s/it]
 44%|████▍     | 10828/24420 [4:30:24<5:36:46,  1.49s/it]
 44%|████▍     | 10829/24420 [4:30:25<5:36:27,  1.49s/it]
 44%|████▍     | 10830/24420 [4:30:26<5:36:35,  1.49s/it]
                                                         
{'loss': 0.1033, 'learning_rate': 5e-05, 'epoch': 2.22}

 44%|████▍     | 10830/24420 [4:30:26<5:36:35,  1.49s/it]
 44%|████▍     | 10831/24420 [4:30:28<5:36:32,  1.49s/it]
 44%|████▍     | 10832/24420 [4:30:29<5:36:55,  1.49s/it]
 44%|████▍     | 10833/24420 [4:30:31<5:36:58,  1.49s/it]
 44%|████▍     | 10834/24420 [4:30:32<5:37:11,  1.49s/it]
 44%|████▍     | 10835/24420 [4:30:34<5:37:02,  1.49s/it]
                                                         
{'loss': 0.0793, 'learning_rate': 5e-05, 'epoch': 2.22}

 44%|████▍     | 10835/24420 [4:30:34<5:37:02,  1.49s/it]
 44%|████▍     | 10836/24420 [4:30:35<5:36:45,  1.49s/it]
 44%|████▍     | 10837/24420 [4:30:37<5:36:18,  1.49s/it]
 44%|████▍     | 10838/24420 [4:30:38<5:36:03,  1.48s/it]
 44%|████▍     | 10839/24420 [4:30:40<5:35:57,  1.48s/it]
 44%|████▍     | 10840/24420 [4:30:41<5:36:41,  1.49s/it]
                                                         
{'loss': 0.101, 'learning_rate': 5e-05, 'epoch': 2.22}

 44%|████▍     | 10840/24420 [4:30:41<5:36:41,  1.49s/it]
 44%|████▍     | 10841/24420 [4:30:43<5:36:45,  1.49s/it]
 44%|████▍     | 10842/24420 [4:30:44<5:36:00,  1.48s/it]
 44%|████▍     | 10843/24420 [4:30:46<5:35:51,  1.48s/it]
 44%|████▍     | 10844/24420 [4:30:47<5:36:18,  1.49s/it]
 44%|████▍     | 10845/24420 [4:30:49<5:36:21,  1.49s/it]
                                                         
{'loss': 0.0955, 'learning_rate': 5e-05, 'epoch': 2.22}

 44%|████▍     | 10845/24420 [4:30:49<5:36:21,  1.49s/it]
 44%|████▍     | 10846/24420 [4:30:50<5:36:20,  1.49s/it]
 44%|████▍     | 10847/24420 [4:30:52<5:36:34,  1.49s/it]
 44%|████▍     | 10848/24420 [4:30:53<5:36:32,  1.49s/it]
 44%|████▍     | 10849/24420 [4:30:55<5:36:34,  1.49s/it]
 44%|████▍     | 10850/24420 [4:30:56<5:35:41,  1.48s/it]
                                                         
{'loss': 0.0742, 'learning_rate': 5e-05, 'epoch': 2.22}

 44%|████▍     | 10850/24420 [4:30:56<5:35:41,  1.48s/it]
 44%|████▍     | 10851/24420 [4:30:58<5:36:18,  1.49s/it]
 44%|████▍     | 10852/24420 [4:30:59<5:36:58,  1.49s/it]
 44%|████▍     | 10853/24420 [4:31:01<5:36:33,  1.49s/it]
 44%|████▍     | 10854/24420 [4:31:02<5:36:32,  1.49s/it]
 44%|████▍     | 10855/24420 [4:31:04<5:36:40,  1.49s/it]
                                                         
{'loss': 0.1064, 'learning_rate': 5e-05, 'epoch': 2.22}

 44%|████▍     | 10855/24420 [4:31:04<5:36:40,  1.49s/it]
 44%|████▍     | 10856/24420 [4:31:05<5:36:33,  1.49s/it]
 44%|████▍     | 10857/24420 [4:31:07<5:36:13,  1.49s/it]
 44%|████▍     | 10858/24420 [4:31:08<5:36:25,  1.49s/it]
 44%|████▍     | 10859/24420 [4:31:10<5:36:42,  1.49s/it]
 44%|████▍     | 10860/24420 [4:31:11<5:36:39,  1.49s/it]
                                                         
{'loss': 0.0872, 'learning_rate': 5e-05, 'epoch': 2.22}

 44%|████▍     | 10860/24420 [4:31:11<5:36:39,  1.49s/it]
 44%|████▍     | 10861/24420 [4:31:13<5:36:41,  1.49s/it]
 44%|████▍     | 10862/24420 [4:31:14<5:37:53,  1.50s/it]
 44%|████▍     | 10863/24420 [4:31:16<5:38:25,  1.50s/it]
 44%|████▍     | 10864/24420 [4:31:17<5:37:50,  1.50s/it]
 44%|████▍     | 10865/24420 [4:31:19<5:37:46,  1.50s/it]
                                                         
{'loss': 0.1136, 'learning_rate': 5e-05, 'epoch': 2.22}

 44%|████▍     | 10865/24420 [4:31:19<5:37:46,  1.50s/it]
 44%|████▍     | 10866/24420 [4:31:20<5:37:23,  1.49s/it]
 45%|████▍     | 10867/24420 [4:31:22<5:36:39,  1.49s/it]
 45%|████▍     | 10868/24420 [4:31:23<5:36:33,  1.49s/it]
 45%|████▍     | 10869/24420 [4:31:25<5:36:23,  1.49s/it]
 45%|████▍     | 10870/24420 [4:31:26<5:35:48,  1.49s/it]
                                                         
{'loss': 0.1169, 'learning_rate': 5e-05, 'epoch': 2.23}

 45%|████▍     | 10870/24420 [4:31:26<5:35:48,  1.49s/it]
 45%|████▍     | 10871/24420 [4:31:28<5:36:11,  1.49s/it]
 45%|████▍     | 10872/24420 [4:31:29<5:36:17,  1.49s/it]
 45%|████▍     | 10873/24420 [4:31:30<5:35:27,  1.49s/it]
 45%|████▍     | 10874/24420 [4:31:32<5:36:01,  1.49s/it]
 45%|████▍     | 10875/24420 [4:31:33<5:35:40,  1.49s/it]
                                                         
{'loss': 0.1082, 'learning_rate': 5e-05, 'epoch': 2.23}

 45%|████▍     | 10875/24420 [4:31:33<5:35:40,  1.49s/it]
 45%|████▍     | 10876/24420 [4:31:35<5:35:25,  1.49s/it]
 45%|████▍     | 10877/24420 [4:31:36<5:34:24,  1.48s/it]
 45%|████▍     | 10878/24420 [4:31:38<5:34:39,  1.48s/it]
 45%|████▍     | 10879/24420 [4:31:39<5:34:29,  1.48s/it]
 45%|████▍     | 10880/24420 [4:31:41<5:34:27,  1.48s/it]
                                                         
{'loss': 0.089, 'learning_rate': 5e-05, 'epoch': 2.23}

 45%|████▍     | 10880/24420 [4:31:41<5:34:27,  1.48s/it]
 45%|████▍     | 10881/24420 [4:31:42<5:35:14,  1.49s/it]
 45%|████▍     | 10882/24420 [4:31:44<5:35:10,  1.49s/it]
 45%|████▍     | 10883/24420 [4:31:45<5:38:20,  1.50s/it]
 45%|████▍     | 10884/24420 [4:31:47<5:37:47,  1.50s/it]
 45%|████▍     | 10885/24420 [4:31:48<5:37:59,  1.50s/it]
                                                         
{'loss': 0.1163, 'learning_rate': 5e-05, 'epoch': 2.23}

 45%|████▍     | 10885/24420 [4:31:48<5:37:59,  1.50s/it]
 45%|████▍     | 10886/24420 [4:31:50<5:37:42,  1.50s/it]
 45%|████▍     | 10887/24420 [4:31:51<5:36:35,  1.49s/it]
 45%|████▍     | 10888/24420 [4:31:53<5:35:52,  1.49s/it]
 45%|████▍     | 10889/24420 [4:31:54<5:42:48,  1.52s/it]
 45%|████▍     | 10890/24420 [4:31:56<5:41:21,  1.51s/it]
                                                         
{'loss': 0.1242, 'learning_rate': 5e-05, 'epoch': 2.23}

 45%|████▍     | 10890/24420 [4:31:56<5:41:21,  1.51s/it]
 45%|████▍     | 10891/24420 [4:31:57<5:39:42,  1.51s/it]
 45%|████▍     | 10892/24420 [4:31:59<5:38:31,  1.50s/it]
 45%|████▍     | 10893/24420 [4:32:00<5:37:59,  1.50s/it]
 45%|████▍     | 10894/24420 [4:32:02<5:37:39,  1.50s/it]
 45%|████▍     | 10895/24420 [4:32:03<5:37:44,  1.50s/it]
                                                         
{'loss': 0.1075, 'learning_rate': 5e-05, 'epoch': 2.23}

 45%|████▍     | 10895/24420 [4:32:03<5:37:44,  1.50s/it]
 45%|████▍     | 10896/24420 [4:32:05<5:37:23,  1.50s/it]
 45%|████▍     | 10897/24420 [4:32:06<5:36:30,  1.49s/it]
 45%|████▍     | 10898/24420 [4:32:08<5:35:52,  1.49s/it]
 45%|████▍     | 10899/24420 [4:32:09<5:35:27,  1.49s/it]
 45%|████▍     | 10900/24420 [4:32:11<5:34:25,  1.48s/it]
                                                         
{'loss': 0.0778, 'learning_rate': 5e-05, 'epoch': 2.23}

 45%|████▍     | 10900/24420 [4:32:11<5:34:25,  1.48s/it]
 45%|████▍     | 10901/24420 [4:32:12<5:35:11,  1.49s/it]
 45%|████▍     | 10902/24420 [4:32:14<5:35:05,  1.49s/it]
 45%|████▍     | 10903/24420 [4:32:15<5:35:58,  1.49s/it]
 45%|████▍     | 10904/24420 [4:32:17<5:35:47,  1.49s/it]
 45%|████▍     | 10905/24420 [4:32:18<5:36:05,  1.49s/it]
                                                         
{'loss': 0.0912, 'learning_rate': 5e-05, 'epoch': 2.23}

 45%|████▍     | 10905/24420 [4:32:18<5:36:05,  1.49s/it]
 45%|████▍     | 10906/24420 [4:32:20<5:36:06,  1.49s/it]
 45%|████▍     | 10907/24420 [4:32:21<5:37:31,  1.50s/it]
 45%|████▍     | 10908/24420 [4:32:23<5:37:14,  1.50s/it]
 45%|████▍     | 10909/24420 [4:32:24<5:36:54,  1.50s/it]
 45%|████▍     | 10910/24420 [4:32:26<5:36:03,  1.49s/it]
                                                         
{'loss': 0.1082, 'learning_rate': 5e-05, 'epoch': 2.23}

 45%|████▍     | 10910/24420 [4:32:26<5:36:03,  1.49s/it]
 45%|████▍     | 10911/24420 [4:32:27<5:36:20,  1.49s/it]
 45%|████▍     | 10912/24420 [4:32:29<5:36:23,  1.49s/it]
 45%|████▍     | 10913/24420 [4:32:30<5:36:00,  1.49s/it]
 45%|████▍     | 10914/24420 [4:32:32<5:35:06,  1.49s/it]
 45%|████▍     | 10915/24420 [4:32:33<5:35:17,  1.49s/it]
                                                         
{'loss': 0.0836, 'learning_rate': 5e-05, 'epoch': 2.23}

 45%|████▍     | 10915/24420 [4:32:33<5:35:17,  1.49s/it]
 45%|████▍     | 10916/24420 [4:32:35<5:34:56,  1.49s/it]
 45%|████▍     | 10917/24420 [4:32:36<5:35:40,  1.49s/it]
 45%|████▍     | 10918/24420 [4:32:38<5:35:16,  1.49s/it]
 45%|████▍     | 10919/24420 [4:32:39<5:36:10,  1.49s/it]
 45%|████▍     | 10920/24420 [4:32:41<5:35:20,  1.49s/it]
                                                         
{'loss': 0.1064, 'learning_rate': 5e-05, 'epoch': 2.24}

 45%|████▍     | 10920/24420 [4:32:41<5:35:20,  1.49s/it]
 45%|████▍     | 10921/24420 [4:32:42<5:34:41,  1.49s/it]
 45%|████▍     | 10922/24420 [4:32:44<5:34:20,  1.49s/it]
 45%|████▍     | 10923/24420 [4:32:45<5:34:22,  1.49s/it]
 45%|████▍     | 10924/24420 [4:32:47<5:34:17,  1.49s/it]
 45%|████▍     | 10925/24420 [4:32:48<5:34:33,  1.49s/it]
                                                         
{'loss': 0.1343, 'learning_rate': 5e-05, 'epoch': 2.24}

 45%|████▍     | 10925/24420 [4:32:48<5:34:33,  1.49s/it]
 45%|████▍     | 10926/24420 [4:32:50<5:34:44,  1.49s/it]
 45%|████▍     | 10927/24420 [4:32:51<5:34:47,  1.49s/it]
 45%|████▍     | 10928/24420 [4:32:53<5:34:28,  1.49s/it]
 45%|████▍     | 10929/24420 [4:32:54<5:33:42,  1.48s/it]
 45%|████▍     | 10930/24420 [4:32:56<5:33:57,  1.49s/it]
                                                         
{'loss': 0.0953, 'learning_rate': 5e-05, 'epoch': 2.24}

 45%|████▍     | 10930/24420 [4:32:56<5:33:57,  1.49s/it]
 45%|████▍     | 10931/24420 [4:32:57<5:34:02,  1.49s/it]
 45%|████▍     | 10932/24420 [4:32:58<5:33:56,  1.49s/it]
 45%|████▍     | 10933/24420 [4:33:00<5:34:29,  1.49s/it]
 45%|████▍     | 10934/24420 [4:33:01<5:34:13,  1.49s/it]
 45%|████▍     | 10935/24420 [4:33:03<5:34:24,  1.49s/it]
                                                         
{'loss': 0.08, 'learning_rate': 5e-05, 'epoch': 2.24}

 45%|████▍     | 10935/24420 [4:33:03<5:34:24,  1.49s/it]
 45%|████▍     | 10936/24420 [4:33:04<5:35:07,  1.49s/it]
 45%|████▍     | 10937/24420 [4:33:06<5:34:54,  1.49s/it]
 45%|████▍     | 10938/24420 [4:33:07<5:36:17,  1.50s/it]
 45%|████▍     | 10939/24420 [4:33:09<5:35:27,  1.49s/it]
 45%|████▍     | 10940/24420 [4:33:10<5:34:44,  1.49s/it]
                                                         
{'loss': 0.0864, 'learning_rate': 5e-05, 'epoch': 2.24}

 45%|████▍     | 10940/24420 [4:33:10<5:34:44,  1.49s/it]
 45%|████▍     | 10941/24420 [4:33:12<5:35:21,  1.49s/it]
 45%|████▍     | 10942/24420 [4:33:13<5:35:11,  1.49s/it]
 45%|████▍     | 10943/24420 [4:33:15<5:34:21,  1.49s/it]
 45%|████▍     | 10944/24420 [4:33:16<5:34:04,  1.49s/it]
 45%|████▍     | 10945/24420 [4:33:18<5:34:48,  1.49s/it]
                                                         
{'loss': 0.0881, 'learning_rate': 5e-05, 'epoch': 2.24}

 45%|████▍     | 10945/24420 [4:33:18<5:34:48,  1.49s/it]
 45%|████▍     | 10946/24420 [4:33:19<5:34:13,  1.49s/it]
 45%|████▍     | 10947/24420 [4:33:21<5:33:24,  1.48s/it]
 45%|████▍     | 10948/24420 [4:33:22<5:33:03,  1.48s/it]
 45%|████▍     | 10949/24420 [4:33:24<5:33:37,  1.49s/it]
 45%|████▍     | 10950/24420 [4:33:25<5:33:15,  1.48s/it]
                                                         
{'loss': 0.0905, 'learning_rate': 5e-05, 'epoch': 2.24}

 45%|████▍     | 10950/24420 [4:33:25<5:33:15,  1.48s/it]
 45%|████▍     | 10951/24420 [4:33:27<5:34:10,  1.49s/it]
 45%|████▍     | 10952/24420 [4:33:28<5:33:32,  1.49s/it]
 45%|████▍     | 10953/24420 [4:33:30<5:34:04,  1.49s/it]
 45%|████▍     | 10954/24420 [4:33:31<5:33:46,  1.49s/it]
 45%|████▍     | 10955/24420 [4:33:33<5:33:46,  1.49s/it]
                                                         
{'loss': 0.0911, 'learning_rate': 5e-05, 'epoch': 2.24}

 45%|████▍     | 10955/24420 [4:33:33<5:33:46,  1.49s/it]
 45%|████▍     | 10956/24420 [4:33:34<5:33:53,  1.49s/it]
 45%|████▍     | 10957/24420 [4:33:36<5:34:48,  1.49s/it]
 45%|████▍     | 10958/24420 [4:33:37<5:34:45,  1.49s/it]
 45%|████▍     | 10959/24420 [4:33:39<5:33:57,  1.49s/it]
 45%|████▍     | 10960/24420 [4:33:40<5:34:11,  1.49s/it]
                                                         
{'loss': 0.108, 'learning_rate': 5e-05, 'epoch': 2.24}

 45%|████▍     | 10960/24420 [4:33:40<5:34:11,  1.49s/it]
 45%|████▍     | 10961/24420 [4:33:42<5:34:38,  1.49s/it]
 45%|████▍     | 10962/24420 [4:33:43<5:34:25,  1.49s/it]
 45%|████▍     | 10963/24420 [4:33:45<5:34:41,  1.49s/it]
 45%|████▍     | 10964/24420 [4:33:46<5:34:09,  1.49s/it]
 45%|████▍     | 10965/24420 [4:33:48<5:33:52,  1.49s/it]
                                                         
{'loss': 0.0905, 'learning_rate': 5e-05, 'epoch': 2.25}

 45%|████▍     | 10965/24420 [4:33:48<5:33:52,  1.49s/it]
 45%|████▍     | 10966/24420 [4:33:49<5:34:17,  1.49s/it]
 45%|████▍     | 10967/24420 [4:33:51<5:33:53,  1.49s/it]
 45%|████▍     | 10968/24420 [4:33:52<5:33:45,  1.49s/it]
 45%|████▍     | 10969/24420 [4:33:54<5:34:26,  1.49s/it]
 45%|████▍     | 10970/24420 [4:33:55<5:34:31,  1.49s/it]
                                                         
{'loss': 0.0982, 'learning_rate': 5e-05, 'epoch': 2.25}

 45%|████▍     | 10970/24420 [4:33:55<5:34:31,  1.49s/it]
 45%|████▍     | 10971/24420 [4:33:57<5:35:02,  1.49s/it]
 45%|████▍     | 10972/24420 [4:33:58<5:34:46,  1.49s/it]
 45%|████▍     | 10973/24420 [4:34:00<5:34:14,  1.49s/it]
 45%|████▍     | 10974/24420 [4:34:01<5:34:36,  1.49s/it]
 45%|████▍     | 10975/24420 [4:34:03<5:34:36,  1.49s/it]
                                                         
{'loss': 0.0929, 'learning_rate': 5e-05, 'epoch': 2.25}

 45%|████▍     | 10975/24420 [4:34:03<5:34:36,  1.49s/it]
 45%|████▍     | 10976/24420 [4:34:04<5:34:51,  1.49s/it]
 45%|████▍     | 10977/24420 [4:34:06<5:35:08,  1.50s/it]
 45%|████▍     | 10978/24420 [4:34:07<5:34:50,  1.49s/it]
 45%|████▍     | 10979/24420 [4:34:09<5:34:39,  1.49s/it]
 45%|████▍     | 10980/24420 [4:34:10<5:34:42,  1.49s/it]
                                                         
{'loss': 0.1025, 'learning_rate': 5e-05, 'epoch': 2.25}

 45%|████▍     | 10980/24420 [4:34:10<5:34:42,  1.49s/it]
 45%|████▍     | 10981/24420 [4:34:12<5:35:03,  1.50s/it]
 45%|████▍     | 10982/24420 [4:34:13<5:35:21,  1.50s/it]
 45%|████▍     | 10983/24420 [4:34:15<5:34:26,  1.49s/it]
 45%|████▍     | 10984/24420 [4:34:16<5:33:32,  1.49s/it]
 45%|████▍     | 10985/24420 [4:34:18<5:33:15,  1.49s/it]
                                                         
{'loss': 0.0923, 'learning_rate': 5e-05, 'epoch': 2.25}

 45%|████▍     | 10985/24420 [4:34:18<5:33:15,  1.49s/it]
 45%|████▍     | 10986/24420 [4:34:19<5:33:12,  1.49s/it]
 45%|████▍     | 10987/24420 [4:34:20<5:32:53,  1.49s/it]
 45%|████▍     | 10988/24420 [4:34:22<5:33:37,  1.49s/it]
 45%|████▌     | 10989/24420 [4:34:23<5:33:08,  1.49s/it]
 45%|████▌     | 10990/24420 [4:34:25<5:33:30,  1.49s/it]
                                                         
{'loss': 0.0945, 'learning_rate': 5e-05, 'epoch': 2.25}

 45%|████▌     | 10990/24420 [4:34:25<5:33:30,  1.49s/it]
 45%|████▌     | 10991/24420 [4:34:26<5:33:34,  1.49s/it]
 45%|████▌     | 10992/24420 [4:34:28<5:33:31,  1.49s/it]
 45%|████▌     | 10993/24420 [4:34:29<5:33:42,  1.49s/it]
 45%|████▌     | 10994/24420 [4:34:31<5:33:07,  1.49s/it]
 45%|████▌     | 10995/24420 [4:34:32<5:32:23,  1.49s/it]
                                                         
{'loss': 0.1046, 'learning_rate': 5e-05, 'epoch': 2.25}

 45%|████▌     | 10995/24420 [4:34:32<5:32:23,  1.49s/it]
 45%|████▌     | 10996/24420 [4:34:34<5:32:43,  1.49s/it]
 45%|████▌     | 10997/24420 [4:34:35<5:32:50,  1.49s/it]
 45%|████▌     | 10998/24420 [4:34:37<5:32:52,  1.49s/it]
 45%|████▌     | 10999/24420 [4:34:38<5:32:51,  1.49s/it]
 45%|████▌     | 11000/24420 [4:34:40<5:32:36,  1.49s/it]
                                                         
{'loss': 0.0996, 'learning_rate': 5e-05, 'epoch': 2.25}

 45%|████▌     | 11000/24420 [4:34:40<5:32:36,  1.49s/it]
 45%|████▌     | 11001/24420 [4:34:44<9:01:21,  2.42s/it]
 45%|████▌     | 11002/24420 [4:34:46<7:58:43,  2.14s/it]
 45%|████▌     | 11003/24420 [4:34:47<7:14:02,  1.94s/it]
 45%|████▌     | 11004/24420 [4:34:49<6:43:19,  1.80s/it]
 45%|████▌     | 11005/24420 [4:34:50<6:21:33,  1.71s/it]
                                                         
{'loss': 0.0856, 'learning_rate': 5e-05, 'epoch': 2.25}

 45%|████▌     | 11005/24420 [4:34:50<6:21:33,  1.71s/it]
 45%|████▌     | 11006/24420 [4:34:52<6:06:29,  1.64s/it]
 45%|████▌     | 11007/24420 [4:34:53<5:56:23,  1.59s/it]
 45%|████▌     | 11008/24420 [4:34:55<5:49:25,  1.56s/it]
 45%|████▌     | 11009/24420 [4:34:56<5:45:56,  1.55s/it]
 45%|████▌     | 11010/24420 [4:34:58<5:41:33,  1.53s/it]
                                                         
{'loss': 0.097, 'learning_rate': 5e-05, 'epoch': 2.25}

 45%|████▌     | 11010/24420 [4:34:58<5:41:33,  1.53s/it]
 45%|████▌     | 11011/24420 [4:34:59<5:40:12,  1.52s/it]
 45%|████▌     | 11012/24420 [4:35:01<5:38:16,  1.51s/it]
 45%|████▌     | 11013/24420 [4:35:02<5:36:54,  1.51s/it]
 45%|████▌     | 11014/24420 [4:35:04<5:36:22,  1.51s/it]
 45%|████▌     | 11015/24420 [4:35:05<5:34:46,  1.50s/it]
                                                         
{'loss': 0.097, 'learning_rate': 5e-05, 'epoch': 2.26}

 45%|████▌     | 11015/24420 [4:35:05<5:34:46,  1.50s/it]
 45%|████▌     | 11016/24420 [4:35:07<5:34:03,  1.50s/it]
 45%|████▌     | 11017/24420 [4:35:08<5:33:16,  1.49s/it]
 45%|████▌     | 11018/24420 [4:35:10<5:32:49,  1.49s/it]
 45%|████▌     | 11019/24420 [4:35:11<5:33:14,  1.49s/it]
 45%|████▌     | 11020/24420 [4:35:13<5:32:52,  1.49s/it]
                                                         
{'loss': 0.0965, 'learning_rate': 5e-05, 'epoch': 2.26}

 45%|████▌     | 11020/24420 [4:35:13<5:32:52,  1.49s/it]
 45%|████▌     | 11021/24420 [4:35:14<5:33:56,  1.50s/it]
 45%|████▌     | 11022/24420 [4:35:16<5:33:20,  1.49s/it]
 45%|████▌     | 11023/24420 [4:35:17<5:32:49,  1.49s/it]
 45%|████▌     | 11024/24420 [4:35:19<5:32:54,  1.49s/it]
 45%|████▌     | 11025/24420 [4:35:20<5:32:44,  1.49s/it]
                                                         
{'loss': 0.0912, 'learning_rate': 5e-05, 'epoch': 2.26}

 45%|████▌     | 11025/24420 [4:35:20<5:32:44,  1.49s/it]
 45%|████▌     | 11026/24420 [4:35:22<5:32:11,  1.49s/it]
 45%|████▌     | 11027/24420 [4:35:23<5:32:10,  1.49s/it]
 45%|████▌     | 11028/24420 [4:35:25<5:32:26,  1.49s/it]
 45%|████▌     | 11029/24420 [4:35:26<5:32:26,  1.49s/it]
 45%|████▌     | 11030/24420 [4:35:28<5:32:15,  1.49s/it]
                                                         
{'loss': 0.0848, 'learning_rate': 5e-05, 'epoch': 2.26}

 45%|████▌     | 11030/24420 [4:35:28<5:32:15,  1.49s/it]
 45%|████▌     | 11031/24420 [4:35:29<5:31:54,  1.49s/it]
 45%|████▌     | 11032/24420 [4:35:31<5:31:38,  1.49s/it]
 45%|████▌     | 11033/24420 [4:35:32<5:32:41,  1.49s/it]
 45%|████▌     | 11034/24420 [4:35:34<5:32:20,  1.49s/it]
 45%|████▌     | 11035/24420 [4:35:35<5:31:16,  1.48s/it]
                                                         
{'loss': 0.0907, 'learning_rate': 5e-05, 'epoch': 2.26}

 45%|████▌     | 11035/24420 [4:35:35<5:31:16,  1.48s/it]
 45%|████▌     | 11036/24420 [4:35:37<5:31:11,  1.48s/it]
 45%|████▌     | 11037/24420 [4:35:38<5:31:16,  1.49s/it]
 45%|████▌     | 11038/24420 [4:35:40<5:30:59,  1.48s/it]
 45%|████▌     | 11039/24420 [4:35:41<5:31:13,  1.49s/it]
 45%|████▌     | 11040/24420 [4:35:42<5:31:05,  1.48s/it]
                                                         
{'loss': 0.0912, 'learning_rate': 5e-05, 'epoch': 2.26}

 45%|████▌     | 11040/24420 [4:35:42<5:31:05,  1.48s/it]
 45%|████▌     | 11041/24420 [4:35:44<5:31:15,  1.49s/it]
 45%|████▌     | 11042/24420 [4:35:45<5:31:09,  1.49s/it]
 45%|████▌     | 11043/24420 [4:35:47<5:31:12,  1.49s/it]
 45%|████▌     | 11044/24420 [4:35:48<5:31:21,  1.49s/it]
 45%|████▌     | 11045/24420 [4:35:50<5:30:22,  1.48s/it]
                                                         
{'loss': 0.0986, 'learning_rate': 5e-05, 'epoch': 2.26}

 45%|████▌     | 11045/24420 [4:35:50<5:30:22,  1.48s/it]
 45%|████▌     | 11046/24420 [4:35:51<5:31:04,  1.49s/it]
 45%|████▌     | 11047/24420 [4:35:53<5:30:56,  1.48s/it]
 45%|████▌     | 11048/24420 [4:35:54<5:31:03,  1.49s/it]
 45%|████▌     | 11049/24420 [4:35:56<5:31:23,  1.49s/it]
 45%|████▌     | 11050/24420 [4:35:57<5:30:54,  1.49s/it]
                                                         
{'loss': 0.0921, 'learning_rate': 5e-05, 'epoch': 2.26}

 45%|████▌     | 11050/24420 [4:35:57<5:30:54,  1.49s/it]
 45%|████▌     | 11051/24420 [4:35:59<5:31:21,  1.49s/it]
 45%|████▌     | 11052/24420 [4:36:00<5:31:36,  1.49s/it]
 45%|████▌     | 11053/24420 [4:36:02<5:31:24,  1.49s/it]
 45%|████▌     | 11054/24420 [4:36:03<5:32:00,  1.49s/it]
 45%|████▌     | 11055/24420 [4:36:05<5:31:37,  1.49s/it]
                                                         
{'loss': 0.0807, 'learning_rate': 5e-05, 'epoch': 2.26}

 45%|████▌     | 11055/24420 [4:36:05<5:31:37,  1.49s/it]
 45%|████▌     | 11056/24420 [4:36:06<5:31:50,  1.49s/it]
 45%|████▌     | 11057/24420 [4:36:08<5:34:00,  1.50s/it]
 45%|████▌     | 11058/24420 [4:36:09<5:33:11,  1.50s/it]
 45%|████▌     | 11059/24420 [4:36:11<5:32:55,  1.50s/it]
 45%|████▌     | 11060/24420 [4:36:12<5:31:59,  1.49s/it]
                                                         
{'loss': 0.0861, 'learning_rate': 5e-05, 'epoch': 2.26}

 45%|████▌     | 11060/24420 [4:36:12<5:31:59,  1.49s/it]
 45%|████▌     | 11061/24420 [4:36:14<5:31:29,  1.49s/it]
 45%|████▌     | 11062/24420 [4:36:15<5:31:04,  1.49s/it]
 45%|████▌     | 11063/24420 [4:36:17<5:30:50,  1.49s/it]
 45%|████▌     | 11064/24420 [4:36:18<5:30:40,  1.49s/it]
 45%|████▌     | 11065/24420 [4:36:20<5:30:17,  1.48s/it]
                                                         
{'loss': 0.0884, 'learning_rate': 5e-05, 'epoch': 2.27}

 45%|████▌     | 11065/24420 [4:36:20<5:30:17,  1.48s/it]
 45%|████▌     | 11066/24420 [4:36:21<5:31:15,  1.49s/it]
 45%|████▌     | 11067/24420 [4:36:23<5:31:41,  1.49s/it]
 45%|████▌     | 11068/24420 [4:36:24<5:32:40,  1.49s/it]
 45%|████▌     | 11069/24420 [4:36:26<5:32:03,  1.49s/it]
 45%|████▌     | 11070/24420 [4:36:27<5:31:48,  1.49s/it]
                                                         
{'loss': 0.1006, 'learning_rate': 5e-05, 'epoch': 2.27}

 45%|████▌     | 11070/24420 [4:36:27<5:31:48,  1.49s/it]
 45%|████▌     | 11071/24420 [4:36:29<5:31:27,  1.49s/it]
 45%|████▌     | 11072/24420 [4:36:30<5:30:55,  1.49s/it]
 45%|████▌     | 11073/24420 [4:36:32<5:31:05,  1.49s/it]
 45%|████▌     | 11074/24420 [4:36:33<5:30:38,  1.49s/it]
 45%|████▌     | 11075/24420 [4:36:35<5:31:43,  1.49s/it]
                                                         
{'loss': 0.1001, 'learning_rate': 5e-05, 'epoch': 2.27}

 45%|████▌     | 11075/24420 [4:36:35<5:31:43,  1.49s/it]
 45%|████▌     | 11076/24420 [4:36:36<5:32:16,  1.49s/it]
 45%|████▌     | 11077/24420 [4:36:38<5:31:26,  1.49s/it]
 45%|████▌     | 11078/24420 [4:36:39<5:30:40,  1.49s/it]
 45%|████▌     | 11079/24420 [4:36:41<5:30:05,  1.48s/it]
 45%|████▌     | 11080/24420 [4:36:42<5:30:08,  1.48s/it]
                                                         
{'loss': 0.096, 'learning_rate': 5e-05, 'epoch': 2.27}

 45%|████▌     | 11080/24420 [4:36:42<5:30:08,  1.48s/it]
 45%|████▌     | 11081/24420 [4:36:44<5:30:24,  1.49s/it]
 45%|████▌     | 11082/24420 [4:36:45<5:30:24,  1.49s/it]
 45%|████▌     | 11083/24420 [4:36:46<5:30:47,  1.49s/it]
 45%|████▌     | 11084/24420 [4:36:48<5:30:32,  1.49s/it]
 45%|████▌     | 11085/24420 [4:36:49<5:30:43,  1.49s/it]
                                                         
{'loss': 0.0977, 'learning_rate': 5e-05, 'epoch': 2.27}

 45%|████▌     | 11085/24420 [4:36:49<5:30:43,  1.49s/it]
 45%|████▌     | 11086/24420 [4:36:51<5:30:09,  1.49s/it]
 45%|████▌     | 11087/24420 [4:36:52<5:29:33,  1.48s/it]
 45%|████▌     | 11088/24420 [4:36:54<5:29:18,  1.48s/it]
 45%|████▌     | 11089/24420 [4:36:55<5:29:29,  1.48s/it]
 45%|████▌     | 11090/24420 [4:36:57<5:29:25,  1.48s/it]
                                                         
{'loss': 0.072, 'learning_rate': 5e-05, 'epoch': 2.27}

 45%|████▌     | 11090/24420 [4:36:57<5:29:25,  1.48s/it]
 45%|████▌     | 11091/24420 [4:36:58<5:29:23,  1.48s/it]
 45%|████▌     | 11092/24420 [4:37:00<5:30:07,  1.49s/it]
 45%|████▌     | 11093/24420 [4:37:01<5:30:23,  1.49s/it]
 45%|████▌     | 11094/24420 [4:37:03<5:30:08,  1.49s/it]
 45%|████▌     | 11095/24420 [4:37:04<5:29:52,  1.49s/it]
                                                         
{'loss': 0.0928, 'learning_rate': 5e-05, 'epoch': 2.27}

 45%|████▌     | 11095/24420 [4:37:04<5:29:52,  1.49s/it]
 45%|████▌     | 11096/24420 [4:37:06<5:30:31,  1.49s/it]
 45%|████▌     | 11097/24420 [4:37:07<5:30:28,  1.49s/it]
 45%|████▌     | 11098/24420 [4:37:09<5:30:21,  1.49s/it]
 45%|████▌     | 11099/24420 [4:37:10<5:30:15,  1.49s/it]
 45%|████▌     | 11100/24420 [4:37:12<5:30:26,  1.49s/it]
                                                         
{'loss': 0.1163, 'learning_rate': 5e-05, 'epoch': 2.27}

 45%|████▌     | 11100/24420 [4:37:12<5:30:26,  1.49s/it]
 45%|████▌     | 11101/24420 [4:37:13<5:30:14,  1.49s/it]
 45%|████▌     | 11102/24420 [4:37:15<5:29:46,  1.49s/it]
 45%|████▌     | 11103/24420 [4:37:16<5:29:44,  1.49s/it]
 45%|████▌     | 11104/24420 [4:37:18<5:29:32,  1.48s/it]
 45%|████▌     | 11105/24420 [4:37:19<5:29:15,  1.48s/it]
                                                         
{'loss': 0.0838, 'learning_rate': 5e-05, 'epoch': 2.27}

 45%|████▌     | 11105/24420 [4:37:19<5:29:15,  1.48s/it]
 45%|████▌     | 11106/24420 [4:37:21<5:29:35,  1.49s/it]
 45%|████▌     | 11107/24420 [4:37:22<5:29:36,  1.49s/it]
 45%|████▌     | 11108/24420 [4:37:24<5:29:39,  1.49s/it]
 45%|████▌     | 11109/24420 [4:37:25<5:29:33,  1.49s/it]
 45%|████▌     | 11110/24420 [4:37:27<5:29:08,  1.48s/it]
                                                         
{'loss': 0.0702, 'learning_rate': 5e-05, 'epoch': 2.27}

 45%|████▌     | 11110/24420 [4:37:27<5:29:08,  1.48s/it]
 45%|████▌     | 11111/24420 [4:37:28<5:29:10,  1.48s/it]
 46%|████▌     | 11112/24420 [4:37:30<5:29:14,  1.48s/it]
 46%|████▌     | 11113/24420 [4:37:31<5:29:45,  1.49s/it]
 46%|████▌     | 11114/24420 [4:37:33<5:29:43,  1.49s/it]
 46%|████▌     | 11115/24420 [4:37:34<5:29:16,  1.48s/it]
                                                         
{'loss': 0.0973, 'learning_rate': 5e-05, 'epoch': 2.28}

 46%|████▌     | 11115/24420 [4:37:34<5:29:16,  1.48s/it]
 46%|████▌     | 11116/24420 [4:37:36<5:28:59,  1.48s/it]
 46%|████▌     | 11117/24420 [4:37:37<5:28:55,  1.48s/it]
 46%|████▌     | 11118/24420 [4:37:38<5:29:11,  1.48s/it]
 46%|████▌     | 11119/24420 [4:37:40<5:29:13,  1.49s/it]
 46%|████▌     | 11120/24420 [4:37:41<5:29:40,  1.49s/it]
                                                         
{'loss': 0.097, 'learning_rate': 5e-05, 'epoch': 2.28}

 46%|████▌     | 11120/24420 [4:37:41<5:29:40,  1.49s/it]
 46%|████▌     | 11121/24420 [4:37:43<5:29:22,  1.49s/it]
 46%|████▌     | 11122/24420 [4:37:44<5:29:26,  1.49s/it]
 46%|████▌     | 11123/24420 [4:37:46<5:29:22,  1.49s/it]
 46%|████▌     | 11124/24420 [4:37:47<5:29:34,  1.49s/it]
 46%|████▌     | 11125/24420 [4:37:49<5:29:35,  1.49s/it]
                                                         
{'loss': 0.0879, 'learning_rate': 5e-05, 'epoch': 2.28}

 46%|████▌     | 11125/24420 [4:37:49<5:29:35,  1.49s/it]
 46%|████▌     | 11126/24420 [4:37:50<5:30:37,  1.49s/it]
 46%|████▌     | 11127/24420 [4:37:52<5:30:48,  1.49s/it]
 46%|████▌     | 11128/24420 [4:37:53<5:30:21,  1.49s/it]
 46%|████▌     | 11129/24420 [4:37:55<5:29:36,  1.49s/it]
 46%|████▌     | 11130/24420 [4:37:56<5:30:13,  1.49s/it]
                                                         
{'loss': 0.081, 'learning_rate': 5e-05, 'epoch': 2.28}

 46%|████▌     | 11130/24420 [4:37:56<5:30:13,  1.49s/it]
 46%|████▌     | 11131/24420 [4:37:58<5:29:59,  1.49s/it]
 46%|████▌     | 11132/24420 [4:37:59<5:30:08,  1.49s/it]
 46%|████▌     | 11133/24420 [4:38:01<5:29:23,  1.49s/it]
 46%|████▌     | 11134/24420 [4:38:02<5:28:28,  1.48s/it]
 46%|████▌     | 11135/24420 [4:38:04<5:28:17,  1.48s/it]
                                                         
{'loss': 0.0839, 'learning_rate': 5e-05, 'epoch': 2.28}

 46%|████▌     | 11135/24420 [4:38:04<5:28:17,  1.48s/it]
 46%|████▌     | 11136/24420 [4:38:05<5:28:29,  1.48s/it]
 46%|████▌     | 11137/24420 [4:38:07<5:28:29,  1.48s/it]
 46%|████▌     | 11138/24420 [4:38:08<5:28:31,  1.48s/it]
 46%|████▌     | 11139/24420 [4:38:10<5:28:43,  1.49s/it]
 46%|████▌     | 11140/24420 [4:38:11<5:29:00,  1.49s/it]
                                                         
{'loss': 0.0886, 'learning_rate': 5e-05, 'epoch': 2.28}

 46%|████▌     | 11140/24420 [4:38:11<5:29:00,  1.49s/it]
 46%|████▌     | 11141/24420 [4:38:13<5:29:21,  1.49s/it]
 46%|████▌     | 11142/24420 [4:38:14<5:29:28,  1.49s/it]
 46%|████▌     | 11143/24420 [4:38:16<5:29:11,  1.49s/it]
 46%|████▌     | 11144/24420 [4:38:17<5:28:57,  1.49s/it]
 46%|████▌     | 11145/24420 [4:38:19<5:28:54,  1.49s/it]
                                                         
{'loss': 0.0954, 'learning_rate': 5e-05, 'epoch': 2.28}

 46%|████▌     | 11145/24420 [4:38:19<5:28:54,  1.49s/it]
 46%|████▌     | 11146/24420 [4:38:20<5:28:58,  1.49s/it]
 46%|████▌     | 11147/24420 [4:38:22<5:28:50,  1.49s/it]
 46%|████▌     | 11148/24420 [4:38:23<5:29:36,  1.49s/it]
 46%|████▌     | 11149/24420 [4:38:25<5:29:40,  1.49s/it]
 46%|████▌     | 11150/24420 [4:38:26<5:30:18,  1.49s/it]
                                                         
{'loss': 0.0795, 'learning_rate': 5e-05, 'epoch': 2.28}

 46%|████▌     | 11150/24420 [4:38:26<5:30:18,  1.49s/it]
 46%|████▌     | 11151/24420 [4:38:28<5:29:19,  1.49s/it]
 46%|████▌     | 11152/24420 [4:38:29<5:29:12,  1.49s/it]
 46%|████▌     | 11153/24420 [4:38:31<5:29:28,  1.49s/it]
 46%|████▌     | 11154/24420 [4:38:32<5:28:55,  1.49s/it]
 46%|████▌     | 11155/24420 [4:38:34<5:28:41,  1.49s/it]
                                                         
{'loss': 0.0801, 'learning_rate': 5e-05, 'epoch': 2.28}

 46%|████▌     | 11155/24420 [4:38:34<5:28:41,  1.49s/it]
 46%|████▌     | 11156/24420 [4:38:35<5:29:18,  1.49s/it]
 46%|████▌     | 11157/24420 [4:38:37<5:29:01,  1.49s/it]
 46%|████▌     | 11158/24420 [4:38:38<5:28:53,  1.49s/it]
 46%|████▌     | 11159/24420 [4:38:39<5:28:11,  1.48s/it]
 46%|████▌     | 11160/24420 [4:38:41<5:28:32,  1.49s/it]
                                                         
{'loss': 0.1086, 'learning_rate': 5e-05, 'epoch': 2.29}

 46%|████▌     | 11160/24420 [4:38:41<5:28:32,  1.49s/it]
 46%|████▌     | 11161/24420 [4:38:42<5:29:07,  1.49s/it]
 46%|████▌     | 11162/24420 [4:38:44<5:28:48,  1.49s/it]
 46%|████▌     | 11163/24420 [4:38:45<5:28:37,  1.49s/it]
 46%|████▌     | 11164/24420 [4:38:47<5:28:56,  1.49s/it]
 46%|████▌     | 11165/24420 [4:38:48<5:29:03,  1.49s/it]
                                                         
{'loss': 0.0881, 'learning_rate': 5e-05, 'epoch': 2.29}

 46%|████▌     | 11165/24420 [4:38:48<5:29:03,  1.49s/it]
 46%|████▌     | 11166/24420 [4:38:50<5:28:53,  1.49s/it]
 46%|████▌     | 11167/24420 [4:38:51<5:28:42,  1.49s/it]
 46%|████▌     | 11168/24420 [4:38:53<5:28:54,  1.49s/it]
 46%|████▌     | 11169/24420 [4:38:54<5:28:44,  1.49s/it]
 46%|████▌     | 11170/24420 [4:38:56<5:29:11,  1.49s/it]
                                                         
{'loss': 0.0993, 'learning_rate': 5e-05, 'epoch': 2.29}

 46%|████▌     | 11170/24420 [4:38:56<5:29:11,  1.49s/it]
 46%|████▌     | 11171/24420 [4:38:57<5:29:27,  1.49s/it]
 46%|████▌     | 11172/24420 [4:38:59<5:29:25,  1.49s/it]
 46%|████▌     | 11173/24420 [4:39:00<5:28:54,  1.49s/it]
 46%|████▌     | 11174/24420 [4:39:02<5:28:51,  1.49s/it]
 46%|████▌     | 11175/24420 [4:39:03<5:28:52,  1.49s/it]
                                                         
{'loss': 0.083, 'learning_rate': 5e-05, 'epoch': 2.29}

 46%|████▌     | 11175/24420 [4:39:03<5:28:52,  1.49s/it]
 46%|████▌     | 11176/24420 [4:39:05<5:28:27,  1.49s/it]
 46%|████▌     | 11177/24420 [4:39:06<5:29:07,  1.49s/it]
 46%|████▌     | 11178/24420 [4:39:08<5:29:22,  1.49s/it]
 46%|████▌     | 11179/24420 [4:39:09<5:29:30,  1.49s/it]
 46%|████▌     | 11180/24420 [4:39:11<5:28:59,  1.49s/it]
                                                         
{'loss': 0.0917, 'learning_rate': 5e-05, 'epoch': 2.29}

 46%|████▌     | 11180/24420 [4:39:11<5:28:59,  1.49s/it]
 46%|████▌     | 11181/24420 [4:39:12<5:29:02,  1.49s/it]
 46%|████▌     | 11182/24420 [4:39:14<5:29:23,  1.49s/it]
 46%|████▌     | 11183/24420 [4:39:15<5:28:55,  1.49s/it]
 46%|████▌     | 11184/24420 [4:39:17<5:28:35,  1.49s/it]
 46%|████▌     | 11185/24420 [4:39:18<5:28:06,  1.49s/it]
                                                         
{'loss': 0.0917, 'learning_rate': 5e-05, 'epoch': 2.29}

 46%|████▌     | 11185/24420 [4:39:18<5:28:06,  1.49s/it]
 46%|████▌     | 11186/24420 [4:39:20<5:28:08,  1.49s/it]
 46%|████▌     | 11187/24420 [4:39:21<5:27:44,  1.49s/it]
 46%|████▌     | 11188/24420 [4:39:23<5:27:31,  1.49s/it]
 46%|████▌     | 11189/24420 [4:39:24<5:27:44,  1.49s/it]
 46%|████▌     | 11190/24420 [4:39:26<5:27:48,  1.49s/it]
                                                         
{'loss': 0.1024, 'learning_rate': 5e-05, 'epoch': 2.29}

 46%|████▌     | 11190/24420 [4:39:26<5:27:48,  1.49s/it]
 46%|████▌     | 11191/24420 [4:39:27<5:28:37,  1.49s/it]
 46%|████▌     | 11192/24420 [4:39:29<5:28:09,  1.49s/it]
 46%|████▌     | 11193/24420 [4:39:30<5:27:41,  1.49s/it]
 46%|████▌     | 11194/24420 [4:39:32<5:41:08,  1.55s/it]
 46%|████▌     | 11195/24420 [4:39:33<5:36:37,  1.53s/it]
                                                         
{'loss': 0.0987, 'learning_rate': 5e-05, 'epoch': 2.29}

 46%|████▌     | 11195/24420 [4:39:33<5:36:37,  1.53s/it]
 46%|████▌     | 11196/24420 [4:39:35<5:33:39,  1.51s/it]
 46%|████▌     | 11197/24420 [4:39:36<5:32:07,  1.51s/it]
 46%|████▌     | 11198/24420 [4:39:38<5:30:39,  1.50s/it]
 46%|████▌     | 11199/24420 [4:39:39<5:29:55,  1.50s/it]
 46%|████▌     | 11200/24420 [4:39:41<5:29:23,  1.49s/it]
                                                         
{'loss': 0.0873, 'learning_rate': 5e-05, 'epoch': 2.29}

 46%|████▌     | 11200/24420 [4:39:41<5:29:23,  1.49s/it]
 46%|████▌     | 11201/24420 [4:39:46<9:08:10,  2.49s/it]
 46%|████▌     | 11202/24420 [4:39:47<8:01:41,  2.19s/it]
 46%|████▌     | 11203/24420 [4:39:48<7:15:20,  1.98s/it]
 46%|████▌     | 11204/24420 [4:39:50<6:43:13,  1.83s/it]
 46%|████▌     | 11205/24420 [4:39:51<6:21:23,  1.73s/it]
                                                         
{'loss': 0.0789, 'learning_rate': 5e-05, 'epoch': 2.29}

 46%|████▌     | 11205/24420 [4:39:51<6:21:23,  1.73s/it]
 46%|████▌     | 11206/24420 [4:39:53<6:04:52,  1.66s/it]
 46%|████▌     | 11207/24420 [4:39:54<5:53:26,  1.60s/it]
 46%|████▌     | 11208/24420 [4:39:56<5:45:51,  1.57s/it]
 46%|████▌     | 11209/24420 [4:39:57<5:40:39,  1.55s/it]
 46%|████▌     | 11210/24420 [4:39:59<5:36:19,  1.53s/it]
                                                         
{'loss': 0.077, 'learning_rate': 5e-05, 'epoch': 2.3}

 46%|████▌     | 11210/24420 [4:39:59<5:36:19,  1.53s/it]
 46%|████▌     | 11211/24420 [4:40:00<5:33:52,  1.52s/it]
 46%|████▌     | 11212/24420 [4:40:02<5:32:09,  1.51s/it]
 46%|████▌     | 11213/24420 [4:40:03<5:30:29,  1.50s/it]
 46%|████▌     | 11214/24420 [4:40:05<5:30:39,  1.50s/it]
 46%|████▌     | 11215/24420 [4:40:06<5:29:22,  1.50s/it]
                                                         
{'loss': 0.085, 'learning_rate': 5e-05, 'epoch': 2.3}

 46%|████▌     | 11215/24420 [4:40:06<5:29:22,  1.50s/it]
 46%|████▌     | 11216/24420 [4:40:08<5:28:42,  1.49s/it]
 46%|████▌     | 11217/24420 [4:40:09<5:28:10,  1.49s/it]
 46%|████▌     | 11218/24420 [4:40:11<5:28:22,  1.49s/it]
 46%|████▌     | 11219/24420 [4:40:12<5:27:56,  1.49s/it]
 46%|████▌     | 11220/24420 [4:40:14<5:28:26,  1.49s/it]
                                                         
{'loss': 0.0983, 'learning_rate': 5e-05, 'epoch': 2.3}

 46%|████▌     | 11220/24420 [4:40:14<5:28:26,  1.49s/it]
 46%|████▌     | 11221/24420 [4:40:15<5:28:12,  1.49s/it]
 46%|████▌     | 11222/24420 [4:40:17<5:27:46,  1.49s/it]
 46%|████▌     | 11223/24420 [4:40:18<5:27:03,  1.49s/it]
 46%|████▌     | 11224/24420 [4:40:20<5:30:07,  1.50s/it]
 46%|████▌     | 11225/24420 [4:40:21<5:29:26,  1.50s/it]
                                                         
{'loss': 0.116, 'learning_rate': 5e-05, 'epoch': 2.3}

 46%|████▌     | 11225/24420 [4:40:21<5:29:26,  1.50s/it]
 46%|████▌     | 11226/24420 [4:40:23<5:30:21,  1.50s/it]
 46%|████▌     | 11227/24420 [4:40:24<5:29:05,  1.50s/it]
 46%|████▌     | 11228/24420 [4:40:26<5:30:08,  1.50s/it]
 46%|████▌     | 11229/24420 [4:40:27<5:29:43,  1.50s/it]
 46%|████▌     | 11230/24420 [4:40:29<5:29:35,  1.50s/it]
                                                         
{'loss': 0.079, 'learning_rate': 5e-05, 'epoch': 2.3}

 46%|████▌     | 11230/24420 [4:40:29<5:29:35,  1.50s/it]
 46%|████▌     | 11231/24420 [4:40:30<5:29:18,  1.50s/it]
 46%|████▌     | 11232/24420 [4:40:32<5:28:00,  1.49s/it]
 46%|████▌     | 11233/24420 [4:40:33<5:29:08,  1.50s/it]
 46%|████▌     | 11234/24420 [4:40:35<5:28:41,  1.50s/it]
 46%|████▌     | 11235/24420 [4:40:36<5:27:40,  1.49s/it]
                                                         
{'loss': 0.0968, 'learning_rate': 5e-05, 'epoch': 2.3}

 46%|████▌     | 11235/24420 [4:40:36<5:27:40,  1.49s/it]
 46%|████▌     | 11236/24420 [4:40:38<5:28:03,  1.49s/it]
 46%|████▌     | 11237/24420 [4:40:39<5:28:07,  1.49s/it]
 46%|████▌     | 11238/24420 [4:40:41<5:27:32,  1.49s/it]
 46%|████▌     | 11239/24420 [4:40:42<5:28:09,  1.49s/it]
 46%|████▌     | 11240/24420 [4:40:44<5:28:36,  1.50s/it]
                                                         
{'loss': 0.0875, 'learning_rate': 5e-05, 'epoch': 2.3}

 46%|████▌     | 11240/24420 [4:40:44<5:28:36,  1.50s/it]
 46%|████▌     | 11241/24420 [4:40:45<5:28:10,  1.49s/it]
 46%|████▌     | 11242/24420 [4:40:47<5:28:02,  1.49s/it]
 46%|████▌     | 11243/24420 [4:40:48<5:27:05,  1.49s/it]
 46%|████▌     | 11244/24420 [4:40:50<5:27:07,  1.49s/it]
 46%|████▌     | 11245/24420 [4:40:51<5:26:38,  1.49s/it]
                                                         
{'loss': 0.0924, 'learning_rate': 5e-05, 'epoch': 2.3}

 46%|████▌     | 11245/24420 [4:40:51<5:26:38,  1.49s/it]
 46%|████▌     | 11246/24420 [4:40:53<5:26:23,  1.49s/it]
 46%|████▌     | 11247/24420 [4:40:54<5:25:51,  1.48s/it]
 46%|████▌     | 11248/24420 [4:40:56<5:26:03,  1.49s/it]
 46%|████▌     | 11249/24420 [4:40:57<5:26:07,  1.49s/it]
 46%|████▌     | 11250/24420 [4:40:59<5:26:16,  1.49s/it]
                                                         
{'loss': 0.0884, 'learning_rate': 5e-05, 'epoch': 2.3}

 46%|████▌     | 11250/24420 [4:40:59<5:26:16,  1.49s/it]
 46%|████▌     | 11251/24420 [4:41:00<5:27:40,  1.49s/it]
 46%|████▌     | 11252/24420 [4:41:02<5:27:23,  1.49s/it]
 46%|████▌     | 11253/24420 [4:41:03<5:27:16,  1.49s/it]
 46%|████▌     | 11254/24420 [4:41:05<5:26:47,  1.49s/it]
 46%|████▌     | 11255/24420 [4:41:06<5:26:31,  1.49s/it]
                                                         
{'loss': 0.103, 'learning_rate': 5e-05, 'epoch': 2.3}

 46%|████▌     | 11255/24420 [4:41:06<5:26:31,  1.49s/it]
 46%|████▌     | 11256/24420 [4:41:08<5:27:00,  1.49s/it]
 46%|████▌     | 11257/24420 [4:41:09<5:26:59,  1.49s/it]
 46%|████▌     | 11258/24420 [4:41:11<5:26:40,  1.49s/it]
 46%|████▌     | 11259/24420 [4:41:12<5:26:26,  1.49s/it]
 46%|████▌     | 11260/24420 [4:41:14<5:26:35,  1.49s/it]
                                                         
{'loss': 0.1051, 'learning_rate': 5e-05, 'epoch': 2.31}

 46%|████▌     | 11260/24420 [4:41:14<5:26:35,  1.49s/it]
 46%|████▌     | 11261/24420 [4:41:15<5:26:12,  1.49s/it]
 46%|████▌     | 11262/24420 [4:41:16<5:25:44,  1.49s/it]
 46%|████▌     | 11263/24420 [4:41:18<5:25:28,  1.48s/it]
 46%|████▌     | 11264/24420 [4:41:19<5:25:38,  1.49s/it]
 46%|████▌     | 11265/24420 [4:41:21<5:26:01,  1.49s/it]
                                                         
{'loss': 0.0839, 'learning_rate': 5e-05, 'epoch': 2.31}

 46%|████▌     | 11265/24420 [4:41:21<5:26:01,  1.49s/it]
 46%|████▌     | 11266/24420 [4:41:22<5:26:16,  1.49s/it]
 46%|████▌     | 11267/24420 [4:41:24<5:25:53,  1.49s/it]
 46%|████▌     | 11268/24420 [4:41:25<5:26:11,  1.49s/it]
 46%|████▌     | 11269/24420 [4:41:27<5:27:16,  1.49s/it]
 46%|████▌     | 11270/24420 [4:41:28<5:27:48,  1.50s/it]
                                                         
{'loss': 0.0877, 'learning_rate': 5e-05, 'epoch': 2.31}

 46%|████▌     | 11270/24420 [4:41:28<5:27:48,  1.50s/it]
 46%|████▌     | 11271/24420 [4:41:30<5:27:47,  1.50s/it]
 46%|████▌     | 11272/24420 [4:41:31<5:27:05,  1.49s/it]
 46%|████▌     | 11273/24420 [4:41:33<5:27:54,  1.50s/it]
 46%|████▌     | 11274/24420 [4:41:34<5:28:06,  1.50s/it]
 46%|████▌     | 11275/24420 [4:41:36<5:28:35,  1.50s/it]
                                                         
{'loss': 0.0977, 'learning_rate': 5e-05, 'epoch': 2.31}

 46%|████▌     | 11275/24420 [4:41:36<5:28:35,  1.50s/it]
 46%|████▌     | 11276/24420 [4:41:37<5:28:24,  1.50s/it]
 46%|████▌     | 11277/24420 [4:41:39<5:28:26,  1.50s/it]
 46%|████▌     | 11278/24420 [4:41:40<5:27:49,  1.50s/it]
 46%|████▌     | 11279/24420 [4:41:42<5:27:51,  1.50s/it]
 46%|████▌     | 11280/24420 [4:41:43<5:27:34,  1.50s/it]
                                                         
{'loss': 0.0914, 'learning_rate': 5e-05, 'epoch': 2.31}

 46%|████▌     | 11280/24420 [4:41:43<5:27:34,  1.50s/it]
 46%|████▌     | 11281/24420 [4:41:45<5:27:37,  1.50s/it]
 46%|████▌     | 11282/24420 [4:41:46<5:28:50,  1.50s/it]
 46%|████▌     | 11283/24420 [4:41:48<5:28:14,  1.50s/it]
 46%|████▌     | 11284/24420 [4:41:49<5:27:45,  1.50s/it]
 46%|████▌     | 11285/24420 [4:41:51<5:27:26,  1.50s/it]
                                                         
{'loss': 0.1071, 'learning_rate': 5e-05, 'epoch': 2.31}

 46%|████▌     | 11285/24420 [4:41:51<5:27:26,  1.50s/it]
 46%|████▌     | 11286/24420 [4:41:52<5:27:19,  1.50s/it]
 46%|████▌     | 11287/24420 [4:41:54<5:27:32,  1.50s/it]
 46%|████▌     | 11288/24420 [4:41:55<5:27:06,  1.49s/it]
 46%|████▌     | 11289/24420 [4:41:57<5:27:01,  1.49s/it]
 46%|████▌     | 11290/24420 [4:41:58<5:27:07,  1.49s/it]
                                                         
{'loss': 0.0927, 'learning_rate': 5e-05, 'epoch': 2.31}

 46%|████▌     | 11290/24420 [4:41:58<5:27:07,  1.49s/it]
 46%|████▌     | 11291/24420 [4:42:00<5:27:03,  1.49s/it]
 46%|████▌     | 11292/24420 [4:42:01<5:28:07,  1.50s/it]
 46%|████▌     | 11293/24420 [4:42:03<5:27:49,  1.50s/it]
 46%|████▌     | 11294/24420 [4:42:04<5:28:10,  1.50s/it]
 46%|████▋     | 11295/24420 [4:42:06<5:28:11,  1.50s/it]
                                                         
{'loss': 0.1356, 'learning_rate': 5e-05, 'epoch': 2.31}

 46%|████▋     | 11295/24420 [4:42:06<5:28:11,  1.50s/it]
 46%|████▋     | 11296/24420 [4:42:07<5:27:59,  1.50s/it]
 46%|████▋     | 11297/24420 [4:42:09<5:27:59,  1.50s/it]
 46%|████▋     | 11298/24420 [4:42:10<5:28:14,  1.50s/it]
 46%|████▋     | 11299/24420 [4:42:12<5:28:02,  1.50s/it]
 46%|████▋     | 11300/24420 [4:42:13<5:28:31,  1.50s/it]
                                                         
{'loss': 0.0904, 'learning_rate': 5e-05, 'epoch': 2.31}

 46%|████▋     | 11300/24420 [4:42:13<5:28:31,  1.50s/it]
 46%|████▋     | 11301/24420 [4:42:15<5:28:15,  1.50s/it]
 46%|████▋     | 11302/24420 [4:42:16<5:27:14,  1.50s/it]
 46%|████▋     | 11303/24420 [4:42:18<5:27:01,  1.50s/it]
 46%|████▋     | 11304/24420 [4:42:19<5:26:34,  1.49s/it]
 46%|████▋     | 11305/24420 [4:42:21<5:25:44,  1.49s/it]
                                                         
{'loss': 0.1098, 'learning_rate': 5e-05, 'epoch': 2.31}

 46%|████▋     | 11305/24420 [4:42:21<5:25:44,  1.49s/it]
 46%|████▋     | 11306/24420 [4:42:22<5:25:56,  1.49s/it]
 46%|████▋     | 11307/24420 [4:42:24<5:26:19,  1.49s/it]
 46%|████▋     | 11308/24420 [4:42:25<5:26:09,  1.49s/it]
 46%|████▋     | 11309/24420 [4:42:27<5:26:03,  1.49s/it]
 46%|████▋     | 11310/24420 [4:42:28<5:25:49,  1.49s/it]
                                                         
{'loss': 0.0837, 'learning_rate': 5e-05, 'epoch': 2.32}

 46%|████▋     | 11310/24420 [4:42:28<5:25:49,  1.49s/it]
 46%|████▋     | 11311/24420 [4:42:30<5:25:32,  1.49s/it]
 46%|████▋     | 11312/24420 [4:42:31<5:25:48,  1.49s/it]
 46%|████▋     | 11313/24420 [4:42:33<5:25:38,  1.49s/it]
 46%|████▋     | 11314/24420 [4:42:34<5:25:45,  1.49s/it]
 46%|████▋     | 11315/24420 [4:42:36<5:25:58,  1.49s/it]
                                                         
{'loss': 0.0913, 'learning_rate': 5e-05, 'epoch': 2.32}

 46%|████▋     | 11315/24420 [4:42:36<5:25:58,  1.49s/it]
 46%|████▋     | 11316/24420 [4:42:37<5:26:05,  1.49s/it]
 46%|████▋     | 11317/24420 [4:42:39<5:25:51,  1.49s/it]
 46%|████▋     | 11318/24420 [4:42:40<5:26:11,  1.49s/it]
 46%|████▋     | 11319/24420 [4:42:42<5:25:49,  1.49s/it]
 46%|████▋     | 11320/24420 [4:42:43<5:26:27,  1.50s/it]
                                                         
{'loss': 0.1008, 'learning_rate': 5e-05, 'epoch': 2.32}

 46%|████▋     | 11320/24420 [4:42:43<5:26:27,  1.50s/it]
 46%|████▋     | 11321/24420 [4:42:45<5:26:55,  1.50s/it]
 46%|████▋     | 11322/24420 [4:42:46<5:27:07,  1.50s/it]
 46%|████▋     | 11323/24420 [4:42:48<5:27:04,  1.50s/it]
 46%|████▋     | 11324/24420 [4:42:49<5:26:17,  1.49s/it]
 46%|████▋     | 11325/24420 [4:42:51<5:26:10,  1.49s/it]
                                                         
{'loss': 0.0964, 'learning_rate': 5e-05, 'epoch': 2.32}

 46%|████▋     | 11325/24420 [4:42:51<5:26:10,  1.49s/it]
 46%|████▋     | 11326/24420 [4:42:52<5:26:21,  1.50s/it]
 46%|████▋     | 11327/24420 [4:42:54<5:26:09,  1.49s/it]
 46%|████▋     | 11328/24420 [4:42:55<5:26:00,  1.49s/it]
 46%|████▋     | 11329/24420 [4:42:57<5:25:43,  1.49s/it]
 46%|████▋     | 11330/24420 [4:42:58<5:25:49,  1.49s/it]
                                                         
{'loss': 0.0899, 'learning_rate': 5e-05, 'epoch': 2.32}

 46%|████▋     | 11330/24420 [4:42:58<5:25:49,  1.49s/it]
 46%|████▋     | 11331/24420 [4:43:00<5:25:45,  1.49s/it]
 46%|████▋     | 11332/24420 [4:43:01<5:26:26,  1.50s/it]
 46%|████▋     | 11333/24420 [4:43:03<5:25:44,  1.49s/it]
 46%|████▋     | 11334/24420 [4:43:04<5:25:25,  1.49s/it]
 46%|████▋     | 11335/24420 [4:43:06<5:25:07,  1.49s/it]
                                                         
{'loss': 0.0829, 'learning_rate': 5e-05, 'epoch': 2.32}

 46%|████▋     | 11335/24420 [4:43:06<5:25:07,  1.49s/it]
 46%|████▋     | 11336/24420 [4:43:07<5:25:10,  1.49s/it]
 46%|████▋     | 11337/24420 [4:43:09<5:24:59,  1.49s/it]
 46%|████▋     | 11338/24420 [4:43:10<5:25:09,  1.49s/it]
 46%|████▋     | 11339/24420 [4:43:12<5:24:53,  1.49s/it]
 46%|████▋     | 11340/24420 [4:43:13<5:24:29,  1.49s/it]
                                                         
{'loss': 0.0969, 'learning_rate': 5e-05, 'epoch': 2.32}

 46%|████▋     | 11340/24420 [4:43:13<5:24:29,  1.49s/it]
 46%|████▋     | 11341/24420 [4:43:15<5:25:12,  1.49s/it]
 46%|████▋     | 11342/24420 [4:43:16<5:25:04,  1.49s/it]
 46%|████▋     | 11343/24420 [4:43:18<5:25:45,  1.49s/it]
 46%|████▋     | 11344/24420 [4:43:19<5:25:39,  1.49s/it]
 46%|████▋     | 11345/24420 [4:43:21<5:26:07,  1.50s/it]
                                                         
{'loss': 0.1049, 'learning_rate': 5e-05, 'epoch': 2.32}

 46%|████▋     | 11345/24420 [4:43:21<5:26:07,  1.50s/it]
 46%|████▋     | 11346/24420 [4:43:22<5:25:56,  1.50s/it]
 46%|████▋     | 11347/24420 [4:43:24<5:25:26,  1.49s/it]
 46%|████▋     | 11348/24420 [4:43:25<5:24:48,  1.49s/it]
 46%|████▋     | 11349/24420 [4:43:26<5:24:31,  1.49s/it]
 46%|████▋     | 11350/24420 [4:43:28<5:24:45,  1.49s/it]
                                                         
{'loss': 0.0883, 'learning_rate': 5e-05, 'epoch': 2.32}

 46%|████▋     | 11350/24420 [4:43:28<5:24:45,  1.49s/it]
 46%|████▋     | 11351/24420 [4:43:29<5:24:33,  1.49s/it]
 46%|████▋     | 11352/24420 [4:43:31<5:23:58,  1.49s/it]
 46%|████▋     | 11353/24420 [4:43:32<5:23:51,  1.49s/it]
 46%|████▋     | 11354/24420 [4:43:34<5:24:09,  1.49s/it]
 46%|████▋     | 11355/24420 [4:43:35<5:23:41,  1.49s/it]
                                                         
{'loss': 0.0895, 'learning_rate': 5e-05, 'epoch': 2.32}

 46%|████▋     | 11355/24420 [4:43:35<5:23:41,  1.49s/it]
 47%|████▋     | 11356/24420 [4:43:37<5:24:38,  1.49s/it]
 47%|████▋     | 11357/24420 [4:43:38<5:24:59,  1.49s/it]
 47%|████▋     | 11358/24420 [4:43:40<5:25:05,  1.49s/it]
 47%|████▋     | 11359/24420 [4:43:41<5:27:14,  1.50s/it]
 47%|████▋     | 11360/24420 [4:43:43<5:27:30,  1.50s/it]
                                                         
{'loss': 0.102, 'learning_rate': 5e-05, 'epoch': 2.33}

 47%|████▋     | 11360/24420 [4:43:43<5:27:30,  1.50s/it]
 47%|████▋     | 11361/24420 [4:43:44<5:27:47,  1.51s/it]
 47%|████▋     | 11362/24420 [4:43:46<5:27:21,  1.50s/it]
 47%|████▋     | 11363/24420 [4:43:47<5:26:34,  1.50s/it]
 47%|████▋     | 11364/24420 [4:43:49<5:25:59,  1.50s/it]
 47%|████▋     | 11365/24420 [4:43:50<5:25:43,  1.50s/it]
                                                         
{'loss': 0.0843, 'learning_rate': 5e-05, 'epoch': 2.33}

 47%|████▋     | 11365/24420 [4:43:50<5:25:43,  1.50s/it]
 47%|████▋     | 11366/24420 [4:43:52<5:26:06,  1.50s/it]
 47%|████▋     | 11367/24420 [4:43:53<5:26:35,  1.50s/it]
 47%|████▋     | 11368/24420 [4:43:55<5:26:56,  1.50s/it]
 47%|████▋     | 11369/24420 [4:43:56<5:26:06,  1.50s/it]
 47%|████▋     | 11370/24420 [4:43:58<5:25:21,  1.50s/it]
                                                         
{'loss': 0.0899, 'learning_rate': 5e-05, 'epoch': 2.33}

 47%|████▋     | 11370/24420 [4:43:58<5:25:21,  1.50s/it]
 47%|████▋     | 11371/24420 [4:43:59<5:25:36,  1.50s/it]
 47%|████▋     | 11372/24420 [4:44:01<5:25:59,  1.50s/it]
 47%|████▋     | 11373/24420 [4:44:02<5:26:12,  1.50s/it]
 47%|████▋     | 11374/24420 [4:44:04<5:25:39,  1.50s/it]
 47%|████▋     | 11375/24420 [4:44:05<5:24:57,  1.49s/it]
                                                         
{'loss': 0.1007, 'learning_rate': 5e-05, 'epoch': 2.33}

 47%|████▋     | 11375/24420 [4:44:05<5:24:57,  1.49s/it]
 47%|████▋     | 11376/24420 [4:44:07<5:25:02,  1.50s/it]
 47%|████▋     | 11377/24420 [4:44:08<5:24:33,  1.49s/it]
 47%|████▋     | 11378/24420 [4:44:10<5:24:25,  1.49s/it]
 47%|████▋     | 11379/24420 [4:44:11<5:24:23,  1.49s/it]
 47%|████▋     | 11380/24420 [4:44:13<5:26:22,  1.50s/it]
                                                         
{'loss': 0.0855, 'learning_rate': 5e-05, 'epoch': 2.33}

 47%|████▋     | 11380/24420 [4:44:13<5:26:22,  1.50s/it]
 47%|████▋     | 11381/24420 [4:44:14<5:25:29,  1.50s/it]
 47%|████▋     | 11382/24420 [4:44:16<5:24:51,  1.49s/it]
 47%|████▋     | 11383/24420 [4:44:17<5:24:46,  1.49s/it]
 47%|████▋     | 11384/24420 [4:44:19<5:24:55,  1.50s/it]
 47%|████▋     | 11385/24420 [4:44:20<5:25:00,  1.50s/it]
                                                         
{'loss': 0.0938, 'learning_rate': 5e-05, 'epoch': 2.33}

 47%|████▋     | 11385/24420 [4:44:20<5:25:00,  1.50s/it]
 47%|████▋     | 11386/24420 [4:44:22<5:24:29,  1.49s/it]
 47%|████▋     | 11387/24420 [4:44:23<5:23:28,  1.49s/it]
 47%|████▋     | 11388/24420 [4:44:25<5:23:13,  1.49s/it]
 47%|████▋     | 11389/24420 [4:44:26<5:23:32,  1.49s/it]
 47%|████▋     | 11390/24420 [4:44:28<5:23:45,  1.49s/it]
                                                         
{'loss': 0.0801, 'learning_rate': 5e-05, 'epoch': 2.33}

 47%|████▋     | 11390/24420 [4:44:28<5:23:45,  1.49s/it]
 47%|████▋     | 11391/24420 [4:44:29<5:23:43,  1.49s/it]
 47%|████▋     | 11392/24420 [4:44:31<5:23:44,  1.49s/it]
 47%|████▋     | 11393/24420 [4:44:32<5:23:30,  1.49s/it]
 47%|████▋     | 11394/24420 [4:44:34<5:23:04,  1.49s/it]
 47%|████▋     | 11395/24420 [4:44:35<5:22:59,  1.49s/it]
                                                         
{'loss': 0.0864, 'learning_rate': 5e-05, 'epoch': 2.33}

 47%|████▋     | 11395/24420 [4:44:35<5:22:59,  1.49s/it]
 47%|████▋     | 11396/24420 [4:44:37<5:23:33,  1.49s/it]
 47%|████▋     | 11397/24420 [4:44:38<5:23:17,  1.49s/it]
 47%|████▋     | 11398/24420 [4:44:40<5:23:19,  1.49s/it]
 47%|████▋     | 11399/24420 [4:44:41<5:22:47,  1.49s/it]
 47%|████▋     | 11400/24420 [4:44:43<5:22:39,  1.49s/it]
                                                         
{'loss': 0.0812, 'learning_rate': 5e-05, 'epoch': 2.33}

 47%|████▋     | 11400/24420 [4:44:43<5:22:39,  1.49s/it]
 47%|████▋     | 11401/24420 [4:44:48<9:12:18,  2.55s/it]
 47%|████▋     | 11402/24420 [4:44:49<8:03:10,  2.23s/it]
 47%|████▋     | 11403/24420 [4:44:51<7:15:39,  2.01s/it]
 47%|████▋     | 11404/24420 [4:44:52<6:41:54,  1.85s/it]
 47%|████▋     | 11405/24420 [4:44:54<6:18:36,  1.75s/it]
                                                         
{'loss': 0.0962, 'learning_rate': 5e-05, 'epoch': 2.34}

 47%|████▋     | 11405/24420 [4:44:54<6:18:36,  1.75s/it]
 47%|████▋     | 11406/24420 [4:44:55<6:01:47,  1.67s/it]
 47%|████▋     | 11407/24420 [4:44:57<5:50:21,  1.62s/it]
 47%|████▋     | 11408/24420 [4:44:58<5:42:26,  1.58s/it]
 47%|████▋     | 11409/24420 [4:45:00<5:36:40,  1.55s/it]
 47%|████▋     | 11410/24420 [4:45:01<5:32:44,  1.53s/it]
                                                         
{'loss': 0.1049, 'learning_rate': 5e-05, 'epoch': 2.34}

 47%|████▋     | 11410/24420 [4:45:01<5:32:44,  1.53s/it]
 47%|████▋     | 11411/24420 [4:45:03<5:30:03,  1.52s/it]
 47%|████▋     | 11412/24420 [4:45:04<5:28:23,  1.51s/it]
 47%|████▋     | 11413/24420 [4:45:06<5:26:34,  1.51s/it]
 47%|████▋     | 11414/24420 [4:45:07<5:25:06,  1.50s/it]
 47%|████▋     | 11415/24420 [4:45:09<5:24:21,  1.50s/it]
                                                         
{'loss': 0.1011, 'learning_rate': 5e-05, 'epoch': 2.34}

 47%|████▋     | 11415/24420 [4:45:09<5:24:21,  1.50s/it]
 47%|████▋     | 11416/24420 [4:45:10<5:23:58,  1.49s/it]
 47%|████▋     | 11417/24420 [4:45:12<5:23:42,  1.49s/it]
 47%|████▋     | 11418/24420 [4:45:13<5:23:30,  1.49s/it]
 47%|████▋     | 11419/24420 [4:45:15<5:23:25,  1.49s/it]
 47%|████▋     | 11420/24420 [4:45:16<5:24:41,  1.50s/it]
                                                         
{'loss': 0.0936, 'learning_rate': 5e-05, 'epoch': 2.34}

 47%|████▋     | 11420/24420 [4:45:16<5:24:41,  1.50s/it]
 47%|████▋     | 11421/24420 [4:45:18<5:24:42,  1.50s/it]
 47%|████▋     | 11422/24420 [4:45:19<5:24:14,  1.50s/it]
 47%|████▋     | 11423/24420 [4:45:21<5:23:43,  1.49s/it]
 47%|████▋     | 11424/24420 [4:45:22<5:23:06,  1.49s/it]
 47%|████▋     | 11425/24420 [4:45:24<5:22:46,  1.49s/it]
                                                         
{'loss': 0.0962, 'learning_rate': 5e-05, 'epoch': 2.34}

 47%|████▋     | 11425/24420 [4:45:24<5:22:46,  1.49s/it]
 47%|████▋     | 11426/24420 [4:45:25<5:22:26,  1.49s/it]
 47%|████▋     | 11427/24420 [4:45:26<5:22:15,  1.49s/it]
 47%|████▋     | 11428/24420 [4:45:28<5:21:33,  1.49s/it]
 47%|████▋     | 11429/24420 [4:45:29<5:21:29,  1.48s/it]
 47%|████▋     | 11430/24420 [4:45:31<5:21:22,  1.48s/it]
                                                         
{'loss': 0.0732, 'learning_rate': 5e-05, 'epoch': 2.34}

 47%|████▋     | 11430/24420 [4:45:31<5:21:22,  1.48s/it]
 47%|████▋     | 11431/24420 [4:45:32<5:21:20,  1.48s/it]
 47%|████▋     | 11432/24420 [4:45:34<5:21:32,  1.49s/it]
 47%|████▋     | 11433/24420 [4:45:35<5:22:13,  1.49s/it]
 47%|████▋     | 11434/24420 [4:45:37<5:22:20,  1.49s/it]
 47%|████▋     | 11435/24420 [4:45:38<5:21:55,  1.49s/it]
                                                         
{'loss': 0.0944, 'learning_rate': 5e-05, 'epoch': 2.34}

 47%|████▋     | 11435/24420 [4:45:38<5:21:55,  1.49s/it]
 47%|████▋     | 11436/24420 [4:45:40<5:21:54,  1.49s/it]
 47%|████▋     | 11437/24420 [4:45:41<5:22:28,  1.49s/it]
 47%|████▋     | 11438/24420 [4:45:43<5:22:15,  1.49s/it]
 47%|████▋     | 11439/24420 [4:45:44<5:22:59,  1.49s/it]
 47%|████▋     | 11440/24420 [4:45:46<5:22:10,  1.49s/it]
                                                         
{'loss': 0.0779, 'learning_rate': 5e-05, 'epoch': 2.34}

 47%|████▋     | 11440/24420 [4:45:46<5:22:10,  1.49s/it]
 47%|████▋     | 11441/24420 [4:45:47<5:22:42,  1.49s/it]
 47%|████▋     | 11442/24420 [4:45:49<5:22:37,  1.49s/it]
 47%|████▋     | 11443/24420 [4:45:50<5:21:59,  1.49s/it]
 47%|████▋     | 11444/24420 [4:45:52<5:21:52,  1.49s/it]
 47%|████▋     | 11445/24420 [4:45:53<5:21:52,  1.49s/it]
                                                         
{'loss': 0.0838, 'learning_rate': 5e-05, 'epoch': 2.34}

 47%|████▋     | 11445/24420 [4:45:53<5:21:52,  1.49s/it]
 47%|████▋     | 11446/24420 [4:45:55<5:21:45,  1.49s/it]
 47%|████▋     | 11447/24420 [4:45:56<5:21:43,  1.49s/it]
 47%|████▋     | 11448/24420 [4:45:58<5:21:51,  1.49s/it]
 47%|████▋     | 11449/24420 [4:45:59<5:21:28,  1.49s/it]
 47%|████▋     | 11450/24420 [4:46:01<5:21:34,  1.49s/it]
                                                         
{'loss': 0.0811, 'learning_rate': 5e-05, 'epoch': 2.34}

 47%|████▋     | 11450/24420 [4:46:01<5:21:34,  1.49s/it]
 47%|████▋     | 11451/24420 [4:46:02<5:22:04,  1.49s/it]
 47%|████▋     | 11452/24420 [4:46:04<5:21:59,  1.49s/it]
 47%|████▋     | 11453/24420 [4:46:05<5:21:47,  1.49s/it]
 47%|████▋     | 11454/24420 [4:46:07<5:21:51,  1.49s/it]
 47%|████▋     | 11455/24420 [4:46:08<5:22:01,  1.49s/it]
                                                         
{'loss': 0.099, 'learning_rate': 5e-05, 'epoch': 2.35}

 47%|████▋     | 11455/24420 [4:46:08<5:22:01,  1.49s/it]
 47%|████▋     | 11456/24420 [4:46:10<5:21:51,  1.49s/it]
 47%|████▋     | 11457/24420 [4:46:11<5:21:50,  1.49s/it]
 47%|████▋     | 11458/24420 [4:46:13<5:21:48,  1.49s/it]
 47%|████▋     | 11459/24420 [4:46:14<5:21:25,  1.49s/it]
 47%|████▋     | 11460/24420 [4:46:16<5:21:41,  1.49s/it]
                                                         
{'loss': 0.0859, 'learning_rate': 5e-05, 'epoch': 2.35}

 47%|████▋     | 11460/24420 [4:46:16<5:21:41,  1.49s/it]
 47%|████▋     | 11461/24420 [4:46:17<5:22:18,  1.49s/it]
 47%|████▋     | 11462/24420 [4:46:19<5:22:47,  1.49s/it]
 47%|████▋     | 11463/24420 [4:46:20<5:22:08,  1.49s/it]
 47%|████▋     | 11464/24420 [4:46:22<5:21:43,  1.49s/it]
 47%|████▋     | 11465/24420 [4:46:23<5:22:09,  1.49s/it]
                                                         
{'loss': 0.085, 'learning_rate': 5e-05, 'epoch': 2.35}

 47%|████▋     | 11465/24420 [4:46:23<5:22:09,  1.49s/it]
 47%|████▋     | 11466/24420 [4:46:25<5:22:20,  1.49s/it]
 47%|████▋     | 11467/24420 [4:46:26<5:22:15,  1.49s/it]
 47%|████▋     | 11468/24420 [4:46:28<5:21:35,  1.49s/it]
 47%|████▋     | 11469/24420 [4:46:29<5:21:54,  1.49s/it]
 47%|████▋     | 11470/24420 [4:46:31<5:21:39,  1.49s/it]
                                                         
{'loss': 0.086, 'learning_rate': 5e-05, 'epoch': 2.35}

 47%|████▋     | 11470/24420 [4:46:31<5:21:39,  1.49s/it]
 47%|████▋     | 11471/24420 [4:46:32<5:22:54,  1.50s/it]
 47%|████▋     | 11472/24420 [4:46:34<5:23:20,  1.50s/it]
 47%|████▋     | 11473/24420 [4:46:35<5:22:53,  1.50s/it]
 47%|████▋     | 11474/24420 [4:46:37<5:22:45,  1.50s/it]
 47%|████▋     | 11475/24420 [4:46:38<5:22:30,  1.49s/it]
                                                         
{'loss': 0.0859, 'learning_rate': 5e-05, 'epoch': 2.35}

 47%|████▋     | 11475/24420 [4:46:38<5:22:30,  1.49s/it]
 47%|████▋     | 11476/24420 [4:46:40<5:22:31,  1.50s/it]
 47%|████▋     | 11477/24420 [4:46:41<5:24:04,  1.50s/it]
 47%|████▋     | 11478/24420 [4:46:43<5:24:14,  1.50s/it]
 47%|████▋     | 11479/24420 [4:46:44<5:23:19,  1.50s/it]
 47%|████▋     | 11480/24420 [4:46:46<5:23:52,  1.50s/it]
                                                         
{'loss': 0.08, 'learning_rate': 5e-05, 'epoch': 2.35}

 47%|████▋     | 11480/24420 [4:46:46<5:23:52,  1.50s/it]
 47%|████▋     | 11481/24420 [4:46:47<5:23:03,  1.50s/it]
 47%|████▋     | 11482/24420 [4:46:49<5:22:35,  1.50s/it]
 47%|████▋     | 11483/24420 [4:46:50<5:22:49,  1.50s/it]
 47%|████▋     | 11484/24420 [4:46:52<5:23:34,  1.50s/it]
 47%|████▋     | 11485/24420 [4:46:53<5:23:27,  1.50s/it]
                                                         
{'loss': 0.1046, 'learning_rate': 5e-05, 'epoch': 2.35}

 47%|████▋     | 11485/24420 [4:46:53<5:23:27,  1.50s/it]
 47%|████▋     | 11486/24420 [4:46:55<5:24:24,  1.50s/it]
 47%|████▋     | 11487/24420 [4:46:56<5:23:25,  1.50s/it]
 47%|████▋     | 11488/24420 [4:46:58<5:22:57,  1.50s/it]
 47%|████▋     | 11489/24420 [4:46:59<5:22:15,  1.50s/it]
 47%|████▋     | 11490/24420 [4:47:01<5:21:58,  1.49s/it]
                                                         
{'loss': 0.0726, 'learning_rate': 5e-05, 'epoch': 2.35}

 47%|████▋     | 11490/24420 [4:47:01<5:21:58,  1.49s/it]
 47%|████▋     | 11491/24420 [4:47:02<5:22:10,  1.50s/it]
 47%|████▋     | 11492/24420 [4:47:03<5:21:59,  1.49s/it]
 47%|████▋     | 11493/24420 [4:47:05<5:21:15,  1.49s/it]
 47%|████▋     | 11494/24420 [4:47:06<5:21:10,  1.49s/it]
 47%|████▋     | 11495/24420 [4:47:08<5:21:09,  1.49s/it]
                                                         
{'loss': 0.0985, 'learning_rate': 5e-05, 'epoch': 2.35}

 47%|████▋     | 11495/24420 [4:47:08<5:21:09,  1.49s/it]
 47%|████▋     | 11496/24420 [4:47:09<5:21:24,  1.49s/it]
 47%|████▋     | 11497/24420 [4:47:11<5:21:36,  1.49s/it]
 47%|████▋     | 11498/24420 [4:47:12<5:21:36,  1.49s/it]
 47%|████▋     | 11499/24420 [4:47:14<5:22:04,  1.50s/it]
 47%|████▋     | 11500/24420 [4:47:15<5:22:05,  1.50s/it]
                                                         
{'loss': 0.1151, 'learning_rate': 5e-05, 'epoch': 2.35}

 47%|████▋     | 11500/24420 [4:47:15<5:22:05,  1.50s/it]
 47%|████▋     | 11501/24420 [4:47:17<5:22:14,  1.50s/it]
 47%|████▋     | 11502/24420 [4:47:18<5:22:13,  1.50s/it]
 47%|████▋     | 11503/24420 [4:47:20<5:21:28,  1.49s/it]
 47%|████▋     | 11504/24420 [4:47:21<5:21:55,  1.50s/it]
 47%|████▋     | 11505/24420 [4:47:23<5:21:48,  1.50s/it]
                                                         
{'loss': 0.0874, 'learning_rate': 5e-05, 'epoch': 2.36}

 47%|████▋     | 11505/24420 [4:47:23<5:21:48,  1.50s/it]
 47%|████▋     | 11506/24420 [4:47:24<5:21:50,  1.50s/it]
 47%|████▋     | 11507/24420 [4:47:26<5:22:18,  1.50s/it]
 47%|████▋     | 11508/24420 [4:47:27<5:24:59,  1.51s/it]
 47%|████▋     | 11509/24420 [4:47:29<5:24:04,  1.51s/it]
 47%|████▋     | 11510/24420 [4:47:30<5:23:14,  1.50s/it]
                                                         
{'loss': 0.0866, 'learning_rate': 5e-05, 'epoch': 2.36}

 47%|████▋     | 11510/24420 [4:47:30<5:23:14,  1.50s/it]
 47%|████▋     | 11511/24420 [4:47:32<5:22:44,  1.50s/it]
 47%|████▋     | 11512/24420 [4:47:33<5:21:45,  1.50s/it]
 47%|████▋     | 11513/24420 [4:47:35<5:21:42,  1.50s/it]
 47%|████▋     | 11514/24420 [4:47:36<5:21:51,  1.50s/it]
 47%|████▋     | 11515/24420 [4:47:38<5:22:07,  1.50s/it]
                                                         
{'loss': 0.0936, 'learning_rate': 5e-05, 'epoch': 2.36}

 47%|████▋     | 11515/24420 [4:47:38<5:22:07,  1.50s/it]
 47%|████▋     | 11516/24420 [4:47:39<5:21:20,  1.49s/it]
 47%|████▋     | 11517/24420 [4:47:41<5:21:28,  1.49s/it]
 47%|████▋     | 11518/24420 [4:47:42<5:20:59,  1.49s/it]
 47%|████▋     | 11519/24420 [4:47:44<5:21:34,  1.50s/it]
 47%|████▋     | 11520/24420 [4:47:45<5:21:53,  1.50s/it]
                                                         
{'loss': 0.0849, 'learning_rate': 5e-05, 'epoch': 2.36}

 47%|████▋     | 11520/24420 [4:47:45<5:21:53,  1.50s/it]
 47%|████▋     | 11521/24420 [4:47:47<5:22:17,  1.50s/it]
 47%|████▋     | 11522/24420 [4:47:48<5:21:30,  1.50s/it]
 47%|████▋     | 11523/24420 [4:47:50<5:21:15,  1.49s/it]
 47%|████▋     | 11524/24420 [4:47:51<5:20:25,  1.49s/it]
 47%|████▋     | 11525/24420 [4:47:53<5:19:27,  1.49s/it]
                                                         
{'loss': 0.0744, 'learning_rate': 5e-05, 'epoch': 2.36}

 47%|████▋     | 11525/24420 [4:47:53<5:19:27,  1.49s/it]
 47%|████▋     | 11526/24420 [4:47:54<5:19:49,  1.49s/it]
 47%|████▋     | 11527/24420 [4:47:56<5:19:49,  1.49s/it]
 47%|████▋     | 11528/24420 [4:47:57<5:20:31,  1.49s/it]
 47%|████▋     | 11529/24420 [4:47:59<5:20:16,  1.49s/it]
 47%|████▋     | 11530/24420 [4:48:00<5:20:44,  1.49s/it]
                                                         
{'loss': 0.0851, 'learning_rate': 5e-05, 'epoch': 2.36}

 47%|████▋     | 11530/24420 [4:48:00<5:20:44,  1.49s/it]
 47%|████▋     | 11531/24420 [4:48:02<5:21:09,  1.50s/it]
 47%|████▋     | 11532/24420 [4:48:03<5:22:15,  1.50s/it]
 47%|████▋     | 11533/24420 [4:48:05<5:21:41,  1.50s/it]
 47%|████▋     | 11534/24420 [4:48:06<5:21:06,  1.50s/it]
 47%|████▋     | 11535/24420 [4:48:08<5:20:49,  1.49s/it]
                                                         
{'loss': 0.0865, 'learning_rate': 5e-05, 'epoch': 2.36}

 47%|████▋     | 11535/24420 [4:48:08<5:20:49,  1.49s/it]
 47%|████▋     | 11536/24420 [4:48:09<5:20:23,  1.49s/it]
 47%|████▋     | 11537/24420 [4:48:11<5:20:19,  1.49s/it]
 47%|████▋     | 11538/24420 [4:48:12<5:19:58,  1.49s/it]
 47%|████▋     | 11539/24420 [4:48:14<5:20:09,  1.49s/it]
 47%|████▋     | 11540/24420 [4:48:15<5:19:49,  1.49s/it]
                                                         
{'loss': 0.1024, 'learning_rate': 5e-05, 'epoch': 2.36}

 47%|████▋     | 11540/24420 [4:48:15<5:19:49,  1.49s/it]
 47%|████▋     | 11541/24420 [4:48:17<5:20:33,  1.49s/it]
 47%|████▋     | 11542/24420 [4:48:18<5:21:20,  1.50s/it]
 47%|████▋     | 11543/24420 [4:48:20<5:20:54,  1.50s/it]
 47%|████▋     | 11544/24420 [4:48:21<5:20:55,  1.50s/it]
 47%|████▋     | 11545/24420 [4:48:23<5:20:45,  1.49s/it]
                                                         
{'loss': 0.0945, 'learning_rate': 5e-05, 'epoch': 2.36}

 47%|████▋     | 11545/24420 [4:48:23<5:20:45,  1.49s/it]
 47%|████▋     | 11546/24420 [4:48:24<5:20:45,  1.49s/it]
 47%|████▋     | 11547/24420 [4:48:26<5:21:19,  1.50s/it]
 47%|████▋     | 11548/24420 [4:48:27<5:21:14,  1.50s/it]
 47%|████▋     | 11549/24420 [4:48:29<5:22:15,  1.50s/it]
 47%|████▋     | 11550/24420 [4:48:30<5:21:32,  1.50s/it]
                                                         
{'loss': 0.0966, 'learning_rate': 5e-05, 'epoch': 2.36}

 47%|████▋     | 11550/24420 [4:48:30<5:21:32,  1.50s/it]
 47%|████▋     | 11551/24420 [4:48:32<5:21:31,  1.50s/it]
 47%|████▋     | 11552/24420 [4:48:33<5:21:16,  1.50s/it]
 47%|████▋     | 11553/24420 [4:48:35<5:21:27,  1.50s/it]
 47%|████▋     | 11554/24420 [4:48:36<5:20:41,  1.50s/it]
 47%|████▋     | 11555/24420 [4:48:38<5:19:57,  1.49s/it]
                                                         
{'loss': 0.0927, 'learning_rate': 5e-05, 'epoch': 2.37}

 47%|████▋     | 11555/24420 [4:48:38<5:19:57,  1.49s/it]
 47%|████▋     | 11556/24420 [4:48:39<5:20:37,  1.50s/it]
 47%|████▋     | 11557/24420 [4:48:41<5:21:03,  1.50s/it]
 47%|████▋     | 11558/24420 [4:48:42<5:21:01,  1.50s/it]
 47%|████▋     | 11559/24420 [4:48:44<5:20:54,  1.50s/it]
 47%|████▋     | 11560/24420 [4:48:45<5:20:30,  1.50s/it]
                                                         
{'loss': 0.0683, 'learning_rate': 5e-05, 'epoch': 2.37}

 47%|████▋     | 11560/24420 [4:48:45<5:20:30,  1.50s/it]
 47%|████▋     | 11561/24420 [4:48:47<5:21:23,  1.50s/it]
 47%|████▋     | 11562/24420 [4:48:48<5:21:20,  1.50s/it]
 47%|████▋     | 11563/24420 [4:48:50<5:20:58,  1.50s/it]
 47%|████▋     | 11564/24420 [4:48:51<5:20:41,  1.50s/it]
 47%|████▋     | 11565/24420 [4:48:53<5:19:55,  1.49s/it]
                                                         
{'loss': 0.0941, 'learning_rate': 5e-05, 'epoch': 2.37}

 47%|████▋     | 11565/24420 [4:48:53<5:19:55,  1.49s/it]
 47%|████▋     | 11566/24420 [4:48:54<5:20:04,  1.49s/it]
 47%|████▋     | 11567/24420 [4:48:56<5:20:37,  1.50s/it]
 47%|████▋     | 11568/24420 [4:48:57<5:21:45,  1.50s/it]
 47%|████▋     | 11569/24420 [4:48:59<5:21:11,  1.50s/it]
 47%|████▋     | 11570/24420 [4:49:00<5:20:41,  1.50s/it]
                                                         
{'loss': 0.0837, 'learning_rate': 5e-05, 'epoch': 2.37}

 47%|████▋     | 11570/24420 [4:49:00<5:20:41,  1.50s/it]
 47%|████▋     | 11571/24420 [4:49:02<5:20:51,  1.50s/it]
 47%|████▋     | 11572/24420 [4:49:03<5:23:29,  1.51s/it]
 47%|████▋     | 11573/24420 [4:49:05<5:22:49,  1.51s/it]
 47%|████▋     | 11574/24420 [4:49:06<5:21:54,  1.50s/it]
 47%|████▋     | 11575/24420 [4:49:08<5:20:49,  1.50s/it]
                                                         
{'loss': 0.1013, 'learning_rate': 5e-05, 'epoch': 2.37}

 47%|████▋     | 11575/24420 [4:49:08<5:20:49,  1.50s/it]
 47%|████▋     | 11576/24420 [4:49:09<5:19:47,  1.49s/it]
 47%|████▋     | 11577/24420 [4:49:11<5:20:20,  1.50s/it]
 47%|████▋     | 11578/24420 [4:49:12<5:20:26,  1.50s/it]
 47%|████▋     | 11579/24420 [4:49:14<5:20:32,  1.50s/it]
 47%|████▋     | 11580/24420 [4:49:15<5:20:40,  1.50s/it]
                                                         
{'loss': 0.0985, 'learning_rate': 5e-05, 'epoch': 2.37}

 47%|████▋     | 11580/24420 [4:49:15<5:20:40,  1.50s/it]
 47%|████▋     | 11581/24420 [4:49:17<5:20:26,  1.50s/it]
 47%|████▋     | 11582/24420 [4:49:18<5:19:57,  1.50s/it]
 47%|████▋     | 11583/24420 [4:49:20<5:19:29,  1.49s/it]
 47%|████▋     | 11584/24420 [4:49:21<5:19:14,  1.49s/it]
 47%|████▋     | 11585/24420 [4:49:23<5:20:08,  1.50s/it]
                                                         
{'loss': 0.0857, 'learning_rate': 5e-05, 'epoch': 2.37}

 47%|████▋     | 11585/24420 [4:49:23<5:20:08,  1.50s/it]
 47%|████▋     | 11586/24420 [4:49:24<5:20:04,  1.50s/it]
 47%|████▋     | 11587/24420 [4:49:26<5:19:14,  1.49s/it]
 47%|████▋     | 11588/24420 [4:49:27<5:18:43,  1.49s/it]
 47%|████▋     | 11589/24420 [4:49:29<5:19:09,  1.49s/it]
 47%|████▋     | 11590/24420 [4:49:30<5:19:03,  1.49s/it]
                                                         
{'loss': 0.084, 'learning_rate': 5e-05, 'epoch': 2.37}

 47%|████▋     | 11590/24420 [4:49:30<5:19:03,  1.49s/it]
 47%|████▋     | 11591/24420 [4:49:32<5:19:20,  1.49s/it]
 47%|████▋     | 11592/24420 [4:49:33<5:18:39,  1.49s/it]
 47%|████▋     | 11593/24420 [4:49:35<5:18:24,  1.49s/it]
 47%|████▋     | 11594/24420 [4:49:36<5:17:41,  1.49s/it]
 47%|████▋     | 11595/24420 [4:49:38<5:17:44,  1.49s/it]
                                                         
{'loss': 0.0932, 'learning_rate': 5e-05, 'epoch': 2.37}

 47%|████▋     | 11595/24420 [4:49:38<5:17:44,  1.49s/it]
 47%|████▋     | 11596/24420 [4:49:39<5:17:44,  1.49s/it]
 47%|████▋     | 11597/24420 [4:49:40<5:17:19,  1.48s/it]
 47%|████▋     | 11598/24420 [4:49:42<5:17:48,  1.49s/it]
 47%|████▋     | 11599/24420 [4:49:43<5:18:38,  1.49s/it]
 48%|████▊     | 11600/24420 [4:49:45<5:18:18,  1.49s/it]
                                                         
{'loss': 0.0855, 'learning_rate': 5e-05, 'epoch': 2.38}

 48%|████▊     | 11600/24420 [4:49:45<5:18:18,  1.49s/it]
 48%|████▊     | 11601/24420 [4:49:50<8:47:44,  2.47s/it]
 48%|████▊     | 11602/24420 [4:49:51<7:44:22,  2.17s/it]
 48%|████▊     | 11603/24420 [4:49:53<7:01:35,  1.97s/it]
 48%|████▊     | 11604/24420 [4:49:54<6:30:58,  1.83s/it]
 48%|████▊     | 11605/24420 [4:49:56<6:08:54,  1.73s/it]
                                                         
{'loss': 0.0962, 'learning_rate': 5e-05, 'epoch': 2.38}

 48%|████▊     | 11605/24420 [4:49:56<6:08:54,  1.73s/it]
 48%|████▊     | 11606/24420 [4:49:57<5:53:54,  1.66s/it]
 48%|████▊     | 11607/24420 [4:49:59<5:43:07,  1.61s/it]
 48%|████▊     | 11608/24420 [4:50:00<5:36:23,  1.58s/it]
 48%|████▊     | 11609/24420 [4:50:02<5:30:51,  1.55s/it]
 48%|████▊     | 11610/24420 [4:50:03<5:27:01,  1.53s/it]
                                                         
{'loss': 0.0918, 'learning_rate': 5e-05, 'epoch': 2.38}

 48%|████▊     | 11610/24420 [4:50:03<5:27:01,  1.53s/it]
 48%|████▊     | 11611/24420 [4:50:05<5:24:23,  1.52s/it]
 48%|████▊     | 11612/24420 [4:50:06<5:23:21,  1.51s/it]
 48%|████▊     | 11613/24420 [4:50:08<5:21:28,  1.51s/it]
 48%|████▊     | 11614/24420 [4:50:09<5:20:05,  1.50s/it]
 48%|████▊     | 11615/24420 [4:50:11<5:19:39,  1.50s/it]
                                                         
{'loss': 0.0811, 'learning_rate': 5e-05, 'epoch': 2.38}

 48%|████▊     | 11615/24420 [4:50:11<5:19:39,  1.50s/it]
 48%|████▊     | 11616/24420 [4:50:12<5:19:11,  1.50s/it]
 48%|████▊     | 11617/24420 [4:50:14<5:18:43,  1.49s/it]
 48%|████▊     | 11618/24420 [4:50:15<5:18:48,  1.49s/it]
 48%|████▊     | 11619/24420 [4:50:17<5:18:48,  1.49s/it]
 48%|████▊     | 11620/24420 [4:50:18<5:18:48,  1.49s/it]
                                                         
{'loss': 0.0873, 'learning_rate': 5e-05, 'epoch': 2.38}

 48%|████▊     | 11620/24420 [4:50:18<5:18:48,  1.49s/it]
 48%|████▊     | 11621/24420 [4:50:20<5:18:53,  1.49s/it]
 48%|████▊     | 11622/24420 [4:50:21<5:18:56,  1.50s/it]
 48%|████▊     | 11623/24420 [4:50:23<5:18:23,  1.49s/it]
 48%|████▊     | 11624/24420 [4:50:24<5:17:42,  1.49s/it]
 48%|████▊     | 11625/24420 [4:50:26<5:17:39,  1.49s/it]
                                                         
{'loss': 0.0947, 'learning_rate': 5e-05, 'epoch': 2.38}

 48%|████▊     | 11625/24420 [4:50:26<5:17:39,  1.49s/it]
 48%|████▊     | 11626/24420 [4:50:27<5:18:21,  1.49s/it]
 48%|████▊     | 11627/24420 [4:50:29<5:18:15,  1.49s/it]
 48%|████▊     | 11628/24420 [4:50:30<5:18:05,  1.49s/it]
 48%|████▊     | 11629/24420 [4:50:32<5:18:01,  1.49s/it]
 48%|████▊     | 11630/24420 [4:50:33<5:18:12,  1.49s/it]
                                                         
{'loss': 0.1046, 'learning_rate': 5e-05, 'epoch': 2.38}

 48%|████▊     | 11630/24420 [4:50:33<5:18:12,  1.49s/it]
 48%|████▊     | 11631/24420 [4:50:35<5:18:41,  1.50s/it]
 48%|████▊     | 11632/24420 [4:50:36<5:17:53,  1.49s/it]
 48%|████▊     | 11633/24420 [4:50:37<5:18:45,  1.50s/it]
 48%|████▊     | 11634/24420 [4:50:39<5:18:02,  1.49s/it]
 48%|████▊     | 11635/24420 [4:50:40<5:17:43,  1.49s/it]
                                                         
{'loss': 0.0795, 'learning_rate': 5e-05, 'epoch': 2.38}

 48%|████▊     | 11635/24420 [4:50:40<5:17:43,  1.49s/it]
 48%|████▊     | 11636/24420 [4:50:42<5:18:03,  1.49s/it]
 48%|████▊     | 11637/24420 [4:50:43<5:17:39,  1.49s/it]
 48%|████▊     | 11638/24420 [4:50:45<5:18:00,  1.49s/it]
 48%|████▊     | 11639/24420 [4:50:46<5:17:46,  1.49s/it]
 48%|████▊     | 11640/24420 [4:50:48<5:17:44,  1.49s/it]
                                                         
{'loss': 0.1055, 'learning_rate': 5e-05, 'epoch': 2.38}

 48%|████▊     | 11640/24420 [4:50:48<5:17:44,  1.49s/it]
 48%|████▊     | 11641/24420 [4:50:49<5:17:36,  1.49s/it]
 48%|████▊     | 11642/24420 [4:50:51<5:17:37,  1.49s/it]
 48%|████▊     | 11643/24420 [4:50:52<5:17:49,  1.49s/it]
 48%|████▊     | 11644/24420 [4:50:54<5:17:45,  1.49s/it]
 48%|████▊     | 11645/24420 [4:50:55<5:17:45,  1.49s/it]
                                                         
{'loss': 0.0827, 'learning_rate': 5e-05, 'epoch': 2.38}

 48%|████▊     | 11645/24420 [4:50:55<5:17:45,  1.49s/it]
 48%|████▊     | 11646/24420 [4:50:57<5:17:39,  1.49s/it]
 48%|████▊     | 11647/24420 [4:50:58<5:17:34,  1.49s/it]
 48%|████▊     | 11648/24420 [4:51:00<5:17:16,  1.49s/it]
 48%|████▊     | 11649/24420 [4:51:01<5:17:06,  1.49s/it]
 48%|████▊     | 11650/24420 [4:51:03<5:17:37,  1.49s/it]
                                                         
{'loss': 0.0922, 'learning_rate': 5e-05, 'epoch': 2.39}

 48%|████▊     | 11650/24420 [4:51:03<5:17:37,  1.49s/it]
 48%|████▊     | 11651/24420 [4:51:04<5:17:37,  1.49s/it]
 48%|████▊     | 11652/24420 [4:51:06<5:17:02,  1.49s/it]
 48%|████▊     | 11653/24420 [4:51:07<5:17:03,  1.49s/it]
 48%|████▊     | 11654/24420 [4:51:09<5:17:01,  1.49s/it]
 48%|████▊     | 11655/24420 [4:51:10<5:16:40,  1.49s/it]
                                                         
{'loss': 0.089, 'learning_rate': 5e-05, 'epoch': 2.39}

 48%|████▊     | 11655/24420 [4:51:10<5:16:40,  1.49s/it]
 48%|████▊     | 11656/24420 [4:51:12<5:16:31,  1.49s/it]
 48%|████▊     | 11657/24420 [4:51:13<5:15:58,  1.49s/it]
 48%|████▊     | 11658/24420 [4:51:15<5:16:01,  1.49s/it]
 48%|████▊     | 11659/24420 [4:51:16<5:16:24,  1.49s/it]
 48%|████▊     | 11660/24420 [4:51:18<5:16:36,  1.49s/it]
                                                         
{'loss': 0.0798, 'learning_rate': 5e-05, 'epoch': 2.39}

 48%|████▊     | 11660/24420 [4:51:18<5:16:36,  1.49s/it]
 48%|████▊     | 11661/24420 [4:51:19<5:17:03,  1.49s/it]
 48%|████▊     | 11662/24420 [4:51:21<5:17:33,  1.49s/it]
 48%|████▊     | 11663/24420 [4:51:22<5:17:21,  1.49s/it]
 48%|████▊     | 11664/24420 [4:51:24<5:16:53,  1.49s/it]
 48%|████▊     | 11665/24420 [4:51:25<5:17:13,  1.49s/it]
                                                         
{'loss': 0.098, 'learning_rate': 5e-05, 'epoch': 2.39}

 48%|████▊     | 11665/24420 [4:51:25<5:17:13,  1.49s/it]
 48%|████▊     | 11666/24420 [4:51:27<5:16:58,  1.49s/it]
 48%|████▊     | 11667/24420 [4:51:28<5:16:39,  1.49s/it]
 48%|████▊     | 11668/24420 [4:51:30<5:16:47,  1.49s/it]
 48%|████▊     | 11669/24420 [4:51:31<5:17:05,  1.49s/it]
 48%|████▊     | 11670/24420 [4:51:33<5:17:22,  1.49s/it]
                                                         
{'loss': 0.0755, 'learning_rate': 5e-05, 'epoch': 2.39}

 48%|████▊     | 11670/24420 [4:51:33<5:17:22,  1.49s/it]
 48%|████▊     | 11671/24420 [4:51:34<5:17:38,  1.49s/it]
 48%|████▊     | 11672/24420 [4:51:36<5:17:33,  1.49s/it]
 48%|████▊     | 11673/24420 [4:51:37<5:17:13,  1.49s/it]
 48%|████▊     | 11674/24420 [4:51:39<5:17:13,  1.49s/it]
 48%|████▊     | 11675/24420 [4:51:40<5:18:07,  1.50s/it]
                                                         
{'loss': 0.1051, 'learning_rate': 5e-05, 'epoch': 2.39}

 48%|████▊     | 11675/24420 [4:51:40<5:18:07,  1.50s/it]
 48%|████▊     | 11676/24420 [4:51:42<5:18:41,  1.50s/it]
 48%|████▊     | 11677/24420 [4:51:43<5:18:39,  1.50s/it]
 48%|████▊     | 11678/24420 [4:51:45<5:17:54,  1.50s/it]
 48%|████▊     | 11679/24420 [4:51:46<5:17:46,  1.50s/it]
 48%|████▊     | 11680/24420 [4:51:48<5:17:21,  1.49s/it]
                                                         
{'loss': 0.0782, 'learning_rate': 5e-05, 'epoch': 2.39}

 48%|████▊     | 11680/24420 [4:51:48<5:17:21,  1.49s/it]
 48%|████▊     | 11681/24420 [4:51:49<5:17:06,  1.49s/it]
 48%|████▊     | 11682/24420 [4:51:51<5:17:05,  1.49s/it]
 48%|████▊     | 11683/24420 [4:51:52<5:16:30,  1.49s/it]
 48%|████▊     | 11684/24420 [4:51:54<5:16:23,  1.49s/it]
 48%|████▊     | 11685/24420 [4:51:55<5:16:56,  1.49s/it]
                                                         
{'loss': 0.092, 'learning_rate': 5e-05, 'epoch': 2.39}

 48%|████▊     | 11685/24420 [4:51:55<5:16:56,  1.49s/it]
 48%|████▊     | 11686/24420 [4:51:57<5:16:56,  1.49s/it]
 48%|████▊     | 11687/24420 [4:51:58<5:16:47,  1.49s/it]
 48%|████▊     | 11688/24420 [4:52:00<5:16:35,  1.49s/it]
 48%|████▊     | 11689/24420 [4:52:01<5:16:36,  1.49s/it]
 48%|████▊     | 11690/24420 [4:52:03<5:16:50,  1.49s/it]
                                                         
{'loss': 0.1077, 'learning_rate': 5e-05, 'epoch': 2.39}

 48%|████▊     | 11690/24420 [4:52:03<5:16:50,  1.49s/it]
 48%|████▊     | 11691/24420 [4:52:04<5:16:56,  1.49s/it]
 48%|████▊     | 11692/24420 [4:52:06<5:17:01,  1.49s/it]
 48%|████▊     | 11693/24420 [4:52:07<5:16:45,  1.49s/it]
 48%|████▊     | 11694/24420 [4:52:09<5:17:54,  1.50s/it]
 48%|████▊     | 11695/24420 [4:52:10<5:18:14,  1.50s/it]
                                                         
{'loss': 0.0818, 'learning_rate': 5e-05, 'epoch': 2.39}

 48%|████▊     | 11695/24420 [4:52:10<5:18:14,  1.50s/it]
 48%|████▊     | 11696/24420 [4:52:12<5:18:04,  1.50s/it]
 48%|████▊     | 11697/24420 [4:52:13<5:17:56,  1.50s/it]
 48%|████▊     | 11698/24420 [4:52:15<5:20:15,  1.51s/it]
 48%|████▊     | 11699/24420 [4:52:16<5:18:37,  1.50s/it]
 48%|████▊     | 11700/24420 [4:52:18<5:17:55,  1.50s/it]
                                                         
{'loss': 0.0907, 'learning_rate': 5e-05, 'epoch': 2.4}

 48%|████▊     | 11700/24420 [4:52:18<5:17:55,  1.50s/it]
 48%|████▊     | 11701/24420 [4:52:19<5:17:56,  1.50s/it]
 48%|████▊     | 11702/24420 [4:52:21<5:17:26,  1.50s/it]
 48%|████▊     | 11703/24420 [4:52:22<5:17:19,  1.50s/it]
 48%|████▊     | 11704/24420 [4:52:24<5:17:07,  1.50s/it]
 48%|████▊     | 11705/24420 [4:52:25<5:17:38,  1.50s/it]
                                                         
{'loss': 0.0883, 'learning_rate': 5e-05, 'epoch': 2.4}

 48%|████▊     | 11705/24420 [4:52:25<5:17:38,  1.50s/it]
 48%|████▊     | 11706/24420 [4:52:27<5:18:02,  1.50s/it]
 48%|████▊     | 11707/24420 [4:52:28<5:17:33,  1.50s/it]
 48%|████▊     | 11708/24420 [4:52:30<5:18:08,  1.50s/it]
 48%|████▊     | 11709/24420 [4:52:31<5:18:01,  1.50s/it]
 48%|████▊     | 11710/24420 [4:52:33<5:17:14,  1.50s/it]
                                                         
{'loss': 0.091, 'learning_rate': 5e-05, 'epoch': 2.4}

 48%|████▊     | 11710/24420 [4:52:33<5:17:14,  1.50s/it]
 48%|████▊     | 11711/24420 [4:52:34<5:16:51,  1.50s/it]
 48%|████▊     | 11712/24420 [4:52:36<5:17:07,  1.50s/it]
 48%|████▊     | 11713/24420 [4:52:37<5:16:31,  1.49s/it]
 48%|████▊     | 11714/24420 [4:52:38<5:15:35,  1.49s/it]
 48%|████▊     | 11715/24420 [4:52:40<5:15:25,  1.49s/it]
                                                         
{'loss': 0.0914, 'learning_rate': 5e-05, 'epoch': 2.4}

 48%|████▊     | 11715/24420 [4:52:40<5:15:25,  1.49s/it]
 48%|████▊     | 11716/24420 [4:52:41<5:15:56,  1.49s/it]
 48%|████▊     | 11717/24420 [4:52:43<5:19:05,  1.51s/it]
 48%|████▊     | 11718/24420 [4:52:45<5:18:05,  1.50s/it]
 48%|████▊     | 11719/24420 [4:52:46<5:17:28,  1.50s/it]
 48%|████▊     | 11720/24420 [4:52:47<5:16:52,  1.50s/it]
                                                         
{'loss': 0.0914, 'learning_rate': 5e-05, 'epoch': 2.4}

 48%|████▊     | 11720/24420 [4:52:48<5:16:52,  1.50s/it]
 48%|████▊     | 11721/24420 [4:52:49<5:17:33,  1.50s/it]
 48%|████▊     | 11722/24420 [4:52:51<5:17:06,  1.50s/it]
 48%|████▊     | 11723/24420 [4:52:52<5:16:51,  1.50s/it]
 48%|████▊     | 11724/24420 [4:52:53<5:16:53,  1.50s/it]
 48%|████▊     | 11725/24420 [4:52:55<5:16:58,  1.50s/it]
                                                         
{'loss': 0.1028, 'learning_rate': 5e-05, 'epoch': 2.4}

 48%|████▊     | 11725/24420 [4:52:55<5:16:58,  1.50s/it]
 48%|████▊     | 11726/24420 [4:52:56<5:16:16,  1.49s/it]
 48%|████▊     | 11727/24420 [4:52:58<5:16:10,  1.49s/it]
 48%|████▊     | 11728/24420 [4:52:59<5:15:30,  1.49s/it]
 48%|████▊     | 11729/24420 [4:53:01<5:15:36,  1.49s/it]
 48%|████▊     | 11730/24420 [4:53:02<5:14:55,  1.49s/it]
                                                         
{'loss': 0.0858, 'learning_rate': 5e-05, 'epoch': 2.4}

 48%|████▊     | 11730/24420 [4:53:02<5:14:55,  1.49s/it]
 48%|████▊     | 11731/24420 [4:53:04<5:15:44,  1.49s/it]
 48%|████▊     | 11732/24420 [4:53:05<5:16:06,  1.49s/it]
 48%|████▊     | 11733/24420 [4:53:07<5:16:55,  1.50s/it]
 48%|████▊     | 11734/24420 [4:53:08<5:16:37,  1.50s/it]
 48%|████▊     | 11735/24420 [4:53:10<5:16:26,  1.50s/it]
                                                         
{'loss': 0.1091, 'learning_rate': 5e-05, 'epoch': 2.4}

 48%|████▊     | 11735/24420 [4:53:10<5:16:26,  1.50s/it]
 48%|████▊     | 11736/24420 [4:53:11<5:16:05,  1.50s/it]
 48%|████▊     | 11737/24420 [4:53:13<5:16:38,  1.50s/it]
 48%|████▊     | 11738/24420 [4:53:14<5:17:00,  1.50s/it]
 48%|████▊     | 11739/24420 [4:53:16<5:16:32,  1.50s/it]
 48%|████▊     | 11740/24420 [4:53:17<5:15:50,  1.49s/it]
                                                         
{'loss': 0.1373, 'learning_rate': 5e-05, 'epoch': 2.4}

 48%|████▊     | 11740/24420 [4:53:17<5:15:50,  1.49s/it]
 48%|████▊     | 11741/24420 [4:53:19<5:15:42,  1.49s/it]
 48%|████▊     | 11742/24420 [4:53:20<5:15:10,  1.49s/it]
 48%|████▊     | 11743/24420 [4:53:22<5:15:40,  1.49s/it]
 48%|████▊     | 11744/24420 [4:53:23<5:16:11,  1.50s/it]
 48%|████▊     | 11745/24420 [4:53:25<5:16:12,  1.50s/it]
                                                         
{'loss': 0.0943, 'learning_rate': 5e-05, 'epoch': 2.4}

 48%|████▊     | 11745/24420 [4:53:25<5:16:12,  1.50s/it]
 48%|████▊     | 11746/24420 [4:53:26<5:16:30,  1.50s/it]
 48%|████▊     | 11747/24420 [4:53:28<5:15:47,  1.50s/it]
 48%|████▊     | 11748/24420 [4:53:29<5:15:18,  1.49s/it]
 48%|████▊     | 11749/24420 [4:53:31<5:15:38,  1.49s/it]
 48%|████▊     | 11750/24420 [4:53:32<5:15:38,  1.49s/it]
                                                         
{'loss': 0.0882, 'learning_rate': 5e-05, 'epoch': 2.41}

 48%|████▊     | 11750/24420 [4:53:32<5:15:38,  1.49s/it]
 48%|████▊     | 11751/24420 [4:53:34<5:16:36,  1.50s/it]
 48%|████▊     | 11752/24420 [4:53:35<5:16:44,  1.50s/it]
 48%|████▊     | 11753/24420 [4:53:37<5:15:58,  1.50s/it]
 48%|████▊     | 11754/24420 [4:53:38<5:17:45,  1.51s/it]
 48%|████▊     | 11755/24420 [4:53:40<5:17:08,  1.50s/it]
                                                         
{'loss': 0.0933, 'learning_rate': 5e-05, 'epoch': 2.41}

 48%|████▊     | 11755/24420 [4:53:40<5:17:08,  1.50s/it]
 48%|████▊     | 11756/24420 [4:53:41<5:17:50,  1.51s/it]
 48%|████▊     | 11757/24420 [4:53:43<5:18:32,  1.51s/it]
 48%|████▊     | 11758/24420 [4:53:44<5:18:28,  1.51s/it]
 48%|████▊     | 11759/24420 [4:53:46<5:17:28,  1.50s/it]
 48%|████▊     | 11760/24420 [4:53:47<5:16:39,  1.50s/it]
                                                         
{'loss': 0.1007, 'learning_rate': 5e-05, 'epoch': 2.41}

 48%|████▊     | 11760/24420 [4:53:47<5:16:39,  1.50s/it]
 48%|████▊     | 11761/24420 [4:53:49<5:16:41,  1.50s/it]
 48%|████▊     | 11762/24420 [4:53:50<5:16:08,  1.50s/it]
 48%|████▊     | 11763/24420 [4:53:52<5:15:34,  1.50s/it]
 48%|████▊     | 11764/24420 [4:53:53<5:15:00,  1.49s/it]
 48%|████▊     | 11765/24420 [4:53:55<5:14:51,  1.49s/it]
                                                         
{'loss': 0.0732, 'learning_rate': 5e-05, 'epoch': 2.41}

 48%|████▊     | 11765/24420 [4:53:55<5:14:51,  1.49s/it]
 48%|████▊     | 11766/24420 [4:53:56<5:14:30,  1.49s/it]
 48%|████▊     | 11767/24420 [4:53:58<5:14:12,  1.49s/it]
 48%|████▊     | 11768/24420 [4:53:59<5:15:21,  1.50s/it]
 48%|████▊     | 11769/24420 [4:54:01<5:16:01,  1.50s/it]
 48%|████▊     | 11770/24420 [4:54:02<5:16:02,  1.50s/it]
                                                         
{'loss': 0.0864, 'learning_rate': 5e-05, 'epoch': 2.41}

 48%|████▊     | 11770/24420 [4:54:02<5:16:02,  1.50s/it]
 48%|████▊     | 11771/24420 [4:54:04<5:15:54,  1.50s/it]
 48%|████▊     | 11772/24420 [4:54:05<5:16:16,  1.50s/it]
 48%|████▊     | 11773/24420 [4:54:07<5:16:43,  1.50s/it]
 48%|████▊     | 11774/24420 [4:54:08<5:16:17,  1.50s/it]
 48%|████▊     | 11775/24420 [4:54:10<5:15:27,  1.50s/it]
                                                         
{'loss': 0.0876, 'learning_rate': 5e-05, 'epoch': 2.41}

 48%|████▊     | 11775/24420 [4:54:10<5:15:27,  1.50s/it]
 48%|████▊     | 11776/24420 [4:54:11<5:15:55,  1.50s/it]
 48%|████▊     | 11777/24420 [4:54:13<5:15:37,  1.50s/it]
 48%|████▊     | 11778/24420 [4:54:14<5:15:45,  1.50s/it]
 48%|████▊     | 11779/24420 [4:54:16<5:15:24,  1.50s/it]
 48%|████▊     | 11780/24420 [4:54:17<5:15:44,  1.50s/it]
                                                         
{'loss': 0.0826, 'learning_rate': 5e-05, 'epoch': 2.41}

 48%|████▊     | 11780/24420 [4:54:17<5:15:44,  1.50s/it]
 48%|████▊     | 11781/24420 [4:54:19<5:17:20,  1.51s/it]
 48%|████▊     | 11782/24420 [4:54:20<5:16:08,  1.50s/it]
 48%|████▊     | 11783/24420 [4:54:22<5:15:32,  1.50s/it]
 48%|████▊     | 11784/24420 [4:54:23<5:15:57,  1.50s/it]
 48%|████▊     | 11785/24420 [4:54:25<5:15:52,  1.50s/it]
                                                         
{'loss': 0.085, 'learning_rate': 5e-05, 'epoch': 2.41}

 48%|████▊     | 11785/24420 [4:54:25<5:15:52,  1.50s/it]
 48%|████▊     | 11786/24420 [4:54:26<5:16:36,  1.50s/it]
 48%|████▊     | 11787/24420 [4:54:28<5:16:28,  1.50s/it]
 48%|████▊     | 11788/24420 [4:54:29<5:16:08,  1.50s/it]
 48%|████▊     | 11789/24420 [4:54:31<5:16:26,  1.50s/it]
 48%|████▊     | 11790/24420 [4:54:32<5:16:00,  1.50s/it]
                                                         
{'loss': 0.0768, 'learning_rate': 5e-05, 'epoch': 2.41}

 48%|████▊     | 11790/24420 [4:54:32<5:16:00,  1.50s/it]
 48%|████▊     | 11791/24420 [4:54:34<5:15:50,  1.50s/it]
 48%|████▊     | 11792/24420 [4:54:35<5:15:56,  1.50s/it]
 48%|████▊     | 11793/24420 [4:54:37<5:16:31,  1.50s/it]
 48%|████▊     | 11794/24420 [4:54:38<5:15:24,  1.50s/it]
 48%|████▊     | 11795/24420 [4:54:40<5:15:03,  1.50s/it]
                                                         
{'loss': 0.089, 'learning_rate': 5e-05, 'epoch': 2.42}

 48%|████▊     | 11795/24420 [4:54:40<5:15:03,  1.50s/it]
 48%|████▊     | 11796/24420 [4:54:41<5:14:43,  1.50s/it]
 48%|████▊     | 11797/24420 [4:54:43<5:15:22,  1.50s/it]
 48%|████▊     | 11798/24420 [4:54:44<5:15:12,  1.50s/it]
 48%|████▊     | 11799/24420 [4:54:46<5:14:54,  1.50s/it]
 48%|████▊     | 11800/24420 [4:54:47<5:14:46,  1.50s/it]
                                                         
{'loss': 0.0928, 'learning_rate': 5e-05, 'epoch': 2.42}

 48%|████▊     | 11800/24420 [4:54:47<5:14:46,  1.50s/it]
 48%|████▊     | 11801/24420 [4:54:54<10:11:15,  2.91s/it]
 48%|████▊     | 11802/24420 [4:54:55<8:42:15,  2.48s/it] 
 48%|████▊     | 11803/24420 [4:54:57<7:40:26,  2.19s/it]
 48%|████▊     | 11804/24420 [4:54:58<6:56:44,  1.98s/it]
 48%|████▊     | 11805/24420 [4:55:00<6:25:20,  1.83s/it]
                                                         
{'loss': 0.095, 'learning_rate': 5e-05, 'epoch': 2.42}

 48%|████▊     | 11805/24420 [4:55:00<6:25:20,  1.83s/it]
 48%|████▊     | 11806/24420 [4:55:01<6:03:50,  1.73s/it]
 48%|████▊     | 11807/24420 [4:55:03<5:49:54,  1.66s/it]
 48%|████▊     | 11808/24420 [4:55:04<5:39:52,  1.62s/it]
 48%|████▊     | 11809/24420 [4:55:06<5:32:50,  1.58s/it]
 48%|████▊     | 11810/24420 [4:55:07<5:26:51,  1.56s/it]
                                                         
{'loss': 0.0963, 'learning_rate': 5e-05, 'epoch': 2.42}

 48%|████▊     | 11810/24420 [4:55:07<5:26:51,  1.56s/it]
 48%|████▊     | 11811/24420 [4:55:09<5:23:40,  1.54s/it]
 48%|████▊     | 11812/24420 [4:55:10<5:21:10,  1.53s/it]
 48%|████▊     | 11813/24420 [4:55:12<5:18:12,  1.51s/it]
 48%|████▊     | 11814/24420 [4:55:13<5:16:48,  1.51s/it]
 48%|████▊     | 11815/24420 [4:55:15<5:16:35,  1.51s/it]
                                                         
{'loss': 0.0873, 'learning_rate': 5e-05, 'epoch': 2.42}

 48%|████▊     | 11815/24420 [4:55:15<5:16:35,  1.51s/it]
 48%|████▊     | 11816/24420 [4:55:16<5:16:20,  1.51s/it]
 48%|████▊     | 11817/24420 [4:55:18<5:15:30,  1.50s/it]
 48%|████▊     | 11818/24420 [4:55:19<5:14:53,  1.50s/it]
 48%|████▊     | 11819/24420 [4:55:20<5:14:06,  1.50s/it]
 48%|████▊     | 11820/24420 [4:55:22<5:14:41,  1.50s/it]
                                                         
{'loss': 0.0861, 'learning_rate': 5e-05, 'epoch': 2.42}

 48%|████▊     | 11820/24420 [4:55:22<5:14:41,  1.50s/it]
 48%|████▊     | 11821/24420 [4:55:24<5:15:14,  1.50s/it]
 48%|████▊     | 11822/24420 [4:55:25<5:15:34,  1.50s/it]
 48%|████▊     | 11823/24420 [4:55:27<5:14:59,  1.50s/it]
 48%|████▊     | 11824/24420 [4:55:28<5:15:00,  1.50s/it]
 48%|████▊     | 11825/24420 [4:55:30<5:14:48,  1.50s/it]
                                                         
{'loss': 0.0867, 'learning_rate': 5e-05, 'epoch': 2.42}

 48%|████▊     | 11825/24420 [4:55:30<5:14:48,  1.50s/it]
 48%|████▊     | 11826/24420 [4:55:31<5:14:31,  1.50s/it]
 48%|████▊     | 11827/24420 [4:55:33<5:14:20,  1.50s/it]
 48%|████▊     | 11828/24420 [4:55:34<5:14:04,  1.50s/it]
 48%|████▊     | 11829/24420 [4:55:35<5:14:17,  1.50s/it]
 48%|████▊     | 11830/24420 [4:55:37<5:14:58,  1.50s/it]
                                                         
{'loss': 0.1032, 'learning_rate': 5e-05, 'epoch': 2.42}

 48%|████▊     | 11830/24420 [4:55:37<5:14:58,  1.50s/it]
 48%|████▊     | 11831/24420 [4:55:39<5:14:35,  1.50s/it]
 48%|████▊     | 11832/24420 [4:55:40<5:14:45,  1.50s/it]
 48%|████▊     | 11833/24420 [4:55:41<5:14:09,  1.50s/it]
 48%|████▊     | 11834/24420 [4:55:43<5:15:12,  1.50s/it]
 48%|████▊     | 11835/24420 [4:55:45<5:14:25,  1.50s/it]
                                                         
{'loss': 0.0937, 'learning_rate': 5e-05, 'epoch': 2.42}

 48%|████▊     | 11835/24420 [4:55:45<5:14:25,  1.50s/it]
 48%|████▊     | 11836/24420 [4:55:46<5:15:12,  1.50s/it]
 48%|████▊     | 11837/24420 [4:55:48<5:14:19,  1.50s/it]
 48%|████▊     | 11838/24420 [4:55:49<5:13:52,  1.50s/it]
 48%|████▊     | 11839/24420 [4:55:50<5:12:51,  1.49s/it]
 48%|████▊     | 11840/24420 [4:55:52<5:12:30,  1.49s/it]
                                                         
{'loss': 0.1011, 'learning_rate': 5e-05, 'epoch': 2.42}

 48%|████▊     | 11840/24420 [4:55:52<5:12:30,  1.49s/it]
 48%|████▊     | 11841/24420 [4:55:53<5:12:42,  1.49s/it]
 48%|████▊     | 11842/24420 [4:55:55<5:13:00,  1.49s/it]
 48%|████▊     | 11843/24420 [4:55:56<5:12:32,  1.49s/it]
 49%|████▊     | 11844/24420 [4:55:58<5:12:54,  1.49s/it]
 49%|████▊     | 11845/24420 [4:55:59<5:12:32,  1.49s/it]
                                                         
{'loss': 0.0841, 'learning_rate': 5e-05, 'epoch': 2.43}

 49%|████▊     | 11845/24420 [4:55:59<5:12:32,  1.49s/it]
 49%|████▊     | 11846/24420 [4:56:01<5:12:23,  1.49s/it]
 49%|████▊     | 11847/24420 [4:56:02<5:12:14,  1.49s/it]
 49%|████▊     | 11848/24420 [4:56:04<5:11:38,  1.49s/it]
 49%|████▊     | 11849/24420 [4:56:05<5:11:46,  1.49s/it]
 49%|████▊     | 11850/24420 [4:56:07<5:12:15,  1.49s/it]
                                                         
{'loss': 0.0978, 'learning_rate': 5e-05, 'epoch': 2.43}

 49%|████▊     | 11850/24420 [4:56:07<5:12:15,  1.49s/it]
 49%|████▊     | 11851/24420 [4:56:08<5:11:52,  1.49s/it]
 49%|████▊     | 11852/24420 [4:56:10<5:12:11,  1.49s/it]
 49%|████▊     | 11853/24420 [4:56:11<5:11:49,  1.49s/it]
 49%|████▊     | 11854/24420 [4:56:13<5:11:38,  1.49s/it]
 49%|████▊     | 11855/24420 [4:56:14<5:11:57,  1.49s/it]
                                                         
{'loss': 0.0979, 'learning_rate': 5e-05, 'epoch': 2.43}

 49%|████▊     | 11855/24420 [4:56:14<5:11:57,  1.49s/it]
 49%|████▊     | 11856/24420 [4:56:16<5:12:58,  1.49s/it]
 49%|████▊     | 11857/24420 [4:56:17<5:12:36,  1.49s/it]
 49%|████▊     | 11858/24420 [4:56:19<5:12:38,  1.49s/it]
 49%|████▊     | 11859/24420 [4:56:20<5:11:48,  1.49s/it]
 49%|████▊     | 11860/24420 [4:56:22<5:11:43,  1.49s/it]
                                                         
{'loss': 0.0891, 'learning_rate': 5e-05, 'epoch': 2.43}

 49%|████▊     | 11860/24420 [4:56:22<5:11:43,  1.49s/it]
 49%|████▊     | 11861/24420 [4:56:23<5:11:55,  1.49s/it]
 49%|████▊     | 11862/24420 [4:56:25<5:12:08,  1.49s/it]
 49%|████▊     | 11863/24420 [4:56:26<5:11:58,  1.49s/it]
 49%|████▊     | 11864/24420 [4:56:28<5:11:58,  1.49s/it]
 49%|████▊     | 11865/24420 [4:56:29<5:11:49,  1.49s/it]
                                                         
{'loss': 0.0876, 'learning_rate': 5e-05, 'epoch': 2.43}

 49%|████▊     | 11865/24420 [4:56:29<5:11:49,  1.49s/it]
 49%|████▊     | 11866/24420 [4:56:31<5:11:45,  1.49s/it]
 49%|████▊     | 11867/24420 [4:56:32<5:11:14,  1.49s/it]
 49%|████▊     | 11868/24420 [4:56:34<5:10:49,  1.49s/it]
 49%|████▊     | 11869/24420 [4:56:35<5:10:42,  1.49s/it]
 49%|████▊     | 11870/24420 [4:56:37<5:10:57,  1.49s/it]
                                                         
{'loss': 0.0827, 'learning_rate': 5e-05, 'epoch': 2.43}

 49%|████▊     | 11870/24420 [4:56:37<5:10:57,  1.49s/it]
 49%|████▊     | 11871/24420 [4:56:38<5:11:45,  1.49s/it]
 49%|████▊     | 11872/24420 [4:56:40<5:11:41,  1.49s/it]
 49%|████▊     | 11873/24420 [4:56:41<5:11:25,  1.49s/it]
 49%|████▊     | 11874/24420 [4:56:43<5:11:39,  1.49s/it]
 49%|████▊     | 11875/24420 [4:56:44<5:11:17,  1.49s/it]
                                                         
{'loss': 0.0847, 'learning_rate': 5e-05, 'epoch': 2.43}

 49%|████▊     | 11875/24420 [4:56:44<5:11:17,  1.49s/it]
 49%|████▊     | 11876/24420 [4:56:46<5:11:12,  1.49s/it]
 49%|████▊     | 11877/24420 [4:56:47<5:11:12,  1.49s/it]
 49%|████▊     | 11878/24420 [4:56:49<5:10:51,  1.49s/it]
 49%|████▊     | 11879/24420 [4:56:50<5:10:40,  1.49s/it]
 49%|████▊     | 11880/24420 [4:56:52<5:10:46,  1.49s/it]
                                                         
{'loss': 0.092, 'learning_rate': 5e-05, 'epoch': 2.43}

 49%|████▊     | 11880/24420 [4:56:52<5:10:46,  1.49s/it]
 49%|████▊     | 11881/24420 [4:56:53<5:10:53,  1.49s/it]
 49%|████▊     | 11882/24420 [4:56:55<5:10:55,  1.49s/it]
 49%|████▊     | 11883/24420 [4:56:56<5:10:19,  1.49s/it]
 49%|████▊     | 11884/24420 [4:56:57<5:10:09,  1.48s/it]
 49%|████▊     | 11885/24420 [4:56:59<5:09:54,  1.48s/it]
                                                         
{'loss': 0.0978, 'learning_rate': 5e-05, 'epoch': 2.43}

 49%|████▊     | 11885/24420 [4:56:59<5:09:54,  1.48s/it]
 49%|████▊     | 11886/24420 [4:57:00<5:10:32,  1.49s/it]
 49%|████▊     | 11887/24420 [4:57:02<5:10:46,  1.49s/it]
 49%|████▊     | 11888/24420 [4:57:03<5:10:37,  1.49s/it]
 49%|████▊     | 11889/24420 [4:57:05<5:10:35,  1.49s/it]
 49%|████▊     | 11890/24420 [4:57:06<5:10:47,  1.49s/it]
                                                         
{'loss': 0.0798, 'learning_rate': 5e-05, 'epoch': 2.43}

 49%|████▊     | 11890/24420 [4:57:06<5:10:47,  1.49s/it]
 49%|████▊     | 11891/24420 [4:57:08<5:11:10,  1.49s/it]
 49%|████▊     | 11892/24420 [4:57:09<5:10:52,  1.49s/it]
 49%|████▊     | 11893/24420 [4:57:11<5:10:33,  1.49s/it]
 49%|████▊     | 11894/24420 [4:57:12<5:10:29,  1.49s/it]
 49%|████▊     | 11895/24420 [4:57:14<5:10:29,  1.49s/it]
                                                         
{'loss': 0.0869, 'learning_rate': 5e-05, 'epoch': 2.44}

 49%|████▊     | 11895/24420 [4:57:14<5:10:29,  1.49s/it]
 49%|████▊     | 11896/24420 [4:57:15<5:10:40,  1.49s/it]
 49%|████▊     | 11897/24420 [4:57:17<5:10:46,  1.49s/it]
 49%|████▊     | 11898/24420 [4:57:18<5:11:23,  1.49s/it]
 49%|████▊     | 11899/24420 [4:57:20<5:11:16,  1.49s/it]
 49%|████▊     | 11900/24420 [4:57:21<5:11:13,  1.49s/it]
                                                         
{'loss': 0.0888, 'learning_rate': 5e-05, 'epoch': 2.44}

 49%|████▊     | 11900/24420 [4:57:21<5:11:13,  1.49s/it]
 49%|████▊     | 11901/24420 [4:57:23<5:10:37,  1.49s/it]
 49%|████▊     | 11902/24420 [4:57:24<5:10:23,  1.49s/it]
 49%|████▊     | 11903/24420 [4:57:26<5:10:43,  1.49s/it]
 49%|████▊     | 11904/24420 [4:57:27<5:10:34,  1.49s/it]
 49%|████▉     | 11905/24420 [4:57:29<5:11:01,  1.49s/it]
                                                         
{'loss': 0.0832, 'learning_rate': 5e-05, 'epoch': 2.44}

 49%|████▉     | 11905/24420 [4:57:29<5:11:01,  1.49s/it]
 49%|████▉     | 11906/24420 [4:57:30<5:11:36,  1.49s/it]
 49%|████▉     | 11907/24420 [4:57:32<5:11:36,  1.49s/it]
 49%|████▉     | 11908/24420 [4:57:33<5:11:10,  1.49s/it]
 49%|████▉     | 11909/24420 [4:57:35<5:10:27,  1.49s/it]
 49%|████▉     | 11910/24420 [4:57:36<5:11:02,  1.49s/it]
                                                         
{'loss': 0.0973, 'learning_rate': 5e-05, 'epoch': 2.44}

 49%|████▉     | 11910/24420 [4:57:36<5:11:02,  1.49s/it]
 49%|████▉     | 11911/24420 [4:57:38<5:11:21,  1.49s/it]
 49%|████▉     | 11912/24420 [4:57:39<5:10:50,  1.49s/it]
 49%|████▉     | 11913/24420 [4:57:41<5:10:57,  1.49s/it]
 49%|████▉     | 11914/24420 [4:57:42<5:10:47,  1.49s/it]
 49%|████▉     | 11915/24420 [4:57:44<5:10:40,  1.49s/it]
                                                         
{'loss': 0.0994, 'learning_rate': 5e-05, 'epoch': 2.44}

 49%|████▉     | 11915/24420 [4:57:44<5:10:40,  1.49s/it]
 49%|████▉     | 11916/24420 [4:57:45<5:10:21,  1.49s/it]
 49%|████▉     | 11917/24420 [4:57:47<5:10:36,  1.49s/it]
 49%|████▉     | 11918/24420 [4:57:48<5:10:40,  1.49s/it]
 49%|████▉     | 11919/24420 [4:57:50<5:10:03,  1.49s/it]
 49%|████▉     | 11920/24420 [4:57:51<5:09:53,  1.49s/it]
                                                         
{'loss': 0.0999, 'learning_rate': 5e-05, 'epoch': 2.44}

 49%|████▉     | 11920/24420 [4:57:51<5:09:53,  1.49s/it]
 49%|████▉     | 11921/24420 [4:57:53<5:09:46,  1.49s/it]
 49%|████▉     | 11922/24420 [4:57:54<5:10:38,  1.49s/it]
 49%|████▉     | 11923/24420 [4:57:56<5:10:38,  1.49s/it]
 49%|████▉     | 11924/24420 [4:57:57<5:10:00,  1.49s/it]
 49%|████▉     | 11925/24420 [4:57:59<5:10:01,  1.49s/it]
                                                         
{'loss': 0.084, 'learning_rate': 5e-05, 'epoch': 2.44}

 49%|████▉     | 11925/24420 [4:57:59<5:10:01,  1.49s/it]
 49%|████▉     | 11926/24420 [4:58:00<5:09:59,  1.49s/it]
 49%|████▉     | 11927/24420 [4:58:02<5:09:42,  1.49s/it]
 49%|████▉     | 11928/24420 [4:58:03<5:10:13,  1.49s/it]
 49%|████▉     | 11929/24420 [4:58:05<5:09:20,  1.49s/it]
 49%|████▉     | 11930/24420 [4:58:06<5:10:19,  1.49s/it]
                                                         
{'loss': 0.0934, 'learning_rate': 5e-05, 'epoch': 2.44}

 49%|████▉     | 11930/24420 [4:58:06<5:10:19,  1.49s/it]
 49%|████▉     | 11931/24420 [4:58:08<5:10:18,  1.49s/it]
 49%|████▉     | 11932/24420 [4:58:09<5:10:29,  1.49s/it]
 49%|████▉     | 11933/24420 [4:58:10<5:11:07,  1.49s/it]
 49%|████▉     | 11934/24420 [4:58:12<5:11:29,  1.50s/it]
 49%|████▉     | 11935/24420 [4:58:13<5:11:34,  1.50s/it]
                                                         
{'loss': 0.0985, 'learning_rate': 5e-05, 'epoch': 2.44}

 49%|████▉     | 11935/24420 [4:58:14<5:11:34,  1.50s/it]
 49%|████▉     | 11936/24420 [4:58:15<5:11:06,  1.50s/it]
 49%|████▉     | 11937/24420 [4:58:16<5:10:59,  1.49s/it]
 49%|████▉     | 11938/24420 [4:58:18<5:10:30,  1.49s/it]
 49%|████▉     | 11939/24420 [4:58:19<5:09:43,  1.49s/it]
 49%|████▉     | 11940/24420 [4:58:21<5:09:32,  1.49s/it]
                                                         
{'loss': 0.0901, 'learning_rate': 5e-05, 'epoch': 2.44}

 49%|████▉     | 11940/24420 [4:58:21<5:09:32,  1.49s/it]
 49%|████▉     | 11941/24420 [4:58:22<5:09:58,  1.49s/it]
 49%|████▉     | 11942/24420 [4:58:24<5:09:38,  1.49s/it]
 49%|████▉     | 11943/24420 [4:58:25<5:09:41,  1.49s/it]
 49%|████▉     | 11944/24420 [4:58:27<5:10:06,  1.49s/it]
 49%|████▉     | 11945/24420 [4:58:28<5:10:00,  1.49s/it]
                                                         
{'loss': 0.0985, 'learning_rate': 5e-05, 'epoch': 2.45}

 49%|████▉     | 11945/24420 [4:58:28<5:10:00,  1.49s/it]
 49%|████▉     | 11946/24420 [4:58:30<5:09:53,  1.49s/it]
 49%|████▉     | 11947/24420 [4:58:31<5:09:51,  1.49s/it]
 49%|████▉     | 11948/24420 [4:58:33<5:09:40,  1.49s/it]
 49%|████▉     | 11949/24420 [4:58:34<5:09:58,  1.49s/it]
 49%|████▉     | 11950/24420 [4:58:36<5:10:01,  1.49s/it]
                                                         
{'loss': 0.0921, 'learning_rate': 5e-05, 'epoch': 2.45}

 49%|████▉     | 11950/24420 [4:58:36<5:10:01,  1.49s/it]
 49%|████▉     | 11951/24420 [4:58:37<5:10:07,  1.49s/it]
 49%|████▉     | 11952/24420 [4:58:39<5:09:54,  1.49s/it]
 49%|████▉     | 11953/24420 [4:58:40<5:09:29,  1.49s/it]
 49%|████▉     | 11954/24420 [4:58:42<5:09:08,  1.49s/it]
 49%|████▉     | 11955/24420 [4:58:43<5:08:59,  1.49s/it]
                                                         
{'loss': 0.0744, 'learning_rate': 5e-05, 'epoch': 2.45}

 49%|████▉     | 11955/24420 [4:58:43<5:08:59,  1.49s/it]
 49%|████▉     | 11956/24420 [4:58:45<5:09:30,  1.49s/it]
 49%|████▉     | 11957/24420 [4:58:46<5:09:33,  1.49s/it]
 49%|████▉     | 11958/24420 [4:58:48<5:09:46,  1.49s/it]
 49%|████▉     | 11959/24420 [4:58:49<5:09:29,  1.49s/it]
 49%|████▉     | 11960/24420 [4:58:51<5:08:56,  1.49s/it]
                                                         
{'loss': 0.0697, 'learning_rate': 5e-05, 'epoch': 2.45}

 49%|████▉     | 11960/24420 [4:58:51<5:08:56,  1.49s/it]
 49%|████▉     | 11961/24420 [4:58:52<5:10:11,  1.49s/it]
 49%|████▉     | 11962/24420 [4:58:54<5:09:35,  1.49s/it]
 49%|████▉     | 11963/24420 [4:58:55<5:10:16,  1.49s/it]
 49%|████▉     | 11964/24420 [4:58:57<5:10:40,  1.50s/it]
 49%|████▉     | 11965/24420 [4:58:58<5:10:55,  1.50s/it]
                                                         
{'loss': 0.1333, 'learning_rate': 5e-05, 'epoch': 2.45}

 49%|████▉     | 11965/24420 [4:58:58<5:10:55,  1.50s/it]
 49%|████▉     | 11966/24420 [4:59:00<5:10:26,  1.50s/it]
 49%|████▉     | 11967/24420 [4:59:01<5:10:38,  1.50s/it]
 49%|████▉     | 11968/24420 [4:59:03<5:10:22,  1.50s/it]
 49%|████▉     | 11969/24420 [4:59:04<5:09:58,  1.49s/it]
 49%|████▉     | 11970/24420 [4:59:06<5:12:25,  1.51s/it]
                                                         
{'loss': 0.0859, 'learning_rate': 5e-05, 'epoch': 2.45}

 49%|████▉     | 11970/24420 [4:59:06<5:12:25,  1.51s/it]
 49%|████▉     | 11971/24420 [4:59:07<5:11:48,  1.50s/it]
 49%|████▉     | 11972/24420 [4:59:09<5:10:55,  1.50s/it]
 49%|████▉     | 11973/24420 [4:59:10<5:10:09,  1.50s/it]
 49%|████▉     | 11974/24420 [4:59:12<5:10:00,  1.49s/it]
 49%|████▉     | 11975/24420 [4:59:13<5:09:15,  1.49s/it]
                                                         
{'loss': 0.0947, 'learning_rate': 5e-05, 'epoch': 2.45}

 49%|████▉     | 11975/24420 [4:59:13<5:09:15,  1.49s/it]
 49%|████▉     | 11976/24420 [4:59:15<5:09:46,  1.49s/it]
 49%|████▉     | 11977/24420 [4:59:16<5:09:28,  1.49s/it]
 49%|████▉     | 11978/24420 [4:59:18<5:09:31,  1.49s/it]
 49%|████▉     | 11979/24420 [4:59:19<5:08:44,  1.49s/it]
 49%|████▉     | 11980/24420 [4:59:21<5:08:24,  1.49s/it]
                                                         
{'loss': 0.0934, 'learning_rate': 5e-05, 'epoch': 2.45}

 49%|████▉     | 11980/24420 [4:59:21<5:08:24,  1.49s/it]
 49%|████▉     | 11981/24420 [4:59:22<5:08:09,  1.49s/it]
 49%|████▉     | 11982/24420 [4:59:24<5:08:55,  1.49s/it]
 49%|████▉     | 11983/24420 [4:59:25<5:09:26,  1.49s/it]
 49%|████▉     | 11984/24420 [4:59:27<5:09:03,  1.49s/it]
 49%|████▉     | 11985/24420 [4:59:28<5:08:48,  1.49s/it]
                                                         
{'loss': 0.1026, 'learning_rate': 5e-05, 'epoch': 2.45}

 49%|████▉     | 11985/24420 [4:59:28<5:08:48,  1.49s/it]
 49%|████▉     | 11986/24420 [4:59:30<5:08:51,  1.49s/it]
 49%|████▉     | 11987/24420 [4:59:31<5:08:30,  1.49s/it]
 49%|████▉     | 11988/24420 [4:59:33<5:09:42,  1.49s/it]
 49%|████▉     | 11989/24420 [4:59:34<5:09:14,  1.49s/it]
 49%|████▉     | 11990/24420 [4:59:36<5:09:16,  1.49s/it]
                                                         
{'loss': 0.0855, 'learning_rate': 5e-05, 'epoch': 2.45}

 49%|████▉     | 11990/24420 [4:59:36<5:09:16,  1.49s/it]
 49%|████▉     | 11991/24420 [4:59:37<5:09:58,  1.50s/it]
 49%|████▉     | 11992/24420 [4:59:39<5:10:00,  1.50s/it]
 49%|████▉     | 11993/24420 [4:59:40<5:09:20,  1.49s/it]
 49%|████▉     | 11994/24420 [4:59:42<5:09:07,  1.49s/it]
 49%|████▉     | 11995/24420 [4:59:43<5:08:59,  1.49s/it]
                                                         
{'loss': 0.0953, 'learning_rate': 5e-05, 'epoch': 2.46}

 49%|████▉     | 11995/24420 [4:59:43<5:08:59,  1.49s/it]
 49%|████▉     | 11996/24420 [4:59:45<5:09:14,  1.49s/it]
 49%|████▉     | 11997/24420 [4:59:46<5:08:40,  1.49s/it]
 49%|████▉     | 11998/24420 [4:59:48<5:08:54,  1.49s/it]
 49%|████▉     | 11999/24420 [4:59:49<5:08:24,  1.49s/it]
 49%|████▉     | 12000/24420 [4:59:50<5:08:33,  1.49s/it]
                                                         
{'loss': 0.0814, 'learning_rate': 5e-05, 'epoch': 2.46}

 49%|████▉     | 12000/24420 [4:59:50<5:08:33,  1.49s/it]
 49%|████▉     | 12001/24420 [4:59:55<8:25:11,  2.44s/it]
 49%|████▉     | 12002/24420 [4:59:57<7:25:44,  2.15s/it]
 49%|████▉     | 12003/24420 [4:59:58<6:44:14,  1.95s/it]
 49%|████▉     | 12004/24420 [5:00:00<6:15:47,  1.82s/it]
 49%|████▉     | 12005/24420 [5:00:01<5:55:21,  1.72s/it]
                                                         
{'loss': 0.0913, 'learning_rate': 5e-05, 'epoch': 2.46}

 49%|████▉     | 12005/24420 [5:00:01<5:55:21,  1.72s/it]
 49%|████▉     | 12006/24420 [5:00:03<5:41:30,  1.65s/it]
 49%|████▉     | 12007/24420 [5:00:04<5:32:19,  1.61s/it]
 49%|████▉     | 12008/24420 [5:00:06<5:24:58,  1.57s/it]
 49%|████▉     | 12009/24420 [5:00:07<5:20:03,  1.55s/it]
 49%|████▉     | 12010/24420 [5:00:09<5:16:28,  1.53s/it]
                                                         
{'loss': 0.1006, 'learning_rate': 5e-05, 'epoch': 2.46}

 49%|████▉     | 12010/24420 [5:00:09<5:16:28,  1.53s/it]
 49%|████▉     | 12011/24420 [5:00:10<5:13:37,  1.52s/it]
 49%|████▉     | 12012/24420 [5:00:12<5:13:04,  1.51s/it]
 49%|████▉     | 12013/24420 [5:00:13<5:11:26,  1.51s/it]
 49%|████▉     | 12014/24420 [5:00:15<5:10:22,  1.50s/it]
 49%|████▉     | 12015/24420 [5:00:16<5:09:09,  1.50s/it]
                                                         
{'loss': 0.0791, 'learning_rate': 5e-05, 'epoch': 2.46}

 49%|████▉     | 12015/24420 [5:00:16<5:09:09,  1.50s/it]
 49%|████▉     | 12016/24420 [5:00:18<5:08:45,  1.49s/it]
 49%|████▉     | 12017/24420 [5:00:19<5:08:11,  1.49s/it]
 49%|████▉     | 12018/24420 [5:00:20<5:07:42,  1.49s/it]
 49%|████▉     | 12019/24420 [5:00:22<5:07:28,  1.49s/it]
 49%|████▉     | 12020/24420 [5:00:23<5:07:01,  1.49s/it]
                                                         
{'loss': 0.0874, 'learning_rate': 5e-05, 'epoch': 2.46}

 49%|████▉     | 12020/24420 [5:00:23<5:07:01,  1.49s/it]
 49%|████▉     | 12021/24420 [5:00:25<5:07:09,  1.49s/it]
 49%|████▉     | 12022/24420 [5:00:26<5:07:02,  1.49s/it]
 49%|████▉     | 12023/24420 [5:00:28<5:07:11,  1.49s/it]
 49%|████▉     | 12024/24420 [5:00:29<5:06:51,  1.49s/it]
 49%|████▉     | 12025/24420 [5:00:31<5:06:51,  1.49s/it]
                                                         
{'loss': 0.0898, 'learning_rate': 5e-05, 'epoch': 2.46}

 49%|████▉     | 12025/24420 [5:00:31<5:06:51,  1.49s/it]
 49%|████▉     | 12026/24420 [5:00:32<5:07:08,  1.49s/it]
 49%|████▉     | 12027/24420 [5:00:34<5:06:52,  1.49s/it]
 49%|████▉     | 12028/24420 [5:00:35<5:07:16,  1.49s/it]
 49%|████▉     | 12029/24420 [5:00:37<5:07:21,  1.49s/it]
 49%|████▉     | 12030/24420 [5:00:38<5:07:10,  1.49s/it]
                                                         
{'loss': 0.1217, 'learning_rate': 5e-05, 'epoch': 2.46}

 49%|████▉     | 12030/24420 [5:00:38<5:07:10,  1.49s/it]
 49%|████▉     | 12031/24420 [5:00:40<5:07:46,  1.49s/it]
 49%|████▉     | 12032/24420 [5:00:41<5:07:18,  1.49s/it]
 49%|████▉     | 12033/24420 [5:00:43<5:07:12,  1.49s/it]
 49%|████▉     | 12034/24420 [5:00:44<5:06:55,  1.49s/it]
 49%|████▉     | 12035/24420 [5:00:46<5:07:06,  1.49s/it]
                                                         
{'loss': 0.099, 'learning_rate': 5e-05, 'epoch': 2.46}

 49%|████▉     | 12035/24420 [5:00:46<5:07:06,  1.49s/it]
 49%|████▉     | 12036/24420 [5:00:47<5:06:59,  1.49s/it]
 49%|████▉     | 12037/24420 [5:00:49<5:07:14,  1.49s/it]
 49%|████▉     | 12038/24420 [5:00:50<5:07:29,  1.49s/it]
 49%|████▉     | 12039/24420 [5:00:52<5:07:56,  1.49s/it]
 49%|████▉     | 12040/24420 [5:00:53<5:07:25,  1.49s/it]
                                                         
{'loss': 0.0879, 'learning_rate': 5e-05, 'epoch': 2.47}

 49%|████▉     | 12040/24420 [5:00:53<5:07:25,  1.49s/it]
 49%|████▉     | 12041/24420 [5:00:55<5:07:20,  1.49s/it]
 49%|████▉     | 12042/24420 [5:00:56<5:10:56,  1.51s/it]
 49%|████▉     | 12043/24420 [5:00:58<5:09:42,  1.50s/it]
 49%|████▉     | 12044/24420 [5:00:59<5:08:57,  1.50s/it]
 49%|████▉     | 12045/24420 [5:01:01<5:08:13,  1.49s/it]
                                                         
{'loss': 0.0877, 'learning_rate': 5e-05, 'epoch': 2.47}

 49%|████▉     | 12045/24420 [5:01:01<5:08:13,  1.49s/it]
 49%|████▉     | 12046/24420 [5:01:02<5:08:02,  1.49s/it]
 49%|████▉     | 12047/24420 [5:01:04<5:07:19,  1.49s/it]
 49%|████▉     | 12048/24420 [5:01:05<5:06:47,  1.49s/it]
 49%|████▉     | 12049/24420 [5:01:07<5:06:27,  1.49s/it]
 49%|████▉     | 12050/24420 [5:01:08<5:06:58,  1.49s/it]
                                                         
{'loss': 0.0855, 'learning_rate': 5e-05, 'epoch': 2.47}

 49%|████▉     | 12050/24420 [5:01:08<5:06:58,  1.49s/it]
 49%|████▉     | 12051/24420 [5:01:10<5:06:56,  1.49s/it]
 49%|████▉     | 12052/24420 [5:01:11<5:06:17,  1.49s/it]
 49%|████▉     | 12053/24420 [5:01:13<5:06:11,  1.49s/it]
 49%|████▉     | 12054/24420 [5:01:14<5:06:22,  1.49s/it]
 49%|████▉     | 12055/24420 [5:01:16<5:06:51,  1.49s/it]
                                                         
{'loss': 0.0836, 'learning_rate': 5e-05, 'epoch': 2.47}

 49%|████▉     | 12055/24420 [5:01:16<5:06:51,  1.49s/it]
 49%|████▉     | 12056/24420 [5:01:17<5:07:10,  1.49s/it]
 49%|████▉     | 12057/24420 [5:01:19<5:06:55,  1.49s/it]
 49%|████▉     | 12058/24420 [5:01:20<5:07:05,  1.49s/it]
 49%|████▉     | 12059/24420 [5:01:22<5:06:41,  1.49s/it]
 49%|████▉     | 12060/24420 [5:01:23<5:06:49,  1.49s/it]
                                                         
{'loss': 0.0942, 'learning_rate': 5e-05, 'epoch': 2.47}

 49%|████▉     | 12060/24420 [5:01:23<5:06:49,  1.49s/it]
 49%|████▉     | 12061/24420 [5:01:25<5:07:19,  1.49s/it]
 49%|████▉     | 12062/24420 [5:01:26<5:06:42,  1.49s/it]
 49%|████▉     | 12063/24420 [5:01:27<5:06:24,  1.49s/it]
 49%|████▉     | 12064/24420 [5:01:29<5:06:53,  1.49s/it]
 49%|████▉     | 12065/24420 [5:01:30<5:06:23,  1.49s/it]
                                                         
{'loss': 0.0885, 'learning_rate': 5e-05, 'epoch': 2.47}

 49%|████▉     | 12065/24420 [5:01:30<5:06:23,  1.49s/it]
 49%|████▉     | 12066/24420 [5:01:32<5:06:50,  1.49s/it]
 49%|████▉     | 12067/24420 [5:01:33<5:07:18,  1.49s/it]
 49%|████▉     | 12068/24420 [5:01:35<5:07:46,  1.50s/it]
 49%|████▉     | 12069/24420 [5:01:36<5:07:52,  1.50s/it]
 49%|████▉     | 12070/24420 [5:01:38<5:07:39,  1.49s/it]
                                                         
{'loss': 0.1057, 'learning_rate': 5e-05, 'epoch': 2.47}

 49%|████▉     | 12070/24420 [5:01:38<5:07:39,  1.49s/it]
 49%|████▉     | 12071/24420 [5:01:39<5:07:30,  1.49s/it]
 49%|████▉     | 12072/24420 [5:01:41<5:06:47,  1.49s/it]
 49%|████▉     | 12073/24420 [5:01:42<5:06:13,  1.49s/it]
 49%|████▉     | 12074/24420 [5:01:44<5:06:25,  1.49s/it]
 49%|████▉     | 12075/24420 [5:01:45<5:06:13,  1.49s/it]
                                                         
{'loss': 0.092, 'learning_rate': 5e-05, 'epoch': 2.47}

 49%|████▉     | 12075/24420 [5:01:45<5:06:13,  1.49s/it]
 49%|████▉     | 12076/24420 [5:01:47<5:06:29,  1.49s/it]
 49%|████▉     | 12077/24420 [5:01:48<5:06:09,  1.49s/it]
 49%|████▉     | 12078/24420 [5:01:50<5:06:55,  1.49s/it]
 49%|████▉     | 12079/24420 [5:01:51<5:07:10,  1.49s/it]
 49%|████▉     | 12080/24420 [5:01:53<5:07:15,  1.49s/it]
                                                         
{'loss': 0.0951, 'learning_rate': 5e-05, 'epoch': 2.47}

 49%|████▉     | 12080/24420 [5:01:53<5:07:15,  1.49s/it]
 49%|████▉     | 12081/24420 [5:01:54<5:09:46,  1.51s/it]
 49%|████▉     | 12082/24420 [5:01:56<5:09:08,  1.50s/it]
 49%|████▉     | 12083/24420 [5:01:57<5:08:13,  1.50s/it]
 49%|████▉     | 12084/24420 [5:01:59<5:08:00,  1.50s/it]
 49%|████▉     | 12085/24420 [5:02:00<5:07:07,  1.49s/it]
                                                         
{'loss': 0.0862, 'learning_rate': 5e-05, 'epoch': 2.47}

 49%|████▉     | 12085/24420 [5:02:00<5:07:07,  1.49s/it]
 49%|████▉     | 12086/24420 [5:02:02<5:07:50,  1.50s/it]
 49%|████▉     | 12087/24420 [5:02:03<5:07:24,  1.50s/it]
 50%|████▉     | 12088/24420 [5:02:05<5:07:05,  1.49s/it]
 50%|████▉     | 12089/24420 [5:02:06<5:06:51,  1.49s/it]
 50%|████▉     | 12090/24420 [5:02:08<5:06:45,  1.49s/it]
                                                         
{'loss': 0.1028, 'learning_rate': 5e-05, 'epoch': 2.48}

 50%|████▉     | 12090/24420 [5:02:08<5:06:45,  1.49s/it]
 50%|████▉     | 12091/24420 [5:02:09<5:07:04,  1.49s/it]
 50%|████▉     | 12092/24420 [5:02:11<5:07:19,  1.50s/it]
 50%|████▉     | 12093/24420 [5:02:12<5:07:28,  1.50s/it]
 50%|████▉     | 12094/24420 [5:02:14<5:06:45,  1.49s/it]
 50%|████▉     | 12095/24420 [5:02:15<5:06:50,  1.49s/it]
                                                         
{'loss': 0.0983, 'learning_rate': 5e-05, 'epoch': 2.48}

 50%|████▉     | 12095/24420 [5:02:15<5:06:50,  1.49s/it]
 50%|████▉     | 12096/24420 [5:02:17<5:06:17,  1.49s/it]
 50%|████▉     | 12097/24420 [5:02:18<5:06:05,  1.49s/it]
 50%|████▉     | 12098/24420 [5:02:20<5:06:34,  1.49s/it]
 50%|████▉     | 12099/24420 [5:02:21<5:06:02,  1.49s/it]
 50%|████▉     | 12100/24420 [5:02:23<5:05:55,  1.49s/it]
                                                         
{'loss': 0.1023, 'learning_rate': 5e-05, 'epoch': 2.48}

 50%|████▉     | 12100/24420 [5:02:23<5:05:55,  1.49s/it]
 50%|████▉     | 12101/24420 [5:02:24<5:05:54,  1.49s/it]
 50%|████▉     | 12102/24420 [5:02:26<5:05:25,  1.49s/it]
 50%|████▉     | 12103/24420 [5:02:27<5:05:59,  1.49s/it]
 50%|████▉     | 12104/24420 [5:02:29<5:06:11,  1.49s/it]
 50%|████▉     | 12105/24420 [5:02:30<5:05:15,  1.49s/it]
                                                         
{'loss': 0.1103, 'learning_rate': 5e-05, 'epoch': 2.48}

 50%|████▉     | 12105/24420 [5:02:30<5:05:15,  1.49s/it]
 50%|████▉     | 12106/24420 [5:02:32<5:05:56,  1.49s/it]
 50%|████▉     | 12107/24420 [5:02:33<5:06:01,  1.49s/it]
 50%|████▉     | 12108/24420 [5:02:35<5:05:08,  1.49s/it]
 50%|████▉     | 12109/24420 [5:02:36<5:04:46,  1.49s/it]
 50%|████▉     | 12110/24420 [5:02:38<5:04:19,  1.48s/it]
                                                         
{'loss': 0.0842, 'learning_rate': 5e-05, 'epoch': 2.48}

 50%|████▉     | 12110/24420 [5:02:38<5:04:19,  1.48s/it]
 50%|████▉     | 12111/24420 [5:02:39<5:05:06,  1.49s/it]
 50%|████▉     | 12112/24420 [5:02:41<5:04:56,  1.49s/it]
 50%|████▉     | 12113/24420 [5:02:42<5:04:40,  1.49s/it]
 50%|████▉     | 12114/24420 [5:02:44<5:05:32,  1.49s/it]
 50%|████▉     | 12115/24420 [5:02:45<5:05:28,  1.49s/it]
                                                         
{'loss': 0.0878, 'learning_rate': 5e-05, 'epoch': 2.48}

 50%|████▉     | 12115/24420 [5:02:45<5:05:28,  1.49s/it]
 50%|████▉     | 12116/24420 [5:02:47<5:05:48,  1.49s/it]
 50%|████▉     | 12117/24420 [5:02:48<5:05:23,  1.49s/it]
 50%|████▉     | 12118/24420 [5:02:50<5:05:25,  1.49s/it]
 50%|████▉     | 12119/24420 [5:02:51<5:05:13,  1.49s/it]
 50%|████▉     | 12120/24420 [5:02:53<5:05:21,  1.49s/it]
                                                         
{'loss': 0.0919, 'learning_rate': 5e-05, 'epoch': 2.48}

 50%|████▉     | 12120/24420 [5:02:53<5:05:21,  1.49s/it]
 50%|████▉     | 12121/24420 [5:02:54<5:06:09,  1.49s/it]
 50%|████▉     | 12122/24420 [5:02:56<5:05:40,  1.49s/it]
 50%|████▉     | 12123/24420 [5:02:57<5:05:35,  1.49s/it]
 50%|████▉     | 12124/24420 [5:02:58<5:05:01,  1.49s/it]
 50%|████▉     | 12125/24420 [5:03:00<5:05:04,  1.49s/it]
                                                         
{'loss': 0.1118, 'learning_rate': 5e-05, 'epoch': 2.48}

 50%|████▉     | 12125/24420 [5:03:00<5:05:04,  1.49s/it]
 50%|████▉     | 12126/24420 [5:03:01<5:04:53,  1.49s/it]
 50%|████▉     | 12127/24420 [5:03:03<5:04:21,  1.49s/it]
 50%|████▉     | 12128/24420 [5:03:04<5:04:39,  1.49s/it]
 50%|████▉     | 12129/24420 [5:03:06<5:04:24,  1.49s/it]
 50%|████▉     | 12130/24420 [5:03:07<5:04:43,  1.49s/it]
                                                         
{'loss': 0.0884, 'learning_rate': 5e-05, 'epoch': 2.48}

 50%|████▉     | 12130/24420 [5:03:07<5:04:43,  1.49s/it]
 50%|████▉     | 12131/24420 [5:03:09<5:04:45,  1.49s/it]
 50%|████▉     | 12132/24420 [5:03:10<5:04:16,  1.49s/it]
 50%|████▉     | 12133/24420 [5:03:12<5:04:05,  1.48s/it]
 50%|████▉     | 12134/24420 [5:03:13<5:04:31,  1.49s/it]
 50%|████▉     | 12135/24420 [5:03:15<5:04:40,  1.49s/it]
                                                         
{'loss': 0.0789, 'learning_rate': 5e-05, 'epoch': 2.48}

 50%|████▉     | 12135/24420 [5:03:15<5:04:40,  1.49s/it]
 50%|████▉     | 12136/24420 [5:03:16<5:04:46,  1.49s/it]
 50%|████▉     | 12137/24420 [5:03:18<5:05:23,  1.49s/it]
 50%|████▉     | 12138/24420 [5:03:19<5:05:00,  1.49s/it]
 50%|████▉     | 12139/24420 [5:03:21<5:05:40,  1.49s/it]
 50%|████▉     | 12140/24420 [5:03:22<5:05:12,  1.49s/it]
                                                         
{'loss': 0.098, 'learning_rate': 5e-05, 'epoch': 2.49}

 50%|████▉     | 12140/24420 [5:03:22<5:05:12,  1.49s/it]
 50%|████▉     | 12141/24420 [5:03:24<5:05:31,  1.49s/it]
 50%|████▉     | 12142/24420 [5:03:25<5:04:56,  1.49s/it]
 50%|████▉     | 12143/24420 [5:03:27<5:04:57,  1.49s/it]
 50%|████▉     | 12144/24420 [5:03:28<5:04:44,  1.49s/it]
 50%|████▉     | 12145/24420 [5:03:30<5:04:48,  1.49s/it]
                                                         
{'loss': 0.0959, 'learning_rate': 5e-05, 'epoch': 2.49}

 50%|████▉     | 12145/24420 [5:03:30<5:04:48,  1.49s/it]
 50%|████▉     | 12146/24420 [5:03:31<5:04:49,  1.49s/it]
 50%|████▉     | 12147/24420 [5:03:33<5:04:43,  1.49s/it]
 50%|████▉     | 12148/24420 [5:03:34<5:04:32,  1.49s/it]
 50%|████▉     | 12149/24420 [5:03:36<5:04:41,  1.49s/it]
 50%|████▉     | 12150/24420 [5:03:37<5:04:21,  1.49s/it]
                                                         
{'loss': 0.0764, 'learning_rate': 5e-05, 'epoch': 2.49}

 50%|████▉     | 12150/24420 [5:03:37<5:04:21,  1.49s/it]
 50%|████▉     | 12151/24420 [5:03:39<5:04:45,  1.49s/it]
 50%|████▉     | 12152/24420 [5:03:40<5:04:24,  1.49s/it]
 50%|████▉     | 12153/24420 [5:03:42<5:04:21,  1.49s/it]
 50%|████▉     | 12154/24420 [5:03:43<5:04:18,  1.49s/it]
 50%|████▉     | 12155/24420 [5:03:45<5:03:34,  1.49s/it]
                                                         
{'loss': 0.1037, 'learning_rate': 5e-05, 'epoch': 2.49}

 50%|████▉     | 12155/24420 [5:03:45<5:03:34,  1.49s/it]
 50%|████▉     | 12156/24420 [5:03:46<5:03:31,  1.48s/it]
 50%|████▉     | 12157/24420 [5:03:48<5:03:44,  1.49s/it]
 50%|████▉     | 12158/24420 [5:03:49<5:04:04,  1.49s/it]
 50%|████▉     | 12159/24420 [5:03:51<5:04:20,  1.49s/it]
 50%|████▉     | 12160/24420 [5:03:52<5:03:56,  1.49s/it]
                                                         
{'loss': 0.0943, 'learning_rate': 5e-05, 'epoch': 2.49}

 50%|████▉     | 12160/24420 [5:03:52<5:03:56,  1.49s/it]
 50%|████▉     | 12161/24420 [5:03:54<5:04:01,  1.49s/it]
 50%|████▉     | 12162/24420 [5:03:55<5:03:42,  1.49s/it]
 50%|████▉     | 12163/24420 [5:03:57<5:04:29,  1.49s/it]
 50%|████▉     | 12164/24420 [5:03:58<5:04:09,  1.49s/it]
 50%|████▉     | 12165/24420 [5:04:00<5:04:19,  1.49s/it]
                                                         
{'loss': 0.0866, 'learning_rate': 5e-05, 'epoch': 2.49}

 50%|████▉     | 12165/24420 [5:04:00<5:04:19,  1.49s/it]
 50%|████▉     | 12166/24420 [5:04:01<5:03:51,  1.49s/it]
 50%|████▉     | 12167/24420 [5:04:02<5:04:28,  1.49s/it]
 50%|████▉     | 12168/24420 [5:04:04<5:06:54,  1.50s/it]
 50%|████▉     | 12169/24420 [5:04:06<5:06:13,  1.50s/it]
 50%|████▉     | 12170/24420 [5:04:07<5:05:37,  1.50s/it]
                                                         
{'loss': 0.094, 'learning_rate': 5e-05, 'epoch': 2.49}

 50%|████▉     | 12170/24420 [5:04:07<5:05:37,  1.50s/it]
 50%|████▉     | 12171/24420 [5:04:09<5:05:19,  1.50s/it]
 50%|████▉     | 12172/24420 [5:04:10<5:04:54,  1.49s/it]
 50%|████▉     | 12173/24420 [5:04:11<5:04:13,  1.49s/it]
 50%|████▉     | 12174/24420 [5:04:13<5:04:15,  1.49s/it]
 50%|████▉     | 12175/24420 [5:04:14<5:04:37,  1.49s/it]
                                                         
{'loss': 0.078, 'learning_rate': 5e-05, 'epoch': 2.49}

 50%|████▉     | 12175/24420 [5:04:14<5:04:37,  1.49s/it]
 50%|████▉     | 12176/24420 [5:04:16<5:03:59,  1.49s/it]
 50%|████▉     | 12177/24420 [5:04:17<5:04:37,  1.49s/it]
 50%|████▉     | 12178/24420 [5:04:19<5:04:58,  1.49s/it]
 50%|████▉     | 12179/24420 [5:04:20<5:04:07,  1.49s/it]
 50%|████▉     | 12180/24420 [5:04:22<5:03:39,  1.49s/it]
                                                         
{'loss': 0.0822, 'learning_rate': 5e-05, 'epoch': 2.49}

 50%|████▉     | 12180/24420 [5:04:22<5:03:39,  1.49s/it]
 50%|████▉     | 12181/24420 [5:04:23<5:03:28,  1.49s/it]
 50%|████▉     | 12182/24420 [5:04:25<5:03:19,  1.49s/it]
 50%|████▉     | 12183/24420 [5:04:26<5:04:35,  1.49s/it]
 50%|████▉     | 12184/24420 [5:04:28<5:04:09,  1.49s/it]
 50%|████▉     | 12185/24420 [5:04:29<5:04:04,  1.49s/it]
                                                         
{'loss': 0.0966, 'learning_rate': 5e-05, 'epoch': 2.49}

 50%|████▉     | 12185/24420 [5:04:29<5:04:04,  1.49s/it]
 50%|████▉     | 12186/24420 [5:04:31<5:03:32,  1.49s/it]
 50%|████▉     | 12187/24420 [5:04:32<5:03:11,  1.49s/it]
 50%|████▉     | 12188/24420 [5:04:34<5:03:19,  1.49s/it]
 50%|████▉     | 12189/24420 [5:04:35<5:03:37,  1.49s/it]
 50%|████▉     | 12190/24420 [5:04:37<5:03:41,  1.49s/it]
                                                         
{'loss': 0.0997, 'learning_rate': 5e-05, 'epoch': 2.5}

 50%|████▉     | 12190/24420 [5:04:37<5:03:41,  1.49s/it]
 50%|████▉     | 12191/24420 [5:04:38<5:04:57,  1.50s/it]
 50%|████▉     | 12192/24420 [5:04:40<5:04:46,  1.50s/it]
 50%|████▉     | 12193/24420 [5:04:41<5:04:26,  1.49s/it]
 50%|████▉     | 12194/24420 [5:04:43<5:04:04,  1.49s/it]
 50%|████▉     | 12195/24420 [5:04:44<5:03:56,  1.49s/it]
                                                         
{'loss': 0.0936, 'learning_rate': 5e-05, 'epoch': 2.5}

 50%|████▉     | 12195/24420 [5:04:44<5:03:56,  1.49s/it]
 50%|████▉     | 12196/24420 [5:04:46<5:03:33,  1.49s/it]
 50%|████▉     | 12197/24420 [5:04:47<5:03:21,  1.49s/it]
 50%|████▉     | 12198/24420 [5:04:49<5:04:26,  1.49s/it]
 50%|████▉     | 12199/24420 [5:04:50<5:04:25,  1.49s/it]
 50%|████▉     | 12200/24420 [5:04:52<5:04:17,  1.49s/it]
                                                         
{'loss': 0.0908, 'learning_rate': 5e-05, 'epoch': 2.5}

 50%|████▉     | 12200/24420 [5:04:52<5:04:17,  1.49s/it]
 50%|████▉     | 12201/24420 [5:04:57<8:27:00,  2.49s/it]
 50%|████▉     | 12202/24420 [5:04:58<7:26:27,  2.19s/it]
 50%|████▉     | 12203/24420 [5:05:00<6:43:12,  1.98s/it]
 50%|████▉     | 12204/24420 [5:05:01<6:12:56,  1.83s/it]
 50%|████▉     | 12205/24420 [5:05:03<5:51:55,  1.73s/it]
                                                         
{'loss': 0.0781, 'learning_rate': 5e-05, 'epoch': 2.5}

 50%|████▉     | 12205/24420 [5:05:03<5:51:55,  1.73s/it]
 50%|████▉     | 12206/24420 [5:05:04<5:37:26,  1.66s/it]
 50%|████▉     | 12207/24420 [5:05:06<5:27:05,  1.61s/it]
 50%|████▉     | 12208/24420 [5:05:07<5:19:46,  1.57s/it]
 50%|████▉     | 12209/24420 [5:05:08<5:14:58,  1.55s/it]
 50%|█████     | 12210/24420 [5:05:10<5:11:16,  1.53s/it]
                                                         
{'loss': 0.0969, 'learning_rate': 5e-05, 'epoch': 2.5}

 50%|█████     | 12210/24420 [5:05:10<5:11:16,  1.53s/it]
 50%|█████     | 12211/24420 [5:05:11<5:09:24,  1.52s/it]
 50%|█████     | 12212/24420 [5:05:13<5:07:38,  1.51s/it]
 50%|█████     | 12213/24420 [5:05:14<5:05:57,  1.50s/it]
 50%|█████     | 12214/24420 [5:05:16<5:04:46,  1.50s/it]
 50%|█████     | 12215/24420 [5:05:17<5:04:23,  1.50s/it]
                                                         
{'loss': 0.0762, 'learning_rate': 5e-05, 'epoch': 2.5}

 50%|█████     | 12215/24420 [5:05:17<5:04:23,  1.50s/it]
 50%|█████     | 12216/24420 [5:05:19<5:03:41,  1.49s/it]
 50%|█████     | 12217/24420 [5:05:20<5:03:38,  1.49s/it]
 50%|█████     | 12218/24420 [5:05:22<5:03:30,  1.49s/it]
 50%|█████     | 12219/24420 [5:05:23<5:02:40,  1.49s/it]
 50%|█████     | 12220/24420 [5:05:25<5:03:11,  1.49s/it]
                                                         
{'loss': 0.0813, 'learning_rate': 5e-05, 'epoch': 2.5}

 50%|█████     | 12220/24420 [5:05:25<5:03:11,  1.49s/it]
 50%|█████     | 12221/24420 [5:05:26<5:03:32,  1.49s/it]
 50%|█████     | 12222/24420 [5:05:28<5:03:22,  1.49s/it]
 50%|█████     | 12223/24420 [5:05:29<5:03:03,  1.49s/it]
 50%|█████     | 12224/24420 [5:05:31<5:02:49,  1.49s/it]
 50%|█████     | 12225/24420 [5:05:32<5:02:53,  1.49s/it]
                                                         
{'loss': 0.0871, 'learning_rate': 5e-05, 'epoch': 2.5}

 50%|█████     | 12225/24420 [5:05:32<5:02:53,  1.49s/it]
 50%|█████     | 12226/24420 [5:05:34<5:03:10,  1.49s/it]
 50%|█████     | 12227/24420 [5:05:35<5:02:39,  1.49s/it]
 50%|█████     | 12228/24420 [5:05:37<5:02:23,  1.49s/it]
 50%|█████     | 12229/24420 [5:05:38<5:02:15,  1.49s/it]
 50%|█████     | 12230/24420 [5:05:40<5:02:27,  1.49s/it]
                                                         
{'loss': 0.1068, 'learning_rate': 5e-05, 'epoch': 2.5}

 50%|█████     | 12230/24420 [5:05:40<5:02:27,  1.49s/it]
 50%|█████     | 12231/24420 [5:05:41<5:02:32,  1.49s/it]
 50%|█████     | 12232/24420 [5:05:43<5:02:35,  1.49s/it]
 50%|█████     | 12233/24420 [5:05:44<5:02:13,  1.49s/it]
 50%|█████     | 12234/24420 [5:05:46<5:01:43,  1.49s/it]
 50%|█████     | 12235/24420 [5:05:47<5:02:14,  1.49s/it]
                                                         
{'loss': 0.0973, 'learning_rate': 5e-05, 'epoch': 2.51}

 50%|█████     | 12235/24420 [5:05:47<5:02:14,  1.49s/it]
 50%|█████     | 12236/24420 [5:05:49<5:02:11,  1.49s/it]
 50%|█████     | 12237/24420 [5:05:50<5:02:17,  1.49s/it]
 50%|█████     | 12238/24420 [5:05:52<5:02:38,  1.49s/it]
 50%|█████     | 12239/24420 [5:05:53<5:02:15,  1.49s/it]
 50%|█████     | 12240/24420 [5:05:55<5:01:38,  1.49s/it]
                                                         
{'loss': 0.0756, 'learning_rate': 5e-05, 'epoch': 2.51}

 50%|█████     | 12240/24420 [5:05:55<5:01:38,  1.49s/it]
 50%|█████     | 12241/24420 [5:05:56<5:02:10,  1.49s/it]
 50%|█████     | 12242/24420 [5:05:58<5:02:30,  1.49s/it]
 50%|█████     | 12243/24420 [5:05:59<5:01:56,  1.49s/it]
 50%|█████     | 12244/24420 [5:06:01<5:02:01,  1.49s/it]
 50%|█████     | 12245/24420 [5:06:02<5:01:32,  1.49s/it]
                                                         
{'loss': 0.1005, 'learning_rate': 5e-05, 'epoch': 2.51}

 50%|█████     | 12245/24420 [5:06:02<5:01:32,  1.49s/it]
 50%|█████     | 12246/24420 [5:06:04<5:01:41,  1.49s/it]
 50%|█████     | 12247/24420 [5:06:05<5:02:13,  1.49s/it]
 50%|█████     | 12248/24420 [5:06:07<5:01:43,  1.49s/it]
 50%|█████     | 12249/24420 [5:06:08<5:01:47,  1.49s/it]
 50%|█████     | 12250/24420 [5:06:10<5:02:18,  1.49s/it]
                                                         
{'loss': 0.0877, 'learning_rate': 5e-05, 'epoch': 2.51}

 50%|█████     | 12250/24420 [5:06:10<5:02:18,  1.49s/it]
 50%|█████     | 12251/24420 [5:06:11<5:03:26,  1.50s/it]
 50%|█████     | 12252/24420 [5:06:13<5:02:42,  1.49s/it]
 50%|█████     | 12253/24420 [5:06:14<5:02:09,  1.49s/it]
 50%|█████     | 12254/24420 [5:06:16<5:02:00,  1.49s/it]
 50%|█████     | 12255/24420 [5:06:17<5:01:58,  1.49s/it]
                                                         
{'loss': 0.1, 'learning_rate': 5e-05, 'epoch': 2.51}

 50%|█████     | 12255/24420 [5:06:17<5:01:58,  1.49s/it]
 50%|█████     | 12256/24420 [5:06:18<5:02:04,  1.49s/it]
 50%|█████     | 12257/24420 [5:06:20<5:01:54,  1.49s/it]
 50%|█████     | 12258/24420 [5:06:21<5:02:05,  1.49s/it]
 50%|█████     | 12259/24420 [5:06:23<5:01:49,  1.49s/it]
 50%|█████     | 12260/24420 [5:06:24<5:02:05,  1.49s/it]
                                                         
{'loss': 0.0825, 'learning_rate': 5e-05, 'epoch': 2.51}

 50%|█████     | 12260/24420 [5:06:24<5:02:05,  1.49s/it]
 50%|█████     | 12261/24420 [5:06:26<5:02:13,  1.49s/it]
 50%|█████     | 12262/24420 [5:06:27<5:02:29,  1.49s/it]
 50%|█████     | 12263/24420 [5:06:29<5:01:57,  1.49s/it]
 50%|█████     | 12264/24420 [5:06:30<5:01:53,  1.49s/it]
 50%|█████     | 12265/24420 [5:06:32<5:01:34,  1.49s/it]
                                                         
{'loss': 0.0843, 'learning_rate': 5e-05, 'epoch': 2.51}

 50%|█████     | 12265/24420 [5:06:32<5:01:34,  1.49s/it]
 50%|█████     | 12266/24420 [5:06:33<5:02:08,  1.49s/it]
 50%|█████     | 12267/24420 [5:06:35<5:01:24,  1.49s/it]
 50%|█████     | 12268/24420 [5:06:36<5:01:34,  1.49s/it]
 50%|█████     | 12269/24420 [5:06:38<5:01:36,  1.49s/it]
 50%|█████     | 12270/24420 [5:06:39<5:01:19,  1.49s/it]
                                                         
{'loss': 0.0851, 'learning_rate': 5e-05, 'epoch': 2.51}

 50%|█████     | 12270/24420 [5:06:39<5:01:19,  1.49s/it]
 50%|█████     | 12271/24420 [5:06:41<5:02:00,  1.49s/it]
 50%|█████     | 12272/24420 [5:06:42<5:02:25,  1.49s/it]
 50%|█████     | 12273/24420 [5:06:44<5:02:42,  1.50s/it]
 50%|█████     | 12274/24420 [5:06:45<5:03:56,  1.50s/it]
 50%|█████     | 12275/24420 [5:06:47<5:03:16,  1.50s/it]
                                                         
{'loss': 0.1012, 'learning_rate': 5e-05, 'epoch': 2.51}

 50%|█████     | 12275/24420 [5:06:47<5:03:16,  1.50s/it]
 50%|█████     | 12276/24420 [5:06:48<5:03:24,  1.50s/it]
 50%|█████     | 12277/24420 [5:06:50<5:03:33,  1.50s/it]
 50%|█████     | 12278/24420 [5:06:51<5:02:32,  1.50s/it]
 50%|█████     | 12279/24420 [5:06:53<5:02:14,  1.49s/it]
 50%|█████     | 12280/24420 [5:06:54<5:01:35,  1.49s/it]
                                                         
{'loss': 0.0866, 'learning_rate': 5e-05, 'epoch': 2.51}

 50%|█████     | 12280/24420 [5:06:54<5:01:35,  1.49s/it]
 50%|█████     | 12281/24420 [5:06:56<5:09:04,  1.53s/it]
 50%|█████     | 12282/24420 [5:06:57<5:06:00,  1.51s/it]
 50%|█████     | 12283/24420 [5:06:59<5:04:36,  1.51s/it]
 50%|█████     | 12284/24420 [5:07:00<5:03:08,  1.50s/it]
 50%|█████     | 12285/24420 [5:07:02<5:01:56,  1.49s/it]
                                                         
{'loss': 0.0851, 'learning_rate': 5e-05, 'epoch': 2.52}

 50%|█████     | 12285/24420 [5:07:02<5:01:56,  1.49s/it]
 50%|█████     | 12286/24420 [5:07:03<5:01:43,  1.49s/it]
 50%|█████     | 12287/24420 [5:07:05<5:01:31,  1.49s/it]
 50%|█████     | 12288/24420 [5:07:06<5:02:02,  1.49s/it]
 50%|█████     | 12289/24420 [5:07:08<5:02:02,  1.49s/it]
 50%|█████     | 12290/24420 [5:07:09<5:01:48,  1.49s/it]
                                                         
{'loss': 0.0781, 'learning_rate': 5e-05, 'epoch': 2.52}

 50%|█████     | 12290/24420 [5:07:09<5:01:48,  1.49s/it]
 50%|█████     | 12291/24420 [5:07:11<5:02:13,  1.50s/it]
 50%|█████     | 12292/24420 [5:07:12<5:01:51,  1.49s/it]
 50%|█████     | 12293/24420 [5:07:14<5:00:57,  1.49s/it]
 50%|█████     | 12294/24420 [5:07:15<5:00:38,  1.49s/it]
 50%|█████     | 12295/24420 [5:07:17<5:00:35,  1.49s/it]
                                                         
{'loss': 0.0842, 'learning_rate': 5e-05, 'epoch': 2.52}

 50%|█████     | 12295/24420 [5:07:17<5:00:35,  1.49s/it]
 50%|█████     | 12296/24420 [5:07:18<5:00:34,  1.49s/it]
 50%|█████     | 12297/24420 [5:07:20<5:00:49,  1.49s/it]
 50%|█████     | 12298/24420 [5:07:21<5:00:32,  1.49s/it]
 50%|█████     | 12299/24420 [5:07:23<5:00:28,  1.49s/it]
 50%|█████     | 12300/24420 [5:07:24<5:00:24,  1.49s/it]
                                                         
{'loss': 0.0913, 'learning_rate': 5e-05, 'epoch': 2.52}

 50%|█████     | 12300/24420 [5:07:24<5:00:24,  1.49s/it]
 50%|█████     | 12301/24420 [5:07:26<5:01:37,  1.49s/it]
 50%|█████     | 12302/24420 [5:07:27<5:02:28,  1.50s/it]
 50%|█████     | 12303/24420 [5:07:29<5:01:32,  1.49s/it]
 50%|█████     | 12304/24420 [5:07:30<5:01:02,  1.49s/it]
 50%|█████     | 12305/24420 [5:07:32<5:01:28,  1.49s/it]
                                                         
{'loss': 0.0824, 'learning_rate': 5e-05, 'epoch': 2.52}

 50%|█████     | 12305/24420 [5:07:32<5:01:28,  1.49s/it]
 50%|█████     | 12306/24420 [5:07:33<5:02:09,  1.50s/it]
 50%|█████     | 12307/24420 [5:07:35<5:01:52,  1.50s/it]
 50%|█████     | 12308/24420 [5:07:36<5:02:13,  1.50s/it]
 50%|█████     | 12309/24420 [5:07:38<5:02:18,  1.50s/it]
 50%|█████     | 12310/24420 [5:07:39<5:01:37,  1.49s/it]
                                                         
{'loss': 0.103, 'learning_rate': 5e-05, 'epoch': 2.52}

 50%|█████     | 12310/24420 [5:07:39<5:01:37,  1.49s/it]
 50%|█████     | 12311/24420 [5:07:41<5:02:27,  1.50s/it]
 50%|█████     | 12312/24420 [5:07:42<5:01:57,  1.50s/it]
 50%|█████     | 12313/24420 [5:07:44<5:00:59,  1.49s/it]
 50%|█████     | 12314/24420 [5:07:45<5:00:48,  1.49s/it]
 50%|█████     | 12315/24420 [5:07:47<5:00:02,  1.49s/it]
                                                         
{'loss': 0.0824, 'learning_rate': 5e-05, 'epoch': 2.52}

 50%|█████     | 12315/24420 [5:07:47<5:00:02,  1.49s/it]
 50%|█████     | 12316/24420 [5:07:48<5:00:49,  1.49s/it]
 50%|█████     | 12317/24420 [5:07:50<5:00:23,  1.49s/it]
 50%|█████     | 12318/24420 [5:07:51<5:00:21,  1.49s/it]
 50%|█████     | 12319/24420 [5:07:53<5:00:36,  1.49s/it]
 50%|█████     | 12320/24420 [5:07:54<4:59:56,  1.49s/it]
                                                         
{'loss': 0.0993, 'learning_rate': 5e-05, 'epoch': 2.52}

 50%|█████     | 12320/24420 [5:07:54<4:59:56,  1.49s/it]
 50%|█████     | 12321/24420 [5:07:56<5:00:05,  1.49s/it]
 50%|█████     | 12322/24420 [5:07:57<5:00:03,  1.49s/it]
 50%|█████     | 12323/24420 [5:07:59<4:59:38,  1.49s/it]
 50%|█████     | 12324/24420 [5:08:00<4:59:40,  1.49s/it]
 50%|█████     | 12325/24420 [5:08:01<4:59:49,  1.49s/it]
                                                         
{'loss': 0.0807, 'learning_rate': 5e-05, 'epoch': 2.52}

 50%|█████     | 12325/24420 [5:08:01<4:59:49,  1.49s/it]
 50%|█████     | 12326/24420 [5:08:03<5:00:13,  1.49s/it]
 50%|█████     | 12327/24420 [5:08:04<4:59:57,  1.49s/it]
 50%|█████     | 12328/24420 [5:08:06<4:59:47,  1.49s/it]
 50%|█████     | 12329/24420 [5:08:07<4:59:48,  1.49s/it]
 50%|█████     | 12330/24420 [5:08:09<4:59:39,  1.49s/it]
                                                         
{'loss': 0.0773, 'learning_rate': 5e-05, 'epoch': 2.52}

 50%|█████     | 12330/24420 [5:08:09<4:59:39,  1.49s/it]
 50%|█████     | 12331/24420 [5:08:10<5:00:04,  1.49s/it]
 50%|█████     | 12332/24420 [5:08:12<5:00:09,  1.49s/it]
 51%|█████     | 12333/24420 [5:08:13<5:00:11,  1.49s/it]
 51%|█████     | 12334/24420 [5:08:15<4:59:58,  1.49s/it]
 51%|█████     | 12335/24420 [5:08:16<4:59:30,  1.49s/it]
                                                         
{'loss': 0.078, 'learning_rate': 5e-05, 'epoch': 2.53}

 51%|█████     | 12335/24420 [5:08:16<4:59:30,  1.49s/it]
 51%|█████     | 12336/24420 [5:08:18<4:59:43,  1.49s/it]
 51%|█████     | 12337/24420 [5:08:19<5:00:04,  1.49s/it]
 51%|█████     | 12338/24420 [5:08:21<5:00:31,  1.49s/it]
 51%|█████     | 12339/24420 [5:08:22<5:00:10,  1.49s/it]
 51%|█████     | 12340/24420 [5:08:24<5:00:02,  1.49s/it]
                                                         
{'loss': 0.0864, 'learning_rate': 5e-05, 'epoch': 2.53}

 51%|█████     | 12340/24420 [5:08:24<5:00:02,  1.49s/it]
 51%|█████     | 12341/24420 [5:08:25<5:00:14,  1.49s/it]
 51%|█████     | 12342/24420 [5:08:27<5:00:37,  1.49s/it]
 51%|█████     | 12343/24420 [5:08:28<5:00:49,  1.49s/it]
 51%|█████     | 12344/24420 [5:08:30<4:59:54,  1.49s/it]
 51%|█████     | 12345/24420 [5:08:31<5:00:17,  1.49s/it]
                                                         
{'loss': 0.0853, 'learning_rate': 5e-05, 'epoch': 2.53}

 51%|█████     | 12345/24420 [5:08:31<5:00:17,  1.49s/it]
 51%|█████     | 12346/24420 [5:08:33<5:00:21,  1.49s/it]
 51%|█████     | 12347/24420 [5:08:34<4:59:26,  1.49s/it]
 51%|█████     | 12348/24420 [5:08:36<4:59:12,  1.49s/it]
 51%|█████     | 12349/24420 [5:08:37<4:59:09,  1.49s/it]
 51%|█████     | 12350/24420 [5:08:39<4:59:11,  1.49s/it]
                                                         
{'loss': 0.08, 'learning_rate': 5e-05, 'epoch': 2.53}

 51%|█████     | 12350/24420 [5:08:39<4:59:11,  1.49s/it]
 51%|█████     | 12351/24420 [5:08:40<4:59:02,  1.49s/it]
 51%|█████     | 12352/24420 [5:08:42<4:58:55,  1.49s/it]
 51%|█████     | 12353/24420 [5:08:43<4:58:24,  1.48s/it]
 51%|█████     | 12354/24420 [5:08:45<4:58:32,  1.48s/it]
 51%|█████     | 12355/24420 [5:08:46<4:58:04,  1.48s/it]
                                                         
{'loss': 0.0835, 'learning_rate': 5e-05, 'epoch': 2.53}

 51%|█████     | 12355/24420 [5:08:46<4:58:04,  1.48s/it]
 51%|█████     | 12356/24420 [5:08:48<4:58:10,  1.48s/it]
 51%|█████     | 12357/24420 [5:08:49<4:58:15,  1.48s/it]
 51%|█████     | 12358/24420 [5:08:51<4:58:37,  1.49s/it]
 51%|█████     | 12359/24420 [5:08:52<4:59:18,  1.49s/it]
 51%|█████     | 12360/24420 [5:08:54<4:59:21,  1.49s/it]
                                                         
{'loss': 0.0874, 'learning_rate': 5e-05, 'epoch': 2.53}

 51%|█████     | 12360/24420 [5:08:54<4:59:21,  1.49s/it]
 51%|█████     | 12361/24420 [5:08:55<4:59:31,  1.49s/it]
 51%|█████     | 12362/24420 [5:08:57<4:59:24,  1.49s/it]
 51%|█████     | 12363/24420 [5:08:58<4:58:44,  1.49s/it]
 51%|█████     | 12364/24420 [5:09:00<4:58:53,  1.49s/it]
 51%|█████     | 12365/24420 [5:09:01<4:58:25,  1.49s/it]
                                                         
{'loss': 0.104, 'learning_rate': 5e-05, 'epoch': 2.53}

 51%|█████     | 12365/24420 [5:09:01<4:58:25,  1.49s/it]
 51%|█████     | 12366/24420 [5:09:02<4:57:42,  1.48s/it]
 51%|█████     | 12367/24420 [5:09:04<4:58:21,  1.49s/it]
 51%|█████     | 12368/24420 [5:09:05<4:58:18,  1.49s/it]
 51%|█████     | 12369/24420 [5:09:07<4:58:28,  1.49s/it]
 51%|█████     | 12370/24420 [5:09:08<4:58:21,  1.49s/it]
                                                         
{'loss': 0.0915, 'learning_rate': 5e-05, 'epoch': 2.53}

 51%|█████     | 12370/24420 [5:09:08<4:58:21,  1.49s/it]
 51%|█████     | 12371/24420 [5:09:10<4:58:32,  1.49s/it]
 51%|█████     | 12372/24420 [5:09:11<4:58:17,  1.49s/it]
 51%|█████     | 12373/24420 [5:09:13<4:58:22,  1.49s/it]
 51%|█████     | 12374/24420 [5:09:14<4:58:12,  1.49s/it]
 51%|█████     | 12375/24420 [5:09:16<4:58:10,  1.49s/it]
                                                         
{'loss': 0.1075, 'learning_rate': 5e-05, 'epoch': 2.53}

 51%|█████     | 12375/24420 [5:09:16<4:58:10,  1.49s/it]
 51%|█████     | 12376/24420 [5:09:17<4:58:29,  1.49s/it]
 51%|█████     | 12377/24420 [5:09:19<4:58:11,  1.49s/it]
 51%|█████     | 12378/24420 [5:09:20<4:58:15,  1.49s/it]
 51%|█████     | 12379/24420 [5:09:22<4:58:02,  1.49s/it]
 51%|█████     | 12380/24420 [5:09:23<4:57:48,  1.48s/it]
                                                         
{'loss': 0.0877, 'learning_rate': 5e-05, 'epoch': 2.53}

 51%|█████     | 12380/24420 [5:09:23<4:57:48,  1.48s/it]
 51%|█████     | 12381/24420 [5:09:25<4:57:51,  1.48s/it]
 51%|█████     | 12382/24420 [5:09:26<4:57:59,  1.49s/it]
 51%|█████     | 12383/24420 [5:09:28<4:57:47,  1.48s/it]
 51%|█████     | 12384/24420 [5:09:29<4:57:23,  1.48s/it]
 51%|█████     | 12385/24420 [5:09:31<4:57:22,  1.48s/it]
                                                         
{'loss': 0.1151, 'learning_rate': 5e-05, 'epoch': 2.54}

 51%|█████     | 12385/24420 [5:09:31<4:57:22,  1.48s/it]
 51%|█████     | 12386/24420 [5:09:32<4:57:59,  1.49s/it]
 51%|█████     | 12387/24420 [5:09:34<4:58:31,  1.49s/it]
 51%|█████     | 12388/24420 [5:09:35<4:58:18,  1.49s/it]
 51%|█████     | 12389/24420 [5:09:37<4:57:42,  1.48s/it]
 51%|█████     | 12390/24420 [5:09:38<4:57:35,  1.48s/it]
                                                         
{'loss': 0.0895, 'learning_rate': 5e-05, 'epoch': 2.54}

 51%|█████     | 12390/24420 [5:09:38<4:57:35,  1.48s/it]
 51%|█████     | 12391/24420 [5:09:40<4:57:41,  1.48s/it]
 51%|█████     | 12392/24420 [5:09:41<4:57:26,  1.48s/it]
 51%|█████     | 12393/24420 [5:09:43<4:58:24,  1.49s/it]
 51%|█████     | 12394/24420 [5:09:44<4:57:42,  1.49s/it]
 51%|█████     | 12395/24420 [5:09:46<4:57:50,  1.49s/it]
                                                         
{'loss': 0.071, 'learning_rate': 5e-05, 'epoch': 2.54}

 51%|█████     | 12395/24420 [5:09:46<4:57:50,  1.49s/it]
 51%|█████     | 12396/24420 [5:09:47<4:57:59,  1.49s/it]
 51%|█████     | 12397/24420 [5:09:49<4:57:42,  1.49s/it]
 51%|█████     | 12398/24420 [5:09:50<4:57:46,  1.49s/it]
 51%|█████     | 12399/24420 [5:09:52<4:57:14,  1.48s/it]
 51%|█████     | 12400/24420 [5:09:53<4:57:00,  1.48s/it]
                                                         
{'loss': 0.0838, 'learning_rate': 5e-05, 'epoch': 2.54}

 51%|█████     | 12400/24420 [5:09:53<4:57:00,  1.48s/it]
 51%|█████     | 12401/24420 [5:09:58<8:12:48,  2.46s/it]
 51%|█████     | 12402/24420 [5:09:59<7:13:49,  2.17s/it]
 51%|█████     | 12403/24420 [5:10:01<6:32:34,  1.96s/it]
 51%|█████     | 12404/24420 [5:10:02<6:03:40,  1.82s/it]
 51%|█████     | 12405/24420 [5:10:04<5:44:04,  1.72s/it]
                                                         
{'loss': 0.0797, 'learning_rate': 5e-05, 'epoch': 2.54}

 51%|█████     | 12405/24420 [5:10:04<5:44:04,  1.72s/it]
 51%|█████     | 12406/24420 [5:10:05<5:29:56,  1.65s/it]
 51%|█████     | 12407/24420 [5:10:07<5:20:29,  1.60s/it]
 51%|█████     | 12408/24420 [5:10:08<5:13:24,  1.57s/it]
 51%|█████     | 12409/24420 [5:10:10<5:08:11,  1.54s/it]
 51%|█████     | 12410/24420 [5:10:11<5:04:30,  1.52s/it]
                                                         
{'loss': 0.0956, 'learning_rate': 5e-05, 'epoch': 2.54}

 51%|█████     | 12410/24420 [5:10:11<5:04:30,  1.52s/it]
 51%|█████     | 12411/24420 [5:10:13<5:02:05,  1.51s/it]
 51%|█████     | 12412/24420 [5:10:14<5:00:32,  1.50s/it]
 51%|█████     | 12413/24420 [5:10:16<4:59:33,  1.50s/it]
 51%|█████     | 12414/24420 [5:10:17<4:58:49,  1.49s/it]
 51%|█████     | 12415/24420 [5:10:19<4:58:52,  1.49s/it]
                                                         
{'loss': 0.0931, 'learning_rate': 5e-05, 'epoch': 2.54}

 51%|█████     | 12415/24420 [5:10:19<4:58:52,  1.49s/it]
 51%|█████     | 12416/24420 [5:10:20<4:58:17,  1.49s/it]
 51%|█████     | 12417/24420 [5:10:21<4:58:43,  1.49s/it]
 51%|█████     | 12418/24420 [5:10:23<4:57:59,  1.49s/it]
 51%|█████     | 12419/24420 [5:10:24<4:58:18,  1.49s/it]
 51%|█████     | 12420/24420 [5:10:26<4:57:41,  1.49s/it]
                                                         
{'loss': 0.0893, 'learning_rate': 5e-05, 'epoch': 2.54}

 51%|█████     | 12420/24420 [5:10:26<4:57:41,  1.49s/it]
 51%|█████     | 12421/24420 [5:10:27<4:57:43,  1.49s/it]
 51%|█████     | 12422/24420 [5:10:29<4:57:30,  1.49s/it]
 51%|█████     | 12423/24420 [5:10:30<4:56:56,  1.49s/it]
 51%|█████     | 12424/24420 [5:10:32<4:57:04,  1.49s/it]
 51%|█████     | 12425/24420 [5:10:33<4:56:54,  1.49s/it]
                                                         
{'loss': 0.0854, 'learning_rate': 5e-05, 'epoch': 2.54}

 51%|█████     | 12425/24420 [5:10:33<4:56:54,  1.49s/it]
 51%|█████     | 12426/24420 [5:10:35<4:57:11,  1.49s/it]
 51%|█████     | 12427/24420 [5:10:36<4:56:58,  1.49s/it]
 51%|█████     | 12428/24420 [5:10:38<4:59:00,  1.50s/it]
 51%|█████     | 12429/24420 [5:10:39<4:57:57,  1.49s/it]
 51%|█████     | 12430/24420 [5:10:41<4:57:42,  1.49s/it]
                                                         
{'loss': 0.0928, 'learning_rate': 5e-05, 'epoch': 2.55}

 51%|█████     | 12430/24420 [5:10:41<4:57:42,  1.49s/it]
 51%|█████     | 12431/24420 [5:10:42<4:57:31,  1.49s/it]
 51%|█████     | 12432/24420 [5:10:44<4:57:00,  1.49s/it]
 51%|█████     | 12433/24420 [5:10:45<4:57:10,  1.49s/it]
 51%|█████     | 12434/24420 [5:10:47<4:56:34,  1.48s/it]
 51%|█████     | 12435/24420 [5:10:48<4:56:57,  1.49s/it]
                                                         
{'loss': 0.0901, 'learning_rate': 5e-05, 'epoch': 2.55}

 51%|█████     | 12435/24420 [5:10:48<4:56:57,  1.49s/it]
 51%|█████     | 12436/24420 [5:10:50<4:57:02,  1.49s/it]
 51%|█████     | 12437/24420 [5:10:51<4:56:48,  1.49s/it]
 51%|█████     | 12438/24420 [5:10:53<4:56:28,  1.48s/it]
 51%|█████     | 12439/24420 [5:10:54<4:56:40,  1.49s/it]
 51%|█████     | 12440/24420 [5:10:56<4:56:27,  1.48s/it]
                                                         
{'loss': 0.0738, 'learning_rate': 5e-05, 'epoch': 2.55}

 51%|█████     | 12440/24420 [5:10:56<4:56:27,  1.48s/it]
 51%|█████     | 12441/24420 [5:10:57<4:56:04,  1.48s/it]
 51%|█████     | 12442/24420 [5:10:59<4:56:09,  1.48s/it]
 51%|█████     | 12443/24420 [5:11:00<4:56:39,  1.49s/it]
 51%|█████     | 12444/24420 [5:11:02<4:56:23,  1.48s/it]
 51%|█████     | 12445/24420 [5:11:03<4:59:35,  1.50s/it]
                                                         
{'loss': 0.0908, 'learning_rate': 5e-05, 'epoch': 2.55}

 51%|█████     | 12445/24420 [5:11:03<4:59:35,  1.50s/it]
 51%|█████     | 12446/24420 [5:11:05<4:59:01,  1.50s/it]
 51%|█████     | 12447/24420 [5:11:06<4:58:39,  1.50s/it]
 51%|█████     | 12448/24420 [5:11:08<4:57:27,  1.49s/it]
 51%|█████     | 12449/24420 [5:11:09<4:57:17,  1.49s/it]
 51%|█████     | 12450/24420 [5:11:11<4:56:57,  1.49s/it]
                                                         
{'loss': 0.0935, 'learning_rate': 5e-05, 'epoch': 2.55}

 51%|█████     | 12450/24420 [5:11:11<4:56:57,  1.49s/it]
 51%|█████     | 12451/24420 [5:11:12<4:56:56,  1.49s/it]
 51%|█████     | 12452/24420 [5:11:14<4:56:24,  1.49s/it]
 51%|█████     | 12453/24420 [5:11:15<4:57:24,  1.49s/it]
 51%|█████     | 12454/24420 [5:11:17<4:57:35,  1.49s/it]
 51%|█████     | 12455/24420 [5:11:18<4:57:00,  1.49s/it]
                                                         
{'loss': 0.0873, 'learning_rate': 5e-05, 'epoch': 2.55}

 51%|█████     | 12455/24420 [5:11:18<4:57:00,  1.49s/it]
 51%|█████     | 12456/24420 [5:11:20<4:56:24,  1.49s/it]
 51%|█████     | 12457/24420 [5:11:21<4:56:17,  1.49s/it]
 51%|█████     | 12458/24420 [5:11:22<4:56:07,  1.49s/it]
 51%|█████     | 12459/24420 [5:11:24<4:55:48,  1.48s/it]
 51%|█████     | 12460/24420 [5:11:25<4:56:09,  1.49s/it]
                                                         
{'loss': 0.0883, 'learning_rate': 5e-05, 'epoch': 2.55}

 51%|█████     | 12460/24420 [5:11:25<4:56:09,  1.49s/it]
 51%|█████     | 12461/24420 [5:11:27<4:56:15,  1.49s/it]
 51%|█████     | 12462/24420 [5:11:28<4:56:24,  1.49s/it]
 51%|█████     | 12463/24420 [5:11:30<4:56:23,  1.49s/it]
 51%|█████     | 12464/24420 [5:11:31<4:56:21,  1.49s/it]
 51%|█████     | 12465/24420 [5:11:33<4:56:28,  1.49s/it]
                                                         
{'loss': 0.0956, 'learning_rate': 5e-05, 'epoch': 2.55}

 51%|█████     | 12465/24420 [5:11:33<4:56:28,  1.49s/it]
 51%|█████     | 12466/24420 [5:11:34<4:56:24,  1.49s/it]
 51%|█████     | 12467/24420 [5:11:36<4:56:14,  1.49s/it]
 51%|█████     | 12468/24420 [5:11:37<4:56:37,  1.49s/it]
 51%|█████     | 12469/24420 [5:11:39<4:56:13,  1.49s/it]
 51%|█████     | 12470/24420 [5:11:40<4:56:15,  1.49s/it]
                                                         
{'loss': 0.0894, 'learning_rate': 5e-05, 'epoch': 2.55}

 51%|█████     | 12470/24420 [5:11:40<4:56:15,  1.49s/it]
 51%|█████     | 12471/24420 [5:11:42<4:56:02,  1.49s/it]
 51%|█████     | 12472/24420 [5:11:43<4:55:34,  1.48s/it]
 51%|█████     | 12473/24420 [5:11:45<4:56:16,  1.49s/it]
 51%|█████     | 12474/24420 [5:11:46<4:55:29,  1.48s/it]
 51%|█████     | 12475/24420 [5:11:48<4:55:42,  1.49s/it]
                                                         
{'loss': 0.0971, 'learning_rate': 5e-05, 'epoch': 2.55}

 51%|█████     | 12475/24420 [5:11:48<4:55:42,  1.49s/it]
 51%|█████     | 12476/24420 [5:11:49<4:55:51,  1.49s/it]
 51%|█████     | 12477/24420 [5:11:51<4:55:43,  1.49s/it]
 51%|█████     | 12478/24420 [5:11:52<4:55:45,  1.49s/it]
 51%|█████     | 12479/24420 [5:11:54<4:55:43,  1.49s/it]
 51%|█████     | 12480/24420 [5:11:55<4:55:41,  1.49s/it]
                                                         
{'loss': 0.0879, 'learning_rate': 5e-05, 'epoch': 2.56}

 51%|█████     | 12480/24420 [5:11:55<4:55:41,  1.49s/it]
 51%|█████     | 12481/24420 [5:11:57<4:56:15,  1.49s/it]
 51%|█████     | 12482/24420 [5:11:58<4:55:16,  1.48s/it]
 51%|█████     | 12483/24420 [5:12:00<4:55:11,  1.48s/it]
 51%|█████     | 12484/24420 [5:12:01<4:56:16,  1.49s/it]
 51%|█████     | 12485/24420 [5:12:03<4:56:44,  1.49s/it]
                                                         
{'loss': 0.0765, 'learning_rate': 5e-05, 'epoch': 2.56}

 51%|█████     | 12485/24420 [5:12:03<4:56:44,  1.49s/it]
 51%|█████     | 12486/24420 [5:12:04<4:56:41,  1.49s/it]
 51%|█████     | 12487/24420 [5:12:06<4:55:54,  1.49s/it]
 51%|█████     | 12488/24420 [5:12:07<4:55:48,  1.49s/it]
 51%|█████     | 12489/24420 [5:12:09<4:55:24,  1.49s/it]
 51%|█████     | 12490/24420 [5:12:10<4:55:52,  1.49s/it]
                                                         
{'loss': 0.1108, 'learning_rate': 5e-05, 'epoch': 2.56}

 51%|█████     | 12490/24420 [5:12:10<4:55:52,  1.49s/it]
 51%|█████     | 12491/24420 [5:12:12<4:55:13,  1.48s/it]
 51%|█████     | 12492/24420 [5:12:13<4:55:37,  1.49s/it]
 51%|█████     | 12493/24420 [5:12:15<4:55:55,  1.49s/it]
 51%|█████     | 12494/24420 [5:12:16<4:56:08,  1.49s/it]
 51%|█████     | 12495/24420 [5:12:18<4:56:21,  1.49s/it]
                                                         
{'loss': 0.0931, 'learning_rate': 5e-05, 'epoch': 2.56}

 51%|█████     | 12495/24420 [5:12:18<4:56:21,  1.49s/it]
 51%|█████     | 12496/24420 [5:12:19<4:55:28,  1.49s/it]
 51%|█████     | 12497/24420 [5:12:20<4:55:36,  1.49s/it]
 51%|█████     | 12498/24420 [5:12:22<4:55:49,  1.49s/it]
 51%|█████     | 12499/24420 [5:12:23<4:55:24,  1.49s/it]
 51%|█████     | 12500/24420 [5:12:25<4:55:45,  1.49s/it]
                                                         
{'loss': 0.0722, 'learning_rate': 5e-05, 'epoch': 2.56}

 51%|█████     | 12500/24420 [5:12:25<4:55:45,  1.49s/it]
 51%|█████     | 12501/24420 [5:12:26<4:55:51,  1.49s/it]
 51%|█████     | 12502/24420 [5:12:28<4:55:35,  1.49s/it]
 51%|█████     | 12503/24420 [5:12:29<4:55:49,  1.49s/it]
 51%|█████     | 12504/24420 [5:12:31<4:55:00,  1.49s/it]
 51%|█████     | 12505/24420 [5:12:32<4:55:16,  1.49s/it]
                                                         
{'loss': 0.0821, 'learning_rate': 5e-05, 'epoch': 2.56}

 51%|█████     | 12505/24420 [5:12:32<4:55:16,  1.49s/it]
 51%|█████     | 12506/24420 [5:12:34<4:55:22,  1.49s/it]
 51%|█████     | 12507/24420 [5:12:35<4:55:10,  1.49s/it]
 51%|█████     | 12508/24420 [5:12:37<4:55:40,  1.49s/it]
 51%|█████     | 12509/24420 [5:12:38<4:55:05,  1.49s/it]
 51%|█████     | 12510/24420 [5:12:40<4:54:45,  1.48s/it]
                                                         
{'loss': 0.1166, 'learning_rate': 5e-05, 'epoch': 2.56}

 51%|█████     | 12510/24420 [5:12:40<4:54:45,  1.48s/it]
 51%|█████     | 12511/24420 [5:12:41<4:54:59,  1.49s/it]
 51%|█████     | 12512/24420 [5:12:43<4:54:46,  1.49s/it]
 51%|█████     | 12513/24420 [5:12:44<4:54:43,  1.49s/it]
 51%|█████     | 12514/24420 [5:12:46<4:54:50,  1.49s/it]
 51%|█████     | 12515/24420 [5:12:47<4:54:25,  1.48s/it]
                                                         
{'loss': 0.0963, 'learning_rate': 5e-05, 'epoch': 2.56}

 51%|█████     | 12515/24420 [5:12:47<4:54:25,  1.48s/it]
 51%|█████▏    | 12516/24420 [5:12:49<4:54:47,  1.49s/it]
 51%|█████▏    | 12517/24420 [5:12:50<4:55:27,  1.49s/it]
 51%|█████▏    | 12518/24420 [5:12:52<4:55:25,  1.49s/it]
 51%|█████▏    | 12519/24420 [5:12:53<4:55:32,  1.49s/it]
 51%|█████▏    | 12520/24420 [5:12:55<4:56:03,  1.49s/it]
                                                         
{'loss': 0.0878, 'learning_rate': 5e-05, 'epoch': 2.56}

 51%|█████▏    | 12520/24420 [5:12:55<4:56:03,  1.49s/it]
 51%|█████▏    | 12521/24420 [5:12:56<4:56:26,  1.49s/it]
 51%|█████▏    | 12522/24420 [5:12:58<4:56:47,  1.50s/it]
 51%|█████▏    | 12523/24420 [5:12:59<4:56:58,  1.50s/it]
 51%|█████▏    | 12524/24420 [5:13:01<4:56:57,  1.50s/it]
 51%|█████▏    | 12525/24420 [5:13:02<4:56:27,  1.50s/it]
                                                         
{'loss': 0.0971, 'learning_rate': 5e-05, 'epoch': 2.56}

 51%|█████▏    | 12525/24420 [5:13:02<4:56:27,  1.50s/it]
 51%|█████▏    | 12526/24420 [5:13:04<4:56:06,  1.49s/it]
 51%|█████▏    | 12527/24420 [5:13:05<4:55:13,  1.49s/it]
 51%|█████▏    | 12528/24420 [5:13:07<4:54:36,  1.49s/it]
 51%|█████▏    | 12529/24420 [5:13:08<4:54:56,  1.49s/it]
 51%|█████▏    | 12530/24420 [5:13:10<4:54:37,  1.49s/it]
                                                         
{'loss': 0.079, 'learning_rate': 5e-05, 'epoch': 2.57}

 51%|█████▏    | 12530/24420 [5:13:10<4:54:37,  1.49s/it]
 51%|█████▏    | 12531/24420 [5:13:11<4:54:21,  1.49s/it]
 51%|█████▏    | 12532/24420 [5:13:13<4:54:33,  1.49s/it]
 51%|█████▏    | 12533/24420 [5:13:14<4:54:04,  1.48s/it]
 51%|█████▏    | 12534/24420 [5:13:16<4:53:38,  1.48s/it]
 51%|█████▏    | 12535/24420 [5:13:17<4:53:37,  1.48s/it]
                                                         
{'loss': 0.0713, 'learning_rate': 5e-05, 'epoch': 2.57}

 51%|█████▏    | 12535/24420 [5:13:17<4:53:37,  1.48s/it]
 51%|█████▏    | 12536/24420 [5:13:19<4:53:21,  1.48s/it]
 51%|█████▏    | 12537/24420 [5:13:20<4:53:54,  1.48s/it]
 51%|█████▏    | 12538/24420 [5:13:22<4:54:01,  1.48s/it]
 51%|█████▏    | 12539/24420 [5:13:23<4:54:00,  1.48s/it]
 51%|█████▏    | 12540/24420 [5:13:24<4:54:01,  1.48s/it]
                                                         
{'loss': 0.084, 'learning_rate': 5e-05, 'epoch': 2.57}

 51%|█████▏    | 12540/24420 [5:13:24<4:54:01,  1.48s/it]
 51%|█████▏    | 12541/24420 [5:13:26<4:54:32,  1.49s/it]
 51%|█████▏    | 12542/24420 [5:13:27<4:55:14,  1.49s/it]
 51%|█████▏    | 12543/24420 [5:13:29<4:55:11,  1.49s/it]
 51%|█████▏    | 12544/24420 [5:13:30<4:54:55,  1.49s/it]
 51%|█████▏    | 12545/24420 [5:13:32<4:54:58,  1.49s/it]
                                                         
{'loss': 0.0876, 'learning_rate': 5e-05, 'epoch': 2.57}

 51%|█████▏    | 12545/24420 [5:13:32<4:54:58,  1.49s/it]
 51%|█████▏    | 12546/24420 [5:13:33<4:55:20,  1.49s/it]
 51%|█████▏    | 12547/24420 [5:13:35<4:54:47,  1.49s/it]
 51%|█████▏    | 12548/24420 [5:13:36<4:54:49,  1.49s/it]
 51%|█████▏    | 12549/24420 [5:13:38<4:54:08,  1.49s/it]
 51%|█████▏    | 12550/24420 [5:13:39<4:54:14,  1.49s/it]
                                                         
{'loss': 0.0796, 'learning_rate': 5e-05, 'epoch': 2.57}

 51%|█████▏    | 12550/24420 [5:13:39<4:54:14,  1.49s/it]
 51%|█████▏    | 12551/24420 [5:13:41<4:54:37,  1.49s/it]
 51%|█████▏    | 12552/24420 [5:13:42<4:54:31,  1.49s/it]
 51%|█████▏    | 12553/24420 [5:13:44<4:54:20,  1.49s/it]
 51%|█████▏    | 12554/24420 [5:13:45<4:54:26,  1.49s/it]
 51%|█████▏    | 12555/24420 [5:13:47<4:54:18,  1.49s/it]
                                                         
{'loss': 0.0906, 'learning_rate': 5e-05, 'epoch': 2.57}

 51%|█████▏    | 12555/24420 [5:13:47<4:54:18,  1.49s/it]
 51%|█████▏    | 12556/24420 [5:13:48<4:54:06,  1.49s/it]
 51%|█████▏    | 12557/24420 [5:13:50<4:54:01,  1.49s/it]
 51%|█████▏    | 12558/24420 [5:13:51<4:54:19,  1.49s/it]
 51%|█████▏    | 12559/24420 [5:13:53<4:54:00,  1.49s/it]
 51%|█████▏    | 12560/24420 [5:13:54<4:53:43,  1.49s/it]
                                                         
{'loss': 0.0931, 'learning_rate': 5e-05, 'epoch': 2.57}

 51%|█████▏    | 12560/24420 [5:13:54<4:53:43,  1.49s/it]
 51%|█████▏    | 12561/24420 [5:13:56<4:53:58,  1.49s/it]
 51%|█████▏    | 12562/24420 [5:13:57<4:53:32,  1.49s/it]
 51%|█████▏    | 12563/24420 [5:13:59<4:53:27,  1.49s/it]
 51%|█████▏    | 12564/24420 [5:14:00<4:53:25,  1.48s/it]
 51%|█████▏    | 12565/24420 [5:14:02<4:54:05,  1.49s/it]
                                                         
{'loss': 0.0866, 'learning_rate': 5e-05, 'epoch': 2.57}

 51%|█████▏    | 12565/24420 [5:14:02<4:54:05,  1.49s/it]
 51%|█████▏    | 12566/24420 [5:14:03<4:54:32,  1.49s/it]
 51%|█████▏    | 12567/24420 [5:14:05<4:54:36,  1.49s/it]
 51%|█████▏    | 12568/24420 [5:14:06<4:54:18,  1.49s/it]
 51%|█████▏    | 12569/24420 [5:14:08<4:53:42,  1.49s/it]
 51%|█████▏    | 12570/24420 [5:14:09<4:54:08,  1.49s/it]
                                                         
{'loss': 0.1012, 'learning_rate': 5e-05, 'epoch': 2.57}

 51%|█████▏    | 12570/24420 [5:14:09<4:54:08,  1.49s/it]
 51%|█████▏    | 12571/24420 [5:14:11<4:53:51,  1.49s/it]
 51%|█████▏    | 12572/24420 [5:14:12<4:54:21,  1.49s/it]
 51%|█████▏    | 12573/24420 [5:14:14<4:54:29,  1.49s/it]
 51%|█████▏    | 12574/24420 [5:14:15<4:54:43,  1.49s/it]
 51%|█████▏    | 12575/24420 [5:14:17<4:55:03,  1.49s/it]
                                                         
{'loss': 0.0869, 'learning_rate': 5e-05, 'epoch': 2.57}

 51%|█████▏    | 12575/24420 [5:14:17<4:55:03,  1.49s/it]
 51%|█████▏    | 12576/24420 [5:14:18<4:55:06,  1.49s/it]
 52%|█████▏    | 12577/24420 [5:14:20<4:55:35,  1.50s/it]
 52%|█████▏    | 12578/24420 [5:14:21<4:55:12,  1.50s/it]
 52%|█████▏    | 12579/24420 [5:14:23<4:54:46,  1.49s/it]
 52%|█████▏    | 12580/24420 [5:14:24<4:54:18,  1.49s/it]
                                                         
{'loss': 0.0896, 'learning_rate': 5e-05, 'epoch': 2.58}

 52%|█████▏    | 12580/24420 [5:14:24<4:54:18,  1.49s/it]
 52%|█████▏    | 12581/24420 [5:14:26<4:54:12,  1.49s/it]
 52%|█████▏    | 12582/24420 [5:14:27<4:53:56,  1.49s/it]
 52%|█████▏    | 12583/24420 [5:14:29<4:53:45,  1.49s/it]
 52%|█████▏    | 12584/24420 [5:14:30<4:53:44,  1.49s/it]
 52%|█████▏    | 12585/24420 [5:14:32<4:53:20,  1.49s/it]
                                                         
{'loss': 0.098, 'learning_rate': 5e-05, 'epoch': 2.58}

 52%|█████▏    | 12585/24420 [5:14:32<4:53:20,  1.49s/it]
 52%|█████▏    | 12586/24420 [5:14:33<4:53:24,  1.49s/it]
 52%|█████▏    | 12587/24420 [5:14:34<4:53:13,  1.49s/it]
 52%|█████▏    | 12588/24420 [5:14:36<4:53:12,  1.49s/it]
 52%|█████▏    | 12589/24420 [5:14:37<4:53:01,  1.49s/it]
 52%|█████▏    | 12590/24420 [5:14:39<4:53:14,  1.49s/it]
                                                         
{'loss': 0.1048, 'learning_rate': 5e-05, 'epoch': 2.58}

 52%|█████▏    | 12590/24420 [5:14:39<4:53:14,  1.49s/it]
 52%|█████▏    | 12591/24420 [5:14:40<4:53:07,  1.49s/it]
 52%|█████▏    | 12592/24420 [5:14:42<4:52:54,  1.49s/it]
 52%|█████▏    | 12593/24420 [5:14:43<4:52:40,  1.48s/it]
 52%|█████▏    | 12594/24420 [5:14:45<4:52:38,  1.48s/it]
 52%|█████▏    | 12595/24420 [5:14:46<4:52:54,  1.49s/it]
                                                         
{'loss': 0.0881, 'learning_rate': 5e-05, 'epoch': 2.58}

 52%|█████▏    | 12595/24420 [5:14:46<4:52:54,  1.49s/it]
 52%|█████▏    | 12596/24420 [5:14:48<4:53:17,  1.49s/it]
 52%|█████▏    | 12597/24420 [5:14:49<4:53:50,  1.49s/it]
 52%|█████▏    | 12598/24420 [5:14:51<4:53:26,  1.49s/it]
 52%|█████▏    | 12599/24420 [5:14:52<4:53:10,  1.49s/it]
 52%|█████▏    | 12600/24420 [5:14:54<4:52:57,  1.49s/it]
                                                         
{'loss': 0.0806, 'learning_rate': 5e-05, 'epoch': 2.58}

 52%|█████▏    | 12600/24420 [5:14:54<4:52:57,  1.49s/it]
 52%|█████▏    | 12601/24420 [5:15:00<9:05:41,  2.77s/it]
 52%|█████▏    | 12602/24420 [5:15:01<7:49:54,  2.39s/it]
 52%|█████▏    | 12603/24420 [5:15:03<6:56:48,  2.12s/it]
 52%|█████▏    | 12604/24420 [5:15:04<6:20:18,  1.93s/it]
 52%|█████▏    | 12605/24420 [5:15:06<5:53:49,  1.80s/it]
                                                         
{'loss': 0.0927, 'learning_rate': 5e-05, 'epoch': 2.58}

 52%|█████▏    | 12605/24420 [5:15:06<5:53:49,  1.80s/it]
 52%|█████▏    | 12606/24420 [5:15:07<5:35:44,  1.71s/it]
 52%|█████▏    | 12607/24420 [5:15:09<5:22:28,  1.64s/it]
 52%|█████▏    | 12608/24420 [5:15:10<5:13:41,  1.59s/it]
 52%|█████▏    | 12609/24420 [5:15:11<5:07:09,  1.56s/it]
 52%|█████▏    | 12610/24420 [5:15:13<5:02:28,  1.54s/it]
                                                         
{'loss': 0.0944, 'learning_rate': 5e-05, 'epoch': 2.58}

 52%|█████▏    | 12610/24420 [5:15:13<5:02:28,  1.54s/it]
 52%|█████▏    | 12611/24420 [5:15:14<4:59:59,  1.52s/it]
 52%|█████▏    | 12612/24420 [5:15:16<4:58:52,  1.52s/it]
 52%|█████▏    | 12613/24420 [5:15:17<4:56:57,  1.51s/it]
 52%|█████▏    | 12614/24420 [5:15:19<4:55:36,  1.50s/it]
 52%|█████▏    | 12615/24420 [5:15:20<4:54:23,  1.50s/it]
                                                         
{'loss': 0.0884, 'learning_rate': 5e-05, 'epoch': 2.58}

 52%|█████▏    | 12615/24420 [5:15:20<4:54:23,  1.50s/it]
 52%|█████▏    | 12616/24420 [5:15:22<4:53:52,  1.49s/it]
 52%|█████▏    | 12617/24420 [5:15:23<4:53:46,  1.49s/it]
 52%|█████▏    | 12618/24420 [5:15:25<4:53:16,  1.49s/it]
 52%|█████▏    | 12619/24420 [5:15:26<4:52:52,  1.49s/it]
 52%|█████▏    | 12620/24420 [5:15:28<4:52:04,  1.49s/it]
                                                         
{'loss': 0.0867, 'learning_rate': 5e-05, 'epoch': 2.58}

 52%|█████▏    | 12620/24420 [5:15:28<4:52:04,  1.49s/it]
 52%|█████▏    | 12621/24420 [5:15:29<4:52:01,  1.48s/it]
 52%|█████▏    | 12622/24420 [5:15:31<4:51:41,  1.48s/it]
 52%|█████▏    | 12623/24420 [5:15:32<4:51:36,  1.48s/it]
 52%|█████▏    | 12624/24420 [5:15:34<4:51:35,  1.48s/it]
 52%|█████▏    | 12625/24420 [5:15:35<4:51:23,  1.48s/it]
                                                         
{'loss': 0.1025, 'learning_rate': 5e-05, 'epoch': 2.58}

 52%|█████▏    | 12625/24420 [5:15:35<4:51:23,  1.48s/it]
 52%|█████▏    | 12626/24420 [5:15:37<4:51:14,  1.48s/it]
 52%|█████▏    | 12627/24420 [5:15:38<4:51:57,  1.49s/it]
 52%|█████▏    | 12628/24420 [5:15:40<4:51:19,  1.48s/it]
 52%|█████▏    | 12629/24420 [5:15:41<4:51:28,  1.48s/it]
 52%|█████▏    | 12630/24420 [5:15:43<4:51:21,  1.48s/it]
                                                         
{'loss': 0.085, 'learning_rate': 5e-05, 'epoch': 2.59}

 52%|█████▏    | 12630/24420 [5:15:43<4:51:21,  1.48s/it]
 52%|█████▏    | 12631/24420 [5:15:44<4:51:11,  1.48s/it]
 52%|█████▏    | 12632/24420 [5:15:46<4:51:51,  1.49s/it]
 52%|█████▏    | 12633/24420 [5:15:47<4:51:24,  1.48s/it]
 52%|█████▏    | 12634/24420 [5:15:49<4:51:34,  1.48s/it]
 52%|█████▏    | 12635/24420 [5:15:50<4:51:50,  1.49s/it]
                                                         
{'loss': 0.0995, 'learning_rate': 5e-05, 'epoch': 2.59}

 52%|█████▏    | 12635/24420 [5:15:50<4:51:50,  1.49s/it]
 52%|█████▏    | 12636/24420 [5:15:52<4:52:23,  1.49s/it]
 52%|█████▏    | 12637/24420 [5:15:53<4:52:11,  1.49s/it]
 52%|█████▏    | 12638/24420 [5:15:55<4:51:48,  1.49s/it]
 52%|█████▏    | 12639/24420 [5:15:56<4:51:46,  1.49s/it]
 52%|█████▏    | 12640/24420 [5:15:58<4:51:52,  1.49s/it]
                                                         
{'loss': 0.0899, 'learning_rate': 5e-05, 'epoch': 2.59}

 52%|█████▏    | 12640/24420 [5:15:58<4:51:52,  1.49s/it]
 52%|█████▏    | 12641/24420 [5:15:59<4:52:01,  1.49s/it]
 52%|█████▏    | 12642/24420 [5:16:01<4:52:39,  1.49s/it]
 52%|█████▏    | 12643/24420 [5:16:02<4:52:21,  1.49s/it]
 52%|█████▏    | 12644/24420 [5:16:03<4:51:41,  1.49s/it]
 52%|█████▏    | 12645/24420 [5:16:05<4:51:43,  1.49s/it]
                                                         
{'loss': 0.085, 'learning_rate': 5e-05, 'epoch': 2.59}

 52%|█████▏    | 12645/24420 [5:16:05<4:51:43,  1.49s/it]
 52%|█████▏    | 12646/24420 [5:16:06<4:52:31,  1.49s/it]
 52%|█████▏    | 12647/24420 [5:16:08<4:52:22,  1.49s/it]
 52%|█████▏    | 12648/24420 [5:16:09<4:52:05,  1.49s/it]
 52%|█████▏    | 12649/24420 [5:16:11<4:51:45,  1.49s/it]
 52%|█████▏    | 12650/24420 [5:16:12<4:51:20,  1.49s/it]
                                                         
{'loss': 0.0812, 'learning_rate': 5e-05, 'epoch': 2.59}

 52%|█████▏    | 12650/24420 [5:16:12<4:51:20,  1.49s/it]
 52%|█████▏    | 12651/24420 [5:16:14<4:51:48,  1.49s/it]
 52%|█████▏    | 12652/24420 [5:16:15<4:51:59,  1.49s/it]
 52%|█████▏    | 12653/24420 [5:16:17<4:51:41,  1.49s/it]
 52%|█████▏    | 12654/24420 [5:16:18<4:51:26,  1.49s/it]
 52%|█████▏    | 12655/24420 [5:16:20<4:51:58,  1.49s/it]
                                                         
{'loss': 0.0927, 'learning_rate': 5e-05, 'epoch': 2.59}

 52%|█████▏    | 12655/24420 [5:16:20<4:51:58,  1.49s/it]
 52%|█████▏    | 12656/24420 [5:16:21<4:51:53,  1.49s/it]
 52%|█████▏    | 12657/24420 [5:16:23<4:52:29,  1.49s/it]
 52%|█████▏    | 12658/24420 [5:16:24<4:52:19,  1.49s/it]
 52%|█████▏    | 12659/24420 [5:16:26<4:52:20,  1.49s/it]
 52%|█████▏    | 12660/24420 [5:16:27<4:52:15,  1.49s/it]
                                                         
{'loss': 0.0829, 'learning_rate': 5e-05, 'epoch': 2.59}

 52%|█████▏    | 12660/24420 [5:16:27<4:52:15,  1.49s/it]
 52%|█████▏    | 12661/24420 [5:16:29<4:51:59,  1.49s/it]
 52%|█████▏    | 12662/24420 [5:16:30<4:51:29,  1.49s/it]
 52%|█████▏    | 12663/24420 [5:16:32<4:51:05,  1.49s/it]
 52%|█████▏    | 12664/24420 [5:16:33<4:51:15,  1.49s/it]
 52%|█████▏    | 12665/24420 [5:16:35<4:51:01,  1.49s/it]
                                                         
{'loss': 0.0661, 'learning_rate': 5e-05, 'epoch': 2.59}

 52%|█████▏    | 12665/24420 [5:16:35<4:51:01,  1.49s/it]
 52%|█████▏    | 12666/24420 [5:16:36<4:51:35,  1.49s/it]
 52%|█████▏    | 12667/24420 [5:16:38<4:52:06,  1.49s/it]
 52%|█████▏    | 12668/24420 [5:16:39<4:52:19,  1.49s/it]
 52%|█████▏    | 12669/24420 [5:16:41<4:51:58,  1.49s/it]
 52%|█████▏    | 12670/24420 [5:16:42<4:51:17,  1.49s/it]
                                                         
{'loss': 0.0883, 'learning_rate': 5e-05, 'epoch': 2.59}

 52%|█████▏    | 12670/24420 [5:16:42<4:51:17,  1.49s/it]
 52%|█████▏    | 12671/24420 [5:16:44<4:51:23,  1.49s/it]
 52%|█████▏    | 12672/24420 [5:16:45<4:51:36,  1.49s/it]
 52%|█████▏    | 12673/24420 [5:16:47<4:52:07,  1.49s/it]
 52%|█████▏    | 12674/24420 [5:16:48<4:51:46,  1.49s/it]
 52%|█████▏    | 12675/24420 [5:16:50<4:51:56,  1.49s/it]
                                                         
{'loss': 0.1064, 'learning_rate': 5e-05, 'epoch': 2.6}

 52%|█████▏    | 12675/24420 [5:16:50<4:51:56,  1.49s/it]
 52%|█████▏    | 12676/24420 [5:16:51<4:51:55,  1.49s/it]
 52%|█████▏    | 12677/24420 [5:16:53<4:51:33,  1.49s/it]
 52%|█████▏    | 12678/24420 [5:16:55<5:13:18,  1.60s/it]
 52%|█████▏    | 12679/24420 [5:16:56<5:06:40,  1.57s/it]
 52%|█████▏    | 12680/24420 [5:16:57<5:02:07,  1.54s/it]
                                                         
{'loss': 0.0883, 'learning_rate': 5e-05, 'epoch': 2.6}

 52%|█████▏    | 12680/24420 [5:16:57<5:02:07,  1.54s/it]
 52%|█████▏    | 12681/24420 [5:16:59<4:58:59,  1.53s/it]
 52%|█████▏    | 12682/24420 [5:17:00<4:56:16,  1.51s/it]
 52%|█████▏    | 12683/24420 [5:17:02<4:54:21,  1.50s/it]
 52%|█████▏    | 12684/24420 [5:17:03<4:53:15,  1.50s/it]
 52%|█████▏    | 12685/24420 [5:17:05<4:52:30,  1.50s/it]
                                                         
{'loss': 0.0881, 'learning_rate': 5e-05, 'epoch': 2.6}

 52%|█████▏    | 12685/24420 [5:17:05<4:52:30,  1.50s/it]
 52%|█████▏    | 12686/24420 [5:17:06<4:51:45,  1.49s/it]
 52%|█████▏    | 12687/24420 [5:17:08<4:51:36,  1.49s/it]
 52%|█████▏    | 12688/24420 [5:17:09<4:51:11,  1.49s/it]
 52%|█████▏    | 12689/24420 [5:17:11<4:51:16,  1.49s/it]
 52%|█████▏    | 12690/24420 [5:17:12<4:51:45,  1.49s/it]
                                                         
{'loss': 0.0772, 'learning_rate': 5e-05, 'epoch': 2.6}

 52%|█████▏    | 12690/24420 [5:17:12<4:51:45,  1.49s/it]
 52%|█████▏    | 12691/24420 [5:17:14<4:51:23,  1.49s/it]
 52%|█████▏    | 12692/24420 [5:17:15<4:51:30,  1.49s/it]
 52%|█████▏    | 12693/24420 [5:17:17<4:50:58,  1.49s/it]
 52%|█████▏    | 12694/24420 [5:17:18<4:50:46,  1.49s/it]
 52%|█████▏    | 12695/24420 [5:17:20<4:50:58,  1.49s/it]
                                                         
{'loss': 0.0858, 'learning_rate': 5e-05, 'epoch': 2.6}

 52%|█████▏    | 12695/24420 [5:17:20<4:50:58,  1.49s/it]
 52%|█████▏    | 12696/24420 [5:17:21<4:50:33,  1.49s/it]
 52%|█████▏    | 12697/24420 [5:17:23<4:51:09,  1.49s/it]
 52%|█████▏    | 12698/24420 [5:17:24<4:50:37,  1.49s/it]
 52%|█████▏    | 12699/24420 [5:17:26<4:50:22,  1.49s/it]
 52%|█████▏    | 12700/24420 [5:17:27<4:50:23,  1.49s/it]
                                                         
{'loss': 0.0826, 'learning_rate': 5e-05, 'epoch': 2.6}

 52%|█████▏    | 12700/24420 [5:17:27<4:50:23,  1.49s/it]
 52%|█████▏    | 12701/24420 [5:17:29<4:50:21,  1.49s/it]
 52%|█████▏    | 12702/24420 [5:17:30<4:49:58,  1.48s/it]
 52%|█████▏    | 12703/24420 [5:17:32<4:49:52,  1.48s/it]
 52%|█████▏    | 12704/24420 [5:17:33<4:50:25,  1.49s/it]
 52%|█████▏    | 12705/24420 [5:17:35<4:50:27,  1.49s/it]
                                                         
{'loss': 0.0829, 'learning_rate': 5e-05, 'epoch': 2.6}

 52%|█████▏    | 12705/24420 [5:17:35<4:50:27,  1.49s/it]
 52%|█████▏    | 12706/24420 [5:17:36<4:50:22,  1.49s/it]
 52%|█████▏    | 12707/24420 [5:17:38<4:50:32,  1.49s/it]
 52%|█████▏    | 12708/24420 [5:17:39<4:50:28,  1.49s/it]
 52%|█████▏    | 12709/24420 [5:17:41<4:50:13,  1.49s/it]
 52%|█████▏    | 12710/24420 [5:17:42<4:50:44,  1.49s/it]
                                                         
{'loss': 0.1054, 'learning_rate': 5e-05, 'epoch': 2.6}

 52%|█████▏    | 12710/24420 [5:17:42<4:50:44,  1.49s/it]
 52%|█████▏    | 12711/24420 [5:17:44<4:50:25,  1.49s/it]
 52%|█████▏    | 12712/24420 [5:17:45<4:51:16,  1.49s/it]
 52%|█████▏    | 12713/24420 [5:17:47<4:50:45,  1.49s/it]
 52%|█████▏    | 12714/24420 [5:17:48<4:50:22,  1.49s/it]
 52%|█████▏    | 12715/24420 [5:17:50<4:49:45,  1.49s/it]
                                                         
{'loss': 0.1049, 'learning_rate': 5e-05, 'epoch': 2.6}

 52%|█████▏    | 12715/24420 [5:17:50<4:49:45,  1.49s/it]
 52%|█████▏    | 12716/24420 [5:17:51<4:49:41,  1.49s/it]
 52%|█████▏    | 12717/24420 [5:17:53<4:49:49,  1.49s/it]
 52%|█████▏    | 12718/24420 [5:17:54<4:49:38,  1.49s/it]
 52%|█████▏    | 12719/24420 [5:17:55<4:49:11,  1.48s/it]
 52%|█████▏    | 12720/24420 [5:17:57<4:49:05,  1.48s/it]
                                                         
{'loss': 0.073, 'learning_rate': 5e-05, 'epoch': 2.6}

 52%|█████▏    | 12720/24420 [5:17:57<4:49:05,  1.48s/it]
 52%|█████▏    | 12721/24420 [5:17:58<4:48:59,  1.48s/it]
 52%|█████▏    | 12722/24420 [5:18:00<4:49:49,  1.49s/it]
 52%|█████▏    | 12723/24420 [5:18:01<4:49:48,  1.49s/it]
 52%|█████▏    | 12724/24420 [5:18:03<4:49:49,  1.49s/it]
 52%|█████▏    | 12725/24420 [5:18:04<4:49:33,  1.49s/it]
                                                         
{'loss': 0.0831, 'learning_rate': 5e-05, 'epoch': 2.61}

 52%|█████▏    | 12725/24420 [5:18:04<4:49:33,  1.49s/it]
 52%|█████▏    | 12726/24420 [5:18:06<4:49:32,  1.49s/it]
 52%|█████▏    | 12727/24420 [5:18:07<4:49:37,  1.49s/it]
 52%|█████▏    | 12728/24420 [5:18:09<4:49:35,  1.49s/it]
 52%|█████▏    | 12729/24420 [5:18:10<4:49:21,  1.49s/it]
 52%|█████▏    | 12730/24420 [5:18:12<4:49:31,  1.49s/it]
                                                         
{'loss': 0.0842, 'learning_rate': 5e-05, 'epoch': 2.61}

 52%|█████▏    | 12730/24420 [5:18:12<4:49:31,  1.49s/it]
 52%|█████▏    | 12731/24420 [5:18:13<4:49:22,  1.49s/it]
 52%|█████▏    | 12732/24420 [5:18:15<4:48:52,  1.48s/it]
 52%|█████▏    | 12733/24420 [5:18:16<4:49:06,  1.48s/it]
 52%|█████▏    | 12734/24420 [5:18:18<4:49:09,  1.48s/it]
 52%|█████▏    | 12735/24420 [5:18:19<4:49:43,  1.49s/it]
                                                         
{'loss': 0.0877, 'learning_rate': 5e-05, 'epoch': 2.61}

 52%|█████▏    | 12735/24420 [5:18:19<4:49:43,  1.49s/it]
 52%|█████▏    | 12736/24420 [5:18:21<4:49:36,  1.49s/it]
 52%|█████▏    | 12737/24420 [5:18:22<4:49:26,  1.49s/it]
 52%|█████▏    | 12738/24420 [5:18:24<4:49:44,  1.49s/it]
 52%|█████▏    | 12739/24420 [5:18:25<4:49:17,  1.49s/it]
 52%|█████▏    | 12740/24420 [5:18:27<4:49:10,  1.49s/it]
                                                         
{'loss': 0.0771, 'learning_rate': 5e-05, 'epoch': 2.61}

 52%|█████▏    | 12740/24420 [5:18:27<4:49:10,  1.49s/it]
 52%|█████▏    | 12741/24420 [5:18:28<4:48:57,  1.48s/it]
 52%|█████▏    | 12742/24420 [5:18:30<4:49:03,  1.49s/it]
 52%|█████▏    | 12743/24420 [5:18:31<4:49:35,  1.49s/it]
 52%|█████▏    | 12744/24420 [5:18:33<4:49:03,  1.49s/it]
 52%|█████▏    | 12745/24420 [5:18:34<4:48:55,  1.48s/it]
                                                         
{'loss': 0.0895, 'learning_rate': 5e-05, 'epoch': 2.61}

 52%|█████▏    | 12745/24420 [5:18:34<4:48:55,  1.48s/it]
 52%|█████▏    | 12746/24420 [5:18:36<4:49:05,  1.49s/it]
 52%|█████▏    | 12747/24420 [5:18:37<4:49:33,  1.49s/it]
 52%|█████▏    | 12748/24420 [5:18:39<4:49:25,  1.49s/it]
 52%|█████▏    | 12749/24420 [5:18:40<4:49:21,  1.49s/it]
 52%|█████▏    | 12750/24420 [5:18:42<4:49:16,  1.49s/it]
                                                         
{'loss': 0.1007, 'learning_rate': 5e-05, 'epoch': 2.61}

 52%|█████▏    | 12750/24420 [5:18:42<4:49:16,  1.49s/it]
 52%|█████▏    | 12751/24420 [5:18:43<4:49:15,  1.49s/it]
 52%|█████▏    | 12752/24420 [5:18:45<4:49:57,  1.49s/it]
 52%|█████▏    | 12753/24420 [5:18:46<4:49:27,  1.49s/it]
 52%|█████▏    | 12754/24420 [5:18:48<4:49:57,  1.49s/it]
 52%|█████▏    | 12755/24420 [5:18:49<4:49:55,  1.49s/it]
                                                         
{'loss': 0.093, 'learning_rate': 5e-05, 'epoch': 2.61}

 52%|█████▏    | 12755/24420 [5:18:49<4:49:55,  1.49s/it]
 52%|█████▏    | 12756/24420 [5:18:50<4:50:01,  1.49s/it]
 52%|█████▏    | 12757/24420 [5:18:52<4:49:34,  1.49s/it]
 52%|█████▏    | 12758/24420 [5:18:53<4:49:15,  1.49s/it]
 52%|█████▏    | 12759/24420 [5:18:55<4:49:02,  1.49s/it]
 52%|█████▏    | 12760/24420 [5:18:56<4:48:59,  1.49s/it]
                                                         
{'loss': 0.0738, 'learning_rate': 5e-05, 'epoch': 2.61}

 52%|█████▏    | 12760/24420 [5:18:56<4:48:59,  1.49s/it]
 52%|█████▏    | 12761/24420 [5:18:58<4:49:08,  1.49s/it]
 52%|█████▏    | 12762/24420 [5:18:59<4:48:54,  1.49s/it]
 52%|█████▏    | 12763/24420 [5:19:01<4:48:44,  1.49s/it]
 52%|█████▏    | 12764/24420 [5:19:02<4:48:30,  1.49s/it]
 52%|█████▏    | 12765/24420 [5:19:04<4:48:25,  1.48s/it]
                                                         
{'loss': 0.0739, 'learning_rate': 5e-05, 'epoch': 2.61}

 52%|█████▏    | 12765/24420 [5:19:04<4:48:25,  1.48s/it]
 52%|█████▏    | 12766/24420 [5:19:05<4:49:15,  1.49s/it]
 52%|█████▏    | 12767/24420 [5:19:07<4:48:30,  1.49s/it]
 52%|█████▏    | 12768/24420 [5:19:08<4:48:59,  1.49s/it]
 52%|█████▏    | 12769/24420 [5:19:10<4:49:14,  1.49s/it]
 52%|█████▏    | 12770/24420 [5:19:11<4:48:50,  1.49s/it]
                                                         
{'loss': 0.0927, 'learning_rate': 5e-05, 'epoch': 2.61}

 52%|█████▏    | 12770/24420 [5:19:11<4:48:50,  1.49s/it]
 52%|█████▏    | 12771/24420 [5:19:13<4:48:38,  1.49s/it]
 52%|█████▏    | 12772/24420 [5:19:14<4:48:47,  1.49s/it]
 52%|█████▏    | 12773/24420 [5:19:16<4:49:19,  1.49s/it]
 52%|█████▏    | 12774/24420 [5:19:17<4:49:18,  1.49s/it]
 52%|█████▏    | 12775/24420 [5:19:19<4:49:11,  1.49s/it]
                                                         
{'loss': 0.0698, 'learning_rate': 5e-05, 'epoch': 2.62}

 52%|█████▏    | 12775/24420 [5:19:19<4:49:11,  1.49s/it]
 52%|█████▏    | 12776/24420 [5:19:20<4:49:03,  1.49s/it]
 52%|█████▏    | 12777/24420 [5:19:22<4:49:33,  1.49s/it]
 52%|█████▏    | 12778/24420 [5:19:23<4:49:15,  1.49s/it]
 52%|█████▏    | 12779/24420 [5:19:25<4:49:00,  1.49s/it]
 52%|█████▏    | 12780/24420 [5:19:26<4:48:43,  1.49s/it]
                                                         
{'loss': 0.0841, 'learning_rate': 5e-05, 'epoch': 2.62}

 52%|█████▏    | 12780/24420 [5:19:26<4:48:43,  1.49s/it]
 52%|█████▏    | 12781/24420 [5:19:28<4:49:12,  1.49s/it]
 52%|█████▏    | 12782/24420 [5:19:29<4:49:04,  1.49s/it]
 52%|█████▏    | 12783/24420 [5:19:31<4:48:13,  1.49s/it]
 52%|█████▏    | 12784/24420 [5:19:32<4:48:38,  1.49s/it]
 52%|█████▏    | 12785/24420 [5:19:34<4:48:02,  1.49s/it]
                                                         
{'loss': 0.0797, 'learning_rate': 5e-05, 'epoch': 2.62}

 52%|█████▏    | 12785/24420 [5:19:34<4:48:02,  1.49s/it]
 52%|█████▏    | 12786/24420 [5:19:35<4:48:13,  1.49s/it]
 52%|█████▏    | 12787/24420 [5:19:37<4:48:11,  1.49s/it]
 52%|█████▏    | 12788/24420 [5:19:38<4:48:07,  1.49s/it]
 52%|█████▏    | 12789/24420 [5:19:40<4:48:02,  1.49s/it]
 52%|█████▏    | 12790/24420 [5:19:41<4:48:30,  1.49s/it]
                                                         
{'loss': 0.0967, 'learning_rate': 5e-05, 'epoch': 2.62}

 52%|█████▏    | 12790/24420 [5:19:41<4:48:30,  1.49s/it]
 52%|█████▏    | 12791/24420 [5:19:43<4:48:50,  1.49s/it]
 52%|█████▏    | 12792/24420 [5:19:44<4:48:19,  1.49s/it]
 52%|█████▏    | 12793/24420 [5:19:46<4:48:06,  1.49s/it]
 52%|█████▏    | 12794/24420 [5:19:47<4:48:58,  1.49s/it]
 52%|█████▏    | 12795/24420 [5:19:49<4:48:26,  1.49s/it]
                                                         
{'loss': 0.0902, 'learning_rate': 5e-05, 'epoch': 2.62}

 52%|█████▏    | 12795/24420 [5:19:49<4:48:26,  1.49s/it]
 52%|█████▏    | 12796/24420 [5:19:50<4:48:20,  1.49s/it]
 52%|█████▏    | 12797/24420 [5:19:51<4:47:44,  1.49s/it]
 52%|█████▏    | 12798/24420 [5:19:53<4:47:33,  1.48s/it]
 52%|█████▏    | 12799/24420 [5:19:54<4:47:37,  1.49s/it]
 52%|█████▏    | 12800/24420 [5:19:56<4:47:31,  1.48s/it]
                                                         
{'loss': 0.0928, 'learning_rate': 5e-05, 'epoch': 2.62}

 52%|█████▏    | 12800/24420 [5:19:56<4:47:31,  1.48s/it]
 52%|█████▏    | 12801/24420 [5:20:01<7:58:52,  2.47s/it]
 52%|█████▏    | 12802/24420 [5:20:02<7:01:37,  2.18s/it]
 52%|█████▏    | 12803/24420 [5:20:04<6:21:10,  1.97s/it]
 52%|█████▏    | 12804/24420 [5:20:05<5:53:07,  1.82s/it]
 52%|█████▏    | 12805/24420 [5:20:07<5:33:03,  1.72s/it]
                                                         
{'loss': 0.0999, 'learning_rate': 5e-05, 'epoch': 2.62}

 52%|█████▏    | 12805/24420 [5:20:07<5:33:03,  1.72s/it]
 52%|█████▏    | 12806/24420 [5:20:08<5:20:13,  1.65s/it]
 52%|█████▏    | 12807/24420 [5:20:10<5:10:34,  1.60s/it]
 52%|█████▏    | 12808/24420 [5:20:11<5:03:25,  1.57s/it]
 52%|█████▏    | 12809/24420 [5:20:13<4:58:27,  1.54s/it]
 52%|█████▏    | 12810/24420 [5:20:14<4:55:31,  1.53s/it]
                                                         
{'loss': 0.1005, 'learning_rate': 5e-05, 'epoch': 2.62}

 52%|█████▏    | 12810/24420 [5:20:14<4:55:31,  1.53s/it]
 52%|█████▏    | 12811/24420 [5:20:16<4:53:38,  1.52s/it]
 52%|█████▏    | 12812/24420 [5:20:17<4:51:25,  1.51s/it]
 52%|█████▏    | 12813/24420 [5:20:19<4:50:21,  1.50s/it]
 52%|█████▏    | 12814/24420 [5:20:20<4:49:07,  1.49s/it]
 52%|█████▏    | 12815/24420 [5:20:22<4:48:24,  1.49s/it]
                                                         
{'loss': 0.0832, 'learning_rate': 5e-05, 'epoch': 2.62}

 52%|█████▏    | 12815/24420 [5:20:22<4:48:24,  1.49s/it]
 52%|█████▏    | 12816/24420 [5:20:23<4:48:02,  1.49s/it]
 52%|█████▏    | 12817/24420 [5:20:25<4:48:06,  1.49s/it]
 52%|█████▏    | 12818/24420 [5:20:26<4:47:11,  1.49s/it]
 52%|█████▏    | 12819/24420 [5:20:27<4:47:09,  1.49s/it]
 52%|█████▏    | 12820/24420 [5:20:29<4:46:42,  1.48s/it]
                                                         
{'loss': 0.0832, 'learning_rate': 5e-05, 'epoch': 2.62}

 52%|█████▏    | 12820/24420 [5:20:29<4:46:42,  1.48s/it]
 53%|█████▎    | 12821/24420 [5:20:30<4:46:57,  1.48s/it]
 53%|█████▎    | 12822/24420 [5:20:32<4:46:48,  1.48s/it]
 53%|█████▎    | 12823/24420 [5:20:33<4:47:33,  1.49s/it]
 53%|█████▎    | 12824/24420 [5:20:35<4:47:29,  1.49s/it]
 53%|█████▎    | 12825/24420 [5:20:36<4:48:12,  1.49s/it]
                                                         
{'loss': 0.1064, 'learning_rate': 5e-05, 'epoch': 2.63}

 53%|█████▎    | 12825/24420 [5:20:36<4:48:12,  1.49s/it]
 53%|█████▎    | 12826/24420 [5:20:38<4:48:17,  1.49s/it]
 53%|█████▎    | 12827/24420 [5:20:39<4:48:19,  1.49s/it]
 53%|█████▎    | 12828/24420 [5:20:41<4:48:32,  1.49s/it]
 53%|█████▎    | 12829/24420 [5:20:42<4:47:51,  1.49s/it]
 53%|█████▎    | 12830/24420 [5:20:44<4:48:07,  1.49s/it]
                                                         
{'loss': 0.1322, 'learning_rate': 5e-05, 'epoch': 2.63}

 53%|█████▎    | 12830/24420 [5:20:44<4:48:07,  1.49s/it]
 53%|█████▎    | 12831/24420 [5:20:45<4:47:45,  1.49s/it]
 53%|█████▎    | 12832/24420 [5:20:47<4:48:22,  1.49s/it]
 53%|█████▎    | 12833/24420 [5:20:48<4:48:24,  1.49s/it]
 53%|█████▎    | 12834/24420 [5:20:50<4:47:56,  1.49s/it]
 53%|█████▎    | 12835/24420 [5:20:51<4:48:04,  1.49s/it]
                                                         
{'loss': 0.0869, 'learning_rate': 5e-05, 'epoch': 2.63}

 53%|█████▎    | 12835/24420 [5:20:51<4:48:04,  1.49s/it]
 53%|█████▎    | 12836/24420 [5:20:53<4:47:36,  1.49s/it]
 53%|█████▎    | 12837/24420 [5:20:54<4:47:23,  1.49s/it]
 53%|█████▎    | 12838/24420 [5:20:56<4:47:07,  1.49s/it]
 53%|█████▎    | 12839/24420 [5:20:57<4:47:34,  1.49s/it]
 53%|█████▎    | 12840/24420 [5:20:59<4:48:00,  1.49s/it]
                                                         
{'loss': 0.0992, 'learning_rate': 5e-05, 'epoch': 2.63}

 53%|█████▎    | 12840/24420 [5:20:59<4:48:00,  1.49s/it]
 53%|█████▎    | 12841/24420 [5:21:00<4:47:56,  1.49s/it]
 53%|█████▎    | 12842/24420 [5:21:02<4:47:11,  1.49s/it]
 53%|█████▎    | 12843/24420 [5:21:03<4:47:07,  1.49s/it]
 53%|█████▎    | 12844/24420 [5:21:05<4:46:17,  1.48s/it]
 53%|█████▎    | 12845/24420 [5:21:06<4:46:50,  1.49s/it]
                                                         
{'loss': 0.1006, 'learning_rate': 5e-05, 'epoch': 2.63}

 53%|█████▎    | 12845/24420 [5:21:06<4:46:50,  1.49s/it]
 53%|█████▎    | 12846/24420 [5:21:08<4:46:50,  1.49s/it]
 53%|█████▎    | 12847/24420 [5:21:09<4:46:38,  1.49s/it]
 53%|█████▎    | 12848/24420 [5:21:11<4:46:58,  1.49s/it]
 53%|█████▎    | 12849/24420 [5:21:12<4:46:37,  1.49s/it]
 53%|█████▎    | 12850/24420 [5:21:14<4:46:51,  1.49s/it]
                                                         
{'loss': 0.0848, 'learning_rate': 5e-05, 'epoch': 2.63}

 53%|█████▎    | 12850/24420 [5:21:14<4:46:51,  1.49s/it]
 53%|█████▎    | 12851/24420 [5:21:15<4:46:27,  1.49s/it]
 53%|█████▎    | 12852/24420 [5:21:17<4:46:22,  1.49s/it]
 53%|█████▎    | 12853/24420 [5:21:18<4:46:29,  1.49s/it]
 53%|█████▎    | 12854/24420 [5:21:20<4:46:42,  1.49s/it]
 53%|█████▎    | 12855/24420 [5:21:21<4:47:03,  1.49s/it]
                                                         
{'loss': 0.0914, 'learning_rate': 5e-05, 'epoch': 2.63}

 53%|█████▎    | 12855/24420 [5:21:21<4:47:03,  1.49s/it]
 53%|█████▎    | 12856/24420 [5:21:23<4:47:21,  1.49s/it]
 53%|█████▎    | 12857/24420 [5:21:24<4:47:10,  1.49s/it]
 53%|█████▎    | 12858/24420 [5:21:26<4:47:20,  1.49s/it]
 53%|█████▎    | 12859/24420 [5:21:27<4:46:59,  1.49s/it]
 53%|█████▎    | 12860/24420 [5:21:29<4:46:58,  1.49s/it]
                                                         
{'loss': 0.0837, 'learning_rate': 5e-05, 'epoch': 2.63}

 53%|█████▎    | 12860/24420 [5:21:29<4:46:58,  1.49s/it]
 53%|█████▎    | 12861/24420 [5:21:30<4:47:19,  1.49s/it]
 53%|█████▎    | 12862/24420 [5:21:32<4:47:56,  1.49s/it]
 53%|█████▎    | 12863/24420 [5:21:33<4:47:05,  1.49s/it]
 53%|█████▎    | 12864/24420 [5:21:34<4:47:18,  1.49s/it]
 53%|█████▎    | 12865/24420 [5:21:36<4:46:15,  1.49s/it]
                                                         
{'loss': 0.0955, 'learning_rate': 5e-05, 'epoch': 2.63}

 53%|█████▎    | 12865/24420 [5:21:36<4:46:15,  1.49s/it]
 53%|█████▎    | 12866/24420 [5:21:37<4:47:00,  1.49s/it]
 53%|█████▎    | 12867/24420 [5:21:39<4:46:47,  1.49s/it]
 53%|█████▎    | 12868/24420 [5:21:40<4:46:37,  1.49s/it]
 53%|█████▎    | 12869/24420 [5:21:42<4:46:32,  1.49s/it]
 53%|█████▎    | 12870/24420 [5:21:43<4:46:48,  1.49s/it]
                                                         
{'loss': 0.0768, 'learning_rate': 5e-05, 'epoch': 2.64}

 53%|█████▎    | 12870/24420 [5:21:43<4:46:48,  1.49s/it]
 53%|█████▎    | 12871/24420 [5:21:45<4:47:03,  1.49s/it]
 53%|█████▎    | 12872/24420 [5:21:46<4:46:22,  1.49s/it]
 53%|█████▎    | 12873/24420 [5:21:48<4:46:12,  1.49s/it]
 53%|█████▎    | 12874/24420 [5:21:49<4:46:13,  1.49s/it]
 53%|█████▎    | 12875/24420 [5:21:51<4:46:03,  1.49s/it]
                                                         
{'loss': 0.0889, 'learning_rate': 5e-05, 'epoch': 2.64}

 53%|█████▎    | 12875/24420 [5:21:51<4:46:03,  1.49s/it]
 53%|█████▎    | 12876/24420 [5:21:52<4:45:52,  1.49s/it]
 53%|█████▎    | 12877/24420 [5:21:54<4:45:49,  1.49s/it]
 53%|█████▎    | 12878/24420 [5:21:55<4:45:31,  1.48s/it]
 53%|█████▎    | 12879/24420 [5:21:57<4:45:40,  1.49s/it]
 53%|█████▎    | 12880/24420 [5:21:58<4:45:38,  1.49s/it]
                                                         
{'loss': 0.0932, 'learning_rate': 5e-05, 'epoch': 2.64}

 53%|█████▎    | 12880/24420 [5:21:58<4:45:38,  1.49s/it]
 53%|█████▎    | 12881/24420 [5:22:00<4:45:33,  1.48s/it]
 53%|█████▎    | 12882/24420 [5:22:01<4:46:27,  1.49s/it]
 53%|█████▎    | 12883/24420 [5:22:03<4:46:35,  1.49s/it]
 53%|█████▎    | 12884/24420 [5:22:04<4:46:03,  1.49s/it]
 53%|█████▎    | 12885/24420 [5:22:06<4:45:55,  1.49s/it]
                                                         
{'loss': 0.0818, 'learning_rate': 5e-05, 'epoch': 2.64}

 53%|█████▎    | 12885/24420 [5:22:06<4:45:55,  1.49s/it]
 53%|█████▎    | 12886/24420 [5:22:07<4:45:34,  1.49s/it]
 53%|█████▎    | 12887/24420 [5:22:09<4:45:06,  1.48s/it]
 53%|█████▎    | 12888/24420 [5:22:10<4:45:20,  1.48s/it]
 53%|█████▎    | 12889/24420 [5:22:12<4:45:10,  1.48s/it]
 53%|█████▎    | 12890/24420 [5:22:13<4:45:14,  1.48s/it]
                                                         
{'loss': 0.0945, 'learning_rate': 5e-05, 'epoch': 2.64}

 53%|█████▎    | 12890/24420 [5:22:13<4:45:14,  1.48s/it]
 53%|█████▎    | 12891/24420 [5:22:15<4:45:24,  1.49s/it]
 53%|█████▎    | 12892/24420 [5:22:16<4:44:40,  1.48s/it]
 53%|█████▎    | 12893/24420 [5:22:18<4:44:12,  1.48s/it]
 53%|█████▎    | 12894/24420 [5:22:19<4:44:14,  1.48s/it]
 53%|█████▎    | 12895/24420 [5:22:21<4:44:24,  1.48s/it]
                                                         
{'loss': 0.0722, 'learning_rate': 5e-05, 'epoch': 2.64}

 53%|█████▎    | 12895/24420 [5:22:21<4:44:24,  1.48s/it]
 53%|█████▎    | 12896/24420 [5:22:22<4:45:11,  1.48s/it]
 53%|█████▎    | 12897/24420 [5:22:24<4:44:31,  1.48s/it]
 53%|█████▎    | 12898/24420 [5:22:25<4:44:29,  1.48s/it]
 53%|█████▎    | 12899/24420 [5:22:26<4:44:10,  1.48s/it]
 53%|█████▎    | 12900/24420 [5:22:28<4:44:32,  1.48s/it]
                                                         
{'loss': 0.0934, 'learning_rate': 5e-05, 'epoch': 2.64}

 53%|█████▎    | 12900/24420 [5:22:28<4:44:32,  1.48s/it]
 53%|█████▎    | 12901/24420 [5:22:29<4:44:38,  1.48s/it]
 53%|█████▎    | 12902/24420 [5:22:31<4:44:48,  1.48s/it]
 53%|█████▎    | 12903/24420 [5:22:32<4:45:32,  1.49s/it]
 53%|█████▎    | 12904/24420 [5:22:34<4:45:31,  1.49s/it]
 53%|█████▎    | 12905/24420 [5:22:35<4:44:58,  1.48s/it]
                                                         
{'loss': 0.0744, 'learning_rate': 5e-05, 'epoch': 2.64}

 53%|█████▎    | 12905/24420 [5:22:35<4:44:58,  1.48s/it]
 53%|█████▎    | 12906/24420 [5:22:37<4:45:10,  1.49s/it]
 53%|█████▎    | 12907/24420 [5:22:38<4:45:27,  1.49s/it]
 53%|█████▎    | 12908/24420 [5:22:40<4:45:22,  1.49s/it]
 53%|█████▎    | 12909/24420 [5:22:41<4:45:36,  1.49s/it]
 53%|█████▎    | 12910/24420 [5:22:43<4:45:30,  1.49s/it]
                                                         
{'loss': 0.0925, 'learning_rate': 5e-05, 'epoch': 2.64}

 53%|█████▎    | 12910/24420 [5:22:43<4:45:30,  1.49s/it]
 53%|█████▎    | 12911/24420 [5:22:44<4:45:32,  1.49s/it]
 53%|█████▎    | 12912/24420 [5:22:46<4:45:28,  1.49s/it]
 53%|█████▎    | 12913/24420 [5:22:47<4:45:29,  1.49s/it]
 53%|█████▎    | 12914/24420 [5:22:49<4:45:09,  1.49s/it]
 53%|█████▎    | 12915/24420 [5:22:50<4:45:16,  1.49s/it]
                                                         
{'loss': 0.1102, 'learning_rate': 5e-05, 'epoch': 2.64}

 53%|█████▎    | 12915/24420 [5:22:50<4:45:16,  1.49s/it]
 53%|█████▎    | 12916/24420 [5:22:52<4:45:29,  1.49s/it]
 53%|█████▎    | 12917/24420 [5:22:53<4:45:01,  1.49s/it]
 53%|█████▎    | 12918/24420 [5:22:55<4:45:01,  1.49s/it]
 53%|█████▎    | 12919/24420 [5:22:56<4:45:31,  1.49s/it]
 53%|█████▎    | 12920/24420 [5:22:58<4:44:42,  1.49s/it]
                                                         
{'loss': 0.089, 'learning_rate': 5e-05, 'epoch': 2.65}

 53%|█████▎    | 12920/24420 [5:22:58<4:44:42,  1.49s/it]
 53%|█████▎    | 12921/24420 [5:22:59<4:44:36,  1.49s/it]
 53%|█████▎    | 12922/24420 [5:23:01<4:44:30,  1.48s/it]
 53%|█████▎    | 12923/24420 [5:23:02<4:44:36,  1.49s/it]
 53%|█████▎    | 12924/24420 [5:23:04<4:44:38,  1.49s/it]
 53%|█████▎    | 12925/24420 [5:23:05<4:44:40,  1.49s/it]
                                                         
{'loss': 0.0801, 'learning_rate': 5e-05, 'epoch': 2.65}

 53%|█████▎    | 12925/24420 [5:23:05<4:44:40,  1.49s/it]
 53%|█████▎    | 12926/24420 [5:23:07<4:44:56,  1.49s/it]
 53%|█████▎    | 12927/24420 [5:23:08<4:45:06,  1.49s/it]
 53%|█████▎    | 12928/24420 [5:23:10<4:44:59,  1.49s/it]
 53%|█████▎    | 12929/24420 [5:23:11<4:44:12,  1.48s/it]
 53%|█████▎    | 12930/24420 [5:23:13<4:44:05,  1.48s/it]
                                                         
{'loss': 0.0845, 'learning_rate': 5e-05, 'epoch': 2.65}

 53%|█████▎    | 12930/24420 [5:23:13<4:44:05,  1.48s/it]
 53%|█████▎    | 12931/24420 [5:23:14<4:44:06,  1.48s/it]
 53%|█████▎    | 12932/24420 [5:23:16<4:43:57,  1.48s/it]
 53%|█████▎    | 12933/24420 [5:23:17<4:44:25,  1.49s/it]
 53%|█████▎    | 12934/24420 [5:23:19<4:44:36,  1.49s/it]
 53%|█████▎    | 12935/24420 [5:23:20<4:45:05,  1.49s/it]
                                                         
{'loss': 0.0818, 'learning_rate': 5e-05, 'epoch': 2.65}

 53%|█████▎    | 12935/24420 [5:23:20<4:45:05,  1.49s/it]
 53%|█████▎    | 12936/24420 [5:23:21<4:45:06,  1.49s/it]
 53%|█████▎    | 12937/24420 [5:23:23<4:45:17,  1.49s/it]
 53%|█████▎    | 12938/24420 [5:23:24<4:44:59,  1.49s/it]
 53%|█████▎    | 12939/24420 [5:23:26<4:44:35,  1.49s/it]
 53%|█████▎    | 12940/24420 [5:23:27<4:44:14,  1.49s/it]
                                                         
{'loss': 0.0829, 'learning_rate': 5e-05, 'epoch': 2.65}

 53%|█████▎    | 12940/24420 [5:23:27<4:44:14,  1.49s/it]
 53%|█████▎    | 12941/24420 [5:23:29<4:44:14,  1.49s/it]
 53%|█████▎    | 12942/24420 [5:23:30<4:43:56,  1.48s/it]
 53%|█████▎    | 12943/24420 [5:23:32<4:44:06,  1.49s/it]
 53%|█████▎    | 12944/24420 [5:23:33<4:43:42,  1.48s/it]
 53%|█████▎    | 12945/24420 [5:23:35<4:43:50,  1.48s/it]
                                                         
{'loss': 0.0978, 'learning_rate': 5e-05, 'epoch': 2.65}

 53%|█████▎    | 12945/24420 [5:23:35<4:43:50,  1.48s/it]
 53%|█████▎    | 12946/24420 [5:23:36<4:44:37,  1.49s/it]
 53%|█████▎    | 12947/24420 [5:23:38<4:45:01,  1.49s/it]
 53%|█████▎    | 12948/24420 [5:23:39<4:45:05,  1.49s/it]
 53%|█████▎    | 12949/24420 [5:23:41<4:45:03,  1.49s/it]
 53%|█████▎    | 12950/24420 [5:23:42<4:44:37,  1.49s/it]
                                                         
{'loss': 0.0869, 'learning_rate': 5e-05, 'epoch': 2.65}

 53%|█████▎    | 12950/24420 [5:23:42<4:44:37,  1.49s/it]
 53%|█████▎    | 12951/24420 [5:23:44<4:44:22,  1.49s/it]
 53%|█████▎    | 12952/24420 [5:23:45<4:43:59,  1.49s/it]
 53%|█████▎    | 12953/24420 [5:23:47<4:43:48,  1.48s/it]
 53%|█████▎    | 12954/24420 [5:23:48<4:43:42,  1.48s/it]
 53%|█████▎    | 12955/24420 [5:23:50<4:43:21,  1.48s/it]
                                                         
{'loss': 0.0955, 'learning_rate': 5e-05, 'epoch': 2.65}

 53%|█████▎    | 12955/24420 [5:23:50<4:43:21,  1.48s/it]
 53%|█████▎    | 12956/24420 [5:23:51<4:43:57,  1.49s/it]
 53%|█████▎    | 12957/24420 [5:23:53<4:43:58,  1.49s/it]
 53%|█████▎    | 12958/24420 [5:23:54<4:43:25,  1.48s/it]
 53%|█████▎    | 12959/24420 [5:23:56<4:43:43,  1.49s/it]
 53%|█████▎    | 12960/24420 [5:23:57<4:44:01,  1.49s/it]
                                                         
{'loss': 0.1152, 'learning_rate': 5e-05, 'epoch': 2.65}

 53%|█████▎    | 12960/24420 [5:23:57<4:44:01,  1.49s/it]
 53%|█████▎    | 12961/24420 [5:23:59<4:43:47,  1.49s/it]
 53%|█████▎    | 12962/24420 [5:24:00<4:44:10,  1.49s/it]
 53%|█████▎    | 12963/24420 [5:24:02<4:44:12,  1.49s/it]
 53%|█████▎    | 12964/24420 [5:24:03<4:44:06,  1.49s/it]
 53%|█████▎    | 12965/24420 [5:24:05<4:44:03,  1.49s/it]
                                                         
{'loss': 0.0996, 'learning_rate': 5e-05, 'epoch': 2.65}

 53%|█████▎    | 12965/24420 [5:24:05<4:44:03,  1.49s/it]
 53%|█████▎    | 12966/24420 [5:24:06<4:44:21,  1.49s/it]
 53%|█████▎    | 12967/24420 [5:24:08<4:43:50,  1.49s/it]
 53%|█████▎    | 12968/24420 [5:24:09<4:44:00,  1.49s/it]
 53%|█████▎    | 12969/24420 [5:24:11<4:44:02,  1.49s/it]
 53%|█████▎    | 12970/24420 [5:24:12<4:44:43,  1.49s/it]
                                                         
{'loss': 0.0882, 'learning_rate': 5e-05, 'epoch': 2.66}

 53%|█████▎    | 12970/24420 [5:24:12<4:44:43,  1.49s/it]
 53%|█████▎    | 12971/24420 [5:24:14<4:44:45,  1.49s/it]
 53%|█████▎    | 12972/24420 [5:24:15<4:44:42,  1.49s/it]
 53%|█████▎    | 12973/24420 [5:24:17<4:44:23,  1.49s/it]
 53%|█████▎    | 12974/24420 [5:24:18<4:43:59,  1.49s/it]
 53%|█████▎    | 12975/24420 [5:24:19<4:43:31,  1.49s/it]
                                                         
{'loss': 0.0808, 'learning_rate': 5e-05, 'epoch': 2.66}

 53%|█████▎    | 12975/24420 [5:24:19<4:43:31,  1.49s/it]
 53%|█████▎    | 12976/24420 [5:24:21<4:43:17,  1.49s/it]
 53%|█████▎    | 12977/24420 [5:24:22<4:42:56,  1.48s/it]
 53%|█████▎    | 12978/24420 [5:24:24<4:43:06,  1.48s/it]
 53%|█████▎    | 12979/24420 [5:24:25<4:42:26,  1.48s/it]
 53%|█████▎    | 12980/24420 [5:24:27<4:42:59,  1.48s/it]
                                                         
{'loss': 0.1033, 'learning_rate': 5e-05, 'epoch': 2.66}

 53%|█████▎    | 12980/24420 [5:24:27<4:42:59,  1.48s/it]
 53%|█████▎    | 12981/24420 [5:24:28<4:43:18,  1.49s/it]
 53%|█████▎    | 12982/24420 [5:24:30<4:43:18,  1.49s/it]
 53%|█████▎    | 12983/24420 [5:24:31<4:44:16,  1.49s/it]
 53%|█████▎    | 12984/24420 [5:24:33<4:44:03,  1.49s/it]
 53%|█████▎    | 12985/24420 [5:24:34<4:44:03,  1.49s/it]
                                                         
{'loss': 0.0888, 'learning_rate': 5e-05, 'epoch': 2.66}

 53%|█████▎    | 12985/24420 [5:24:34<4:44:03,  1.49s/it]
 53%|█████▎    | 12986/24420 [5:24:36<4:44:02,  1.49s/it]
 53%|█████▎    | 12987/24420 [5:24:37<4:43:46,  1.49s/it]
 53%|█████▎    | 12988/24420 [5:24:39<4:43:53,  1.49s/it]
 53%|█████▎    | 12989/24420 [5:24:40<4:43:23,  1.49s/it]
 53%|█████▎    | 12990/24420 [5:24:42<4:43:22,  1.49s/it]
                                                         
{'loss': 0.0892, 'learning_rate': 5e-05, 'epoch': 2.66}

 53%|█████▎    | 12990/24420 [5:24:42<4:43:22,  1.49s/it]
 53%|█████▎    | 12991/24420 [5:24:43<4:43:04,  1.49s/it]
 53%|█████▎    | 12992/24420 [5:24:45<4:42:53,  1.49s/it]
 53%|█████▎    | 12993/24420 [5:24:46<4:42:51,  1.49s/it]
 53%|█████▎    | 12994/24420 [5:24:48<4:42:18,  1.48s/it]
 53%|█████▎    | 12995/24420 [5:24:49<4:42:04,  1.48s/it]
                                                         
{'loss': 0.0864, 'learning_rate': 5e-05, 'epoch': 2.66}

 53%|█████▎    | 12995/24420 [5:24:49<4:42:04,  1.48s/it]
 53%|█████▎    | 12996/24420 [5:24:51<4:42:20,  1.48s/it]
 53%|█████▎    | 12997/24420 [5:24:52<4:41:50,  1.48s/it]
 53%|█████▎    | 12998/24420 [5:24:54<4:42:32,  1.48s/it]
 53%|█████▎    | 12999/24420 [5:24:55<4:42:12,  1.48s/it]
 53%|█████▎    | 13000/24420 [5:24:57<4:42:20,  1.48s/it]
                                                         
{'loss': 0.0823, 'learning_rate': 5e-05, 'epoch': 2.66}

 53%|█████▎    | 13000/24420 [5:24:57<4:42:20,  1.48s/it]
 53%|█████▎    | 13001/24420 [5:25:01<7:38:16,  2.41s/it]
 53%|█████▎    | 13002/24420 [5:25:03<6:45:42,  2.13s/it]
 53%|█████▎    | 13003/24420 [5:25:04<6:08:29,  1.94s/it]
 53%|█████▎    | 13004/24420 [5:25:06<5:42:57,  1.80s/it]
 53%|█████▎    | 13005/24420 [5:25:07<5:24:11,  1.70s/it]
                                                         
{'loss': 0.087, 'learning_rate': 5e-05, 'epoch': 2.66}

 53%|█████▎    | 13005/24420 [5:25:07<5:24:11,  1.70s/it]
 53%|█████▎    | 13006/24420 [5:25:09<5:11:41,  1.64s/it]
 53%|█████▎    | 13007/24420 [5:25:10<5:02:49,  1.59s/it]
 53%|█████▎    | 13008/24420 [5:25:12<4:56:08,  1.56s/it]
 53%|█████▎    | 13009/24420 [5:25:13<4:52:19,  1.54s/it]
 53%|█████▎    | 13010/24420 [5:25:15<4:49:06,  1.52s/it]
                                                         
{'loss': 0.08, 'learning_rate': 5e-05, 'epoch': 2.66}

 53%|█████▎    | 13010/24420 [5:25:15<4:49:06,  1.52s/it]
 53%|█████▎    | 13011/24420 [5:25:16<4:47:34,  1.51s/it]
 53%|█████▎    | 13012/24420 [5:25:18<4:46:04,  1.50s/it]
 53%|█████▎    | 13013/24420 [5:25:19<4:44:26,  1.50s/it]
 53%|█████▎    | 13014/24420 [5:25:20<4:44:32,  1.50s/it]
 53%|█████▎    | 13015/24420 [5:25:22<4:43:45,  1.49s/it]
                                                         
{'loss': 0.0981, 'learning_rate': 5e-05, 'epoch': 2.66}

 53%|█████▎    | 13015/24420 [5:25:22<4:43:45,  1.49s/it]
 53%|█████▎    | 13016/24420 [5:25:23<4:43:31,  1.49s/it]
 53%|█████▎    | 13017/24420 [5:25:25<4:42:46,  1.49s/it]
 53%|█████▎    | 13018/24420 [5:25:26<4:42:25,  1.49s/it]
 53%|█████▎    | 13019/24420 [5:25:28<4:42:22,  1.49s/it]
 53%|█████▎    | 13020/24420 [5:25:29<4:42:49,  1.49s/it]
                                                         
{'loss': 0.0848, 'learning_rate': 5e-05, 'epoch': 2.67}

 53%|█████▎    | 13020/24420 [5:25:29<4:42:49,  1.49s/it]
 53%|█████▎    | 13021/24420 [5:25:31<4:43:05,  1.49s/it]
 53%|█████▎    | 13022/24420 [5:25:32<4:43:03,  1.49s/it]
 53%|█████▎    | 13023/24420 [5:25:34<4:43:35,  1.49s/it]
 53%|█████▎    | 13024/24420 [5:25:35<4:43:44,  1.49s/it]
 53%|█████▎    | 13025/24420 [5:25:37<4:44:00,  1.50s/it]
                                                         
{'loss': 0.1027, 'learning_rate': 5e-05, 'epoch': 2.67}

 53%|█████▎    | 13025/24420 [5:25:37<4:44:00,  1.50s/it]
 53%|█████▎    | 13026/24420 [5:25:38<4:43:44,  1.49s/it]
 53%|█████▎    | 13027/24420 [5:25:40<4:43:21,  1.49s/it]
 53%|█████▎    | 13028/24420 [5:25:41<4:43:05,  1.49s/it]
 53%|█████▎    | 13029/24420 [5:25:43<4:42:39,  1.49s/it]
 53%|█████▎    | 13030/24420 [5:25:44<4:42:19,  1.49s/it]
                                                         
{'loss': 0.0921, 'learning_rate': 5e-05, 'epoch': 2.67}

 53%|█████▎    | 13030/24420 [5:25:44<4:42:19,  1.49s/it]
 53%|█████▎    | 13031/24420 [5:25:46<4:42:25,  1.49s/it]
 53%|█████▎    | 13032/24420 [5:25:47<4:42:16,  1.49s/it]
 53%|█████▎    | 13033/24420 [5:25:49<4:42:03,  1.49s/it]
 53%|█████▎    | 13034/24420 [5:25:50<4:41:40,  1.48s/it]
 53%|█████▎    | 13035/24420 [5:25:52<4:41:44,  1.48s/it]
                                                         
{'loss': 0.0845, 'learning_rate': 5e-05, 'epoch': 2.67}

 53%|█████▎    | 13035/24420 [5:25:52<4:41:44,  1.48s/it]
 53%|█████▎    | 13036/24420 [5:25:53<4:41:38,  1.48s/it]
 53%|█████▎    | 13037/24420 [5:25:55<4:41:54,  1.49s/it]
 53%|█████▎    | 13038/24420 [5:25:56<4:41:48,  1.49s/it]
 53%|█████▎    | 13039/24420 [5:25:58<4:42:45,  1.49s/it]
 53%|█████▎    | 13040/24420 [5:25:59<4:42:07,  1.49s/it]
                                                         
{'loss': 0.0888, 'learning_rate': 5e-05, 'epoch': 2.67}

 53%|█████▎    | 13040/24420 [5:25:59<4:42:07,  1.49s/it]
 53%|█████▎    | 13041/24420 [5:26:01<4:42:03,  1.49s/it]
 53%|█████▎    | 13042/24420 [5:26:02<4:42:18,  1.49s/it]
 53%|█████▎    | 13043/24420 [5:26:04<4:42:09,  1.49s/it]
 53%|█████▎    | 13044/24420 [5:26:05<4:41:50,  1.49s/it]
 53%|█████▎    | 13045/24420 [5:26:07<5:17:00,  1.67s/it]
                                                         
{'loss': 0.0943, 'learning_rate': 5e-05, 'epoch': 2.67}

 53%|█████▎    | 13045/24420 [5:26:07<5:17:00,  1.67s/it]
 53%|█████▎    | 13046/24420 [5:26:09<5:06:43,  1.62s/it]
 53%|█████▎    | 13047/24420 [5:26:10<4:58:51,  1.58s/it]
 53%|█████▎    | 13048/24420 [5:26:12<4:53:23,  1.55s/it]
 53%|█████▎    | 13049/24420 [5:26:13<4:49:37,  1.53s/it]
 53%|█████▎    | 13050/24420 [5:26:15<4:50:47,  1.53s/it]
                                                         
{'loss': 0.0893, 'learning_rate': 5e-05, 'epoch': 2.67}

 53%|█████▎    | 13050/24420 [5:26:15<4:50:47,  1.53s/it]
 53%|█████▎    | 13051/24420 [5:26:16<4:47:58,  1.52s/it]
 53%|█████▎    | 13052/24420 [5:26:18<4:46:07,  1.51s/it]
 53%|█████▎    | 13053/24420 [5:26:19<4:44:37,  1.50s/it]
 53%|█████▎    | 13054/24420 [5:26:21<4:43:48,  1.50s/it]
 53%|█████▎    | 13055/24420 [5:26:22<4:44:08,  1.50s/it]
                                                         
{'loss': 0.1091, 'learning_rate': 5e-05, 'epoch': 2.67}

 53%|█████▎    | 13055/24420 [5:26:22<4:44:08,  1.50s/it]
 53%|█████▎    | 13056/24420 [5:26:24<4:43:24,  1.50s/it]
 53%|█████▎    | 13057/24420 [5:26:25<4:42:24,  1.49s/it]
 53%|█████▎    | 13058/24420 [5:26:27<4:42:23,  1.49s/it]
 53%|█████▎    | 13059/24420 [5:26:28<4:42:44,  1.49s/it]
 53%|█████▎    | 13060/24420 [5:26:30<4:43:04,  1.50s/it]
                                                         
{'loss': 0.0887, 'learning_rate': 5e-05, 'epoch': 2.67}

 53%|█████▎    | 13060/24420 [5:26:30<4:43:04,  1.50s/it]
 53%|█████▎    | 13061/24420 [5:26:31<4:42:24,  1.49s/it]
 53%|█████▎    | 13062/24420 [5:26:33<4:42:07,  1.49s/it]
 53%|█████▎    | 13063/24420 [5:26:34<4:42:14,  1.49s/it]
 53%|█████▎    | 13064/24420 [5:26:36<4:41:33,  1.49s/it]
 54%|█████▎    | 13065/24420 [5:26:37<4:40:55,  1.48s/it]
                                                         
{'loss': 0.1097, 'learning_rate': 5e-05, 'epoch': 2.68}

 54%|█████▎    | 13065/24420 [5:26:37<4:40:55,  1.48s/it]
 54%|█████▎    | 13066/24420 [5:26:39<4:40:46,  1.48s/it]
 54%|█████▎    | 13067/24420 [5:26:40<4:43:57,  1.50s/it]
 54%|█████▎    | 13068/24420 [5:26:42<4:42:56,  1.50s/it]
 54%|█████▎    | 13069/24420 [5:26:43<4:42:05,  1.49s/it]
 54%|█████▎    | 13070/24420 [5:26:45<4:41:54,  1.49s/it]
                                                         
{'loss': 0.089, 'learning_rate': 5e-05, 'epoch': 2.68}

 54%|█████▎    | 13070/24420 [5:26:45<4:41:54,  1.49s/it]
 54%|█████▎    | 13071/24420 [5:26:46<4:42:09,  1.49s/it]
 54%|█████▎    | 13072/24420 [5:26:48<4:42:32,  1.49s/it]
 54%|█████▎    | 13073/24420 [5:26:49<4:43:12,  1.50s/it]
 54%|█████▎    | 13074/24420 [5:26:51<4:42:26,  1.49s/it]
 54%|█████▎    | 13075/24420 [5:26:52<4:42:10,  1.49s/it]
                                                         
{'loss': 0.1107, 'learning_rate': 5e-05, 'epoch': 2.68}

 54%|█████▎    | 13075/24420 [5:26:52<4:42:10,  1.49s/it]
 54%|█████▎    | 13076/24420 [5:26:53<4:42:15,  1.49s/it]
 54%|█████▎    | 13077/24420 [5:26:55<4:42:07,  1.49s/it]
 54%|█████▎    | 13078/24420 [5:26:56<4:42:13,  1.49s/it]
 54%|█████▎    | 13079/24420 [5:26:58<4:42:17,  1.49s/it]
 54%|█████▎    | 13080/24420 [5:26:59<4:41:56,  1.49s/it]
                                                         
{'loss': 0.0961, 'learning_rate': 5e-05, 'epoch': 2.68}

 54%|█████▎    | 13080/24420 [5:26:59<4:41:56,  1.49s/it]
 54%|█████▎    | 13081/24420 [5:27:01<4:41:30,  1.49s/it]
 54%|█████▎    | 13082/24420 [5:27:02<4:40:48,  1.49s/it]
 54%|█████▎    | 13083/24420 [5:27:04<4:40:52,  1.49s/it]
 54%|█████▎    | 13084/24420 [5:27:05<4:40:56,  1.49s/it]
 54%|█████▎    | 13085/24420 [5:27:07<4:40:31,  1.48s/it]
                                                         
{'loss': 0.0873, 'learning_rate': 5e-05, 'epoch': 2.68}

 54%|█████▎    | 13085/24420 [5:27:07<4:40:31,  1.48s/it]
 54%|█████▎    | 13086/24420 [5:27:08<4:40:16,  1.48s/it]
 54%|█████▎    | 13087/24420 [5:27:10<4:40:33,  1.49s/it]
 54%|█████▎    | 13088/24420 [5:27:11<4:40:41,  1.49s/it]
 54%|█████▎    | 13089/24420 [5:27:13<4:40:36,  1.49s/it]
 54%|█████▎    | 13090/24420 [5:27:14<4:41:24,  1.49s/it]
                                                         
{'loss': 0.0881, 'learning_rate': 5e-05, 'epoch': 2.68}

 54%|█████▎    | 13090/24420 [5:27:14<4:41:24,  1.49s/it]
 54%|█████▎    | 13091/24420 [5:27:16<4:41:13,  1.49s/it]
 54%|█████▎    | 13092/24420 [5:27:17<4:40:35,  1.49s/it]
 54%|█████▎    | 13093/24420 [5:27:19<4:40:44,  1.49s/it]
 54%|█████▎    | 13094/24420 [5:27:20<4:41:38,  1.49s/it]
 54%|█████▎    | 13095/24420 [5:27:22<4:42:06,  1.49s/it]
                                                         
{'loss': 0.109, 'learning_rate': 5e-05, 'epoch': 2.68}

 54%|█████▎    | 13095/24420 [5:27:22<4:42:06,  1.49s/it]
 54%|█████▎    | 13096/24420 [5:27:23<4:41:44,  1.49s/it]
 54%|█████▎    | 13097/24420 [5:27:25<4:41:19,  1.49s/it]
 54%|█████▎    | 13098/24420 [5:27:26<4:40:52,  1.49s/it]
 54%|█████▎    | 13099/24420 [5:27:28<4:40:40,  1.49s/it]
 54%|█████▎    | 13100/24420 [5:27:29<4:40:27,  1.49s/it]
                                                         
{'loss': 0.1029, 'learning_rate': 5e-05, 'epoch': 2.68}

 54%|█████▎    | 13100/24420 [5:27:29<4:40:27,  1.49s/it]
 54%|█████▎    | 13101/24420 [5:27:31<4:41:02,  1.49s/it]
 54%|█████▎    | 13102/24420 [5:27:32<4:40:51,  1.49s/it]
 54%|█████▎    | 13103/24420 [5:27:34<4:40:51,  1.49s/it]
 54%|█████▎    | 13104/24420 [5:27:35<4:40:28,  1.49s/it]
 54%|█████▎    | 13105/24420 [5:27:37<4:40:47,  1.49s/it]
                                                         
{'loss': 0.0929, 'learning_rate': 5e-05, 'epoch': 2.68}

 54%|█████▎    | 13105/24420 [5:27:37<4:40:47,  1.49s/it]
 54%|█████▎    | 13106/24420 [5:27:38<4:41:19,  1.49s/it]
 54%|█████▎    | 13107/24420 [5:27:40<4:41:10,  1.49s/it]
 54%|█████▎    | 13108/24420 [5:27:41<4:41:10,  1.49s/it]
 54%|█████▎    | 13109/24420 [5:27:43<4:40:21,  1.49s/it]
 54%|█████▎    | 13110/24420 [5:27:44<4:40:00,  1.49s/it]
                                                         
{'loss': 0.1038, 'learning_rate': 5e-05, 'epoch': 2.68}

 54%|█████▎    | 13110/24420 [5:27:44<4:40:00,  1.49s/it]
 54%|█████▎    | 13111/24420 [5:27:46<4:40:03,  1.49s/it]
 54%|█████▎    | 13112/24420 [5:27:47<4:40:03,  1.49s/it]
 54%|█████▎    | 13113/24420 [5:27:49<4:40:04,  1.49s/it]
 54%|█████▎    | 13114/24420 [5:27:50<4:40:18,  1.49s/it]
 54%|█████▎    | 13115/24420 [5:27:52<4:40:11,  1.49s/it]
                                                         
{'loss': 0.0986, 'learning_rate': 5e-05, 'epoch': 2.69}

 54%|█████▎    | 13115/24420 [5:27:52<4:40:11,  1.49s/it]
 54%|█████▎    | 13116/24420 [5:27:53<4:39:49,  1.49s/it]
 54%|█████▎    | 13117/24420 [5:27:55<4:39:34,  1.48s/it]
 54%|█████▎    | 13118/24420 [5:27:56<4:38:59,  1.48s/it]
 54%|█████▎    | 13119/24420 [5:27:57<4:39:27,  1.48s/it]
 54%|█████▎    | 13120/24420 [5:27:59<4:40:01,  1.49s/it]
                                                         
{'loss': 0.0764, 'learning_rate': 5e-05, 'epoch': 2.69}

 54%|█████▎    | 13120/24420 [5:27:59<4:40:01,  1.49s/it]
 54%|█████▎    | 13121/24420 [5:28:00<4:40:06,  1.49s/it]
 54%|█████▎    | 13122/24420 [5:28:02<4:40:26,  1.49s/it]
 54%|█████▎    | 13123/24420 [5:28:03<4:40:20,  1.49s/it]
 54%|█████▎    | 13124/24420 [5:28:05<4:39:58,  1.49s/it]
 54%|█████▎    | 13125/24420 [5:28:06<4:39:35,  1.49s/it]
                                                         
{'loss': 0.0787, 'learning_rate': 5e-05, 'epoch': 2.69}

 54%|█████▎    | 13125/24420 [5:28:06<4:39:35,  1.49s/it]
 54%|█████▍    | 13126/24420 [5:28:08<4:39:13,  1.48s/it]
 54%|█████▍    | 13127/24420 [5:28:09<4:38:59,  1.48s/it]
 54%|█████▍    | 13128/24420 [5:28:11<4:38:28,  1.48s/it]
 54%|█████▍    | 13129/24420 [5:28:12<4:38:08,  1.48s/it]
 54%|█████▍    | 13130/24420 [5:28:14<4:38:25,  1.48s/it]
                                                         
{'loss': 0.081, 'learning_rate': 5e-05, 'epoch': 2.69}

 54%|█████▍    | 13130/24420 [5:28:14<4:38:25,  1.48s/it]
 54%|█████▍    | 13131/24420 [5:28:15<4:38:37,  1.48s/it]
 54%|█████▍    | 13132/24420 [5:28:17<4:39:02,  1.48s/it]
 54%|█████▍    | 13133/24420 [5:28:18<4:39:46,  1.49s/it]
 54%|█████▍    | 13134/24420 [5:28:20<4:39:33,  1.49s/it]
 54%|█████▍    | 13135/24420 [5:28:21<4:39:29,  1.49s/it]
                                                         
{'loss': 0.099, 'learning_rate': 5e-05, 'epoch': 2.69}

 54%|█████▍    | 13135/24420 [5:28:21<4:39:29,  1.49s/it]
 54%|█████▍    | 13136/24420 [5:28:23<4:39:31,  1.49s/it]
 54%|█████▍    | 13137/24420 [5:28:24<4:39:26,  1.49s/it]
 54%|█████▍    | 13138/24420 [5:28:26<4:38:59,  1.48s/it]
 54%|█████▍    | 13139/24420 [5:28:27<4:39:14,  1.49s/it]
 54%|█████▍    | 13140/24420 [5:28:29<4:39:13,  1.49s/it]
                                                         
{'loss': 0.0879, 'learning_rate': 5e-05, 'epoch': 2.69}

 54%|█████▍    | 13140/24420 [5:28:29<4:39:13,  1.49s/it]
 54%|█████▍    | 13141/24420 [5:28:30<4:39:20,  1.49s/it]
 54%|█████▍    | 13142/24420 [5:28:32<4:39:07,  1.49s/it]
 54%|█████▍    | 13143/24420 [5:28:33<4:39:03,  1.48s/it]
 54%|█████▍    | 13144/24420 [5:28:35<4:39:50,  1.49s/it]
 54%|█████▍    | 13145/24420 [5:28:36<4:39:34,  1.49s/it]
                                                         
{'loss': 0.0852, 'learning_rate': 5e-05, 'epoch': 2.69}

 54%|█████▍    | 13145/24420 [5:28:36<4:39:34,  1.49s/it]
 54%|█████▍    | 13146/24420 [5:28:38<4:39:39,  1.49s/it]
 54%|█████▍    | 13147/24420 [5:28:39<4:40:19,  1.49s/it]
 54%|█████▍    | 13148/24420 [5:28:41<4:39:50,  1.49s/it]
 54%|█████▍    | 13149/24420 [5:28:42<4:39:50,  1.49s/it]
 54%|█████▍    | 13150/24420 [5:28:44<4:39:55,  1.49s/it]
                                                         
{'loss': 0.0932, 'learning_rate': 5e-05, 'epoch': 2.69}

 54%|█████▍    | 13150/24420 [5:28:44<4:39:55,  1.49s/it]
 54%|█████▍    | 13151/24420 [5:28:45<4:39:38,  1.49s/it]
 54%|█████▍    | 13152/24420 [5:28:47<4:39:33,  1.49s/it]
 54%|█████▍    | 13153/24420 [5:28:48<4:39:13,  1.49s/it]
 54%|█████▍    | 13154/24420 [5:28:49<4:38:52,  1.49s/it]
 54%|█████▍    | 13155/24420 [5:28:51<4:38:36,  1.48s/it]
                                                         
{'loss': 0.0862, 'learning_rate': 5e-05, 'epoch': 2.69}

 54%|█████▍    | 13155/24420 [5:28:51<4:38:36,  1.48s/it]
 54%|█████▍    | 13156/24420 [5:28:52<4:38:38,  1.48s/it]
 54%|█████▍    | 13157/24420 [5:28:54<4:38:36,  1.48s/it]
 54%|█████▍    | 13158/24420 [5:28:55<4:38:16,  1.48s/it]
 54%|█████▍    | 13159/24420 [5:28:57<4:38:18,  1.48s/it]
 54%|█████▍    | 13160/24420 [5:28:58<4:37:59,  1.48s/it]
                                                         
{'loss': 0.1059, 'learning_rate': 5e-05, 'epoch': 2.69}

 54%|█████▍    | 13160/24420 [5:28:58<4:37:59,  1.48s/it]
 54%|█████▍    | 13161/24420 [5:29:00<4:37:48,  1.48s/it]
 54%|█████▍    | 13162/24420 [5:29:01<4:38:42,  1.49s/it]
 54%|█████▍    | 13163/24420 [5:29:03<4:38:07,  1.48s/it]
 54%|█████▍    | 13164/24420 [5:29:04<4:38:31,  1.48s/it]
 54%|█████▍    | 13165/24420 [5:29:06<4:38:28,  1.48s/it]
                                                         
{'loss': 0.065, 'learning_rate': 5e-05, 'epoch': 2.7}

 54%|█████▍    | 13165/24420 [5:29:06<4:38:28,  1.48s/it]
 54%|█████▍    | 13166/24420 [5:29:07<4:38:37,  1.49s/it]
 54%|█████▍    | 13167/24420 [5:29:09<4:38:33,  1.49s/it]
 54%|█████▍    | 13168/24420 [5:29:10<4:38:06,  1.48s/it]
 54%|█████▍    | 13169/24420 [5:29:12<4:38:00,  1.48s/it]
 54%|█████▍    | 13170/24420 [5:29:13<4:38:15,  1.48s/it]
                                                         
{'loss': 0.0881, 'learning_rate': 5e-05, 'epoch': 2.7}

 54%|█████▍    | 13170/24420 [5:29:13<4:38:15,  1.48s/it]
 54%|█████▍    | 13171/24420 [5:29:15<4:39:01,  1.49s/it]
 54%|█████▍    | 13172/24420 [5:29:16<4:39:01,  1.49s/it]
 54%|█████▍    | 13173/24420 [5:29:18<4:38:32,  1.49s/it]
 54%|█████▍    | 13174/24420 [5:29:19<4:38:31,  1.49s/it]
 54%|█████▍    | 13175/24420 [5:29:21<4:38:50,  1.49s/it]
                                                         
{'loss': 0.085, 'learning_rate': 5e-05, 'epoch': 2.7}

 54%|█████▍    | 13175/24420 [5:29:21<4:38:50,  1.49s/it]
 54%|█████▍    | 13176/24420 [5:29:22<4:38:47,  1.49s/it]
 54%|█████▍    | 13177/24420 [5:29:24<4:38:22,  1.49s/it]
 54%|█████▍    | 13178/24420 [5:29:25<4:38:17,  1.49s/it]
 54%|█████▍    | 13179/24420 [5:29:27<4:38:21,  1.49s/it]
 54%|█████▍    | 13180/24420 [5:29:28<4:37:43,  1.48s/it]
                                                         
{'loss': 0.0819, 'learning_rate': 5e-05, 'epoch': 2.7}

 54%|█████▍    | 13180/24420 [5:29:28<4:37:43,  1.48s/it]
 54%|█████▍    | 13181/24420 [5:29:30<4:37:52,  1.48s/it]
 54%|█████▍    | 13182/24420 [5:29:31<4:37:56,  1.48s/it]
 54%|█████▍    | 13183/24420 [5:29:33<4:37:51,  1.48s/it]
 54%|█████▍    | 13184/24420 [5:29:34<4:37:39,  1.48s/it]
 54%|█████▍    | 13185/24420 [5:29:35<4:38:07,  1.49s/it]
                                                         
{'loss': 0.0962, 'learning_rate': 5e-05, 'epoch': 2.7}

 54%|█████▍    | 13185/24420 [5:29:36<4:38:07,  1.49s/it]
 54%|█████▍    | 13186/24420 [5:29:37<4:38:43,  1.49s/it]
 54%|█████▍    | 13187/24420 [5:29:38<4:38:55,  1.49s/it]
 54%|█████▍    | 13188/24420 [5:29:40<4:38:15,  1.49s/it]
 54%|█████▍    | 13189/24420 [5:29:41<4:38:25,  1.49s/it]
 54%|█████▍    | 13190/24420 [5:29:43<4:37:54,  1.48s/it]
                                                         
{'loss': 0.0834, 'learning_rate': 5e-05, 'epoch': 2.7}

 54%|█████▍    | 13190/24420 [5:29:43<4:37:54,  1.48s/it]
 54%|█████▍    | 13191/24420 [5:29:44<4:38:05,  1.49s/it]
 54%|█████▍    | 13192/24420 [5:29:46<4:37:37,  1.48s/it]
 54%|█████▍    | 13193/24420 [5:29:47<4:37:36,  1.48s/it]
 54%|█████▍    | 13194/24420 [5:29:49<4:37:32,  1.48s/it]
 54%|█████▍    | 13195/24420 [5:29:50<4:37:39,  1.48s/it]
                                                         
{'loss': 0.1025, 'learning_rate': 5e-05, 'epoch': 2.7}

 54%|█████▍    | 13195/24420 [5:29:50<4:37:39,  1.48s/it]
 54%|█████▍    | 13196/24420 [5:29:52<4:37:44,  1.48s/it]
 54%|█████▍    | 13197/24420 [5:29:53<4:37:52,  1.49s/it]
 54%|█████▍    | 13198/24420 [5:29:55<4:37:46,  1.49s/it]
 54%|█████▍    | 13199/24420 [5:29:56<4:37:22,  1.48s/it]
 54%|█████▍    | 13200/24420 [5:29:58<4:37:32,  1.48s/it]
                                                         
{'loss': 0.1047, 'learning_rate': 5e-05, 'epoch': 2.7}

 54%|█████▍    | 13200/24420 [5:29:58<4:37:32,  1.48s/it]
 54%|█████▍    | 13201/24420 [5:30:02<7:31:38,  2.42s/it]
 54%|█████▍    | 13202/24420 [5:30:04<6:39:33,  2.14s/it]
 54%|█████▍    | 13203/24420 [5:30:05<6:02:40,  1.94s/it]
 54%|█████▍    | 13204/24420 [5:30:07<5:36:58,  1.80s/it]
 54%|█████▍    | 13205/24420 [5:30:08<5:19:38,  1.71s/it]
                                                         
{'loss': 0.0806, 'learning_rate': 5e-05, 'epoch': 2.7}

 54%|█████▍    | 13205/24420 [5:30:08<5:19:38,  1.71s/it]
 54%|█████▍    | 13206/24420 [5:30:10<5:08:15,  1.65s/it]
 54%|█████▍    | 13207/24420 [5:30:11<4:59:14,  1.60s/it]
 54%|█████▍    | 13208/24420 [5:30:13<4:52:17,  1.56s/it]
 54%|█████▍    | 13209/24420 [5:30:14<4:47:42,  1.54s/it]
 54%|█████▍    | 13210/24420 [5:30:16<4:44:11,  1.52s/it]
                                                         
{'loss': 0.082, 'learning_rate': 5e-05, 'epoch': 2.7}

 54%|█████▍    | 13210/24420 [5:30:16<4:44:11,  1.52s/it]
 54%|█████▍    | 13211/24420 [5:30:17<4:42:23,  1.51s/it]
 54%|█████▍    | 13212/24420 [5:30:19<4:40:26,  1.50s/it]
 54%|█████▍    | 13213/24420 [5:30:20<4:39:59,  1.50s/it]
 54%|█████▍    | 13214/24420 [5:30:22<4:39:14,  1.50s/it]
 54%|█████▍    | 13215/24420 [5:30:23<4:38:49,  1.49s/it]
                                                         
{'loss': 0.0828, 'learning_rate': 5e-05, 'epoch': 2.71}

 54%|█████▍    | 13215/24420 [5:30:23<4:38:49,  1.49s/it]
 54%|█████▍    | 13216/24420 [5:30:25<4:40:05,  1.50s/it]
 54%|█████▍    | 13217/24420 [5:30:26<4:39:01,  1.49s/it]
 54%|█████▍    | 13218/24420 [5:30:28<4:37:59,  1.49s/it]
 54%|█████▍    | 13219/24420 [5:30:29<4:37:38,  1.49s/it]
 54%|█████▍    | 13220/24420 [5:30:31<4:37:00,  1.48s/it]
                                                         
{'loss': 0.0866, 'learning_rate': 5e-05, 'epoch': 2.71}

 54%|█████▍    | 13220/24420 [5:30:31<4:37:00,  1.48s/it]
 54%|█████▍    | 13221/24420 [5:30:32<4:37:30,  1.49s/it]
 54%|█████▍    | 13222/24420 [5:30:34<4:37:32,  1.49s/it]
 54%|█████▍    | 13223/24420 [5:30:35<4:37:33,  1.49s/it]
 54%|█████▍    | 13224/24420 [5:30:37<4:37:52,  1.49s/it]
 54%|█████▍    | 13225/24420 [5:30:38<4:38:04,  1.49s/it]
                                                         
{'loss': 0.1007, 'learning_rate': 5e-05, 'epoch': 2.71}

 54%|█████▍    | 13225/24420 [5:30:38<4:38:04,  1.49s/it]
 54%|█████▍    | 13226/24420 [5:30:40<4:37:42,  1.49s/it]
 54%|█████▍    | 13227/24420 [5:30:41<4:37:53,  1.49s/it]
 54%|█████▍    | 13228/24420 [5:30:43<4:38:00,  1.49s/it]
 54%|█████▍    | 13229/24420 [5:30:44<4:37:46,  1.49s/it]
 54%|█████▍    | 13230/24420 [5:30:46<4:38:05,  1.49s/it]
                                                         
{'loss': 0.0998, 'learning_rate': 5e-05, 'epoch': 2.71}

 54%|█████▍    | 13230/24420 [5:30:46<4:38:05,  1.49s/it]
 54%|█████▍    | 13231/24420 [5:30:47<4:38:41,  1.49s/it]
 54%|█████▍    | 13232/24420 [5:30:48<4:37:47,  1.49s/it]
 54%|█████▍    | 13233/24420 [5:30:50<4:37:25,  1.49s/it]
 54%|█████▍    | 13234/24420 [5:30:51<4:37:41,  1.49s/it]
 54%|█████▍    | 13235/24420 [5:30:53<4:37:00,  1.49s/it]
                                                         
{'loss': 0.0765, 'learning_rate': 5e-05, 'epoch': 2.71}

 54%|█████▍    | 13235/24420 [5:30:53<4:37:00,  1.49s/it]
 54%|█████▍    | 13236/24420 [5:30:54<4:37:25,  1.49s/it]
 54%|█████▍    | 13237/24420 [5:30:56<4:37:24,  1.49s/it]
 54%|█████▍    | 13238/24420 [5:30:57<4:37:00,  1.49s/it]
 54%|█████▍    | 13239/24420 [5:30:59<4:36:32,  1.48s/it]
 54%|█████▍    | 13240/24420 [5:31:00<4:36:18,  1.48s/it]
                                                         
{'loss': 0.082, 'learning_rate': 5e-05, 'epoch': 2.71}

 54%|█████▍    | 13240/24420 [5:31:00<4:36:18,  1.48s/it]
 54%|█████▍    | 13241/24420 [5:31:02<4:36:23,  1.48s/it]
 54%|█████▍    | 13242/24420 [5:31:03<4:36:29,  1.48s/it]
 54%|█████▍    | 13243/24420 [5:31:05<4:36:06,  1.48s/it]
 54%|█████▍    | 13244/24420 [5:31:06<4:36:34,  1.48s/it]
 54%|█████▍    | 13245/24420 [5:31:08<4:37:00,  1.49s/it]
                                                         
{'loss': 0.0826, 'learning_rate': 5e-05, 'epoch': 2.71}

 54%|█████▍    | 13245/24420 [5:31:08<4:37:00,  1.49s/it]
 54%|█████▍    | 13246/24420 [5:31:09<4:36:32,  1.48s/it]
 54%|█████▍    | 13247/24420 [5:31:11<4:37:45,  1.49s/it]
 54%|█████▍    | 13248/24420 [5:31:12<4:37:17,  1.49s/it]
 54%|█████▍    | 13249/24420 [5:31:14<4:36:55,  1.49s/it]
 54%|█████▍    | 13250/24420 [5:31:15<4:36:34,  1.49s/it]
                                                         
{'loss': 0.075, 'learning_rate': 5e-05, 'epoch': 2.71}

 54%|█████▍    | 13250/24420 [5:31:15<4:36:34,  1.49s/it]
 54%|█████▍    | 13251/24420 [5:31:17<4:37:10,  1.49s/it]
 54%|█████▍    | 13252/24420 [5:31:18<4:37:19,  1.49s/it]
 54%|█████▍    | 13253/24420 [5:31:20<4:38:51,  1.50s/it]
 54%|█████▍    | 13254/24420 [5:31:21<4:37:26,  1.49s/it]
 54%|█████▍    | 13255/24420 [5:31:23<4:36:57,  1.49s/it]
                                                         
{'loss': 0.1154, 'learning_rate': 5e-05, 'epoch': 2.71}

 54%|█████▍    | 13255/24420 [5:31:23<4:36:57,  1.49s/it]
 54%|█████▍    | 13256/24420 [5:31:24<4:36:39,  1.49s/it]
 54%|█████▍    | 13257/24420 [5:31:26<4:36:33,  1.49s/it]
 54%|█████▍    | 13258/24420 [5:31:27<4:36:15,  1.48s/it]
 54%|█████▍    | 13259/24420 [5:31:29<4:36:26,  1.49s/it]
 54%|█████▍    | 13260/24420 [5:31:30<4:36:18,  1.49s/it]
                                                         
{'loss': 0.1209, 'learning_rate': 5e-05, 'epoch': 2.71}

 54%|█████▍    | 13260/24420 [5:31:30<4:36:18,  1.49s/it]
 54%|█████▍    | 13261/24420 [5:31:32<4:36:15,  1.49s/it]
 54%|█████▍    | 13262/24420 [5:31:33<4:36:18,  1.49s/it]
 54%|█████▍    | 13263/24420 [5:31:35<4:36:18,  1.49s/it]
 54%|█████▍    | 13264/24420 [5:31:36<4:36:26,  1.49s/it]
 54%|█████▍    | 13265/24420 [5:31:38<4:36:27,  1.49s/it]
                                                         
{'loss': 0.0991, 'learning_rate': 5e-05, 'epoch': 2.72}

 54%|█████▍    | 13265/24420 [5:31:38<4:36:27,  1.49s/it]
 54%|█████▍    | 13266/24420 [5:31:39<4:36:34,  1.49s/it]
 54%|█████▍    | 13267/24420 [5:31:41<4:37:50,  1.49s/it]
 54%|█████▍    | 13268/24420 [5:31:42<4:37:22,  1.49s/it]
 54%|█████▍    | 13269/24420 [5:31:44<4:36:44,  1.49s/it]
 54%|█████▍    | 13270/24420 [5:31:45<4:36:35,  1.49s/it]
                                                         
{'loss': 0.0903, 'learning_rate': 5e-05, 'epoch': 2.72}

 54%|█████▍    | 13270/24420 [5:31:45<4:36:35,  1.49s/it]
 54%|█████▍    | 13271/24420 [5:31:47<4:36:34,  1.49s/it]
 54%|█████▍    | 13272/24420 [5:31:48<4:36:32,  1.49s/it]
 54%|█████▍    | 13273/24420 [5:31:49<4:36:16,  1.49s/it]
 54%|█████▍    | 13274/24420 [5:31:51<4:35:48,  1.48s/it]
 54%|█████▍    | 13275/24420 [5:31:52<4:35:24,  1.48s/it]
                                                         
{'loss': 0.094, 'learning_rate': 5e-05, 'epoch': 2.72}

 54%|█████▍    | 13275/24420 [5:31:52<4:35:24,  1.48s/it]
 54%|█████▍    | 13276/24420 [5:31:54<4:35:41,  1.48s/it]
 54%|█████▍    | 13277/24420 [5:31:55<4:36:09,  1.49s/it]
 54%|█████▍    | 13278/24420 [5:31:57<4:36:14,  1.49s/it]
 54%|█████▍    | 13279/24420 [5:31:58<4:36:06,  1.49s/it]
 54%|█████▍    | 13280/24420 [5:32:00<4:36:10,  1.49s/it]
                                                         
{'loss': 0.0872, 'learning_rate': 5e-05, 'epoch': 2.72}

 54%|█████▍    | 13280/24420 [5:32:00<4:36:10,  1.49s/it]
 54%|█████▍    | 13281/24420 [5:32:01<4:36:19,  1.49s/it]
 54%|█████▍    | 13282/24420 [5:32:03<4:36:12,  1.49s/it]
 54%|█████▍    | 13283/24420 [5:32:04<4:36:05,  1.49s/it]
 54%|█████▍    | 13284/24420 [5:32:06<4:36:27,  1.49s/it]
 54%|█████▍    | 13285/24420 [5:32:07<4:36:05,  1.49s/it]
                                                         
{'loss': 0.0804, 'learning_rate': 5e-05, 'epoch': 2.72}

 54%|█████▍    | 13285/24420 [5:32:07<4:36:05,  1.49s/it]
 54%|█████▍    | 13286/24420 [5:32:09<4:36:11,  1.49s/it]
 54%|█████▍    | 13287/24420 [5:32:10<4:35:58,  1.49s/it]
 54%|█████▍    | 13288/24420 [5:32:12<4:35:36,  1.49s/it]
 54%|█████▍    | 13289/24420 [5:32:13<4:36:10,  1.49s/it]
 54%|█████▍    | 13290/24420 [5:32:15<4:35:43,  1.49s/it]
                                                         
{'loss': 0.07, 'learning_rate': 5e-05, 'epoch': 2.72}

 54%|█████▍    | 13290/24420 [5:32:15<4:35:43,  1.49s/it]
 54%|█████▍    | 13291/24420 [5:32:16<4:35:46,  1.49s/it]
 54%|█████▍    | 13292/24420 [5:32:18<4:35:54,  1.49s/it]
 54%|█████▍    | 13293/24420 [5:32:19<4:36:03,  1.49s/it]
 54%|█████▍    | 13294/24420 [5:32:21<4:36:24,  1.49s/it]
 54%|█████▍    | 13295/24420 [5:32:22<4:36:32,  1.49s/it]
                                                         
{'loss': 0.0797, 'learning_rate': 5e-05, 'epoch': 2.72}

 54%|█████▍    | 13295/24420 [5:32:22<4:36:32,  1.49s/it]
 54%|█████▍    | 13296/24420 [5:32:24<4:36:30,  1.49s/it]
 54%|█████▍    | 13297/24420 [5:32:25<4:36:01,  1.49s/it]
 54%|█████▍    | 13298/24420 [5:32:27<4:35:42,  1.49s/it]
 54%|█████▍    | 13299/24420 [5:32:28<4:35:55,  1.49s/it]
 54%|█████▍    | 13300/24420 [5:32:30<4:36:11,  1.49s/it]
                                                         
{'loss': 0.11, 'learning_rate': 5e-05, 'epoch': 2.72}

 54%|█████▍    | 13300/24420 [5:32:30<4:36:11,  1.49s/it]
 54%|█████▍    | 13301/24420 [5:32:31<4:36:03,  1.49s/it]
 54%|█████▍    | 13302/24420 [5:32:33<4:36:15,  1.49s/it]
 54%|█████▍    | 13303/24420 [5:32:34<4:36:02,  1.49s/it]
 54%|█████▍    | 13304/24420 [5:32:36<4:36:15,  1.49s/it]
 54%|█████▍    | 13305/24420 [5:32:37<4:35:51,  1.49s/it]
                                                         
{'loss': 0.0981, 'learning_rate': 5e-05, 'epoch': 2.72}

 54%|█████▍    | 13305/24420 [5:32:37<4:35:51,  1.49s/it]
 54%|█████▍    | 13306/24420 [5:32:39<4:35:27,  1.49s/it]
 54%|█████▍    | 13307/24420 [5:32:40<4:35:18,  1.49s/it]
 54%|█████▍    | 13308/24420 [5:32:42<4:35:18,  1.49s/it]
 55%|█████▍    | 13309/24420 [5:32:43<4:35:25,  1.49s/it]
 55%|█████▍    | 13310/24420 [5:32:45<4:34:45,  1.48s/it]
                                                         
{'loss': 0.0872, 'learning_rate': 5e-05, 'epoch': 2.73}

 55%|█████▍    | 13310/24420 [5:32:45<4:34:45,  1.48s/it]
 55%|█████▍    | 13311/24420 [5:32:46<4:34:45,  1.48s/it]
 55%|█████▍    | 13312/24420 [5:32:47<4:34:29,  1.48s/it]
 55%|█████▍    | 13313/24420 [5:32:49<4:34:30,  1.48s/it]
 55%|█████▍    | 13314/24420 [5:32:50<4:35:04,  1.49s/it]
 55%|█████▍    | 13315/24420 [5:32:52<4:34:57,  1.49s/it]
                                                         
{'loss': 0.0935, 'learning_rate': 5e-05, 'epoch': 2.73}

 55%|█████▍    | 13315/24420 [5:32:52<4:34:57,  1.49s/it]
 55%|█████▍    | 13316/24420 [5:32:53<4:35:39,  1.49s/it]
 55%|█████▍    | 13317/24420 [5:32:55<4:35:42,  1.49s/it]
 55%|█████▍    | 13318/24420 [5:32:56<4:35:55,  1.49s/it]
 55%|█████▍    | 13319/24420 [5:32:58<4:35:47,  1.49s/it]
 55%|█████▍    | 13320/24420 [5:32:59<4:35:19,  1.49s/it]
                                                         
{'loss': 0.081, 'learning_rate': 5e-05, 'epoch': 2.73}

 55%|█████▍    | 13320/24420 [5:32:59<4:35:19,  1.49s/it]
 55%|█████▍    | 13321/24420 [5:33:01<4:35:06,  1.49s/it]
 55%|█████▍    | 13322/24420 [5:33:02<4:35:17,  1.49s/it]
 55%|█████▍    | 13323/24420 [5:33:04<4:35:25,  1.49s/it]
 55%|█████▍    | 13324/24420 [5:33:05<4:35:30,  1.49s/it]
 55%|█████▍    | 13325/24420 [5:33:07<4:35:34,  1.49s/it]
                                                         
{'loss': 0.0831, 'learning_rate': 5e-05, 'epoch': 2.73}

 55%|█████▍    | 13325/24420 [5:33:07<4:35:34,  1.49s/it]
 55%|█████▍    | 13326/24420 [5:33:08<4:35:41,  1.49s/it]
 55%|█████▍    | 13327/24420 [5:33:10<4:35:15,  1.49s/it]
 55%|█████▍    | 13328/24420 [5:33:11<4:35:00,  1.49s/it]
 55%|█████▍    | 13329/24420 [5:33:13<4:35:05,  1.49s/it]
 55%|█████▍    | 13330/24420 [5:33:14<4:35:18,  1.49s/it]
                                                         
{'loss': 0.0905, 'learning_rate': 5e-05, 'epoch': 2.73}

 55%|█████▍    | 13330/24420 [5:33:14<4:35:18,  1.49s/it]
 55%|█████▍    | 13331/24420 [5:33:16<4:35:14,  1.49s/it]
 55%|█████▍    | 13332/24420 [5:33:17<4:34:49,  1.49s/it]
 55%|█████▍    | 13333/24420 [5:33:19<4:34:50,  1.49s/it]
 55%|█████▍    | 13334/24420 [5:33:20<4:34:41,  1.49s/it]
 55%|█████▍    | 13335/24420 [5:33:22<4:35:00,  1.49s/it]
                                                         
{'loss': 0.0751, 'learning_rate': 5e-05, 'epoch': 2.73}

 55%|█████▍    | 13335/24420 [5:33:22<4:35:00,  1.49s/it]
 55%|█████▍    | 13336/24420 [5:33:23<4:34:39,  1.49s/it]
 55%|█████▍    | 13337/24420 [5:33:25<4:34:35,  1.49s/it]
 55%|█████▍    | 13338/24420 [5:33:26<4:34:28,  1.49s/it]
 55%|█████▍    | 13339/24420 [5:33:28<4:34:18,  1.49s/it]
 55%|█████▍    | 13340/24420 [5:33:29<4:34:37,  1.49s/it]
                                                         
{'loss': 0.0789, 'learning_rate': 5e-05, 'epoch': 2.73}

 55%|█████▍    | 13340/24420 [5:33:29<4:34:37,  1.49s/it]
 55%|█████▍    | 13341/24420 [5:33:31<4:34:25,  1.49s/it]
 55%|█████▍    | 13342/24420 [5:33:32<4:33:47,  1.48s/it]
 55%|█████▍    | 13343/24420 [5:33:34<4:34:12,  1.49s/it]
 55%|█████▍    | 13344/24420 [5:33:35<4:33:51,  1.48s/it]
 55%|█████▍    | 13345/24420 [5:33:37<4:33:47,  1.48s/it]
                                                         
{'loss': 0.0829, 'learning_rate': 5e-05, 'epoch': 2.73}

 55%|█████▍    | 13345/24420 [5:33:37<4:33:47,  1.48s/it]
 55%|█████▍    | 13346/24420 [5:33:38<4:34:01,  1.48s/it]
 55%|█████▍    | 13347/24420 [5:33:40<4:34:07,  1.49s/it]
 55%|█████▍    | 13348/24420 [5:33:41<4:34:34,  1.49s/it]
 55%|█████▍    | 13349/24420 [5:33:43<4:34:35,  1.49s/it]
 55%|█████▍    | 13350/24420 [5:33:44<4:34:39,  1.49s/it]
                                                         
{'loss': 0.0901, 'learning_rate': 5e-05, 'epoch': 2.73}

 55%|█████▍    | 13350/24420 [5:33:44<4:34:39,  1.49s/it]
 55%|█████▍    | 13351/24420 [5:33:46<4:34:42,  1.49s/it]
 55%|█████▍    | 13352/24420 [5:33:47<4:34:21,  1.49s/it]
 55%|█████▍    | 13353/24420 [5:33:48<4:33:43,  1.48s/it]
 55%|█████▍    | 13354/24420 [5:33:50<4:34:09,  1.49s/it]
 55%|█████▍    | 13355/24420 [5:33:51<4:33:57,  1.49s/it]
                                                         
{'loss': 0.0816, 'learning_rate': 5e-05, 'epoch': 2.73}

 55%|█████▍    | 13355/24420 [5:33:51<4:33:57,  1.49s/it]
 55%|█████▍    | 13356/24420 [5:33:53<4:34:23,  1.49s/it]
 55%|█████▍    | 13357/24420 [5:33:54<4:34:35,  1.49s/it]
 55%|█████▍    | 13358/24420 [5:33:56<4:34:41,  1.49s/it]
 55%|█████▍    | 13359/24420 [5:33:57<4:34:37,  1.49s/it]
 55%|█████▍    | 13360/24420 [5:33:59<4:34:29,  1.49s/it]
                                                         
{'loss': 0.094, 'learning_rate': 5e-05, 'epoch': 2.74}

 55%|█████▍    | 13360/24420 [5:33:59<4:34:29,  1.49s/it]
 55%|█████▍    | 13361/24420 [5:34:00<4:33:55,  1.49s/it]
 55%|█████▍    | 13362/24420 [5:34:02<4:33:28,  1.48s/it]
 55%|█████▍    | 13363/24420 [5:34:03<4:32:45,  1.48s/it]
 55%|█████▍    | 13364/24420 [5:34:05<4:33:09,  1.48s/it]
 55%|█████▍    | 13365/24420 [5:34:06<4:33:08,  1.48s/it]
                                                         
{'loss': 0.0768, 'learning_rate': 5e-05, 'epoch': 2.74}

 55%|█████▍    | 13365/24420 [5:34:06<4:33:08,  1.48s/it]
 55%|█████▍    | 13366/24420 [5:34:08<4:33:11,  1.48s/it]
 55%|█████▍    | 13367/24420 [5:34:09<4:35:24,  1.50s/it]
 55%|█████▍    | 13368/24420 [5:34:11<4:34:41,  1.49s/it]
 55%|█████▍    | 13369/24420 [5:34:12<4:34:18,  1.49s/it]
 55%|█████▍    | 13370/24420 [5:34:14<4:34:02,  1.49s/it]
                                                         
{'loss': 0.0865, 'learning_rate': 5e-05, 'epoch': 2.74}

 55%|█████▍    | 13370/24420 [5:34:14<4:34:02,  1.49s/it]
 55%|█████▍    | 13371/24420 [5:34:15<4:34:07,  1.49s/it]
 55%|█████▍    | 13372/24420 [5:34:17<4:34:07,  1.49s/it]
 55%|█████▍    | 13373/24420 [5:34:18<4:34:12,  1.49s/it]
 55%|█████▍    | 13374/24420 [5:34:20<4:33:58,  1.49s/it]
 55%|█████▍    | 13375/24420 [5:34:21<4:33:37,  1.49s/it]
                                                         
{'loss': 0.0887, 'learning_rate': 5e-05, 'epoch': 2.74}

 55%|█████▍    | 13375/24420 [5:34:21<4:33:37,  1.49s/it]
 55%|█████▍    | 13376/24420 [5:34:23<4:33:54,  1.49s/it]
 55%|█████▍    | 13377/24420 [5:34:24<4:33:52,  1.49s/it]
 55%|█████▍    | 13378/24420 [5:34:26<4:33:13,  1.48s/it]
 55%|█████▍    | 13379/24420 [5:34:27<4:32:58,  1.48s/it]
 55%|█████▍    | 13380/24420 [5:34:29<4:33:16,  1.49s/it]
                                                         
{'loss': 0.0862, 'learning_rate': 5e-05, 'epoch': 2.74}

 55%|█████▍    | 13380/24420 [5:34:29<4:33:16,  1.49s/it]
 55%|█████▍    | 13381/24420 [5:34:30<4:33:27,  1.49s/it]
 55%|█████▍    | 13382/24420 [5:34:32<4:33:07,  1.48s/it]
 55%|█████▍    | 13383/24420 [5:34:33<4:32:46,  1.48s/it]
 55%|█████▍    | 13384/24420 [5:34:35<4:32:57,  1.48s/it]
 55%|█████▍    | 13385/24420 [5:34:36<4:33:15,  1.49s/it]
                                                         
{'loss': 0.0803, 'learning_rate': 5e-05, 'epoch': 2.74}

 55%|█████▍    | 13385/24420 [5:34:36<4:33:15,  1.49s/it]
 55%|█████▍    | 13386/24420 [5:34:38<4:33:19,  1.49s/it]
 55%|█████▍    | 13387/24420 [5:34:39<4:33:45,  1.49s/it]
 55%|█████▍    | 13388/24420 [5:34:41<4:33:34,  1.49s/it]
 55%|█████▍    | 13389/24420 [5:34:42<4:33:50,  1.49s/it]
 55%|█████▍    | 13390/24420 [5:34:43<4:33:26,  1.49s/it]
                                                         
{'loss': 0.0845, 'learning_rate': 5e-05, 'epoch': 2.74}

 55%|█████▍    | 13390/24420 [5:34:43<4:33:26,  1.49s/it]
 55%|█████▍    | 13391/24420 [5:34:45<4:33:30,  1.49s/it]
 55%|█████▍    | 13392/24420 [5:34:46<4:33:27,  1.49s/it]
 55%|█████▍    | 13393/24420 [5:34:48<4:33:19,  1.49s/it]
 55%|█████▍    | 13394/24420 [5:34:49<4:33:29,  1.49s/it]
 55%|█████▍    | 13395/24420 [5:34:51<4:33:36,  1.49s/it]
                                                         
{'loss': 0.0895, 'learning_rate': 5e-05, 'epoch': 2.74}

 55%|█████▍    | 13395/24420 [5:34:51<4:33:36,  1.49s/it]
 55%|█████▍    | 13396/24420 [5:34:52<4:33:37,  1.49s/it]
 55%|█████▍    | 13397/24420 [5:34:54<4:33:03,  1.49s/it]
 55%|█████▍    | 13398/24420 [5:34:55<4:32:58,  1.49s/it]
 55%|█████▍    | 13399/24420 [5:34:57<4:32:55,  1.49s/it]
 55%|█████▍    | 13400/24420 [5:34:58<4:33:03,  1.49s/it]
                                                         
{'loss': 0.0945, 'learning_rate': 5e-05, 'epoch': 2.74}

 55%|█████▍    | 13400/24420 [5:34:58<4:33:03,  1.49s/it]
 55%|█████▍    | 13401/24420 [5:35:03<7:35:08,  2.48s/it]
 55%|█████▍    | 13402/24420 [5:35:05<6:41:20,  2.19s/it]
 55%|█████▍    | 13403/24420 [5:35:06<6:03:06,  1.98s/it]
 55%|█████▍    | 13404/24420 [5:35:08<5:35:55,  1.83s/it]
 55%|█████▍    | 13405/24420 [5:35:09<5:16:42,  1.73s/it]
                                                         
{'loss': 0.0973, 'learning_rate': 5e-05, 'epoch': 2.74}

 55%|█████▍    | 13405/24420 [5:35:09<5:16:42,  1.73s/it]
 55%|█████▍    | 13406/24420 [5:35:11<5:03:23,  1.65s/it]
 55%|█████▍    | 13407/24420 [5:35:12<4:54:25,  1.60s/it]
 55%|█████▍    | 13408/24420 [5:35:14<4:47:37,  1.57s/it]
 55%|█████▍    | 13409/24420 [5:35:15<4:42:42,  1.54s/it]
 55%|█████▍    | 13410/24420 [5:35:17<4:39:34,  1.52s/it]
                                                         
{'loss': 0.0811, 'learning_rate': 5e-05, 'epoch': 2.75}

 55%|█████▍    | 13410/24420 [5:35:17<4:39:34,  1.52s/it]
 55%|█████▍    | 13411/24420 [5:35:18<4:37:05,  1.51s/it]
 55%|█████▍    | 13412/24420 [5:35:20<4:37:54,  1.51s/it]
 55%|█████▍    | 13413/24420 [5:35:21<4:36:13,  1.51s/it]
 55%|█████▍    | 13414/24420 [5:35:23<4:34:58,  1.50s/it]
 55%|█████▍    | 13415/24420 [5:35:24<4:33:56,  1.49s/it]
                                                         
{'loss': 0.0797, 'learning_rate': 5e-05, 'epoch': 2.75}

 55%|█████▍    | 13415/24420 [5:35:24<4:33:56,  1.49s/it]
 55%|█████▍    | 13416/24420 [5:35:25<4:33:24,  1.49s/it]
 55%|█████▍    | 13417/24420 [5:35:27<4:33:18,  1.49s/it]
 55%|█████▍    | 13418/24420 [5:35:28<4:33:06,  1.49s/it]
 55%|█████▍    | 13419/24420 [5:35:30<4:32:52,  1.49s/it]
 55%|█████▍    | 13420/24420 [5:35:31<4:32:46,  1.49s/it]
                                                         
{'loss': 0.0846, 'learning_rate': 5e-05, 'epoch': 2.75}

 55%|█████▍    | 13420/24420 [5:35:31<4:32:46,  1.49s/it]
 55%|█████▍    | 13421/24420 [5:35:33<4:32:46,  1.49s/it]
 55%|█████▍    | 13422/24420 [5:35:34<4:32:19,  1.49s/it]
 55%|█████▍    | 13423/24420 [5:35:36<4:32:14,  1.49s/it]
 55%|█████▍    | 13424/24420 [5:35:37<4:32:01,  1.48s/it]
 55%|█████▍    | 13425/24420 [5:35:39<4:32:35,  1.49s/it]
                                                         
{'loss': 0.0883, 'learning_rate': 5e-05, 'epoch': 2.75}

 55%|█████▍    | 13425/24420 [5:35:39<4:32:35,  1.49s/it]
 55%|█████▍    | 13426/24420 [5:35:40<4:32:33,  1.49s/it]
 55%|█████▍    | 13427/24420 [5:35:42<4:32:55,  1.49s/it]
 55%|█████▍    | 13428/24420 [5:35:43<4:32:52,  1.49s/it]
 55%|█████▍    | 13429/24420 [5:35:45<4:32:21,  1.49s/it]
 55%|█████▍    | 13430/24420 [5:35:46<4:32:07,  1.49s/it]
                                                         
{'loss': 0.1015, 'learning_rate': 5e-05, 'epoch': 2.75}

 55%|█████▍    | 13430/24420 [5:35:46<4:32:07,  1.49s/it]
 55%|█████▌    | 13431/24420 [5:35:48<4:32:11,  1.49s/it]
 55%|█████▌    | 13432/24420 [5:35:49<4:32:04,  1.49s/it]
 55%|█████▌    | 13433/24420 [5:35:51<4:31:51,  1.48s/it]
 55%|█████▌    | 13434/24420 [5:35:52<4:31:43,  1.48s/it]
 55%|█████▌    | 13435/24420 [5:35:54<4:31:46,  1.48s/it]
                                                         
{'loss': 0.1017, 'learning_rate': 5e-05, 'epoch': 2.75}

 55%|█████▌    | 13435/24420 [5:35:54<4:31:46,  1.48s/it]
 55%|█████▌    | 13436/24420 [5:35:55<4:31:57,  1.49s/it]
 55%|█████▌    | 13437/24420 [5:35:57<4:32:21,  1.49s/it]
 55%|█████▌    | 13438/24420 [5:35:58<4:32:31,  1.49s/it]
 55%|█████▌    | 13439/24420 [5:36:00<4:32:41,  1.49s/it]
 55%|█████▌    | 13440/24420 [5:36:01<4:32:10,  1.49s/it]
                                                         
{'loss': 0.0714, 'learning_rate': 5e-05, 'epoch': 2.75}

 55%|█████▌    | 13440/24420 [5:36:01<4:32:10,  1.49s/it]
 55%|█████▌    | 13441/24420 [5:36:03<4:31:41,  1.48s/it]
 55%|█████▌    | 13442/24420 [5:36:04<4:32:03,  1.49s/it]
 55%|█████▌    | 13443/24420 [5:36:06<4:31:51,  1.49s/it]
 55%|█████▌    | 13444/24420 [5:36:07<4:31:51,  1.49s/it]
 55%|█████▌    | 13445/24420 [5:36:09<4:31:37,  1.48s/it]
                                                         
{'loss': 0.0828, 'learning_rate': 5e-05, 'epoch': 2.75}

 55%|█████▌    | 13445/24420 [5:36:09<4:31:37,  1.48s/it]
 55%|█████▌    | 13446/24420 [5:36:10<4:31:59,  1.49s/it]
 55%|█████▌    | 13447/24420 [5:36:12<4:31:46,  1.49s/it]
 55%|█████▌    | 13448/24420 [5:36:13<4:31:29,  1.48s/it]
 55%|█████▌    | 13449/24420 [5:36:15<4:31:45,  1.49s/it]
 55%|█████▌    | 13450/24420 [5:36:16<4:31:27,  1.48s/it]
                                                         
{'loss': 0.0905, 'learning_rate': 5e-05, 'epoch': 2.75}

 55%|█████▌    | 13450/24420 [5:36:16<4:31:27,  1.48s/it]
 55%|█████▌    | 13451/24420 [5:36:17<4:31:31,  1.49s/it]
 55%|█████▌    | 13452/24420 [5:36:19<4:31:09,  1.48s/it]
 55%|█████▌    | 13453/24420 [5:36:20<4:31:29,  1.49s/it]
 55%|█████▌    | 13454/24420 [5:36:22<4:31:09,  1.48s/it]
 55%|█████▌    | 13455/24420 [5:36:23<4:31:58,  1.49s/it]
                                                         
{'loss': 0.083, 'learning_rate': 5e-05, 'epoch': 2.75}

 55%|█████▌    | 13455/24420 [5:36:23<4:31:58,  1.49s/it]
 55%|█████▌    | 13456/24420 [5:36:25<4:33:29,  1.50s/it]
 55%|█████▌    | 13457/24420 [5:36:26<4:32:58,  1.49s/it]
 55%|█████▌    | 13458/24420 [5:36:28<4:32:26,  1.49s/it]
 55%|█████▌    | 13459/24420 [5:36:29<4:32:06,  1.49s/it]
 55%|█████▌    | 13460/24420 [5:36:31<4:32:10,  1.49s/it]
                                                         
{'loss': 0.0933, 'learning_rate': 5e-05, 'epoch': 2.76}

 55%|█████▌    | 13460/24420 [5:36:31<4:32:10,  1.49s/it]
 55%|█████▌    | 13461/24420 [5:36:32<4:32:17,  1.49s/it]
 55%|█████▌    | 13462/24420 [5:36:34<4:32:05,  1.49s/it]
 55%|█████▌    | 13463/24420 [5:36:35<4:31:59,  1.49s/it]
 55%|█████▌    | 13464/24420 [5:36:37<4:31:28,  1.49s/it]
 55%|█████▌    | 13465/24420 [5:36:38<4:31:42,  1.49s/it]
                                                         
{'loss': 0.0869, 'learning_rate': 5e-05, 'epoch': 2.76}

 55%|█████▌    | 13465/24420 [5:36:38<4:31:42,  1.49s/it]
 55%|█████▌    | 13466/24420 [5:36:40<4:31:15,  1.49s/it]
 55%|█████▌    | 13467/24420 [5:36:41<4:31:43,  1.49s/it]
 55%|█████▌    | 13468/24420 [5:36:43<4:33:12,  1.50s/it]
 55%|█████▌    | 13469/24420 [5:36:44<4:32:49,  1.49s/it]
 55%|█████▌    | 13470/24420 [5:36:46<4:31:49,  1.49s/it]
                                                         
{'loss': 0.0883, 'learning_rate': 5e-05, 'epoch': 2.76}

 55%|█████▌    | 13470/24420 [5:36:46<4:31:49,  1.49s/it]
 55%|█████▌    | 13471/24420 [5:36:47<4:32:10,  1.49s/it]
 55%|█████▌    | 13472/24420 [5:36:49<4:32:36,  1.49s/it]
 55%|█████▌    | 13473/24420 [5:36:50<4:31:56,  1.49s/it]
 55%|█████▌    | 13474/24420 [5:36:52<4:31:35,  1.49s/it]
 55%|█████▌    | 13475/24420 [5:36:53<4:31:55,  1.49s/it]
                                                         
{'loss': 0.0815, 'learning_rate': 5e-05, 'epoch': 2.76}

 55%|█████▌    | 13475/24420 [5:36:53<4:31:55,  1.49s/it]
 55%|█████▌    | 13476/24420 [5:36:55<4:31:22,  1.49s/it]
 55%|█████▌    | 13477/24420 [5:36:56<4:30:54,  1.49s/it]
 55%|█████▌    | 13478/24420 [5:36:58<4:30:52,  1.49s/it]
 55%|█████▌    | 13479/24420 [5:36:59<4:31:17,  1.49s/it]
 55%|█████▌    | 13480/24420 [5:37:01<4:31:19,  1.49s/it]
                                                         
{'loss': 0.0834, 'learning_rate': 5e-05, 'epoch': 2.76}

 55%|█████▌    | 13480/24420 [5:37:01<4:31:19,  1.49s/it]
 55%|█████▌    | 13481/24420 [5:37:02<4:31:29,  1.49s/it]
 55%|█████▌    | 13482/24420 [5:37:04<4:31:24,  1.49s/it]
 55%|█████▌    | 13483/24420 [5:37:05<4:31:12,  1.49s/it]
 55%|█████▌    | 13484/24420 [5:37:07<4:30:32,  1.48s/it]
 55%|█████▌    | 13485/24420 [5:37:08<4:31:09,  1.49s/it]
                                                         
{'loss': 0.0888, 'learning_rate': 5e-05, 'epoch': 2.76}

 55%|█████▌    | 13485/24420 [5:37:08<4:31:09,  1.49s/it]
 55%|█████▌    | 13486/24420 [5:37:10<4:30:55,  1.49s/it]
 55%|█████▌    | 13487/24420 [5:37:11<4:31:12,  1.49s/it]
 55%|█████▌    | 13488/24420 [5:37:13<4:30:54,  1.49s/it]
 55%|█████▌    | 13489/24420 [5:37:14<4:30:39,  1.49s/it]
 55%|█████▌    | 13490/24420 [5:37:16<4:30:15,  1.48s/it]
                                                         
{'loss': 0.102, 'learning_rate': 5e-05, 'epoch': 2.76}

 55%|█████▌    | 13490/24420 [5:37:16<4:30:15,  1.48s/it]
 55%|█████▌    | 13491/24420 [5:37:17<4:30:03,  1.48s/it]
 55%|█████▌    | 13492/24420 [5:37:19<4:30:14,  1.48s/it]
 55%|█████▌    | 13493/24420 [5:37:20<4:30:36,  1.49s/it]
 55%|█████▌    | 13494/24420 [5:37:21<4:30:33,  1.49s/it]
 55%|█████▌    | 13495/24420 [5:37:23<4:30:24,  1.49s/it]
                                                         
{'loss': 0.0677, 'learning_rate': 5e-05, 'epoch': 2.76}

 55%|█████▌    | 13495/24420 [5:37:23<4:30:24,  1.49s/it]
 55%|█████▌    | 13496/24420 [5:37:24<4:30:53,  1.49s/it]
 55%|█████▌    | 13497/24420 [5:37:26<4:30:43,  1.49s/it]
 55%|█████▌    | 13498/24420 [5:37:27<4:30:25,  1.49s/it]
 55%|█████▌    | 13499/24420 [5:37:29<4:30:53,  1.49s/it]
 55%|█████▌    | 13500/24420 [5:37:30<4:30:50,  1.49s/it]
                                                         
{'loss': 0.0979, 'learning_rate': 5e-05, 'epoch': 2.76}

 55%|█████▌    | 13500/24420 [5:37:30<4:30:50,  1.49s/it]
 55%|█████▌    | 13501/24420 [5:37:32<4:34:44,  1.51s/it]
 55%|█████▌    | 13502/24420 [5:37:33<4:33:30,  1.50s/it]
 55%|█████▌    | 13503/24420 [5:37:35<4:32:23,  1.50s/it]
 55%|█████▌    | 13504/24420 [5:37:36<4:31:19,  1.49s/it]
 55%|█████▌    | 13505/24420 [5:37:38<4:30:48,  1.49s/it]
                                                         
{'loss': 0.1074, 'learning_rate': 5e-05, 'epoch': 2.77}

 55%|█████▌    | 13505/24420 [5:37:38<4:30:48,  1.49s/it]
 55%|█████▌    | 13506/24420 [5:37:39<4:31:02,  1.49s/it]
 55%|█████▌    | 13507/24420 [5:37:41<4:30:39,  1.49s/it]
 55%|█████▌    | 13508/24420 [5:37:42<4:30:45,  1.49s/it]
 55%|█████▌    | 13509/24420 [5:37:44<4:30:32,  1.49s/it]
 55%|█████▌    | 13510/24420 [5:37:45<4:30:36,  1.49s/it]
                                                         
{'loss': 0.0794, 'learning_rate': 5e-05, 'epoch': 2.77}

 55%|█████▌    | 13510/24420 [5:37:45<4:30:36,  1.49s/it]
 55%|█████▌    | 13511/24420 [5:37:47<4:30:51,  1.49s/it]
 55%|█████▌    | 13512/24420 [5:37:48<4:30:12,  1.49s/it]
 55%|█████▌    | 13513/24420 [5:37:50<4:29:42,  1.48s/it]
 55%|█████▌    | 13514/24420 [5:37:51<4:29:53,  1.48s/it]
 55%|█████▌    | 13515/24420 [5:37:53<4:29:55,  1.49s/it]
                                                         
{'loss': 0.09, 'learning_rate': 5e-05, 'epoch': 2.77}

 55%|█████▌    | 13515/24420 [5:37:53<4:29:55,  1.49s/it]
 55%|█████▌    | 13516/24420 [5:37:54<4:29:57,  1.49s/it]
 55%|█████▌    | 13517/24420 [5:37:56<4:29:57,  1.49s/it]
 55%|█████▌    | 13518/24420 [5:37:57<4:29:51,  1.49s/it]
 55%|█████▌    | 13519/24420 [5:37:59<4:29:29,  1.48s/it]
 55%|█████▌    | 13520/24420 [5:38:00<4:29:21,  1.48s/it]
                                                         
{'loss': 0.082, 'learning_rate': 5e-05, 'epoch': 2.77}

 55%|█████▌    | 13520/24420 [5:38:00<4:29:21,  1.48s/it]
 55%|█████▌    | 13521/24420 [5:38:02<4:29:20,  1.48s/it]
 55%|█████▌    | 13522/24420 [5:38:03<4:29:23,  1.48s/it]
 55%|█████▌    | 13523/24420 [5:38:05<4:29:53,  1.49s/it]
 55%|█████▌    | 13524/24420 [5:38:06<4:30:09,  1.49s/it]
 55%|█████▌    | 13525/24420 [5:38:08<4:30:29,  1.49s/it]
                                                         
{'loss': 0.0866, 'learning_rate': 5e-05, 'epoch': 2.77}

 55%|█████▌    | 13525/24420 [5:38:08<4:30:29,  1.49s/it]
 55%|█████▌    | 13526/24420 [5:38:09<4:30:36,  1.49s/it]
 55%|█████▌    | 13527/24420 [5:38:11<4:30:32,  1.49s/it]
 55%|█████▌    | 13528/24420 [5:38:12<4:29:53,  1.49s/it]
 55%|█████▌    | 13529/24420 [5:38:14<4:29:15,  1.48s/it]
 55%|█████▌    | 13530/24420 [5:38:15<4:29:26,  1.48s/it]
                                                         
{'loss': 0.0789, 'learning_rate': 5e-05, 'epoch': 2.77}

 55%|█████▌    | 13530/24420 [5:38:15<4:29:26,  1.48s/it]
 55%|█████▌    | 13531/24420 [5:38:17<4:29:39,  1.49s/it]
 55%|█████▌    | 13532/24420 [5:38:18<4:29:13,  1.48s/it]
 55%|█████▌    | 13533/24420 [5:38:20<4:28:57,  1.48s/it]
 55%|█████▌    | 13534/24420 [5:38:21<4:29:03,  1.48s/it]
 55%|█████▌    | 13535/24420 [5:38:22<4:29:25,  1.49s/it]
                                                         
{'loss': 0.1008, 'learning_rate': 5e-05, 'epoch': 2.77}

 55%|█████▌    | 13535/24420 [5:38:22<4:29:25,  1.49s/it]
 55%|█████▌    | 13536/24420 [5:38:24<4:29:35,  1.49s/it]
 55%|█████▌    | 13537/24420 [5:38:25<4:30:18,  1.49s/it]
 55%|█████▌    | 13538/24420 [5:38:27<4:29:43,  1.49s/it]
 55%|█████▌    | 13539/24420 [5:38:28<4:29:40,  1.49s/it]
 55%|█████▌    | 13540/24420 [5:38:30<4:29:55,  1.49s/it]
                                                         
{'loss': 0.0874, 'learning_rate': 5e-05, 'epoch': 2.77}

 55%|█████▌    | 13540/24420 [5:38:30<4:29:55,  1.49s/it]
 55%|█████▌    | 13541/24420 [5:38:31<4:29:57,  1.49s/it]
 55%|█████▌    | 13542/24420 [5:38:33<4:29:45,  1.49s/it]
 55%|█████▌    | 13543/24420 [5:38:34<4:29:37,  1.49s/it]
 55%|█████▌    | 13544/24420 [5:38:36<4:29:44,  1.49s/it]
 55%|█████▌    | 13545/24420 [5:38:37<4:29:34,  1.49s/it]
                                                         
{'loss': 0.0735, 'learning_rate': 5e-05, 'epoch': 2.77}

 55%|█████▌    | 13545/24420 [5:38:37<4:29:34,  1.49s/it]
 55%|█████▌    | 13546/24420 [5:38:39<4:29:33,  1.49s/it]
 55%|█████▌    | 13547/24420 [5:38:40<4:29:52,  1.49s/it]
 55%|█████▌    | 13548/24420 [5:38:42<4:30:12,  1.49s/it]
 55%|█████▌    | 13549/24420 [5:38:43<4:29:55,  1.49s/it]
 55%|█████▌    | 13550/24420 [5:38:45<4:29:21,  1.49s/it]
                                                         
{'loss': 0.0937, 'learning_rate': 5e-05, 'epoch': 2.77}

 55%|█████▌    | 13550/24420 [5:38:45<4:29:21,  1.49s/it]
 55%|█████▌    | 13551/24420 [5:38:46<4:29:11,  1.49s/it]
 55%|█████▌    | 13552/24420 [5:38:48<4:29:20,  1.49s/it]
 55%|█████▌    | 13553/24420 [5:38:49<4:29:06,  1.49s/it]
 56%|█████▌    | 13554/24420 [5:38:51<4:29:19,  1.49s/it]
 56%|█████▌    | 13555/24420 [5:38:52<4:29:28,  1.49s/it]
                                                         
{'loss': 0.1001, 'learning_rate': 5e-05, 'epoch': 2.78}

 56%|█████▌    | 13555/24420 [5:38:52<4:29:28,  1.49s/it]
 56%|█████▌    | 13556/24420 [5:38:54<4:29:50,  1.49s/it]
 56%|█████▌    | 13557/24420 [5:38:55<4:29:39,  1.49s/it]
 56%|█████▌    | 13558/24420 [5:38:57<4:29:38,  1.49s/it]
 56%|█████▌    | 13559/24420 [5:38:58<4:29:29,  1.49s/it]
 56%|█████▌    | 13560/24420 [5:39:00<4:29:12,  1.49s/it]
                                                         
{'loss': 0.0796, 'learning_rate': 5e-05, 'epoch': 2.78}

 56%|█████▌    | 13560/24420 [5:39:00<4:29:12,  1.49s/it]
 56%|█████▌    | 13561/24420 [5:39:01<4:30:05,  1.49s/it]
 56%|█████▌    | 13562/24420 [5:39:03<4:29:55,  1.49s/it]
 56%|█████▌    | 13563/24420 [5:39:04<4:30:37,  1.50s/it]
 56%|█████▌    | 13564/24420 [5:39:06<4:30:09,  1.49s/it]
 56%|█████▌    | 13565/24420 [5:39:07<4:30:18,  1.49s/it]
                                                         
{'loss': 0.0816, 'learning_rate': 5e-05, 'epoch': 2.78}

 56%|█████▌    | 13565/24420 [5:39:07<4:30:18,  1.49s/it]
 56%|█████▌    | 13566/24420 [5:39:09<4:30:02,  1.49s/it]
 56%|█████▌    | 13567/24420 [5:39:10<4:29:58,  1.49s/it]
 56%|█████▌    | 13568/24420 [5:39:12<4:29:26,  1.49s/it]
 56%|█████▌    | 13569/24420 [5:39:13<4:28:55,  1.49s/it]
 56%|█████▌    | 13570/24420 [5:39:15<4:28:44,  1.49s/it]
                                                         
{'loss': 0.0748, 'learning_rate': 5e-05, 'epoch': 2.78}

 56%|█████▌    | 13570/24420 [5:39:15<4:28:44,  1.49s/it]
 56%|█████▌    | 13571/24420 [5:39:16<4:28:33,  1.49s/it]
 56%|█████▌    | 13572/24420 [5:39:18<4:28:31,  1.49s/it]
 56%|█████▌    | 13573/24420 [5:39:19<4:28:33,  1.49s/it]
 56%|█████▌    | 13574/24420 [5:39:21<4:28:48,  1.49s/it]
 56%|█████▌    | 13575/24420 [5:39:22<4:28:34,  1.49s/it]
                                                         
{'loss': 0.0923, 'learning_rate': 5e-05, 'epoch': 2.78}

 56%|█████▌    | 13575/24420 [5:39:22<4:28:34,  1.49s/it]
 56%|█████▌    | 13576/24420 [5:39:24<4:28:44,  1.49s/it]
 56%|█████▌    | 13577/24420 [5:39:25<4:28:40,  1.49s/it]
 56%|█████▌    | 13578/24420 [5:39:26<4:28:22,  1.49s/it]
 56%|█████▌    | 13579/24420 [5:39:28<4:28:15,  1.48s/it]
 56%|█████▌    | 13580/24420 [5:39:29<4:28:12,  1.48s/it]
                                                         
{'loss': 0.0771, 'learning_rate': 5e-05, 'epoch': 2.78}

 56%|█████▌    | 13580/24420 [5:39:29<4:28:12,  1.48s/it]
 56%|█████▌    | 13581/24420 [5:39:31<4:28:47,  1.49s/it]
 56%|█████▌    | 13582/24420 [5:39:32<4:28:52,  1.49s/it]
 56%|█████▌    | 13583/24420 [5:39:34<4:28:55,  1.49s/it]
 56%|█████▌    | 13584/24420 [5:39:35<4:28:54,  1.49s/it]
 56%|█████▌    | 13585/24420 [5:39:37<4:28:30,  1.49s/it]
                                                         
{'loss': 0.0807, 'learning_rate': 5e-05, 'epoch': 2.78}

 56%|█████▌    | 13585/24420 [5:39:37<4:28:30,  1.49s/it]
 56%|█████▌    | 13586/24420 [5:39:38<4:28:47,  1.49s/it]
 56%|█████▌    | 13587/24420 [5:39:40<4:30:32,  1.50s/it]
 56%|█████▌    | 13588/24420 [5:39:41<4:30:51,  1.50s/it]
 56%|█████▌    | 13589/24420 [5:39:43<4:29:44,  1.49s/it]
 56%|█████▌    | 13590/24420 [5:39:44<4:29:18,  1.49s/it]
                                                         
{'loss': 0.0926, 'learning_rate': 5e-05, 'epoch': 2.78}

 56%|█████▌    | 13590/24420 [5:39:44<4:29:18,  1.49s/it]
 56%|█████▌    | 13591/24420 [5:39:46<4:29:11,  1.49s/it]
 56%|█████▌    | 13592/24420 [5:39:47<4:29:17,  1.49s/it]
 56%|█████▌    | 13593/24420 [5:39:49<4:29:23,  1.49s/it]
 56%|█████▌    | 13594/24420 [5:39:50<4:29:04,  1.49s/it]
 56%|█████▌    | 13595/24420 [5:39:52<4:28:37,  1.49s/it]
                                                         
{'loss': 0.0825, 'learning_rate': 5e-05, 'epoch': 2.78}

 56%|█████▌    | 13595/24420 [5:39:52<4:28:37,  1.49s/it]
 56%|█████▌    | 13596/24420 [5:39:53<4:28:46,  1.49s/it]
 56%|█████▌    | 13597/24420 [5:39:55<4:29:11,  1.49s/it]
 56%|█████▌    | 13598/24420 [5:39:56<4:28:23,  1.49s/it]
 56%|█████▌    | 13599/24420 [5:39:58<4:28:22,  1.49s/it]
 56%|█████▌    | 13600/24420 [5:39:59<4:28:36,  1.49s/it]
                                                         
{'loss': 0.08, 'learning_rate': 5e-05, 'epoch': 2.78}

 56%|█████▌    | 13600/24420 [5:39:59<4:28:36,  1.49s/it]
 56%|█████▌    | 13601/24420 [5:40:04<7:17:23,  2.43s/it]
 56%|█████▌    | 13602/24420 [5:40:05<6:26:52,  2.15s/it]
 56%|█████▌    | 13603/24420 [5:40:07<5:51:42,  1.95s/it]
 56%|█████▌    | 13604/24420 [5:40:08<5:26:34,  1.81s/it]
 56%|█████▌    | 13605/24420 [5:40:10<5:08:51,  1.71s/it]
                                                         
{'loss': 0.0853, 'learning_rate': 5e-05, 'epoch': 2.79}

 56%|█████▌    | 13605/24420 [5:40:10<5:08:51,  1.71s/it]
 56%|█████▌    | 13606/24420 [5:40:11<4:57:07,  1.65s/it]
 56%|█████▌    | 13607/24420 [5:40:13<4:48:36,  1.60s/it]
 56%|█████▌    | 13608/24420 [5:40:14<4:42:50,  1.57s/it]
 56%|█████▌    | 13609/24420 [5:40:16<4:38:30,  1.55s/it]
 56%|█████▌    | 13610/24420 [5:40:17<4:35:31,  1.53s/it]
                                                         
{'loss': 0.0987, 'learning_rate': 5e-05, 'epoch': 2.79}

 56%|█████▌    | 13610/24420 [5:40:17<4:35:31,  1.53s/it]
 56%|█████▌    | 13611/24420 [5:40:19<4:33:07,  1.52s/it]
 56%|█████▌    | 13612/24420 [5:40:20<4:31:07,  1.51s/it]
 56%|█████▌    | 13613/24420 [5:40:22<4:30:01,  1.50s/it]
 56%|█████▌    | 13614/24420 [5:40:23<4:29:02,  1.49s/it]
 56%|█████▌    | 13615/24420 [5:40:25<4:28:40,  1.49s/it]
                                                         
{'loss': 0.0838, 'learning_rate': 5e-05, 'epoch': 2.79}

 56%|█████▌    | 13615/24420 [5:40:25<4:28:40,  1.49s/it]
 56%|█████▌    | 13616/24420 [5:40:26<4:28:41,  1.49s/it]
 56%|█████▌    | 13617/24420 [5:40:28<4:29:07,  1.49s/it]
 56%|█████▌    | 13618/24420 [5:40:29<4:28:51,  1.49s/it]
 56%|█████▌    | 13619/24420 [5:40:31<4:29:11,  1.50s/it]
 56%|█████▌    | 13620/24420 [5:40:32<4:28:41,  1.49s/it]
                                                         
{'loss': 0.0886, 'learning_rate': 5e-05, 'epoch': 2.79}

 56%|█████▌    | 13620/24420 [5:40:32<4:28:41,  1.49s/it]
 56%|█████▌    | 13621/24420 [5:40:34<4:28:23,  1.49s/it]
 56%|█████▌    | 13622/24420 [5:40:35<4:27:41,  1.49s/it]
 56%|█████▌    | 13623/24420 [5:40:37<4:27:33,  1.49s/it]
 56%|█████▌    | 13624/24420 [5:40:38<4:27:35,  1.49s/it]
 56%|█████▌    | 13625/24420 [5:40:40<4:27:44,  1.49s/it]
                                                         
{'loss': 0.0901, 'learning_rate': 5e-05, 'epoch': 2.79}

 56%|█████▌    | 13625/24420 [5:40:40<4:27:44,  1.49s/it]
 56%|█████▌    | 13626/24420 [5:40:41<4:27:59,  1.49s/it]
 56%|█████▌    | 13627/24420 [5:40:43<4:28:24,  1.49s/it]
 56%|█████▌    | 13628/24420 [5:40:44<4:27:41,  1.49s/it]
 56%|█████▌    | 13629/24420 [5:40:46<4:27:10,  1.49s/it]
 56%|█████▌    | 13630/24420 [5:40:47<4:26:37,  1.48s/it]
                                                         
{'loss': 0.0912, 'learning_rate': 5e-05, 'epoch': 2.79}

 56%|█████▌    | 13630/24420 [5:40:47<4:26:37,  1.48s/it]
 56%|█████▌    | 13631/24420 [5:40:49<4:26:50,  1.48s/it]
 56%|█████▌    | 13632/24420 [5:40:50<4:26:41,  1.48s/it]
 56%|█████▌    | 13633/24420 [5:40:52<4:27:03,  1.49s/it]
 56%|█████▌    | 13634/24420 [5:40:53<4:27:34,  1.49s/it]
 56%|█████▌    | 13635/24420 [5:40:55<4:27:06,  1.49s/it]
                                                         
{'loss': 0.0881, 'learning_rate': 5e-05, 'epoch': 2.79}

 56%|█████▌    | 13635/24420 [5:40:55<4:27:06,  1.49s/it]
 56%|█████▌    | 13636/24420 [5:40:56<4:27:07,  1.49s/it]
 56%|█████▌    | 13637/24420 [5:40:57<4:27:29,  1.49s/it]
 56%|█████▌    | 13638/24420 [5:40:59<4:26:54,  1.49s/it]
 56%|█████▌    | 13639/24420 [5:41:00<4:26:49,  1.48s/it]
 56%|█████▌    | 13640/24420 [5:41:02<4:27:01,  1.49s/it]
                                                         
{'loss': 0.0865, 'learning_rate': 5e-05, 'epoch': 2.79}

 56%|█████▌    | 13640/24420 [5:41:02<4:27:01,  1.49s/it]
 56%|█████▌    | 13641/24420 [5:41:03<4:27:27,  1.49s/it]
 56%|█████▌    | 13642/24420 [5:41:05<4:27:24,  1.49s/it]
 56%|█████▌    | 13643/24420 [5:41:06<4:27:13,  1.49s/it]
 56%|█████▌    | 13644/24420 [5:41:08<4:27:13,  1.49s/it]
 56%|█████▌    | 13645/24420 [5:41:09<4:26:55,  1.49s/it]
                                                         
{'loss': 0.0904, 'learning_rate': 5e-05, 'epoch': 2.79}

 56%|█████▌    | 13645/24420 [5:41:09<4:26:55,  1.49s/it]
 56%|█████▌    | 13646/24420 [5:41:11<4:27:14,  1.49s/it]
 56%|█████▌    | 13647/24420 [5:41:12<4:27:12,  1.49s/it]
 56%|█████▌    | 13648/24420 [5:41:14<4:28:03,  1.49s/it]
 56%|█████▌    | 13649/24420 [5:41:15<4:27:23,  1.49s/it]
 56%|█████▌    | 13650/24420 [5:41:17<4:26:37,  1.49s/it]
                                                         
{'loss': 0.0749, 'learning_rate': 5e-05, 'epoch': 2.79}

 56%|█████▌    | 13650/24420 [5:41:17<4:26:37,  1.49s/it]
 56%|█████▌    | 13651/24420 [5:41:18<4:26:47,  1.49s/it]
 56%|█████▌    | 13652/24420 [5:41:20<4:26:47,  1.49s/it]
 56%|█████▌    | 13653/24420 [5:41:21<4:26:49,  1.49s/it]
 56%|█████▌    | 13654/24420 [5:41:23<4:26:24,  1.48s/it]
 56%|█████▌    | 13655/24420 [5:41:24<4:26:16,  1.48s/it]
                                                         
{'loss': 0.0977, 'learning_rate': 5e-05, 'epoch': 2.8}

 56%|█████▌    | 13655/24420 [5:41:24<4:26:16,  1.48s/it]
 56%|█████▌    | 13656/24420 [5:41:26<4:26:06,  1.48s/it]
 56%|█████▌    | 13657/24420 [5:41:27<4:26:24,  1.49s/it]
 56%|█████▌    | 13658/24420 [5:41:29<4:27:16,  1.49s/it]
 56%|█████▌    | 13659/24420 [5:41:30<4:26:48,  1.49s/it]
 56%|█████▌    | 13660/24420 [5:41:32<4:26:22,  1.49s/it]
                                                         
{'loss': 0.1176, 'learning_rate': 5e-05, 'epoch': 2.8}

 56%|█████▌    | 13660/24420 [5:41:32<4:26:22,  1.49s/it]
 56%|█████▌    | 13661/24420 [5:41:33<4:26:42,  1.49s/it]
 56%|█████▌    | 13662/24420 [5:41:35<4:26:32,  1.49s/it]
 56%|█████▌    | 13663/24420 [5:41:36<4:26:35,  1.49s/it]
 56%|█████▌    | 13664/24420 [5:41:38<4:26:12,  1.49s/it]
 56%|█████▌    | 13665/24420 [5:41:39<4:26:23,  1.49s/it]
                                                         
{'loss': 0.0863, 'learning_rate': 5e-05, 'epoch': 2.8}

 56%|█████▌    | 13665/24420 [5:41:39<4:26:23,  1.49s/it]
 56%|█████▌    | 13666/24420 [5:41:41<4:26:24,  1.49s/it]
 56%|█████▌    | 13667/24420 [5:41:42<4:26:32,  1.49s/it]
 56%|█████▌    | 13668/24420 [5:41:44<4:26:35,  1.49s/it]
 56%|█████▌    | 13669/24420 [5:41:45<4:27:23,  1.49s/it]
 56%|█████▌    | 13670/24420 [5:41:47<4:27:57,  1.50s/it]
                                                         
{'loss': 0.0932, 'learning_rate': 5e-05, 'epoch': 2.8}

 56%|█████▌    | 13670/24420 [5:41:47<4:27:57,  1.50s/it]
 56%|█████▌    | 13671/24420 [5:41:48<4:27:53,  1.50s/it]
 56%|█████▌    | 13672/24420 [5:41:50<4:27:15,  1.49s/it]
 56%|█████▌    | 13673/24420 [5:41:51<4:26:48,  1.49s/it]
 56%|█████▌    | 13674/24420 [5:41:53<4:26:22,  1.49s/it]
 56%|█████▌    | 13675/24420 [5:41:54<4:26:01,  1.49s/it]
                                                         
{'loss': 0.0875, 'learning_rate': 5e-05, 'epoch': 2.8}

 56%|█████▌    | 13675/24420 [5:41:54<4:26:01,  1.49s/it]
 56%|█████▌    | 13676/24420 [5:41:55<4:25:53,  1.48s/it]
 56%|█████▌    | 13677/24420 [5:41:57<4:26:41,  1.49s/it]
 56%|█████▌    | 13678/24420 [5:41:58<4:26:25,  1.49s/it]
 56%|█████▌    | 13679/24420 [5:42:00<4:26:05,  1.49s/it]
 56%|█████▌    | 13680/24420 [5:42:01<4:26:08,  1.49s/it]
                                                         
{'loss': 0.0955, 'learning_rate': 5e-05, 'epoch': 2.8}

 56%|█████▌    | 13680/24420 [5:42:01<4:26:08,  1.49s/it]
 56%|█████▌    | 13681/24420 [5:42:03<4:26:07,  1.49s/it]
 56%|█████▌    | 13682/24420 [5:42:04<4:26:06,  1.49s/it]
 56%|█████▌    | 13683/24420 [5:42:06<4:26:15,  1.49s/it]
 56%|█████▌    | 13684/24420 [5:42:07<4:26:06,  1.49s/it]
 56%|█████▌    | 13685/24420 [5:42:09<4:26:38,  1.49s/it]
                                                         
{'loss': 0.0925, 'learning_rate': 5e-05, 'epoch': 2.8}

 56%|█████▌    | 13685/24420 [5:42:09<4:26:38,  1.49s/it]
 56%|█████▌    | 13686/24420 [5:42:10<4:26:57,  1.49s/it]
 56%|█████▌    | 13687/24420 [5:42:12<4:26:38,  1.49s/it]
 56%|█████▌    | 13688/24420 [5:42:13<4:27:22,  1.49s/it]
 56%|█████▌    | 13689/24420 [5:42:15<4:27:46,  1.50s/it]
 56%|█████▌    | 13690/24420 [5:42:16<4:27:31,  1.50s/it]
                                                         
{'loss': 0.1068, 'learning_rate': 5e-05, 'epoch': 2.8}

 56%|█████▌    | 13690/24420 [5:42:16<4:27:31,  1.50s/it]
 56%|█████▌    | 13691/24420 [5:42:18<4:27:24,  1.50s/it]
 56%|█████▌    | 13692/24420 [5:42:19<4:26:37,  1.49s/it]
 56%|█████▌    | 13693/24420 [5:42:21<4:25:53,  1.49s/it]
 56%|█████▌    | 13694/24420 [5:42:22<4:26:38,  1.49s/it]
 56%|█████▌    | 13695/24420 [5:42:24<4:26:29,  1.49s/it]
                                                         
{'loss': 0.0976, 'learning_rate': 5e-05, 'epoch': 2.8}

 56%|█████▌    | 13695/24420 [5:42:24<4:26:29,  1.49s/it]
 56%|█████▌    | 13696/24420 [5:42:25<4:26:07,  1.49s/it]
 56%|█████▌    | 13697/24420 [5:42:27<4:26:33,  1.49s/it]
 56%|█████▌    | 13698/24420 [5:42:28<4:26:20,  1.49s/it]
 56%|█████▌    | 13699/24420 [5:42:30<4:26:43,  1.49s/it]
 56%|█████▌    | 13700/24420 [5:42:31<4:26:30,  1.49s/it]
                                                         
{'loss': 0.0727, 'learning_rate': 5e-05, 'epoch': 2.81}

 56%|█████▌    | 13700/24420 [5:42:31<4:26:30,  1.49s/it]
 56%|█████▌    | 13701/24420 [5:42:33<4:26:35,  1.49s/it]
 56%|█████▌    | 13702/24420 [5:42:34<4:26:08,  1.49s/it]
 56%|█████▌    | 13703/24420 [5:42:36<4:25:40,  1.49s/it]
 56%|█████▌    | 13704/24420 [5:42:37<4:25:32,  1.49s/it]
 56%|█████▌    | 13705/24420 [5:42:39<4:26:10,  1.49s/it]
                                                         
{'loss': 0.0886, 'learning_rate': 5e-05, 'epoch': 2.81}

 56%|█████▌    | 13705/24420 [5:42:39<4:26:10,  1.49s/it]
 56%|█████▌    | 13706/24420 [5:42:40<4:26:41,  1.49s/it]
 56%|█████▌    | 13707/24420 [5:42:42<4:25:56,  1.49s/it]
 56%|█████▌    | 13708/24420 [5:42:43<4:25:40,  1.49s/it]
 56%|█████▌    | 13709/24420 [5:42:45<4:25:42,  1.49s/it]
 56%|█████▌    | 13710/24420 [5:42:46<4:25:53,  1.49s/it]
                                                         
{'loss': 0.0715, 'learning_rate': 5e-05, 'epoch': 2.81}

 56%|█████▌    | 13710/24420 [5:42:46<4:25:53,  1.49s/it]
 56%|█████▌    | 13711/24420 [5:42:48<4:25:55,  1.49s/it]
 56%|█████▌    | 13712/24420 [5:42:49<4:26:15,  1.49s/it]
 56%|█████▌    | 13713/24420 [5:42:51<4:26:33,  1.49s/it]
 56%|█████▌    | 13714/24420 [5:42:52<4:25:52,  1.49s/it]
 56%|█████▌    | 13715/24420 [5:42:54<4:25:33,  1.49s/it]
                                                         
{'loss': 0.0835, 'learning_rate': 5e-05, 'epoch': 2.81}

 56%|█████▌    | 13715/24420 [5:42:54<4:25:33,  1.49s/it]
 56%|█████▌    | 13716/24420 [5:42:55<4:25:34,  1.49s/it]
 56%|█████▌    | 13717/24420 [5:42:57<4:25:18,  1.49s/it]
 56%|█████▌    | 13718/24420 [5:42:58<4:24:47,  1.48s/it]
 56%|█████▌    | 13719/24420 [5:43:00<4:24:46,  1.48s/it]
 56%|█████▌    | 13720/24420 [5:43:01<4:24:55,  1.49s/it]
                                                         
{'loss': 0.0868, 'learning_rate': 5e-05, 'epoch': 2.81}

 56%|█████▌    | 13720/24420 [5:43:01<4:24:55,  1.49s/it]
 56%|█████▌    | 13721/24420 [5:43:03<4:24:54,  1.49s/it]
 56%|█████▌    | 13722/24420 [5:43:04<4:25:02,  1.49s/it]
 56%|█████▌    | 13723/24420 [5:43:06<4:24:47,  1.49s/it]
 56%|█████▌    | 13724/24420 [5:43:07<4:24:40,  1.48s/it]
 56%|█████▌    | 13725/24420 [5:43:08<4:24:36,  1.48s/it]
                                                         
{'loss': 0.0917, 'learning_rate': 5e-05, 'epoch': 2.81}

 56%|█████▌    | 13725/24420 [5:43:08<4:24:36,  1.48s/it]
 56%|█████▌    | 13726/24420 [5:43:10<4:25:14,  1.49s/it]
 56%|█████▌    | 13727/24420 [5:43:11<4:25:06,  1.49s/it]
 56%|█████▌    | 13728/24420 [5:43:13<4:25:27,  1.49s/it]
 56%|█████▌    | 13729/24420 [5:43:14<4:25:05,  1.49s/it]
 56%|█████▌    | 13730/24420 [5:43:16<4:25:31,  1.49s/it]
                                                         
{'loss': 0.1143, 'learning_rate': 5e-05, 'epoch': 2.81}

 56%|█████▌    | 13730/24420 [5:43:16<4:25:31,  1.49s/it]
 56%|█████▌    | 13731/24420 [5:43:17<4:25:24,  1.49s/it]
 56%|█████▌    | 13732/24420 [5:43:19<4:25:00,  1.49s/it]
 56%|█████▌    | 13733/24420 [5:43:20<4:24:50,  1.49s/it]
 56%|█████▌    | 13734/24420 [5:43:22<4:24:56,  1.49s/it]
 56%|█████▌    | 13735/24420 [5:43:23<4:24:44,  1.49s/it]
                                                         
{'loss': 0.0817, 'learning_rate': 5e-05, 'epoch': 2.81}

 56%|█████▌    | 13735/24420 [5:43:23<4:24:44,  1.49s/it]
 56%|█████▌    | 13736/24420 [5:43:25<4:24:55,  1.49s/it]
 56%|█████▋    | 13737/24420 [5:43:26<4:25:14,  1.49s/it]
 56%|█████▋    | 13738/24420 [5:43:28<4:25:14,  1.49s/it]
 56%|█████▋    | 13739/24420 [5:43:29<4:24:46,  1.49s/it]
 56%|█████▋    | 13740/24420 [5:43:31<4:24:39,  1.49s/it]
                                                         
{'loss': 0.0939, 'learning_rate': 5e-05, 'epoch': 2.81}

 56%|█████▋    | 13740/24420 [5:43:31<4:24:39,  1.49s/it]
 56%|█████▋    | 13741/24420 [5:43:32<4:24:48,  1.49s/it]
 56%|█████▋    | 13742/24420 [5:43:34<4:24:40,  1.49s/it]
 56%|█████▋    | 13743/24420 [5:43:35<4:25:14,  1.49s/it]
 56%|█████▋    | 13744/24420 [5:43:37<4:24:38,  1.49s/it]
 56%|█████▋    | 13745/24420 [5:43:38<4:24:43,  1.49s/it]
                                                         
{'loss': 0.0866, 'learning_rate': 5e-05, 'epoch': 2.81}

 56%|█████▋    | 13745/24420 [5:43:38<4:24:43,  1.49s/it]
 56%|█████▋    | 13746/24420 [5:43:40<4:24:57,  1.49s/it]
 56%|█████▋    | 13747/24420 [5:43:41<4:24:57,  1.49s/it]
 56%|█████▋    | 13748/24420 [5:43:43<4:24:30,  1.49s/it]
 56%|█████▋    | 13749/24420 [5:43:44<4:23:50,  1.48s/it]
 56%|█████▋    | 13750/24420 [5:43:46<4:24:00,  1.48s/it]
                                                         
{'loss': 0.072, 'learning_rate': 5e-05, 'epoch': 2.82}

 56%|█████▋    | 13750/24420 [5:43:46<4:24:00,  1.48s/it]
 56%|█████▋    | 13751/24420 [5:43:47<4:24:57,  1.49s/it]
 56%|█████▋    | 13752/24420 [5:43:49<4:24:22,  1.49s/it]
 56%|█████▋    | 13753/24420 [5:43:50<4:24:47,  1.49s/it]
 56%|█████▋    | 13754/24420 [5:43:52<4:24:31,  1.49s/it]
 56%|█████▋    | 13755/24420 [5:43:53<4:24:34,  1.49s/it]
                                                         
{'loss': 0.0914, 'learning_rate': 5e-05, 'epoch': 2.82}

 56%|█████▋    | 13755/24420 [5:43:53<4:24:34,  1.49s/it]
 56%|█████▋    | 13756/24420 [5:43:55<4:24:48,  1.49s/it]
 56%|█████▋    | 13757/24420 [5:43:56<4:25:02,  1.49s/it]
 56%|█████▋    | 13758/24420 [5:43:58<4:25:24,  1.49s/it]
 56%|█████▋    | 13759/24420 [5:43:59<4:24:42,  1.49s/it]
 56%|█████▋    | 13760/24420 [5:44:01<4:25:08,  1.49s/it]
                                                         
{'loss': 0.0901, 'learning_rate': 5e-05, 'epoch': 2.82}

 56%|█████▋    | 13760/24420 [5:44:01<4:25:08,  1.49s/it]
 56%|█████▋    | 13761/24420 [5:44:02<4:25:02,  1.49s/it]
 56%|█████▋    | 13762/24420 [5:44:04<4:25:15,  1.49s/it]
 56%|█████▋    | 13763/24420 [5:44:05<4:26:00,  1.50s/it]
 56%|█████▋    | 13764/24420 [5:44:07<4:25:24,  1.49s/it]
 56%|█████▋    | 13765/24420 [5:44:08<4:25:15,  1.49s/it]
                                                         
{'loss': 0.0897, 'learning_rate': 5e-05, 'epoch': 2.82}

 56%|█████▋    | 13765/24420 [5:44:08<4:25:15,  1.49s/it]
 56%|█████▋    | 13766/24420 [5:44:10<4:26:36,  1.50s/it]
 56%|█████▋    | 13767/24420 [5:44:11<4:26:09,  1.50s/it]
 56%|█████▋    | 13768/24420 [5:44:13<4:25:08,  1.49s/it]
 56%|█████▋    | 13769/24420 [5:44:14<4:24:26,  1.49s/it]
 56%|█████▋    | 13770/24420 [5:44:16<4:25:07,  1.49s/it]
                                                         
{'loss': 0.0844, 'learning_rate': 5e-05, 'epoch': 2.82}

 56%|█████▋    | 13770/24420 [5:44:16<4:25:07,  1.49s/it]
 56%|█████▋    | 13771/24420 [5:44:17<4:24:31,  1.49s/it]
 56%|█████▋    | 13772/24420 [5:44:19<4:24:30,  1.49s/it]
 56%|█████▋    | 13773/24420 [5:44:20<4:24:23,  1.49s/it]
 56%|█████▋    | 13774/24420 [5:44:21<4:23:59,  1.49s/it]
 56%|█████▋    | 13775/24420 [5:44:23<4:23:33,  1.49s/it]
                                                         
{'loss': 0.0827, 'learning_rate': 5e-05, 'epoch': 2.82}

 56%|█████▋    | 13775/24420 [5:44:23<4:23:33,  1.49s/it]
 56%|█████▋    | 13776/24420 [5:44:24<4:23:59,  1.49s/it]
 56%|█████▋    | 13777/24420 [5:44:26<4:24:37,  1.49s/it]
 56%|█████▋    | 13778/24420 [5:44:27<4:24:08,  1.49s/it]
 56%|█████▋    | 13779/24420 [5:44:29<4:24:02,  1.49s/it]
 56%|█████▋    | 13780/24420 [5:44:30<4:23:33,  1.49s/it]
                                                         
{'loss': 0.086, 'learning_rate': 5e-05, 'epoch': 2.82}

 56%|█████▋    | 13780/24420 [5:44:30<4:23:33,  1.49s/it]
 56%|█████▋    | 13781/24420 [5:44:32<4:24:13,  1.49s/it]
 56%|█████▋    | 13782/24420 [5:44:33<4:24:17,  1.49s/it]
 56%|█████▋    | 13783/24420 [5:44:35<4:24:02,  1.49s/it]
 56%|█████▋    | 13784/24420 [5:44:36<4:23:16,  1.49s/it]
 56%|█████▋    | 13785/24420 [5:44:38<4:23:40,  1.49s/it]
                                                         
{'loss': 0.1092, 'learning_rate': 5e-05, 'epoch': 2.82}

 56%|█████▋    | 13785/24420 [5:44:38<4:23:40,  1.49s/it]
 56%|█████▋    | 13786/24420 [5:44:39<4:23:26,  1.49s/it]
 56%|█████▋    | 13787/24420 [5:44:41<4:23:14,  1.49s/it]
 56%|█████▋    | 13788/24420 [5:44:42<4:23:48,  1.49s/it]
 56%|█████▋    | 13789/24420 [5:44:44<4:23:34,  1.49s/it]
 56%|█████▋    | 13790/24420 [5:44:45<4:23:20,  1.49s/it]
                                                         
{'loss': 0.074, 'learning_rate': 5e-05, 'epoch': 2.82}

 56%|█████▋    | 13790/24420 [5:44:45<4:23:20,  1.49s/it]
 56%|█████▋    | 13791/24420 [5:44:47<4:23:18,  1.49s/it]
 56%|█████▋    | 13792/24420 [5:44:48<4:23:36,  1.49s/it]
 56%|█████▋    | 13793/24420 [5:44:50<4:23:28,  1.49s/it]
 56%|█████▋    | 13794/24420 [5:44:51<4:23:25,  1.49s/it]
 56%|█████▋    | 13795/24420 [5:44:53<4:23:13,  1.49s/it]
                                                         
{'loss': 0.0948, 'learning_rate': 5e-05, 'epoch': 2.82}

 56%|█████▋    | 13795/24420 [5:44:53<4:23:13,  1.49s/it]
 56%|█████▋    | 13796/24420 [5:44:54<4:23:49,  1.49s/it]
 56%|█████▋    | 13797/24420 [5:44:56<4:24:03,  1.49s/it]
 57%|█████▋    | 13798/24420 [5:44:57<4:23:27,  1.49s/it]
 57%|█████▋    | 13799/24420 [5:44:59<4:23:31,  1.49s/it]
 57%|█████▋    | 13800/24420 [5:45:00<4:23:20,  1.49s/it]
                                                         
{'loss': 0.0943, 'learning_rate': 5e-05, 'epoch': 2.83}

 57%|█████▋    | 13800/24420 [5:45:00<4:23:20,  1.49s/it]
 57%|█████▋    | 13801/24420 [5:45:06<8:39:39,  2.94s/it]
 57%|█████▋    | 13802/24420 [5:45:08<7:22:09,  2.50s/it]
 57%|█████▋    | 13803/24420 [5:45:09<6:28:20,  2.19s/it]
 57%|█████▋    | 13804/24420 [5:45:11<5:50:42,  1.98s/it]
 57%|█████▋    | 13805/24420 [5:45:12<5:24:23,  1.83s/it]
                                                         
{'loss': 0.0783, 'learning_rate': 5e-05, 'epoch': 2.83}

 57%|█████▋    | 13805/24420 [5:45:12<5:24:23,  1.83s/it]
 57%|█████▋    | 13806/24420 [5:45:14<5:05:40,  1.73s/it]
 57%|█████▋    | 13807/24420 [5:45:15<4:52:40,  1.65s/it]
 57%|█████▋    | 13808/24420 [5:45:17<4:43:27,  1.60s/it]
 57%|█████▋    | 13809/24420 [5:45:18<4:37:11,  1.57s/it]
 57%|█████▋    | 13810/24420 [5:45:20<4:32:49,  1.54s/it]
                                                         
{'loss': 0.0859, 'learning_rate': 5e-05, 'epoch': 2.83}

 57%|█████▋    | 13810/24420 [5:45:20<4:32:49,  1.54s/it]
 57%|█████▋    | 13811/24420 [5:45:21<4:30:03,  1.53s/it]
 57%|█████▋    | 13812/24420 [5:45:23<4:27:39,  1.51s/it]
 57%|█████▋    | 13813/24420 [5:45:24<4:26:47,  1.51s/it]
 57%|█████▋    | 13814/24420 [5:45:26<4:25:37,  1.50s/it]
 57%|█████▋    | 13815/24420 [5:45:27<4:24:43,  1.50s/it]
                                                         
{'loss': 0.0873, 'learning_rate': 5e-05, 'epoch': 2.83}

 57%|█████▋    | 13815/24420 [5:45:27<4:24:43,  1.50s/it]
 57%|█████▋    | 13816/24420 [5:45:29<4:24:28,  1.50s/it]
 57%|█████▋    | 13817/24420 [5:45:30<4:23:21,  1.49s/it]
 57%|█████▋    | 13818/24420 [5:45:32<4:23:05,  1.49s/it]
 57%|█████▋    | 13819/24420 [5:45:33<4:22:54,  1.49s/it]
 57%|█████▋    | 13820/24420 [5:45:35<4:23:00,  1.49s/it]
                                                         
{'loss': 0.086, 'learning_rate': 5e-05, 'epoch': 2.83}

 57%|█████▋    | 13820/24420 [5:45:35<4:23:00,  1.49s/it]
 57%|█████▋    | 13821/24420 [5:45:36<4:23:34,  1.49s/it]
 57%|█████▋    | 13822/24420 [5:45:38<4:23:03,  1.49s/it]
 57%|█████▋    | 13823/24420 [5:45:39<4:23:25,  1.49s/it]
 57%|█████▋    | 13824/24420 [5:45:41<4:23:05,  1.49s/it]
 57%|█████▋    | 13825/24420 [5:45:42<4:22:53,  1.49s/it]
                                                         
{'loss': 0.117, 'learning_rate': 5e-05, 'epoch': 2.83}

 57%|█████▋    | 13825/24420 [5:45:42<4:22:53,  1.49s/it]
 57%|█████▋    | 13826/24420 [5:45:44<4:22:55,  1.49s/it]
 57%|█████▋    | 13827/24420 [5:45:45<4:22:31,  1.49s/it]
 57%|█████▋    | 13828/24420 [5:45:47<4:22:14,  1.49s/it]
 57%|█████▋    | 13829/24420 [5:45:48<4:22:23,  1.49s/it]
 57%|█████▋    | 13830/24420 [5:45:50<4:21:56,  1.48s/it]
                                                         
{'loss': 0.083, 'learning_rate': 5e-05, 'epoch': 2.83}

 57%|█████▋    | 13830/24420 [5:45:50<4:21:56,  1.48s/it]
 57%|█████▋    | 13831/24420 [5:45:51<4:23:03,  1.49s/it]
 57%|█████▋    | 13832/24420 [5:45:53<4:23:02,  1.49s/it]
 57%|█████▋    | 13833/24420 [5:45:54<4:22:31,  1.49s/it]
 57%|█████▋    | 13834/24420 [5:45:56<4:22:09,  1.49s/it]
 57%|█████▋    | 13835/24420 [5:45:57<4:22:05,  1.49s/it]
                                                         
{'loss': 0.0819, 'learning_rate': 5e-05, 'epoch': 2.83}

 57%|█████▋    | 13835/24420 [5:45:57<4:22:05,  1.49s/it]
 57%|█████▋    | 13836/24420 [5:45:59<4:22:22,  1.49s/it]
 57%|█████▋    | 13837/24420 [5:46:00<4:22:29,  1.49s/it]
 57%|█████▋    | 13838/24420 [5:46:02<4:22:20,  1.49s/it]
 57%|█████▋    | 13839/24420 [5:46:03<4:22:02,  1.49s/it]
 57%|█████▋    | 13840/24420 [5:46:04<4:22:18,  1.49s/it]
                                                         
{'loss': 0.1039, 'learning_rate': 5e-05, 'epoch': 2.83}

 57%|█████▋    | 13840/24420 [5:46:04<4:22:18,  1.49s/it]
 57%|█████▋    | 13841/24420 [5:46:06<4:22:25,  1.49s/it]
 57%|█████▋    | 13842/24420 [5:46:07<4:21:50,  1.49s/it]
 57%|█████▋    | 13843/24420 [5:46:09<4:21:34,  1.48s/it]
 57%|█████▋    | 13844/24420 [5:46:10<4:21:34,  1.48s/it]
 57%|█████▋    | 13845/24420 [5:46:12<4:21:36,  1.48s/it]
                                                         
{'loss': 0.0805, 'learning_rate': 5e-05, 'epoch': 2.83}

 57%|█████▋    | 13845/24420 [5:46:12<4:21:36,  1.48s/it]
 57%|█████▋    | 13846/24420 [5:46:13<4:21:38,  1.48s/it]
 57%|█████▋    | 13847/24420 [5:46:15<4:21:59,  1.49s/it]
 57%|█████▋    | 13848/24420 [5:46:16<4:21:45,  1.49s/it]
 57%|█████▋    | 13849/24420 [5:46:18<4:21:41,  1.49s/it]
 57%|█████▋    | 13850/24420 [5:46:19<4:22:00,  1.49s/it]
                                                         
{'loss': 0.0849, 'learning_rate': 5e-05, 'epoch': 2.84}

 57%|█████▋    | 13850/24420 [5:46:19<4:22:00,  1.49s/it]
 57%|█████▋    | 13851/24420 [5:46:21<4:23:13,  1.49s/it]
 57%|█████▋    | 13852/24420 [5:46:22<4:22:46,  1.49s/it]
 57%|█████▋    | 13853/24420 [5:46:24<4:22:20,  1.49s/it]
 57%|█████▋    | 13854/24420 [5:46:25<4:22:00,  1.49s/it]
 57%|█████▋    | 13855/24420 [5:46:27<4:22:08,  1.49s/it]
                                                         
{'loss': 0.082, 'learning_rate': 5e-05, 'epoch': 2.84}

 57%|█████▋    | 13855/24420 [5:46:27<4:22:08,  1.49s/it]
 57%|█████▋    | 13856/24420 [5:46:28<4:21:53,  1.49s/it]
 57%|█████▋    | 13857/24420 [5:46:30<4:21:31,  1.49s/it]
 57%|█████▋    | 13858/24420 [5:46:31<4:21:20,  1.48s/it]
 57%|█████▋    | 13859/24420 [5:46:33<4:21:29,  1.49s/it]
 57%|█████▋    | 13860/24420 [5:46:34<4:21:40,  1.49s/it]
                                                         
{'loss': 0.0809, 'learning_rate': 5e-05, 'epoch': 2.84}

 57%|█████▋    | 13860/24420 [5:46:34<4:21:40,  1.49s/it]
 57%|█████▋    | 13861/24420 [5:46:36<4:21:22,  1.49s/it]
 57%|█████▋    | 13862/24420 [5:46:37<4:21:25,  1.49s/it]
 57%|█████▋    | 13863/24420 [5:46:39<4:21:39,  1.49s/it]
 57%|█████▋    | 13864/24420 [5:46:40<4:21:37,  1.49s/it]
 57%|█████▋    | 13865/24420 [5:46:42<4:21:13,  1.48s/it]
                                                         
{'loss': 0.089, 'learning_rate': 5e-05, 'epoch': 2.84}

 57%|█████▋    | 13865/24420 [5:46:42<4:21:13,  1.48s/it]
 57%|█████▋    | 13866/24420 [5:46:43<4:21:02,  1.48s/it]
 57%|█████▋    | 13867/24420 [5:46:45<4:20:41,  1.48s/it]
 57%|█████▋    | 13868/24420 [5:46:46<4:21:05,  1.48s/it]
 57%|█████▋    | 13869/24420 [5:46:48<4:21:43,  1.49s/it]
 57%|█████▋    | 13870/24420 [5:46:49<4:21:33,  1.49s/it]
                                                         
{'loss': 0.1058, 'learning_rate': 5e-05, 'epoch': 2.84}

 57%|█████▋    | 13870/24420 [5:46:49<4:21:33,  1.49s/it]
 57%|█████▋    | 13871/24420 [5:46:51<4:22:31,  1.49s/it]
 57%|█████▋    | 13872/24420 [5:46:52<4:22:10,  1.49s/it]
 57%|█████▋    | 13873/24420 [5:46:54<4:21:58,  1.49s/it]
 57%|█████▋    | 13874/24420 [5:46:55<4:21:55,  1.49s/it]
 57%|█████▋    | 13875/24420 [5:46:57<4:21:33,  1.49s/it]
                                                         
{'loss': 0.0796, 'learning_rate': 5e-05, 'epoch': 2.84}

 57%|█████▋    | 13875/24420 [5:46:57<4:21:33,  1.49s/it]
 57%|█████▋    | 13876/24420 [5:46:58<4:21:21,  1.49s/it]
 57%|█████▋    | 13877/24420 [5:46:59<4:21:07,  1.49s/it]
 57%|█████▋    | 13878/24420 [5:47:01<4:21:09,  1.49s/it]
 57%|█████▋    | 13879/24420 [5:47:02<4:21:33,  1.49s/it]
 57%|█████▋    | 13880/24420 [5:47:04<4:20:56,  1.49s/it]
                                                         
{'loss': 0.0835, 'learning_rate': 5e-05, 'epoch': 2.84}

 57%|█████▋    | 13880/24420 [5:47:04<4:20:56,  1.49s/it]
 57%|█████▋    | 13881/24420 [5:47:05<4:20:41,  1.48s/it]
 57%|█████▋    | 13882/24420 [5:47:07<4:20:46,  1.48s/it]
 57%|█████▋    | 13883/24420 [5:47:08<4:21:15,  1.49s/it]
 57%|█████▋    | 13884/24420 [5:47:10<4:21:52,  1.49s/it]
 57%|█████▋    | 13885/24420 [5:47:11<4:21:42,  1.49s/it]
                                                         
{'loss': 0.0868, 'learning_rate': 5e-05, 'epoch': 2.84}

 57%|█████▋    | 13885/24420 [5:47:11<4:21:42,  1.49s/it]
 57%|█████▋    | 13886/24420 [5:47:13<4:22:22,  1.49s/it]
 57%|█████▋    | 13887/24420 [5:47:14<4:22:11,  1.49s/it]
 57%|█████▋    | 13888/24420 [5:47:16<4:21:28,  1.49s/it]
 57%|█████▋    | 13889/24420 [5:47:17<4:21:03,  1.49s/it]
 57%|█████▋    | 13890/24420 [5:47:19<4:20:59,  1.49s/it]
                                                         
{'loss': 0.0914, 'learning_rate': 5e-05, 'epoch': 2.84}

 57%|█████▋    | 13890/24420 [5:47:19<4:20:59,  1.49s/it]
 57%|█████▋    | 13891/24420 [5:47:20<4:20:54,  1.49s/it]
 57%|█████▋    | 13892/24420 [5:47:22<4:20:56,  1.49s/it]
 57%|█████▋    | 13893/24420 [5:47:23<4:20:54,  1.49s/it]
 57%|█████▋    | 13894/24420 [5:47:25<4:20:25,  1.48s/it]
 57%|█████▋    | 13895/24420 [5:47:26<4:20:58,  1.49s/it]
                                                         
{'loss': 0.0888, 'learning_rate': 5e-05, 'epoch': 2.85}

 57%|█████▋    | 13895/24420 [5:47:26<4:20:58,  1.49s/it]
 57%|█████▋    | 13896/24420 [5:47:28<4:21:00,  1.49s/it]
 57%|█████▋    | 13897/24420 [5:47:29<4:21:14,  1.49s/it]
 57%|█████▋    | 13898/24420 [5:47:31<4:21:18,  1.49s/it]
 57%|█████▋    | 13899/24420 [5:47:32<4:21:22,  1.49s/it]
 57%|█████▋    | 13900/24420 [5:47:34<4:20:53,  1.49s/it]
                                                         
{'loss': 0.0856, 'learning_rate': 5e-05, 'epoch': 2.85}

 57%|█████▋    | 13900/24420 [5:47:34<4:20:53,  1.49s/it]
 57%|█████▋    | 13901/24420 [5:47:35<4:20:43,  1.49s/it]
 57%|█████▋    | 13902/24420 [5:47:37<4:20:45,  1.49s/it]
 57%|█████▋    | 13903/24420 [5:47:38<4:20:21,  1.49s/it]
 57%|█████▋    | 13904/24420 [5:47:40<4:20:17,  1.49s/it]
 57%|█████▋    | 13905/24420 [5:47:41<4:20:12,  1.48s/it]
                                                         
{'loss': 0.0857, 'learning_rate': 5e-05, 'epoch': 2.85}

 57%|█████▋    | 13905/24420 [5:47:41<4:20:12,  1.48s/it]
 57%|█████▋    | 13906/24420 [5:47:43<4:29:52,  1.54s/it]
 57%|█████▋    | 13907/24420 [5:47:44<4:26:44,  1.52s/it]
 57%|█████▋    | 13908/24420 [5:47:46<4:24:18,  1.51s/it]
 57%|█████▋    | 13909/24420 [5:47:47<4:22:50,  1.50s/it]
 57%|█████▋    | 13910/24420 [5:47:49<4:21:59,  1.50s/it]
                                                         
{'loss': 0.0812, 'learning_rate': 5e-05, 'epoch': 2.85}

 57%|█████▋    | 13910/24420 [5:47:49<4:21:59,  1.50s/it]
 57%|█████▋    | 13911/24420 [5:47:50<4:22:16,  1.50s/it]
 57%|█████▋    | 13912/24420 [5:47:52<4:21:48,  1.49s/it]
 57%|█████▋    | 13913/24420 [5:47:53<4:21:29,  1.49s/it]
 57%|█████▋    | 13914/24420 [5:47:55<4:21:29,  1.49s/it]
 57%|█████▋    | 13915/24420 [5:47:56<4:21:02,  1.49s/it]
                                                         
{'loss': 0.1039, 'learning_rate': 5e-05, 'epoch': 2.85}

 57%|█████▋    | 13915/24420 [5:47:56<4:21:02,  1.49s/it]
 57%|█████▋    | 13916/24420 [5:47:58<4:21:08,  1.49s/it]
 57%|█████▋    | 13917/24420 [5:47:59<4:21:21,  1.49s/it]
 57%|█████▋    | 13918/24420 [5:48:01<4:20:49,  1.49s/it]
 57%|█████▋    | 13919/24420 [5:48:02<4:21:07,  1.49s/it]
 57%|█████▋    | 13920/24420 [5:48:04<4:20:40,  1.49s/it]
                                                         
{'loss': 0.1053, 'learning_rate': 5e-05, 'epoch': 2.85}

 57%|█████▋    | 13920/24420 [5:48:04<4:20:40,  1.49s/it]
 57%|█████▋    | 13921/24420 [5:48:05<4:21:29,  1.49s/it]
 57%|█████▋    | 13922/24420 [5:48:07<4:21:23,  1.49s/it]
 57%|█████▋    | 13923/24420 [5:48:08<4:21:18,  1.49s/it]
 57%|█████▋    | 13924/24420 [5:48:10<4:20:50,  1.49s/it]
 57%|█████▋    | 13925/24420 [5:48:11<4:20:56,  1.49s/it]
                                                         
{'loss': 0.1008, 'learning_rate': 5e-05, 'epoch': 2.85}

 57%|█████▋    | 13925/24420 [5:48:11<4:20:56,  1.49s/it]
 57%|█████▋    | 13926/24420 [5:48:13<4:21:11,  1.49s/it]
 57%|█████▋    | 13927/24420 [5:48:14<4:20:41,  1.49s/it]
 57%|█████▋    | 13928/24420 [5:48:16<4:20:32,  1.49s/it]
 57%|█████▋    | 13929/24420 [5:48:17<4:20:10,  1.49s/it]
 57%|█████▋    | 13930/24420 [5:48:19<4:19:56,  1.49s/it]
                                                         
{'loss': 0.0732, 'learning_rate': 5e-05, 'epoch': 2.85}

 57%|█████▋    | 13930/24420 [5:48:19<4:19:56,  1.49s/it]
 57%|█████▋    | 13931/24420 [5:48:20<4:19:39,  1.49s/it]
 57%|█████▋    | 13932/24420 [5:48:22<4:19:40,  1.49s/it]
 57%|█████▋    | 13933/24420 [5:48:23<4:20:25,  1.49s/it]
 57%|█████▋    | 13934/24420 [5:48:25<4:20:23,  1.49s/it]
 57%|█████▋    | 13935/24420 [5:48:26<4:20:09,  1.49s/it]
                                                         
{'loss': 0.09, 'learning_rate': 5e-05, 'epoch': 2.85}

 57%|█████▋    | 13935/24420 [5:48:26<4:20:09,  1.49s/it]
 57%|█████▋    | 13936/24420 [5:48:28<4:24:47,  1.52s/it]
 57%|█████▋    | 13937/24420 [5:48:29<4:22:52,  1.50s/it]
 57%|█████▋    | 13938/24420 [5:48:31<4:21:58,  1.50s/it]
 57%|█████▋    | 13939/24420 [5:48:32<4:21:25,  1.50s/it]
 57%|█████▋    | 13940/24420 [5:48:34<4:20:58,  1.49s/it]
                                                         
{'loss': 0.0837, 'learning_rate': 5e-05, 'epoch': 2.85}

 57%|█████▋    | 13940/24420 [5:48:34<4:20:58,  1.49s/it]
 57%|█████▋    | 13941/24420 [5:48:35<4:20:33,  1.49s/it]
 57%|█████▋    | 13942/24420 [5:48:37<4:20:30,  1.49s/it]
 57%|█████▋    | 13943/24420 [5:48:38<4:20:08,  1.49s/it]
 57%|█████▋    | 13944/24420 [5:48:39<4:20:15,  1.49s/it]
 57%|█████▋    | 13945/24420 [5:48:41<4:20:27,  1.49s/it]
                                                         
{'loss': 0.0898, 'learning_rate': 5e-05, 'epoch': 2.86}

 57%|█████▋    | 13945/24420 [5:48:41<4:20:27,  1.49s/it]
 57%|█████▋    | 13946/24420 [5:48:42<4:20:41,  1.49s/it]
 57%|█████▋    | 13947/24420 [5:48:44<4:20:10,  1.49s/it]
 57%|█████▋    | 13948/24420 [5:48:45<4:19:38,  1.49s/it]
 57%|█████▋    | 13949/24420 [5:48:47<4:19:46,  1.49s/it]
 57%|█████▋    | 13950/24420 [5:48:48<4:19:42,  1.49s/it]
                                                         
{'loss': 0.0881, 'learning_rate': 5e-05, 'epoch': 2.86}

 57%|█████▋    | 13950/24420 [5:48:48<4:19:42,  1.49s/it]
 57%|█████▋    | 13951/24420 [5:48:50<4:20:26,  1.49s/it]
 57%|█████▋    | 13952/24420 [5:48:51<4:20:02,  1.49s/it]
 57%|█████▋    | 13953/24420 [5:48:53<4:19:30,  1.49s/it]
 57%|█████▋    | 13954/24420 [5:48:54<4:19:28,  1.49s/it]
 57%|█████▋    | 13955/24420 [5:48:56<4:19:35,  1.49s/it]
                                                         
{'loss': 0.0759, 'learning_rate': 5e-05, 'epoch': 2.86}

 57%|█████▋    | 13955/24420 [5:48:56<4:19:35,  1.49s/it]
 57%|█████▋    | 13956/24420 [5:48:57<4:19:55,  1.49s/it]
 57%|█████▋    | 13957/24420 [5:48:59<4:19:33,  1.49s/it]
 57%|█████▋    | 13958/24420 [5:49:00<4:19:51,  1.49s/it]
 57%|█████▋    | 13959/24420 [5:49:02<4:20:02,  1.49s/it]
 57%|█████▋    | 13960/24420 [5:49:03<4:20:00,  1.49s/it]
                                                         
{'loss': 0.0774, 'learning_rate': 5e-05, 'epoch': 2.86}

 57%|█████▋    | 13960/24420 [5:49:03<4:20:00,  1.49s/it]
 57%|█████▋    | 13961/24420 [5:49:05<4:19:25,  1.49s/it]
 57%|█████▋    | 13962/24420 [5:49:06<4:19:46,  1.49s/it]
 57%|█████▋    | 13963/24420 [5:49:08<4:19:02,  1.49s/it]
 57%|█████▋    | 13964/24420 [5:49:09<4:18:25,  1.48s/it]
 57%|█████▋    | 13965/24420 [5:49:11<4:18:33,  1.48s/it]
                                                         
{'loss': 0.0782, 'learning_rate': 5e-05, 'epoch': 2.86}

 57%|█████▋    | 13965/24420 [5:49:11<4:18:33,  1.48s/it]
 57%|█████▋    | 13966/24420 [5:49:12<4:18:38,  1.48s/it]
 57%|█████▋    | 13967/24420 [5:49:14<4:18:58,  1.49s/it]
 57%|█████▋    | 13968/24420 [5:49:15<4:19:19,  1.49s/it]
 57%|█████▋    | 13969/24420 [5:49:17<4:18:55,  1.49s/it]
 57%|█████▋    | 13970/24420 [5:49:18<4:19:01,  1.49s/it]
                                                         
{'loss': 0.0906, 'learning_rate': 5e-05, 'epoch': 2.86}

 57%|█████▋    | 13970/24420 [5:49:18<4:19:01,  1.49s/it]
 57%|█████▋    | 13971/24420 [5:49:20<4:19:25,  1.49s/it]
 57%|█████▋    | 13972/24420 [5:49:21<4:19:00,  1.49s/it]
 57%|█████▋    | 13973/24420 [5:49:23<4:18:38,  1.49s/it]
 57%|█████▋    | 13974/24420 [5:49:24<4:18:01,  1.48s/it]
 57%|█████▋    | 13975/24420 [5:49:26<4:17:52,  1.48s/it]
                                                         
{'loss': 0.0815, 'learning_rate': 5e-05, 'epoch': 2.86}

 57%|█████▋    | 13975/24420 [5:49:26<4:17:52,  1.48s/it]
 57%|█████▋    | 13976/24420 [5:49:27<4:18:20,  1.48s/it]
 57%|█████▋    | 13977/24420 [5:49:29<4:18:38,  1.49s/it]
 57%|█████▋    | 13978/24420 [5:49:30<4:18:37,  1.49s/it]
 57%|█████▋    | 13979/24420 [5:49:32<4:18:20,  1.48s/it]
 57%|█████▋    | 13980/24420 [5:49:33<4:18:17,  1.48s/it]
                                                         
{'loss': 0.0793, 'learning_rate': 5e-05, 'epoch': 2.86}

 57%|█████▋    | 13980/24420 [5:49:33<4:18:17,  1.48s/it]
 57%|█████▋    | 13981/24420 [5:49:35<4:18:23,  1.49s/it]
 57%|█████▋    | 13982/24420 [5:49:36<4:18:14,  1.48s/it]
 57%|█████▋    | 13983/24420 [5:49:38<4:19:05,  1.49s/it]
 57%|█████▋    | 13984/24420 [5:49:39<4:19:02,  1.49s/it]
 57%|█████▋    | 13985/24420 [5:49:40<4:18:38,  1.49s/it]
                                                         
{'loss': 0.0788, 'learning_rate': 5e-05, 'epoch': 2.86}

 57%|█████▋    | 13985/24420 [5:49:40<4:18:38,  1.49s/it]
 57%|█████▋    | 13986/24420 [5:49:42<4:18:46,  1.49s/it]
 57%|█████▋    | 13987/24420 [5:49:43<4:18:32,  1.49s/it]
 57%|█████▋    | 13988/24420 [5:49:45<4:18:17,  1.49s/it]
 57%|█████▋    | 13989/24420 [5:49:46<4:18:14,  1.49s/it]
 57%|█████▋    | 13990/24420 [5:49:48<4:17:50,  1.48s/it]
                                                         
{'loss': 0.0898, 'learning_rate': 5e-05, 'epoch': 2.86}

 57%|█████▋    | 13990/24420 [5:49:48<4:17:50,  1.48s/it]
 57%|█████▋    | 13991/24420 [5:49:49<4:17:42,  1.48s/it]
 57%|█████▋    | 13992/24420 [5:49:51<4:18:05,  1.49s/it]
 57%|█████▋    | 13993/24420 [5:49:52<4:18:16,  1.49s/it]
 57%|█████▋    | 13994/24420 [5:49:54<4:18:11,  1.49s/it]
 57%|█████▋    | 13995/24420 [5:49:55<4:17:55,  1.48s/it]
                                                         
{'loss': 0.0792, 'learning_rate': 5e-05, 'epoch': 2.87}

 57%|█████▋    | 13995/24420 [5:49:55<4:17:55,  1.48s/it]
 57%|█████▋    | 13996/24420 [5:49:57<4:17:54,  1.48s/it]
 57%|█████▋    | 13997/24420 [5:49:58<4:18:18,  1.49s/it]
 57%|█████▋    | 13998/24420 [5:50:00<4:18:38,  1.49s/it]
 57%|█████▋    | 13999/24420 [5:50:01<4:18:21,  1.49s/it]
 57%|█████▋    | 14000/24420 [5:50:03<4:18:09,  1.49s/it]
                                                         
{'loss': 0.0921, 'learning_rate': 5e-05, 'epoch': 2.87}

 57%|█████▋    | 14000/24420 [5:50:03<4:18:09,  1.49s/it]
 57%|█████▋    | 14001/24420 [5:50:07<7:07:26,  2.46s/it]
 57%|█████▋    | 14002/24420 [5:50:09<6:16:16,  2.17s/it]
 57%|█████▋    | 14003/24420 [5:50:10<5:40:49,  1.96s/it]
 57%|█████▋    | 14004/24420 [5:50:12<5:16:41,  1.82s/it]
 57%|█████▋    | 14005/24420 [5:50:13<4:58:50,  1.72s/it]
                                                         
{'loss': 0.0796, 'learning_rate': 5e-05, 'epoch': 2.87}

 57%|█████▋    | 14005/24420 [5:50:13<4:58:50,  1.72s/it]
 57%|█████▋    | 14006/24420 [5:50:15<4:46:49,  1.65s/it]
 57%|█████▋    | 14007/24420 [5:50:16<4:38:10,  1.60s/it]
 57%|█████▋    | 14008/24420 [5:50:18<4:31:50,  1.57s/it]
 57%|█████▋    | 14009/24420 [5:50:19<4:27:52,  1.54s/it]
 57%|█████▋    | 14010/24420 [5:50:21<4:25:26,  1.53s/it]
                                                         
{'loss': 0.0853, 'learning_rate': 5e-05, 'epoch': 2.87}

 57%|█████▋    | 14010/24420 [5:50:21<4:25:26,  1.53s/it]
 57%|█████▋    | 14011/24420 [5:50:22<4:23:14,  1.52s/it]
 57%|█████▋    | 14012/24420 [5:50:24<4:21:32,  1.51s/it]
 57%|█████▋    | 14013/24420 [5:50:25<4:20:27,  1.50s/it]
 57%|█████▋    | 14014/24420 [5:50:27<4:19:59,  1.50s/it]
 57%|█████▋    | 14015/24420 [5:50:28<4:19:26,  1.50s/it]
                                                         
{'loss': 0.0972, 'learning_rate': 5e-05, 'epoch': 2.87}

 57%|█████▋    | 14015/24420 [5:50:28<4:19:26,  1.50s/it]
 57%|█████▋    | 14016/24420 [5:50:30<4:18:57,  1.49s/it]
 57%|█████▋    | 14017/24420 [5:50:31<4:17:56,  1.49s/it]
 57%|█████▋    | 14018/24420 [5:50:33<4:17:35,  1.49s/it]
 57%|█████▋    | 14019/24420 [5:50:34<4:17:10,  1.48s/it]
 57%|█████▋    | 14020/24420 [5:50:36<4:17:23,  1.48s/it]
                                                         
{'loss': 0.0724, 'learning_rate': 5e-05, 'epoch': 2.87}

 57%|█████▋    | 14020/24420 [5:50:36<4:17:23,  1.48s/it]
 57%|█████▋    | 14021/24420 [5:50:37<4:17:33,  1.49s/it]
 57%|█████▋    | 14022/24420 [5:50:39<4:17:46,  1.49s/it]
 57%|█████▋    | 14023/24420 [5:50:40<4:17:55,  1.49s/it]
 57%|█████▋    | 14024/24420 [5:50:42<4:18:00,  1.49s/it]
 57%|█████▋    | 14025/24420 [5:50:43<4:17:44,  1.49s/it]
                                                         
{'loss': 0.0764, 'learning_rate': 5e-05, 'epoch': 2.87}

 57%|█████▋    | 14025/24420 [5:50:43<4:17:44,  1.49s/it]
 57%|█████▋    | 14026/24420 [5:50:45<4:17:33,  1.49s/it]
 57%|█████▋    | 14027/24420 [5:50:46<4:18:34,  1.49s/it]
 57%|█████▋    | 14028/24420 [5:50:48<4:18:23,  1.49s/it]
 57%|█████▋    | 14029/24420 [5:50:49<4:18:23,  1.49s/it]
 57%|█████▋    | 14030/24420 [5:50:51<4:17:52,  1.49s/it]
                                                         
{'loss': 0.084, 'learning_rate': 5e-05, 'epoch': 2.87}

 57%|█████▋    | 14030/24420 [5:50:51<4:17:52,  1.49s/it]
 57%|█████▋    | 14031/24420 [5:50:52<4:18:12,  1.49s/it]
 57%|█████▋    | 14032/24420 [5:50:54<4:18:29,  1.49s/it]
 57%|█████▋    | 14033/24420 [5:50:55<4:18:09,  1.49s/it]
 57%|█████▋    | 14034/24420 [5:50:57<4:18:18,  1.49s/it]
 57%|█████▋    | 14035/24420 [5:50:58<4:17:46,  1.49s/it]
                                                         
{'loss': 0.0985, 'learning_rate': 5e-05, 'epoch': 2.87}

 57%|█████▋    | 14035/24420 [5:50:58<4:17:46,  1.49s/it]
 57%|█████▋    | 14036/24420 [5:51:00<4:18:20,  1.49s/it]
 57%|█████▋    | 14037/24420 [5:51:01<4:18:15,  1.49s/it]
 57%|█████▋    | 14038/24420 [5:51:03<4:18:45,  1.50s/it]
 57%|█████▋    | 14039/24420 [5:51:04<4:18:36,  1.49s/it]
 57%|█████▋    | 14040/24420 [5:51:06<4:17:57,  1.49s/it]
                                                         
{'loss': 0.0725, 'learning_rate': 5e-05, 'epoch': 2.87}

 57%|█████▋    | 14040/24420 [5:51:06<4:17:57,  1.49s/it]
 57%|█████▋    | 14041/24420 [5:51:07<4:17:28,  1.49s/it]
 58%|█████▊    | 14042/24420 [5:51:09<4:17:15,  1.49s/it]
 58%|█████▊    | 14043/24420 [5:51:10<4:17:13,  1.49s/it]
 58%|█████▊    | 14044/24420 [5:51:12<4:17:01,  1.49s/it]
 58%|█████▊    | 14045/24420 [5:51:13<4:16:49,  1.49s/it]
                                                         
{'loss': 0.0728, 'learning_rate': 5e-05, 'epoch': 2.88}

 58%|█████▊    | 14045/24420 [5:51:13<4:16:49,  1.49s/it]
 58%|█████▊    | 14046/24420 [5:51:14<4:16:41,  1.48s/it]
 58%|█████▊    | 14047/24420 [5:51:16<4:16:47,  1.49s/it]
 58%|█████▊    | 14048/24420 [5:51:17<4:16:47,  1.49s/it]
 58%|█████▊    | 14049/24420 [5:51:19<4:16:38,  1.48s/it]
 58%|█████▊    | 14050/24420 [5:51:20<4:16:51,  1.49s/it]
                                                         
{'loss': 0.071, 'learning_rate': 5e-05, 'epoch': 2.88}

 58%|█████▊    | 14050/24420 [5:51:20<4:16:51,  1.49s/it]
 58%|█████▊    | 14051/24420 [5:51:22<4:16:44,  1.49s/it]
 58%|█████▊    | 14052/24420 [5:51:23<4:16:27,  1.48s/it]
 58%|█████▊    | 14053/24420 [5:51:25<4:16:44,  1.49s/it]
 58%|█████▊    | 14054/24420 [5:51:26<4:17:36,  1.49s/it]
 58%|█████▊    | 14055/24420 [5:51:28<4:17:02,  1.49s/it]
                                                         
{'loss': 0.0854, 'learning_rate': 5e-05, 'epoch': 2.88}

 58%|█████▊    | 14055/24420 [5:51:28<4:17:02,  1.49s/it]
 58%|█████▊    | 14056/24420 [5:51:29<4:17:57,  1.49s/it]
 58%|█████▊    | 14057/24420 [5:51:31<4:18:04,  1.49s/it]
 58%|█████▊    | 14058/24420 [5:51:32<4:17:22,  1.49s/it]
 58%|█████▊    | 14059/24420 [5:51:34<4:16:51,  1.49s/it]
 58%|█████▊    | 14060/24420 [5:51:35<4:16:38,  1.49s/it]
                                                         
{'loss': 0.0653, 'learning_rate': 5e-05, 'epoch': 2.88}

 58%|█████▊    | 14060/24420 [5:51:35<4:16:38,  1.49s/it]
 58%|█████▊    | 14061/24420 [5:51:37<4:16:46,  1.49s/it]
 58%|█████▊    | 14062/24420 [5:51:38<4:16:42,  1.49s/it]
 58%|█████▊    | 14063/24420 [5:51:40<4:16:57,  1.49s/it]
 58%|█████▊    | 14064/24420 [5:51:41<4:16:32,  1.49s/it]
 58%|█████▊    | 14065/24420 [5:51:43<4:16:42,  1.49s/it]
                                                         
{'loss': 0.0821, 'learning_rate': 5e-05, 'epoch': 2.88}

 58%|█████▊    | 14065/24420 [5:51:43<4:16:42,  1.49s/it]
 58%|█████▊    | 14066/24420 [5:51:44<4:16:50,  1.49s/it]
 58%|█████▊    | 14067/24420 [5:51:46<4:17:05,  1.49s/it]
 58%|█████▊    | 14068/24420 [5:51:47<4:17:23,  1.49s/it]
 58%|█████▊    | 14069/24420 [5:51:49<4:16:57,  1.49s/it]
 58%|█████▊    | 14070/24420 [5:51:50<4:16:54,  1.49s/it]
                                                         
{'loss': 0.078, 'learning_rate': 5e-05, 'epoch': 2.88}

 58%|█████▊    | 14070/24420 [5:51:50<4:16:54,  1.49s/it]
 58%|█████▊    | 14071/24420 [5:51:52<4:16:48,  1.49s/it]
 58%|█████▊    | 14072/24420 [5:51:53<4:16:55,  1.49s/it]
 58%|█████▊    | 14073/24420 [5:51:55<4:17:06,  1.49s/it]
 58%|█████▊    | 14074/24420 [5:51:56<4:16:55,  1.49s/it]
 58%|█████▊    | 14075/24420 [5:51:58<4:16:44,  1.49s/it]
                                                         
{'loss': 0.0718, 'learning_rate': 5e-05, 'epoch': 2.88}

 58%|█████▊    | 14075/24420 [5:51:58<4:16:44,  1.49s/it]
 58%|█████▊    | 14076/24420 [5:51:59<4:16:34,  1.49s/it]
 58%|█████▊    | 14077/24420 [5:52:01<4:16:04,  1.49s/it]
 58%|█████▊    | 14078/24420 [5:52:02<4:15:50,  1.48s/it]
 58%|█████▊    | 14079/24420 [5:52:04<4:15:43,  1.48s/it]
 58%|█████▊    | 14080/24420 [5:52:05<4:15:55,  1.49s/it]
                                                         
{'loss': 0.0831, 'learning_rate': 5e-05, 'epoch': 2.88}

 58%|█████▊    | 14080/24420 [5:52:05<4:15:55,  1.49s/it]
 58%|█████▊    | 14081/24420 [5:52:07<4:16:06,  1.49s/it]
 58%|█████▊    | 14082/24420 [5:52:08<4:16:27,  1.49s/it]
 58%|█████▊    | 14083/24420 [5:52:10<4:16:18,  1.49s/it]
 58%|█████▊    | 14084/24420 [5:52:11<4:15:39,  1.48s/it]
 58%|█████▊    | 14085/24420 [5:52:13<4:15:52,  1.49s/it]
                                                         
{'loss': 0.0698, 'learning_rate': 5e-05, 'epoch': 2.88}

 58%|█████▊    | 14085/24420 [5:52:13<4:15:52,  1.49s/it]
 58%|█████▊    | 14086/24420 [5:52:14<4:15:59,  1.49s/it]
 58%|█████▊    | 14087/24420 [5:52:15<4:16:28,  1.49s/it]
 58%|█████▊    | 14088/24420 [5:52:17<4:17:11,  1.49s/it]
 58%|█████▊    | 14089/24420 [5:52:18<4:16:44,  1.49s/it]
 58%|█████▊    | 14090/24420 [5:52:20<4:17:02,  1.49s/it]
                                                         
{'loss': 0.0841, 'learning_rate': 5e-05, 'epoch': 2.88}

 58%|█████▊    | 14090/24420 [5:52:20<4:17:02,  1.49s/it]
 58%|█████▊    | 14091/24420 [5:52:21<4:16:49,  1.49s/it]
 58%|█████▊    | 14092/24420 [5:52:23<4:16:31,  1.49s/it]
 58%|█████▊    | 14093/24420 [5:52:24<4:16:47,  1.49s/it]
 58%|█████▊    | 14094/24420 [5:52:26<4:16:29,  1.49s/it]
 58%|█████▊    | 14095/24420 [5:52:27<4:16:06,  1.49s/it]
                                                         
{'loss': 0.0822, 'learning_rate': 5e-05, 'epoch': 2.89}

 58%|█████▊    | 14095/24420 [5:52:27<4:16:06,  1.49s/it]
 58%|█████▊    | 14096/24420 [5:52:29<4:16:09,  1.49s/it]
 58%|█████▊    | 14097/24420 [5:52:30<4:16:25,  1.49s/it]
 58%|█████▊    | 14098/24420 [5:52:32<4:16:41,  1.49s/it]
 58%|█████▊    | 14099/24420 [5:52:33<4:17:00,  1.49s/it]
 58%|█████▊    | 14100/24420 [5:52:35<4:16:55,  1.49s/it]
                                                         
{'loss': 0.0914, 'learning_rate': 5e-05, 'epoch': 2.89}

 58%|█████▊    | 14100/24420 [5:52:35<4:16:55,  1.49s/it]
 58%|█████▊    | 14101/24420 [5:52:36<4:16:32,  1.49s/it]
 58%|█████▊    | 14102/24420 [5:52:38<4:16:27,  1.49s/it]
 58%|█████▊    | 14103/24420 [5:52:39<4:16:39,  1.49s/it]
 58%|█████▊    | 14104/24420 [5:52:41<4:16:49,  1.49s/it]
 58%|█████▊    | 14105/24420 [5:52:42<4:17:14,  1.50s/it]
                                                         
{'loss': 0.0951, 'learning_rate': 5e-05, 'epoch': 2.89}

 58%|█████▊    | 14105/24420 [5:52:42<4:17:14,  1.50s/it]
 58%|█████▊    | 14106/24420 [5:52:44<4:17:33,  1.50s/it]
 58%|█████▊    | 14107/24420 [5:52:45<4:17:03,  1.50s/it]
 58%|█████▊    | 14108/24420 [5:52:47<4:17:08,  1.50s/it]
 58%|█████▊    | 14109/24420 [5:52:48<4:16:39,  1.49s/it]
 58%|█████▊    | 14110/24420 [5:52:50<4:16:27,  1.49s/it]
                                                         
{'loss': 0.1023, 'learning_rate': 5e-05, 'epoch': 2.89}

 58%|█████▊    | 14110/24420 [5:52:50<4:16:27,  1.49s/it]
 58%|█████▊    | 14111/24420 [5:52:51<4:16:00,  1.49s/it]
 58%|█████▊    | 14112/24420 [5:52:53<4:16:00,  1.49s/it]
 58%|█████▊    | 14113/24420 [5:52:54<4:15:23,  1.49s/it]
 58%|█████▊    | 14114/24420 [5:52:56<4:15:28,  1.49s/it]
 58%|█████▊    | 14115/24420 [5:52:57<4:15:35,  1.49s/it]
                                                         
{'loss': 0.0818, 'learning_rate': 5e-05, 'epoch': 2.89}

 58%|█████▊    | 14115/24420 [5:52:57<4:15:35,  1.49s/it]
 58%|█████▊    | 14116/24420 [5:52:59<4:15:51,  1.49s/it]
 58%|█████▊    | 14117/24420 [5:53:00<4:16:04,  1.49s/it]
 58%|█████▊    | 14118/24420 [5:53:02<4:15:59,  1.49s/it]
 58%|█████▊    | 14119/24420 [5:53:03<4:15:47,  1.49s/it]
 58%|█████▊    | 14120/24420 [5:53:05<4:15:37,  1.49s/it]
                                                         
{'loss': 0.0946, 'learning_rate': 5e-05, 'epoch': 2.89}

 58%|█████▊    | 14120/24420 [5:53:05<4:15:37,  1.49s/it]
 58%|█████▊    | 14121/24420 [5:53:06<4:16:10,  1.49s/it]
 58%|█████▊    | 14122/24420 [5:53:08<4:16:07,  1.49s/it]
 58%|█████▊    | 14123/24420 [5:53:09<4:16:27,  1.49s/it]
 58%|█████▊    | 14124/24420 [5:53:11<4:16:32,  1.50s/it]
 58%|█████▊    | 14125/24420 [5:53:12<4:16:13,  1.49s/it]
                                                         
{'loss': 0.0903, 'learning_rate': 5e-05, 'epoch': 2.89}

 58%|█████▊    | 14125/24420 [5:53:12<4:16:13,  1.49s/it]
 58%|█████▊    | 14126/24420 [5:53:14<4:15:53,  1.49s/it]
 58%|█████▊    | 14127/24420 [5:53:15<4:15:50,  1.49s/it]
 58%|█████▊    | 14128/24420 [5:53:17<4:15:32,  1.49s/it]
 58%|█████▊    | 14129/24420 [5:53:18<4:15:43,  1.49s/it]
 58%|█████▊    | 14130/24420 [5:53:20<4:15:50,  1.49s/it]
                                                         
{'loss': 0.0807, 'learning_rate': 5e-05, 'epoch': 2.89}

 58%|█████▊    | 14130/24420 [5:53:20<4:15:50,  1.49s/it]
 58%|█████▊    | 14131/24420 [5:53:21<4:15:47,  1.49s/it]
 58%|█████▊    | 14132/24420 [5:53:23<4:15:21,  1.49s/it]
 58%|█████▊    | 14133/24420 [5:53:24<4:16:26,  1.50s/it]
 58%|█████▊    | 14134/24420 [5:53:26<4:16:15,  1.49s/it]
 58%|█████▊    | 14135/24420 [5:53:27<4:16:10,  1.49s/it]
                                                         
{'loss': 0.0854, 'learning_rate': 5e-05, 'epoch': 2.89}

 58%|█████▊    | 14135/24420 [5:53:27<4:16:10,  1.49s/it]
 58%|█████▊    | 14136/24420 [5:53:29<4:16:15,  1.50s/it]
 58%|█████▊    | 14137/24420 [5:53:30<4:15:45,  1.49s/it]
 58%|█████▊    | 14138/24420 [5:53:32<4:16:00,  1.49s/it]
 58%|█████▊    | 14139/24420 [5:53:33<4:16:47,  1.50s/it]
 58%|█████▊    | 14140/24420 [5:53:35<4:16:39,  1.50s/it]
                                                         
{'loss': 0.0816, 'learning_rate': 5e-05, 'epoch': 2.9}

 58%|█████▊    | 14140/24420 [5:53:35<4:16:39,  1.50s/it]
 58%|█████▊    | 14141/24420 [5:53:36<4:16:35,  1.50s/it]
 58%|█████▊    | 14142/24420 [5:53:38<4:16:15,  1.50s/it]
 58%|█████▊    | 14143/24420 [5:53:39<4:15:48,  1.49s/it]
 58%|█████▊    | 14144/24420 [5:53:41<4:16:16,  1.50s/it]
 58%|█████▊    | 14145/24420 [5:53:42<4:16:07,  1.50s/it]
                                                         
{'loss': 0.0933, 'learning_rate': 5e-05, 'epoch': 2.9}

 58%|█████▊    | 14145/24420 [5:53:42<4:16:07,  1.50s/it]
 58%|█████▊    | 14146/24420 [5:53:44<4:16:02,  1.50s/it]
 58%|█████▊    | 14147/24420 [5:53:45<4:15:24,  1.49s/it]
 58%|█████▊    | 14148/24420 [5:53:47<4:16:48,  1.50s/it]
 58%|█████▊    | 14149/24420 [5:53:48<4:15:54,  1.49s/it]
 58%|█████▊    | 14150/24420 [5:53:50<4:15:30,  1.49s/it]
                                                         
{'loss': 0.0869, 'learning_rate': 5e-05, 'epoch': 2.9}

 58%|█████▊    | 14150/24420 [5:53:50<4:15:30,  1.49s/it]
 58%|█████▊    | 14151/24420 [5:53:51<4:14:48,  1.49s/it]
 58%|█████▊    | 14152/24420 [5:53:53<4:14:55,  1.49s/it]
 58%|█████▊    | 14153/24420 [5:53:54<4:15:08,  1.49s/it]
 58%|█████▊    | 14154/24420 [5:53:55<4:15:05,  1.49s/it]
 58%|█████▊    | 14155/24420 [5:53:57<4:14:58,  1.49s/it]
                                                         
{'loss': 0.0728, 'learning_rate': 5e-05, 'epoch': 2.9}

 58%|█████▊    | 14155/24420 [5:53:57<4:14:58,  1.49s/it]
 58%|█████▊    | 14156/24420 [5:53:58<4:14:46,  1.49s/it]
 58%|█████▊    | 14157/24420 [5:54:00<4:14:46,  1.49s/it]
 58%|█████▊    | 14158/24420 [5:54:01<4:15:53,  1.50s/it]
 58%|█████▊    | 14159/24420 [5:54:03<4:15:15,  1.49s/it]
 58%|█████▊    | 14160/24420 [5:54:04<4:15:03,  1.49s/it]
                                                         
{'loss': 0.0941, 'learning_rate': 5e-05, 'epoch': 2.9}

 58%|█████▊    | 14160/24420 [5:54:04<4:15:03,  1.49s/it]
 58%|█████▊    | 14161/24420 [5:54:06<4:15:41,  1.50s/it]
 58%|█████▊    | 14162/24420 [5:54:07<4:15:36,  1.50s/it]
 58%|█████▊    | 14163/24420 [5:54:09<4:20:40,  1.52s/it]
 58%|█████▊    | 14164/24420 [5:54:11<4:18:46,  1.51s/it]
 58%|█████▊    | 14165/24420 [5:54:12<4:17:17,  1.51s/it]
                                                         
{'loss': 0.1259, 'learning_rate': 5e-05, 'epoch': 2.9}

 58%|█████▊    | 14165/24420 [5:54:12<4:17:17,  1.51s/it]
 58%|█████▊    | 14166/24420 [5:54:13<4:16:15,  1.50s/it]
 58%|█████▊    | 14167/24420 [5:54:15<4:15:13,  1.49s/it]
 58%|█████▊    | 14168/24420 [5:54:16<4:15:10,  1.49s/it]
 58%|█████▊    | 14169/24420 [5:54:18<4:15:23,  1.49s/it]
 58%|█████▊    | 14170/24420 [5:54:19<4:14:56,  1.49s/it]
                                                         
{'loss': 0.0832, 'learning_rate': 5e-05, 'epoch': 2.9}

 58%|█████▊    | 14170/24420 [5:54:19<4:14:56,  1.49s/it]
 58%|█████▊    | 14171/24420 [5:54:21<4:15:08,  1.49s/it]
 58%|█████▊    | 14172/24420 [5:54:22<4:14:54,  1.49s/it]
 58%|█████▊    | 14173/24420 [5:54:24<4:15:20,  1.50s/it]
 58%|█████▊    | 14174/24420 [5:54:25<4:15:31,  1.50s/it]
 58%|█████▊    | 14175/24420 [5:54:27<4:14:52,  1.49s/it]
                                                         
{'loss': 0.0905, 'learning_rate': 5e-05, 'epoch': 2.9}

 58%|█████▊    | 14175/24420 [5:54:27<4:14:52,  1.49s/it]
 58%|█████▊    | 14176/24420 [5:54:28<4:14:29,  1.49s/it]
 58%|█████▊    | 14177/24420 [5:54:30<4:13:57,  1.49s/it]
 58%|█████▊    | 14178/24420 [5:54:31<4:13:34,  1.49s/it]
 58%|█████▊    | 14179/24420 [5:54:33<4:13:39,  1.49s/it]
 58%|█████▊    | 14180/24420 [5:54:34<4:13:28,  1.49s/it]
                                                         
{'loss': 0.0759, 'learning_rate': 5e-05, 'epoch': 2.9}

 58%|█████▊    | 14180/24420 [5:54:34<4:13:28,  1.49s/it]
 58%|█████▊    | 14181/24420 [5:54:36<4:13:55,  1.49s/it]
 58%|█████▊    | 14182/24420 [5:54:37<4:13:37,  1.49s/it]
 58%|█████▊    | 14183/24420 [5:54:39<4:13:22,  1.49s/it]
 58%|█████▊    | 14184/24420 [5:54:40<4:13:13,  1.48s/it]
 58%|█████▊    | 14185/24420 [5:54:42<4:13:20,  1.49s/it]
                                                         
{'loss': 0.0772, 'learning_rate': 5e-05, 'epoch': 2.9}

 58%|█████▊    | 14185/24420 [5:54:42<4:13:20,  1.49s/it]
 58%|█████▊    | 14186/24420 [5:54:43<4:13:13,  1.48s/it]
 58%|█████▊    | 14187/24420 [5:54:45<4:13:18,  1.49s/it]
 58%|█████▊    | 14188/24420 [5:54:46<4:13:06,  1.48s/it]
 58%|█████▊    | 14189/24420 [5:54:48<4:13:20,  1.49s/it]
 58%|█████▊    | 14190/24420 [5:54:49<4:13:25,  1.49s/it]
                                                         
{'loss': 0.0828, 'learning_rate': 5e-05, 'epoch': 2.91}

 58%|█████▊    | 14190/24420 [5:54:49<4:13:25,  1.49s/it]
 58%|█████▊    | 14191/24420 [5:54:51<4:13:38,  1.49s/it]
 58%|█████▊    | 14192/24420 [5:54:52<4:13:48,  1.49s/it]
 58%|█████▊    | 14193/24420 [5:54:54<4:13:24,  1.49s/it]
 58%|█████▊    | 14194/24420 [5:54:55<4:14:01,  1.49s/it]
 58%|█████▊    | 14195/24420 [5:54:57<4:13:26,  1.49s/it]
                                                         
{'loss': 0.1038, 'learning_rate': 5e-05, 'epoch': 2.91}

 58%|█████▊    | 14195/24420 [5:54:57<4:13:26,  1.49s/it]
 58%|█████▊    | 14196/24420 [5:54:58<4:14:12,  1.49s/it]
 58%|█████▊    | 14197/24420 [5:55:00<4:13:58,  1.49s/it]
 58%|█████▊    | 14198/24420 [5:55:01<4:13:20,  1.49s/it]
 58%|█████▊    | 14199/24420 [5:55:03<4:13:23,  1.49s/it]
 58%|█████▊    | 14200/24420 [5:55:04<4:13:15,  1.49s/it]
                                                         
{'loss': 0.0776, 'learning_rate': 5e-05, 'epoch': 2.91}

 58%|█████▊    | 14200/24420 [5:55:04<4:13:15,  1.49s/it]
 58%|█████▊    | 14201/24420 [5:55:09<7:09:54,  2.52s/it]
 58%|█████▊    | 14202/24420 [5:55:11<6:16:53,  2.21s/it]
 58%|█████▊    | 14203/24420 [5:55:12<5:39:34,  1.99s/it]
 58%|█████▊    | 14204/24420 [5:55:13<5:13:48,  1.84s/it]
 58%|█████▊    | 14205/24420 [5:55:15<4:55:25,  1.74s/it]
                                                         
{'loss': 0.1093, 'learning_rate': 5e-05, 'epoch': 2.91}

 58%|█████▊    | 14205/24420 [5:55:15<4:55:25,  1.74s/it]
 58%|█████▊    | 14206/24420 [5:55:16<4:42:16,  1.66s/it]
 58%|█████▊    | 14207/24420 [5:55:18<4:33:10,  1.60s/it]
 58%|█████▊    | 14208/24420 [5:55:19<4:26:51,  1.57s/it]
 58%|█████▊    | 14209/24420 [5:55:21<4:22:36,  1.54s/it]
 58%|█████▊    | 14210/24420 [5:55:22<4:19:56,  1.53s/it]
                                                         
{'loss': 0.0933, 'learning_rate': 5e-05, 'epoch': 2.91}

 58%|█████▊    | 14210/24420 [5:55:22<4:19:56,  1.53s/it]
 58%|█████▊    | 14211/24420 [5:55:24<4:18:26,  1.52s/it]
 58%|█████▊    | 14212/24420 [5:55:25<4:16:49,  1.51s/it]
 58%|█████▊    | 14213/24420 [5:55:27<4:16:18,  1.51s/it]
 58%|█████▊    | 14214/24420 [5:55:28<4:15:52,  1.50s/it]
 58%|█████▊    | 14215/24420 [5:55:30<4:15:06,  1.50s/it]
                                                         
{'loss': 0.0963, 'learning_rate': 5e-05, 'epoch': 2.91}

 58%|█████▊    | 14215/24420 [5:55:30<4:15:06,  1.50s/it]
 58%|█████▊    | 14216/24420 [5:55:31<4:14:27,  1.50s/it]
 58%|█████▊    | 14217/24420 [5:55:33<4:14:02,  1.49s/it]
 58%|█████▊    | 14218/24420 [5:55:34<4:14:01,  1.49s/it]
 58%|█████▊    | 14219/24420 [5:55:36<4:13:24,  1.49s/it]
 58%|█████▊    | 14220/24420 [5:55:37<4:12:50,  1.49s/it]
                                                         
{'loss': 0.0773, 'learning_rate': 5e-05, 'epoch': 2.91}

 58%|█████▊    | 14220/24420 [5:55:37<4:12:50,  1.49s/it]
 58%|█████▊    | 14221/24420 [5:55:39<4:12:55,  1.49s/it]
 58%|█████▊    | 14222/24420 [5:55:40<4:13:17,  1.49s/it]
 58%|█████▊    | 14223/24420 [5:55:42<4:13:35,  1.49s/it]
 58%|█████▊    | 14224/24420 [5:55:43<4:13:28,  1.49s/it]
 58%|█████▊    | 14225/24420 [5:55:45<4:13:08,  1.49s/it]
                                                         
{'loss': 0.0944, 'learning_rate': 5e-05, 'epoch': 2.91}

 58%|█████▊    | 14225/24420 [5:55:45<4:13:08,  1.49s/it]
 58%|█████▊    | 14226/24420 [5:55:46<4:12:42,  1.49s/it]
 58%|█████▊    | 14227/24420 [5:55:48<4:12:38,  1.49s/it]
 58%|█████▊    | 14228/24420 [5:55:49<4:12:39,  1.49s/it]
 58%|█████▊    | 14229/24420 [5:55:51<4:12:14,  1.49s/it]
 58%|█████▊    | 14230/24420 [5:55:52<4:12:10,  1.48s/it]
                                                         
{'loss': 0.0748, 'learning_rate': 5e-05, 'epoch': 2.91}

 58%|█████▊    | 14230/24420 [5:55:52<4:12:10,  1.48s/it]
 58%|█████▊    | 14231/24420 [5:55:54<4:12:21,  1.49s/it]
 58%|█████▊    | 14232/24420 [5:55:55<4:12:19,  1.49s/it]
 58%|█████▊    | 14233/24420 [5:55:57<4:13:05,  1.49s/it]
 58%|█████▊    | 14234/24420 [5:55:58<4:13:10,  1.49s/it]
 58%|█████▊    | 14235/24420 [5:56:00<4:12:55,  1.49s/it]
                                                         
{'loss': 0.0786, 'learning_rate': 5e-05, 'epoch': 2.91}

 58%|█████▊    | 14235/24420 [5:56:00<4:12:55,  1.49s/it]
 58%|█████▊    | 14236/24420 [5:56:01<4:12:51,  1.49s/it]
 58%|█████▊    | 14237/24420 [5:56:03<4:12:22,  1.49s/it]
 58%|█████▊    | 14238/24420 [5:56:04<4:12:16,  1.49s/it]
 58%|█████▊    | 14239/24420 [5:56:06<4:12:10,  1.49s/it]
 58%|█████▊    | 14240/24420 [5:56:07<4:11:53,  1.48s/it]
                                                         
{'loss': 0.0814, 'learning_rate': 5e-05, 'epoch': 2.92}

 58%|█████▊    | 14240/24420 [5:56:07<4:11:53,  1.48s/it]
 58%|█████▊    | 14241/24420 [5:56:09<4:12:09,  1.49s/it]
 58%|█████▊    | 14242/24420 [5:56:10<4:12:06,  1.49s/it]
 58%|█████▊    | 14243/24420 [5:56:12<4:12:09,  1.49s/it]
 58%|█████▊    | 14244/24420 [5:56:13<4:12:02,  1.49s/it]
 58%|█████▊    | 14245/24420 [5:56:15<4:12:58,  1.49s/it]
                                                         
{'loss': 0.0886, 'learning_rate': 5e-05, 'epoch': 2.92}

 58%|█████▊    | 14245/24420 [5:56:15<4:12:58,  1.49s/it]
 58%|█████▊    | 14246/24420 [5:56:16<4:13:22,  1.49s/it]
 58%|█████▊    | 14247/24420 [5:56:17<4:12:54,  1.49s/it]
 58%|█████▊    | 14248/24420 [5:56:19<4:12:32,  1.49s/it]
 58%|█████▊    | 14249/24420 [5:56:20<4:12:14,  1.49s/it]
 58%|█████▊    | 14250/24420 [5:56:22<4:12:00,  1.49s/it]
                                                         
{'loss': 0.071, 'learning_rate': 5e-05, 'epoch': 2.92}

 58%|█████▊    | 14250/24420 [5:56:22<4:12:00,  1.49s/it]
 58%|█████▊    | 14251/24420 [5:56:23<4:12:56,  1.49s/it]
 58%|█████▊    | 14252/24420 [5:56:25<4:12:28,  1.49s/it]
 58%|█████▊    | 14253/24420 [5:56:26<4:12:51,  1.49s/it]
 58%|█████▊    | 14254/24420 [5:56:28<4:12:31,  1.49s/it]
 58%|█████▊    | 14255/24420 [5:56:29<4:12:43,  1.49s/it]
                                                         
{'loss': 0.1005, 'learning_rate': 5e-05, 'epoch': 2.92}

 58%|█████▊    | 14255/24420 [5:56:29<4:12:43,  1.49s/it]
 58%|█████▊    | 14256/24420 [5:56:31<4:12:29,  1.49s/it]
 58%|█████▊    | 14257/24420 [5:56:32<4:13:00,  1.49s/it]
 58%|█████▊    | 14258/24420 [5:56:34<4:12:43,  1.49s/it]
 58%|█████▊    | 14259/24420 [5:56:35<4:12:48,  1.49s/it]
 58%|█████▊    | 14260/24420 [5:56:37<4:12:13,  1.49s/it]
                                                         
{'loss': 0.0725, 'learning_rate': 5e-05, 'epoch': 2.92}

 58%|█████▊    | 14260/24420 [5:56:37<4:12:13,  1.49s/it]
 58%|█████▊    | 14261/24420 [5:56:38<4:12:05,  1.49s/it]
 58%|█████▊    | 14262/24420 [5:56:40<4:11:27,  1.49s/it]
 58%|█████▊    | 14263/24420 [5:56:41<4:11:10,  1.48s/it]
 58%|█████▊    | 14264/24420 [5:56:43<4:11:38,  1.49s/it]
 58%|█████▊    | 14265/24420 [5:56:44<4:11:43,  1.49s/it]
                                                         
{'loss': 0.0803, 'learning_rate': 5e-05, 'epoch': 2.92}

 58%|█████▊    | 14265/24420 [5:56:44<4:11:43,  1.49s/it]
 58%|█████▊    | 14266/24420 [5:56:46<4:12:22,  1.49s/it]
 58%|█████▊    | 14267/24420 [5:56:47<4:12:27,  1.49s/it]
 58%|█████▊    | 14268/24420 [5:56:49<4:11:51,  1.49s/it]
 58%|█████▊    | 14269/24420 [5:56:50<4:11:18,  1.49s/it]
 58%|█████▊    | 14270/24420 [5:56:52<4:11:36,  1.49s/it]
                                                         
{'loss': 0.0785, 'learning_rate': 5e-05, 'epoch': 2.92}

 58%|█████▊    | 14270/24420 [5:56:52<4:11:36,  1.49s/it]
 58%|█████▊    | 14271/24420 [5:56:53<4:11:28,  1.49s/it]
 58%|█████▊    | 14272/24420 [5:56:55<4:11:07,  1.48s/it]
 58%|█████▊    | 14273/24420 [5:56:56<4:11:36,  1.49s/it]
 58%|█████▊    | 14274/24420 [5:56:58<4:11:29,  1.49s/it]
 58%|█████▊    | 14275/24420 [5:56:59<4:11:32,  1.49s/it]
                                                         
{'loss': 0.0744, 'learning_rate': 5e-05, 'epoch': 2.92}

 58%|█████▊    | 14275/24420 [5:56:59<4:11:32,  1.49s/it]
 58%|█████▊    | 14276/24420 [5:57:01<4:11:59,  1.49s/it]
 58%|█████▊    | 14277/24420 [5:57:02<4:11:52,  1.49s/it]
 58%|█████▊    | 14278/24420 [5:57:04<4:11:57,  1.49s/it]
 58%|█████▊    | 14279/24420 [5:57:05<4:11:54,  1.49s/it]
 58%|█████▊    | 14280/24420 [5:57:07<4:11:26,  1.49s/it]
                                                         
{'loss': 0.0828, 'learning_rate': 5e-05, 'epoch': 2.92}

 58%|█████▊    | 14280/24420 [5:57:07<4:11:26,  1.49s/it]
 58%|█████▊    | 14281/24420 [5:57:08<4:11:29,  1.49s/it]
 58%|█████▊    | 14282/24420 [5:57:10<4:11:05,  1.49s/it]
 58%|█████▊    | 14283/24420 [5:57:11<4:11:24,  1.49s/it]
 58%|█████▊    | 14284/24420 [5:57:13<4:11:24,  1.49s/it]
 58%|█████▊    | 14285/24420 [5:57:14<4:11:34,  1.49s/it]
                                                         
{'loss': 0.0827, 'learning_rate': 5e-05, 'epoch': 2.92}

 58%|█████▊    | 14285/24420 [5:57:14<4:11:34,  1.49s/it]
 59%|█████▊    | 14286/24420 [5:57:16<4:11:16,  1.49s/it]
 59%|█████▊    | 14287/24420 [5:57:17<4:11:04,  1.49s/it]
 59%|█████▊    | 14288/24420 [5:57:19<4:10:37,  1.48s/it]
 59%|█████▊    | 14289/24420 [5:57:20<4:10:00,  1.48s/it]
 59%|█████▊    | 14290/24420 [5:57:21<4:10:18,  1.48s/it]
                                                         
{'loss': 0.0895, 'learning_rate': 5e-05, 'epoch': 2.93}

 59%|█████▊    | 14290/24420 [5:57:21<4:10:18,  1.48s/it]
 59%|█████▊    | 14291/24420 [5:57:23<4:10:15,  1.48s/it]
 59%|█████▊    | 14292/24420 [5:57:24<4:10:29,  1.48s/it]
 59%|█████▊    | 14293/24420 [5:57:26<4:10:16,  1.48s/it]
 59%|█████▊    | 14294/24420 [5:57:27<4:10:32,  1.48s/it]
 59%|█████▊    | 14295/24420 [5:57:29<4:10:38,  1.49s/it]
                                                         
{'loss': 0.0919, 'learning_rate': 5e-05, 'epoch': 2.93}

 59%|█████▊    | 14295/24420 [5:57:29<4:10:38,  1.49s/it]
 59%|█████▊    | 14296/24420 [5:57:30<4:10:22,  1.48s/it]
 59%|█████▊    | 14297/24420 [5:57:32<4:10:59,  1.49s/it]
 59%|█████▊    | 14298/24420 [5:57:33<4:10:53,  1.49s/it]
 59%|█████▊    | 14299/24420 [5:57:35<4:10:28,  1.48s/it]
 59%|█████▊    | 14300/24420 [5:57:36<4:10:22,  1.48s/it]
                                                         
{'loss': 0.0817, 'learning_rate': 5e-05, 'epoch': 2.93}

 59%|█████▊    | 14300/24420 [5:57:36<4:10:22,  1.48s/it]
 59%|█████▊    | 14301/24420 [5:57:38<4:10:21,  1.48s/it]
 59%|█████▊    | 14302/24420 [5:57:39<4:10:40,  1.49s/it]
 59%|█████▊    | 14303/24420 [5:57:41<4:10:43,  1.49s/it]
 59%|█████▊    | 14304/24420 [5:57:42<4:11:12,  1.49s/it]
 59%|█████▊    | 14305/24420 [5:57:44<4:11:07,  1.49s/it]
                                                         
{'loss': 0.077, 'learning_rate': 5e-05, 'epoch': 2.93}

 59%|█████▊    | 14305/24420 [5:57:44<4:11:07,  1.49s/it]
 59%|█████▊    | 14306/24420 [5:57:45<4:11:33,  1.49s/it]
 59%|█████▊    | 14307/24420 [5:57:47<4:11:34,  1.49s/it]
 59%|█████▊    | 14308/24420 [5:57:48<4:11:20,  1.49s/it]
 59%|█████▊    | 14309/24420 [5:57:50<4:10:33,  1.49s/it]
 59%|█████▊    | 14310/24420 [5:57:51<4:10:12,  1.48s/it]
                                                         
{'loss': 0.0719, 'learning_rate': 5e-05, 'epoch': 2.93}

 59%|█████▊    | 14310/24420 [5:57:51<4:10:12,  1.48s/it]
 59%|█████▊    | 14311/24420 [5:57:53<4:10:20,  1.49s/it]
 59%|█████▊    | 14312/24420 [5:57:54<4:11:14,  1.49s/it]
 59%|█████▊    | 14313/24420 [5:57:56<4:10:45,  1.49s/it]
 59%|█████▊    | 14314/24420 [5:57:57<4:10:38,  1.49s/it]
 59%|█████▊    | 14315/24420 [5:57:59<4:10:22,  1.49s/it]
                                                         
{'loss': 0.0999, 'learning_rate': 5e-05, 'epoch': 2.93}

 59%|█████▊    | 14315/24420 [5:57:59<4:10:22,  1.49s/it]
 59%|█████▊    | 14316/24420 [5:58:00<4:10:32,  1.49s/it]
 59%|█████▊    | 14317/24420 [5:58:02<4:10:26,  1.49s/it]
 59%|█████▊    | 14318/24420 [5:58:03<4:10:00,  1.48s/it]
 59%|█████▊    | 14319/24420 [5:58:05<4:09:58,  1.48s/it]
 59%|█████▊    | 14320/24420 [5:58:06<4:10:24,  1.49s/it]
                                                         
{'loss': 0.0833, 'learning_rate': 5e-05, 'epoch': 2.93}

 59%|█████▊    | 14320/24420 [5:58:06<4:10:24,  1.49s/it]
 59%|█████▊    | 14321/24420 [5:58:08<4:10:41,  1.49s/it]
 59%|█████▊    | 14322/24420 [5:58:09<4:10:54,  1.49s/it]
 59%|█████▊    | 14323/24420 [5:58:11<4:10:35,  1.49s/it]
 59%|█████▊    | 14324/24420 [5:58:12<4:10:29,  1.49s/it]
 59%|█████▊    | 14325/24420 [5:58:14<4:10:20,  1.49s/it]
                                                         
{'loss': 0.0745, 'learning_rate': 5e-05, 'epoch': 2.93}

 59%|█████▊    | 14325/24420 [5:58:14<4:10:20,  1.49s/it]
 59%|█████▊    | 14326/24420 [5:58:15<4:10:23,  1.49s/it]
 59%|█████▊    | 14327/24420 [5:58:17<4:11:01,  1.49s/it]
 59%|█████▊    | 14328/24420 [5:58:18<4:10:56,  1.49s/it]
 59%|█████▊    | 14329/24420 [5:58:20<4:10:34,  1.49s/it]
 59%|█████▊    | 14330/24420 [5:58:21<4:10:45,  1.49s/it]
                                                         
{'loss': 0.0732, 'learning_rate': 5e-05, 'epoch': 2.93}

 59%|█████▊    | 14330/24420 [5:58:21<4:10:45,  1.49s/it]
 59%|█████▊    | 14331/24420 [5:58:22<4:10:34,  1.49s/it]
 59%|█████▊    | 14332/24420 [5:58:24<4:10:42,  1.49s/it]
 59%|█████▊    | 14333/24420 [5:58:25<4:10:11,  1.49s/it]
 59%|█████▊    | 14334/24420 [5:58:27<4:10:01,  1.49s/it]
 59%|█████▊    | 14335/24420 [5:58:28<4:09:57,  1.49s/it]
                                                         
{'loss': 0.0794, 'learning_rate': 5e-05, 'epoch': 2.94}

 59%|█████▊    | 14335/24420 [5:58:28<4:09:57,  1.49s/it]
 59%|█████▊    | 14336/24420 [5:58:30<4:10:15,  1.49s/it]
 59%|█████▊    | 14337/24420 [5:58:31<4:09:54,  1.49s/it]
 59%|█████▊    | 14338/24420 [5:58:33<4:10:03,  1.49s/it]
 59%|█████▊    | 14339/24420 [5:58:34<4:09:36,  1.49s/it]
 59%|█████▊    | 14340/24420 [5:58:36<4:09:15,  1.48s/it]
                                                         
{'loss': 0.0921, 'learning_rate': 5e-05, 'epoch': 2.94}

 59%|█████▊    | 14340/24420 [5:58:36<4:09:15,  1.48s/it]
 59%|█████▊    | 14341/24420 [5:58:37<4:09:51,  1.49s/it]
 59%|█████▊    | 14342/24420 [5:58:39<4:09:55,  1.49s/it]
 59%|█████▊    | 14343/24420 [5:58:40<4:09:37,  1.49s/it]
 59%|█████▊    | 14344/24420 [5:58:42<4:09:35,  1.49s/it]
 59%|█████▊    | 14345/24420 [5:58:43<4:09:33,  1.49s/it]
                                                         
{'loss': 0.0952, 'learning_rate': 5e-05, 'epoch': 2.94}

 59%|█████▊    | 14345/24420 [5:58:43<4:09:33,  1.49s/it]
 59%|█████▊    | 14346/24420 [5:58:45<4:09:50,  1.49s/it]
 59%|█████▉    | 14347/24420 [5:58:46<4:09:43,  1.49s/it]
 59%|█████▉    | 14348/24420 [5:58:48<4:10:17,  1.49s/it]
 59%|█████▉    | 14349/24420 [5:58:49<4:09:51,  1.49s/it]
 59%|█████▉    | 14350/24420 [5:58:51<4:09:36,  1.49s/it]
                                                         
{'loss': 0.0851, 'learning_rate': 5e-05, 'epoch': 2.94}

 59%|█████▉    | 14350/24420 [5:58:51<4:09:36,  1.49s/it]
 59%|█████▉    | 14351/24420 [5:58:52<4:09:27,  1.49s/it]
 59%|█████▉    | 14352/24420 [5:58:54<4:09:10,  1.48s/it]
 59%|█████▉    | 14353/24420 [5:58:55<4:08:59,  1.48s/it]
 59%|█████▉    | 14354/24420 [5:58:57<4:09:24,  1.49s/it]
 59%|█████▉    | 14355/24420 [5:58:58<4:09:17,  1.49s/it]
                                                         
{'loss': 0.0857, 'learning_rate': 5e-05, 'epoch': 2.94}

 59%|█████▉    | 14355/24420 [5:58:58<4:09:17,  1.49s/it]
 59%|█████▉    | 14356/24420 [5:59:00<4:09:12,  1.49s/it]
 59%|█████▉    | 14357/24420 [5:59:01<4:09:34,  1.49s/it]
 59%|█████▉    | 14358/24420 [5:59:03<4:09:55,  1.49s/it]
 59%|█████▉    | 14359/24420 [5:59:04<4:09:35,  1.49s/it]
 59%|█████▉    | 14360/24420 [5:59:06<4:09:49,  1.49s/it]
                                                         
{'loss': 0.0765, 'learning_rate': 5e-05, 'epoch': 2.94}

 59%|█████▉    | 14360/24420 [5:59:06<4:09:49,  1.49s/it]
 59%|█████▉    | 14361/24420 [5:59:07<4:09:10,  1.49s/it]
 59%|█████▉    | 14362/24420 [5:59:09<4:08:49,  1.48s/it]
 59%|█████▉    | 14363/24420 [5:59:10<4:09:08,  1.49s/it]
 59%|█████▉    | 14364/24420 [5:59:12<4:09:15,  1.49s/it]
 59%|█████▉    | 14365/24420 [5:59:13<4:09:08,  1.49s/it]
                                                         
{'loss': 0.0816, 'learning_rate': 5e-05, 'epoch': 2.94}

 59%|█████▉    | 14365/24420 [5:59:13<4:09:08,  1.49s/it]
 59%|█████▉    | 14366/24420 [5:59:15<4:09:36,  1.49s/it]
 59%|█████▉    | 14367/24420 [5:59:16<4:09:34,  1.49s/it]
 59%|█████▉    | 14368/24420 [5:59:18<4:09:05,  1.49s/it]
 59%|█████▉    | 14369/24420 [5:59:19<4:09:12,  1.49s/it]
 59%|█████▉    | 14370/24420 [5:59:20<4:09:19,  1.49s/it]
                                                         
{'loss': 0.083, 'learning_rate': 5e-05, 'epoch': 2.94}

 59%|█████▉    | 14370/24420 [5:59:20<4:09:19,  1.49s/it]
 59%|█████▉    | 14371/24420 [5:59:22<4:09:45,  1.49s/it]
 59%|█████▉    | 14372/24420 [5:59:23<4:09:39,  1.49s/it]
 59%|█████▉    | 14373/24420 [5:59:25<4:09:26,  1.49s/it]
 59%|█████▉    | 14374/24420 [5:59:26<4:09:24,  1.49s/it]
 59%|█████▉    | 14375/24420 [5:59:28<4:09:12,  1.49s/it]
                                                         
{'loss': 0.0806, 'learning_rate': 5e-05, 'epoch': 2.94}

 59%|█████▉    | 14375/24420 [5:59:28<4:09:12,  1.49s/it]
 59%|█████▉    | 14376/24420 [5:59:29<4:09:04,  1.49s/it]
 59%|█████▉    | 14377/24420 [5:59:31<4:08:38,  1.49s/it]
 59%|█████▉    | 14378/24420 [5:59:32<4:08:21,  1.48s/it]
 59%|█████▉    | 14379/24420 [5:59:34<4:08:19,  1.48s/it]
 59%|█████▉    | 14380/24420 [5:59:35<4:08:35,  1.49s/it]
                                                         
{'loss': 0.085, 'learning_rate': 5e-05, 'epoch': 2.94}

 59%|█████▉    | 14380/24420 [5:59:35<4:08:35,  1.49s/it]
 59%|█████▉    | 14381/24420 [5:59:37<4:08:47,  1.49s/it]
 59%|█████▉    | 14382/24420 [5:59:38<4:08:51,  1.49s/it]
 59%|█████▉    | 14383/24420 [5:59:40<4:08:44,  1.49s/it]
 59%|█████▉    | 14384/24420 [5:59:41<4:08:43,  1.49s/it]
 59%|█████▉    | 14385/24420 [5:59:43<4:08:53,  1.49s/it]
                                                         
{'loss': 0.0861, 'learning_rate': 5e-05, 'epoch': 2.95}

 59%|█████▉    | 14385/24420 [5:59:43<4:08:53,  1.49s/it]
 59%|█████▉    | 14386/24420 [5:59:44<4:09:05,  1.49s/it]
 59%|█████▉    | 14387/24420 [5:59:46<4:09:18,  1.49s/it]
 59%|█████▉    | 14388/24420 [5:59:47<4:09:13,  1.49s/it]
 59%|█████▉    | 14389/24420 [5:59:49<4:09:29,  1.49s/it]
 59%|█████▉    | 14390/24420 [5:59:50<4:09:26,  1.49s/it]
                                                         
{'loss': 0.0997, 'learning_rate': 5e-05, 'epoch': 2.95}

 59%|█████▉    | 14390/24420 [5:59:50<4:09:26,  1.49s/it]
 59%|█████▉    | 14391/24420 [5:59:52<4:08:46,  1.49s/it]
 59%|█████▉    | 14392/24420 [5:59:53<4:08:53,  1.49s/it]
 59%|█████▉    | 14393/24420 [5:59:55<4:08:57,  1.49s/it]
 59%|█████▉    | 14394/24420 [5:59:56<4:08:47,  1.49s/it]
 59%|█████▉    | 14395/24420 [5:59:58<4:08:56,  1.49s/it]
                                                         
{'loss': 0.0677, 'learning_rate': 5e-05, 'epoch': 2.95}

 59%|█████▉    | 14395/24420 [5:59:58<4:08:56,  1.49s/it]
 59%|█████▉    | 14396/24420 [5:59:59<4:09:04,  1.49s/it]
 59%|█████▉    | 14397/24420 [6:00:01<4:08:55,  1.49s/it]
 59%|█████▉    | 14398/24420 [6:00:02<4:08:57,  1.49s/it]
 59%|█████▉    | 14399/24420 [6:00:04<4:09:04,  1.49s/it]
 59%|█████▉    | 14400/24420 [6:00:05<4:08:38,  1.49s/it]
                                                         
{'loss': 0.0733, 'learning_rate': 5e-05, 'epoch': 2.95}

 59%|█████▉    | 14400/24420 [6:00:05<4:08:38,  1.49s/it]
 59%|█████▉    | 14401/24420 [6:00:10<6:43:36,  2.42s/it]
 59%|█████▉    | 14402/24420 [6:00:11<5:56:56,  2.14s/it]
 59%|█████▉    | 14403/24420 [6:00:13<5:24:24,  1.94s/it]
 59%|█████▉    | 14404/24420 [6:00:14<5:01:49,  1.81s/it]
 59%|█████▉    | 14405/24420 [6:00:16<4:45:37,  1.71s/it]
                                                         
{'loss': 0.0876, 'learning_rate': 5e-05, 'epoch': 2.95}

 59%|█████▉    | 14405/24420 [6:00:16<4:45:37,  1.71s/it]
 59%|█████▉    | 14406/24420 [6:00:17<4:34:48,  1.65s/it]
 59%|█████▉    | 14407/24420 [6:00:19<5:07:15,  1.84s/it]
 59%|█████▉    | 14408/24420 [6:00:21<4:49:24,  1.73s/it]
 59%|█████▉    | 14409/24420 [6:00:22<4:36:51,  1.66s/it]
 59%|█████▉    | 14410/24420 [6:00:24<4:28:07,  1.61s/it]
                                                         
{'loss': 0.0808, 'learning_rate': 5e-05, 'epoch': 2.95}

 59%|█████▉    | 14410/24420 [6:00:24<4:28:07,  1.61s/it]
 59%|█████▉    | 14411/24420 [6:00:25<4:22:35,  1.57s/it]
 59%|█████▉    | 14412/24420 [6:00:27<4:18:23,  1.55s/it]
 59%|█████▉    | 14413/24420 [6:00:28<4:15:11,  1.53s/it]
 59%|█████▉    | 14414/24420 [6:00:30<4:12:55,  1.52s/it]
 59%|█████▉    | 14415/24420 [6:00:31<4:11:48,  1.51s/it]
                                                         
{'loss': 0.0771, 'learning_rate': 5e-05, 'epoch': 2.95}

 59%|█████▉    | 14415/24420 [6:00:31<4:11:48,  1.51s/it]
 59%|█████▉    | 14416/24420 [6:00:33<4:10:20,  1.50s/it]
 59%|█████▉    | 14417/24420 [6:00:34<4:09:14,  1.49s/it]
 59%|█████▉    | 14418/24420 [6:00:36<4:08:38,  1.49s/it]
 59%|█████▉    | 14419/24420 [6:00:37<4:08:20,  1.49s/it]
 59%|█████▉    | 14420/24420 [6:00:39<4:08:21,  1.49s/it]
                                                         
{'loss': 0.0891, 'learning_rate': 5e-05, 'epoch': 2.95}

 59%|█████▉    | 14420/24420 [6:00:39<4:08:21,  1.49s/it]
 59%|█████▉    | 14421/24420 [6:00:40<4:08:13,  1.49s/it]
 59%|█████▉    | 14422/24420 [6:00:42<4:07:52,  1.49s/it]
 59%|█████▉    | 14423/24420 [6:00:43<4:07:41,  1.49s/it]
 59%|█████▉    | 14424/24420 [6:00:45<4:08:04,  1.49s/it]
 59%|█████▉    | 14425/24420 [6:00:46<4:08:05,  1.49s/it]
                                                         
{'loss': 0.0774, 'learning_rate': 5e-05, 'epoch': 2.95}

 59%|█████▉    | 14425/24420 [6:00:46<4:08:05,  1.49s/it]
 59%|█████▉    | 14426/24420 [6:00:48<4:07:44,  1.49s/it]
 59%|█████▉    | 14427/24420 [6:00:49<4:07:19,  1.49s/it]
 59%|█████▉    | 14428/24420 [6:00:51<4:07:05,  1.48s/it]
 59%|█████▉    | 14429/24420 [6:00:52<4:07:08,  1.48s/it]
 59%|█████▉    | 14430/24420 [6:00:54<4:07:15,  1.49s/it]
                                                         
{'loss': 0.0832, 'learning_rate': 5e-05, 'epoch': 2.95}

 59%|█████▉    | 14430/24420 [6:00:54<4:07:15,  1.49s/it]
 59%|█████▉    | 14431/24420 [6:00:55<4:07:34,  1.49s/it]
 59%|█████▉    | 14432/24420 [6:00:57<4:07:38,  1.49s/it]
 59%|█████▉    | 14433/24420 [6:00:58<4:07:39,  1.49s/it]
 59%|█████▉    | 14434/24420 [6:01:00<4:07:15,  1.49s/it]
 59%|█████▉    | 14435/24420 [6:01:01<4:07:15,  1.49s/it]
                                                         
{'loss': 0.082, 'learning_rate': 5e-05, 'epoch': 2.96}

 59%|█████▉    | 14435/24420 [6:01:01<4:07:15,  1.49s/it]
 59%|█████▉    | 14436/24420 [6:01:03<4:07:35,  1.49s/it]
 59%|█████▉    | 14437/24420 [6:01:04<4:07:12,  1.49s/it]
 59%|█████▉    | 14438/24420 [6:01:06<4:07:08,  1.49s/it]
 59%|█████▉    | 14439/24420 [6:01:07<4:07:36,  1.49s/it]
 59%|█████▉    | 14440/24420 [6:01:09<4:07:21,  1.49s/it]
                                                         
{'loss': 0.091, 'learning_rate': 5e-05, 'epoch': 2.96}

 59%|█████▉    | 14440/24420 [6:01:09<4:07:21,  1.49s/it]
 59%|█████▉    | 14441/24420 [6:01:10<4:07:32,  1.49s/it]
 59%|█████▉    | 14442/24420 [6:01:12<4:07:16,  1.49s/it]
 59%|█████▉    | 14443/24420 [6:01:13<4:06:36,  1.48s/it]
 59%|█████▉    | 14444/24420 [6:01:14<4:06:21,  1.48s/it]
 59%|█████▉    | 14445/24420 [6:01:16<4:06:37,  1.48s/it]
                                                         
{'loss': 0.083, 'learning_rate': 5e-05, 'epoch': 2.96}

 59%|█████▉    | 14445/24420 [6:01:16<4:06:37,  1.48s/it]
 59%|█████▉    | 14446/24420 [6:01:17<4:06:45,  1.48s/it]
 59%|█████▉    | 14447/24420 [6:01:19<4:06:46,  1.48s/it]
 59%|█████▉    | 14448/24420 [6:01:20<4:06:55,  1.49s/it]
 59%|█████▉    | 14449/24420 [6:01:22<4:07:29,  1.49s/it]
 59%|█████▉    | 14450/24420 [6:01:23<4:07:36,  1.49s/it]
                                                         
{'loss': 0.093, 'learning_rate': 5e-05, 'epoch': 2.96}

 59%|█████▉    | 14450/24420 [6:01:23<4:07:36,  1.49s/it]
 59%|█████▉    | 14451/24420 [6:01:25<4:07:28,  1.49s/it]
 59%|█████▉    | 14452/24420 [6:01:26<4:07:05,  1.49s/it]
 59%|█████▉    | 14453/24420 [6:01:28<4:06:45,  1.49s/it]
 59%|█████▉    | 14454/24420 [6:01:29<4:06:27,  1.48s/it]
 59%|█████▉    | 14455/24420 [6:01:31<4:06:23,  1.48s/it]
                                                         
{'loss': 0.0774, 'learning_rate': 5e-05, 'epoch': 2.96}

 59%|█████▉    | 14455/24420 [6:01:31<4:06:23,  1.48s/it]
 59%|█████▉    | 14456/24420 [6:01:32<4:07:04,  1.49s/it]
 59%|█████▉    | 14457/24420 [6:01:34<4:07:20,  1.49s/it]
 59%|█████▉    | 14458/24420 [6:01:35<4:07:37,  1.49s/it]
 59%|█████▉    | 14459/24420 [6:01:37<4:07:11,  1.49s/it]
 59%|█████▉    | 14460/24420 [6:01:38<4:06:58,  1.49s/it]
                                                         
{'loss': 0.0943, 'learning_rate': 5e-05, 'epoch': 2.96}

 59%|█████▉    | 14460/24420 [6:01:38<4:06:58,  1.49s/it]
 59%|█████▉    | 14461/24420 [6:01:40<4:07:13,  1.49s/it]
 59%|█████▉    | 14462/24420 [6:01:41<4:07:16,  1.49s/it]
 59%|█████▉    | 14463/24420 [6:01:43<4:07:08,  1.49s/it]
 59%|█████▉    | 14464/24420 [6:01:44<4:06:40,  1.49s/it]
 59%|█████▉    | 14465/24420 [6:01:46<4:06:35,  1.49s/it]
                                                         
{'loss': 0.091, 'learning_rate': 5e-05, 'epoch': 2.96}

 59%|█████▉    | 14465/24420 [6:01:46<4:06:35,  1.49s/it]
 59%|█████▉    | 14466/24420 [6:01:47<4:06:59,  1.49s/it]
 59%|█████▉    | 14467/24420 [6:01:49<4:06:43,  1.49s/it]
 59%|█████▉    | 14468/24420 [6:01:50<4:06:39,  1.49s/it]
 59%|█████▉    | 14469/24420 [6:01:52<4:06:24,  1.49s/it]
 59%|█████▉    | 14470/24420 [6:01:53<4:06:57,  1.49s/it]
                                                         
{'loss': 0.0831, 'learning_rate': 5e-05, 'epoch': 2.96}

 59%|█████▉    | 14470/24420 [6:01:53<4:06:57,  1.49s/it]
 59%|█████▉    | 14471/24420 [6:01:55<4:06:53,  1.49s/it]
 59%|█████▉    | 14472/24420 [6:01:56<4:07:07,  1.49s/it]
 59%|█████▉    | 14473/24420 [6:01:58<4:06:32,  1.49s/it]
 59%|█████▉    | 14474/24420 [6:01:59<4:06:13,  1.49s/it]
 59%|█████▉    | 14475/24420 [6:02:01<4:06:16,  1.49s/it]
                                                         
{'loss': 0.0722, 'learning_rate': 5e-05, 'epoch': 2.96}

 59%|█████▉    | 14475/24420 [6:02:01<4:06:16,  1.49s/it]
 59%|█████▉    | 14476/24420 [6:02:02<4:06:35,  1.49s/it]
 59%|█████▉    | 14477/24420 [6:02:04<4:06:35,  1.49s/it]
 59%|█████▉    | 14478/24420 [6:02:05<4:06:03,  1.49s/it]
 59%|█████▉    | 14479/24420 [6:02:07<4:06:32,  1.49s/it]
 59%|█████▉    | 14480/24420 [6:02:08<4:06:34,  1.49s/it]
                                                         
{'loss': 0.0846, 'learning_rate': 5e-05, 'epoch': 2.96}

 59%|█████▉    | 14480/24420 [6:02:08<4:06:34,  1.49s/it]
 59%|█████▉    | 14481/24420 [6:02:10<4:06:26,  1.49s/it]
 59%|█████▉    | 14482/24420 [6:02:11<4:06:30,  1.49s/it]
 59%|█████▉    | 14483/24420 [6:02:12<4:06:17,  1.49s/it]
 59%|█████▉    | 14484/24420 [6:02:14<4:06:37,  1.49s/it]
 59%|█████▉    | 14485/24420 [6:02:15<4:06:23,  1.49s/it]
                                                         
{'loss': 0.0822, 'learning_rate': 5e-05, 'epoch': 2.97}

 59%|█████▉    | 14485/24420 [6:02:15<4:06:23,  1.49s/it]
 59%|█████▉    | 14486/24420 [6:02:17<4:06:07,  1.49s/it]
 59%|█████▉    | 14487/24420 [6:02:18<4:06:15,  1.49s/it]
 59%|█████▉    | 14488/24420 [6:02:20<4:07:03,  1.49s/it]
 59%|█████▉    | 14489/24420 [6:02:21<4:06:43,  1.49s/it]
 59%|█████▉    | 14490/24420 [6:02:23<4:06:02,  1.49s/it]
                                                         
{'loss': 0.1018, 'learning_rate': 5e-05, 'epoch': 2.97}

 59%|█████▉    | 14490/24420 [6:02:23<4:06:02,  1.49s/it]
 59%|█████▉    | 14491/24420 [6:02:24<4:06:22,  1.49s/it]
 59%|█████▉    | 14492/24420 [6:02:26<4:06:14,  1.49s/it]
 59%|█████▉    | 14493/24420 [6:02:27<4:06:13,  1.49s/it]
 59%|█████▉    | 14494/24420 [6:02:29<4:06:29,  1.49s/it]
 59%|█████▉    | 14495/24420 [6:02:30<4:06:22,  1.49s/it]
                                                         
{'loss': 0.0866, 'learning_rate': 5e-05, 'epoch': 2.97}

 59%|█████▉    | 14495/24420 [6:02:30<4:06:22,  1.49s/it]
 59%|█████▉    | 14496/24420 [6:02:32<4:06:45,  1.49s/it]
 59%|█████▉    | 14497/24420 [6:02:33<4:06:43,  1.49s/it]
 59%|█████▉    | 14498/24420 [6:02:35<4:06:28,  1.49s/it]
 59%|█████▉    | 14499/24420 [6:02:36<4:06:31,  1.49s/it]
 59%|█████▉    | 14500/24420 [6:02:38<4:06:24,  1.49s/it]
                                                         
{'loss': 0.0776, 'learning_rate': 5e-05, 'epoch': 2.97}

 59%|█████▉    | 14500/24420 [6:02:38<4:06:24,  1.49s/it]
 59%|█████▉    | 14501/24420 [6:02:39<4:06:03,  1.49s/it]
 59%|█████▉    | 14502/24420 [6:02:41<4:06:10,  1.49s/it]
 59%|█████▉    | 14503/24420 [6:02:42<4:06:19,  1.49s/it]
 59%|█████▉    | 14504/24420 [6:02:44<4:06:23,  1.49s/it]
 59%|█████▉    | 14505/24420 [6:02:45<4:06:45,  1.49s/it]
                                                         
{'loss': 0.0939, 'learning_rate': 5e-05, 'epoch': 2.97}

 59%|█████▉    | 14505/24420 [6:02:45<4:06:45,  1.49s/it]
 59%|█████▉    | 14506/24420 [6:02:47<4:06:44,  1.49s/it]
 59%|█████▉    | 14507/24420 [6:02:48<4:06:08,  1.49s/it]
 59%|█████▉    | 14508/24420 [6:02:50<4:05:41,  1.49s/it]
 59%|█████▉    | 14509/24420 [6:02:51<4:05:35,  1.49s/it]
 59%|█████▉    | 14510/24420 [6:02:53<4:06:00,  1.49s/it]
                                                         
{'loss': 0.0867, 'learning_rate': 5e-05, 'epoch': 2.97}

 59%|█████▉    | 14510/24420 [6:02:53<4:06:00,  1.49s/it]
 59%|█████▉    | 14511/24420 [6:02:54<4:06:15,  1.49s/it]
 59%|█████▉    | 14512/24420 [6:02:56<4:06:08,  1.49s/it]
 59%|█████▉    | 14513/24420 [6:02:57<4:06:27,  1.49s/it]
 59%|█████▉    | 14514/24420 [6:02:59<4:06:00,  1.49s/it]
 59%|█████▉    | 14515/24420 [6:03:00<4:06:27,  1.49s/it]
                                                         
{'loss': 0.0859, 'learning_rate': 5e-05, 'epoch': 2.97}

 59%|█████▉    | 14515/24420 [6:03:00<4:06:27,  1.49s/it]
 59%|█████▉    | 14516/24420 [6:03:02<4:06:33,  1.49s/it]
 59%|█████▉    | 14517/24420 [6:03:03<4:06:50,  1.50s/it]
 59%|█████▉    | 14518/24420 [6:03:05<4:06:21,  1.49s/it]
 59%|█████▉    | 14519/24420 [6:03:06<4:05:48,  1.49s/it]
 59%|█████▉    | 14520/24420 [6:03:08<4:05:44,  1.49s/it]
                                                         
{'loss': 0.0847, 'learning_rate': 5e-05, 'epoch': 2.97}

 59%|█████▉    | 14520/24420 [6:03:08<4:05:44,  1.49s/it]
 59%|█████▉    | 14521/24420 [6:03:09<4:05:48,  1.49s/it]
 59%|█████▉    | 14522/24420 [6:03:11<4:06:02,  1.49s/it]
 59%|█████▉    | 14523/24420 [6:03:12<4:06:38,  1.50s/it]
 59%|█████▉    | 14524/24420 [6:03:14<4:06:36,  1.50s/it]
 59%|█████▉    | 14525/24420 [6:03:15<4:06:14,  1.49s/it]
                                                         
{'loss': 0.1075, 'learning_rate': 5e-05, 'epoch': 2.97}

 59%|█████▉    | 14525/24420 [6:03:15<4:06:14,  1.49s/it]
 59%|█████▉    | 14526/24420 [6:03:17<4:06:41,  1.50s/it]
 59%|█████▉    | 14527/24420 [6:03:18<4:06:16,  1.49s/it]
 59%|█████▉    | 14528/24420 [6:03:20<4:05:36,  1.49s/it]
 59%|█████▉    | 14529/24420 [6:03:21<4:05:06,  1.49s/it]
 60%|█████▉    | 14530/24420 [6:03:23<4:04:53,  1.49s/it]
                                                         
{'loss': 0.0745, 'learning_rate': 5e-05, 'epoch': 2.98}

 60%|█████▉    | 14530/24420 [6:03:23<4:04:53,  1.49s/it]
 60%|█████▉    | 14531/24420 [6:03:24<4:05:04,  1.49s/it]
 60%|█████▉    | 14532/24420 [6:03:26<4:04:59,  1.49s/it]
 60%|█████▉    | 14533/24420 [6:03:27<4:06:05,  1.49s/it]
 60%|█████▉    | 14534/24420 [6:03:29<4:05:58,  1.49s/it]
 60%|█████▉    | 14535/24420 [6:03:30<4:05:58,  1.49s/it]
                                                         
{'loss': 0.0758, 'learning_rate': 5e-05, 'epoch': 2.98}

 60%|█████▉    | 14535/24420 [6:03:30<4:05:58,  1.49s/it]
 60%|█████▉    | 14536/24420 [6:03:32<4:05:35,  1.49s/it]
 60%|█████▉    | 14537/24420 [6:03:33<4:05:17,  1.49s/it]
 60%|█████▉    | 14538/24420 [6:03:34<4:05:12,  1.49s/it]
 60%|█████▉    | 14539/24420 [6:03:36<4:05:37,  1.49s/it]
 60%|█████▉    | 14540/24420 [6:03:37<4:05:50,  1.49s/it]
                                                         
{'loss': 0.0751, 'learning_rate': 5e-05, 'epoch': 2.98}

 60%|█████▉    | 14540/24420 [6:03:37<4:05:50,  1.49s/it]
 60%|█████▉    | 14541/24420 [6:03:39<4:05:41,  1.49s/it]
 60%|█████▉    | 14542/24420 [6:03:40<4:06:27,  1.50s/it]
 60%|█████▉    | 14543/24420 [6:03:42<4:06:14,  1.50s/it]
 60%|█████▉    | 14544/24420 [6:03:43<4:05:50,  1.49s/it]
 60%|█████▉    | 14545/24420 [6:03:45<4:05:34,  1.49s/it]
                                                         
{'loss': 0.0855, 'learning_rate': 5e-05, 'epoch': 2.98}

 60%|█████▉    | 14545/24420 [6:03:45<4:05:34,  1.49s/it]
 60%|█████▉    | 14546/24420 [6:03:46<4:06:20,  1.50s/it]
 60%|█████▉    | 14547/24420 [6:03:48<4:06:30,  1.50s/it]
 60%|█████▉    | 14548/24420 [6:03:49<4:05:36,  1.49s/it]
 60%|█████▉    | 14549/24420 [6:03:51<4:05:03,  1.49s/it]
 60%|█████▉    | 14550/24420 [6:03:52<4:04:56,  1.49s/it]
                                                         
{'loss': 0.0825, 'learning_rate': 5e-05, 'epoch': 2.98}

 60%|█████▉    | 14550/24420 [6:03:52<4:04:56,  1.49s/it]
 60%|█████▉    | 14551/24420 [6:03:54<4:05:37,  1.49s/it]
 60%|█████▉    | 14552/24420 [6:03:55<4:06:36,  1.50s/it]
 60%|█████▉    | 14553/24420 [6:03:57<4:06:13,  1.50s/it]
 60%|█████▉    | 14554/24420 [6:03:58<4:06:07,  1.50s/it]
 60%|█████▉    | 14555/24420 [6:04:00<4:05:56,  1.50s/it]
                                                         
{'loss': 0.0815, 'learning_rate': 5e-05, 'epoch': 2.98}

 60%|█████▉    | 14555/24420 [6:04:00<4:05:56,  1.50s/it]
 60%|█████▉    | 14556/24420 [6:04:01<4:05:47,  1.50s/it]
 60%|█████▉    | 14557/24420 [6:04:03<4:05:22,  1.49s/it]
 60%|█████▉    | 14558/24420 [6:04:04<4:05:03,  1.49s/it]
 60%|█████▉    | 14559/24420 [6:04:06<4:04:57,  1.49s/it]
 60%|█████▉    | 14560/24420 [6:04:07<4:05:21,  1.49s/it]
                                                         
{'loss': 0.079, 'learning_rate': 5e-05, 'epoch': 2.98}

 60%|█████▉    | 14560/24420 [6:04:07<4:05:21,  1.49s/it]
 60%|█████▉    | 14561/24420 [6:04:09<4:04:57,  1.49s/it]
 60%|█████▉    | 14562/24420 [6:04:10<4:04:49,  1.49s/it]
 60%|█████▉    | 14563/24420 [6:04:12<4:04:10,  1.49s/it]
 60%|█████▉    | 14564/24420 [6:04:13<4:04:08,  1.49s/it]
 60%|█████▉    | 14565/24420 [6:04:15<4:03:39,  1.48s/it]
                                                         
{'loss': 0.0852, 'learning_rate': 5e-05, 'epoch': 2.98}

 60%|█████▉    | 14565/24420 [6:04:15<4:03:39,  1.48s/it]
 60%|█████▉    | 14566/24420 [6:04:16<4:03:49,  1.48s/it]
 60%|█████▉    | 14567/24420 [6:04:18<4:04:25,  1.49s/it]
 60%|█████▉    | 14568/24420 [6:04:19<4:04:15,  1.49s/it]
 60%|█████▉    | 14569/24420 [6:04:21<4:04:19,  1.49s/it]
 60%|█████▉    | 14570/24420 [6:04:22<4:05:12,  1.49s/it]
                                                         
{'loss': 0.0818, 'learning_rate': 5e-05, 'epoch': 2.98}

 60%|█████▉    | 14570/24420 [6:04:22<4:05:12,  1.49s/it]
 60%|█████▉    | 14571/24420 [6:04:24<4:04:35,  1.49s/it]
 60%|█████▉    | 14572/24420 [6:04:25<4:04:22,  1.49s/it]
 60%|█████▉    | 14573/24420 [6:04:27<4:04:20,  1.49s/it]
 60%|█████▉    | 14574/24420 [6:04:28<4:04:22,  1.49s/it]
 60%|█████▉    | 14575/24420 [6:04:30<4:04:29,  1.49s/it]
                                                         
{'loss': 0.0869, 'learning_rate': 5e-05, 'epoch': 2.98}

 60%|█████▉    | 14575/24420 [6:04:30<4:04:29,  1.49s/it]
 60%|█████▉    | 14576/24420 [6:04:31<4:05:08,  1.49s/it]
 60%|█████▉    | 14577/24420 [6:04:33<4:04:42,  1.49s/it]
 60%|█████▉    | 14578/24420 [6:04:34<4:04:58,  1.49s/it]
 60%|█████▉    | 14579/24420 [6:04:36<4:04:35,  1.49s/it]
 60%|█████▉    | 14580/24420 [6:04:37<4:04:12,  1.49s/it]
                                                         
{'loss': 0.1085, 'learning_rate': 5e-05, 'epoch': 2.99}

 60%|█████▉    | 14580/24420 [6:04:37<4:04:12,  1.49s/it]
 60%|█████▉    | 14581/24420 [6:04:39<4:04:44,  1.49s/it]
 60%|█████▉    | 14582/24420 [6:04:40<4:04:18,  1.49s/it]
 60%|█████▉    | 14583/24420 [6:04:42<4:04:04,  1.49s/it]
 60%|█████▉    | 14584/24420 [6:04:43<4:04:12,  1.49s/it]
 60%|█████▉    | 14585/24420 [6:04:45<4:04:58,  1.49s/it]
                                                         
{'loss': 0.0683, 'learning_rate': 5e-05, 'epoch': 2.99}

 60%|█████▉    | 14585/24420 [6:04:45<4:04:58,  1.49s/it]
 60%|█████▉    | 14586/24420 [6:04:46<4:04:45,  1.49s/it]
 60%|█████▉    | 14587/24420 [6:04:48<4:05:07,  1.50s/it]
 60%|█████▉    | 14588/24420 [6:04:49<4:05:18,  1.50s/it]
 60%|█████▉    | 14589/24420 [6:04:51<4:04:37,  1.49s/it]
 60%|█████▉    | 14590/24420 [6:04:52<4:04:19,  1.49s/it]
                                                         
{'loss': 0.0874, 'learning_rate': 5e-05, 'epoch': 2.99}

 60%|█████▉    | 14590/24420 [6:04:52<4:04:19,  1.49s/it]
 60%|█████▉    | 14591/24420 [6:04:54<4:04:31,  1.49s/it]
 60%|█████▉    | 14592/24420 [6:04:55<4:04:26,  1.49s/it]
 60%|█████▉    | 14593/24420 [6:04:57<4:04:31,  1.49s/it]
 60%|█████▉    | 14594/24420 [6:04:58<4:03:23,  1.49s/it]
 60%|█████▉    | 14595/24420 [6:04:59<4:03:18,  1.49s/it]
                                                         
{'loss': 0.0858, 'learning_rate': 5e-05, 'epoch': 2.99}

 60%|█████▉    | 14595/24420 [6:05:00<4:03:18,  1.49s/it]
 60%|█████▉    | 14596/24420 [6:05:01<4:03:13,  1.49s/it]
 60%|█████▉    | 14597/24420 [6:05:02<4:03:17,  1.49s/it]
 60%|█████▉    | 14598/24420 [6:05:04<4:03:15,  1.49s/it]
 60%|█████▉    | 14599/24420 [6:05:05<4:03:38,  1.49s/it]
 60%|█████▉    | 14600/24420 [6:05:07<4:03:44,  1.49s/it]
                                                         
{'loss': 0.0923, 'learning_rate': 5e-05, 'epoch': 2.99}

 60%|█████▉    | 14600/24420 [6:05:07<4:03:44,  1.49s/it]
 60%|█████▉    | 14601/24420 [6:05:12<6:36:32,  2.42s/it]
 60%|█████▉    | 14602/24420 [6:05:13<5:49:57,  2.14s/it]
 60%|█████▉    | 14603/24420 [6:05:14<5:17:28,  1.94s/it]
 60%|█████▉    | 14604/24420 [6:05:16<4:54:54,  1.80s/it]
 60%|█████▉    | 14605/24420 [6:05:17<4:39:23,  1.71s/it]
                                                         
{'loss': 0.0755, 'learning_rate': 5e-05, 'epoch': 2.99}

 60%|█████▉    | 14605/24420 [6:05:17<4:39:23,  1.71s/it]
 60%|█████▉    | 14606/24420 [6:05:19<4:28:40,  1.64s/it]
 60%|█████▉    | 14607/24420 [6:05:20<4:20:53,  1.60s/it]
 60%|█████▉    | 14608/24420 [6:05:22<4:15:27,  1.56s/it]
 60%|█████▉    | 14609/24420 [6:05:23<4:11:34,  1.54s/it]
 60%|█████▉    | 14610/24420 [6:05:25<4:09:03,  1.52s/it]
                                                         
{'loss': 0.092, 'learning_rate': 5e-05, 'epoch': 2.99}

 60%|█████▉    | 14610/24420 [6:05:25<4:09:03,  1.52s/it]
 60%|█████▉    | 14611/24420 [6:05:26<4:07:14,  1.51s/it]
 60%|█████▉    | 14612/24420 [6:05:28<4:06:38,  1.51s/it]
 60%|█████▉    | 14613/24420 [6:05:29<4:05:22,  1.50s/it]
 60%|█████▉    | 14614/24420 [6:05:31<4:04:26,  1.50s/it]
 60%|█████▉    | 14615/24420 [6:05:32<4:03:54,  1.49s/it]
                                                         
{'loss': 0.0946, 'learning_rate': 5e-05, 'epoch': 2.99}

 60%|█████▉    | 14615/24420 [6:05:32<4:03:54,  1.49s/it]
 60%|█████▉    | 14616/24420 [6:05:34<4:03:45,  1.49s/it]
 60%|█████▉    | 14617/24420 [6:05:35<4:03:29,  1.49s/it]
 60%|█████▉    | 14618/24420 [6:05:37<4:03:15,  1.49s/it]
 60%|█████▉    | 14619/24420 [6:05:38<4:03:00,  1.49s/it]
 60%|█████▉    | 14620/24420 [6:05:40<4:03:11,  1.49s/it]
                                                         
{'loss': 0.0772, 'learning_rate': 5e-05, 'epoch': 2.99}

 60%|█████▉    | 14620/24420 [6:05:40<4:03:11,  1.49s/it]
 60%|█████▉    | 14621/24420 [6:05:41<4:03:41,  1.49s/it]
 60%|█████▉    | 14622/24420 [6:05:43<4:03:14,  1.49s/it]
 60%|█████▉    | 14623/24420 [6:05:44<4:03:13,  1.49s/it]
 60%|█████▉    | 14624/24420 [6:05:46<4:03:19,  1.49s/it]
 60%|█████▉    | 14625/24420 [6:05:47<4:02:49,  1.49s/it]
                                                         
{'loss': 0.0998, 'learning_rate': 5e-05, 'epoch': 2.99}

 60%|█████▉    | 14625/24420 [6:05:47<4:02:49,  1.49s/it]
 60%|█████▉    | 14626/24420 [6:05:49<4:02:27,  1.49s/it]
 60%|█████▉    | 14627/24420 [6:05:50<4:03:06,  1.49s/it]
 60%|█████▉    | 14628/24420 [6:05:52<4:03:12,  1.49s/it]
 60%|█████▉    | 14629/24420 [6:05:53<4:03:09,  1.49s/it]
 60%|█████▉    | 14630/24420 [6:05:55<4:03:29,  1.49s/it]
                                                         
{'loss': 0.085, 'learning_rate': 5e-05, 'epoch': 3.0}

 60%|█████▉    | 14630/24420 [6:05:55<4:03:29,  1.49s/it]
 60%|█████▉    | 14631/24420 [6:05:56<4:03:09,  1.49s/it]
 60%|█████▉    | 14632/24420 [6:05:58<4:03:09,  1.49s/it]
 60%|█████▉    | 14633/24420 [6:05:59<4:02:48,  1.49s/it]
 60%|█████▉    | 14634/24420 [6:06:01<4:03:06,  1.49s/it]
 60%|█████▉    | 14635/24420 [6:06:02<4:02:38,  1.49s/it]
                                                         
{'loss': 0.0706, 'learning_rate': 5e-05, 'epoch': 3.0}

 60%|█████▉    | 14635/24420 [6:06:02<4:02:38,  1.49s/it]
 60%|█████▉    | 14636/24420 [6:06:04<4:03:02,  1.49s/it]
 60%|█████▉    | 14637/24420 [6:06:05<4:03:24,  1.49s/it]
 60%|█████▉    | 14638/24420 [6:06:07<4:03:19,  1.49s/it]
 60%|█████▉    | 14639/24420 [6:06:08<4:03:02,  1.49s/it]
 60%|█████▉    | 14640/24420 [6:06:10<4:02:32,  1.49s/it]
                                                         
{'loss': 0.1095, 'learning_rate': 5e-05, 'epoch': 3.0}

 60%|█████▉    | 14640/24420 [6:06:10<4:02:32,  1.49s/it]
 60%|█████▉    | 14641/24420 [6:06:11<4:02:21,  1.49s/it]
 60%|█████▉    | 14642/24420 [6:06:13<4:02:21,  1.49s/it]
 60%|█████▉    | 14643/24420 [6:06:14<4:02:29,  1.49s/it]
 60%|█████▉    | 14644/24420 [6:06:16<4:02:19,  1.49s/it]
 60%|█████▉    | 14645/24420 [6:06:17<4:02:10,  1.49s/it]
                                                         
{'loss': 0.0897, 'learning_rate': 5e-05, 'epoch': 3.0}

 60%|█████▉    | 14645/24420 [6:06:17<4:02:10,  1.49s/it]
 60%|█████▉    | 14646/24420 [6:06:18<4:02:18,  1.49s/it]
 60%|█████▉    | 14647/24420 [6:06:20<4:02:46,  1.49s/it]
 60%|█████▉    | 14648/24420 [6:06:21<4:02:55,  1.49s/it]
 60%|█████▉    | 14649/24420 [6:06:23<4:02:40,  1.49s/it]
 60%|█████▉    | 14650/24420 [6:06:24<4:02:38,  1.49s/it]
                                                         
{'loss': 0.0859, 'learning_rate': 5e-05, 'epoch': 3.0}

 60%|█████▉    | 14650/24420 [6:06:24<4:02:38,  1.49s/it]
 60%|█████▉    | 14651/24420 [6:06:26<4:02:17,  1.49s/it]
 60%|██████    | 14652/24420 [6:06:27<4:02:06,  1.49s/it]
 60%|██████    | 14653/24420 [6:06:29<4:03:25,  1.50s/it]
 60%|██████    | 14654/24420 [6:06:30<4:03:02,  1.49s/it]
 60%|██████    | 14655/24420 [6:06:32<4:02:41,  1.49s/it]
                                                         
{'loss': 0.0761, 'learning_rate': 5e-05, 'epoch': 3.0}

 60%|██████    | 14655/24420 [6:06:32<4:02:41,  1.49s/it]
 60%|██████    | 14656/24420 [6:06:33<4:02:17,  1.49s/it]
 60%|██████    | 14657/24420 [6:06:35<4:01:48,  1.49s/it]
 60%|██████    | 14658/24420 [6:06:36<4:01:44,  1.49s/it]
 60%|██████    | 14659/24420 [6:06:38<4:01:49,  1.49s/it]
 60%|██████    | 14660/24420 [6:06:39<4:01:41,  1.49s/it]
                                                         
{'loss': 0.0917, 'learning_rate': 5e-05, 'epoch': 3.0}

 60%|██████    | 14660/24420 [6:06:39<4:01:41,  1.49s/it]
 60%|██████    | 14661/24420 [6:06:41<4:02:06,  1.49s/it]
 60%|██████    | 14662/24420 [6:06:42<4:02:22,  1.49s/it]
 60%|██████    | 14663/24420 [6:06:44<4:02:38,  1.49s/it]
 60%|██████    | 14664/24420 [6:06:45<4:02:37,  1.49s/it]
 60%|██████    | 14665/24420 [6:06:47<4:02:35,  1.49s/it]
                                                         
{'loss': 0.1033, 'learning_rate': 5e-05, 'epoch': 3.0}

 60%|██████    | 14665/24420 [6:06:47<4:02:35,  1.49s/it]
 60%|██████    | 14666/24420 [6:06:48<4:02:34,  1.49s/it]
 60%|██████    | 14667/24420 [6:06:50<4:02:44,  1.49s/it]
 60%|██████    | 14668/24420 [6:06:51<4:03:01,  1.50s/it]
 60%|██████    | 14669/24420 [6:06:53<4:02:55,  1.49s/it]
 60%|██████    | 14670/24420 [6:06:54<4:02:23,  1.49s/it]
                                                         
{'loss': 0.0967, 'learning_rate': 5e-05, 'epoch': 3.0}

 60%|██████    | 14670/24420 [6:06:54<4:02:23,  1.49s/it]
 60%|██████    | 14671/24420 [6:06:56<4:02:22,  1.49s/it]
 60%|██████    | 14672/24420 [6:06:57<4:02:06,  1.49s/it]
 60%|██████    | 14673/24420 [6:06:59<4:01:46,  1.49s/it]
 60%|██████    | 14674/24420 [6:07:00<4:01:45,  1.49s/it]
 60%|██████    | 14675/24420 [6:07:02<4:02:01,  1.49s/it]
                                                         
{'loss': 0.079, 'learning_rate': 5e-05, 'epoch': 3.0}

 60%|██████    | 14675/24420 [6:07:02<4:02:01,  1.49s/it]
 60%|██████    | 14676/24420 [6:07:03<4:02:22,  1.49s/it]
 60%|██████    | 14677/24420 [6:07:05<4:02:17,  1.49s/it]
 60%|██████    | 14678/24420 [6:07:06<4:02:19,  1.49s/it]
 60%|██████    | 14679/24420 [6:07:08<4:02:10,  1.49s/it]
 60%|██████    | 14680/24420 [6:07:09<4:01:44,  1.49s/it]
                                                         
{'loss': 0.0779, 'learning_rate': 5e-05, 'epoch': 3.01}

 60%|██████    | 14680/24420 [6:07:09<4:01:44,  1.49s/it]
 60%|██████    | 14681/24420 [6:07:11<4:01:37,  1.49s/it]
 60%|██████    | 14682/24420 [6:07:12<4:01:26,  1.49s/it]
 60%|██████    | 14683/24420 [6:07:14<4:01:15,  1.49s/it]
 60%|██████    | 14684/24420 [6:07:15<4:01:07,  1.49s/it]
 60%|██████    | 14685/24420 [6:07:17<4:01:18,  1.49s/it]
                                                         
{'loss': 0.0747, 'learning_rate': 5e-05, 'epoch': 3.01}

 60%|██████    | 14685/24420 [6:07:17<4:01:18,  1.49s/it]
 60%|██████    | 14686/24420 [6:07:18<4:01:21,  1.49s/it]
 60%|██████    | 14687/24420 [6:07:20<4:01:05,  1.49s/it]
 60%|██████    | 14688/24420 [6:07:21<4:01:11,  1.49s/it]
 60%|██████    | 14689/24420 [6:07:23<4:01:07,  1.49s/it]
 60%|██████    | 14690/24420 [6:07:24<4:01:15,  1.49s/it]
                                                         
{'loss': 0.0738, 'learning_rate': 5e-05, 'epoch': 3.01}

 60%|██████    | 14690/24420 [6:07:24<4:01:15,  1.49s/it]
 60%|██████    | 14691/24420 [6:07:26<4:01:24,  1.49s/it]
 60%|██████    | 14692/24420 [6:07:27<4:01:11,  1.49s/it]
 60%|██████    | 14693/24420 [6:07:29<4:01:23,  1.49s/it]
 60%|██████    | 14694/24420 [6:07:30<4:01:19,  1.49s/it]
 60%|██████    | 14695/24420 [6:07:31<4:00:50,  1.49s/it]
                                                         
{'loss': 0.0731, 'learning_rate': 5e-05, 'epoch': 3.01}

 60%|██████    | 14695/24420 [6:07:31<4:00:50,  1.49s/it]
 60%|██████    | 14696/24420 [6:07:33<4:01:08,  1.49s/it]
 60%|██████    | 14697/24420 [6:07:34<4:01:28,  1.49s/it]
 60%|██████    | 14698/24420 [6:07:36<4:01:27,  1.49s/it]
 60%|██████    | 14699/24420 [6:07:37<4:01:01,  1.49s/it]
 60%|██████    | 14700/24420 [6:07:39<4:01:26,  1.49s/it]
                                                         
{'loss': 0.0703, 'learning_rate': 5e-05, 'epoch': 3.01}

 60%|██████    | 14700/24420 [6:07:39<4:01:26,  1.49s/it]
 60%|██████    | 14701/24420 [6:07:40<4:01:36,  1.49s/it]
 60%|██████    | 14702/24420 [6:07:42<4:01:50,  1.49s/it]
 60%|██████    | 14703/24420 [6:07:43<4:01:44,  1.49s/it]
 60%|██████    | 14704/24420 [6:07:45<4:01:20,  1.49s/it]
 60%|██████    | 14705/24420 [6:07:46<4:01:16,  1.49s/it]
                                                         
{'loss': 0.0738, 'learning_rate': 5e-05, 'epoch': 3.01}

 60%|██████    | 14705/24420 [6:07:46<4:01:16,  1.49s/it]
 60%|██████    | 14706/24420 [6:07:48<4:01:19,  1.49s/it]
 60%|██████    | 14707/24420 [6:07:49<4:01:20,  1.49s/it]
 60%|██████    | 14708/24420 [6:07:51<4:01:43,  1.49s/it]
 60%|██████    | 14709/24420 [6:07:52<4:01:09,  1.49s/it]
 60%|██████    | 14710/24420 [6:07:54<4:01:12,  1.49s/it]
                                                         
{'loss': 0.0845, 'learning_rate': 5e-05, 'epoch': 3.01}

 60%|██████    | 14710/24420 [6:07:54<4:01:12,  1.49s/it]
 60%|██████    | 14711/24420 [6:07:55<4:01:27,  1.49s/it]
 60%|██████    | 14712/24420 [6:07:57<4:01:31,  1.49s/it]
 60%|██████    | 14713/24420 [6:07:58<4:01:25,  1.49s/it]
 60%|██████    | 14714/24420 [6:08:00<4:02:04,  1.50s/it]
 60%|██████    | 14715/24420 [6:08:01<4:01:52,  1.50s/it]
                                                         
{'loss': 0.0867, 'learning_rate': 5e-05, 'epoch': 3.01}

 60%|██████    | 14715/24420 [6:08:01<4:01:52,  1.50s/it]
 60%|██████    | 14716/24420 [6:08:03<4:01:36,  1.49s/it]
 60%|██████    | 14717/24420 [6:08:04<4:00:58,  1.49s/it]
 60%|██████    | 14718/24420 [6:08:06<4:01:00,  1.49s/it]
 60%|██████    | 14719/24420 [6:08:07<4:00:36,  1.49s/it]
 60%|██████    | 14720/24420 [6:08:09<4:00:59,  1.49s/it]
                                                         
{'loss': 0.0829, 'learning_rate': 5e-05, 'epoch': 3.01}

 60%|██████    | 14720/24420 [6:08:09<4:00:59,  1.49s/it]
 60%|██████    | 14721/24420 [6:08:10<4:00:44,  1.49s/it]
 60%|██████    | 14722/24420 [6:08:12<4:00:44,  1.49s/it]
 60%|██████    | 14723/24420 [6:08:13<4:01:04,  1.49s/it]
 60%|██████    | 14724/24420 [6:08:15<4:00:49,  1.49s/it]
 60%|██████    | 14725/24420 [6:08:16<4:00:49,  1.49s/it]
                                                         
{'loss': 0.0973, 'learning_rate': 5e-05, 'epoch': 3.01}

 60%|██████    | 14725/24420 [6:08:16<4:00:49,  1.49s/it]
 60%|██████    | 14726/24420 [6:08:18<4:01:02,  1.49s/it]
 60%|██████    | 14727/24420 [6:08:19<4:00:51,  1.49s/it]
 60%|██████    | 14728/24420 [6:08:21<4:00:49,  1.49s/it]
 60%|██████    | 14729/24420 [6:08:22<4:01:06,  1.49s/it]
 60%|██████    | 14730/24420 [6:08:24<4:01:19,  1.49s/it]
                                                         
{'loss': 0.0767, 'learning_rate': 5e-05, 'epoch': 3.02}

 60%|██████    | 14730/24420 [6:08:24<4:01:19,  1.49s/it]
 60%|██████    | 14731/24420 [6:08:25<4:01:46,  1.50s/it]
 60%|██████    | 14732/24420 [6:08:27<4:01:07,  1.49s/it]
 60%|██████    | 14733/24420 [6:08:28<4:00:49,  1.49s/it]
 60%|██████    | 14734/24420 [6:08:30<4:01:42,  1.50s/it]
 60%|██████    | 14735/24420 [6:08:31<4:01:17,  1.49s/it]
                                                         
{'loss': 0.0777, 'learning_rate': 5e-05, 'epoch': 3.02}

 60%|██████    | 14735/24420 [6:08:31<4:01:17,  1.49s/it]
 60%|██████    | 14736/24420 [6:08:33<4:01:06,  1.49s/it]
 60%|██████    | 14737/24420 [6:08:34<4:01:49,  1.50s/it]
 60%|██████    | 14738/24420 [6:08:36<4:01:20,  1.50s/it]
 60%|██████    | 14739/24420 [6:08:37<4:01:52,  1.50s/it]
 60%|██████    | 14740/24420 [6:08:39<4:01:02,  1.49s/it]
                                                         
{'loss': 0.0794, 'learning_rate': 5e-05, 'epoch': 3.02}

 60%|██████    | 14740/24420 [6:08:39<4:01:02,  1.49s/it]
 60%|██████    | 14741/24420 [6:08:40<4:00:48,  1.49s/it]
 60%|██████    | 14742/24420 [6:08:42<4:00:12,  1.49s/it]
 60%|██████    | 14743/24420 [6:08:43<4:00:15,  1.49s/it]
 60%|██████    | 14744/24420 [6:08:45<4:00:38,  1.49s/it]
 60%|██████    | 14745/24420 [6:08:46<4:01:13,  1.50s/it]
                                                         
{'loss': 0.0736, 'learning_rate': 5e-05, 'epoch': 3.02}

 60%|██████    | 14745/24420 [6:08:46<4:01:13,  1.50s/it]
 60%|██████    | 14746/24420 [6:08:48<4:01:21,  1.50s/it]
 60%|██████    | 14747/24420 [6:08:49<4:01:11,  1.50s/it]
 60%|██████    | 14748/24420 [6:08:51<4:00:58,  1.49s/it]
 60%|██████    | 14749/24420 [6:08:52<4:00:38,  1.49s/it]
 60%|██████    | 14750/24420 [6:08:54<4:00:13,  1.49s/it]
                                                         
{'loss': 0.0715, 'learning_rate': 5e-05, 'epoch': 3.02}

 60%|██████    | 14750/24420 [6:08:54<4:00:13,  1.49s/it]
 60%|██████    | 14751/24420 [6:08:55<4:00:19,  1.49s/it]
 60%|██████    | 14752/24420 [6:08:57<3:59:57,  1.49s/it]
 60%|██████    | 14753/24420 [6:08:58<4:00:09,  1.49s/it]
 60%|██████    | 14754/24420 [6:09:00<3:59:57,  1.49s/it]
 60%|██████    | 14755/24420 [6:09:01<3:59:52,  1.49s/it]
                                                         
{'loss': 0.0746, 'learning_rate': 5e-05, 'epoch': 3.02}

 60%|██████    | 14755/24420 [6:09:01<3:59:52,  1.49s/it]
 60%|██████    | 14756/24420 [6:09:03<4:00:08,  1.49s/it]
 60%|██████    | 14757/24420 [6:09:04<4:00:20,  1.49s/it]
 60%|██████    | 14758/24420 [6:09:06<4:00:17,  1.49s/it]
 60%|██████    | 14759/24420 [6:09:07<3:59:53,  1.49s/it]
 60%|██████    | 14760/24420 [6:09:08<4:00:05,  1.49s/it]
                                                         
{'loss': 0.0765, 'learning_rate': 5e-05, 'epoch': 3.02}

 60%|██████    | 14760/24420 [6:09:08<4:00:05,  1.49s/it]
 60%|██████    | 14761/24420 [6:09:10<3:59:49,  1.49s/it]
 60%|██████    | 14762/24420 [6:09:11<3:59:58,  1.49s/it]
 60%|██████    | 14763/24420 [6:09:13<4:00:10,  1.49s/it]
 60%|██████    | 14764/24420 [6:09:14<4:00:05,  1.49s/it]
 60%|██████    | 14765/24420 [6:09:16<3:59:41,  1.49s/it]
                                                         
{'loss': 0.0791, 'learning_rate': 5e-05, 'epoch': 3.02}

 60%|██████    | 14765/24420 [6:09:16<3:59:41,  1.49s/it]
 60%|██████    | 14766/24420 [6:09:17<3:59:58,  1.49s/it]
 60%|██████    | 14767/24420 [6:09:19<3:59:44,  1.49s/it]
 60%|██████    | 14768/24420 [6:09:20<3:59:30,  1.49s/it]
 60%|██████    | 14769/24420 [6:09:22<3:59:36,  1.49s/it]
 60%|██████    | 14770/24420 [6:09:23<3:59:58,  1.49s/it]
                                                         
{'loss': 0.0888, 'learning_rate': 5e-05, 'epoch': 3.02}

 60%|██████    | 14770/24420 [6:09:23<3:59:58,  1.49s/it]
 60%|██████    | 14771/24420 [6:09:25<3:59:42,  1.49s/it]
 60%|██████    | 14772/24420 [6:09:26<3:59:22,  1.49s/it]
 60%|██████    | 14773/24420 [6:09:28<4:00:17,  1.49s/it]
 60%|██████    | 14774/24420 [6:09:29<4:00:08,  1.49s/it]
 61%|██████    | 14775/24420 [6:09:31<4:00:30,  1.50s/it]
                                                         
{'loss': 0.085, 'learning_rate': 5e-05, 'epoch': 3.03}

 61%|██████    | 14775/24420 [6:09:31<4:00:30,  1.50s/it]
 61%|██████    | 14776/24420 [6:09:32<4:00:13,  1.49s/it]
 61%|██████    | 14777/24420 [6:09:34<4:00:01,  1.49s/it]
 61%|██████    | 14778/24420 [6:09:35<3:59:33,  1.49s/it]
 61%|██████    | 14779/24420 [6:09:37<3:59:06,  1.49s/it]
 61%|██████    | 14780/24420 [6:09:38<3:59:01,  1.49s/it]
                                                         
{'loss': 0.078, 'learning_rate': 5e-05, 'epoch': 3.03}

 61%|██████    | 14780/24420 [6:09:38<3:59:01,  1.49s/it]
 61%|██████    | 14781/24420 [6:09:40<3:59:24,  1.49s/it]
 61%|██████    | 14782/24420 [6:09:41<3:59:52,  1.49s/it]
 61%|██████    | 14783/24420 [6:09:43<3:59:48,  1.49s/it]
 61%|██████    | 14784/24420 [6:09:44<3:59:38,  1.49s/it]
 61%|██████    | 14785/24420 [6:09:46<4:00:07,  1.50s/it]
                                                         
{'loss': 0.1064, 'learning_rate': 5e-05, 'epoch': 3.03}

 61%|██████    | 14785/24420 [6:09:46<4:00:07,  1.50s/it]
 61%|██████    | 14786/24420 [6:09:47<3:59:51,  1.49s/it]
 61%|██████    | 14787/24420 [6:09:49<3:59:52,  1.49s/it]
 61%|██████    | 14788/24420 [6:09:50<3:59:27,  1.49s/it]
 61%|██████    | 14789/24420 [6:09:52<3:59:09,  1.49s/it]
 61%|██████    | 14790/24420 [6:09:53<3:59:32,  1.49s/it]
                                                         
{'loss': 0.0818, 'learning_rate': 5e-05, 'epoch': 3.03}

 61%|██████    | 14790/24420 [6:09:53<3:59:32,  1.49s/it]
 61%|██████    | 14791/24420 [6:09:55<3:59:04,  1.49s/it]
 61%|██████    | 14792/24420 [6:09:56<3:58:22,  1.49s/it]
 61%|██████    | 14793/24420 [6:09:58<3:58:46,  1.49s/it]
 61%|██████    | 14794/24420 [6:09:59<3:58:22,  1.49s/it]
 61%|██████    | 14795/24420 [6:10:01<3:58:22,  1.49s/it]
                                                         
{'loss': 0.0885, 'learning_rate': 5e-05, 'epoch': 3.03}

 61%|██████    | 14795/24420 [6:10:01<3:58:22,  1.49s/it]
 61%|██████    | 14796/24420 [6:10:02<3:58:21,  1.49s/it]
 61%|██████    | 14797/24420 [6:10:04<3:58:31,  1.49s/it]
 61%|██████    | 14798/24420 [6:10:05<3:58:26,  1.49s/it]
 61%|██████    | 14799/24420 [6:10:07<3:58:32,  1.49s/it]
 61%|██████    | 14800/24420 [6:10:08<3:58:23,  1.49s/it]
                                                         
{'loss': 0.0836, 'learning_rate': 5e-05, 'epoch': 3.03}

 61%|██████    | 14800/24420 [6:10:08<3:58:23,  1.49s/it]
 61%|██████    | 14801/24420 [6:10:13<6:43:43,  2.52s/it]
 61%|██████    | 14802/24420 [6:10:15<5:54:11,  2.21s/it]
 61%|██████    | 14803/24420 [6:10:16<5:19:37,  1.99s/it]
 61%|██████    | 14804/24420 [6:10:17<4:55:21,  1.84s/it]
 61%|██████    | 14805/24420 [6:10:19<4:38:18,  1.74s/it]
                                                         
{'loss': 0.0899, 'learning_rate': 5e-05, 'epoch': 3.03}

 61%|██████    | 14805/24420 [6:10:19<4:38:18,  1.74s/it]
 61%|██████    | 14806/24420 [6:10:20<4:26:22,  1.66s/it]
 61%|██████    | 14807/24420 [6:10:22<4:17:25,  1.61s/it]
 61%|██████    | 14808/24420 [6:10:23<4:11:35,  1.57s/it]
 61%|██████    | 14809/24420 [6:10:25<4:07:33,  1.55s/it]
 61%|██████    | 14810/24420 [6:10:26<4:04:35,  1.53s/it]
                                                         
{'loss': 0.0773, 'learning_rate': 5e-05, 'epoch': 3.03}

 61%|██████    | 14810/24420 [6:10:26<4:04:35,  1.53s/it]
 61%|██████    | 14811/24420 [6:10:28<4:02:53,  1.52s/it]
 61%|██████    | 14812/24420 [6:10:29<4:01:16,  1.51s/it]
 61%|██████    | 14813/24420 [6:10:31<4:01:32,  1.51s/it]
 61%|██████    | 14814/24420 [6:10:32<4:01:02,  1.51s/it]
 61%|██████    | 14815/24420 [6:10:34<4:00:05,  1.50s/it]
                                                         
{'loss': 0.0929, 'learning_rate': 5e-05, 'epoch': 3.03}

 61%|██████    | 14815/24420 [6:10:34<4:00:05,  1.50s/it]
 61%|██████    | 14816/24420 [6:10:35<3:59:24,  1.50s/it]
 61%|██████    | 14817/24420 [6:10:37<3:58:50,  1.49s/it]
 61%|██████    | 14818/24420 [6:10:38<3:58:26,  1.49s/it]
 61%|██████    | 14819/24420 [6:10:40<3:58:42,  1.49s/it]
 61%|██████    | 14820/24420 [6:10:41<3:58:24,  1.49s/it]
                                                         
{'loss': 0.0783, 'learning_rate': 5e-05, 'epoch': 3.03}

 61%|██████    | 14820/24420 [6:10:41<3:58:24,  1.49s/it]
 61%|██████    | 14821/24420 [6:10:43<3:58:38,  1.49s/it]
 61%|██████    | 14822/24420 [6:10:44<3:58:09,  1.49s/it]
 61%|██████    | 14823/24420 [6:10:46<3:58:36,  1.49s/it]
 61%|██████    | 14824/24420 [6:10:47<3:58:22,  1.49s/it]
 61%|██████    | 14825/24420 [6:10:49<3:58:12,  1.49s/it]
                                                         
{'loss': 0.0946, 'learning_rate': 5e-05, 'epoch': 3.04}

 61%|██████    | 14825/24420 [6:10:49<3:58:12,  1.49s/it]
 61%|██████    | 14826/24420 [6:10:50<3:58:08,  1.49s/it]
 61%|██████    | 14827/24420 [6:10:52<3:58:20,  1.49s/it]
 61%|██████    | 14828/24420 [6:10:53<3:58:21,  1.49s/it]
 61%|██████    | 14829/24420 [6:10:55<3:58:26,  1.49s/it]
 61%|██████    | 14830/24420 [6:10:56<3:58:07,  1.49s/it]
                                                         
{'loss': 0.0732, 'learning_rate': 5e-05, 'epoch': 3.04}

 61%|██████    | 14830/24420 [6:10:56<3:58:07,  1.49s/it]
 61%|██████    | 14831/24420 [6:10:58<3:58:17,  1.49s/it]
 61%|██████    | 14832/24420 [6:10:59<3:58:17,  1.49s/it]
 61%|██████    | 14833/24420 [6:11:01<3:57:40,  1.49s/it]
 61%|██████    | 14834/24420 [6:11:02<3:57:42,  1.49s/it]
 61%|██████    | 14835/24420 [6:11:04<3:57:42,  1.49s/it]
                                                         
{'loss': 0.0722, 'learning_rate': 5e-05, 'epoch': 3.04}

 61%|██████    | 14835/24420 [6:11:04<3:57:42,  1.49s/it]
 61%|██████    | 14836/24420 [6:11:05<3:58:03,  1.49s/it]
 61%|██████    | 14837/24420 [6:11:07<3:57:55,  1.49s/it]
 61%|██████    | 14838/24420 [6:11:08<3:58:11,  1.49s/it]
 61%|██████    | 14839/24420 [6:11:10<3:58:32,  1.49s/it]
 61%|██████    | 14840/24420 [6:11:11<3:58:01,  1.49s/it]
                                                         
{'loss': 0.0784, 'learning_rate': 5e-05, 'epoch': 3.04}

 61%|██████    | 14840/24420 [6:11:11<3:58:01,  1.49s/it]
 61%|██████    | 14841/24420 [6:11:13<3:58:23,  1.49s/it]
 61%|██████    | 14842/24420 [6:11:14<3:58:06,  1.49s/it]
 61%|██████    | 14843/24420 [6:11:16<3:57:29,  1.49s/it]
 61%|██████    | 14844/24420 [6:11:17<3:57:25,  1.49s/it]
 61%|██████    | 14845/24420 [6:11:19<3:57:37,  1.49s/it]
                                                         
{'loss': 0.0749, 'learning_rate': 5e-05, 'epoch': 3.04}

 61%|██████    | 14845/24420 [6:11:19<3:57:37,  1.49s/it]
 61%|██████    | 14846/24420 [6:11:20<3:57:44,  1.49s/it]
 61%|██████    | 14847/24420 [6:11:22<3:57:45,  1.49s/it]
 61%|██████    | 14848/24420 [6:11:23<3:57:49,  1.49s/it]
 61%|██████    | 14849/24420 [6:11:25<3:57:55,  1.49s/it]
 61%|██████    | 14850/24420 [6:11:26<3:58:13,  1.49s/it]
                                                         
{'loss': 0.0798, 'learning_rate': 5e-05, 'epoch': 3.04}

 61%|██████    | 14850/24420 [6:11:26<3:58:13,  1.49s/it]
 61%|██████    | 14851/24420 [6:11:28<3:57:40,  1.49s/it]
 61%|██████    | 14852/24420 [6:11:29<3:57:49,  1.49s/it]
 61%|██████    | 14853/24420 [6:11:31<3:58:08,  1.49s/it]
 61%|██████    | 14854/24420 [6:11:32<3:58:21,  1.50s/it]
 61%|██████    | 14855/24420 [6:11:33<3:58:23,  1.50s/it]
                                                         
{'loss': 0.0689, 'learning_rate': 5e-05, 'epoch': 3.04}

 61%|██████    | 14855/24420 [6:11:34<3:58:23,  1.50s/it]
 61%|██████    | 14856/24420 [6:11:35<3:58:37,  1.50s/it]
 61%|██████    | 14857/24420 [6:11:36<3:58:11,  1.49s/it]
 61%|██████    | 14858/24420 [6:11:38<3:58:00,  1.49s/it]
 61%|██████    | 14859/24420 [6:11:39<3:58:02,  1.49s/it]
 61%|██████    | 14860/24420 [6:11:41<3:57:30,  1.49s/it]
                                                         
{'loss': 0.0829, 'learning_rate': 5e-05, 'epoch': 3.04}

 61%|██████    | 14860/24420 [6:11:41<3:57:30,  1.49s/it]
 61%|██████    | 14861/24420 [6:11:42<3:58:03,  1.49s/it]
 61%|██████    | 14862/24420 [6:11:44<3:57:25,  1.49s/it]
 61%|██████    | 14863/24420 [6:11:45<3:57:21,  1.49s/it]
 61%|██████    | 14864/24420 [6:11:47<3:58:03,  1.49s/it]
 61%|██████    | 14865/24420 [6:11:48<3:57:49,  1.49s/it]
                                                         
{'loss': 0.1276, 'learning_rate': 5e-05, 'epoch': 3.04}

 61%|██████    | 14865/24420 [6:11:48<3:57:49,  1.49s/it]
 61%|██████    | 14866/24420 [6:11:50<3:57:32,  1.49s/it]
 61%|██████    | 14867/24420 [6:11:51<3:57:22,  1.49s/it]
 61%|██████    | 14868/24420 [6:11:53<3:57:47,  1.49s/it]
 61%|██████    | 14869/24420 [6:11:54<3:57:13,  1.49s/it]
 61%|██████    | 14870/24420 [6:11:56<3:57:30,  1.49s/it]
                                                         
{'loss': 0.077, 'learning_rate': 5e-05, 'epoch': 3.04}

 61%|██████    | 14870/24420 [6:11:56<3:57:30,  1.49s/it]
 61%|██████    | 14871/24420 [6:11:57<3:57:32,  1.49s/it]
 61%|██████    | 14872/24420 [6:11:59<3:57:22,  1.49s/it]
 61%|██████    | 14873/24420 [6:12:00<3:57:13,  1.49s/it]
 61%|██████    | 14874/24420 [6:12:02<3:57:03,  1.49s/it]
 61%|██████    | 14875/24420 [6:12:03<3:57:06,  1.49s/it]
                                                         
{'loss': 0.0711, 'learning_rate': 5e-05, 'epoch': 3.05}

 61%|██████    | 14875/24420 [6:12:03<3:57:06,  1.49s/it]
 61%|██████    | 14876/24420 [6:12:05<3:57:44,  1.49s/it]
 61%|██████    | 14877/24420 [6:12:06<3:57:45,  1.49s/it]
 61%|██████    | 14878/24420 [6:12:08<3:57:01,  1.49s/it]
 61%|██████    | 14879/24420 [6:12:09<3:57:04,  1.49s/it]
 61%|██████    | 14880/24420 [6:12:11<3:57:04,  1.49s/it]
                                                         
{'loss': 0.087, 'learning_rate': 5e-05, 'epoch': 3.05}

 61%|██████    | 14880/24420 [6:12:11<3:57:04,  1.49s/it]
 61%|██████    | 14881/24420 [6:12:12<3:56:59,  1.49s/it]
 61%|██████    | 14882/24420 [6:12:14<3:57:05,  1.49s/it]
 61%|██████    | 14883/24420 [6:12:15<3:57:21,  1.49s/it]
 61%|██████    | 14884/24420 [6:12:17<3:57:15,  1.49s/it]
 61%|██████    | 14885/24420 [6:12:18<3:56:56,  1.49s/it]
                                                         
{'loss': 0.0881, 'learning_rate': 5e-05, 'epoch': 3.05}

 61%|██████    | 14885/24420 [6:12:18<3:56:56,  1.49s/it]
 61%|██████    | 14886/24420 [6:12:20<3:57:06,  1.49s/it]
 61%|██████    | 14887/24420 [6:12:21<3:58:01,  1.50s/it]
 61%|██████    | 14888/24420 [6:12:23<3:57:34,  1.50s/it]
 61%|██████    | 14889/24420 [6:12:24<3:57:10,  1.49s/it]
 61%|██████    | 14890/24420 [6:12:26<3:57:34,  1.50s/it]
                                                         
{'loss': 0.1026, 'learning_rate': 5e-05, 'epoch': 3.05}

 61%|██████    | 14890/24420 [6:12:26<3:57:34,  1.50s/it]
 61%|██████    | 14891/24420 [6:12:27<3:57:05,  1.49s/it]
 61%|██████    | 14892/24420 [6:12:29<3:56:44,  1.49s/it]
 61%|██████    | 14893/24420 [6:12:30<3:56:54,  1.49s/it]
 61%|██████    | 14894/24420 [6:12:32<3:57:57,  1.50s/it]
 61%|██████    | 14895/24420 [6:12:33<3:57:58,  1.50s/it]
                                                         
{'loss': 0.0706, 'learning_rate': 5e-05, 'epoch': 3.05}

 61%|██████    | 14895/24420 [6:12:33<3:57:58,  1.50s/it]
 61%|██████    | 14896/24420 [6:12:35<3:57:49,  1.50s/it]
 61%|██████    | 14897/24420 [6:12:36<3:57:31,  1.50s/it]
 61%|██████    | 14898/24420 [6:12:38<3:57:02,  1.49s/it]
 61%|██████    | 14899/24420 [6:12:39<3:57:02,  1.49s/it]
 61%|██████    | 14900/24420 [6:12:41<3:56:41,  1.49s/it]
                                                         
{'loss': 0.0773, 'learning_rate': 5e-05, 'epoch': 3.05}

 61%|██████    | 14900/24420 [6:12:41<3:56:41,  1.49s/it]
 61%|██████    | 14901/24420 [6:12:42<3:56:58,  1.49s/it]
 61%|██████    | 14902/24420 [6:12:44<3:57:09,  1.49s/it]
 61%|██████    | 14903/24420 [6:12:45<3:57:27,  1.50s/it]
 61%|██████    | 14904/24420 [6:12:47<3:57:24,  1.50s/it]
 61%|██████    | 14905/24420 [6:12:48<3:57:15,  1.50s/it]
                                                         
{'loss': 0.074, 'learning_rate': 5e-05, 'epoch': 3.05}

 61%|██████    | 14905/24420 [6:12:48<3:57:15,  1.50s/it]
 61%|██████    | 14906/24420 [6:12:50<3:57:10,  1.50s/it]
 61%|██████    | 14907/24420 [6:12:51<3:56:34,  1.49s/it]
 61%|██████    | 14908/24420 [6:12:53<3:56:05,  1.49s/it]
 61%|██████    | 14909/24420 [6:12:54<3:55:54,  1.49s/it]
 61%|██████    | 14910/24420 [6:12:56<3:56:00,  1.49s/it]
                                                         
{'loss': 0.0735, 'learning_rate': 5e-05, 'epoch': 3.05}

 61%|██████    | 14910/24420 [6:12:56<3:56:00,  1.49s/it]
 61%|██████    | 14911/24420 [6:12:57<3:56:07,  1.49s/it]
 61%|██████    | 14912/24420 [6:12:59<3:56:06,  1.49s/it]
 61%|██████    | 14913/24420 [6:13:00<3:56:41,  1.49s/it]
 61%|██████    | 14914/24420 [6:13:02<3:56:04,  1.49s/it]
 61%|██████    | 14915/24420 [6:13:03<3:56:14,  1.49s/it]
                                                         
{'loss': 0.0945, 'learning_rate': 5e-05, 'epoch': 3.05}

 61%|██████    | 14915/24420 [6:13:03<3:56:14,  1.49s/it]
 61%|██████    | 14916/24420 [6:13:05<3:55:58,  1.49s/it]
 61%|██████    | 14917/24420 [6:13:06<3:55:47,  1.49s/it]
 61%|██████    | 14918/24420 [6:13:08<3:55:46,  1.49s/it]
 61%|██████    | 14919/24420 [6:13:09<3:55:58,  1.49s/it]
 61%|██████    | 14920/24420 [6:13:11<3:56:20,  1.49s/it]
                                                         
{'loss': 0.0792, 'learning_rate': 5e-05, 'epoch': 3.05}

 61%|██████    | 14920/24420 [6:13:11<3:56:20,  1.49s/it]
 61%|██████    | 14921/24420 [6:13:12<3:56:37,  1.49s/it]
 61%|██████    | 14922/24420 [6:13:14<3:56:41,  1.50s/it]
 61%|██████    | 14923/24420 [6:13:15<3:56:33,  1.49s/it]
 61%|██████    | 14924/24420 [6:13:16<3:56:13,  1.49s/it]
 61%|██████    | 14925/24420 [6:13:18<3:55:58,  1.49s/it]
                                                         
{'loss': 0.091, 'learning_rate': 5e-05, 'epoch': 3.06}

 61%|██████    | 14925/24420 [6:13:18<3:55:58,  1.49s/it]
 61%|██████    | 14926/24420 [6:13:19<3:55:51,  1.49s/it]
 61%|██████    | 14927/24420 [6:13:21<3:55:54,  1.49s/it]
 61%|██████    | 14928/24420 [6:13:22<3:55:43,  1.49s/it]
 61%|██████    | 14929/24420 [6:13:24<3:55:48,  1.49s/it]
 61%|██████    | 14930/24420 [6:13:25<3:55:25,  1.49s/it]
                                                         
{'loss': 0.0827, 'learning_rate': 5e-05, 'epoch': 3.06}

 61%|██████    | 14930/24420 [6:13:25<3:55:25,  1.49s/it]
 61%|██████    | 14931/24420 [6:13:27<3:55:49,  1.49s/it]
 61%|██████    | 14932/24420 [6:13:28<3:55:55,  1.49s/it]
 61%|██████    | 14933/24420 [6:13:30<3:55:52,  1.49s/it]
 61%|██████    | 14934/24420 [6:13:31<3:55:35,  1.49s/it]
 61%|██████    | 14935/24420 [6:13:33<3:55:22,  1.49s/it]
                                                         
{'loss': 0.0782, 'learning_rate': 5e-05, 'epoch': 3.06}

 61%|██████    | 14935/24420 [6:13:33<3:55:22,  1.49s/it]
 61%|██████    | 14936/24420 [6:13:34<3:55:26,  1.49s/it]
 61%|██████    | 14937/24420 [6:13:36<3:55:29,  1.49s/it]
 61%|██████    | 14938/24420 [6:13:37<3:55:59,  1.49s/it]
 61%|██████    | 14939/24420 [6:13:39<3:56:08,  1.49s/it]
 61%|██████    | 14940/24420 [6:13:40<3:55:56,  1.49s/it]
                                                         
{'loss': 0.0719, 'learning_rate': 5e-05, 'epoch': 3.06}

 61%|██████    | 14940/24420 [6:13:40<3:55:56,  1.49s/it]
 61%|██████    | 14941/24420 [6:13:42<3:55:33,  1.49s/it]
 61%|██████    | 14942/24420 [6:13:43<3:55:33,  1.49s/it]
 61%|██████    | 14943/24420 [6:13:45<3:55:03,  1.49s/it]
 61%|██████    | 14944/24420 [6:13:46<3:54:57,  1.49s/it]
 61%|██████    | 14945/24420 [6:13:48<3:54:59,  1.49s/it]
                                                         
{'loss': 0.0716, 'learning_rate': 5e-05, 'epoch': 3.06}

 61%|██████    | 14945/24420 [6:13:48<3:54:59,  1.49s/it]
 61%|██████    | 14946/24420 [6:13:49<3:55:26,  1.49s/it]
 61%|██████    | 14947/24420 [6:13:51<3:55:26,  1.49s/it]
 61%|██████    | 14948/24420 [6:13:52<3:56:17,  1.50s/it]
 61%|██████    | 14949/24420 [6:13:54<3:56:02,  1.50s/it]
 61%|██████    | 14950/24420 [6:13:55<3:55:05,  1.49s/it]
                                                         
{'loss': 0.0852, 'learning_rate': 5e-05, 'epoch': 3.06}

 61%|██████    | 14950/24420 [6:13:55<3:55:05,  1.49s/it]
 61%|██████    | 14951/24420 [6:13:57<3:55:01,  1.49s/it]
 61%|██████    | 14952/24420 [6:13:58<3:54:51,  1.49s/it]
 61%|██████    | 14953/24420 [6:14:00<3:55:03,  1.49s/it]
 61%|██████    | 14954/24420 [6:14:01<3:55:31,  1.49s/it]
 61%|██████    | 14955/24420 [6:14:03<3:55:30,  1.49s/it]
                                                         
{'loss': 0.0751, 'learning_rate': 5e-05, 'epoch': 3.06}

 61%|██████    | 14955/24420 [6:14:03<3:55:30,  1.49s/it]
 61%|██████    | 14956/24420 [6:14:04<3:55:31,  1.49s/it]
 61%|██████    | 14957/24420 [6:14:06<3:55:24,  1.49s/it]
 61%|██████▏   | 14958/24420 [6:14:07<3:55:23,  1.49s/it]
 61%|██████▏   | 14959/24420 [6:14:09<3:54:57,  1.49s/it]
 61%|██████▏   | 14960/24420 [6:14:10<3:54:43,  1.49s/it]
                                                         
{'loss': 0.0803, 'learning_rate': 5e-05, 'epoch': 3.06}

 61%|██████▏   | 14960/24420 [6:14:10<3:54:43,  1.49s/it]
 61%|██████▏   | 14961/24420 [6:14:12<3:55:15,  1.49s/it]
 61%|██████▏   | 14962/24420 [6:14:13<3:55:05,  1.49s/it]
 61%|██████▏   | 14963/24420 [6:14:15<3:55:10,  1.49s/it]
 61%|██████▏   | 14964/24420 [6:14:16<3:55:10,  1.49s/it]
 61%|██████▏   | 14965/24420 [6:14:18<3:54:40,  1.49s/it]
                                                         
{'loss': 0.0812, 'learning_rate': 5e-05, 'epoch': 3.06}

 61%|██████▏   | 14965/24420 [6:14:18<3:54:40,  1.49s/it]
 61%|██████▏   | 14966/24420 [6:14:19<3:54:40,  1.49s/it]
 61%|██████▏   | 14967/24420 [6:14:21<3:54:48,  1.49s/it]
 61%|██████▏   | 14968/24420 [6:14:22<3:54:23,  1.49s/it]
 61%|██████▏   | 14969/24420 [6:14:24<3:53:48,  1.48s/it]
 61%|██████▏   | 14970/24420 [6:14:25<3:54:06,  1.49s/it]
                                                         
{'loss': 0.0798, 'learning_rate': 5e-05, 'epoch': 3.07}

 61%|██████▏   | 14970/24420 [6:14:25<3:54:06,  1.49s/it]
 61%|██████▏   | 14971/24420 [6:14:27<3:53:58,  1.49s/it]
 61%|██████▏   | 14972/24420 [6:14:28<3:54:01,  1.49s/it]
 61%|██████▏   | 14973/24420 [6:14:30<3:54:18,  1.49s/it]
 61%|██████▏   | 14974/24420 [6:14:31<3:54:31,  1.49s/it]
 61%|██████▏   | 14975/24420 [6:14:33<3:54:25,  1.49s/it]
                                                         
{'loss': 0.0951, 'learning_rate': 5e-05, 'epoch': 3.07}

 61%|██████▏   | 14975/24420 [6:14:33<3:54:25,  1.49s/it]
 61%|██████▏   | 14976/24420 [6:14:34<3:54:16,  1.49s/it]
 61%|██████▏   | 14977/24420 [6:14:35<3:54:28,  1.49s/it]
 61%|██████▏   | 14978/24420 [6:14:37<3:54:17,  1.49s/it]
 61%|██████▏   | 14979/24420 [6:14:38<3:54:16,  1.49s/it]
 61%|██████▏   | 14980/24420 [6:14:40<3:54:15,  1.49s/it]
                                                         
{'loss': 0.0827, 'learning_rate': 5e-05, 'epoch': 3.07}

 61%|██████▏   | 14980/24420 [6:14:40<3:54:15,  1.49s/it]
 61%|██████▏   | 14981/24420 [6:14:41<3:54:33,  1.49s/it]
 61%|██████▏   | 14982/24420 [6:14:43<3:54:14,  1.49s/it]
 61%|██████▏   | 14983/24420 [6:14:44<3:53:58,  1.49s/it]
 61%|██████▏   | 14984/24420 [6:14:46<3:53:46,  1.49s/it]
 61%|██████▏   | 14985/24420 [6:14:47<3:54:12,  1.49s/it]
                                                         
{'loss': 0.0729, 'learning_rate': 5e-05, 'epoch': 3.07}

 61%|██████▏   | 14985/24420 [6:14:47<3:54:12,  1.49s/it]
 61%|██████▏   | 14986/24420 [6:14:49<3:54:19,  1.49s/it]
 61%|██████▏   | 14987/24420 [6:14:50<3:54:04,  1.49s/it]
 61%|██████▏   | 14988/24420 [6:14:52<3:54:02,  1.49s/it]
 61%|██████▏   | 14989/24420 [6:14:53<3:53:51,  1.49s/it]
 61%|██████▏   | 14990/24420 [6:14:55<3:53:52,  1.49s/it]
                                                         
{'loss': 0.0843, 'learning_rate': 5e-05, 'epoch': 3.07}

 61%|██████▏   | 14990/24420 [6:14:55<3:53:52,  1.49s/it]
 61%|██████▏   | 14991/24420 [6:14:56<3:55:05,  1.50s/it]
 61%|██████▏   | 14992/24420 [6:14:58<3:55:05,  1.50s/it]
 61%|██████▏   | 14993/24420 [6:14:59<3:54:31,  1.49s/it]
 61%|██████▏   | 14994/24420 [6:15:01<3:54:28,  1.49s/it]
 61%|██████▏   | 14995/24420 [6:15:02<3:54:25,  1.49s/it]
                                                         
{'loss': 0.0776, 'learning_rate': 5e-05, 'epoch': 3.07}

 61%|██████▏   | 14995/24420 [6:15:02<3:54:25,  1.49s/it]
 61%|██████▏   | 14996/24420 [6:15:04<3:54:23,  1.49s/it]
 61%|██████▏   | 14997/24420 [6:15:05<3:53:59,  1.49s/it]
 61%|██████▏   | 14998/24420 [6:15:07<3:54:10,  1.49s/it]
 61%|██████▏   | 14999/24420 [6:15:08<3:53:52,  1.49s/it]
 61%|██████▏   | 15000/24420 [6:15:10<3:53:58,  1.49s/it]
                                                         
{'loss': 0.0722, 'learning_rate': 5e-05, 'epoch': 3.07}

 61%|██████▏   | 15000/24420 [6:15:10<3:53:58,  1.49s/it]
 61%|██████▏   | 15001/24420 [6:15:14<6:26:01,  2.46s/it]
 61%|██████▏   | 15002/24420 [6:15:16<5:40:33,  2.17s/it]
 61%|██████▏   | 15003/24420 [6:15:17<5:08:36,  1.97s/it]
 61%|██████▏   | 15004/24420 [6:15:19<4:45:57,  1.82s/it]
 61%|██████▏   | 15005/24420 [6:15:20<4:30:09,  1.72s/it]
                                                         
{'loss': 0.0967, 'learning_rate': 5e-05, 'epoch': 3.07}

 61%|██████▏   | 15005/24420 [6:15:20<4:30:09,  1.72s/it]
 61%|██████▏   | 15006/24420 [6:15:22<4:19:12,  1.65s/it]
 61%|██████▏   | 15007/24420 [6:15:23<4:11:56,  1.61s/it]
 61%|██████▏   | 15008/24420 [6:15:25<4:06:57,  1.57s/it]
 61%|██████▏   | 15009/24420 [6:15:26<4:03:13,  1.55s/it]
 61%|██████▏   | 15010/24420 [6:15:28<4:00:21,  1.53s/it]
                                                         
{'loss': 0.0804, 'learning_rate': 5e-05, 'epoch': 3.07}

 61%|██████▏   | 15010/24420 [6:15:28<4:00:21,  1.53s/it]
 61%|██████▏   | 15011/24420 [6:15:29<3:58:30,  1.52s/it]
 61%|██████▏   | 15012/24420 [6:15:31<3:56:59,  1.51s/it]
 61%|██████▏   | 15013/24420 [6:15:32<3:55:58,  1.51s/it]
 61%|██████▏   | 15014/24420 [6:15:34<3:55:09,  1.50s/it]
 61%|██████▏   | 15015/24420 [6:15:35<3:54:33,  1.50s/it]
                                                         
{'loss': 0.0749, 'learning_rate': 5e-05, 'epoch': 3.07}

 61%|██████▏   | 15015/24420 [6:15:35<3:54:33,  1.50s/it]
 61%|██████▏   | 15016/24420 [6:15:37<3:55:38,  1.50s/it]
 61%|██████▏   | 15017/24420 [6:15:38<3:54:34,  1.50s/it]
 61%|██████▏   | 15018/24420 [6:15:40<3:54:19,  1.50s/it]
 62%|██████▏   | 15019/24420 [6:15:41<3:53:57,  1.49s/it]
 62%|██████▏   | 15020/24420 [6:15:43<3:53:10,  1.49s/it]
                                                         
{'loss': 0.0793, 'learning_rate': 5e-05, 'epoch': 3.08}

 62%|██████▏   | 15020/24420 [6:15:43<3:53:10,  1.49s/it]
 62%|██████▏   | 15021/24420 [6:15:44<3:53:11,  1.49s/it]
 62%|██████▏   | 15022/24420 [6:15:46<3:53:41,  1.49s/it]
 62%|██████▏   | 15023/24420 [6:15:47<3:53:38,  1.49s/it]
 62%|██████▏   | 15024/24420 [6:15:49<3:53:30,  1.49s/it]
 62%|██████▏   | 15025/24420 [6:15:50<3:54:25,  1.50s/it]
                                                         
{'loss': 0.0809, 'learning_rate': 5e-05, 'epoch': 3.08}

 62%|██████▏   | 15025/24420 [6:15:50<3:54:25,  1.50s/it]
 62%|██████▏   | 15026/24420 [6:15:52<3:53:49,  1.49s/it]
 62%|██████▏   | 15027/24420 [6:15:53<3:53:38,  1.49s/it]
 62%|██████▏   | 15028/24420 [6:15:55<3:53:58,  1.49s/it]
 62%|██████▏   | 15029/24420 [6:15:56<3:53:59,  1.49s/it]
 62%|██████▏   | 15030/24420 [6:15:58<3:53:28,  1.49s/it]
                                                         
{'loss': 0.0809, 'learning_rate': 5e-05, 'epoch': 3.08}

 62%|██████▏   | 15030/24420 [6:15:58<3:53:28,  1.49s/it]
 62%|██████▏   | 15031/24420 [6:15:59<3:53:47,  1.49s/it]
 62%|██████▏   | 15032/24420 [6:16:01<3:54:31,  1.50s/it]
 62%|██████▏   | 15033/24420 [6:16:02<3:55:18,  1.50s/it]
 62%|██████▏   | 15034/24420 [6:16:04<3:54:47,  1.50s/it]
 62%|██████▏   | 15035/24420 [6:16:05<3:54:28,  1.50s/it]
                                                         
{'loss': 0.091, 'learning_rate': 5e-05, 'epoch': 3.08}

 62%|██████▏   | 15035/24420 [6:16:05<3:54:28,  1.50s/it]
 62%|██████▏   | 15036/24420 [6:16:07<3:54:29,  1.50s/it]
 62%|██████▏   | 15037/24420 [6:16:08<3:54:03,  1.50s/it]
 62%|██████▏   | 15038/24420 [6:16:10<3:53:43,  1.49s/it]
 62%|██████▏   | 15039/24420 [6:16:11<3:53:29,  1.49s/it]
 62%|██████▏   | 15040/24420 [6:16:13<3:53:34,  1.49s/it]
                                                         
{'loss': 0.0614, 'learning_rate': 5e-05, 'epoch': 3.08}

 62%|██████▏   | 15040/24420 [6:16:13<3:53:34,  1.49s/it]
 62%|██████▏   | 15041/24420 [6:16:14<3:53:31,  1.49s/it]
 62%|██████▏   | 15042/24420 [6:16:16<3:53:50,  1.50s/it]
 62%|██████▏   | 15043/24420 [6:16:17<3:53:56,  1.50s/it]
 62%|██████▏   | 15044/24420 [6:16:19<3:54:19,  1.50s/it]
 62%|██████▏   | 15045/24420 [6:16:20<3:53:51,  1.50s/it]
                                                         
{'loss': 0.0832, 'learning_rate': 5e-05, 'epoch': 3.08}

 62%|██████▏   | 15045/24420 [6:16:20<3:53:51,  1.50s/it]
 62%|██████▏   | 15046/24420 [6:16:22<3:53:35,  1.50s/it]
 62%|██████▏   | 15047/24420 [6:16:23<3:53:36,  1.50s/it]
 62%|██████▏   | 15048/24420 [6:16:25<3:53:11,  1.49s/it]
 62%|██████▏   | 15049/24420 [6:16:26<3:53:02,  1.49s/it]
 62%|██████▏   | 15050/24420 [6:16:28<3:53:11,  1.49s/it]
                                                         
{'loss': 0.0919, 'learning_rate': 5e-05, 'epoch': 3.08}

 62%|██████▏   | 15050/24420 [6:16:28<3:53:11,  1.49s/it]
 62%|██████▏   | 15051/24420 [6:16:29<3:52:54,  1.49s/it]
 62%|██████▏   | 15052/24420 [6:16:31<3:53:17,  1.49s/it]
 62%|██████▏   | 15053/24420 [6:16:32<3:53:18,  1.49s/it]
 62%|██████▏   | 15054/24420 [6:16:34<3:53:16,  1.49s/it]
 62%|██████▏   | 15055/24420 [6:16:35<3:52:44,  1.49s/it]
                                                         
{'loss': 0.082, 'learning_rate': 5e-05, 'epoch': 3.08}

 62%|██████▏   | 15055/24420 [6:16:35<3:52:44,  1.49s/it]
 62%|██████▏   | 15056/24420 [6:16:37<3:52:59,  1.49s/it]
 62%|██████▏   | 15057/24420 [6:16:38<3:52:34,  1.49s/it]
 62%|██████▏   | 15058/24420 [6:16:40<3:52:31,  1.49s/it]
 62%|██████▏   | 15059/24420 [6:16:41<3:52:22,  1.49s/it]
 62%|██████▏   | 15060/24420 [6:16:43<3:52:39,  1.49s/it]
                                                         
{'loss': 0.0808, 'learning_rate': 5e-05, 'epoch': 3.08}

 62%|██████▏   | 15060/24420 [6:16:43<3:52:39,  1.49s/it]
 62%|██████▏   | 15061/24420 [6:16:44<3:53:09,  1.49s/it]
 62%|██████▏   | 15062/24420 [6:16:46<3:52:35,  1.49s/it]
 62%|██████▏   | 15063/24420 [6:16:47<3:52:41,  1.49s/it]
 62%|██████▏   | 15064/24420 [6:16:49<3:52:44,  1.49s/it]
 62%|██████▏   | 15065/24420 [6:16:50<3:52:49,  1.49s/it]
                                                         
{'loss': 0.0908, 'learning_rate': 5e-05, 'epoch': 3.08}

 62%|██████▏   | 15065/24420 [6:16:50<3:52:49,  1.49s/it]
 62%|██████▏   | 15066/24420 [6:16:52<3:52:54,  1.49s/it]
 62%|██████▏   | 15067/24420 [6:16:53<3:52:47,  1.49s/it]
 62%|██████▏   | 15068/24420 [6:16:55<3:53:18,  1.50s/it]
 62%|██████▏   | 15069/24420 [6:16:56<3:53:09,  1.50s/it]
 62%|██████▏   | 15070/24420 [6:16:58<3:52:39,  1.49s/it]
                                                         
{'loss': 0.0814, 'learning_rate': 5e-05, 'epoch': 3.09}

 62%|██████▏   | 15070/24420 [6:16:58<3:52:39,  1.49s/it]
 62%|██████▏   | 15071/24420 [6:16:59<3:52:48,  1.49s/it]
 62%|██████▏   | 15072/24420 [6:17:01<3:52:27,  1.49s/it]
 62%|██████▏   | 15073/24420 [6:17:02<3:52:15,  1.49s/it]
 62%|██████▏   | 15074/24420 [6:17:04<3:52:19,  1.49s/it]
 62%|██████▏   | 15075/24420 [6:17:05<3:52:35,  1.49s/it]
                                                         
{'loss': 0.0759, 'learning_rate': 5e-05, 'epoch': 3.09}

 62%|██████▏   | 15075/24420 [6:17:05<3:52:35,  1.49s/it]
 62%|██████▏   | 15076/24420 [6:17:06<3:52:32,  1.49s/it]
 62%|██████▏   | 15077/24420 [6:17:08<3:52:28,  1.49s/it]
 62%|██████▏   | 15078/24420 [6:17:09<3:52:20,  1.49s/it]
 62%|██████▏   | 15079/24420 [6:17:11<3:52:19,  1.49s/it]
 62%|██████▏   | 15080/24420 [6:17:12<3:52:07,  1.49s/it]
                                                         
{'loss': 0.0881, 'learning_rate': 5e-05, 'epoch': 3.09}

 62%|██████▏   | 15080/24420 [6:17:12<3:52:07,  1.49s/it]
 62%|██████▏   | 15081/24420 [6:17:14<3:51:54,  1.49s/it]
 62%|██████▏   | 15082/24420 [6:17:15<3:51:38,  1.49s/it]
 62%|██████▏   | 15083/24420 [6:17:17<3:51:28,  1.49s/it]
 62%|██████▏   | 15084/24420 [6:17:18<3:52:00,  1.49s/it]
 62%|██████▏   | 15085/24420 [6:17:20<3:51:55,  1.49s/it]
                                                         
{'loss': 0.0774, 'learning_rate': 5e-05, 'epoch': 3.09}

 62%|██████▏   | 15085/24420 [6:17:20<3:51:55,  1.49s/it]
 62%|██████▏   | 15086/24420 [6:17:21<3:51:58,  1.49s/it]
 62%|██████▏   | 15087/24420 [6:17:23<3:52:03,  1.49s/it]
 62%|██████▏   | 15088/24420 [6:17:24<3:51:35,  1.49s/it]
 62%|██████▏   | 15089/24420 [6:17:26<3:51:31,  1.49s/it]
 62%|██████▏   | 15090/24420 [6:17:27<3:51:45,  1.49s/it]
                                                         
{'loss': 0.0881, 'learning_rate': 5e-05, 'epoch': 3.09}

 62%|██████▏   | 15090/24420 [6:17:27<3:51:45,  1.49s/it]
 62%|██████▏   | 15091/24420 [6:17:29<3:51:35,  1.49s/it]
 62%|██████▏   | 15092/24420 [6:17:30<3:51:34,  1.49s/it]
 62%|██████▏   | 15093/24420 [6:17:32<3:51:15,  1.49s/it]
 62%|██████▏   | 15094/24420 [6:17:33<3:51:14,  1.49s/it]
 62%|██████▏   | 15095/24420 [6:17:35<3:51:16,  1.49s/it]
                                                         
{'loss': 0.0835, 'learning_rate': 5e-05, 'epoch': 3.09}

 62%|██████▏   | 15095/24420 [6:17:35<3:51:16,  1.49s/it]
 62%|██████▏   | 15096/24420 [6:17:36<3:50:57,  1.49s/it]
 62%|██████▏   | 15097/24420 [6:17:38<3:51:10,  1.49s/it]
 62%|██████▏   | 15098/24420 [6:17:39<3:51:14,  1.49s/it]
 62%|██████▏   | 15099/24420 [6:17:41<3:51:15,  1.49s/it]
 62%|██████▏   | 15100/24420 [6:17:42<3:50:54,  1.49s/it]
                                                         
{'loss': 0.0688, 'learning_rate': 5e-05, 'epoch': 3.09}

 62%|██████▏   | 15100/24420 [6:17:42<3:50:54,  1.49s/it]
 62%|██████▏   | 15101/24420 [6:17:44<3:51:22,  1.49s/it]
 62%|██████▏   | 15102/24420 [6:17:45<3:51:18,  1.49s/it]
 62%|██████▏   | 15103/24420 [6:17:47<3:51:26,  1.49s/it]
 62%|██████▏   | 15104/24420 [6:17:48<3:50:53,  1.49s/it]
 62%|██████▏   | 15105/24420 [6:17:50<3:51:14,  1.49s/it]
                                                         
{'loss': 0.0916, 'learning_rate': 5e-05, 'epoch': 3.09}

 62%|██████▏   | 15105/24420 [6:17:50<3:51:14,  1.49s/it]
 62%|██████▏   | 15106/24420 [6:17:51<3:51:31,  1.49s/it]
 62%|██████▏   | 15107/24420 [6:17:53<3:51:56,  1.49s/it]
 62%|██████▏   | 15108/24420 [6:17:54<3:51:56,  1.49s/it]
 62%|██████▏   | 15109/24420 [6:17:56<3:51:38,  1.49s/it]
 62%|██████▏   | 15110/24420 [6:17:57<3:51:13,  1.49s/it]
                                                         
{'loss': 0.0759, 'learning_rate': 5e-05, 'epoch': 3.09}

 62%|██████▏   | 15110/24420 [6:17:57<3:51:13,  1.49s/it]
 62%|██████▏   | 15111/24420 [6:17:59<3:51:08,  1.49s/it]
 62%|██████▏   | 15112/24420 [6:18:00<3:51:01,  1.49s/it]
 62%|██████▏   | 15113/24420 [6:18:02<3:50:31,  1.49s/it]
 62%|██████▏   | 15114/24420 [6:18:03<3:50:25,  1.49s/it]
 62%|██████▏   | 15115/24420 [6:18:05<3:50:31,  1.49s/it]
                                                         
{'loss': 0.0806, 'learning_rate': 5e-05, 'epoch': 3.09}

 62%|██████▏   | 15115/24420 [6:18:05<3:50:31,  1.49s/it]
 62%|██████▏   | 15116/24420 [6:18:06<3:50:28,  1.49s/it]
 62%|██████▏   | 15117/24420 [6:18:08<3:50:36,  1.49s/it]
 62%|██████▏   | 15118/24420 [6:18:09<3:50:54,  1.49s/it]
 62%|██████▏   | 15119/24420 [6:18:11<3:50:11,  1.48s/it]
 62%|██████▏   | 15120/24420 [6:18:12<3:50:02,  1.48s/it]
                                                         
{'loss': 0.086, 'learning_rate': 5e-05, 'epoch': 3.1}

 62%|██████▏   | 15120/24420 [6:18:12<3:50:02,  1.48s/it]
 62%|██████▏   | 15121/24420 [6:18:13<3:50:09,  1.49s/it]
 62%|██████▏   | 15122/24420 [6:18:15<3:50:21,  1.49s/it]
 62%|██████▏   | 15123/24420 [6:18:16<3:50:21,  1.49s/it]
 62%|██████▏   | 15124/24420 [6:18:18<3:49:51,  1.48s/it]
 62%|██████▏   | 15125/24420 [6:18:19<3:49:48,  1.48s/it]
                                                         
{'loss': 0.0809, 'learning_rate': 5e-05, 'epoch': 3.1}

 62%|██████▏   | 15125/24420 [6:18:19<3:49:48,  1.48s/it]
 62%|██████▏   | 15126/24420 [6:18:21<3:49:48,  1.48s/it]
 62%|██████▏   | 15127/24420 [6:18:22<3:49:51,  1.48s/it]
 62%|██████▏   | 15128/24420 [6:18:24<3:50:15,  1.49s/it]
 62%|██████▏   | 15129/24420 [6:18:25<3:50:32,  1.49s/it]
 62%|██████▏   | 15130/24420 [6:18:27<3:50:42,  1.49s/it]
                                                         
{'loss': 0.0909, 'learning_rate': 5e-05, 'epoch': 3.1}

 62%|██████▏   | 15130/24420 [6:18:27<3:50:42,  1.49s/it]
 62%|██████▏   | 15131/24420 [6:18:28<3:50:36,  1.49s/it]
 62%|██████▏   | 15132/24420 [6:18:30<3:51:17,  1.49s/it]
 62%|██████▏   | 15133/24420 [6:18:31<3:51:34,  1.50s/it]
 62%|██████▏   | 15134/24420 [6:18:33<3:51:32,  1.50s/it]
 62%|██████▏   | 15135/24420 [6:18:34<3:51:31,  1.50s/it]
                                                         
{'loss': 0.0685, 'learning_rate': 5e-05, 'epoch': 3.1}

 62%|██████▏   | 15135/24420 [6:18:34<3:51:31,  1.50s/it]
 62%|██████▏   | 15136/24420 [6:18:36<3:50:48,  1.49s/it]
 62%|██████▏   | 15137/24420 [6:18:37<3:51:05,  1.49s/it]
 62%|██████▏   | 15138/24420 [6:18:39<3:50:55,  1.49s/it]
 62%|██████▏   | 15139/24420 [6:18:40<3:51:05,  1.49s/it]
 62%|██████▏   | 15140/24420 [6:18:42<3:50:52,  1.49s/it]
                                                         
{'loss': 0.0779, 'learning_rate': 5e-05, 'epoch': 3.1}

 62%|██████▏   | 15140/24420 [6:18:42<3:50:52,  1.49s/it]
 62%|██████▏   | 15141/24420 [6:18:43<3:50:58,  1.49s/it]
 62%|██████▏   | 15142/24420 [6:18:45<3:50:44,  1.49s/it]
 62%|██████▏   | 15143/24420 [6:18:46<3:51:30,  1.50s/it]
 62%|██████▏   | 15144/24420 [6:18:48<3:51:07,  1.49s/it]
 62%|██████▏   | 15145/24420 [6:18:49<3:50:46,  1.49s/it]
                                                         
{'loss': 0.0827, 'learning_rate': 5e-05, 'epoch': 3.1}

 62%|██████▏   | 15145/24420 [6:18:49<3:50:46,  1.49s/it]
 62%|██████▏   | 15146/24420 [6:18:51<3:50:50,  1.49s/it]
 62%|██████▏   | 15147/24420 [6:18:52<3:50:31,  1.49s/it]
 62%|██████▏   | 15148/24420 [6:18:54<3:50:49,  1.49s/it]
 62%|██████▏   | 15149/24420 [6:18:55<3:50:38,  1.49s/it]
 62%|██████▏   | 15150/24420 [6:18:57<3:50:17,  1.49s/it]
                                                         
{'loss': 0.0846, 'learning_rate': 5e-05, 'epoch': 3.1}

 62%|██████▏   | 15150/24420 [6:18:57<3:50:17,  1.49s/it]
 62%|██████▏   | 15151/24420 [6:18:58<3:50:26,  1.49s/it]
 62%|██████▏   | 15152/24420 [6:19:00<3:50:35,  1.49s/it]
 62%|██████▏   | 15153/24420 [6:19:01<3:50:30,  1.49s/it]
 62%|██████▏   | 15154/24420 [6:19:03<3:50:09,  1.49s/it]
 62%|██████▏   | 15155/24420 [6:19:04<3:49:59,  1.49s/it]
                                                         
{'loss': 0.0897, 'learning_rate': 5e-05, 'epoch': 3.1}

 62%|██████▏   | 15155/24420 [6:19:04<3:49:59,  1.49s/it]
 62%|██████▏   | 15156/24420 [6:19:06<3:49:58,  1.49s/it]
 62%|██████▏   | 15157/24420 [6:19:07<3:51:14,  1.50s/it]
 62%|██████▏   | 15158/24420 [6:19:09<3:51:07,  1.50s/it]
 62%|██████▏   | 15159/24420 [6:19:10<3:51:00,  1.50s/it]
 62%|██████▏   | 15160/24420 [6:19:12<3:50:38,  1.49s/it]
                                                         
{'loss': 0.0834, 'learning_rate': 5e-05, 'epoch': 3.1}

 62%|██████▏   | 15160/24420 [6:19:12<3:50:38,  1.49s/it]
 62%|██████▏   | 15161/24420 [6:19:13<3:50:10,  1.49s/it]
 62%|██████▏   | 15162/24420 [6:19:15<3:50:28,  1.49s/it]
 62%|██████▏   | 15163/24420 [6:19:16<3:50:23,  1.49s/it]
 62%|██████▏   | 15164/24420 [6:19:18<3:50:50,  1.50s/it]
 62%|██████▏   | 15165/24420 [6:19:19<3:50:42,  1.50s/it]
                                                         
{'loss': 0.0785, 'learning_rate': 5e-05, 'epoch': 3.11}

 62%|██████▏   | 15165/24420 [6:19:19<3:50:42,  1.50s/it]
 62%|██████▏   | 15166/24420 [6:19:21<3:50:10,  1.49s/it]
 62%|██████▏   | 15167/24420 [6:19:22<3:50:17,  1.49s/it]
 62%|██████▏   | 15168/24420 [6:19:24<3:50:22,  1.49s/it]
 62%|██████▏   | 15169/24420 [6:19:25<3:50:16,  1.49s/it]
 62%|██████▏   | 15170/24420 [6:19:27<3:49:38,  1.49s/it]
                                                         
{'loss': 0.0888, 'learning_rate': 5e-05, 'epoch': 3.11}

 62%|██████▏   | 15170/24420 [6:19:27<3:49:38,  1.49s/it]
 62%|██████▏   | 15171/24420 [6:19:28<3:49:22,  1.49s/it]
 62%|██████▏   | 15172/24420 [6:19:30<3:49:47,  1.49s/it]
 62%|██████▏   | 15173/24420 [6:19:31<3:49:40,  1.49s/it]
 62%|██████▏   | 15174/24420 [6:19:33<3:49:20,  1.49s/it]
 62%|██████▏   | 15175/24420 [6:19:34<3:49:23,  1.49s/it]
                                                         
{'loss': 0.0784, 'learning_rate': 5e-05, 'epoch': 3.11}

 62%|██████▏   | 15175/24420 [6:19:34<3:49:23,  1.49s/it]
 62%|██████▏   | 15176/24420 [6:19:36<3:50:13,  1.49s/it]
 62%|██████▏   | 15177/24420 [6:19:37<3:49:38,  1.49s/it]
 62%|██████▏   | 15178/24420 [6:19:39<3:49:10,  1.49s/it]
 62%|██████▏   | 15179/24420 [6:19:40<3:49:14,  1.49s/it]
 62%|██████▏   | 15180/24420 [6:19:41<3:48:57,  1.49s/it]
                                                         
{'loss': 0.0779, 'learning_rate': 5e-05, 'epoch': 3.11}

 62%|██████▏   | 15180/24420 [6:19:41<3:48:57,  1.49s/it]
 62%|██████▏   | 15181/24420 [6:19:43<3:49:09,  1.49s/it]
 62%|██████▏   | 15182/24420 [6:19:44<3:49:04,  1.49s/it]
 62%|██████▏   | 15183/24420 [6:19:46<3:49:05,  1.49s/it]
 62%|██████▏   | 15184/24420 [6:19:47<3:48:43,  1.49s/it]
 62%|██████▏   | 15185/24420 [6:19:49<3:49:03,  1.49s/it]
                                                         
{'loss': 0.0821, 'learning_rate': 5e-05, 'epoch': 3.11}

 62%|██████▏   | 15185/24420 [6:19:49<3:49:03,  1.49s/it]
 62%|██████▏   | 15186/24420 [6:19:50<3:48:43,  1.49s/it]
 62%|██████▏   | 15187/24420 [6:19:52<3:48:47,  1.49s/it]
 62%|██████▏   | 15188/24420 [6:19:53<3:48:50,  1.49s/it]
 62%|██████▏   | 15189/24420 [6:19:55<3:48:39,  1.49s/it]
 62%|██████▏   | 15190/24420 [6:19:56<3:48:44,  1.49s/it]
                                                         
{'loss': 0.087, 'learning_rate': 5e-05, 'epoch': 3.11}

 62%|██████▏   | 15190/24420 [6:19:56<3:48:44,  1.49s/it]
 62%|██████▏   | 15191/24420 [6:19:58<3:49:20,  1.49s/it]
 62%|██████▏   | 15192/24420 [6:19:59<3:49:20,  1.49s/it]
 62%|██████▏   | 15193/24420 [6:20:01<3:50:07,  1.50s/it]
 62%|██████▏   | 15194/24420 [6:20:02<3:49:49,  1.49s/it]
 62%|██████▏   | 15195/24420 [6:20:04<3:49:17,  1.49s/it]
                                                         
{'loss': 0.0848, 'learning_rate': 5e-05, 'epoch': 3.11}

 62%|██████▏   | 15195/24420 [6:20:04<3:49:17,  1.49s/it]
 62%|██████▏   | 15196/24420 [6:20:05<3:49:07,  1.49s/it]
 62%|██████▏   | 15197/24420 [6:20:07<3:49:04,  1.49s/it]
 62%|██████▏   | 15198/24420 [6:20:08<3:49:17,  1.49s/it]
 62%|██████▏   | 15199/24420 [6:20:10<3:49:15,  1.49s/it]
 62%|██████▏   | 15200/24420 [6:20:11<3:48:51,  1.49s/it]
                                                         
{'loss': 0.0872, 'learning_rate': 5e-05, 'epoch': 3.11}

 62%|██████▏   | 15200/24420 [6:20:11<3:48:51,  1.49s/it]
 62%|██████▏   | 15201/24420 [6:20:16<6:19:31,  2.47s/it]
 62%|██████▏   | 15202/24420 [6:20:18<5:33:56,  2.17s/it]
 62%|██████▏   | 15203/24420 [6:20:19<5:02:19,  1.97s/it]
 62%|██████▏   | 15204/24420 [6:20:21<4:40:29,  1.83s/it]
 62%|██████▏   | 15205/24420 [6:20:22<4:24:50,  1.72s/it]
                                                         
{'loss': 0.0872, 'learning_rate': 5e-05, 'epoch': 3.11}

 62%|██████▏   | 15205/24420 [6:20:22<4:24:50,  1.72s/it]
 62%|██████▏   | 15206/24420 [6:20:23<4:13:54,  1.65s/it]
 62%|██████▏   | 15207/24420 [6:20:25<4:06:13,  1.60s/it]
 62%|██████▏   | 15208/24420 [6:20:26<4:01:07,  1.57s/it]
 62%|██████▏   | 15209/24420 [6:20:28<3:57:20,  1.55s/it]
 62%|██████▏   | 15210/24420 [6:20:29<3:54:52,  1.53s/it]
                                                         
{'loss': 0.0904, 'learning_rate': 5e-05, 'epoch': 3.11}

 62%|██████▏   | 15210/24420 [6:20:29<3:54:52,  1.53s/it]
 62%|██████▏   | 15211/24420 [6:20:31<3:53:10,  1.52s/it]
 62%|██████▏   | 15212/24420 [6:20:32<3:51:31,  1.51s/it]
 62%|██████▏   | 15213/24420 [6:20:34<3:50:39,  1.50s/it]
 62%|██████▏   | 15214/24420 [6:20:35<3:50:18,  1.50s/it]
 62%|██████▏   | 15215/24420 [6:20:37<3:49:54,  1.50s/it]
                                                         
{'loss': 0.0782, 'learning_rate': 5e-05, 'epoch': 3.12}

 62%|██████▏   | 15215/24420 [6:20:37<3:49:54,  1.50s/it]
 62%|██████▏   | 15216/24420 [6:20:38<3:49:32,  1.50s/it]
 62%|██████▏   | 15217/24420 [6:20:40<3:49:19,  1.50s/it]
 62%|██████▏   | 15218/24420 [6:20:41<3:49:28,  1.50s/it]
 62%|██████▏   | 15219/24420 [6:20:43<3:49:16,  1.50s/it]
 62%|██████▏   | 15220/24420 [6:20:44<3:49:12,  1.49s/it]
                                                         
{'loss': 0.0852, 'learning_rate': 5e-05, 'epoch': 3.12}

 62%|██████▏   | 15220/24420 [6:20:44<3:49:12,  1.49s/it]
 62%|██████▏   | 15221/24420 [6:20:46<3:49:18,  1.50s/it]
 62%|██████▏   | 15222/24420 [6:20:47<3:48:28,  1.49s/it]
 62%|██████▏   | 15223/24420 [6:20:49<3:48:13,  1.49s/it]
 62%|██████▏   | 15224/24420 [6:20:50<3:48:15,  1.49s/it]
 62%|██████▏   | 15225/24420 [6:20:52<3:48:23,  1.49s/it]
                                                         
{'loss': 0.0689, 'learning_rate': 5e-05, 'epoch': 3.12}

 62%|██████▏   | 15225/24420 [6:20:52<3:48:23,  1.49s/it]
 62%|██████▏   | 15226/24420 [6:20:53<3:48:46,  1.49s/it]
 62%|██████▏   | 15227/24420 [6:20:55<3:48:52,  1.49s/it]
 62%|██████▏   | 15228/24420 [6:20:56<3:49:14,  1.50s/it]
 62%|██████▏   | 15229/24420 [6:20:58<3:49:48,  1.50s/it]
 62%|██████▏   | 15230/24420 [6:20:59<3:49:48,  1.50s/it]
                                                         
{'loss': 0.0764, 'learning_rate': 5e-05, 'epoch': 3.12}

 62%|██████▏   | 15230/24420 [6:20:59<3:49:48,  1.50s/it]
 62%|██████▏   | 15231/24420 [6:21:01<3:49:21,  1.50s/it]
 62%|██████▏   | 15232/24420 [6:21:02<3:48:40,  1.49s/it]
 62%|██████▏   | 15233/24420 [6:21:04<3:48:45,  1.49s/it]
 62%|██████▏   | 15234/24420 [6:21:05<3:48:34,  1.49s/it]
 62%|██████▏   | 15235/24420 [6:21:07<3:48:16,  1.49s/it]
                                                         
{'loss': 0.0795, 'learning_rate': 5e-05, 'epoch': 3.12}

 62%|██████▏   | 15235/24420 [6:21:07<3:48:16,  1.49s/it]
 62%|██████▏   | 15236/24420 [6:21:08<3:48:29,  1.49s/it]
 62%|██████▏   | 15237/24420 [6:21:10<3:48:44,  1.49s/it]
 62%|██████▏   | 15238/24420 [6:21:11<3:48:18,  1.49s/it]
 62%|██████▏   | 15239/24420 [6:21:13<3:48:08,  1.49s/it]
 62%|██████▏   | 15240/24420 [6:21:14<3:48:08,  1.49s/it]
                                                         
{'loss': 0.0666, 'learning_rate': 5e-05, 'epoch': 3.12}

 62%|██████▏   | 15240/24420 [6:21:14<3:48:08,  1.49s/it]
 62%|██████▏   | 15241/24420 [6:21:16<3:47:56,  1.49s/it]
 62%|██████▏   | 15242/24420 [6:21:17<3:47:59,  1.49s/it]
 62%|██████▏   | 15243/24420 [6:21:19<3:48:32,  1.49s/it]
 62%|██████▏   | 15244/24420 [6:21:20<3:48:07,  1.49s/it]
 62%|██████▏   | 15245/24420 [6:21:22<3:47:38,  1.49s/it]
                                                         
{'loss': 0.0795, 'learning_rate': 5e-05, 'epoch': 3.12}

 62%|██████▏   | 15245/24420 [6:21:22<3:47:38,  1.49s/it]
 62%|██████▏   | 15246/24420 [6:21:23<3:47:41,  1.49s/it]
 62%|██████▏   | 15247/24420 [6:21:25<3:47:30,  1.49s/it]
 62%|██████▏   | 15248/24420 [6:21:26<3:47:46,  1.49s/it]
 62%|██████▏   | 15249/24420 [6:21:28<3:47:39,  1.49s/it]
 62%|██████▏   | 15250/24420 [6:21:29<3:47:58,  1.49s/it]
                                                         
{'loss': 0.0786, 'learning_rate': 5e-05, 'epoch': 3.12}

 62%|██████▏   | 15250/24420 [6:21:29<3:47:58,  1.49s/it]
 62%|██████▏   | 15251/24420 [6:21:31<3:47:54,  1.49s/it]
 62%|██████▏   | 15252/24420 [6:21:32<3:47:55,  1.49s/it]
 62%|██████▏   | 15253/24420 [6:21:34<3:47:54,  1.49s/it]
 62%|██████▏   | 15254/24420 [6:21:35<3:47:30,  1.49s/it]
 62%|██████▏   | 15255/24420 [6:21:37<3:47:35,  1.49s/it]
                                                         
{'loss': 0.0675, 'learning_rate': 5e-05, 'epoch': 3.12}

 62%|██████▏   | 15255/24420 [6:21:37<3:47:35,  1.49s/it]
 62%|██████▏   | 15256/24420 [6:21:38<3:47:45,  1.49s/it]
 62%|██████▏   | 15257/24420 [6:21:40<3:47:38,  1.49s/it]
 62%|██████▏   | 15258/24420 [6:21:41<3:47:47,  1.49s/it]
 62%|██████▏   | 15259/24420 [6:21:43<3:48:34,  1.50s/it]
 62%|██████▏   | 15260/24420 [6:21:44<3:48:02,  1.49s/it]
                                                         
{'loss': 0.1044, 'learning_rate': 5e-05, 'epoch': 3.12}

 62%|██████▏   | 15260/24420 [6:21:44<3:48:02,  1.49s/it]
 62%|██████▏   | 15261/24420 [6:21:46<3:48:31,  1.50s/it]
 62%|██████▏   | 15262/24420 [6:21:47<3:48:38,  1.50s/it]
 63%|██████▎   | 15263/24420 [6:21:49<3:48:23,  1.50s/it]
 63%|██████▎   | 15264/24420 [6:21:50<3:48:08,  1.49s/it]
 63%|██████▎   | 15265/24420 [6:21:52<3:48:06,  1.49s/it]
                                                         
{'loss': 0.066, 'learning_rate': 5e-05, 'epoch': 3.13}

 63%|██████▎   | 15265/24420 [6:21:52<3:48:06,  1.49s/it]
 63%|██████▎   | 15266/24420 [6:21:53<3:48:16,  1.50s/it]
 63%|██████▎   | 15267/24420 [6:21:55<3:48:02,  1.49s/it]
 63%|██████▎   | 15268/24420 [6:21:56<3:48:30,  1.50s/it]
 63%|██████▎   | 15269/24420 [6:21:58<3:47:47,  1.49s/it]
 63%|██████▎   | 15270/24420 [6:21:59<3:47:54,  1.49s/it]
                                                         
{'loss': 0.068, 'learning_rate': 5e-05, 'epoch': 3.13}

 63%|██████▎   | 15270/24420 [6:21:59<3:47:54,  1.49s/it]
 63%|██████▎   | 15271/24420 [6:22:01<3:48:15,  1.50s/it]
 63%|██████▎   | 15272/24420 [6:22:02<3:48:18,  1.50s/it]
 63%|██████▎   | 15273/24420 [6:22:04<3:49:59,  1.51s/it]
 63%|██████▎   | 15274/24420 [6:22:05<3:49:11,  1.50s/it]
 63%|██████▎   | 15275/24420 [6:22:07<3:48:48,  1.50s/it]
                                                         
{'loss': 0.0671, 'learning_rate': 5e-05, 'epoch': 3.13}

 63%|██████▎   | 15275/24420 [6:22:07<3:48:48,  1.50s/it]
 63%|██████▎   | 15276/24420 [6:22:08<3:48:58,  1.50s/it]
 63%|██████▎   | 15277/24420 [6:22:10<3:47:57,  1.50s/it]
 63%|██████▎   | 15278/24420 [6:22:11<3:48:30,  1.50s/it]
 63%|██████▎   | 15279/24420 [6:22:13<3:48:03,  1.50s/it]
 63%|██████▎   | 15280/24420 [6:22:14<3:47:40,  1.49s/it]
                                                         
{'loss': 0.0881, 'learning_rate': 5e-05, 'epoch': 3.13}

 63%|██████▎   | 15280/24420 [6:22:14<3:47:40,  1.49s/it]
 63%|██████▎   | 15281/24420 [6:22:16<3:47:06,  1.49s/it]
 63%|██████▎   | 15282/24420 [6:22:17<3:47:08,  1.49s/it]
 63%|██████▎   | 15283/24420 [6:22:18<3:46:53,  1.49s/it]
 63%|██████▎   | 15284/24420 [6:22:20<3:46:55,  1.49s/it]
 63%|██████▎   | 15285/24420 [6:22:21<3:47:09,  1.49s/it]
                                                         
{'loss': 0.1009, 'learning_rate': 5e-05, 'epoch': 3.13}

 63%|██████▎   | 15285/24420 [6:22:21<3:47:09,  1.49s/it]
 63%|██████▎   | 15286/24420 [6:22:23<3:47:04,  1.49s/it]
 63%|██████▎   | 15287/24420 [6:22:24<3:48:05,  1.50s/it]
 63%|██████▎   | 15288/24420 [6:22:26<3:48:14,  1.50s/it]
 63%|██████▎   | 15289/24420 [6:22:27<3:47:35,  1.50s/it]
 63%|██████▎   | 15290/24420 [6:22:29<3:48:06,  1.50s/it]
                                                         
{'loss': 0.0764, 'learning_rate': 5e-05, 'epoch': 3.13}

 63%|██████▎   | 15290/24420 [6:22:29<3:48:06,  1.50s/it]
 63%|██████▎   | 15291/24420 [6:22:30<3:47:45,  1.50s/it]
 63%|██████▎   | 15292/24420 [6:22:32<3:47:03,  1.49s/it]
 63%|██████▎   | 15293/24420 [6:22:33<3:47:18,  1.49s/it]
 63%|██████▎   | 15294/24420 [6:22:35<3:47:10,  1.49s/it]
 63%|██████▎   | 15295/24420 [6:22:36<3:47:15,  1.49s/it]
                                                         
{'loss': 0.0752, 'learning_rate': 5e-05, 'epoch': 3.13}

 63%|██████▎   | 15295/24420 [6:22:36<3:47:15,  1.49s/it]
 63%|██████▎   | 15296/24420 [6:22:38<3:47:45,  1.50s/it]
 63%|██████▎   | 15297/24420 [6:22:39<3:47:16,  1.49s/it]
 63%|██████▎   | 15298/24420 [6:22:41<3:46:54,  1.49s/it]
 63%|██████▎   | 15299/24420 [6:22:42<3:46:54,  1.49s/it]
 63%|██████▎   | 15300/24420 [6:22:44<3:46:21,  1.49s/it]
                                                         
{'loss': 0.0832, 'learning_rate': 5e-05, 'epoch': 3.13}

 63%|██████▎   | 15300/24420 [6:22:44<3:46:21,  1.49s/it]
 63%|██████▎   | 15301/24420 [6:22:45<3:47:14,  1.50s/it]
 63%|██████▎   | 15302/24420 [6:22:47<3:48:06,  1.50s/it]
 63%|██████▎   | 15303/24420 [6:22:48<3:47:48,  1.50s/it]
 63%|██████▎   | 15304/24420 [6:22:50<3:47:18,  1.50s/it]
 63%|██████▎   | 15305/24420 [6:22:51<3:46:39,  1.49s/it]
                                                         
{'loss': 0.0658, 'learning_rate': 5e-05, 'epoch': 3.13}

 63%|██████▎   | 15305/24420 [6:22:51<3:46:39,  1.49s/it]
 63%|██████▎   | 15306/24420 [6:22:53<3:46:29,  1.49s/it]
 63%|██████▎   | 15307/24420 [6:22:54<3:46:08,  1.49s/it]
 63%|██████▎   | 15308/24420 [6:22:56<3:45:55,  1.49s/it]
 63%|██████▎   | 15309/24420 [6:22:57<3:45:49,  1.49s/it]
 63%|██████▎   | 15310/24420 [6:22:59<3:45:12,  1.48s/it]
                                                         
{'loss': 0.0846, 'learning_rate': 5e-05, 'epoch': 3.13}

 63%|██████▎   | 15310/24420 [6:22:59<3:45:12,  1.48s/it]
 63%|██████▎   | 15311/24420 [6:23:00<3:45:29,  1.49s/it]
 63%|██████▎   | 15312/24420 [6:23:02<3:45:39,  1.49s/it]
 63%|██████▎   | 15313/24420 [6:23:03<3:45:46,  1.49s/it]
 63%|██████▎   | 15314/24420 [6:23:05<3:45:28,  1.49s/it]
 63%|██████▎   | 15315/24420 [6:23:06<3:45:21,  1.49s/it]
                                                         
{'loss': 0.0814, 'learning_rate': 5e-05, 'epoch': 3.14}

 63%|██████▎   | 15315/24420 [6:23:06<3:45:21,  1.49s/it]
 63%|██████▎   | 15316/24420 [6:23:08<3:45:28,  1.49s/it]
 63%|██████▎   | 15317/24420 [6:23:09<3:45:15,  1.48s/it]
 63%|██████▎   | 15318/24420 [6:23:11<3:45:13,  1.48s/it]
 63%|██████▎   | 15319/24420 [6:23:12<3:45:21,  1.49s/it]
 63%|██████▎   | 15320/24420 [6:23:14<3:45:35,  1.49s/it]
                                                         
{'loss': 0.0905, 'learning_rate': 5e-05, 'epoch': 3.14}

 63%|██████▎   | 15320/24420 [6:23:14<3:45:35,  1.49s/it]
 63%|██████▎   | 15321/24420 [6:23:15<3:46:03,  1.49s/it]
 63%|██████▎   | 15322/24420 [6:23:17<3:46:00,  1.49s/it]
 63%|██████▎   | 15323/24420 [6:23:18<3:46:19,  1.49s/it]
 63%|██████▎   | 15324/24420 [6:23:20<3:46:11,  1.49s/it]
 63%|██████▎   | 15325/24420 [6:23:21<3:46:03,  1.49s/it]
                                                         
{'loss': 0.0722, 'learning_rate': 5e-05, 'epoch': 3.14}

 63%|██████▎   | 15325/24420 [6:23:21<3:46:03,  1.49s/it]
 63%|██████▎   | 15326/24420 [6:23:23<3:46:01,  1.49s/it]
 63%|██████▎   | 15327/24420 [6:23:24<3:45:41,  1.49s/it]
 63%|██████▎   | 15328/24420 [6:23:26<3:45:53,  1.49s/it]
 63%|██████▎   | 15329/24420 [6:23:27<3:45:39,  1.49s/it]
 63%|██████▎   | 15330/24420 [6:23:29<3:45:15,  1.49s/it]
                                                         
{'loss': 0.0834, 'learning_rate': 5e-05, 'epoch': 3.14}

 63%|██████▎   | 15330/24420 [6:23:29<3:45:15,  1.49s/it]
 63%|██████▎   | 15331/24420 [6:23:30<3:45:25,  1.49s/it]
 63%|██████▎   | 15332/24420 [6:23:32<3:45:21,  1.49s/it]
 63%|██████▎   | 15333/24420 [6:23:33<3:45:31,  1.49s/it]
 63%|██████▎   | 15334/24420 [6:23:35<3:45:29,  1.49s/it]
 63%|██████▎   | 15335/24420 [6:23:36<3:45:55,  1.49s/it]
                                                         
{'loss': 0.0911, 'learning_rate': 5e-05, 'epoch': 3.14}

 63%|██████▎   | 15335/24420 [6:23:36<3:45:55,  1.49s/it]
 63%|██████▎   | 15336/24420 [6:23:38<3:45:35,  1.49s/it]
 63%|██████▎   | 15337/24420 [6:23:39<3:45:28,  1.49s/it]
 63%|██████▎   | 15338/24420 [6:23:40<3:45:17,  1.49s/it]
 63%|██████▎   | 15339/24420 [6:23:42<3:45:38,  1.49s/it]
 63%|██████▎   | 15340/24420 [6:23:43<3:45:41,  1.49s/it]
                                                         
{'loss': 0.0727, 'learning_rate': 5e-05, 'epoch': 3.14}

 63%|██████▎   | 15340/24420 [6:23:43<3:45:41,  1.49s/it]
 63%|██████▎   | 15341/24420 [6:23:45<3:45:46,  1.49s/it]
 63%|██████▎   | 15342/24420 [6:23:46<3:45:25,  1.49s/it]
 63%|██████▎   | 15343/24420 [6:23:48<3:45:08,  1.49s/it]
 63%|██████▎   | 15344/24420 [6:23:49<3:45:06,  1.49s/it]
 63%|██████▎   | 15345/24420 [6:23:51<3:45:05,  1.49s/it]
                                                         
{'loss': 0.0665, 'learning_rate': 5e-05, 'epoch': 3.14}

 63%|██████▎   | 15345/24420 [6:23:51<3:45:05,  1.49s/it]
 63%|██████▎   | 15346/24420 [6:23:52<3:45:43,  1.49s/it]
 63%|██████▎   | 15347/24420 [6:23:54<3:45:31,  1.49s/it]
 63%|██████▎   | 15348/24420 [6:23:55<3:45:27,  1.49s/it]
 63%|██████▎   | 15349/24420 [6:23:57<3:45:33,  1.49s/it]
 63%|██████▎   | 15350/24420 [6:23:58<3:45:22,  1.49s/it]
                                                         
{'loss': 0.0679, 'learning_rate': 5e-05, 'epoch': 3.14}

 63%|██████▎   | 15350/24420 [6:23:58<3:45:22,  1.49s/it]
 63%|██████▎   | 15351/24420 [6:24:00<3:45:35,  1.49s/it]
 63%|██████▎   | 15352/24420 [6:24:01<3:45:41,  1.49s/it]
 63%|██████▎   | 15353/24420 [6:24:03<3:46:01,  1.50s/it]
 63%|██████▎   | 15354/24420 [6:24:04<3:45:53,  1.49s/it]
 63%|██████▎   | 15355/24420 [6:24:06<3:45:33,  1.49s/it]
                                                         
{'loss': 0.0692, 'learning_rate': 5e-05, 'epoch': 3.14}

 63%|██████▎   | 15355/24420 [6:24:06<3:45:33,  1.49s/it]
 63%|██████▎   | 15356/24420 [6:24:07<3:45:54,  1.50s/it]
 63%|██████▎   | 15357/24420 [6:24:09<3:45:53,  1.50s/it]
 63%|██████▎   | 15358/24420 [6:24:10<3:45:36,  1.49s/it]
 63%|██████▎   | 15359/24420 [6:24:12<3:45:10,  1.49s/it]
 63%|██████▎   | 15360/24420 [6:24:13<3:45:04,  1.49s/it]
                                                         
{'loss': 0.0761, 'learning_rate': 5e-05, 'epoch': 3.14}

 63%|██████▎   | 15360/24420 [6:24:13<3:45:04,  1.49s/it]
 63%|██████▎   | 15361/24420 [6:24:15<3:45:34,  1.49s/it]
 63%|██████▎   | 15362/24420 [6:24:16<3:45:12,  1.49s/it]
 63%|██████▎   | 15363/24420 [6:24:18<3:45:23,  1.49s/it]
 63%|██████▎   | 15364/24420 [6:24:19<3:45:16,  1.49s/it]
 63%|██████▎   | 15365/24420 [6:24:21<3:44:59,  1.49s/it]
                                                         
{'loss': 0.0812, 'learning_rate': 5e-05, 'epoch': 3.15}

 63%|██████▎   | 15365/24420 [6:24:21<3:44:59,  1.49s/it]
 63%|██████▎   | 15366/24420 [6:24:22<3:45:22,  1.49s/it]
 63%|██████▎   | 15367/24420 [6:24:24<3:45:23,  1.49s/it]
 63%|██████▎   | 15368/24420 [6:24:25<3:44:58,  1.49s/it]
 63%|██████▎   | 15369/24420 [6:24:27<3:45:03,  1.49s/it]
 63%|██████▎   | 15370/24420 [6:24:28<3:44:58,  1.49s/it]
                                                         
{'loss': 0.0947, 'learning_rate': 5e-05, 'epoch': 3.15}

 63%|██████▎   | 15370/24420 [6:24:28<3:44:58,  1.49s/it]
 63%|██████▎   | 15371/24420 [6:24:30<3:45:12,  1.49s/it]
 63%|██████▎   | 15372/24420 [6:24:31<3:44:53,  1.49s/it]
 63%|██████▎   | 15373/24420 [6:24:33<3:44:46,  1.49s/it]
 63%|██████▎   | 15374/24420 [6:24:34<3:44:32,  1.49s/it]
 63%|██████▎   | 15375/24420 [6:24:36<3:44:20,  1.49s/it]
                                                         
{'loss': 0.0884, 'learning_rate': 5e-05, 'epoch': 3.15}

 63%|██████▎   | 15375/24420 [6:24:36<3:44:20,  1.49s/it]
 63%|██████▎   | 15376/24420 [6:24:37<3:44:38,  1.49s/it]
 63%|██████▎   | 15377/24420 [6:24:39<3:44:38,  1.49s/it]
 63%|██████▎   | 15378/24420 [6:24:40<3:44:43,  1.49s/it]
 63%|██████▎   | 15379/24420 [6:24:42<3:44:43,  1.49s/it]
 63%|██████▎   | 15380/24420 [6:24:43<3:44:36,  1.49s/it]
                                                         
{'loss': 0.1016, 'learning_rate': 5e-05, 'epoch': 3.15}

 63%|██████▎   | 15380/24420 [6:24:43<3:44:36,  1.49s/it]
 63%|██████▎   | 15381/24420 [6:24:45<3:45:10,  1.49s/it]
 63%|██████▎   | 15382/24420 [6:24:46<3:45:05,  1.49s/it]
 63%|██████▎   | 15383/24420 [6:24:48<3:44:42,  1.49s/it]
 63%|██████▎   | 15384/24420 [6:24:49<3:44:17,  1.49s/it]
 63%|██████▎   | 15385/24420 [6:24:51<3:44:35,  1.49s/it]
                                                         
{'loss': 0.083, 'learning_rate': 5e-05, 'epoch': 3.15}

 63%|██████▎   | 15385/24420 [6:24:51<3:44:35,  1.49s/it]
 63%|██████▎   | 15386/24420 [6:24:52<3:45:17,  1.50s/it]
 63%|██████▎   | 15387/24420 [6:24:54<3:45:03,  1.49s/it]
 63%|██████▎   | 15388/24420 [6:24:55<3:44:58,  1.49s/it]
 63%|██████▎   | 15389/24420 [6:24:57<3:44:44,  1.49s/it]
 63%|██████▎   | 15390/24420 [6:24:58<3:44:32,  1.49s/it]
                                                         
{'loss': 0.0921, 'learning_rate': 5e-05, 'epoch': 3.15}

 63%|██████▎   | 15390/24420 [6:24:58<3:44:32,  1.49s/it]
 63%|██████▎   | 15391/24420 [6:25:00<3:44:42,  1.49s/it]
 63%|██████▎   | 15392/24420 [6:25:01<3:44:49,  1.49s/it]
 63%|██████▎   | 15393/24420 [6:25:03<3:44:29,  1.49s/it]
 63%|██████▎   | 15394/24420 [6:25:04<3:44:11,  1.49s/it]
 63%|██████▎   | 15395/24420 [6:25:06<3:44:12,  1.49s/it]
                                                         
{'loss': 0.0675, 'learning_rate': 5e-05, 'epoch': 3.15}

 63%|██████▎   | 15395/24420 [6:25:06<3:44:12,  1.49s/it]
 63%|██████▎   | 15396/24420 [6:25:07<3:44:06,  1.49s/it]
 63%|██████▎   | 15397/24420 [6:25:09<3:44:03,  1.49s/it]
 63%|██████▎   | 15398/24420 [6:25:10<3:44:12,  1.49s/it]
 63%|██████▎   | 15399/24420 [6:25:12<3:44:40,  1.49s/it]
 63%|██████▎   | 15400/24420 [6:25:13<3:44:48,  1.50s/it]
                                                         
{'loss': 0.0866, 'learning_rate': 5e-05, 'epoch': 3.15}

 63%|██████▎   | 15400/24420 [6:25:13<3:44:48,  1.50s/it]
 63%|██████▎   | 15401/24420 [6:25:18<6:11:35,  2.47s/it]
 63%|██████▎   | 15402/24420 [6:25:19<5:27:02,  2.18s/it]
 63%|██████▎   | 15403/24420 [6:25:21<4:56:11,  1.97s/it]
 63%|██████▎   | 15404/24420 [6:25:22<4:34:24,  1.83s/it]
 63%|██████▎   | 15405/24420 [6:25:24<4:19:38,  1.73s/it]
                                                         
{'loss': 0.0751, 'learning_rate': 5e-05, 'epoch': 3.15}

 63%|██████▎   | 15405/24420 [6:25:24<4:19:38,  1.73s/it]
 63%|██████▎   | 15406/24420 [6:25:25<4:08:55,  1.66s/it]
 63%|██████▎   | 15407/24420 [6:25:27<4:01:04,  1.60s/it]
 63%|██████▎   | 15408/24420 [6:25:28<3:56:13,  1.57s/it]
 63%|██████▎   | 15409/24420 [6:25:30<3:52:08,  1.55s/it]
 63%|██████▎   | 15410/24420 [6:25:31<3:49:51,  1.53s/it]
                                                         
{'loss': 0.0782, 'learning_rate': 5e-05, 'epoch': 3.16}

 63%|██████▎   | 15410/24420 [6:25:31<3:49:51,  1.53s/it]
 63%|██████▎   | 15411/24420 [6:25:33<3:47:56,  1.52s/it]
 63%|██████▎   | 15412/24420 [6:25:34<3:46:29,  1.51s/it]
 63%|██████▎   | 15413/24420 [6:25:36<3:45:53,  1.50s/it]
 63%|██████▎   | 15414/24420 [6:25:37<3:45:51,  1.50s/it]
 63%|██████▎   | 15415/24420 [6:25:39<3:45:03,  1.50s/it]
                                                         
{'loss': 0.0804, 'learning_rate': 5e-05, 'epoch': 3.16}

 63%|██████▎   | 15415/24420 [6:25:39<3:45:03,  1.50s/it]
 63%|██████▎   | 15416/24420 [6:25:40<3:44:42,  1.50s/it]
 63%|██████▎   | 15417/24420 [6:25:42<3:44:17,  1.49s/it]
 63%|██████▎   | 15418/24420 [6:25:43<3:43:52,  1.49s/it]
 63%|██████▎   | 15419/24420 [6:25:45<3:43:29,  1.49s/it]
 63%|██████▎   | 15420/24420 [6:25:46<3:43:54,  1.49s/it]
                                                         
{'loss': 0.0773, 'learning_rate': 5e-05, 'epoch': 3.16}

 63%|██████▎   | 15420/24420 [6:25:46<3:43:54,  1.49s/it]
 63%|██████▎   | 15421/24420 [6:25:48<3:44:04,  1.49s/it]
 63%|██████▎   | 15422/24420 [6:25:49<3:43:34,  1.49s/it]
 63%|██████▎   | 15423/24420 [6:25:51<3:43:07,  1.49s/it]
 63%|██████▎   | 15424/24420 [6:25:52<3:43:16,  1.49s/it]
 63%|██████▎   | 15425/24420 [6:25:54<3:43:28,  1.49s/it]
                                                         
{'loss': 0.1016, 'learning_rate': 5e-05, 'epoch': 3.16}

 63%|██████▎   | 15425/24420 [6:25:54<3:43:28,  1.49s/it]
 63%|██████▎   | 15426/24420 [6:25:55<3:43:13,  1.49s/it]
 63%|██████▎   | 15427/24420 [6:25:57<3:43:35,  1.49s/it]
 63%|██████▎   | 15428/24420 [6:25:58<3:43:51,  1.49s/it]
 63%|██████▎   | 15429/24420 [6:26:00<3:43:38,  1.49s/it]
 63%|██████▎   | 15430/24420 [6:26:01<3:43:29,  1.49s/it]
                                                         
{'loss': 0.0727, 'learning_rate': 5e-05, 'epoch': 3.16}

 63%|██████▎   | 15430/24420 [6:26:01<3:43:29,  1.49s/it]
 63%|██████▎   | 15431/24420 [6:26:03<3:43:51,  1.49s/it]
 63%|██████▎   | 15432/24420 [6:26:04<3:44:10,  1.50s/it]
 63%|██████▎   | 15433/24420 [6:26:05<3:43:50,  1.49s/it]
 63%|██████▎   | 15434/24420 [6:26:07<3:43:54,  1.50s/it]
 63%|██████▎   | 15435/24420 [6:26:08<3:43:49,  1.49s/it]
                                                         
{'loss': 0.0823, 'learning_rate': 5e-05, 'epoch': 3.16}

 63%|██████▎   | 15435/24420 [6:26:08<3:43:49,  1.49s/it]
 63%|██████▎   | 15436/24420 [6:26:10<3:43:23,  1.49s/it]
 63%|██████▎   | 15437/24420 [6:26:11<3:43:18,  1.49s/it]
 63%|██████▎   | 15438/24420 [6:26:13<3:43:31,  1.49s/it]
 63%|██████▎   | 15439/24420 [6:26:14<3:43:52,  1.50s/it]
 63%|██████▎   | 15440/24420 [6:26:16<3:43:43,  1.49s/it]
                                                         
{'loss': 0.0907, 'learning_rate': 5e-05, 'epoch': 3.16}

 63%|██████▎   | 15440/24420 [6:26:16<3:43:43,  1.49s/it]
 63%|██████▎   | 15441/24420 [6:26:17<3:43:59,  1.50s/it]
 63%|██████▎   | 15442/24420 [6:26:19<3:43:22,  1.49s/it]
 63%|██████▎   | 15443/24420 [6:26:20<3:43:08,  1.49s/it]
 63%|██████▎   | 15444/24420 [6:26:22<3:42:40,  1.49s/it]
 63%|██████▎   | 15445/24420 [6:26:23<3:42:44,  1.49s/it]
                                                         
{'loss': 0.0932, 'learning_rate': 5e-05, 'epoch': 3.16}

 63%|██████▎   | 15445/24420 [6:26:23<3:42:44,  1.49s/it]
 63%|██████▎   | 15446/24420 [6:26:25<3:42:33,  1.49s/it]
 63%|██████▎   | 15447/24420 [6:26:26<3:42:40,  1.49s/it]
 63%|██████▎   | 15448/24420 [6:26:28<3:42:48,  1.49s/it]
 63%|██████▎   | 15449/24420 [6:26:29<3:42:43,  1.49s/it]
 63%|██████▎   | 15450/24420 [6:26:31<3:42:24,  1.49s/it]
                                                         
{'loss': 0.0814, 'learning_rate': 5e-05, 'epoch': 3.16}

 63%|██████▎   | 15450/24420 [6:26:31<3:42:24,  1.49s/it]
 63%|██████▎   | 15451/24420 [6:26:32<3:42:22,  1.49s/it]
 63%|██████▎   | 15452/24420 [6:26:34<3:42:16,  1.49s/it]
 63%|██████▎   | 15453/24420 [6:26:35<3:42:40,  1.49s/it]
 63%|██████▎   | 15454/24420 [6:26:37<3:42:33,  1.49s/it]
 63%|██████▎   | 15455/24420 [6:26:38<3:42:34,  1.49s/it]
                                                         
{'loss': 0.0732, 'learning_rate': 5e-05, 'epoch': 3.16}

 63%|██████▎   | 15455/24420 [6:26:38<3:42:34,  1.49s/it]
 63%|██████▎   | 15456/24420 [6:26:40<3:42:36,  1.49s/it]
 63%|██████▎   | 15457/24420 [6:26:41<3:42:42,  1.49s/it]
 63%|██████▎   | 15458/24420 [6:26:43<3:42:32,  1.49s/it]
 63%|██████▎   | 15459/24420 [6:26:44<3:42:32,  1.49s/it]
 63%|██████▎   | 15460/24420 [6:26:46<3:42:16,  1.49s/it]
                                                         
{'loss': 0.0863, 'learning_rate': 5e-05, 'epoch': 3.17}

 63%|██████▎   | 15460/24420 [6:26:46<3:42:16,  1.49s/it]
 63%|██████▎   | 15461/24420 [6:26:47<3:42:30,  1.49s/it]
 63%|██████▎   | 15462/24420 [6:26:49<3:42:22,  1.49s/it]
 63%|██████▎   | 15463/24420 [6:26:50<3:42:13,  1.49s/it]
 63%|██████▎   | 15464/24420 [6:26:52<3:42:39,  1.49s/it]
 63%|██████▎   | 15465/24420 [6:26:53<3:42:09,  1.49s/it]
                                                         
{'loss': 0.0773, 'learning_rate': 5e-05, 'epoch': 3.17}

 63%|██████▎   | 15465/24420 [6:26:53<3:42:09,  1.49s/it]
 63%|██████▎   | 15466/24420 [6:26:55<3:42:18,  1.49s/it]
 63%|██████▎   | 15467/24420 [6:26:56<3:42:00,  1.49s/it]
 63%|██████▎   | 15468/24420 [6:26:58<3:41:46,  1.49s/it]
 63%|██████▎   | 15469/24420 [6:26:59<3:42:05,  1.49s/it]
 63%|██████▎   | 15470/24420 [6:27:01<3:41:48,  1.49s/it]
                                                         
{'loss': 0.0949, 'learning_rate': 5e-05, 'epoch': 3.17}

 63%|██████▎   | 15470/24420 [6:27:01<3:41:48,  1.49s/it]
 63%|██████▎   | 15471/24420 [6:27:02<3:42:29,  1.49s/it]
 63%|██████▎   | 15472/24420 [6:27:04<3:42:00,  1.49s/it]
 63%|██████▎   | 15473/24420 [6:27:05<3:41:38,  1.49s/it]
 63%|██████▎   | 15474/24420 [6:27:07<3:41:11,  1.48s/it]
 63%|██████▎   | 15475/24420 [6:27:08<3:41:22,  1.48s/it]
                                                         
{'loss': 0.0938, 'learning_rate': 5e-05, 'epoch': 3.17}

 63%|██████▎   | 15475/24420 [6:27:08<3:41:22,  1.48s/it]
 63%|██████▎   | 15476/24420 [6:27:10<3:41:38,  1.49s/it]
 63%|██████▎   | 15477/24420 [6:27:11<3:41:53,  1.49s/it]
 63%|██████▎   | 15478/24420 [6:27:13<3:41:52,  1.49s/it]
 63%|██████▎   | 15479/24420 [6:27:14<3:41:49,  1.49s/it]
 63%|██████▎   | 15480/24420 [6:27:16<3:42:00,  1.49s/it]
                                                         
{'loss': 0.0745, 'learning_rate': 5e-05, 'epoch': 3.17}

 63%|██████▎   | 15480/24420 [6:27:16<3:42:00,  1.49s/it]
 63%|██████▎   | 15481/24420 [6:27:17<3:41:58,  1.49s/it]
 63%|██████▎   | 15482/24420 [6:27:18<3:42:24,  1.49s/it]
 63%|██████▎   | 15483/24420 [6:27:20<3:41:39,  1.49s/it]
 63%|██████▎   | 15484/24420 [6:27:21<3:41:21,  1.49s/it]
 63%|██████▎   | 15485/24420 [6:27:23<3:41:34,  1.49s/it]
                                                         
{'loss': 0.0702, 'learning_rate': 5e-05, 'epoch': 3.17}

 63%|██████▎   | 15485/24420 [6:27:23<3:41:34,  1.49s/it]
 63%|██████▎   | 15486/24420 [6:27:24<3:41:35,  1.49s/it]
 63%|██████▎   | 15487/24420 [6:27:26<3:41:09,  1.49s/it]
 63%|██████▎   | 15488/24420 [6:27:27<3:41:41,  1.49s/it]
 63%|██████▎   | 15489/24420 [6:27:29<3:41:59,  1.49s/it]
 63%|██████▎   | 15490/24420 [6:27:30<3:41:43,  1.49s/it]
                                                         
{'loss': 0.0739, 'learning_rate': 5e-05, 'epoch': 3.17}

 63%|██████▎   | 15490/24420 [6:27:30<3:41:43,  1.49s/it]
 63%|██████▎   | 15491/24420 [6:27:32<3:41:34,  1.49s/it]
 63%|██████▎   | 15492/24420 [6:27:33<3:41:12,  1.49s/it]
 63%|██████▎   | 15493/24420 [6:27:35<3:41:38,  1.49s/it]
 63%|██████▎   | 15494/24420 [6:27:36<3:41:34,  1.49s/it]
 63%|██████▎   | 15495/24420 [6:27:38<3:41:54,  1.49s/it]
                                                         
{'loss': 0.0834, 'learning_rate': 5e-05, 'epoch': 3.17}

 63%|██████▎   | 15495/24420 [6:27:38<3:41:54,  1.49s/it]
 63%|██████▎   | 15496/24420 [6:27:39<3:42:23,  1.50s/it]
 63%|██████▎   | 15497/24420 [6:27:41<3:41:54,  1.49s/it]
 63%|██████▎   | 15498/24420 [6:27:42<3:42:01,  1.49s/it]
 63%|██████▎   | 15499/24420 [6:27:44<3:42:07,  1.49s/it]
 63%|██████▎   | 15500/24420 [6:27:45<3:41:50,  1.49s/it]
                                                         
{'loss': 0.0889, 'learning_rate': 5e-05, 'epoch': 3.17}

 63%|██████▎   | 15500/24420 [6:27:45<3:41:50,  1.49s/it]
 63%|██████▎   | 15501/24420 [6:27:47<3:41:20,  1.49s/it]
 63%|██████▎   | 15502/24420 [6:27:48<3:41:17,  1.49s/it]
 63%|██████▎   | 15503/24420 [6:27:50<3:41:01,  1.49s/it]
 63%|██████▎   | 15504/24420 [6:27:51<3:40:59,  1.49s/it]
 63%|██████▎   | 15505/24420 [6:27:53<3:43:35,  1.50s/it]
                                                         
{'loss': 0.0768, 'learning_rate': 5e-05, 'epoch': 3.17}

 63%|██████▎   | 15505/24420 [6:27:53<3:43:35,  1.50s/it]
 63%|██████▎   | 15506/24420 [6:27:54<3:43:00,  1.50s/it]
 64%|██████▎   | 15507/24420 [6:27:56<3:42:34,  1.50s/it]
 64%|██████▎   | 15508/24420 [6:27:57<3:41:49,  1.49s/it]
 64%|██████▎   | 15509/24420 [6:27:59<3:41:28,  1.49s/it]
 64%|██████▎   | 15510/24420 [6:28:00<3:41:26,  1.49s/it]
                                                         
{'loss': 0.0825, 'learning_rate': 5e-05, 'epoch': 3.18}

 64%|██████▎   | 15510/24420 [6:28:00<3:41:26,  1.49s/it]
 64%|██████▎   | 15511/24420 [6:28:02<3:41:45,  1.49s/it]
 64%|██████▎   | 15512/24420 [6:28:03<3:41:27,  1.49s/it]
 64%|██████▎   | 15513/24420 [6:28:05<3:41:23,  1.49s/it]
 64%|██████▎   | 15514/24420 [6:28:06<3:41:24,  1.49s/it]
 64%|██████▎   | 15515/24420 [6:28:08<3:41:01,  1.49s/it]
                                                         
{'loss': 0.0717, 'learning_rate': 5e-05, 'epoch': 3.18}

 64%|██████▎   | 15515/24420 [6:28:08<3:41:01,  1.49s/it]
 64%|██████▎   | 15516/24420 [6:28:09<3:40:55,  1.49s/it]
 64%|██████▎   | 15517/24420 [6:28:11<3:40:53,  1.49s/it]
 64%|██████▎   | 15518/24420 [6:28:12<3:40:41,  1.49s/it]
 64%|██████▎   | 15519/24420 [6:28:14<3:40:52,  1.49s/it]
 64%|██████▎   | 15520/24420 [6:28:15<3:40:53,  1.49s/it]
                                                         
{'loss': 0.0672, 'learning_rate': 5e-05, 'epoch': 3.18}

 64%|██████▎   | 15520/24420 [6:28:15<3:40:53,  1.49s/it]
 64%|██████▎   | 15521/24420 [6:28:17<3:41:13,  1.49s/it]
 64%|██████▎   | 15522/24420 [6:28:18<3:41:16,  1.49s/it]
 64%|██████▎   | 15523/24420 [6:28:20<3:41:07,  1.49s/it]
 64%|██████▎   | 15524/24420 [6:28:21<3:40:51,  1.49s/it]
 64%|██████▎   | 15525/24420 [6:28:23<3:40:37,  1.49s/it]
                                                         
{'loss': 0.075, 'learning_rate': 5e-05, 'epoch': 3.18}

 64%|██████▎   | 15525/24420 [6:28:23<3:40:37,  1.49s/it]
 64%|██████▎   | 15526/24420 [6:28:24<3:41:00,  1.49s/it]
 64%|██████▎   | 15527/24420 [6:28:26<3:41:20,  1.49s/it]
 64%|██████▎   | 15528/24420 [6:28:27<3:41:23,  1.49s/it]
 64%|██████▎   | 15529/24420 [6:28:29<3:41:20,  1.49s/it]
 64%|██████▎   | 15530/24420 [6:28:30<3:40:46,  1.49s/it]
                                                         
{'loss': 0.0767, 'learning_rate': 5e-05, 'epoch': 3.18}

 64%|██████▎   | 15530/24420 [6:28:30<3:40:46,  1.49s/it]
 64%|██████▎   | 15531/24420 [6:28:32<3:40:35,  1.49s/it]
 64%|██████▎   | 15532/24420 [6:28:33<3:40:40,  1.49s/it]
 64%|██████▎   | 15533/24420 [6:28:35<3:40:41,  1.49s/it]
 64%|██████▎   | 15534/24420 [6:28:36<3:40:35,  1.49s/it]
 64%|██████▎   | 15535/24420 [6:28:38<3:40:32,  1.49s/it]
                                                         
{'loss': 0.0756, 'learning_rate': 5e-05, 'epoch': 3.18}

 64%|██████▎   | 15535/24420 [6:28:38<3:40:32,  1.49s/it]
 64%|██████▎   | 15536/24420 [6:28:39<3:40:57,  1.49s/it]
 64%|██████▎   | 15537/24420 [6:28:40<3:40:32,  1.49s/it]
 64%|██████▎   | 15538/24420 [6:28:42<3:40:39,  1.49s/it]
 64%|██████▎   | 15539/24420 [6:28:43<3:40:52,  1.49s/it]
 64%|██████▎   | 15540/24420 [6:28:45<3:40:32,  1.49s/it]
                                                         
{'loss': 0.072, 'learning_rate': 5e-05, 'epoch': 3.18}

 64%|██████▎   | 15540/24420 [6:28:45<3:40:32,  1.49s/it]
 64%|██████▎   | 15541/24420 [6:28:46<3:40:36,  1.49s/it]
 64%|██████▎   | 15542/24420 [6:28:48<3:40:43,  1.49s/it]
 64%|██████▎   | 15543/24420 [6:28:49<3:40:39,  1.49s/it]
 64%|██████▎   | 15544/24420 [6:28:51<3:40:07,  1.49s/it]
 64%|██████▎   | 15545/24420 [6:28:52<3:39:54,  1.49s/it]
                                                         
{'loss': 0.078, 'learning_rate': 5e-05, 'epoch': 3.18}

 64%|██████▎   | 15545/24420 [6:28:52<3:39:54,  1.49s/it]
 64%|██████▎   | 15546/24420 [6:28:54<3:40:02,  1.49s/it]
 64%|██████▎   | 15547/24420 [6:28:55<3:40:06,  1.49s/it]
 64%|██████▎   | 15548/24420 [6:28:57<3:39:51,  1.49s/it]
 64%|██████▎   | 15549/24420 [6:28:58<3:39:57,  1.49s/it]
 64%|██████▎   | 15550/24420 [6:29:00<3:39:41,  1.49s/it]
                                                         
{'loss': 0.0765, 'learning_rate': 5e-05, 'epoch': 3.18}

 64%|██████▎   | 15550/24420 [6:29:00<3:39:41,  1.49s/it]
 64%|██████▎   | 15551/24420 [6:29:01<3:39:52,  1.49s/it]
 64%|██████▎   | 15552/24420 [6:29:03<3:40:05,  1.49s/it]
 64%|██████▎   | 15553/24420 [6:29:04<3:40:17,  1.49s/it]
 64%|██████▎   | 15554/24420 [6:29:06<3:40:00,  1.49s/it]
 64%|██████▎   | 15555/24420 [6:29:07<3:40:09,  1.49s/it]
                                                         
{'loss': 0.0812, 'learning_rate': 5e-05, 'epoch': 3.18}

 64%|██████▎   | 15555/24420 [6:29:07<3:40:09,  1.49s/it]
 64%|██████▎   | 15556/24420 [6:29:09<3:40:11,  1.49s/it]
 64%|██████▎   | 15557/24420 [6:29:10<3:40:13,  1.49s/it]
 64%|██████▎   | 15558/24420 [6:29:12<3:40:01,  1.49s/it]
 64%|██████▎   | 15559/24420 [6:29:13<3:40:19,  1.49s/it]
 64%|██████▎   | 15560/24420 [6:29:15<3:39:46,  1.49s/it]
                                                         
{'loss': 0.0788, 'learning_rate': 5e-05, 'epoch': 3.19}

 64%|██████▎   | 15560/24420 [6:29:15<3:39:46,  1.49s/it]
 64%|██████▎   | 15561/24420 [6:29:16<3:40:03,  1.49s/it]
 64%|██████▎   | 15562/24420 [6:29:18<3:40:24,  1.49s/it]
 64%|██████▎   | 15563/24420 [6:29:19<3:40:15,  1.49s/it]
 64%|██████▎   | 15564/24420 [6:29:21<3:40:10,  1.49s/it]
 64%|██████▎   | 15565/24420 [6:29:22<3:40:23,  1.49s/it]
                                                         
{'loss': 0.0758, 'learning_rate': 5e-05, 'epoch': 3.19}

 64%|██████▎   | 15565/24420 [6:29:22<3:40:23,  1.49s/it]
 64%|██████▎   | 15566/24420 [6:29:24<3:40:35,  1.49s/it]
 64%|██████▎   | 15567/24420 [6:29:25<3:40:23,  1.49s/it]
 64%|██████▍   | 15568/24420 [6:29:27<3:40:47,  1.50s/it]
 64%|██████▍   | 15569/24420 [6:29:28<3:41:07,  1.50s/it]
 64%|██████▍   | 15570/24420 [6:29:30<3:41:07,  1.50s/it]
                                                         
{'loss': 0.1038, 'learning_rate': 5e-05, 'epoch': 3.19}

 64%|██████▍   | 15570/24420 [6:29:30<3:41:07,  1.50s/it]
 64%|██████▍   | 15571/24420 [6:29:31<3:41:19,  1.50s/it]
 64%|██████▍   | 15572/24420 [6:29:33<3:40:50,  1.50s/it]
 64%|██████▍   | 15573/24420 [6:29:34<3:40:44,  1.50s/it]
 64%|██████▍   | 15574/24420 [6:29:36<3:40:17,  1.49s/it]
 64%|██████▍   | 15575/24420 [6:29:37<3:40:50,  1.50s/it]
                                                         
{'loss': 0.092, 'learning_rate': 5e-05, 'epoch': 3.19}

 64%|██████▍   | 15575/24420 [6:29:37<3:40:50,  1.50s/it]
 64%|██████▍   | 15576/24420 [6:29:39<3:41:00,  1.50s/it]
 64%|██████▍   | 15577/24420 [6:29:40<3:40:27,  1.50s/it]
 64%|██████▍   | 15578/24420 [6:29:42<3:40:32,  1.50s/it]
 64%|██████▍   | 15579/24420 [6:29:43<3:41:06,  1.50s/it]
 64%|██████▍   | 15580/24420 [6:29:45<3:40:37,  1.50s/it]
                                                         
{'loss': 0.0754, 'learning_rate': 5e-05, 'epoch': 3.19}

 64%|██████▍   | 15580/24420 [6:29:45<3:40:37,  1.50s/it]
 64%|██████▍   | 15581/24420 [6:29:46<3:40:40,  1.50s/it]
 64%|██████▍   | 15582/24420 [6:29:48<3:40:57,  1.50s/it]
 64%|██████▍   | 15583/24420 [6:29:49<3:40:42,  1.50s/it]
 64%|██████▍   | 15584/24420 [6:29:51<3:40:35,  1.50s/it]
 64%|██████▍   | 15585/24420 [6:29:52<3:40:12,  1.50s/it]
                                                         
{'loss': 0.114, 'learning_rate': 5e-05, 'epoch': 3.19}

 64%|██████▍   | 15585/24420 [6:29:52<3:40:12,  1.50s/it]
 64%|██████▍   | 15586/24420 [6:29:54<3:39:52,  1.49s/it]
 64%|██████▍   | 15587/24420 [6:29:55<3:39:53,  1.49s/it]
 64%|██████▍   | 15588/24420 [6:29:57<3:39:36,  1.49s/it]
 64%|██████▍   | 15589/24420 [6:29:58<3:40:07,  1.50s/it]
 64%|██████▍   | 15590/24420 [6:30:00<3:40:01,  1.50s/it]
                                                         
{'loss': 0.0715, 'learning_rate': 5e-05, 'epoch': 3.19}

 64%|██████▍   | 15590/24420 [6:30:00<3:40:01,  1.50s/it]
 64%|██████▍   | 15591/24420 [6:30:01<3:40:17,  1.50s/it]
 64%|██████▍   | 15592/24420 [6:30:03<3:39:26,  1.49s/it]
 64%|██████▍   | 15593/24420 [6:30:04<3:39:37,  1.49s/it]
 64%|██████▍   | 15594/24420 [6:30:06<3:39:59,  1.50s/it]
 64%|██████▍   | 15595/24420 [6:30:07<3:40:05,  1.50s/it]
                                                         
{'loss': 0.0642, 'learning_rate': 5e-05, 'epoch': 3.19}

 64%|██████▍   | 15595/24420 [6:30:07<3:40:05,  1.50s/it]
 64%|██████▍   | 15596/24420 [6:30:09<3:39:48,  1.49s/it]
 64%|██████▍   | 15597/24420 [6:30:10<3:39:44,  1.49s/it]
 64%|██████▍   | 15598/24420 [6:30:12<3:39:53,  1.50s/it]
 64%|██████▍   | 15599/24420 [6:30:13<3:39:44,  1.49s/it]
 64%|██████▍   | 15600/24420 [6:30:15<3:39:34,  1.49s/it]
                                                         
{'loss': 0.0802, 'learning_rate': 5e-05, 'epoch': 3.19}

 64%|██████▍   | 15600/24420 [6:30:15<3:39:34,  1.49s/it]
 64%|██████▍   | 15601/24420 [6:30:19<5:56:31,  2.43s/it]
 64%|██████▍   | 15602/24420 [6:30:21<5:15:02,  2.14s/it]
 64%|██████▍   | 15603/24420 [6:30:22<4:45:56,  1.95s/it]
 64%|██████▍   | 15604/24420 [6:30:24<4:25:47,  1.81s/it]
 64%|██████▍   | 15605/24420 [6:30:25<4:11:30,  1.71s/it]
                                                         
{'loss': 0.0675, 'learning_rate': 5e-05, 'epoch': 3.2}

 64%|██████▍   | 15605/24420 [6:30:25<4:11:30,  1.71s/it]
 64%|██████▍   | 15606/24420 [6:30:27<4:01:51,  1.65s/it]
 64%|██████▍   | 15607/24420 [6:30:28<3:54:55,  1.60s/it]
 64%|██████▍   | 15608/24420 [6:30:30<3:49:56,  1.57s/it]
 64%|██████▍   | 15609/24420 [6:30:31<3:46:44,  1.54s/it]
 64%|██████▍   | 15610/24420 [6:30:33<3:44:08,  1.53s/it]
                                                         
{'loss': 0.0798, 'learning_rate': 5e-05, 'epoch': 3.2}

 64%|██████▍   | 15610/24420 [6:30:33<3:44:08,  1.53s/it]
 64%|██████▍   | 15611/24420 [6:30:34<3:42:27,  1.52s/it]
 64%|██████▍   | 15612/24420 [6:30:36<3:41:05,  1.51s/it]
 64%|██████▍   | 15613/24420 [6:30:37<3:40:22,  1.50s/it]
 64%|██████▍   | 15614/24420 [6:30:39<3:39:32,  1.50s/it]
 64%|██████▍   | 15615/24420 [6:30:40<3:39:14,  1.49s/it]
                                                         
{'loss': 0.0816, 'learning_rate': 5e-05, 'epoch': 3.2}

 64%|██████▍   | 15615/24420 [6:30:40<3:39:14,  1.49s/it]
 64%|██████▍   | 15616/24420 [6:30:42<3:39:13,  1.49s/it]
 64%|██████▍   | 15617/24420 [6:30:43<3:38:52,  1.49s/it]
 64%|██████▍   | 15618/24420 [6:30:44<3:38:30,  1.49s/it]
 64%|██████▍   | 15619/24420 [6:30:46<3:38:45,  1.49s/it]
 64%|██████▍   | 15620/24420 [6:30:47<3:38:56,  1.49s/it]
                                                         
{'loss': 0.0741, 'learning_rate': 5e-05, 'epoch': 3.2}

 64%|██████▍   | 15620/24420 [6:30:47<3:38:56,  1.49s/it]
 64%|██████▍   | 15621/24420 [6:30:49<3:38:48,  1.49s/it]
 64%|██████▍   | 15622/24420 [6:30:50<3:38:40,  1.49s/it]
 64%|██████▍   | 15623/24420 [6:30:52<3:38:25,  1.49s/it]
 64%|██████▍   | 15624/24420 [6:30:53<3:38:31,  1.49s/it]
 64%|██████▍   | 15625/24420 [6:30:55<3:38:08,  1.49s/it]
                                                         
{'loss': 0.0787, 'learning_rate': 5e-05, 'epoch': 3.2}

 64%|██████▍   | 15625/24420 [6:30:55<3:38:08,  1.49s/it]
 64%|██████▍   | 15626/24420 [6:30:56<3:38:20,  1.49s/it]
 64%|██████▍   | 15627/24420 [6:30:58<3:38:36,  1.49s/it]
 64%|██████▍   | 15628/24420 [6:30:59<3:38:16,  1.49s/it]
 64%|██████▍   | 15629/24420 [6:31:01<3:38:33,  1.49s/it]
 64%|██████▍   | 15630/24420 [6:31:02<3:38:08,  1.49s/it]
                                                         
{'loss': 0.0891, 'learning_rate': 5e-05, 'epoch': 3.2}

 64%|██████▍   | 15630/24420 [6:31:02<3:38:08,  1.49s/it]
 64%|██████▍   | 15631/24420 [6:31:04<3:38:09,  1.49s/it]
 64%|██████▍   | 15632/24420 [6:31:05<3:38:00,  1.49s/it]
 64%|██████▍   | 15633/24420 [6:31:07<3:37:38,  1.49s/it]
 64%|██████▍   | 15634/24420 [6:31:08<3:37:39,  1.49s/it]
 64%|██████▍   | 15635/24420 [6:31:10<3:37:46,  1.49s/it]
                                                         
{'loss': 0.0797, 'learning_rate': 5e-05, 'epoch': 3.2}

 64%|██████▍   | 15635/24420 [6:31:10<3:37:46,  1.49s/it]
 64%|██████▍   | 15636/24420 [6:31:11<3:38:19,  1.49s/it]
 64%|██████▍   | 15637/24420 [6:31:13<3:38:17,  1.49s/it]
 64%|██████▍   | 15638/24420 [6:31:14<3:38:43,  1.49s/it]
 64%|██████▍   | 15639/24420 [6:31:16<3:38:28,  1.49s/it]
 64%|██████▍   | 15640/24420 [6:31:17<3:38:12,  1.49s/it]
                                                         
{'loss': 0.0822, 'learning_rate': 5e-05, 'epoch': 3.2}

 64%|██████▍   | 15640/24420 [6:31:17<3:38:12,  1.49s/it]
 64%|██████▍   | 15641/24420 [6:31:19<3:38:28,  1.49s/it]
 64%|██████▍   | 15642/24420 [6:31:20<3:38:09,  1.49s/it]
 64%|██████▍   | 15643/24420 [6:31:22<3:38:15,  1.49s/it]
 64%|██████▍   | 15644/24420 [6:31:23<3:38:19,  1.49s/it]
 64%|██████▍   | 15645/24420 [6:31:25<3:38:24,  1.49s/it]
                                                         
{'loss': 0.0684, 'learning_rate': 5e-05, 'epoch': 3.2}

 64%|██████▍   | 15645/24420 [6:31:25<3:38:24,  1.49s/it]
 64%|██████▍   | 15646/24420 [6:31:26<3:38:22,  1.49s/it]
 64%|██████▍   | 15647/24420 [6:31:28<3:38:22,  1.49s/it]
 64%|██████▍   | 15648/24420 [6:31:29<3:38:16,  1.49s/it]
 64%|██████▍   | 15649/24420 [6:31:31<3:38:04,  1.49s/it]
 64%|██████▍   | 15650/24420 [6:31:32<3:38:04,  1.49s/it]
                                                         
{'loss': 0.0887, 'learning_rate': 5e-05, 'epoch': 3.2}

 64%|██████▍   | 15650/24420 [6:31:32<3:38:04,  1.49s/it]
 64%|██████▍   | 15651/24420 [6:31:34<3:38:26,  1.49s/it]
 64%|██████▍   | 15652/24420 [6:31:35<3:38:43,  1.50s/it]
 64%|██████▍   | 15653/24420 [6:31:37<3:38:26,  1.50s/it]
 64%|██████▍   | 15654/24420 [6:31:38<3:38:44,  1.50s/it]
 64%|██████▍   | 15655/24420 [6:31:40<3:38:39,  1.50s/it]
                                                         
{'loss': 0.0807, 'learning_rate': 5e-05, 'epoch': 3.21}

 64%|██████▍   | 15655/24420 [6:31:40<3:38:39,  1.50s/it]
 64%|██████▍   | 15656/24420 [6:31:41<3:38:36,  1.50s/it]
 64%|██████▍   | 15657/24420 [6:31:43<3:38:27,  1.50s/it]
 64%|██████▍   | 15658/24420 [6:31:44<3:38:05,  1.49s/it]
 64%|██████▍   | 15659/24420 [6:31:46<3:37:44,  1.49s/it]
 64%|██████▍   | 15660/24420 [6:31:47<3:37:02,  1.49s/it]
                                                         
{'loss': 0.0798, 'learning_rate': 5e-05, 'epoch': 3.21}

 64%|██████▍   | 15660/24420 [6:31:47<3:37:02,  1.49s/it]
 64%|██████▍   | 15661/24420 [6:31:49<3:36:59,  1.49s/it]
 64%|██████▍   | 15662/24420 [6:31:50<3:37:14,  1.49s/it]
 64%|██████▍   | 15663/24420 [6:31:52<3:37:29,  1.49s/it]
 64%|██████▍   | 15664/24420 [6:31:53<3:37:38,  1.49s/it]
 64%|██████▍   | 15665/24420 [6:31:55<3:37:21,  1.49s/it]
                                                         
{'loss': 0.0771, 'learning_rate': 5e-05, 'epoch': 3.21}

 64%|██████▍   | 15665/24420 [6:31:55<3:37:21,  1.49s/it]
 64%|██████▍   | 15666/24420 [6:31:56<3:37:11,  1.49s/it]
 64%|██████▍   | 15667/24420 [6:31:58<3:37:30,  1.49s/it]
 64%|██████▍   | 15668/24420 [6:31:59<3:37:34,  1.49s/it]
 64%|██████▍   | 15669/24420 [6:32:01<3:37:12,  1.49s/it]
 64%|██████▍   | 15670/24420 [6:32:02<3:36:55,  1.49s/it]
                                                         
{'loss': 0.0864, 'learning_rate': 5e-05, 'epoch': 3.21}

 64%|██████▍   | 15670/24420 [6:32:02<3:36:55,  1.49s/it]
 64%|██████▍   | 15671/24420 [6:32:04<3:37:43,  1.49s/it]
 64%|██████▍   | 15672/24420 [6:32:05<3:37:49,  1.49s/it]
 64%|██████▍   | 15673/24420 [6:32:07<3:37:24,  1.49s/it]
 64%|██████▍   | 15674/24420 [6:32:08<3:36:59,  1.49s/it]
 64%|██████▍   | 15675/24420 [6:32:09<3:36:42,  1.49s/it]
                                                         
{'loss': 0.0738, 'learning_rate': 5e-05, 'epoch': 3.21}

 64%|██████▍   | 15675/24420 [6:32:09<3:36:42,  1.49s/it]
 64%|██████▍   | 15676/24420 [6:32:11<3:37:06,  1.49s/it]
 64%|██████▍   | 15677/24420 [6:32:12<3:37:15,  1.49s/it]
 64%|██████▍   | 15678/24420 [6:32:14<3:37:24,  1.49s/it]
 64%|██████▍   | 15679/24420 [6:32:15<3:37:03,  1.49s/it]
 64%|██████▍   | 15680/24420 [6:32:17<3:36:46,  1.49s/it]
                                                         
{'loss': 0.0841, 'learning_rate': 5e-05, 'epoch': 3.21}

 64%|██████▍   | 15680/24420 [6:32:17<3:36:46,  1.49s/it]
 64%|██████▍   | 15681/24420 [6:32:18<3:37:03,  1.49s/it]
 64%|██████▍   | 15682/24420 [6:32:20<3:37:10,  1.49s/it]
 64%|██████▍   | 15683/24420 [6:32:21<3:36:55,  1.49s/it]
 64%|██████▍   | 15684/24420 [6:32:23<3:36:28,  1.49s/it]
 64%|██████▍   | 15685/24420 [6:32:24<3:36:24,  1.49s/it]
                                                         
{'loss': 0.0766, 'learning_rate': 5e-05, 'epoch': 3.21}

 64%|██████▍   | 15685/24420 [6:32:24<3:36:24,  1.49s/it]
 64%|██████▍   | 15686/24420 [6:32:26<3:36:46,  1.49s/it]
 64%|██████▍   | 15687/24420 [6:32:27<3:36:34,  1.49s/it]
 64%|██████▍   | 15688/24420 [6:32:29<3:36:39,  1.49s/it]
 64%|██████▍   | 15689/24420 [6:32:30<3:37:21,  1.49s/it]
 64%|██████▍   | 15690/24420 [6:32:32<3:36:56,  1.49s/it]
                                                         
{'loss': 0.0767, 'learning_rate': 5e-05, 'epoch': 3.21}

 64%|██████▍   | 15690/24420 [6:32:32<3:36:56,  1.49s/it]
 64%|██████▍   | 15691/24420 [6:32:33<3:36:42,  1.49s/it]
 64%|██████▍   | 15692/24420 [6:32:35<3:36:38,  1.49s/it]
 64%|██████▍   | 15693/24420 [6:32:36<3:36:17,  1.49s/it]
 64%|██████▍   | 15694/24420 [6:32:38<3:36:03,  1.49s/it]
 64%|██████▍   | 15695/24420 [6:32:39<3:36:13,  1.49s/it]
                                                         
{'loss': 0.0779, 'learning_rate': 5e-05, 'epoch': 3.21}

 64%|██████▍   | 15695/24420 [6:32:39<3:36:13,  1.49s/it]
 64%|██████▍   | 15696/24420 [6:32:41<3:36:13,  1.49s/it]
 64%|██████▍   | 15697/24420 [6:32:42<3:36:30,  1.49s/it]
 64%|██████▍   | 15698/24420 [6:32:44<3:36:39,  1.49s/it]
 64%|██████▍   | 15699/24420 [6:32:45<3:37:03,  1.49s/it]
 64%|██████▍   | 15700/24420 [6:32:47<3:36:44,  1.49s/it]
                                                         
{'loss': 0.0771, 'learning_rate': 5e-05, 'epoch': 3.21}

 64%|██████▍   | 15700/24420 [6:32:47<3:36:44,  1.49s/it]
 64%|██████▍   | 15701/24420 [6:32:48<3:37:01,  1.49s/it]
 64%|██████▍   | 15702/24420 [6:32:50<3:36:43,  1.49s/it]
 64%|██████▍   | 15703/24420 [6:32:51<3:36:44,  1.49s/it]
 64%|██████▍   | 15704/24420 [6:32:53<3:36:47,  1.49s/it]
 64%|██████▍   | 15705/24420 [6:32:54<3:36:42,  1.49s/it]
                                                         
{'loss': 0.0849, 'learning_rate': 5e-05, 'epoch': 3.22}

 64%|██████▍   | 15705/24420 [6:32:54<3:36:42,  1.49s/it]
 64%|██████▍   | 15706/24420 [6:32:56<3:36:47,  1.49s/it]
 64%|██████▍   | 15707/24420 [6:32:57<3:36:39,  1.49s/it]
 64%|██████▍   | 15708/24420 [6:32:59<3:36:30,  1.49s/it]
 64%|██████▍   | 15709/24420 [6:33:00<3:37:07,  1.50s/it]
 64%|██████▍   | 15710/24420 [6:33:02<3:36:55,  1.49s/it]
                                                         
{'loss': 0.0749, 'learning_rate': 5e-05, 'epoch': 3.22}

 64%|██████▍   | 15710/24420 [6:33:02<3:36:55,  1.49s/it]
 64%|██████▍   | 15711/24420 [6:33:03<3:36:32,  1.49s/it]
 64%|██████▍   | 15712/24420 [6:33:05<3:36:22,  1.49s/it]
 64%|██████▍   | 15713/24420 [6:33:06<3:36:16,  1.49s/it]
 64%|██████▍   | 15714/24420 [6:33:08<3:36:04,  1.49s/it]
 64%|██████▍   | 15715/24420 [6:33:09<3:36:04,  1.49s/it]
                                                         
{'loss': 0.071, 'learning_rate': 5e-05, 'epoch': 3.22}

 64%|██████▍   | 15715/24420 [6:33:09<3:36:04,  1.49s/it]
 64%|██████▍   | 15716/24420 [6:33:11<3:36:38,  1.49s/it]
 64%|██████▍   | 15717/24420 [6:33:12<3:36:44,  1.49s/it]
 64%|██████▍   | 15718/24420 [6:33:14<3:36:47,  1.49s/it]
 64%|██████▍   | 15719/24420 [6:33:15<3:36:55,  1.50s/it]
 64%|██████▍   | 15720/24420 [6:33:17<3:36:29,  1.49s/it]
                                                         
{'loss': 0.0809, 'learning_rate': 5e-05, 'epoch': 3.22}

 64%|██████▍   | 15720/24420 [6:33:17<3:36:29,  1.49s/it]
 64%|██████▍   | 15721/24420 [6:33:18<3:36:41,  1.49s/it]
 64%|██████▍   | 15722/24420 [6:33:20<3:36:13,  1.49s/it]
 64%|██████▍   | 15723/24420 [6:33:21<3:35:50,  1.49s/it]
 64%|██████▍   | 15724/24420 [6:33:23<3:36:12,  1.49s/it]
 64%|██████▍   | 15725/24420 [6:33:24<3:36:01,  1.49s/it]
                                                         
{'loss': 0.0907, 'learning_rate': 5e-05, 'epoch': 3.22}

 64%|██████▍   | 15725/24420 [6:33:24<3:36:01,  1.49s/it]
 64%|██████▍   | 15726/24420 [6:33:26<3:36:10,  1.49s/it]
 64%|██████▍   | 15727/24420 [6:33:27<3:36:33,  1.49s/it]
 64%|██████▍   | 15728/24420 [6:33:29<3:36:43,  1.50s/it]
 64%|██████▍   | 15729/24420 [6:33:30<3:36:21,  1.49s/it]
 64%|██████▍   | 15730/24420 [6:33:31<3:35:46,  1.49s/it]
                                                         
{'loss': 0.0812, 'learning_rate': 5e-05, 'epoch': 3.22}

 64%|██████▍   | 15730/24420 [6:33:31<3:35:46,  1.49s/it]
 64%|██████▍   | 15731/24420 [6:33:33<3:36:13,  1.49s/it]
 64%|██████▍   | 15732/24420 [6:33:34<3:36:11,  1.49s/it]
 64%|██████▍   | 15733/24420 [6:33:36<3:36:17,  1.49s/it]
 64%|██████▍   | 15734/24420 [6:33:37<3:36:22,  1.49s/it]
 64%|██████▍   | 15735/24420 [6:33:39<3:36:27,  1.50s/it]
                                                         
{'loss': 0.0892, 'learning_rate': 5e-05, 'epoch': 3.22}

 64%|██████▍   | 15735/24420 [6:33:39<3:36:27,  1.50s/it]
 64%|██████▍   | 15736/24420 [6:33:40<3:36:24,  1.50s/it]
 64%|██████▍   | 15737/24420 [6:33:42<3:36:09,  1.49s/it]
 64%|██████▍   | 15738/24420 [6:33:43<3:35:55,  1.49s/it]
 64%|██████▍   | 15739/24420 [6:33:45<3:35:34,  1.49s/it]
 64%|██████▍   | 15740/24420 [6:33:46<3:35:35,  1.49s/it]
                                                         
{'loss': 0.0738, 'learning_rate': 5e-05, 'epoch': 3.22}

 64%|██████▍   | 15740/24420 [6:33:46<3:35:35,  1.49s/it]
 64%|██████▍   | 15741/24420 [6:33:48<3:35:54,  1.49s/it]
 64%|██████▍   | 15742/24420 [6:33:49<3:36:28,  1.50s/it]
 64%|██████▍   | 15743/24420 [6:33:51<3:36:26,  1.50s/it]
 64%|██████▍   | 15744/24420 [6:33:52<3:36:25,  1.50s/it]
 64%|██████▍   | 15745/24420 [6:33:54<3:35:59,  1.49s/it]
                                                         
{'loss': 0.0734, 'learning_rate': 5e-05, 'epoch': 3.22}

 64%|██████▍   | 15745/24420 [6:33:54<3:35:59,  1.49s/it]
 64%|██████▍   | 15746/24420 [6:33:55<3:35:45,  1.49s/it]
 64%|██████▍   | 15747/24420 [6:33:57<3:35:28,  1.49s/it]
 64%|██████▍   | 15748/24420 [6:33:58<3:35:10,  1.49s/it]
 64%|██████▍   | 15749/24420 [6:34:00<3:35:40,  1.49s/it]
 64%|██████▍   | 15750/24420 [6:34:01<3:35:16,  1.49s/it]
                                                         
{'loss': 0.0774, 'learning_rate': 5e-05, 'epoch': 3.22}

 64%|██████▍   | 15750/24420 [6:34:01<3:35:16,  1.49s/it]
 65%|██████▍   | 15751/24420 [6:34:03<3:35:05,  1.49s/it]
 65%|██████▍   | 15752/24420 [6:34:04<3:34:53,  1.49s/it]
 65%|██████▍   | 15753/24420 [6:34:06<3:35:14,  1.49s/it]
 65%|██████▍   | 15754/24420 [6:34:07<3:34:52,  1.49s/it]
 65%|██████▍   | 15755/24420 [6:34:09<3:34:43,  1.49s/it]
                                                         
{'loss': 0.0818, 'learning_rate': 5e-05, 'epoch': 3.23}

 65%|██████▍   | 15755/24420 [6:34:09<3:34:43,  1.49s/it]
 65%|██████▍   | 15756/24420 [6:34:10<3:34:31,  1.49s/it]
 65%|██████▍   | 15757/24420 [6:34:12<3:34:18,  1.48s/it]
 65%|██████▍   | 15758/24420 [6:34:13<3:44:56,  1.56s/it]
 65%|██████▍   | 15759/24420 [6:34:15<3:41:46,  1.54s/it]
 65%|██████▍   | 15760/24420 [6:34:16<3:40:08,  1.53s/it]
                                                         
{'loss': 0.0761, 'learning_rate': 5e-05, 'epoch': 3.23}

 65%|██████▍   | 15760/24420 [6:34:16<3:40:08,  1.53s/it]
 65%|██████▍   | 15761/24420 [6:34:18<3:38:29,  1.51s/it]
 65%|██████▍   | 15762/24420 [6:34:19<3:37:12,  1.51s/it]
 65%|██████▍   | 15763/24420 [6:34:21<3:36:41,  1.50s/it]
 65%|██████▍   | 15764/24420 [6:34:22<3:35:52,  1.50s/it]
 65%|██████▍   | 15765/24420 [6:34:24<3:35:27,  1.49s/it]
                                                         
{'loss': 0.0794, 'learning_rate': 5e-05, 'epoch': 3.23}

 65%|██████▍   | 15765/24420 [6:34:24<3:35:27,  1.49s/it]
 65%|██████▍   | 15766/24420 [6:34:25<3:35:13,  1.49s/it]
 65%|██████▍   | 15767/24420 [6:34:27<3:35:21,  1.49s/it]
 65%|██████▍   | 15768/24420 [6:34:28<3:35:13,  1.49s/it]
 65%|██████▍   | 15769/24420 [6:34:30<3:35:28,  1.49s/it]
 65%|██████▍   | 15770/24420 [6:34:31<3:35:12,  1.49s/it]
                                                         
{'loss': 0.0818, 'learning_rate': 5e-05, 'epoch': 3.23}

 65%|██████▍   | 15770/24420 [6:34:31<3:35:12,  1.49s/it]
 65%|██████▍   | 15771/24420 [6:34:33<3:35:01,  1.49s/it]
 65%|██████▍   | 15772/24420 [6:34:34<3:35:03,  1.49s/it]
 65%|██████▍   | 15773/24420 [6:34:36<3:34:34,  1.49s/it]
 65%|██████▍   | 15774/24420 [6:34:37<3:34:56,  1.49s/it]
 65%|██████▍   | 15775/24420 [6:34:39<3:35:16,  1.49s/it]
                                                         
{'loss': 0.1054, 'learning_rate': 5e-05, 'epoch': 3.23}

 65%|██████▍   | 15775/24420 [6:34:39<3:35:16,  1.49s/it]
 65%|██████▍   | 15776/24420 [6:34:40<3:35:14,  1.49s/it]
 65%|██████▍   | 15777/24420 [6:34:42<3:35:27,  1.50s/it]
 65%|██████▍   | 15778/24420 [6:34:43<3:34:47,  1.49s/it]
 65%|██████▍   | 15779/24420 [6:34:45<3:34:58,  1.49s/it]
 65%|██████▍   | 15780/24420 [6:34:46<3:34:54,  1.49s/it]
                                                         
{'loss': 0.0793, 'learning_rate': 5e-05, 'epoch': 3.23}

 65%|██████▍   | 15780/24420 [6:34:46<3:34:54,  1.49s/it]
 65%|██████▍   | 15781/24420 [6:34:48<3:35:08,  1.49s/it]
 65%|██████▍   | 15782/24420 [6:34:49<3:34:59,  1.49s/it]
 65%|██████▍   | 15783/24420 [6:34:51<3:34:47,  1.49s/it]
 65%|██████▍   | 15784/24420 [6:34:52<3:34:33,  1.49s/it]
 65%|██████▍   | 15785/24420 [6:34:54<3:34:15,  1.49s/it]
                                                         
{'loss': 0.0821, 'learning_rate': 5e-05, 'epoch': 3.23}

 65%|██████▍   | 15785/24420 [6:34:54<3:34:15,  1.49s/it]
 65%|██████▍   | 15786/24420 [6:34:55<3:34:25,  1.49s/it]
 65%|██████▍   | 15787/24420 [6:34:57<3:34:05,  1.49s/it]
 65%|██████▍   | 15788/24420 [6:34:58<3:34:05,  1.49s/it]
 65%|██████▍   | 15789/24420 [6:35:00<3:34:21,  1.49s/it]
 65%|██████▍   | 15790/24420 [6:35:01<3:34:05,  1.49s/it]
                                                         
{'loss': 0.0793, 'learning_rate': 5e-05, 'epoch': 3.23}

 65%|██████▍   | 15790/24420 [6:35:01<3:34:05,  1.49s/it]
 65%|██████▍   | 15791/24420 [6:35:03<3:34:45,  1.49s/it]
 65%|██████▍   | 15792/24420 [6:35:04<3:34:05,  1.49s/it]
 65%|██████▍   | 15793/24420 [6:35:06<3:34:31,  1.49s/it]
 65%|██████▍   | 15794/24420 [6:35:07<3:34:21,  1.49s/it]
 65%|██████▍   | 15795/24420 [6:35:09<3:34:20,  1.49s/it]
                                                         
{'loss': 0.0819, 'learning_rate': 5e-05, 'epoch': 3.23}

 65%|██████▍   | 15795/24420 [6:35:09<3:34:20,  1.49s/it]
 65%|██████▍   | 15796/24420 [6:35:10<3:34:51,  1.49s/it]
 65%|██████▍   | 15797/24420 [6:35:12<3:34:41,  1.49s/it]
 65%|██████▍   | 15798/24420 [6:35:13<3:34:39,  1.49s/it]
 65%|██████▍   | 15799/24420 [6:35:15<3:35:07,  1.50s/it]
 65%|██████▍   | 15800/24420 [6:35:16<3:34:54,  1.50s/it]
                                                         
{'loss': 0.0788, 'learning_rate': 5e-05, 'epoch': 3.24}

 65%|██████▍   | 15800/24420 [6:35:16<3:34:54,  1.50s/it]
 65%|██████▍   | 15801/24420 [6:35:21<5:54:57,  2.47s/it]
 65%|██████▍   | 15802/24420 [6:35:22<5:12:25,  2.18s/it]
 65%|██████▍   | 15803/24420 [6:35:24<4:42:47,  1.97s/it]
 65%|██████▍   | 15804/24420 [6:35:25<4:22:28,  1.83s/it]
 65%|██████▍   | 15805/24420 [6:35:27<4:08:14,  1.73s/it]
                                                         
{'loss': 0.0851, 'learning_rate': 5e-05, 'epoch': 3.24}

 65%|██████▍   | 15805/24420 [6:35:27<4:08:14,  1.73s/it]
 65%|██████▍   | 15806/24420 [6:35:28<3:57:40,  1.66s/it]
 65%|██████▍   | 15807/24420 [6:35:30<3:50:30,  1.61s/it]
 65%|██████▍   | 15808/24420 [6:35:31<3:45:25,  1.57s/it]
 65%|██████▍   | 15809/24420 [6:35:33<3:41:43,  1.54s/it]
 65%|██████▍   | 15810/24420 [6:35:34<3:39:25,  1.53s/it]
                                                         
{'loss': 0.0587, 'learning_rate': 5e-05, 'epoch': 3.24}

 65%|██████▍   | 15810/24420 [6:35:34<3:39:25,  1.53s/it]
 65%|██████▍   | 15811/24420 [6:35:36<3:37:39,  1.52s/it]
 65%|██████▍   | 15812/24420 [6:35:37<3:36:18,  1.51s/it]
 65%|██████▍   | 15813/24420 [6:35:39<3:35:47,  1.50s/it]
 65%|██████▍   | 15814/24420 [6:35:40<3:35:28,  1.50s/it]
 65%|██████▍   | 15815/24420 [6:35:42<3:35:00,  1.50s/it]
                                                         
{'loss': 0.081, 'learning_rate': 5e-05, 'epoch': 3.24}

 65%|██████▍   | 15815/24420 [6:35:42<3:35:00,  1.50s/it]
 65%|██████▍   | 15816/24420 [6:35:43<3:35:04,  1.50s/it]
 65%|██████▍   | 15817/24420 [6:35:45<3:34:38,  1.50s/it]
 65%|██████▍   | 15818/24420 [6:35:46<3:34:16,  1.49s/it]
 65%|██████▍   | 15819/24420 [6:35:48<3:33:46,  1.49s/it]
 65%|██████▍   | 15820/24420 [6:35:49<3:33:55,  1.49s/it]
                                                         
{'loss': 0.0875, 'learning_rate': 5e-05, 'epoch': 3.24}

 65%|██████▍   | 15820/24420 [6:35:49<3:33:55,  1.49s/it]
 65%|██████▍   | 15821/24420 [6:35:51<3:33:49,  1.49s/it]
 65%|██████▍   | 15822/24420 [6:35:52<3:33:38,  1.49s/it]
 65%|██████▍   | 15823/24420 [6:35:54<3:33:22,  1.49s/it]
 65%|██████▍   | 15824/24420 [6:35:55<3:33:07,  1.49s/it]
 65%|██████▍   | 15825/24420 [6:35:57<3:33:55,  1.49s/it]
                                                         
{'loss': 0.0807, 'learning_rate': 5e-05, 'epoch': 3.24}

 65%|██████▍   | 15825/24420 [6:35:57<3:33:55,  1.49s/it]
 65%|██████▍   | 15826/24420 [6:35:58<3:33:43,  1.49s/it]
 65%|██████▍   | 15827/24420 [6:36:00<3:33:29,  1.49s/it]
 65%|██████▍   | 15828/24420 [6:36:01<3:33:26,  1.49s/it]
 65%|██████▍   | 15829/24420 [6:36:03<3:33:19,  1.49s/it]
 65%|██████▍   | 15830/24420 [6:36:04<3:33:18,  1.49s/it]
                                                         
{'loss': 0.0817, 'learning_rate': 5e-05, 'epoch': 3.24}

 65%|██████▍   | 15830/24420 [6:36:04<3:33:18,  1.49s/it]
 65%|██████▍   | 15831/24420 [6:36:06<3:33:22,  1.49s/it]
 65%|██████▍   | 15832/24420 [6:36:07<3:32:59,  1.49s/it]
 65%|██████▍   | 15833/24420 [6:36:09<3:33:05,  1.49s/it]
 65%|██████▍   | 15834/24420 [6:36:10<3:32:37,  1.49s/it]
 65%|██████▍   | 15835/24420 [6:36:12<3:32:27,  1.48s/it]
                                                         
{'loss': 0.0672, 'learning_rate': 5e-05, 'epoch': 3.24}

 65%|██████▍   | 15835/24420 [6:36:12<3:32:27,  1.48s/it]
 65%|██████▍   | 15836/24420 [6:36:13<3:32:39,  1.49s/it]
 65%|██████▍   | 15837/24420 [6:36:15<3:32:30,  1.49s/it]
 65%|██████▍   | 15838/24420 [6:36:16<3:32:47,  1.49s/it]
 65%|██████▍   | 15839/24420 [6:36:17<3:32:59,  1.49s/it]
 65%|██████▍   | 15840/24420 [6:36:19<3:33:00,  1.49s/it]
                                                         
{'loss': 0.0823, 'learning_rate': 5e-05, 'epoch': 3.24}

 65%|██████▍   | 15840/24420 [6:36:19<3:33:00,  1.49s/it]
 65%|██████▍   | 15841/24420 [6:36:20<3:32:58,  1.49s/it]
 65%|██████▍   | 15842/24420 [6:36:22<3:33:06,  1.49s/it]
 65%|██████▍   | 15843/24420 [6:36:23<3:33:03,  1.49s/it]
 65%|██████▍   | 15844/24420 [6:36:25<3:33:11,  1.49s/it]
 65%|██████▍   | 15845/24420 [6:36:26<3:32:47,  1.49s/it]
                                                         
{'loss': 0.0827, 'learning_rate': 5e-05, 'epoch': 3.24}

 65%|██████▍   | 15845/24420 [6:36:26<3:32:47,  1.49s/it]
 65%|██████▍   | 15846/24420 [6:36:28<3:32:55,  1.49s/it]
 65%|██████▍   | 15847/24420 [6:36:29<3:32:37,  1.49s/it]
 65%|██████▍   | 15848/24420 [6:36:31<3:32:30,  1.49s/it]
 65%|██████▍   | 15849/24420 [6:36:32<3:32:32,  1.49s/it]
 65%|██████▍   | 15850/24420 [6:36:34<3:32:42,  1.49s/it]
                                                         
{'loss': 0.084, 'learning_rate': 5e-05, 'epoch': 3.25}

 65%|██████▍   | 15850/24420 [6:36:34<3:32:42,  1.49s/it]
 65%|██████▍   | 15851/24420 [6:36:35<3:33:06,  1.49s/it]
 65%|██████▍   | 15852/24420 [6:36:37<3:33:36,  1.50s/it]
 65%|██████▍   | 15853/24420 [6:36:38<3:33:31,  1.50s/it]
 65%|██████▍   | 15854/24420 [6:36:40<3:33:29,  1.50s/it]
 65%|██████▍   | 15855/24420 [6:36:41<3:33:11,  1.49s/it]
                                                         
{'loss': 0.103, 'learning_rate': 5e-05, 'epoch': 3.25}

 65%|██████▍   | 15855/24420 [6:36:41<3:33:11,  1.49s/it]
 65%|██████▍   | 15856/24420 [6:36:43<3:33:11,  1.49s/it]
 65%|██████▍   | 15857/24420 [6:36:44<3:33:16,  1.49s/it]
 65%|██████▍   | 15858/24420 [6:36:46<3:33:28,  1.50s/it]
 65%|██████▍   | 15859/24420 [6:36:47<3:33:06,  1.49s/it]
 65%|██████▍   | 15860/24420 [6:36:49<3:32:55,  1.49s/it]
                                                         
{'loss': 0.0708, 'learning_rate': 5e-05, 'epoch': 3.25}

 65%|██████▍   | 15860/24420 [6:36:49<3:32:55,  1.49s/it]
 65%|██████▍   | 15861/24420 [6:36:50<3:32:43,  1.49s/it]
 65%|██████▍   | 15862/24420 [6:36:52<3:33:15,  1.50s/it]
 65%|██████▍   | 15863/24420 [6:36:53<3:33:03,  1.49s/it]
 65%|██████▍   | 15864/24420 [6:36:55<3:32:12,  1.49s/it]
 65%|██████▍   | 15865/24420 [6:36:56<3:31:49,  1.49s/it]
                                                         
{'loss': 0.0972, 'learning_rate': 5e-05, 'epoch': 3.25}

 65%|██████▍   | 15865/24420 [6:36:56<3:31:49,  1.49s/it]
 65%|██████▍   | 15866/24420 [6:36:58<3:31:57,  1.49s/it]
 65%|██████▍   | 15867/24420 [6:36:59<3:31:46,  1.49s/it]
 65%|██████▍   | 15868/24420 [6:37:01<3:32:02,  1.49s/it]
 65%|██████▍   | 15869/24420 [6:37:02<3:32:09,  1.49s/it]
 65%|██████▍   | 15870/24420 [6:37:04<3:32:09,  1.49s/it]
                                                         
{'loss': 0.078, 'learning_rate': 5e-05, 'epoch': 3.25}

 65%|██████▍   | 15870/24420 [6:37:04<3:32:09,  1.49s/it]
 65%|██████▍   | 15871/24420 [6:37:05<3:32:15,  1.49s/it]
 65%|██████▍   | 15872/24420 [6:37:07<3:32:00,  1.49s/it]
 65%|██████▌   | 15873/24420 [6:37:08<3:31:52,  1.49s/it]
 65%|██████▌   | 15874/24420 [6:37:10<3:32:01,  1.49s/it]
 65%|██████▌   | 15875/24420 [6:37:11<3:32:19,  1.49s/it]
                                                         
{'loss': 0.0906, 'learning_rate': 5e-05, 'epoch': 3.25}

 65%|██████▌   | 15875/24420 [6:37:11<3:32:19,  1.49s/it]
 65%|██████▌   | 15876/24420 [6:37:13<3:32:08,  1.49s/it]
 65%|██████▌   | 15877/24420 [6:37:14<3:31:43,  1.49s/it]
 65%|██████▌   | 15878/24420 [6:37:16<3:31:35,  1.49s/it]
 65%|██████▌   | 15879/24420 [6:37:17<3:31:17,  1.48s/it]
 65%|██████▌   | 15880/24420 [6:37:19<3:31:50,  1.49s/it]
                                                         
{'loss': 0.0766, 'learning_rate': 5e-05, 'epoch': 3.25}

 65%|██████▌   | 15880/24420 [6:37:19<3:31:50,  1.49s/it]
 65%|██████▌   | 15881/24420 [6:37:20<3:31:52,  1.49s/it]
 65%|██████▌   | 15882/24420 [6:37:22<3:31:50,  1.49s/it]
 65%|██████▌   | 15883/24420 [6:37:23<3:31:54,  1.49s/it]
 65%|██████▌   | 15884/24420 [6:37:25<3:31:47,  1.49s/it]
 65%|██████▌   | 15885/24420 [6:37:26<3:32:38,  1.49s/it]
                                                         
{'loss': 0.0939, 'learning_rate': 5e-05, 'epoch': 3.25}

 65%|██████▌   | 15885/24420 [6:37:26<3:32:38,  1.49s/it]
 65%|██████▌   | 15886/24420 [6:37:28<3:32:53,  1.50s/it]
 65%|██████▌   | 15887/24420 [6:37:29<3:32:38,  1.50s/it]
 65%|██████▌   | 15888/24420 [6:37:31<3:32:32,  1.49s/it]
 65%|██████▌   | 15889/24420 [6:37:32<3:32:12,  1.49s/it]
 65%|██████▌   | 15890/24420 [6:37:34<3:32:17,  1.49s/it]
                                                         
{'loss': 0.0843, 'learning_rate': 5e-05, 'epoch': 3.25}

 65%|██████▌   | 15890/24420 [6:37:34<3:32:17,  1.49s/it]
 65%|██████▌   | 15891/24420 [6:37:35<3:32:31,  1.50s/it]
 65%|██████▌   | 15892/24420 [6:37:37<3:32:20,  1.49s/it]
 65%|██████▌   | 15893/24420 [6:37:38<3:31:52,  1.49s/it]
 65%|██████▌   | 15894/24420 [6:37:39<3:32:11,  1.49s/it]
 65%|██████▌   | 15895/24420 [6:37:41<3:32:06,  1.49s/it]
                                                         
{'loss': 0.0764, 'learning_rate': 5e-05, 'epoch': 3.25}

 65%|██████▌   | 15895/24420 [6:37:41<3:32:06,  1.49s/it]
 65%|██████▌   | 15896/24420 [6:37:42<3:32:09,  1.49s/it]
 65%|██████▌   | 15897/24420 [6:37:44<3:32:03,  1.49s/it]
 65%|██████▌   | 15898/24420 [6:37:45<3:31:49,  1.49s/it]
 65%|██████▌   | 15899/24420 [6:37:47<3:31:54,  1.49s/it]
 65%|██████▌   | 15900/24420 [6:37:48<3:31:52,  1.49s/it]
                                                         
{'loss': 0.0676, 'learning_rate': 5e-05, 'epoch': 3.26}

 65%|██████▌   | 15900/24420 [6:37:48<3:31:52,  1.49s/it]
 65%|██████▌   | 15901/24420 [6:37:50<3:31:36,  1.49s/it]
 65%|██████▌   | 15902/24420 [6:37:51<3:31:19,  1.49s/it]
 65%|██████▌   | 15903/24420 [6:37:53<3:32:12,  1.49s/it]
 65%|██████▌   | 15904/24420 [6:37:54<3:32:13,  1.50s/it]
 65%|██████▌   | 15905/24420 [6:37:56<3:35:35,  1.52s/it]
                                                         
{'loss': 0.0913, 'learning_rate': 5e-05, 'epoch': 3.26}

 65%|██████▌   | 15905/24420 [6:37:56<3:35:35,  1.52s/it]
 65%|██████▌   | 15906/24420 [6:37:57<3:34:29,  1.51s/it]
 65%|██████▌   | 15907/24420 [6:37:59<3:34:00,  1.51s/it]
 65%|██████▌   | 15908/24420 [6:38:00<3:33:30,  1.50s/it]
 65%|██████▌   | 15909/24420 [6:38:02<3:33:33,  1.51s/it]
 65%|██████▌   | 15910/24420 [6:38:03<3:33:07,  1.50s/it]
                                                         
{'loss': 0.0845, 'learning_rate': 5e-05, 'epoch': 3.26}

 65%|██████▌   | 15910/24420 [6:38:03<3:33:07,  1.50s/it]
 65%|██████▌   | 15911/24420 [6:38:05<3:32:22,  1.50s/it]
 65%|██████▌   | 15912/24420 [6:38:06<3:31:53,  1.49s/it]
 65%|██████▌   | 15913/24420 [6:38:08<3:31:23,  1.49s/it]
 65%|██████▌   | 15914/24420 [6:38:09<3:31:03,  1.49s/it]
 65%|██████▌   | 15915/24420 [6:38:11<3:31:27,  1.49s/it]
                                                         
{'loss': 0.067, 'learning_rate': 5e-05, 'epoch': 3.26}

 65%|██████▌   | 15915/24420 [6:38:11<3:31:27,  1.49s/it]
 65%|██████▌   | 15916/24420 [6:38:12<3:31:36,  1.49s/it]
 65%|██████▌   | 15917/24420 [6:38:14<3:32:17,  1.50s/it]
 65%|██████▌   | 15918/24420 [6:38:15<3:31:44,  1.49s/it]
 65%|██████▌   | 15919/24420 [6:38:17<3:31:27,  1.49s/it]
 65%|██████▌   | 15920/24420 [6:38:18<3:31:11,  1.49s/it]
                                                         
{'loss': 0.0858, 'learning_rate': 5e-05, 'epoch': 3.26}

 65%|██████▌   | 15920/24420 [6:38:18<3:31:11,  1.49s/it]
 65%|██████▌   | 15921/24420 [6:38:20<3:30:57,  1.49s/it]
 65%|██████▌   | 15922/24420 [6:38:21<3:30:40,  1.49s/it]
 65%|██████▌   | 15923/24420 [6:38:23<3:30:19,  1.49s/it]
 65%|██████▌   | 15924/24420 [6:38:24<3:30:24,  1.49s/it]
 65%|██████▌   | 15925/24420 [6:38:26<3:30:37,  1.49s/it]
                                                         
{'loss': 0.0785, 'learning_rate': 5e-05, 'epoch': 3.26}

 65%|██████▌   | 15925/24420 [6:38:26<3:30:37,  1.49s/it]
 65%|██████▌   | 15926/24420 [6:38:27<3:30:36,  1.49s/it]
 65%|██████▌   | 15927/24420 [6:38:29<3:30:27,  1.49s/it]
 65%|██████▌   | 15928/24420 [6:38:30<3:30:32,  1.49s/it]
 65%|██████▌   | 15929/24420 [6:38:32<3:30:38,  1.49s/it]
 65%|██████▌   | 15930/24420 [6:38:33<3:31:06,  1.49s/it]
                                                         
{'loss': 0.0838, 'learning_rate': 5e-05, 'epoch': 3.26}

 65%|██████▌   | 15930/24420 [6:38:33<3:31:06,  1.49s/it]
 65%|██████▌   | 15931/24420 [6:38:35<3:31:15,  1.49s/it]
 65%|██████▌   | 15932/24420 [6:38:36<3:31:24,  1.49s/it]
 65%|██████▌   | 15933/24420 [6:38:38<3:30:55,  1.49s/it]
 65%|██████▌   | 15934/24420 [6:38:39<3:31:05,  1.49s/it]
 65%|██████▌   | 15935/24420 [6:38:41<3:31:06,  1.49s/it]
                                                         
{'loss': 0.0843, 'learning_rate': 5e-05, 'epoch': 3.26}

 65%|██████▌   | 15935/24420 [6:38:41<3:31:06,  1.49s/it]
 65%|██████▌   | 15936/24420 [6:38:42<3:31:20,  1.49s/it]
 65%|██████▌   | 15937/24420 [6:38:44<3:31:23,  1.50s/it]
 65%|██████▌   | 15938/24420 [6:38:45<3:31:08,  1.49s/it]
 65%|██████▌   | 15939/24420 [6:38:47<3:31:05,  1.49s/it]
 65%|██████▌   | 15940/24420 [6:38:48<3:30:40,  1.49s/it]
                                                         
{'loss': 0.0809, 'learning_rate': 5e-05, 'epoch': 3.26}

 65%|██████▌   | 15940/24420 [6:38:48<3:30:40,  1.49s/it]
 65%|██████▌   | 15941/24420 [6:38:50<3:31:09,  1.49s/it]
 65%|██████▌   | 15942/24420 [6:38:51<3:30:51,  1.49s/it]
 65%|██████▌   | 15943/24420 [6:38:53<3:30:58,  1.49s/it]
 65%|██████▌   | 15944/24420 [6:38:54<3:30:30,  1.49s/it]
 65%|██████▌   | 15945/24420 [6:38:56<3:30:23,  1.49s/it]
                                                         
{'loss': 0.0828, 'learning_rate': 5e-05, 'epoch': 3.26}

 65%|██████▌   | 15945/24420 [6:38:56<3:30:23,  1.49s/it]
 65%|██████▌   | 15946/24420 [6:38:57<3:30:29,  1.49s/it]
 65%|██████▌   | 15947/24420 [6:38:59<3:30:11,  1.49s/it]
 65%|██████▌   | 15948/24420 [6:39:00<3:30:02,  1.49s/it]
 65%|██████▌   | 15949/24420 [6:39:02<3:30:28,  1.49s/it]
 65%|██████▌   | 15950/24420 [6:39:03<3:30:38,  1.49s/it]
                                                         
{'loss': 0.0785, 'learning_rate': 5e-05, 'epoch': 3.27}

 65%|██████▌   | 15950/24420 [6:39:03<3:30:38,  1.49s/it]
 65%|██████▌   | 15951/24420 [6:39:05<3:30:35,  1.49s/it]
 65%|██████▌   | 15952/24420 [6:39:06<3:30:15,  1.49s/it]
 65%|██████▌   | 15953/24420 [6:39:08<3:29:58,  1.49s/it]
 65%|██████▌   | 15954/24420 [6:39:09<3:30:10,  1.49s/it]
 65%|██████▌   | 15955/24420 [6:39:11<3:30:09,  1.49s/it]
                                                         
{'loss': 0.0816, 'learning_rate': 5e-05, 'epoch': 3.27}

 65%|██████▌   | 15955/24420 [6:39:11<3:30:09,  1.49s/it]
 65%|██████▌   | 15956/24420 [6:39:12<3:30:32,  1.49s/it]
 65%|██████▌   | 15957/24420 [6:39:14<3:29:57,  1.49s/it]
 65%|██████▌   | 15958/24420 [6:39:15<3:30:07,  1.49s/it]
 65%|██████▌   | 15959/24420 [6:39:17<3:30:07,  1.49s/it]
 65%|██████▌   | 15960/24420 [6:39:18<3:30:04,  1.49s/it]
                                                         
{'loss': 0.0694, 'learning_rate': 5e-05, 'epoch': 3.27}

 65%|██████▌   | 15960/24420 [6:39:18<3:30:04,  1.49s/it]
 65%|██████▌   | 15961/24420 [6:39:19<3:29:56,  1.49s/it]
 65%|██████▌   | 15962/24420 [6:39:21<3:29:41,  1.49s/it]
 65%|██████▌   | 15963/24420 [6:39:22<3:29:31,  1.49s/it]
 65%|██████▌   | 15964/24420 [6:39:24<3:29:33,  1.49s/it]
 65%|██████▌   | 15965/24420 [6:39:25<3:29:32,  1.49s/it]
                                                         
{'loss': 0.0808, 'learning_rate': 5e-05, 'epoch': 3.27}

 65%|██████▌   | 15965/24420 [6:39:25<3:29:32,  1.49s/it]
 65%|██████▌   | 15966/24420 [6:39:27<3:29:49,  1.49s/it]
 65%|██████▌   | 15967/24420 [6:39:28<3:29:55,  1.49s/it]
 65%|██████▌   | 15968/24420 [6:39:30<3:29:54,  1.49s/it]
 65%|██████▌   | 15969/24420 [6:39:31<3:30:26,  1.49s/it]
 65%|██████▌   | 15970/24420 [6:39:33<3:30:44,  1.50s/it]
                                                         
{'loss': 0.0706, 'learning_rate': 5e-05, 'epoch': 3.27}

 65%|██████▌   | 15970/24420 [6:39:33<3:30:44,  1.50s/it]
 65%|██████▌   | 15971/24420 [6:39:34<3:31:18,  1.50s/it]
 65%|██████▌   | 15972/24420 [6:39:36<3:30:43,  1.50s/it]
 65%|██████▌   | 15973/24420 [6:39:37<3:30:29,  1.50s/it]
 65%|██████▌   | 15974/24420 [6:39:39<3:29:59,  1.49s/it]
 65%|██████▌   | 15975/24420 [6:39:40<3:30:23,  1.49s/it]
                                                         
{'loss': 0.0946, 'learning_rate': 5e-05, 'epoch': 3.27}

 65%|██████▌   | 15975/24420 [6:39:40<3:30:23,  1.49s/it]
 65%|██████▌   | 15976/24420 [6:39:42<3:30:42,  1.50s/it]
 65%|██████▌   | 15977/24420 [6:39:43<3:30:31,  1.50s/it]
 65%|██████▌   | 15978/24420 [6:39:45<3:30:50,  1.50s/it]
 65%|██████▌   | 15979/24420 [6:39:46<3:30:18,  1.49s/it]
 65%|██████▌   | 15980/24420 [6:39:48<3:30:22,  1.50s/it]
                                                         
{'loss': 0.077, 'learning_rate': 5e-05, 'epoch': 3.27}

 65%|██████▌   | 15980/24420 [6:39:48<3:30:22,  1.50s/it]
 65%|██████▌   | 15981/24420 [6:39:49<3:30:12,  1.49s/it]
 65%|██████▌   | 15982/24420 [6:39:51<3:30:27,  1.50s/it]
 65%|██████▌   | 15983/24420 [6:39:52<3:29:54,  1.49s/it]
 65%|██████▌   | 15984/24420 [6:39:54<3:29:43,  1.49s/it]
 65%|██████▌   | 15985/24420 [6:39:55<3:29:31,  1.49s/it]
                                                         
{'loss': 0.0881, 'learning_rate': 5e-05, 'epoch': 3.27}

 65%|██████▌   | 15985/24420 [6:39:55<3:29:31,  1.49s/it]
 65%|██████▌   | 15986/24420 [6:39:57<3:29:49,  1.49s/it]
 65%|██████▌   | 15987/24420 [6:39:58<3:29:44,  1.49s/it]
 65%|██████▌   | 15988/24420 [6:40:00<3:29:23,  1.49s/it]
 65%|██████▌   | 15989/24420 [6:40:01<3:29:24,  1.49s/it]
 65%|██████▌   | 15990/24420 [6:40:03<3:29:30,  1.49s/it]
                                                         
{'loss': 0.0836, 'learning_rate': 5e-05, 'epoch': 3.27}

 65%|██████▌   | 15990/24420 [6:40:03<3:29:30,  1.49s/it]
 65%|██████▌   | 15991/24420 [6:40:04<3:29:38,  1.49s/it]
 65%|██████▌   | 15992/24420 [6:40:06<3:29:35,  1.49s/it]
 65%|██████▌   | 15993/24420 [6:40:07<3:29:25,  1.49s/it]
 65%|██████▌   | 15994/24420 [6:40:09<3:29:05,  1.49s/it]
 65%|██████▌   | 15995/24420 [6:40:10<3:29:14,  1.49s/it]
                                                         
{'loss': 0.0823, 'learning_rate': 5e-05, 'epoch': 3.27}

 65%|██████▌   | 15995/24420 [6:40:10<3:29:14,  1.49s/it]
 66%|██████▌   | 15996/24420 [6:40:12<3:29:10,  1.49s/it]
 66%|██████▌   | 15997/24420 [6:40:13<3:28:51,  1.49s/it]
 66%|██████▌   | 15998/24420 [6:40:15<3:28:54,  1.49s/it]
 66%|██████▌   | 15999/24420 [6:40:16<3:29:02,  1.49s/it]
 66%|██████▌   | 16000/24420 [6:40:18<3:28:48,  1.49s/it]
                                                         
{'loss': 0.0727, 'learning_rate': 5e-05, 'epoch': 3.28}

 66%|██████▌   | 16000/24420 [6:40:18<3:28:48,  1.49s/it]
 66%|██████▌   | 16001/24420 [6:40:22<5:38:02,  2.41s/it]
 66%|██████▌   | 16002/24420 [6:40:24<4:59:05,  2.13s/it]
 66%|██████▌   | 16003/24420 [6:40:25<4:32:00,  1.94s/it]
 66%|██████▌   | 16004/24420 [6:40:27<4:13:28,  1.81s/it]
 66%|██████▌   | 16005/24420 [6:40:28<3:59:57,  1.71s/it]
                                                         
{'loss': 0.0778, 'learning_rate': 5e-05, 'epoch': 3.28}

 66%|██████▌   | 16005/24420 [6:40:28<3:59:57,  1.71s/it]
 66%|██████▌   | 16006/24420 [6:40:30<3:51:11,  1.65s/it]
 66%|██████▌   | 16007/24420 [6:40:31<3:44:19,  1.60s/it]
 66%|██████▌   | 16008/24420 [6:40:33<3:39:34,  1.57s/it]
 66%|██████▌   | 16009/24420 [6:40:34<3:36:20,  1.54s/it]
 66%|██████▌   | 16010/24420 [6:40:36<3:34:03,  1.53s/it]
                                                         
{'loss': 0.0753, 'learning_rate': 5e-05, 'epoch': 3.28}

 66%|██████▌   | 16010/24420 [6:40:36<3:34:03,  1.53s/it]
 66%|██████▌   | 16011/24420 [6:40:37<3:32:23,  1.52s/it]
 66%|██████▌   | 16012/24420 [6:40:39<3:31:14,  1.51s/it]
 66%|██████▌   | 16013/24420 [6:40:40<3:30:37,  1.50s/it]
 66%|██████▌   | 16014/24420 [6:40:42<3:29:40,  1.50s/it]
 66%|██████▌   | 16015/24420 [6:40:43<3:29:43,  1.50s/it]
                                                         
{'loss': 0.0839, 'learning_rate': 5e-05, 'epoch': 3.28}

 66%|██████▌   | 16015/24420 [6:40:43<3:29:43,  1.50s/it]
 66%|██████▌   | 16016/24420 [6:40:45<3:29:47,  1.50s/it]
 66%|██████▌   | 16017/24420 [6:40:46<3:30:11,  1.50s/it]
 66%|██████▌   | 16018/24420 [6:40:48<3:29:45,  1.50s/it]
 66%|██████▌   | 16019/24420 [6:40:49<3:29:29,  1.50s/it]
 66%|██████▌   | 16020/24420 [6:40:51<3:29:40,  1.50s/it]
                                                         
{'loss': 0.0832, 'learning_rate': 5e-05, 'epoch': 3.28}

 66%|██████▌   | 16020/24420 [6:40:51<3:29:40,  1.50s/it]
 66%|██████▌   | 16021/24420 [6:40:52<3:29:23,  1.50s/it]
 66%|██████▌   | 16022/24420 [6:40:54<3:29:07,  1.49s/it]
 66%|██████▌   | 16023/24420 [6:40:55<3:28:58,  1.49s/it]
 66%|██████▌   | 16024/24420 [6:40:57<3:28:32,  1.49s/it]
 66%|██████▌   | 16025/24420 [6:40:58<3:28:31,  1.49s/it]
                                                         
{'loss': 0.0721, 'learning_rate': 5e-05, 'epoch': 3.28}

 66%|██████▌   | 16025/24420 [6:40:58<3:28:31,  1.49s/it]
 66%|██████▌   | 16026/24420 [6:41:00<3:28:42,  1.49s/it]
 66%|██████▌   | 16027/24420 [6:41:01<3:28:34,  1.49s/it]
 66%|██████▌   | 16028/24420 [6:41:03<3:29:03,  1.49s/it]
 66%|██████▌   | 16029/24420 [6:41:04<3:28:56,  1.49s/it]
 66%|██████▌   | 16030/24420 [6:41:06<3:28:35,  1.49s/it]
                                                         
{'loss': 0.0723, 'learning_rate': 5e-05, 'epoch': 3.28}

 66%|██████▌   | 16030/24420 [6:41:06<3:28:35,  1.49s/it]
 66%|██████▌   | 16031/24420 [6:41:07<3:28:32,  1.49s/it]
 66%|██████▌   | 16032/24420 [6:41:09<3:29:17,  1.50s/it]
 66%|██████▌   | 16033/24420 [6:41:10<3:29:17,  1.50s/it]
 66%|██████▌   | 16034/24420 [6:41:12<3:29:19,  1.50s/it]
 66%|██████▌   | 16035/24420 [6:41:13<3:29:06,  1.50s/it]
                                                         
{'loss': 0.0871, 'learning_rate': 5e-05, 'epoch': 3.28}

 66%|██████▌   | 16035/24420 [6:41:13<3:29:06,  1.50s/it]
 66%|██████▌   | 16036/24420 [6:41:14<3:29:00,  1.50s/it]
 66%|██████▌   | 16037/24420 [6:41:16<3:28:41,  1.49s/it]
 66%|██████▌   | 16038/24420 [6:41:17<3:29:18,  1.50s/it]
 66%|██████▌   | 16039/24420 [6:41:19<3:28:34,  1.49s/it]
 66%|██████▌   | 16040/24420 [6:41:20<3:28:44,  1.49s/it]
                                                         
{'loss': 0.0635, 'learning_rate': 5e-05, 'epoch': 3.28}

 66%|██████▌   | 16040/24420 [6:41:20<3:28:44,  1.49s/it]
 66%|██████▌   | 16041/24420 [6:41:22<3:29:47,  1.50s/it]
 66%|██████▌   | 16042/24420 [6:41:23<3:29:16,  1.50s/it]
 66%|██████▌   | 16043/24420 [6:41:25<3:29:10,  1.50s/it]
 66%|██████▌   | 16044/24420 [6:41:26<3:29:27,  1.50s/it]
 66%|██████▌   | 16045/24420 [6:41:28<3:29:05,  1.50s/it]
                                                         
{'loss': 0.081, 'learning_rate': 5e-05, 'epoch': 3.29}

 66%|██████▌   | 16045/24420 [6:41:28<3:29:05,  1.50s/it]
 66%|██████▌   | 16046/24420 [6:41:29<3:29:39,  1.50s/it]
 66%|██████▌   | 16047/24420 [6:41:31<3:29:31,  1.50s/it]
 66%|██████▌   | 16048/24420 [6:41:32<3:29:01,  1.50s/it]
 66%|██████▌   | 16049/24420 [6:41:34<3:28:39,  1.50s/it]
 66%|██████▌   | 16050/24420 [6:41:35<3:28:05,  1.49s/it]
                                                         
{'loss': 0.0698, 'learning_rate': 5e-05, 'epoch': 3.29}

 66%|██████▌   | 16050/24420 [6:41:35<3:28:05,  1.49s/it]
 66%|██████▌   | 16051/24420 [6:41:37<3:28:16,  1.49s/it]
 66%|██████▌   | 16052/24420 [6:41:38<3:27:50,  1.49s/it]
 66%|██████▌   | 16053/24420 [6:41:40<3:28:35,  1.50s/it]
 66%|██████▌   | 16054/24420 [6:41:41<3:27:55,  1.49s/it]
 66%|██████▌   | 16055/24420 [6:41:43<3:27:58,  1.49s/it]
                                                         
{'loss': 0.0841, 'learning_rate': 5e-05, 'epoch': 3.29}

 66%|██████▌   | 16055/24420 [6:41:43<3:27:58,  1.49s/it]
 66%|██████▌   | 16056/24420 [6:41:44<3:28:12,  1.49s/it]
 66%|██████▌   | 16057/24420 [6:41:46<3:28:19,  1.49s/it]
 66%|██████▌   | 16058/24420 [6:41:47<3:27:53,  1.49s/it]
 66%|██████▌   | 16059/24420 [6:41:49<3:27:50,  1.49s/it]
 66%|██████▌   | 16060/24420 [6:41:50<3:28:38,  1.50s/it]
                                                         
{'loss': 0.0976, 'learning_rate': 5e-05, 'epoch': 3.29}

 66%|██████▌   | 16060/24420 [6:41:50<3:28:38,  1.50s/it]
 66%|██████▌   | 16061/24420 [6:41:52<3:28:39,  1.50s/it]
 66%|██████▌   | 16062/24420 [6:41:53<3:28:11,  1.49s/it]
 66%|██████▌   | 16063/24420 [6:41:55<3:28:48,  1.50s/it]
 66%|██████▌   | 16064/24420 [6:41:56<3:28:18,  1.50s/it]
 66%|██████▌   | 16065/24420 [6:41:58<3:28:12,  1.50s/it]
                                                         
{'loss': 0.0786, 'learning_rate': 5e-05, 'epoch': 3.29}

 66%|██████▌   | 16065/24420 [6:41:58<3:28:12,  1.50s/it]
 66%|██████▌   | 16066/24420 [6:41:59<3:27:58,  1.49s/it]
 66%|██████▌   | 16067/24420 [6:42:01<3:28:25,  1.50s/it]
 66%|██████▌   | 16068/24420 [6:42:02<3:27:58,  1.49s/it]
 66%|██████▌   | 16069/24420 [6:42:04<3:27:51,  1.49s/it]
 66%|██████▌   | 16070/24420 [6:42:05<3:27:29,  1.49s/it]
                                                         
{'loss': 0.0822, 'learning_rate': 5e-05, 'epoch': 3.29}

 66%|██████▌   | 16070/24420 [6:42:05<3:27:29,  1.49s/it]
 66%|██████▌   | 16071/24420 [6:42:07<3:27:29,  1.49s/it]
 66%|██████▌   | 16072/24420 [6:42:08<3:27:31,  1.49s/it]
 66%|██████▌   | 16073/24420 [6:42:10<3:27:08,  1.49s/it]
 66%|██████▌   | 16074/24420 [6:42:11<3:27:12,  1.49s/it]
 66%|██████▌   | 16075/24420 [6:42:13<3:27:12,  1.49s/it]
                                                         
{'loss': 0.089, 'learning_rate': 5e-05, 'epoch': 3.29}

 66%|██████▌   | 16075/24420 [6:42:13<3:27:12,  1.49s/it]
 66%|██████▌   | 16076/24420 [6:42:14<3:27:25,  1.49s/it]
 66%|██████▌   | 16077/24420 [6:42:16<3:27:04,  1.49s/it]
 66%|██████▌   | 16078/24420 [6:42:17<3:27:11,  1.49s/it]
 66%|██████▌   | 16079/24420 [6:42:19<3:27:13,  1.49s/it]
 66%|██████▌   | 16080/24420 [6:42:20<3:27:09,  1.49s/it]
                                                         
{'loss': 0.0802, 'learning_rate': 5e-05, 'epoch': 3.29}

 66%|██████▌   | 16080/24420 [6:42:20<3:27:09,  1.49s/it]
 66%|██████▌   | 16081/24420 [6:42:22<3:27:42,  1.49s/it]
 66%|██████▌   | 16082/24420 [6:42:23<3:27:35,  1.49s/it]
 66%|██████▌   | 16083/24420 [6:42:25<3:27:17,  1.49s/it]
 66%|██████▌   | 16084/24420 [6:42:26<3:27:37,  1.49s/it]
 66%|██████▌   | 16085/24420 [6:42:28<3:27:31,  1.49s/it]
                                                         
{'loss': 0.0855, 'learning_rate': 5e-05, 'epoch': 3.29}

 66%|██████▌   | 16085/24420 [6:42:28<3:27:31,  1.49s/it]
 66%|██████▌   | 16086/24420 [6:42:29<3:27:27,  1.49s/it]
 66%|██████▌   | 16087/24420 [6:42:31<3:27:03,  1.49s/it]
 66%|██████▌   | 16088/24420 [6:42:32<3:26:53,  1.49s/it]
 66%|██████▌   | 16089/24420 [6:42:34<3:27:03,  1.49s/it]
 66%|██████▌   | 16090/24420 [6:42:35<3:26:55,  1.49s/it]
                                                         
{'loss': 0.0781, 'learning_rate': 5e-05, 'epoch': 3.29}

 66%|██████▌   | 16090/24420 [6:42:35<3:26:55,  1.49s/it]
 66%|██████▌   | 16091/24420 [6:42:37<3:26:51,  1.49s/it]
 66%|██████▌   | 16092/24420 [6:42:38<3:26:50,  1.49s/it]
 66%|██████▌   | 16093/24420 [6:42:40<3:27:05,  1.49s/it]
 66%|██████▌   | 16094/24420 [6:42:41<3:27:23,  1.49s/it]
 66%|██████▌   | 16095/24420 [6:42:43<3:27:05,  1.49s/it]
                                                         
{'loss': 0.0767, 'learning_rate': 5e-05, 'epoch': 3.3}

 66%|██████▌   | 16095/24420 [6:42:43<3:27:05,  1.49s/it]
 66%|██████▌   | 16096/24420 [6:42:44<3:27:28,  1.50s/it]
 66%|██████▌   | 16097/24420 [6:42:46<3:27:18,  1.49s/it]
 66%|██████▌   | 16098/24420 [6:42:47<3:26:56,  1.49s/it]
 66%|██████▌   | 16099/24420 [6:42:49<3:26:28,  1.49s/it]
 66%|██████▌   | 16100/24420 [6:42:50<3:26:39,  1.49s/it]
                                                         
{'loss': 0.0776, 'learning_rate': 5e-05, 'epoch': 3.3}

 66%|██████▌   | 16100/24420 [6:42:50<3:26:39,  1.49s/it]
 66%|██████▌   | 16101/24420 [6:42:52<3:26:29,  1.49s/it]
 66%|██████▌   | 16102/24420 [6:42:53<3:26:29,  1.49s/it]
 66%|██████▌   | 16103/24420 [6:42:55<3:26:15,  1.49s/it]
 66%|██████▌   | 16104/24420 [6:42:56<3:25:48,  1.48s/it]
 66%|██████▌   | 16105/24420 [6:42:58<3:25:49,  1.49s/it]
                                                         
{'loss': 0.089, 'learning_rate': 5e-05, 'epoch': 3.3}

 66%|██████▌   | 16105/24420 [6:42:58<3:25:49,  1.49s/it]
 66%|██████▌   | 16106/24420 [6:42:59<3:26:02,  1.49s/it]
 66%|██████▌   | 16107/24420 [6:43:00<3:25:53,  1.49s/it]
 66%|██████▌   | 16108/24420 [6:43:02<3:25:34,  1.48s/it]
 66%|██████▌   | 16109/24420 [6:43:03<3:25:33,  1.48s/it]
 66%|██████▌   | 16110/24420 [6:43:05<3:25:44,  1.49s/it]
                                                         
{'loss': 0.0995, 'learning_rate': 5e-05, 'epoch': 3.3}

 66%|██████▌   | 16110/24420 [6:43:05<3:25:44,  1.49s/it]
 66%|██████▌   | 16111/24420 [6:43:06<3:25:56,  1.49s/it]
 66%|██████▌   | 16112/24420 [6:43:08<3:25:30,  1.48s/it]
 66%|██████▌   | 16113/24420 [6:43:10<3:34:34,  1.55s/it]
 66%|██████▌   | 16114/24420 [6:43:11<3:31:51,  1.53s/it]
 66%|██████▌   | 16115/24420 [6:43:13<3:30:03,  1.52s/it]
                                                         
{'loss': 0.0763, 'learning_rate': 5e-05, 'epoch': 3.3}

 66%|██████▌   | 16115/24420 [6:43:13<3:30:03,  1.52s/it]
 66%|██████▌   | 16116/24420 [6:43:14<3:28:47,  1.51s/it]
 66%|██████▌   | 16117/24420 [6:43:16<3:28:01,  1.50s/it]
 66%|██████▌   | 16118/24420 [6:43:17<3:27:22,  1.50s/it]
 66%|██████▌   | 16119/24420 [6:43:19<3:26:58,  1.50s/it]
 66%|██████▌   | 16120/24420 [6:43:20<3:27:26,  1.50s/it]
                                                         
{'loss': 0.0745, 'learning_rate': 5e-05, 'epoch': 3.3}

 66%|██████▌   | 16120/24420 [6:43:20<3:27:26,  1.50s/it]
 66%|██████▌   | 16121/24420 [6:43:22<3:27:17,  1.50s/it]
 66%|██████▌   | 16122/24420 [6:43:23<3:26:52,  1.50s/it]
 66%|██████▌   | 16123/24420 [6:43:25<3:26:40,  1.49s/it]
 66%|██████▌   | 16124/24420 [6:43:26<3:26:35,  1.49s/it]
 66%|██████▌   | 16125/24420 [6:43:27<3:26:09,  1.49s/it]
                                                         
{'loss': 0.0875, 'learning_rate': 5e-05, 'epoch': 3.3}

 66%|██████▌   | 16125/24420 [6:43:28<3:26:09,  1.49s/it]
 66%|██████▌   | 16126/24420 [6:43:29<3:25:52,  1.49s/it]
 66%|██████▌   | 16127/24420 [6:43:30<3:25:41,  1.49s/it]
 66%|██████▌   | 16128/24420 [6:43:32<3:25:42,  1.49s/it]
 66%|██████▌   | 16129/24420 [6:43:33<3:25:50,  1.49s/it]
 66%|██████▌   | 16130/24420 [6:43:35<3:26:01,  1.49s/it]
                                                         
{'loss': 0.0784, 'learning_rate': 5e-05, 'epoch': 3.3}

 66%|██████▌   | 16130/24420 [6:43:35<3:26:01,  1.49s/it]
 66%|██████▌   | 16131/24420 [6:43:36<3:26:09,  1.49s/it]
 66%|██████▌   | 16132/24420 [6:43:38<3:26:05,  1.49s/it]
 66%|██████▌   | 16133/24420 [6:43:39<3:26:36,  1.50s/it]
 66%|██████▌   | 16134/24420 [6:43:41<3:26:59,  1.50s/it]
 66%|██████▌   | 16135/24420 [6:43:42<3:26:43,  1.50s/it]
                                                         
{'loss': 0.0769, 'learning_rate': 5e-05, 'epoch': 3.3}

 66%|██████▌   | 16135/24420 [6:43:42<3:26:43,  1.50s/it]
 66%|██████▌   | 16136/24420 [6:43:44<3:26:07,  1.49s/it]
 66%|██████▌   | 16137/24420 [6:43:45<3:25:37,  1.49s/it]
 66%|██████▌   | 16138/24420 [6:43:47<3:25:31,  1.49s/it]
 66%|██████▌   | 16139/24420 [6:43:48<3:25:37,  1.49s/it]
 66%|██████▌   | 16140/24420 [6:43:50<3:25:26,  1.49s/it]
                                                         
{'loss': 0.0917, 'learning_rate': 5e-05, 'epoch': 3.3}

 66%|██████▌   | 16140/24420 [6:43:50<3:25:26,  1.49s/it]
 66%|██████▌   | 16141/24420 [6:43:51<3:25:52,  1.49s/it]
 66%|██████▌   | 16142/24420 [6:43:53<3:25:41,  1.49s/it]
 66%|██████▌   | 16143/24420 [6:43:54<3:25:51,  1.49s/it]
 66%|██████▌   | 16144/24420 [6:43:56<3:25:29,  1.49s/it]
 66%|██████▌   | 16145/24420 [6:43:57<3:25:43,  1.49s/it]
                                                         
{'loss': 0.0782, 'learning_rate': 5e-05, 'epoch': 3.31}

 66%|██████▌   | 16145/24420 [6:43:57<3:25:43,  1.49s/it]
 66%|██████▌   | 16146/24420 [6:43:59<3:25:55,  1.49s/it]
 66%|██████▌   | 16147/24420 [6:44:00<3:25:38,  1.49s/it]
 66%|██████▌   | 16148/24420 [6:44:02<3:25:53,  1.49s/it]
 66%|██████▌   | 16149/24420 [6:44:03<3:25:28,  1.49s/it]
 66%|██████▌   | 16150/24420 [6:44:05<3:25:13,  1.49s/it]
                                                         
{'loss': 0.0819, 'learning_rate': 5e-05, 'epoch': 3.31}

 66%|██████▌   | 16150/24420 [6:44:05<3:25:13,  1.49s/it]
 66%|██████▌   | 16151/24420 [6:44:06<3:25:15,  1.49s/it]
 66%|██████▌   | 16152/24420 [6:44:08<3:25:23,  1.49s/it]
 66%|██████▌   | 16153/24420 [6:44:09<3:25:21,  1.49s/it]
 66%|██████▌   | 16154/24420 [6:44:11<3:25:36,  1.49s/it]
 66%|██████▌   | 16155/24420 [6:44:12<3:25:34,  1.49s/it]
                                                         
{'loss': 0.077, 'learning_rate': 5e-05, 'epoch': 3.31}

 66%|██████▌   | 16155/24420 [6:44:12<3:25:34,  1.49s/it]
 66%|██████▌   | 16156/24420 [6:44:14<3:25:26,  1.49s/it]
 66%|██████▌   | 16157/24420 [6:44:15<3:25:21,  1.49s/it]
 66%|██████▌   | 16158/24420 [6:44:17<3:25:18,  1.49s/it]
 66%|██████▌   | 16159/24420 [6:44:18<3:25:13,  1.49s/it]
 66%|██████▌   | 16160/24420 [6:44:20<3:24:52,  1.49s/it]
                                                         
{'loss': 0.081, 'learning_rate': 5e-05, 'epoch': 3.31}

 66%|██████▌   | 16160/24420 [6:44:20<3:24:52,  1.49s/it]
 66%|██████▌   | 16161/24420 [6:44:21<3:25:07,  1.49s/it]
 66%|██████▌   | 16162/24420 [6:44:23<3:24:47,  1.49s/it]
 66%|██████▌   | 16163/24420 [6:44:24<3:25:01,  1.49s/it]
 66%|██████▌   | 16164/24420 [6:44:26<3:25:16,  1.49s/it]
 66%|██████▌   | 16165/24420 [6:44:27<3:25:06,  1.49s/it]
                                                         
{'loss': 0.0805, 'learning_rate': 5e-05, 'epoch': 3.31}

 66%|██████▌   | 16165/24420 [6:44:27<3:25:06,  1.49s/it]
 66%|██████▌   | 16166/24420 [6:44:29<3:25:29,  1.49s/it]
 66%|██████▌   | 16167/24420 [6:44:30<3:25:31,  1.49s/it]
 66%|██████▌   | 16168/24420 [6:44:32<3:25:17,  1.49s/it]
 66%|██████▌   | 16169/24420 [6:44:33<3:25:10,  1.49s/it]
 66%|██████▌   | 16170/24420 [6:44:35<3:25:02,  1.49s/it]
                                                         
{'loss': 0.0819, 'learning_rate': 5e-05, 'epoch': 3.31}

 66%|██████▌   | 16170/24420 [6:44:35<3:25:02,  1.49s/it]
 66%|██████▌   | 16171/24420 [6:44:36<3:25:06,  1.49s/it]
 66%|██████▌   | 16172/24420 [6:44:38<3:24:41,  1.49s/it]
 66%|██████▌   | 16173/24420 [6:44:39<3:24:42,  1.49s/it]
 66%|██████▌   | 16174/24420 [6:44:41<3:25:00,  1.49s/it]
 66%|██████▌   | 16175/24420 [6:44:42<3:24:56,  1.49s/it]
                                                         
{'loss': 0.0904, 'learning_rate': 5e-05, 'epoch': 3.31}

 66%|██████▌   | 16175/24420 [6:44:42<3:24:56,  1.49s/it]
 66%|██████▌   | 16176/24420 [6:44:44<3:24:39,  1.49s/it]
 66%|██████▌   | 16177/24420 [6:44:45<3:24:46,  1.49s/it]
 66%|██████▌   | 16178/24420 [6:44:47<3:25:45,  1.50s/it]
 66%|██████▋   | 16179/24420 [6:44:48<3:25:13,  1.49s/it]
 66%|██████▋   | 16180/24420 [6:44:50<3:25:13,  1.49s/it]
                                                         
{'loss': 0.0914, 'learning_rate': 5e-05, 'epoch': 3.31}

 66%|██████▋   | 16180/24420 [6:44:50<3:25:13,  1.49s/it]
 66%|██████▋   | 16181/24420 [6:44:51<3:25:08,  1.49s/it]
 66%|██████▋   | 16182/24420 [6:44:53<3:24:53,  1.49s/it]
 66%|██████▋   | 16183/24420 [6:44:54<3:24:34,  1.49s/it]
 66%|██████▋   | 16184/24420 [6:44:55<3:24:41,  1.49s/it]
 66%|██████▋   | 16185/24420 [6:44:57<3:24:17,  1.49s/it]
                                                         
{'loss': 0.0653, 'learning_rate': 5e-05, 'epoch': 3.31}

 66%|██████▋   | 16185/24420 [6:44:57<3:24:17,  1.49s/it]
 66%|██████▋   | 16186/24420 [6:44:58<3:24:17,  1.49s/it]
 66%|██████▋   | 16187/24420 [6:45:00<3:23:51,  1.49s/it]
 66%|██████▋   | 16188/24420 [6:45:01<3:23:39,  1.48s/it]
 66%|██████▋   | 16189/24420 [6:45:03<3:23:31,  1.48s/it]
 66%|██████▋   | 16190/24420 [6:45:04<3:23:46,  1.49s/it]
                                                         
{'loss': 0.0759, 'learning_rate': 5e-05, 'epoch': 3.31}

 66%|██████▋   | 16190/24420 [6:45:04<3:23:46,  1.49s/it]
 66%|██████▋   | 16191/24420 [6:45:06<3:23:56,  1.49s/it]
 66%|██████▋   | 16192/24420 [6:45:07<3:23:51,  1.49s/it]
 66%|██████▋   | 16193/24420 [6:45:09<3:23:52,  1.49s/it]
 66%|██████▋   | 16194/24420 [6:45:10<3:24:21,  1.49s/it]
 66%|██████▋   | 16195/24420 [6:45:12<3:24:07,  1.49s/it]
                                                         
{'loss': 0.0761, 'learning_rate': 5e-05, 'epoch': 3.32}

 66%|██████▋   | 16195/24420 [6:45:12<3:24:07,  1.49s/it]
 66%|██████▋   | 16196/24420 [6:45:13<3:24:22,  1.49s/it]
 66%|██████▋   | 16197/24420 [6:45:15<3:24:13,  1.49s/it]
 66%|██████▋   | 16198/24420 [6:45:16<3:24:21,  1.49s/it]
 66%|██████▋   | 16199/24420 [6:45:18<3:24:05,  1.49s/it]
 66%|██████▋   | 16200/24420 [6:45:19<3:24:21,  1.49s/it]
                                                         
{'loss': 0.0928, 'learning_rate': 5e-05, 'epoch': 3.32}

 66%|██████▋   | 16200/24420 [6:45:19<3:24:21,  1.49s/it]
 66%|██████▋   | 16201/24420 [6:45:24<5:36:18,  2.46s/it]
 66%|██████▋   | 16202/24420 [6:45:25<4:56:30,  2.16s/it]
 66%|██████▋   | 16203/24420 [6:45:27<4:29:11,  1.97s/it]
 66%|██████▋   | 16204/24420 [6:45:28<4:09:26,  1.82s/it]
 66%|██████▋   | 16205/24420 [6:45:30<3:55:18,  1.72s/it]
                                                         
{'loss': 0.0798, 'learning_rate': 5e-05, 'epoch': 3.32}

 66%|██████▋   | 16205/24420 [6:45:30<3:55:18,  1.72s/it]
 66%|██████▋   | 16206/24420 [6:45:31<3:45:58,  1.65s/it]
 66%|██████▋   | 16207/24420 [6:45:33<3:42:56,  1.63s/it]
 66%|██████▋   | 16208/24420 [6:45:35<3:36:53,  1.58s/it]
 66%|██████▋   | 16209/24420 [6:45:36<3:32:45,  1.55s/it]
 66%|██████▋   | 16210/24420 [6:45:37<3:30:12,  1.54s/it]
                                                         
{'loss': 0.0819, 'learning_rate': 5e-05, 'epoch': 3.32}

 66%|██████▋   | 16210/24420 [6:45:37<3:30:12,  1.54s/it]
 66%|██████▋   | 16211/24420 [6:45:39<3:28:52,  1.53s/it]
 66%|██████▋   | 16212/24420 [6:45:40<3:27:16,  1.52s/it]
 66%|██████▋   | 16213/24420 [6:45:42<3:26:16,  1.51s/it]
 66%|██████▋   | 16214/24420 [6:45:43<3:25:24,  1.50s/it]
 66%|██████▋   | 16215/24420 [6:45:45<3:24:42,  1.50s/it]
                                                         
{'loss': 0.0743, 'learning_rate': 5e-05, 'epoch': 3.32}

 66%|██████▋   | 16215/24420 [6:45:45<3:24:42,  1.50s/it]
 66%|██████▋   | 16216/24420 [6:45:46<3:24:18,  1.49s/it]
 66%|██████▋   | 16217/24420 [6:45:48<3:23:53,  1.49s/it]
 66%|██████▋   | 16218/24420 [6:45:49<3:23:58,  1.49s/it]
 66%|██████▋   | 16219/24420 [6:45:51<3:24:14,  1.49s/it]
 66%|██████▋   | 16220/24420 [6:45:52<3:23:47,  1.49s/it]
                                                         
{'loss': 0.0804, 'learning_rate': 5e-05, 'epoch': 3.32}

 66%|██████▋   | 16220/24420 [6:45:52<3:23:47,  1.49s/it]
 66%|██████▋   | 16221/24420 [6:45:54<3:24:54,  1.50s/it]
 66%|██████▋   | 16222/24420 [6:45:55<3:24:08,  1.49s/it]
 66%|██████▋   | 16223/24420 [6:45:57<3:24:11,  1.49s/it]
 66%|██████▋   | 16224/24420 [6:45:58<3:23:59,  1.49s/it]
 66%|██████▋   | 16225/24420 [6:46:00<3:23:50,  1.49s/it]
                                                         
{'loss': 0.0851, 'learning_rate': 5e-05, 'epoch': 3.32}

 66%|██████▋   | 16225/24420 [6:46:00<3:23:50,  1.49s/it]
 66%|██████▋   | 16226/24420 [6:46:01<3:23:35,  1.49s/it]
 66%|██████▋   | 16227/24420 [6:46:03<3:23:39,  1.49s/it]
 66%|██████▋   | 16228/24420 [6:46:04<3:23:19,  1.49s/it]
 66%|██████▋   | 16229/24420 [6:46:06<3:23:11,  1.49s/it]
 66%|██████▋   | 16230/24420 [6:46:07<3:23:13,  1.49s/it]
                                                         
{'loss': 0.0642, 'learning_rate': 5e-05, 'epoch': 3.32}

 66%|██████▋   | 16230/24420 [6:46:07<3:23:13,  1.49s/it]
 66%|██████▋   | 16231/24420 [6:46:09<3:23:04,  1.49s/it]
 66%|██████▋   | 16232/24420 [6:46:10<3:23:26,  1.49s/it]
 66%|██████▋   | 16233/24420 [6:46:12<3:23:39,  1.49s/it]
 66%|██████▋   | 16234/24420 [6:46:13<3:23:45,  1.49s/it]
 66%|██████▋   | 16235/24420 [6:46:15<3:23:59,  1.50s/it]
                                                         
{'loss': 0.072, 'learning_rate': 5e-05, 'epoch': 3.32}

 66%|██████▋   | 16235/24420 [6:46:15<3:23:59,  1.50s/it]
 66%|██████▋   | 16236/24420 [6:46:16<3:23:37,  1.49s/it]
 66%|██████▋   | 16237/24420 [6:46:18<3:23:20,  1.49s/it]
 66%|██████▋   | 16238/24420 [6:46:19<3:23:34,  1.49s/it]
 66%|██████▋   | 16239/24420 [6:46:21<3:23:27,  1.49s/it]
 67%|██████▋   | 16240/24420 [6:46:22<3:23:20,  1.49s/it]
                                                         
{'loss': 0.0844, 'learning_rate': 5e-05, 'epoch': 3.33}

 67%|██████▋   | 16240/24420 [6:46:22<3:23:20,  1.49s/it]
 67%|██████▋   | 16241/24420 [6:46:24<3:23:23,  1.49s/it]
 67%|██████▋   | 16242/24420 [6:46:25<3:23:14,  1.49s/it]
 67%|██████▋   | 16243/24420 [6:46:27<3:23:42,  1.49s/it]
 67%|██████▋   | 16244/24420 [6:46:28<3:23:36,  1.49s/it]
 67%|██████▋   | 16245/24420 [6:46:30<3:23:35,  1.49s/it]
                                                         
{'loss': 0.0792, 'learning_rate': 5e-05, 'epoch': 3.33}

 67%|██████▋   | 16245/24420 [6:46:30<3:23:35,  1.49s/it]
 67%|██████▋   | 16246/24420 [6:46:31<3:23:23,  1.49s/it]
 67%|██████▋   | 16247/24420 [6:46:33<3:23:15,  1.49s/it]
 67%|██████▋   | 16248/24420 [6:46:34<3:22:53,  1.49s/it]
 67%|██████▋   | 16249/24420 [6:46:36<3:23:21,  1.49s/it]
 67%|██████▋   | 16250/24420 [6:46:37<3:23:17,  1.49s/it]
                                                         
{'loss': 0.0914, 'learning_rate': 5e-05, 'epoch': 3.33}

 67%|██████▋   | 16250/24420 [6:46:37<3:23:17,  1.49s/it]
 67%|██████▋   | 16251/24420 [6:46:39<3:23:02,  1.49s/it]
 67%|██████▋   | 16252/24420 [6:46:40<3:23:18,  1.49s/it]
 67%|██████▋   | 16253/24420 [6:46:42<3:23:21,  1.49s/it]
 67%|██████▋   | 16254/24420 [6:46:43<3:23:03,  1.49s/it]
 67%|██████▋   | 16255/24420 [6:46:45<3:22:42,  1.49s/it]
                                                         
{'loss': 0.0759, 'learning_rate': 5e-05, 'epoch': 3.33}

 67%|██████▋   | 16255/24420 [6:46:45<3:22:42,  1.49s/it]
 67%|██████▋   | 16256/24420 [6:46:46<3:22:43,  1.49s/it]
 67%|██████▋   | 16257/24420 [6:46:48<3:22:47,  1.49s/it]
 67%|██████▋   | 16258/24420 [6:46:49<3:22:51,  1.49s/it]
 67%|██████▋   | 16259/24420 [6:46:51<3:23:08,  1.49s/it]
 67%|██████▋   | 16260/24420 [6:46:52<3:23:17,  1.49s/it]
                                                         
{'loss': 0.0733, 'learning_rate': 5e-05, 'epoch': 3.33}

 67%|██████▋   | 16260/24420 [6:46:52<3:23:17,  1.49s/it]
 67%|██████▋   | 16261/24420 [6:46:54<3:23:15,  1.49s/it]
 67%|██████▋   | 16262/24420 [6:46:55<3:22:56,  1.49s/it]
 67%|██████▋   | 16263/24420 [6:46:57<3:22:40,  1.49s/it]
 67%|██████▋   | 16264/24420 [6:46:58<3:22:56,  1.49s/it]
 67%|██████▋   | 16265/24420 [6:47:00<3:22:49,  1.49s/it]
                                                         
{'loss': 0.0743, 'learning_rate': 5e-05, 'epoch': 3.33}

 67%|██████▋   | 16265/24420 [6:47:00<3:22:49,  1.49s/it]
 67%|██████▋   | 16266/24420 [6:47:01<3:22:44,  1.49s/it]
 67%|██████▋   | 16267/24420 [6:47:03<3:22:52,  1.49s/it]
 67%|██████▋   | 16268/24420 [6:47:04<3:22:41,  1.49s/it]
 67%|██████▋   | 16269/24420 [6:47:06<3:22:30,  1.49s/it]
 67%|██████▋   | 16270/24420 [6:47:07<3:22:05,  1.49s/it]
                                                         
{'loss': 0.0734, 'learning_rate': 5e-05, 'epoch': 3.33}

 67%|██████▋   | 16270/24420 [6:47:07<3:22:05,  1.49s/it]
 67%|██████▋   | 16271/24420 [6:47:08<3:22:24,  1.49s/it]
 67%|██████▋   | 16272/24420 [6:47:10<3:22:19,  1.49s/it]
 67%|██████▋   | 16273/24420 [6:47:11<3:22:15,  1.49s/it]
 67%|██████▋   | 16274/24420 [6:47:13<3:22:34,  1.49s/it]
 67%|██████▋   | 16275/24420 [6:47:14<3:22:19,  1.49s/it]
                                                         
{'loss': 0.071, 'learning_rate': 5e-05, 'epoch': 3.33}

 67%|██████▋   | 16275/24420 [6:47:14<3:22:19,  1.49s/it]
 67%|██████▋   | 16276/24420 [6:47:16<3:22:26,  1.49s/it]
 67%|██████▋   | 16277/24420 [6:47:17<3:22:04,  1.49s/it]
 67%|██████▋   | 16278/24420 [6:47:19<3:22:01,  1.49s/it]
 67%|██████▋   | 16279/24420 [6:47:20<3:21:55,  1.49s/it]
 67%|██████▋   | 16280/24420 [6:47:22<3:22:11,  1.49s/it]
                                                         
{'loss': 0.087, 'learning_rate': 5e-05, 'epoch': 3.33}

 67%|██████▋   | 16280/24420 [6:47:22<3:22:11,  1.49s/it]
 67%|██████▋   | 16281/24420 [6:47:23<3:22:05,  1.49s/it]
 67%|██████▋   | 16282/24420 [6:47:25<3:22:00,  1.49s/it]
 67%|██████▋   | 16283/24420 [6:47:26<3:21:37,  1.49s/it]
 67%|██████▋   | 16284/24420 [6:47:28<3:21:25,  1.49s/it]
 67%|██████▋   | 16285/24420 [6:47:29<3:21:30,  1.49s/it]
                                                         
{'loss': 0.0973, 'learning_rate': 5e-05, 'epoch': 3.33}

 67%|██████▋   | 16285/24420 [6:47:29<3:21:30,  1.49s/it]
 67%|██████▋   | 16286/24420 [6:47:31<3:22:03,  1.49s/it]
 67%|██████▋   | 16287/24420 [6:47:32<3:21:57,  1.49s/it]
 67%|██████▋   | 16288/24420 [6:47:34<3:21:53,  1.49s/it]
 67%|██████▋   | 16289/24420 [6:47:35<3:22:17,  1.49s/it]
 67%|██████▋   | 16290/24420 [6:47:37<3:22:02,  1.49s/it]
                                                         
{'loss': 0.0823, 'learning_rate': 5e-05, 'epoch': 3.34}

 67%|██████▋   | 16290/24420 [6:47:37<3:22:02,  1.49s/it]
 67%|██████▋   | 16291/24420 [6:47:38<3:21:44,  1.49s/it]
 67%|██████▋   | 16292/24420 [6:47:40<3:21:58,  1.49s/it]
 67%|██████▋   | 16293/24420 [6:47:41<3:22:22,  1.49s/it]
 67%|██████▋   | 16294/24420 [6:47:43<3:22:33,  1.50s/it]
 67%|██████▋   | 16295/24420 [6:47:44<3:22:34,  1.50s/it]
                                                         
{'loss': 0.0754, 'learning_rate': 5e-05, 'epoch': 3.34}

 67%|██████▋   | 16295/24420 [6:47:44<3:22:34,  1.50s/it]
 67%|██████▋   | 16296/24420 [6:47:46<3:22:34,  1.50s/it]
 67%|██████▋   | 16297/24420 [6:47:47<3:22:22,  1.49s/it]
 67%|██████▋   | 16298/24420 [6:47:49<3:22:12,  1.49s/it]
 67%|██████▋   | 16299/24420 [6:47:50<3:22:17,  1.49s/it]
 67%|██████▋   | 16300/24420 [6:47:52<3:22:58,  1.50s/it]
                                                         
{'loss': 0.0675, 'learning_rate': 5e-05, 'epoch': 3.34}

 67%|██████▋   | 16300/24420 [6:47:52<3:22:58,  1.50s/it]
 67%|██████▋   | 16301/24420 [6:47:53<3:22:32,  1.50s/it]
 67%|██████▋   | 16302/24420 [6:47:55<3:22:15,  1.49s/it]
 67%|██████▋   | 16303/24420 [6:47:56<3:22:11,  1.49s/it]
 67%|██████▋   | 16304/24420 [6:47:58<3:22:01,  1.49s/it]
 67%|██████▋   | 16305/24420 [6:47:59<3:22:22,  1.50s/it]
                                                         
{'loss': 0.0794, 'learning_rate': 5e-05, 'epoch': 3.34}

 67%|██████▋   | 16305/24420 [6:47:59<3:22:22,  1.50s/it]
 67%|██████▋   | 16306/24420 [6:48:01<3:22:06,  1.49s/it]
 67%|██████▋   | 16307/24420 [6:48:02<3:21:59,  1.49s/it]
 67%|██████▋   | 16308/24420 [6:48:04<3:21:41,  1.49s/it]
 67%|██████▋   | 16309/24420 [6:48:05<3:21:28,  1.49s/it]
 67%|██████▋   | 16310/24420 [6:48:07<3:21:26,  1.49s/it]
                                                         
{'loss': 0.0767, 'learning_rate': 5e-05, 'epoch': 3.34}

 67%|██████▋   | 16310/24420 [6:48:07<3:21:26,  1.49s/it]
 67%|██████▋   | 16311/24420 [6:48:08<3:21:17,  1.49s/it]
 67%|██████▋   | 16312/24420 [6:48:10<3:21:11,  1.49s/it]
 67%|██████▋   | 16313/24420 [6:48:11<3:21:09,  1.49s/it]
 67%|██████▋   | 16314/24420 [6:48:13<3:21:07,  1.49s/it]
 67%|██████▋   | 16315/24420 [6:48:14<3:21:09,  1.49s/it]
                                                         
{'loss': 0.0686, 'learning_rate': 5e-05, 'epoch': 3.34}

 67%|██████▋   | 16315/24420 [6:48:14<3:21:09,  1.49s/it]
 67%|██████▋   | 16316/24420 [6:48:16<3:21:13,  1.49s/it]
 67%|██████▋   | 16317/24420 [6:48:17<3:21:19,  1.49s/it]
 67%|██████▋   | 16318/24420 [6:48:19<3:21:19,  1.49s/it]
 67%|██████▋   | 16319/24420 [6:48:20<3:21:09,  1.49s/it]
 67%|██████▋   | 16320/24420 [6:48:22<3:21:05,  1.49s/it]
                                                         
{'loss': 0.0896, 'learning_rate': 5e-05, 'epoch': 3.34}

 67%|██████▋   | 16320/24420 [6:48:22<3:21:05,  1.49s/it]
 67%|██████▋   | 16321/24420 [6:48:23<3:21:08,  1.49s/it]
 67%|██████▋   | 16322/24420 [6:48:25<3:21:03,  1.49s/it]
 67%|██████▋   | 16323/24420 [6:48:26<3:20:54,  1.49s/it]
 67%|██████▋   | 16324/24420 [6:48:28<3:20:50,  1.49s/it]
 67%|██████▋   | 16325/24420 [6:48:29<3:21:19,  1.49s/it]
                                                         
{'loss': 0.0767, 'learning_rate': 5e-05, 'epoch': 3.34}

 67%|██████▋   | 16325/24420 [6:48:29<3:21:19,  1.49s/it]
 67%|██████▋   | 16326/24420 [6:48:31<3:21:14,  1.49s/it]
 67%|██████▋   | 16327/24420 [6:48:32<3:21:18,  1.49s/it]
 67%|██████▋   | 16328/24420 [6:48:34<3:21:05,  1.49s/it]
 67%|██████▋   | 16329/24420 [6:48:35<3:21:21,  1.49s/it]
 67%|██████▋   | 16330/24420 [6:48:36<3:20:56,  1.49s/it]
                                                         
{'loss': 0.0865, 'learning_rate': 5e-05, 'epoch': 3.34}

 67%|██████▋   | 16330/24420 [6:48:36<3:20:56,  1.49s/it]
 67%|██████▋   | 16331/24420 [6:48:38<3:21:17,  1.49s/it]
 67%|██████▋   | 16332/24420 [6:48:39<3:21:04,  1.49s/it]
 67%|██████▋   | 16333/24420 [6:48:41<3:21:09,  1.49s/it]
 67%|██████▋   | 16334/24420 [6:48:42<3:20:51,  1.49s/it]
 67%|██████▋   | 16335/24420 [6:48:44<3:20:53,  1.49s/it]
                                                         
{'loss': 0.1061, 'learning_rate': 5e-05, 'epoch': 3.34}

 67%|██████▋   | 16335/24420 [6:48:44<3:20:53,  1.49s/it]
 67%|██████▋   | 16336/24420 [6:48:45<3:21:03,  1.49s/it]
 67%|██████▋   | 16337/24420 [6:48:47<3:21:22,  1.49s/it]
 67%|██████▋   | 16338/24420 [6:48:48<3:21:03,  1.49s/it]
 67%|██████▋   | 16339/24420 [6:48:50<3:20:51,  1.49s/it]
 67%|██████▋   | 16340/24420 [6:48:51<3:20:31,  1.49s/it]
                                                         
{'loss': 0.0828, 'learning_rate': 5e-05, 'epoch': 3.35}

 67%|██████▋   | 16340/24420 [6:48:51<3:20:31,  1.49s/it]
 67%|██████▋   | 16341/24420 [6:48:53<3:20:40,  1.49s/it]
 67%|██████▋   | 16342/24420 [6:48:54<3:20:45,  1.49s/it]
 67%|██████▋   | 16343/24420 [6:48:56<3:20:46,  1.49s/it]
 67%|██████▋   | 16344/24420 [6:48:57<3:20:38,  1.49s/it]
 67%|██████▋   | 16345/24420 [6:48:59<3:20:23,  1.49s/it]
                                                         
{'loss': 0.0793, 'learning_rate': 5e-05, 'epoch': 3.35}

 67%|██████▋   | 16345/24420 [6:48:59<3:20:23,  1.49s/it]
 67%|██████▋   | 16346/24420 [6:49:00<3:20:32,  1.49s/it]
 67%|██████▋   | 16347/24420 [6:49:02<3:20:28,  1.49s/it]
 67%|██████▋   | 16348/24420 [6:49:03<3:20:01,  1.49s/it]
 67%|██████▋   | 16349/24420 [6:49:05<3:19:48,  1.49s/it]
 67%|██████▋   | 16350/24420 [6:49:06<3:19:26,  1.48s/it]
                                                         
{'loss': 0.0883, 'learning_rate': 5e-05, 'epoch': 3.35}

 67%|██████▋   | 16350/24420 [6:49:06<3:19:26,  1.48s/it]
 67%|██████▋   | 16351/24420 [6:49:08<3:19:43,  1.49s/it]
 67%|██████▋   | 16352/24420 [6:49:09<3:20:25,  1.49s/it]
 67%|██████▋   | 16353/24420 [6:49:11<3:20:35,  1.49s/it]
 67%|██████▋   | 16354/24420 [6:49:12<3:20:12,  1.49s/it]
 67%|██████▋   | 16355/24420 [6:49:14<3:20:06,  1.49s/it]
                                                         
{'loss': 0.0735, 'learning_rate': 5e-05, 'epoch': 3.35}

 67%|██████▋   | 16355/24420 [6:49:14<3:20:06,  1.49s/it]
 67%|██████▋   | 16356/24420 [6:49:15<3:19:54,  1.49s/it]
 67%|██████▋   | 16357/24420 [6:49:17<3:19:35,  1.49s/it]
 67%|██████▋   | 16358/24420 [6:49:18<3:19:55,  1.49s/it]
 67%|██████▋   | 16359/24420 [6:49:20<3:19:54,  1.49s/it]
 67%|██████▋   | 16360/24420 [6:49:21<3:20:24,  1.49s/it]
                                                         
{'loss': 0.0757, 'learning_rate': 5e-05, 'epoch': 3.35}

 67%|██████▋   | 16360/24420 [6:49:21<3:20:24,  1.49s/it]
 67%|██████▋   | 16361/24420 [6:49:23<3:20:42,  1.49s/it]
 67%|██████▋   | 16362/24420 [6:49:24<3:20:38,  1.49s/it]
 67%|██████▋   | 16363/24420 [6:49:26<3:20:20,  1.49s/it]
 67%|██████▋   | 16364/24420 [6:49:27<3:20:39,  1.49s/it]
 67%|██████▋   | 16365/24420 [6:49:29<3:20:28,  1.49s/it]
                                                         
{'loss': 0.0717, 'learning_rate': 5e-05, 'epoch': 3.35}

 67%|██████▋   | 16365/24420 [6:49:29<3:20:28,  1.49s/it]
 67%|██████▋   | 16366/24420 [6:49:30<3:20:33,  1.49s/it]
 67%|██████▋   | 16367/24420 [6:49:32<3:20:33,  1.49s/it]
 67%|██████▋   | 16368/24420 [6:49:33<3:20:05,  1.49s/it]
 67%|██████▋   | 16369/24420 [6:49:35<3:20:03,  1.49s/it]
 67%|██████▋   | 16370/24420 [6:49:36<3:20:03,  1.49s/it]
                                                         
{'loss': 0.0772, 'learning_rate': 5e-05, 'epoch': 3.35}

 67%|██████▋   | 16370/24420 [6:49:36<3:20:03,  1.49s/it]
 67%|██████▋   | 16371/24420 [6:49:38<3:19:58,  1.49s/it]
 67%|██████▋   | 16372/24420 [6:49:39<3:20:04,  1.49s/it]
 67%|██████▋   | 16373/24420 [6:49:41<3:19:59,  1.49s/it]
 67%|██████▋   | 16374/24420 [6:49:42<3:20:22,  1.49s/it]
 67%|██████▋   | 16375/24420 [6:49:44<3:20:09,  1.49s/it]
                                                         
{'loss': 0.1064, 'learning_rate': 5e-05, 'epoch': 3.35}

 67%|██████▋   | 16375/24420 [6:49:44<3:20:09,  1.49s/it]
 67%|██████▋   | 16376/24420 [6:49:45<3:20:10,  1.49s/it]
 67%|██████▋   | 16377/24420 [6:49:47<3:19:45,  1.49s/it]
 67%|██████▋   | 16378/24420 [6:49:48<3:20:02,  1.49s/it]
 67%|██████▋   | 16379/24420 [6:49:50<3:20:12,  1.49s/it]
 67%|██████▋   | 16380/24420 [6:49:51<3:20:18,  1.49s/it]
                                                         
{'loss': 0.0726, 'learning_rate': 5e-05, 'epoch': 3.35}

 67%|██████▋   | 16380/24420 [6:49:51<3:20:18,  1.49s/it]
 67%|██████▋   | 16381/24420 [6:49:53<3:20:26,  1.50s/it]
 67%|██████▋   | 16382/24420 [6:49:54<3:20:16,  1.50s/it]
 67%|██████▋   | 16383/24420 [6:49:56<3:20:36,  1.50s/it]
 67%|██████▋   | 16384/24420 [6:49:57<3:20:17,  1.50s/it]
 67%|██████▋   | 16385/24420 [6:49:59<3:20:17,  1.50s/it]
                                                         
{'loss': 0.0703, 'learning_rate': 5e-05, 'epoch': 3.35}

 67%|██████▋   | 16385/24420 [6:49:59<3:20:17,  1.50s/it]
 67%|██████▋   | 16386/24420 [6:50:00<3:20:00,  1.49s/it]
 67%|██████▋   | 16387/24420 [6:50:02<3:20:06,  1.49s/it]
 67%|██████▋   | 16388/24420 [6:50:03<3:19:51,  1.49s/it]
 67%|██████▋   | 16389/24420 [6:50:04<3:19:47,  1.49s/it]
 67%|██████▋   | 16390/24420 [6:50:06<3:19:33,  1.49s/it]
                                                         
{'loss': 0.0912, 'learning_rate': 5e-05, 'epoch': 3.36}

 67%|██████▋   | 16390/24420 [6:50:06<3:19:33,  1.49s/it]
 67%|██████▋   | 16391/24420 [6:50:07<3:19:42,  1.49s/it]
 67%|██████▋   | 16392/24420 [6:50:09<3:19:25,  1.49s/it]
 67%|██████▋   | 16393/24420 [6:50:10<3:19:36,  1.49s/it]
 67%|██████▋   | 16394/24420 [6:50:12<3:19:26,  1.49s/it]
 67%|██████▋   | 16395/24420 [6:50:13<3:20:45,  1.50s/it]
                                                         
{'loss': 0.0838, 'learning_rate': 5e-05, 'epoch': 3.36}

 67%|██████▋   | 16395/24420 [6:50:13<3:20:45,  1.50s/it]
 67%|██████▋   | 16396/24420 [6:50:15<3:20:36,  1.50s/it]
 67%|██████▋   | 16397/24420 [6:50:16<3:20:14,  1.50s/it]
 67%|██████▋   | 16398/24420 [6:50:18<3:19:40,  1.49s/it]
 67%|██████▋   | 16399/24420 [6:50:19<3:19:36,  1.49s/it]
 67%|██████▋   | 16400/24420 [6:50:21<3:19:26,  1.49s/it]
                                                         
{'loss': 0.0991, 'learning_rate': 5e-05, 'epoch': 3.36}

 67%|██████▋   | 16400/24420 [6:50:21<3:19:26,  1.49s/it]
 67%|██████▋   | 16401/24420 [6:50:27<6:24:45,  2.88s/it]
 67%|██████▋   | 16402/24420 [6:50:29<5:29:27,  2.47s/it]
 67%|██████▋   | 16403/24420 [6:50:30<4:50:14,  2.17s/it]
 67%|██████▋   | 16404/24420 [6:50:32<4:22:53,  1.97s/it]
 67%|██████▋   | 16405/24420 [6:50:33<4:03:43,  1.82s/it]
                                                         
{'loss': 0.095, 'learning_rate': 5e-05, 'epoch': 3.36}

 67%|██████▋   | 16405/24420 [6:50:33<4:03:43,  1.82s/it]
 67%|██████▋   | 16406/24420 [6:50:34<3:50:05,  1.72s/it]
 67%|██████▋   | 16407/24420 [6:50:36<3:40:34,  1.65s/it]
 67%|██████▋   | 16408/24420 [6:50:37<3:34:09,  1.60s/it]
 67%|██████▋   | 16409/24420 [6:50:39<3:29:23,  1.57s/it]
 67%|██████▋   | 16410/24420 [6:50:40<3:26:00,  1.54s/it]
                                                         
{'loss': 0.0705, 'learning_rate': 5e-05, 'epoch': 3.36}

 67%|██████▋   | 16410/24420 [6:50:40<3:26:00,  1.54s/it]
 67%|██████▋   | 16411/24420 [6:50:42<3:23:51,  1.53s/it]
 67%|██████▋   | 16412/24420 [6:50:43<3:22:30,  1.52s/it]
 67%|██████▋   | 16413/24420 [6:50:45<3:20:45,  1.50s/it]
 67%|██████▋   | 16414/24420 [6:50:46<3:20:21,  1.50s/it]
 67%|██████▋   | 16415/24420 [6:50:48<3:19:59,  1.50s/it]
                                                         
{'loss': 0.0784, 'learning_rate': 5e-05, 'epoch': 3.36}

 67%|██████▋   | 16415/24420 [6:50:48<3:19:59,  1.50s/it]
 67%|██████▋   | 16416/24420 [6:50:49<3:20:38,  1.50s/it]
 67%|██████▋   | 16417/24420 [6:50:51<3:19:54,  1.50s/it]
 67%|██████▋   | 16418/24420 [6:50:52<3:19:35,  1.50s/it]
 67%|██████▋   | 16419/24420 [6:50:54<3:19:55,  1.50s/it]
 67%|██████▋   | 16420/24420 [6:50:55<3:19:44,  1.50s/it]
                                                         
{'loss': 0.0833, 'learning_rate': 5e-05, 'epoch': 3.36}

 67%|██████▋   | 16420/24420 [6:50:55<3:19:44,  1.50s/it]
 67%|██████▋   | 16421/24420 [6:50:57<3:19:26,  1.50s/it]
 67%|██████▋   | 16422/24420 [6:50:58<3:19:19,  1.50s/it]
 67%|██████▋   | 16423/24420 [6:51:00<3:19:14,  1.49s/it]
 67%|██████▋   | 16424/24420 [6:51:01<3:18:35,  1.49s/it]
 67%|██████▋   | 16425/24420 [6:51:03<3:18:16,  1.49s/it]
                                                         
{'loss': 0.0826, 'learning_rate': 5e-05, 'epoch': 3.36}

 67%|██████▋   | 16425/24420 [6:51:03<3:18:16,  1.49s/it]
 67%|██████▋   | 16426/24420 [6:51:04<3:18:24,  1.49s/it]
 67%|██████▋   | 16427/24420 [6:51:06<3:18:31,  1.49s/it]
 67%|██████▋   | 16428/24420 [6:51:07<3:18:45,  1.49s/it]
 67%|██████▋   | 16429/24420 [6:51:09<3:18:36,  1.49s/it]
 67%|██████▋   | 16430/24420 [6:51:10<3:18:19,  1.49s/it]
                                                         
{'loss': 0.0783, 'learning_rate': 5e-05, 'epoch': 3.36}

 67%|██████▋   | 16430/24420 [6:51:10<3:18:19,  1.49s/it]
 67%|██████▋   | 16431/24420 [6:51:12<3:18:16,  1.49s/it]
 67%|██████▋   | 16432/24420 [6:51:13<3:17:44,  1.49s/it]
 67%|██████▋   | 16433/24420 [6:51:15<3:17:47,  1.49s/it]
 67%|██████▋   | 16434/24420 [6:51:16<3:18:01,  1.49s/it]
 67%|██████▋   | 16435/24420 [6:51:18<3:17:50,  1.49s/it]
                                                         
{'loss': 0.0783, 'learning_rate': 5e-05, 'epoch': 3.37}

 67%|██████▋   | 16435/24420 [6:51:18<3:17:50,  1.49s/it]
 67%|██████▋   | 16436/24420 [6:51:19<3:17:41,  1.49s/it]
 67%|██████▋   | 16437/24420 [6:51:21<3:17:28,  1.48s/it]
 67%|██████▋   | 16438/24420 [6:51:22<3:17:28,  1.48s/it]
 67%|██████▋   | 16439/24420 [6:51:24<3:17:59,  1.49s/it]
 67%|██████▋   | 16440/24420 [6:51:25<3:18:04,  1.49s/it]
                                                         
{'loss': 0.0807, 'learning_rate': 5e-05, 'epoch': 3.37}

 67%|██████▋   | 16440/24420 [6:51:25<3:18:04,  1.49s/it]
 67%|██████▋   | 16441/24420 [6:51:27<3:18:12,  1.49s/it]
 67%|██████▋   | 16442/24420 [6:51:28<3:17:54,  1.49s/it]
 67%|██████▋   | 16443/24420 [6:51:30<3:17:55,  1.49s/it]
 67%|██████▋   | 16444/24420 [6:51:31<3:18:04,  1.49s/it]
 67%|██████▋   | 16445/24420 [6:51:33<3:17:56,  1.49s/it]
                                                         
{'loss': 0.0653, 'learning_rate': 5e-05, 'epoch': 3.37}

 67%|██████▋   | 16445/24420 [6:51:33<3:17:56,  1.49s/it]
 67%|██████▋   | 16446/24420 [6:51:34<3:18:14,  1.49s/it]
 67%|██████▋   | 16447/24420 [6:51:36<3:18:18,  1.49s/it]
 67%|██████▋   | 16448/24420 [6:51:37<3:18:18,  1.49s/it]
 67%|██████▋   | 16449/24420 [6:51:39<3:18:34,  1.49s/it]
 67%|██████▋   | 16450/24420 [6:51:40<3:18:31,  1.49s/it]
                                                         
{'loss': 0.0712, 'learning_rate': 5e-05, 'epoch': 3.37}

 67%|██████▋   | 16450/24420 [6:51:40<3:18:31,  1.49s/it]
 67%|██████▋   | 16451/24420 [6:51:42<3:18:35,  1.50s/it]
 67%|██████▋   | 16452/24420 [6:51:43<3:18:10,  1.49s/it]
 67%|██████▋   | 16453/24420 [6:51:45<3:17:57,  1.49s/it]
 67%|██████▋   | 16454/24420 [6:51:46<3:17:50,  1.49s/it]
 67%|██████▋   | 16455/24420 [6:51:48<3:17:41,  1.49s/it]
                                                         
{'loss': 0.0714, 'learning_rate': 5e-05, 'epoch': 3.37}

 67%|██████▋   | 16455/24420 [6:51:48<3:17:41,  1.49s/it]
 67%|██████▋   | 16456/24420 [6:51:49<3:17:36,  1.49s/it]
 67%|██████▋   | 16457/24420 [6:51:50<3:17:32,  1.49s/it]
 67%|██████▋   | 16458/24420 [6:51:52<3:17:46,  1.49s/it]
 67%|██████▋   | 16459/24420 [6:51:53<3:18:02,  1.49s/it]
 67%|██████▋   | 16460/24420 [6:51:55<3:18:04,  1.49s/it]
                                                         
{'loss': 0.0759, 'learning_rate': 5e-05, 'epoch': 3.37}

 67%|██████▋   | 16460/24420 [6:51:55<3:18:04,  1.49s/it]
 67%|██████▋   | 16461/24420 [6:51:56<3:18:13,  1.49s/it]
 67%|██████▋   | 16462/24420 [6:51:58<3:18:17,  1.50s/it]
 67%|██████▋   | 16463/24420 [6:51:59<3:18:01,  1.49s/it]
 67%|██████▋   | 16464/24420 [6:52:01<3:17:56,  1.49s/it]
 67%|██████▋   | 16465/24420 [6:52:02<3:17:59,  1.49s/it]
                                                         
{'loss': 0.0777, 'learning_rate': 5e-05, 'epoch': 3.37}

 67%|██████▋   | 16465/24420 [6:52:02<3:17:59,  1.49s/it]
 67%|██████▋   | 16466/24420 [6:52:04<3:17:49,  1.49s/it]
 67%|██████▋   | 16467/24420 [6:52:05<3:17:50,  1.49s/it]
 67%|██████▋   | 16468/24420 [6:52:07<3:17:30,  1.49s/it]
 67%|██████▋   | 16469/24420 [6:52:08<3:17:21,  1.49s/it]
 67%|██████▋   | 16470/24420 [6:52:10<3:17:19,  1.49s/it]
                                                         
{'loss': 0.0854, 'learning_rate': 5e-05, 'epoch': 3.37}

 67%|██████▋   | 16470/24420 [6:52:10<3:17:19,  1.49s/it]
 67%|██████▋   | 16471/24420 [6:52:11<3:17:41,  1.49s/it]
 67%|██████▋   | 16472/24420 [6:52:13<3:17:33,  1.49s/it]
 67%|██████▋   | 16473/24420 [6:52:14<3:17:43,  1.49s/it]
 67%|██████▋   | 16474/24420 [6:52:16<3:17:53,  1.49s/it]
 67%|██████▋   | 16475/24420 [6:52:17<3:18:18,  1.50s/it]
                                                         
{'loss': 0.0745, 'learning_rate': 5e-05, 'epoch': 3.37}

 67%|██████▋   | 16475/24420 [6:52:17<3:18:18,  1.50s/it]
 67%|██████▋   | 16476/24420 [6:52:19<3:17:55,  1.49s/it]
 67%|██████▋   | 16477/24420 [6:52:20<3:17:35,  1.49s/it]
 67%|██████▋   | 16478/24420 [6:52:22<3:17:27,  1.49s/it]
 67%|██████▋   | 16479/24420 [6:52:23<3:17:39,  1.49s/it]
 67%|██████▋   | 16480/24420 [6:52:25<3:17:26,  1.49s/it]
                                                         
{'loss': 0.0747, 'learning_rate': 5e-05, 'epoch': 3.37}

 67%|██████▋   | 16480/24420 [6:52:25<3:17:26,  1.49s/it]
 67%|██████▋   | 16481/24420 [6:52:26<3:17:17,  1.49s/it]
 67%|██████▋   | 16482/24420 [6:52:28<3:17:18,  1.49s/it]
 67%|██████▋   | 16483/24420 [6:52:29<3:17:09,  1.49s/it]
 68%|██████▊   | 16484/24420 [6:52:31<3:16:58,  1.49s/it]
 68%|██████▊   | 16485/24420 [6:52:32<3:17:20,  1.49s/it]
                                                         
{'loss': 0.0768, 'learning_rate': 5e-05, 'epoch': 3.38}

 68%|██████▊   | 16485/24420 [6:52:32<3:17:20,  1.49s/it]
 68%|██████▊   | 16486/24420 [6:52:34<3:17:47,  1.50s/it]
 68%|██████▊   | 16487/24420 [6:52:35<3:17:44,  1.50s/it]
 68%|██████▊   | 16488/24420 [6:52:37<3:17:42,  1.50s/it]
 68%|██████▊   | 16489/24420 [6:52:38<3:17:29,  1.49s/it]
 68%|██████▊   | 16490/24420 [6:52:40<3:17:28,  1.49s/it]
                                                         
{'loss': 0.0678, 'learning_rate': 5e-05, 'epoch': 3.38}

 68%|██████▊   | 16490/24420 [6:52:40<3:17:28,  1.49s/it]
 68%|██████▊   | 16491/24420 [6:52:41<3:17:35,  1.50s/it]
 68%|██████▊   | 16492/24420 [6:52:43<3:17:19,  1.49s/it]
 68%|██████▊   | 16493/24420 [6:52:44<3:17:31,  1.50s/it]
 68%|██████▊   | 16494/24420 [6:52:46<3:17:30,  1.50s/it]
 68%|██████▊   | 16495/24420 [6:52:47<3:17:36,  1.50s/it]
                                                         
{'loss': 0.1094, 'learning_rate': 5e-05, 'epoch': 3.38}

 68%|██████▊   | 16495/24420 [6:52:47<3:17:36,  1.50s/it]
 68%|██████▊   | 16496/24420 [6:52:49<3:17:42,  1.50s/it]
 68%|██████▊   | 16497/24420 [6:52:50<3:17:43,  1.50s/it]
 68%|██████▊   | 16498/24420 [6:52:52<3:17:23,  1.49s/it]
 68%|██████▊   | 16499/24420 [6:52:53<3:17:24,  1.50s/it]
 68%|██████▊   | 16500/24420 [6:52:55<3:17:45,  1.50s/it]
                                                         
{'loss': 0.0756, 'learning_rate': 5e-05, 'epoch': 3.38}

 68%|██████▊   | 16500/24420 [6:52:55<3:17:45,  1.50s/it]
 68%|██████▊   | 16501/24420 [6:52:56<3:17:58,  1.50s/it]
 68%|██████▊   | 16502/24420 [6:52:58<3:17:38,  1.50s/it]
 68%|██████▊   | 16503/24420 [6:52:59<3:17:11,  1.49s/it]
 68%|██████▊   | 16504/24420 [6:53:01<3:17:03,  1.49s/it]
 68%|██████▊   | 16505/24420 [6:53:02<3:17:04,  1.49s/it]
                                                         
{'loss': 0.086, 'learning_rate': 5e-05, 'epoch': 3.38}

 68%|██████▊   | 16505/24420 [6:53:02<3:17:04,  1.49s/it]
 68%|██████▊   | 16506/24420 [6:53:04<3:16:55,  1.49s/it]
 68%|██████▊   | 16507/24420 [6:53:05<3:16:56,  1.49s/it]
 68%|██████▊   | 16508/24420 [6:53:07<3:17:22,  1.50s/it]
 68%|██████▊   | 16509/24420 [6:53:08<3:17:18,  1.50s/it]
 68%|██████▊   | 16510/24420 [6:53:10<3:17:37,  1.50s/it]
                                                         
{'loss': 0.0777, 'learning_rate': 5e-05, 'epoch': 3.38}

 68%|██████▊   | 16510/24420 [6:53:10<3:17:37,  1.50s/it]
 68%|██████▊   | 16511/24420 [6:53:11<3:17:31,  1.50s/it]
 68%|██████▊   | 16512/24420 [6:53:13<3:17:27,  1.50s/it]
 68%|██████▊   | 16513/24420 [6:53:14<3:17:10,  1.50s/it]
 68%|██████▊   | 16514/24420 [6:53:16<3:16:32,  1.49s/it]
 68%|██████▊   | 16515/24420 [6:53:17<3:16:26,  1.49s/it]
                                                         
{'loss': 0.0889, 'learning_rate': 5e-05, 'epoch': 3.38}

 68%|██████▊   | 16515/24420 [6:53:17<3:16:26,  1.49s/it]
 68%|██████▊   | 16516/24420 [6:53:19<3:16:39,  1.49s/it]
 68%|██████▊   | 16517/24420 [6:53:20<3:16:51,  1.49s/it]
 68%|██████▊   | 16518/24420 [6:53:22<3:16:29,  1.49s/it]
 68%|██████▊   | 16519/24420 [6:53:23<3:16:19,  1.49s/it]
 68%|██████▊   | 16520/24420 [6:53:25<3:16:11,  1.49s/it]
                                                         
{'loss': 0.0815, 'learning_rate': 5e-05, 'epoch': 3.38}

 68%|██████▊   | 16520/24420 [6:53:25<3:16:11,  1.49s/it]
 68%|██████▊   | 16521/24420 [6:53:26<3:15:56,  1.49s/it]
 68%|██████▊   | 16522/24420 [6:53:28<3:15:41,  1.49s/it]
 68%|██████▊   | 16523/24420 [6:53:29<3:15:40,  1.49s/it]
 68%|██████▊   | 16524/24420 [6:53:31<3:15:39,  1.49s/it]
 68%|██████▊   | 16525/24420 [6:53:32<3:15:23,  1.48s/it]
                                                         
{'loss': 0.0747, 'learning_rate': 5e-05, 'epoch': 3.38}

 68%|██████▊   | 16525/24420 [6:53:32<3:15:23,  1.48s/it]
 68%|██████▊   | 16526/24420 [6:53:34<3:15:26,  1.49s/it]
 68%|██████▊   | 16527/24420 [6:53:35<3:15:33,  1.49s/it]
 68%|██████▊   | 16528/24420 [6:53:36<3:15:34,  1.49s/it]
 68%|██████▊   | 16529/24420 [6:53:38<3:15:28,  1.49s/it]
 68%|██████▊   | 16530/24420 [6:53:39<3:15:44,  1.49s/it]
                                                         
{'loss': 0.0739, 'learning_rate': 5e-05, 'epoch': 3.38}

 68%|██████▊   | 16530/24420 [6:53:39<3:15:44,  1.49s/it]
 68%|██████▊   | 16531/24420 [6:53:41<3:15:48,  1.49s/it]
 68%|██████▊   | 16532/24420 [6:53:42<3:16:38,  1.50s/it]
 68%|██████▊   | 16533/24420 [6:53:44<3:16:21,  1.49s/it]
 68%|██████▊   | 16534/24420 [6:53:45<3:16:13,  1.49s/it]
 68%|██████▊   | 16535/24420 [6:53:47<3:16:30,  1.50s/it]
                                                         
{'loss': 0.0749, 'learning_rate': 5e-05, 'epoch': 3.39}

 68%|██████▊   | 16535/24420 [6:53:47<3:16:30,  1.50s/it]
 68%|██████▊   | 16536/24420 [6:53:48<3:16:18,  1.49s/it]
 68%|██████▊   | 16537/24420 [6:53:50<3:16:12,  1.49s/it]
 68%|██████▊   | 16538/24420 [6:53:51<3:16:34,  1.50s/it]
 68%|██████▊   | 16539/24420 [6:53:53<3:16:04,  1.49s/it]
 68%|██████▊   | 16540/24420 [6:53:54<3:15:57,  1.49s/it]
                                                         
{'loss': 0.0701, 'learning_rate': 5e-05, 'epoch': 3.39}

 68%|██████▊   | 16540/24420 [6:53:54<3:15:57,  1.49s/it]
 68%|██████▊   | 16541/24420 [6:53:56<3:15:56,  1.49s/it]
 68%|██████▊   | 16542/24420 [6:53:57<3:15:43,  1.49s/it]
 68%|██████▊   | 16543/24420 [6:53:59<3:15:50,  1.49s/it]
 68%|██████▊   | 16544/24420 [6:54:00<3:16:21,  1.50s/it]
 68%|██████▊   | 16545/24420 [6:54:02<3:16:10,  1.49s/it]
                                                         
{'loss': 0.0748, 'learning_rate': 5e-05, 'epoch': 3.39}

 68%|██████▊   | 16545/24420 [6:54:02<3:16:10,  1.49s/it]
 68%|██████▊   | 16546/24420 [6:54:03<3:16:11,  1.49s/it]
 68%|██████▊   | 16547/24420 [6:54:05<3:15:54,  1.49s/it]
 68%|██████▊   | 16548/24420 [6:54:06<3:16:12,  1.50s/it]
 68%|██████▊   | 16549/24420 [6:54:08<3:16:01,  1.49s/it]
 68%|██████▊   | 16550/24420 [6:54:09<3:15:45,  1.49s/it]
                                                         
{'loss': 0.0729, 'learning_rate': 5e-05, 'epoch': 3.39}

 68%|██████▊   | 16550/24420 [6:54:09<3:15:45,  1.49s/it]
 68%|██████▊   | 16551/24420 [6:54:11<3:16:06,  1.50s/it]
 68%|██████▊   | 16552/24420 [6:54:12<3:15:57,  1.49s/it]
 68%|██████▊   | 16553/24420 [6:54:14<3:15:45,  1.49s/it]
 68%|██████▊   | 16554/24420 [6:54:15<3:15:52,  1.49s/it]
 68%|██████▊   | 16555/24420 [6:54:17<3:16:07,  1.50s/it]
                                                         
{'loss': 0.0841, 'learning_rate': 5e-05, 'epoch': 3.39}

 68%|██████▊   | 16555/24420 [6:54:17<3:16:07,  1.50s/it]
 68%|██████▊   | 16556/24420 [6:54:18<3:15:37,  1.49s/it]
 68%|██████▊   | 16557/24420 [6:54:20<3:15:51,  1.49s/it]
 68%|██████▊   | 16558/24420 [6:54:21<3:15:44,  1.49s/it]
 68%|██████▊   | 16559/24420 [6:54:23<3:16:03,  1.50s/it]
 68%|██████▊   | 16560/24420 [6:54:24<3:15:52,  1.50s/it]
                                                         
{'loss': 0.0838, 'learning_rate': 5e-05, 'epoch': 3.39}

 68%|██████▊   | 16560/24420 [6:54:24<3:15:52,  1.50s/it]
 68%|██████▊   | 16561/24420 [6:54:26<3:16:40,  1.50s/it]
 68%|██████▊   | 16562/24420 [6:54:27<3:16:09,  1.50s/it]
 68%|██████▊   | 16563/24420 [6:54:29<3:15:53,  1.50s/it]
 68%|██████▊   | 16564/24420 [6:54:30<3:15:45,  1.50s/it]
 68%|██████▊   | 16565/24420 [6:54:32<3:15:39,  1.49s/it]
                                                         
{'loss': 0.0769, 'learning_rate': 5e-05, 'epoch': 3.39}

 68%|██████▊   | 16565/24420 [6:54:32<3:15:39,  1.49s/it]
 68%|██████▊   | 16566/24420 [6:54:33<3:16:02,  1.50s/it]
 68%|██████▊   | 16567/24420 [6:54:35<3:16:16,  1.50s/it]
 68%|██████▊   | 16568/24420 [6:54:36<3:15:55,  1.50s/it]
 68%|██████▊   | 16569/24420 [6:54:38<3:15:31,  1.49s/it]
 68%|██████▊   | 16570/24420 [6:54:39<3:15:20,  1.49s/it]
                                                         
{'loss': 0.0824, 'learning_rate': 5e-05, 'epoch': 3.39}

 68%|██████▊   | 16570/24420 [6:54:39<3:15:20,  1.49s/it]
 68%|██████▊   | 16571/24420 [6:54:41<3:15:26,  1.49s/it]
 68%|██████▊   | 16572/24420 [6:54:42<3:15:38,  1.50s/it]
 68%|██████▊   | 16573/24420 [6:54:44<3:15:38,  1.50s/it]
 68%|██████▊   | 16574/24420 [6:54:45<3:15:16,  1.49s/it]
 68%|██████▊   | 16575/24420 [6:54:47<3:15:22,  1.49s/it]
                                                         
{'loss': 0.1019, 'learning_rate': 5e-05, 'epoch': 3.39}

 68%|██████▊   | 16575/24420 [6:54:47<3:15:22,  1.49s/it]
 68%|██████▊   | 16576/24420 [6:54:48<3:15:35,  1.50s/it]
 68%|██████▊   | 16577/24420 [6:54:50<3:15:30,  1.50s/it]
 68%|██████▊   | 16578/24420 [6:54:51<3:15:14,  1.49s/it]
 68%|██████▊   | 16579/24420 [6:54:53<3:15:13,  1.49s/it]
 68%|██████▊   | 16580/24420 [6:54:54<3:15:13,  1.49s/it]
                                                         
{'loss': 0.0827, 'learning_rate': 5e-05, 'epoch': 3.39}

 68%|██████▊   | 16580/24420 [6:54:54<3:15:13,  1.49s/it]
 68%|██████▊   | 16581/24420 [6:54:56<3:15:36,  1.50s/it]
 68%|██████▊   | 16582/24420 [6:54:57<3:15:43,  1.50s/it]
 68%|██████▊   | 16583/24420 [6:54:59<3:15:20,  1.50s/it]
 68%|██████▊   | 16584/24420 [6:55:00<3:14:52,  1.49s/it]
 68%|██████▊   | 16585/24420 [6:55:02<3:14:34,  1.49s/it]
                                                         
{'loss': 0.079, 'learning_rate': 5e-05, 'epoch': 3.4}

 68%|██████▊   | 16585/24420 [6:55:02<3:14:34,  1.49s/it]
 68%|██████▊   | 16586/24420 [6:55:03<3:15:07,  1.49s/it]
 68%|██████▊   | 16587/24420 [6:55:05<3:15:23,  1.50s/it]
 68%|██████▊   | 16588/24420 [6:55:06<3:15:32,  1.50s/it]
 68%|██████▊   | 16589/24420 [6:55:08<3:14:55,  1.49s/it]
 68%|██████▊   | 16590/24420 [6:55:09<3:14:56,  1.49s/it]
                                                         
{'loss': 0.0899, 'learning_rate': 5e-05, 'epoch': 3.4}

 68%|██████▊   | 16590/24420 [6:55:09<3:14:56,  1.49s/it]
 68%|██████▊   | 16591/24420 [6:55:11<3:15:00,  1.49s/it]
 68%|██████▊   | 16592/24420 [6:55:12<3:14:53,  1.49s/it]
 68%|██████▊   | 16593/24420 [6:55:14<3:14:39,  1.49s/it]
 68%|██████▊   | 16594/24420 [6:55:15<3:14:40,  1.49s/it]
 68%|██████▊   | 16595/24420 [6:55:17<3:14:22,  1.49s/it]
                                                         
{'loss': 0.0773, 'learning_rate': 5e-05, 'epoch': 3.4}

 68%|██████▊   | 16595/24420 [6:55:17<3:14:22,  1.49s/it]
 68%|██████▊   | 16596/24420 [6:55:18<3:14:41,  1.49s/it]
 68%|██████▊   | 16597/24420 [6:55:20<3:14:48,  1.49s/it]
 68%|██████▊   | 16598/24420 [6:55:21<3:14:37,  1.49s/it]
 68%|██████▊   | 16599/24420 [6:55:23<3:14:37,  1.49s/it]
 68%|██████▊   | 16600/24420 [6:55:24<3:15:00,  1.50s/it]
                                                         
{'loss': 0.0876, 'learning_rate': 5e-05, 'epoch': 3.4}

 68%|██████▊   | 16600/24420 [6:55:24<3:15:00,  1.50s/it]
 68%|██████▊   | 16601/24420 [6:55:29<5:21:27,  2.47s/it]
 68%|██████▊   | 16602/24420 [6:55:30<4:42:58,  2.17s/it]
 68%|██████▊   | 16603/24420 [6:55:32<4:16:34,  1.97s/it]
 68%|██████▊   | 16604/24420 [6:55:33<3:57:56,  1.83s/it]
 68%|██████▊   | 16605/24420 [6:55:35<3:45:15,  1.73s/it]
                                                         
{'loss': 0.0864, 'learning_rate': 5e-05, 'epoch': 3.4}

 68%|██████▊   | 16605/24420 [6:55:35<3:45:15,  1.73s/it]
 68%|██████▊   | 16606/24420 [6:55:36<3:36:12,  1.66s/it]
 68%|██████▊   | 16607/24420 [6:55:38<3:29:22,  1.61s/it]
 68%|██████▊   | 16608/24420 [6:55:39<3:24:39,  1.57s/it]
 68%|██████▊   | 16609/24420 [6:55:41<3:21:16,  1.55s/it]
 68%|██████▊   | 16610/24420 [6:55:42<3:19:05,  1.53s/it]
                                                         
{'loss': 0.0759, 'learning_rate': 5e-05, 'epoch': 3.4}

 68%|██████▊   | 16610/24420 [6:55:42<3:19:05,  1.53s/it]
 68%|██████▊   | 16611/24420 [6:55:44<3:17:24,  1.52s/it]
 68%|██████▊   | 16612/24420 [6:55:45<3:16:04,  1.51s/it]
 68%|██████▊   | 16613/24420 [6:55:47<3:15:02,  1.50s/it]
 68%|██████▊   | 16614/24420 [6:55:48<3:14:47,  1.50s/it]
 68%|██████▊   | 16615/24420 [6:55:50<3:14:38,  1.50s/it]
                                                         
{'loss': 0.0665, 'learning_rate': 5e-05, 'epoch': 3.4}

 68%|██████▊   | 16615/24420 [6:55:50<3:14:38,  1.50s/it]
 68%|██████▊   | 16616/24420 [6:55:51<3:14:34,  1.50s/it]
 68%|██████▊   | 16617/24420 [6:55:53<3:14:10,  1.49s/it]
 68%|██████▊   | 16618/24420 [6:55:54<3:14:03,  1.49s/it]
 68%|██████▊   | 16619/24420 [6:55:56<3:14:05,  1.49s/it]
 68%|██████▊   | 16620/24420 [6:55:57<3:14:02,  1.49s/it]
                                                         
{'loss': 0.0863, 'learning_rate': 5e-05, 'epoch': 3.4}

 68%|██████▊   | 16620/24420 [6:55:57<3:14:02,  1.49s/it]
 68%|██████▊   | 16621/24420 [6:55:59<3:13:25,  1.49s/it]
 68%|██████▊   | 16622/24420 [6:56:00<3:13:48,  1.49s/it]
 68%|██████▊   | 16623/24420 [6:56:02<3:13:30,  1.49s/it]
 68%|██████▊   | 16624/24420 [6:56:03<3:13:32,  1.49s/it]
 68%|██████▊   | 16625/24420 [6:56:05<3:13:26,  1.49s/it]
                                                         
{'loss': 0.0799, 'learning_rate': 5e-05, 'epoch': 3.4}

 68%|██████▊   | 16625/24420 [6:56:05<3:13:26,  1.49s/it]
 68%|██████▊   | 16626/24420 [6:56:06<3:13:38,  1.49s/it]
 68%|██████▊   | 16627/24420 [6:56:08<3:13:45,  1.49s/it]
 68%|██████▊   | 16628/24420 [6:56:09<3:13:44,  1.49s/it]
 68%|██████▊   | 16629/24420 [6:56:11<3:13:38,  1.49s/it]
 68%|██████▊   | 16630/24420 [6:56:12<3:13:15,  1.49s/it]
                                                         
{'loss': 0.0784, 'learning_rate': 5e-05, 'epoch': 3.4}

 68%|██████▊   | 16630/24420 [6:56:12<3:13:15,  1.49s/it]
 68%|██████▊   | 16631/24420 [6:56:14<3:13:09,  1.49s/it]
 68%|██████▊   | 16632/24420 [6:56:15<3:13:01,  1.49s/it]
 68%|██████▊   | 16633/24420 [6:56:17<3:13:15,  1.49s/it]
 68%|██████▊   | 16634/24420 [6:56:18<3:13:25,  1.49s/it]
 68%|██████▊   | 16635/24420 [6:56:19<3:13:17,  1.49s/it]
                                                         
{'loss': 0.0678, 'learning_rate': 5e-05, 'epoch': 3.41}

 68%|██████▊   | 16635/24420 [6:56:19<3:13:17,  1.49s/it]
 68%|██████▊   | 16636/24420 [6:56:21<3:13:29,  1.49s/it]
 68%|██████▊   | 16637/24420 [6:56:22<3:13:48,  1.49s/it]
 68%|██████▊   | 16638/24420 [6:56:24<3:14:11,  1.50s/it]
 68%|██████▊   | 16639/24420 [6:56:25<3:14:01,  1.50s/it]
 68%|██████▊   | 16640/24420 [6:56:27<3:13:54,  1.50s/it]
                                                         
{'loss': 0.0775, 'learning_rate': 5e-05, 'epoch': 3.41}

 68%|██████▊   | 16640/24420 [6:56:27<3:13:54,  1.50s/it]
 68%|██████▊   | 16641/24420 [6:56:28<3:13:42,  1.49s/it]
 68%|██████▊   | 16642/24420 [6:56:30<3:13:15,  1.49s/it]
 68%|██████▊   | 16643/24420 [6:56:31<3:13:00,  1.49s/it]
 68%|██████▊   | 16644/24420 [6:56:33<3:13:32,  1.49s/it]
 68%|██████▊   | 16645/24420 [6:56:34<3:13:32,  1.49s/it]
                                                         
{'loss': 0.0776, 'learning_rate': 5e-05, 'epoch': 3.41}

 68%|██████▊   | 16645/24420 [6:56:34<3:13:32,  1.49s/it]
 68%|██████▊   | 16646/24420 [6:56:36<3:13:26,  1.49s/it]
 68%|██████▊   | 16647/24420 [6:56:37<3:13:00,  1.49s/it]
 68%|██████▊   | 16648/24420 [6:56:39<3:13:00,  1.49s/it]
 68%|██████▊   | 16649/24420 [6:56:40<3:13:02,  1.49s/it]
 68%|██████▊   | 16650/24420 [6:56:42<3:13:31,  1.49s/it]
                                                         
{'loss': 0.0742, 'learning_rate': 5e-05, 'epoch': 3.41}

 68%|██████▊   | 16650/24420 [6:56:42<3:13:31,  1.49s/it]
 68%|██████▊   | 16651/24420 [6:56:43<3:13:30,  1.49s/it]
 68%|██████▊   | 16652/24420 [6:56:45<3:13:07,  1.49s/it]
 68%|██████▊   | 16653/24420 [6:56:46<3:13:31,  1.49s/it]
 68%|██████▊   | 16654/24420 [6:56:48<3:13:17,  1.49s/it]
 68%|██████▊   | 16655/24420 [6:56:49<3:13:04,  1.49s/it]
                                                         
{'loss': 0.0751, 'learning_rate': 5e-05, 'epoch': 3.41}

 68%|██████▊   | 16655/24420 [6:56:49<3:13:04,  1.49s/it]
 68%|██████▊   | 16656/24420 [6:56:51<3:13:09,  1.49s/it]
 68%|██████▊   | 16657/24420 [6:56:52<3:12:57,  1.49s/it]
 68%|██████▊   | 16658/24420 [6:56:54<3:12:46,  1.49s/it]
 68%|██████▊   | 16659/24420 [6:56:55<3:13:01,  1.49s/it]
 68%|██████▊   | 16660/24420 [6:56:57<3:13:02,  1.49s/it]
                                                         
{'loss': 0.0923, 'learning_rate': 5e-05, 'epoch': 3.41}

 68%|██████▊   | 16660/24420 [6:56:57<3:13:02,  1.49s/it]
 68%|██████▊   | 16661/24420 [6:56:58<3:12:55,  1.49s/it]
 68%|██████▊   | 16662/24420 [6:57:00<3:12:42,  1.49s/it]
 68%|██████▊   | 16663/24420 [6:57:01<3:12:49,  1.49s/it]
 68%|██████▊   | 16664/24420 [6:57:03<3:13:14,  1.49s/it]
 68%|██████▊   | 16665/24420 [6:57:04<3:13:22,  1.50s/it]
                                                         
{'loss': 0.0947, 'learning_rate': 5e-05, 'epoch': 3.41}

 68%|██████▊   | 16665/24420 [6:57:04<3:13:22,  1.50s/it]
 68%|██████▊   | 16666/24420 [6:57:06<3:13:25,  1.50s/it]
 68%|██████▊   | 16667/24420 [6:57:07<3:13:00,  1.49s/it]
 68%|██████▊   | 16668/24420 [6:57:09<3:12:29,  1.49s/it]
 68%|██████▊   | 16669/24420 [6:57:10<3:12:18,  1.49s/it]
 68%|██████▊   | 16670/24420 [6:57:12<3:12:18,  1.49s/it]
                                                         
{'loss': 0.0865, 'learning_rate': 5e-05, 'epoch': 3.41}

 68%|██████▊   | 16670/24420 [6:57:12<3:12:18,  1.49s/it]
 68%|██████▊   | 16671/24420 [6:57:13<3:12:15,  1.49s/it]
 68%|██████▊   | 16672/24420 [6:57:15<3:12:05,  1.49s/it]
 68%|██████▊   | 16673/24420 [6:57:16<3:12:14,  1.49s/it]
 68%|██████▊   | 16674/24420 [6:57:18<3:12:06,  1.49s/it]
 68%|██████▊   | 16675/24420 [6:57:19<3:12:13,  1.49s/it]
                                                         
{'loss': 0.0904, 'learning_rate': 5e-05, 'epoch': 3.41}

 68%|██████▊   | 16675/24420 [6:57:19<3:12:13,  1.49s/it]
 68%|██████▊   | 16676/24420 [6:57:21<3:12:30,  1.49s/it]
 68%|██████▊   | 16677/24420 [6:57:22<3:12:21,  1.49s/it]
 68%|██████▊   | 16678/24420 [6:57:24<3:12:12,  1.49s/it]
 68%|██████▊   | 16679/24420 [6:57:25<3:12:10,  1.49s/it]
 68%|██████▊   | 16680/24420 [6:57:27<3:12:50,  1.49s/it]
                                                         
{'loss': 0.0747, 'learning_rate': 5e-05, 'epoch': 3.42}

 68%|██████▊   | 16680/24420 [6:57:27<3:12:50,  1.49s/it]
 68%|██████▊   | 16681/24420 [6:57:28<3:12:32,  1.49s/it]
 68%|██████▊   | 16682/24420 [6:57:30<3:12:11,  1.49s/it]
 68%|██████▊   | 16683/24420 [6:57:31<3:12:19,  1.49s/it]
 68%|██████▊   | 16684/24420 [6:57:33<3:12:04,  1.49s/it]
 68%|██████▊   | 16685/24420 [6:57:34<3:12:13,  1.49s/it]
                                                         
{'loss': 0.0717, 'learning_rate': 5e-05, 'epoch': 3.42}

 68%|██████▊   | 16685/24420 [6:57:34<3:12:13,  1.49s/it]
 68%|██████▊   | 16686/24420 [6:57:36<3:12:04,  1.49s/it]
 68%|██████▊   | 16687/24420 [6:57:37<3:11:45,  1.49s/it]
 68%|██████▊   | 16688/24420 [6:57:39<3:11:46,  1.49s/it]
 68%|██████▊   | 16689/24420 [6:57:40<3:12:50,  1.50s/it]
 68%|██████▊   | 16690/24420 [6:57:42<3:12:14,  1.49s/it]
                                                         
{'loss': 0.0872, 'learning_rate': 5e-05, 'epoch': 3.42}

 68%|██████▊   | 16690/24420 [6:57:42<3:12:14,  1.49s/it]
 68%|██████▊   | 16691/24420 [6:57:43<3:12:34,  1.49s/it]
 68%|██████▊   | 16692/24420 [6:57:45<3:12:15,  1.49s/it]
 68%|██████▊   | 16693/24420 [6:57:46<3:12:16,  1.49s/it]
 68%|██████▊   | 16694/24420 [6:57:48<3:11:51,  1.49s/it]
 68%|██████▊   | 16695/24420 [6:57:49<3:12:15,  1.49s/it]
                                                         
{'loss': 0.1049, 'learning_rate': 5e-05, 'epoch': 3.42}

 68%|██████▊   | 16695/24420 [6:57:49<3:12:15,  1.49s/it]
 68%|██████▊   | 16696/24420 [6:57:51<3:12:02,  1.49s/it]
 68%|██████▊   | 16697/24420 [6:57:52<3:12:09,  1.49s/it]
 68%|██████▊   | 16698/24420 [6:57:53<3:12:11,  1.49s/it]
 68%|██████▊   | 16699/24420 [6:57:55<3:11:49,  1.49s/it]
 68%|██████▊   | 16700/24420 [6:57:56<3:11:49,  1.49s/it]
                                                         
{'loss': 0.0697, 'learning_rate': 5e-05, 'epoch': 3.42}

 68%|██████▊   | 16700/24420 [6:57:56<3:11:49,  1.49s/it]
 68%|██████▊   | 16701/24420 [6:57:58<3:11:59,  1.49s/it]
 68%|██████▊   | 16702/24420 [6:57:59<3:12:26,  1.50s/it]
 68%|██████▊   | 16703/24420 [6:58:01<3:12:07,  1.49s/it]
 68%|██████▊   | 16704/24420 [6:58:02<3:12:19,  1.50s/it]
 68%|██████▊   | 16705/24420 [6:58:04<3:12:14,  1.50s/it]
                                                         
{'loss': 0.0759, 'learning_rate': 5e-05, 'epoch': 3.42}

 68%|██████▊   | 16705/24420 [6:58:04<3:12:14,  1.50s/it]
 68%|██████▊   | 16706/24420 [6:58:05<3:12:19,  1.50s/it]
 68%|██████▊   | 16707/24420 [6:58:07<3:12:11,  1.50s/it]
 68%|██████▊   | 16708/24420 [6:58:08<3:12:00,  1.49s/it]
 68%|██████▊   | 16709/24420 [6:58:10<3:11:38,  1.49s/it]
 68%|██████▊   | 16710/24420 [6:58:11<3:11:21,  1.49s/it]
                                                         
{'loss': 0.0709, 'learning_rate': 5e-05, 'epoch': 3.42}

 68%|██████▊   | 16710/24420 [6:58:11<3:11:21,  1.49s/it]
 68%|██████▊   | 16711/24420 [6:58:13<3:11:24,  1.49s/it]
 68%|██████▊   | 16712/24420 [6:58:14<3:11:38,  1.49s/it]
 68%|██████▊   | 16713/24420 [6:58:16<3:11:45,  1.49s/it]
 68%|██████▊   | 16714/24420 [6:58:17<3:11:20,  1.49s/it]
 68%|██████▊   | 16715/24420 [6:58:19<3:11:04,  1.49s/it]
                                                         
{'loss': 0.076, 'learning_rate': 5e-05, 'epoch': 3.42}

 68%|██████▊   | 16715/24420 [6:58:19<3:11:04,  1.49s/it]
 68%|██████▊   | 16716/24420 [6:58:20<3:11:30,  1.49s/it]
 68%|██████▊   | 16717/24420 [6:58:22<3:11:32,  1.49s/it]
 68%|██████▊   | 16718/24420 [6:58:23<3:11:14,  1.49s/it]
 68%|██████▊   | 16719/24420 [6:58:25<3:11:14,  1.49s/it]
 68%|██████▊   | 16720/24420 [6:58:26<3:10:56,  1.49s/it]
                                                         
{'loss': 0.0849, 'learning_rate': 5e-05, 'epoch': 3.42}

 68%|██████▊   | 16720/24420 [6:58:26<3:10:56,  1.49s/it]
 68%|██████▊   | 16721/24420 [6:58:28<3:11:04,  1.49s/it]
 68%|██████▊   | 16722/24420 [6:58:29<3:11:09,  1.49s/it]
 68%|██████▊   | 16723/24420 [6:58:31<3:11:05,  1.49s/it]
 68%|██████▊   | 16724/24420 [6:58:32<3:11:28,  1.49s/it]
 68%|██████▊   | 16725/24420 [6:58:34<3:11:23,  1.49s/it]
                                                         
{'loss': 0.096, 'learning_rate': 5e-05, 'epoch': 3.42}

 68%|██████▊   | 16725/24420 [6:58:34<3:11:23,  1.49s/it]
 68%|██████▊   | 16726/24420 [6:58:35<3:11:27,  1.49s/it]
 68%|██████▊   | 16727/24420 [6:58:37<3:11:11,  1.49s/it]
 69%|██████▊   | 16728/24420 [6:58:38<3:11:01,  1.49s/it]
 69%|██████▊   | 16729/24420 [6:58:40<3:10:56,  1.49s/it]
 69%|██████▊   | 16730/24420 [6:58:41<3:11:13,  1.49s/it]
                                                         
{'loss': 0.0833, 'learning_rate': 5e-05, 'epoch': 3.43}

 69%|██████▊   | 16730/24420 [6:58:41<3:11:13,  1.49s/it]
 69%|██████▊   | 16731/24420 [6:58:43<3:11:36,  1.50s/it]
 69%|██████▊   | 16732/24420 [6:58:44<3:11:31,  1.49s/it]
 69%|██████▊   | 16733/24420 [6:58:46<3:11:24,  1.49s/it]
 69%|██████▊   | 16734/24420 [6:58:47<3:11:12,  1.49s/it]
 69%|██████▊   | 16735/24420 [6:58:49<3:11:09,  1.49s/it]
                                                         
{'loss': 0.0696, 'learning_rate': 5e-05, 'epoch': 3.43}

 69%|██████▊   | 16735/24420 [6:58:49<3:11:09,  1.49s/it]
 69%|██████▊   | 16736/24420 [6:58:50<3:11:35,  1.50s/it]
 69%|██████▊   | 16737/24420 [6:58:52<3:11:10,  1.49s/it]
 69%|██████▊   | 16738/24420 [6:58:53<3:10:40,  1.49s/it]
 69%|██████▊   | 16739/24420 [6:58:55<3:10:55,  1.49s/it]
 69%|██████▊   | 16740/24420 [6:58:56<3:10:51,  1.49s/it]
                                                         
{'loss': 0.073, 'learning_rate': 5e-05, 'epoch': 3.43}

 69%|██████▊   | 16740/24420 [6:58:56<3:10:51,  1.49s/it]
 69%|██████▊   | 16741/24420 [6:58:58<3:10:58,  1.49s/it]
 69%|██████▊   | 16742/24420 [6:58:59<3:10:41,  1.49s/it]
 69%|██████▊   | 16743/24420 [6:59:01<3:10:28,  1.49s/it]
 69%|██████▊   | 16744/24420 [6:59:02<3:10:40,  1.49s/it]
 69%|██████▊   | 16745/24420 [6:59:04<3:11:12,  1.49s/it]
                                                         
{'loss': 0.0738, 'learning_rate': 5e-05, 'epoch': 3.43}

 69%|██████▊   | 16745/24420 [6:59:04<3:11:12,  1.49s/it]
 69%|██████▊   | 16746/24420 [6:59:05<3:11:01,  1.49s/it]
 69%|██████▊   | 16747/24420 [6:59:07<3:10:34,  1.49s/it]
 69%|██████▊   | 16748/24420 [6:59:08<3:10:15,  1.49s/it]
 69%|██████▊   | 16749/24420 [6:59:10<3:10:43,  1.49s/it]
 69%|██████▊   | 16750/24420 [6:59:11<3:10:56,  1.49s/it]
                                                         
{'loss': 0.0839, 'learning_rate': 5e-05, 'epoch': 3.43}

 69%|██████▊   | 16750/24420 [6:59:11<3:10:56,  1.49s/it]
 69%|██████▊   | 16751/24420 [6:59:13<3:10:21,  1.49s/it]
 69%|██████▊   | 16752/24420 [6:59:14<3:10:46,  1.49s/it]
 69%|██████▊   | 16753/24420 [6:59:16<3:10:36,  1.49s/it]
 69%|██████▊   | 16754/24420 [6:59:17<3:10:57,  1.49s/it]
 69%|██████▊   | 16755/24420 [6:59:19<3:10:56,  1.49s/it]
                                                         
{'loss': 0.0758, 'learning_rate': 5e-05, 'epoch': 3.43}

 69%|██████▊   | 16755/24420 [6:59:19<3:10:56,  1.49s/it]
 69%|██████▊   | 16756/24420 [6:59:20<3:10:56,  1.49s/it]
 69%|██████▊   | 16757/24420 [6:59:22<3:10:42,  1.49s/it]
 69%|██████▊   | 16758/24420 [6:59:23<3:10:21,  1.49s/it]
 69%|██████▊   | 16759/24420 [6:59:24<3:10:14,  1.49s/it]
 69%|██████▊   | 16760/24420 [6:59:26<3:10:23,  1.49s/it]
                                                         
{'loss': 0.0752, 'learning_rate': 5e-05, 'epoch': 3.43}

 69%|██████▊   | 16760/24420 [6:59:26<3:10:23,  1.49s/it]
 69%|██████▊   | 16761/24420 [6:59:27<3:10:07,  1.49s/it]
 69%|██████▊   | 16762/24420 [6:59:29<3:10:05,  1.49s/it]
 69%|██████▊   | 16763/24420 [6:59:30<3:10:18,  1.49s/it]
 69%|██████▊   | 16764/24420 [6:59:32<3:10:13,  1.49s/it]
 69%|██████▊   | 16765/24420 [6:59:33<3:10:06,  1.49s/it]
                                                         
{'loss': 0.0744, 'learning_rate': 5e-05, 'epoch': 3.43}

 69%|██████▊   | 16765/24420 [6:59:33<3:10:06,  1.49s/it]
 69%|██████▊   | 16766/24420 [6:59:35<3:10:29,  1.49s/it]
 69%|██████▊   | 16767/24420 [6:59:36<3:10:04,  1.49s/it]
 69%|██████▊   | 16768/24420 [6:59:38<3:10:27,  1.49s/it]
 69%|██████▊   | 16769/24420 [6:59:39<3:10:22,  1.49s/it]
 69%|██████▊   | 16770/24420 [6:59:41<3:11:16,  1.50s/it]
                                                         
{'loss': 0.0906, 'learning_rate': 5e-05, 'epoch': 3.43}

 69%|██████▊   | 16770/24420 [6:59:41<3:11:16,  1.50s/it]
 69%|██████▊   | 16771/24420 [6:59:42<3:11:14,  1.50s/it]
 69%|██████▊   | 16772/24420 [6:59:44<3:10:35,  1.50s/it]
 69%|██████▊   | 16773/24420 [6:59:45<3:10:01,  1.49s/it]
 69%|██████▊   | 16774/24420 [6:59:47<3:09:51,  1.49s/it]
 69%|██████▊   | 16775/24420 [6:59:48<3:09:37,  1.49s/it]
                                                         
{'loss': 0.0855, 'learning_rate': 5e-05, 'epoch': 3.43}

 69%|██████▊   | 16775/24420 [6:59:48<3:09:37,  1.49s/it]
 69%|██████▊   | 16776/24420 [6:59:50<3:09:28,  1.49s/it]
 69%|██████▊   | 16777/24420 [6:59:51<3:09:26,  1.49s/it]
 69%|██████▊   | 16778/24420 [6:59:53<3:09:24,  1.49s/it]
 69%|██████▊   | 16779/24420 [6:59:54<3:09:26,  1.49s/it]
 69%|██████▊   | 16780/24420 [6:59:56<3:09:34,  1.49s/it]
                                                         
{'loss': 0.0876, 'learning_rate': 5e-05, 'epoch': 3.44}

 69%|██████▊   | 16780/24420 [6:59:56<3:09:34,  1.49s/it]
 69%|██████▊   | 16781/24420 [6:59:57<3:09:40,  1.49s/it]
 69%|██████▊   | 16782/24420 [6:59:59<3:09:41,  1.49s/it]
 69%|██████▊   | 16783/24420 [7:00:00<3:09:35,  1.49s/it]
 69%|██████▊   | 16784/24420 [7:00:02<3:09:18,  1.49s/it]
 69%|██████▊   | 16785/24420 [7:00:03<3:09:15,  1.49s/it]
                                                         
{'loss': 0.0793, 'learning_rate': 5e-05, 'epoch': 3.44}

 69%|██████▊   | 16785/24420 [7:00:03<3:09:15,  1.49s/it]
 69%|██████▊   | 16786/24420 [7:00:05<3:09:36,  1.49s/it]
 69%|██████▊   | 16787/24420 [7:00:06<3:09:37,  1.49s/it]
 69%|██████▊   | 16788/24420 [7:00:08<3:10:03,  1.49s/it]
 69%|██████▉   | 16789/24420 [7:00:09<3:10:02,  1.49s/it]
 69%|██████▉   | 16790/24420 [7:00:11<3:09:55,  1.49s/it]
                                                         
{'loss': 0.0974, 'learning_rate': 5e-05, 'epoch': 3.44}

 69%|██████▉   | 16790/24420 [7:00:11<3:09:55,  1.49s/it]
 69%|██████▉   | 16791/24420 [7:00:12<3:10:19,  1.50s/it]
 69%|██████▉   | 16792/24420 [7:00:14<3:10:03,  1.49s/it]
 69%|██████▉   | 16793/24420 [7:00:15<3:09:54,  1.49s/it]
 69%|██████▉   | 16794/24420 [7:00:17<3:09:30,  1.49s/it]
 69%|██████▉   | 16795/24420 [7:00:18<3:09:41,  1.49s/it]
                                                         
{'loss': 0.0767, 'learning_rate': 5e-05, 'epoch': 3.44}

 69%|██████▉   | 16795/24420 [7:00:18<3:09:41,  1.49s/it]
 69%|██████▉   | 16796/24420 [7:00:20<3:09:49,  1.49s/it]
 69%|██████▉   | 16797/24420 [7:00:21<3:09:26,  1.49s/it]
 69%|██████▉   | 16798/24420 [7:00:23<3:09:05,  1.49s/it]
 69%|██████▉   | 16799/24420 [7:00:24<3:09:24,  1.49s/it]
 69%|██████▉   | 16800/24420 [7:00:26<3:09:15,  1.49s/it]
                                                         
{'loss': 0.0884, 'learning_rate': 5e-05, 'epoch': 3.44}

 69%|██████▉   | 16800/24420 [7:00:26<3:09:15,  1.49s/it]
 69%|██████▉   | 16801/24420 [7:00:30<5:05:57,  2.41s/it]
 69%|██████▉   | 16802/24420 [7:00:32<4:30:54,  2.13s/it]
 69%|██████▉   | 16803/24420 [7:00:33<4:06:16,  1.94s/it]
 69%|██████▉   | 16804/24420 [7:00:35<3:49:00,  1.80s/it]
 69%|██████▉   | 16805/24420 [7:00:36<3:36:51,  1.71s/it]
                                                         
{'loss': 0.0903, 'learning_rate': 5e-05, 'epoch': 3.44}

 69%|██████▉   | 16805/24420 [7:00:36<3:36:51,  1.71s/it]
 69%|██████▉   | 16806/24420 [7:00:38<3:28:51,  1.65s/it]
 69%|██████▉   | 16807/24420 [7:00:39<3:22:31,  1.60s/it]
 69%|██████▉   | 16808/24420 [7:00:41<3:18:34,  1.57s/it]
 69%|██████▉   | 16809/24420 [7:00:42<3:16:09,  1.55s/it]
 69%|██████▉   | 16810/24420 [7:00:44<3:14:19,  1.53s/it]
                                                         
{'loss': 0.0802, 'learning_rate': 5e-05, 'epoch': 3.44}

 69%|██████▉   | 16810/24420 [7:00:44<3:14:19,  1.53s/it]
 69%|██████▉   | 16811/24420 [7:00:45<3:12:52,  1.52s/it]
 69%|██████▉   | 16812/24420 [7:00:47<3:11:47,  1.51s/it]
 69%|██████▉   | 16813/24420 [7:00:48<3:10:43,  1.50s/it]
 69%|██████▉   | 16814/24420 [7:00:50<3:09:59,  1.50s/it]
 69%|██████▉   | 16815/24420 [7:00:51<3:09:25,  1.49s/it]
                                                         
{'loss': 0.0839, 'learning_rate': 5e-05, 'epoch': 3.44}

 69%|██████▉   | 16815/24420 [7:00:51<3:09:25,  1.49s/it]
 69%|██████▉   | 16816/24420 [7:00:53<3:09:25,  1.49s/it]
 69%|██████▉   | 16817/24420 [7:00:54<3:09:01,  1.49s/it]
 69%|██████▉   | 16818/24420 [7:00:56<3:08:44,  1.49s/it]
 69%|██████▉   | 16819/24420 [7:00:57<3:17:02,  1.56s/it]
 69%|██████▉   | 16820/24420 [7:00:59<3:14:46,  1.54s/it]
                                                         
{'loss': 0.0811, 'learning_rate': 5e-05, 'epoch': 3.44}

 69%|██████▉   | 16820/24420 [7:00:59<3:14:46,  1.54s/it]
 69%|██████▉   | 16821/24420 [7:01:00<3:13:13,  1.53s/it]
 69%|██████▉   | 16822/24420 [7:01:02<3:12:01,  1.52s/it]
 69%|██████▉   | 16823/24420 [7:01:03<3:11:01,  1.51s/it]
 69%|██████▉   | 16824/24420 [7:01:05<3:10:41,  1.51s/it]
 69%|██████▉   | 16825/24420 [7:01:06<3:09:52,  1.50s/it]
                                                         
{'loss': 0.0848, 'learning_rate': 5e-05, 'epoch': 3.44}

 69%|██████▉   | 16825/24420 [7:01:06<3:09:52,  1.50s/it]
 69%|██████▉   | 16826/24420 [7:01:08<3:09:22,  1.50s/it]
 69%|██████▉   | 16827/24420 [7:01:09<3:09:22,  1.50s/it]
 69%|██████▉   | 16828/24420 [7:01:11<3:08:47,  1.49s/it]
 69%|██████▉   | 16829/24420 [7:01:12<3:08:39,  1.49s/it]
 69%|██████▉   | 16830/24420 [7:01:14<3:08:34,  1.49s/it]
                                                         
{'loss': 0.0702, 'learning_rate': 5e-05, 'epoch': 3.45}

 69%|██████▉   | 16830/24420 [7:01:14<3:08:34,  1.49s/it]
 69%|██████▉   | 16831/24420 [7:01:15<3:08:38,  1.49s/it]
 69%|██████▉   | 16832/24420 [7:01:17<3:08:37,  1.49s/it]
 69%|██████▉   | 16833/24420 [7:01:18<3:08:27,  1.49s/it]
 69%|██████▉   | 16834/24420 [7:01:20<3:08:26,  1.49s/it]
 69%|██████▉   | 16835/24420 [7:01:21<3:08:17,  1.49s/it]
                                                         
{'loss': 0.0752, 'learning_rate': 5e-05, 'epoch': 3.45}

 69%|██████▉   | 16835/24420 [7:01:21<3:08:17,  1.49s/it]
 69%|██████▉   | 16836/24420 [7:01:23<3:08:19,  1.49s/it]
 69%|██████▉   | 16837/24420 [7:01:24<3:08:34,  1.49s/it]
 69%|██████▉   | 16838/24420 [7:01:26<3:08:50,  1.49s/it]
 69%|██████▉   | 16839/24420 [7:01:27<3:08:25,  1.49s/it]
 69%|██████▉   | 16840/24420 [7:01:29<3:08:14,  1.49s/it]
                                                         
{'loss': 0.0836, 'learning_rate': 5e-05, 'epoch': 3.45}

 69%|██████▉   | 16840/24420 [7:01:29<3:08:14,  1.49s/it]
 69%|██████▉   | 16841/24420 [7:01:30<3:08:05,  1.49s/it]
 69%|██████▉   | 16842/24420 [7:01:32<3:08:20,  1.49s/it]
 69%|██████▉   | 16843/24420 [7:01:33<3:08:05,  1.49s/it]
 69%|██████▉   | 16844/24420 [7:01:35<3:08:07,  1.49s/it]
 69%|██████▉   | 16845/24420 [7:01:36<3:08:06,  1.49s/it]
                                                         
{'loss': 0.0816, 'learning_rate': 5e-05, 'epoch': 3.45}

 69%|██████▉   | 16845/24420 [7:01:36<3:08:06,  1.49s/it]
 69%|██████▉   | 16846/24420 [7:01:38<3:08:23,  1.49s/it]
 69%|██████▉   | 16847/24420 [7:01:39<3:08:08,  1.49s/it]
 69%|██████▉   | 16848/24420 [7:01:40<3:08:29,  1.49s/it]
 69%|██████▉   | 16849/24420 [7:01:42<3:08:12,  1.49s/it]
 69%|██████▉   | 16850/24420 [7:01:43<3:08:01,  1.49s/it]
                                                         
{'loss': 0.0814, 'learning_rate': 5e-05, 'epoch': 3.45}

 69%|██████▉   | 16850/24420 [7:01:43<3:08:01,  1.49s/it]
 69%|██████▉   | 16851/24420 [7:01:45<3:07:49,  1.49s/it]
 69%|██████▉   | 16852/24420 [7:01:46<3:07:43,  1.49s/it]
 69%|██████▉   | 16853/24420 [7:01:48<3:07:56,  1.49s/it]
 69%|██████▉   | 16854/24420 [7:01:49<3:07:42,  1.49s/it]
 69%|██████▉   | 16855/24420 [7:01:51<3:08:00,  1.49s/it]
                                                         
{'loss': 0.0823, 'learning_rate': 5e-05, 'epoch': 3.45}

 69%|██████▉   | 16855/24420 [7:01:51<3:08:00,  1.49s/it]
 69%|██████▉   | 16856/24420 [7:01:52<3:07:22,  1.49s/it]
 69%|██████▉   | 16857/24420 [7:01:54<3:07:20,  1.49s/it]
 69%|██████▉   | 16858/24420 [7:01:55<3:07:50,  1.49s/it]
 69%|██████▉   | 16859/24420 [7:01:57<3:07:54,  1.49s/it]
 69%|██████▉   | 16860/24420 [7:01:58<3:07:45,  1.49s/it]
                                                         
{'loss': 0.0751, 'learning_rate': 5e-05, 'epoch': 3.45}

 69%|██████▉   | 16860/24420 [7:01:58<3:07:45,  1.49s/it]
 69%|██████▉   | 16861/24420 [7:02:00<3:07:35,  1.49s/it]
 69%|██████▉   | 16862/24420 [7:02:01<3:07:30,  1.49s/it]
 69%|██████▉   | 16863/24420 [7:02:03<3:07:36,  1.49s/it]
 69%|██████▉   | 16864/24420 [7:02:04<3:07:32,  1.49s/it]
 69%|██████▉   | 16865/24420 [7:02:06<3:07:48,  1.49s/it]
                                                         
{'loss': 0.0817, 'learning_rate': 5e-05, 'epoch': 3.45}

 69%|██████▉   | 16865/24420 [7:02:06<3:07:48,  1.49s/it]
 69%|██████▉   | 16866/24420 [7:02:07<3:07:54,  1.49s/it]
 69%|██████▉   | 16867/24420 [7:02:09<3:07:45,  1.49s/it]
 69%|██████▉   | 16868/24420 [7:02:10<3:07:36,  1.49s/it]
 69%|██████▉   | 16869/24420 [7:02:12<3:07:15,  1.49s/it]
 69%|██████▉   | 16870/24420 [7:02:13<3:07:05,  1.49s/it]
                                                         
{'loss': 0.0765, 'learning_rate': 5e-05, 'epoch': 3.45}

 69%|██████▉   | 16870/24420 [7:02:13<3:07:05,  1.49s/it]
 69%|██████▉   | 16871/24420 [7:02:15<3:07:42,  1.49s/it]
 69%|██████▉   | 16872/24420 [7:02:16<3:07:31,  1.49s/it]
 69%|██████▉   | 16873/24420 [7:02:18<3:07:19,  1.49s/it]
 69%|██████▉   | 16874/24420 [7:02:19<3:07:02,  1.49s/it]
 69%|██████▉   | 16875/24420 [7:02:21<3:07:36,  1.49s/it]
                                                         
{'loss': 0.0784, 'learning_rate': 5e-05, 'epoch': 3.46}

 69%|██████▉   | 16875/24420 [7:02:21<3:07:36,  1.49s/it]
 69%|██████▉   | 16876/24420 [7:02:22<3:09:30,  1.51s/it]
 69%|██████▉   | 16877/24420 [7:02:24<3:11:35,  1.52s/it]
 69%|██████▉   | 16878/24420 [7:02:25<3:10:54,  1.52s/it]
 69%|██████▉   | 16879/24420 [7:02:27<3:09:56,  1.51s/it]
 69%|██████▉   | 16880/24420 [7:02:28<3:09:11,  1.51s/it]
                                                         
{'loss': 0.073, 'learning_rate': 5e-05, 'epoch': 3.46}

 69%|██████▉   | 16880/24420 [7:02:28<3:09:11,  1.51s/it]
 69%|██████▉   | 16881/24420 [7:02:30<3:08:45,  1.50s/it]
 69%|██████▉   | 16882/24420 [7:02:31<3:09:48,  1.51s/it]
 69%|██████▉   | 16883/24420 [7:02:33<3:10:00,  1.51s/it]
 69%|██████▉   | 16884/24420 [7:02:34<3:10:27,  1.52s/it]
 69%|██████▉   | 16885/24420 [7:02:36<3:10:17,  1.52s/it]
                                                         
{'loss': 0.0838, 'learning_rate': 5e-05, 'epoch': 3.46}

 69%|██████▉   | 16885/24420 [7:02:36<3:10:17,  1.52s/it]
 69%|██████▉   | 16886/24420 [7:02:37<3:10:06,  1.51s/it]
 69%|██████▉   | 16887/24420 [7:02:39<3:09:52,  1.51s/it]
 69%|██████▉   | 16888/24420 [7:02:40<3:09:37,  1.51s/it]
 69%|██████▉   | 16889/24420 [7:02:42<3:09:24,  1.51s/it]
 69%|██████▉   | 16890/24420 [7:02:43<3:09:18,  1.51s/it]
                                                         
{'loss': 0.0808, 'learning_rate': 5e-05, 'epoch': 3.46}

 69%|██████▉   | 16890/24420 [7:02:43<3:09:18,  1.51s/it]
 69%|██████▉   | 16891/24420 [7:02:45<3:09:01,  1.51s/it]
 69%|██████▉   | 16892/24420 [7:02:46<3:08:45,  1.50s/it]
 69%|██████▉   | 16893/24420 [7:02:48<3:09:10,  1.51s/it]
 69%|██████▉   | 16894/24420 [7:02:49<3:08:54,  1.51s/it]
 69%|██████▉   | 16895/24420 [7:02:51<3:08:36,  1.50s/it]
                                                         
{'loss': 0.0826, 'learning_rate': 5e-05, 'epoch': 3.46}

 69%|██████▉   | 16895/24420 [7:02:51<3:08:36,  1.50s/it]
 69%|██████▉   | 16896/24420 [7:02:52<3:09:16,  1.51s/it]
 69%|██████▉   | 16897/24420 [7:02:54<3:08:58,  1.51s/it]
 69%|██████▉   | 16898/24420 [7:02:55<3:09:16,  1.51s/it]
 69%|██████▉   | 16899/24420 [7:02:57<3:09:02,  1.51s/it]
 69%|██████▉   | 16900/24420 [7:02:58<3:09:08,  1.51s/it]
                                                         
{'loss': 0.0753, 'learning_rate': 5e-05, 'epoch': 3.46}

 69%|██████▉   | 16900/24420 [7:02:59<3:09:08,  1.51s/it]
 69%|██████▉   | 16901/24420 [7:03:00<3:09:17,  1.51s/it]
 69%|██████▉   | 16902/24420 [7:03:02<3:09:59,  1.52s/it]
 69%|██████▉   | 16903/24420 [7:03:03<3:11:32,  1.53s/it]
 69%|██████▉   | 16904/24420 [7:03:05<3:10:33,  1.52s/it]
 69%|██████▉   | 16905/24420 [7:03:06<3:10:02,  1.52s/it]
                                                         
{'loss': 0.0602, 'learning_rate': 5e-05, 'epoch': 3.46}

 69%|██████▉   | 16905/24420 [7:03:06<3:10:02,  1.52s/it]
 69%|██████▉   | 16906/24420 [7:03:08<3:09:12,  1.51s/it]
 69%|██████▉   | 16907/24420 [7:03:09<3:08:58,  1.51s/it]
 69%|██████▉   | 16908/24420 [7:03:11<3:08:30,  1.51s/it]
 69%|██████▉   | 16909/24420 [7:03:12<3:08:23,  1.50s/it]
 69%|██████▉   | 16910/24420 [7:03:14<3:08:35,  1.51s/it]
                                                         
{'loss': 0.0864, 'learning_rate': 5e-05, 'epoch': 3.46}

 69%|██████▉   | 16910/24420 [7:03:14<3:08:35,  1.51s/it]
 69%|██████▉   | 16911/24420 [7:03:15<3:08:36,  1.51s/it]
 69%|██████▉   | 16912/24420 [7:03:17<3:08:17,  1.50s/it]
 69%|██████▉   | 16913/24420 [7:03:18<3:08:18,  1.51s/it]
 69%|██████▉   | 16914/24420 [7:03:20<3:08:15,  1.50s/it]
 69%|██████▉   | 16915/24420 [7:03:21<3:08:08,  1.50s/it]
                                                         
{'loss': 0.0715, 'learning_rate': 5e-05, 'epoch': 3.46}

 69%|██████▉   | 16915/24420 [7:03:21<3:08:08,  1.50s/it]
 69%|██████▉   | 16916/24420 [7:03:23<3:08:24,  1.51s/it]
 69%|██████▉   | 16917/24420 [7:03:24<3:08:23,  1.51s/it]
 69%|██████▉   | 16918/24420 [7:03:26<3:08:23,  1.51s/it]
 69%|██████▉   | 16919/24420 [7:03:27<3:08:30,  1.51s/it]
 69%|██████▉   | 16920/24420 [7:03:29<3:08:11,  1.51s/it]
                                                         
{'loss': 0.0774, 'learning_rate': 5e-05, 'epoch': 3.46}

 69%|██████▉   | 16920/24420 [7:03:29<3:08:11,  1.51s/it]
 69%|██████▉   | 16921/24420 [7:03:30<3:08:11,  1.51s/it]
 69%|██████▉   | 16922/24420 [7:03:32<3:08:16,  1.51s/it]
 69%|██████▉   | 16923/24420 [7:03:33<3:08:09,  1.51s/it]
 69%|██████▉   | 16924/24420 [7:03:35<3:07:54,  1.50s/it]
 69%|██████▉   | 16925/24420 [7:03:36<3:08:23,  1.51s/it]
                                                         
{'loss': 0.077, 'learning_rate': 5e-05, 'epoch': 3.47}

 69%|██████▉   | 16925/24420 [7:03:36<3:08:23,  1.51s/it]
 69%|██████▉   | 16926/24420 [7:03:38<3:08:05,  1.51s/it]
 69%|██████▉   | 16927/24420 [7:03:39<3:07:50,  1.50s/it]
 69%|██████▉   | 16928/24420 [7:03:41<3:07:48,  1.50s/it]
 69%|██████▉   | 16929/24420 [7:03:42<3:07:45,  1.50s/it]
 69%|██████▉   | 16930/24420 [7:03:44<3:08:10,  1.51s/it]
                                                         
{'loss': 0.0722, 'learning_rate': 5e-05, 'epoch': 3.47}

 69%|██████▉   | 16930/24420 [7:03:44<3:08:10,  1.51s/it]
 69%|██████▉   | 16931/24420 [7:03:45<3:08:51,  1.51s/it]
 69%|██████▉   | 16932/24420 [7:03:47<3:08:20,  1.51s/it]
 69%|██████▉   | 16933/24420 [7:03:48<3:08:12,  1.51s/it]
 69%|██████▉   | 16934/24420 [7:03:50<3:08:51,  1.51s/it]
 69%|██████▉   | 16935/24420 [7:03:51<3:08:36,  1.51s/it]
                                                         
{'loss': 0.0821, 'learning_rate': 5e-05, 'epoch': 3.47}

 69%|██████▉   | 16935/24420 [7:03:51<3:08:36,  1.51s/it]
 69%|██████▉   | 16936/24420 [7:03:53<3:08:41,  1.51s/it]
 69%|██████▉   | 16937/24420 [7:03:54<3:08:34,  1.51s/it]
 69%|██████▉   | 16938/24420 [7:03:56<3:08:03,  1.51s/it]
 69%|██████▉   | 16939/24420 [7:03:57<3:08:00,  1.51s/it]
 69%|██████▉   | 16940/24420 [7:03:59<3:08:06,  1.51s/it]
                                                         
{'loss': 0.0613, 'learning_rate': 5e-05, 'epoch': 3.47}

 69%|██████▉   | 16940/24420 [7:03:59<3:08:06,  1.51s/it]
 69%|██████▉   | 16941/24420 [7:04:00<3:08:09,  1.51s/it]
 69%|██████▉   | 16942/24420 [7:04:02<3:07:55,  1.51s/it]
 69%|██████▉   | 16943/24420 [7:04:03<3:07:59,  1.51s/it]
 69%|██████▉   | 16944/24420 [7:04:05<3:07:36,  1.51s/it]
 69%|██████▉   | 16945/24420 [7:04:06<3:07:30,  1.51s/it]
                                                         
{'loss': 0.1097, 'learning_rate': 5e-05, 'epoch': 3.47}

 69%|██████▉   | 16945/24420 [7:04:06<3:07:30,  1.51s/it]
 69%|██████▉   | 16946/24420 [7:04:08<3:08:25,  1.51s/it]
 69%|██████▉   | 16947/24420 [7:04:09<3:08:37,  1.51s/it]
 69%|██████▉   | 16948/24420 [7:04:11<3:08:36,  1.51s/it]
 69%|██████▉   | 16949/24420 [7:04:12<3:08:43,  1.52s/it]
 69%|██████▉   | 16950/24420 [7:04:14<3:08:28,  1.51s/it]
                                                         
{'loss': 0.079, 'learning_rate': 5e-05, 'epoch': 3.47}

 69%|██████▉   | 16950/24420 [7:04:14<3:08:28,  1.51s/it]
 69%|██████▉   | 16951/24420 [7:04:15<3:08:41,  1.52s/it]
 69%|██████▉   | 16952/24420 [7:04:17<3:09:06,  1.52s/it]
 69%|██████▉   | 16953/24420 [7:04:19<3:09:20,  1.52s/it]
 69%|██████▉   | 16954/24420 [7:04:20<3:08:53,  1.52s/it]
 69%|██████▉   | 16955/24420 [7:04:22<3:08:35,  1.52s/it]
                                                         
{'loss': 0.0803, 'learning_rate': 5e-05, 'epoch': 3.47}

 69%|██████▉   | 16955/24420 [7:04:22<3:08:35,  1.52s/it]
 69%|██████▉   | 16956/24420 [7:04:23<3:08:19,  1.51s/it]
 69%|██████▉   | 16957/24420 [7:04:25<3:08:31,  1.52s/it]
 69%|██████▉   | 16958/24420 [7:04:26<3:08:13,  1.51s/it]
 69%|██████▉   | 16959/24420 [7:04:28<3:08:10,  1.51s/it]
 69%|██████▉   | 16960/24420 [7:04:29<3:08:11,  1.51s/it]
                                                         
{'loss': 0.0731, 'learning_rate': 5e-05, 'epoch': 3.47}

 69%|██████▉   | 16960/24420 [7:04:29<3:08:11,  1.51s/it]
 69%|██████▉   | 16961/24420 [7:04:31<3:08:01,  1.51s/it]
 69%|██████▉   | 16962/24420 [7:04:32<3:07:38,  1.51s/it]
 69%|██████▉   | 16963/24420 [7:04:34<3:07:39,  1.51s/it]
 69%|██████▉   | 16964/24420 [7:04:35<3:07:32,  1.51s/it]
 69%|██████▉   | 16965/24420 [7:04:37<3:07:20,  1.51s/it]
                                                         
{'loss': 0.075, 'learning_rate': 5e-05, 'epoch': 3.47}

 69%|██████▉   | 16965/24420 [7:04:37<3:07:20,  1.51s/it]
 69%|██████▉   | 16966/24420 [7:04:38<3:07:07,  1.51s/it]
 69%|██████▉   | 16967/24420 [7:04:40<3:06:56,  1.50s/it]
 69%|██████▉   | 16968/24420 [7:04:41<3:07:03,  1.51s/it]
 69%|██████▉   | 16969/24420 [7:04:43<3:07:02,  1.51s/it]
 69%|██████▉   | 16970/24420 [7:04:44<3:06:58,  1.51s/it]
                                                         
{'loss': 0.0758, 'learning_rate': 5e-05, 'epoch': 3.47}

 69%|██████▉   | 16970/24420 [7:04:44<3:06:58,  1.51s/it]
 69%|██████▉   | 16971/24420 [7:04:46<3:06:50,  1.50s/it]
 70%|██████▉   | 16972/24420 [7:04:47<3:06:49,  1.51s/it]
 70%|██████▉   | 16973/24420 [7:04:49<3:06:52,  1.51s/it]
 70%|██████▉   | 16974/24420 [7:04:50<3:06:42,  1.50s/it]
 70%|██████▉   | 16975/24420 [7:04:52<3:06:43,  1.50s/it]
                                                         
{'loss': 0.0707, 'learning_rate': 5e-05, 'epoch': 3.48}

 70%|██████▉   | 16975/24420 [7:04:52<3:06:43,  1.50s/it]
 70%|██████▉   | 16976/24420 [7:04:53<3:07:14,  1.51s/it]
 70%|██████▉   | 16977/24420 [7:04:55<3:07:05,  1.51s/it]
 70%|██████▉   | 16978/24420 [7:04:56<3:06:51,  1.51s/it]
 70%|██████▉   | 16979/24420 [7:04:58<3:07:01,  1.51s/it]
 70%|██████▉   | 16980/24420 [7:04:59<3:06:51,  1.51s/it]
                                                         
{'loss': 0.0799, 'learning_rate': 5e-05, 'epoch': 3.48}

 70%|██████▉   | 16980/24420 [7:04:59<3:06:51,  1.51s/it]
 70%|██████▉   | 16981/24420 [7:05:01<3:06:51,  1.51s/it]
 70%|██████▉   | 16982/24420 [7:05:02<3:07:06,  1.51s/it]
 70%|██████▉   | 16983/24420 [7:05:04<3:06:42,  1.51s/it]
 70%|██████▉   | 16984/24420 [7:05:05<3:06:37,  1.51s/it]
 70%|██████▉   | 16985/24420 [7:05:07<3:06:46,  1.51s/it]
                                                         
{'loss': 0.0939, 'learning_rate': 5e-05, 'epoch': 3.48}

 70%|██████▉   | 16985/24420 [7:05:07<3:06:46,  1.51s/it]
 70%|██████▉   | 16986/24420 [7:05:08<3:07:15,  1.51s/it]
 70%|██████▉   | 16987/24420 [7:05:10<3:07:11,  1.51s/it]
 70%|██████▉   | 16988/24420 [7:05:11<3:06:42,  1.51s/it]
 70%|██████▉   | 16989/24420 [7:05:13<3:06:57,  1.51s/it]
 70%|██████▉   | 16990/24420 [7:05:14<3:07:16,  1.51s/it]
                                                         
{'loss': 0.0624, 'learning_rate': 5e-05, 'epoch': 3.48}

 70%|██████▉   | 16990/24420 [7:05:14<3:07:16,  1.51s/it]
 70%|██████▉   | 16991/24420 [7:05:16<3:07:28,  1.51s/it]
 70%|██████▉   | 16992/24420 [7:05:17<3:06:56,  1.51s/it]
 70%|██████▉   | 16993/24420 [7:05:19<3:06:36,  1.51s/it]
 70%|██████▉   | 16994/24420 [7:05:20<3:06:18,  1.51s/it]
 70%|██████▉   | 16995/24420 [7:05:22<3:05:57,  1.50s/it]
                                                         
{'loss': 0.0745, 'learning_rate': 5e-05, 'epoch': 3.48}

 70%|██████▉   | 16995/24420 [7:05:22<3:05:57,  1.50s/it]
 70%|██████▉   | 16996/24420 [7:05:23<3:06:47,  1.51s/it]
 70%|██████▉   | 16997/24420 [7:05:25<3:06:58,  1.51s/it]
 70%|██████▉   | 16998/24420 [7:05:26<3:07:50,  1.52s/it]
 70%|██████▉   | 16999/24420 [7:05:28<3:07:19,  1.51s/it]
 70%|██████▉   | 17000/24420 [7:05:29<3:06:53,  1.51s/it]
                                                         
{'loss': 0.0891, 'learning_rate': 5e-05, 'epoch': 3.48}

 70%|██████▉   | 17000/24420 [7:05:29<3:06:53,  1.51s/it]
 70%|██████▉   | 17001/24420 [7:05:36<6:10:06,  2.99s/it]
 70%|██████▉   | 17002/24420 [7:05:37<5:14:37,  2.54s/it]
 70%|██████▉   | 17003/24420 [7:05:39<4:35:51,  2.23s/it]
 70%|██████▉   | 17004/24420 [7:05:40<4:09:10,  2.02s/it]
 70%|██████▉   | 17005/24420 [7:05:42<3:50:43,  1.87s/it]
                                                         
{'loss': 0.0735, 'learning_rate': 5e-05, 'epoch': 3.48}

 70%|██████▉   | 17005/24420 [7:05:42<3:50:43,  1.87s/it]
 70%|██████▉   | 17006/24420 [7:05:43<3:37:09,  1.76s/it]
 70%|██████▉   | 17007/24420 [7:05:45<3:27:23,  1.68s/it]
 70%|██████▉   | 17008/24420 [7:05:46<3:21:02,  1.63s/it]
 70%|██████▉   | 17009/24420 [7:05:48<3:16:07,  1.59s/it]
 70%|██████▉   | 17010/24420 [7:05:49<3:13:19,  1.57s/it]
                                                         
{'loss': 0.0729, 'learning_rate': 5e-05, 'epoch': 3.48}

 70%|██████▉   | 17010/24420 [7:05:49<3:13:19,  1.57s/it]
 70%|██████▉   | 17011/24420 [7:05:51<3:11:20,  1.55s/it]
 70%|██████▉   | 17012/24420 [7:05:52<3:09:32,  1.54s/it]
 70%|██████▉   | 17013/24420 [7:05:54<3:08:24,  1.53s/it]
 70%|██████▉   | 17014/24420 [7:05:55<3:07:16,  1.52s/it]
 70%|██████▉   | 17015/24420 [7:05:57<3:06:48,  1.51s/it]
                                                         
{'loss': 0.0905, 'learning_rate': 5e-05, 'epoch': 3.48}

 70%|██████▉   | 17015/24420 [7:05:57<3:06:48,  1.51s/it]
 70%|██████▉   | 17016/24420 [7:05:58<3:06:24,  1.51s/it]
 70%|██████▉   | 17017/24420 [7:06:00<3:06:12,  1.51s/it]
 70%|██████▉   | 17018/24420 [7:06:01<3:06:13,  1.51s/it]
 70%|██████▉   | 17019/24420 [7:06:03<3:05:57,  1.51s/it]
 70%|██████▉   | 17020/24420 [7:06:05<3:06:05,  1.51s/it]
                                                         
{'loss': 0.0879, 'learning_rate': 5e-05, 'epoch': 3.48}

 70%|██████▉   | 17020/24420 [7:06:05<3:06:05,  1.51s/it]
 70%|██████▉   | 17021/24420 [7:06:06<3:05:55,  1.51s/it]
 70%|██████▉   | 17022/24420 [7:06:08<3:05:53,  1.51s/it]
 70%|██████▉   | 17023/24420 [7:06:09<3:05:46,  1.51s/it]
 70%|██████▉   | 17024/24420 [7:06:11<3:05:58,  1.51s/it]
 70%|██████▉   | 17025/24420 [7:06:12<3:06:09,  1.51s/it]
                                                         
{'loss': 0.076, 'learning_rate': 5e-05, 'epoch': 3.49}

 70%|██████▉   | 17025/24420 [7:06:12<3:06:09,  1.51s/it]
 70%|██████▉   | 17026/24420 [7:06:14<3:06:25,  1.51s/it]
 70%|██████▉   | 17027/24420 [7:06:15<3:06:10,  1.51s/it]
 70%|██████▉   | 17028/24420 [7:06:17<3:05:55,  1.51s/it]
 70%|██████▉   | 17029/24420 [7:06:18<3:05:27,  1.51s/it]
 70%|██████▉   | 17030/24420 [7:06:20<3:05:29,  1.51s/it]
                                                         
{'loss': 0.0836, 'learning_rate': 5e-05, 'epoch': 3.49}

 70%|██████▉   | 17030/24420 [7:06:20<3:05:29,  1.51s/it]
 70%|██████▉   | 17031/24420 [7:06:21<3:05:27,  1.51s/it]
 70%|██████▉   | 17032/24420 [7:06:23<3:05:41,  1.51s/it]
 70%|██████▉   | 17033/24420 [7:06:24<3:05:33,  1.51s/it]
 70%|██████▉   | 17034/24420 [7:06:26<3:05:58,  1.51s/it]
 70%|██████▉   | 17035/24420 [7:06:27<3:05:41,  1.51s/it]
                                                         
{'loss': 0.0911, 'learning_rate': 5e-05, 'epoch': 3.49}

 70%|██████▉   | 17035/24420 [7:06:27<3:05:41,  1.51s/it]
 70%|██████▉   | 17036/24420 [7:06:29<3:05:31,  1.51s/it]
 70%|██████▉   | 17037/24420 [7:06:30<3:05:31,  1.51s/it]
 70%|██████▉   | 17038/24420 [7:06:32<3:05:18,  1.51s/it]
 70%|██████▉   | 17039/24420 [7:06:33<3:05:44,  1.51s/it]
 70%|██████▉   | 17040/24420 [7:06:35<3:05:51,  1.51s/it]
                                                         
{'loss': 0.0812, 'learning_rate': 5e-05, 'epoch': 3.49}

 70%|██████▉   | 17040/24420 [7:06:35<3:05:51,  1.51s/it]
 70%|██████▉   | 17041/24420 [7:06:36<3:05:38,  1.51s/it]
 70%|██████▉   | 17042/24420 [7:06:38<3:05:45,  1.51s/it]
 70%|██████▉   | 17043/24420 [7:06:39<3:05:29,  1.51s/it]
 70%|██████▉   | 17044/24420 [7:06:41<3:05:39,  1.51s/it]
 70%|██████▉   | 17045/24420 [7:06:42<3:05:45,  1.51s/it]
                                                         
{'loss': 0.08, 'learning_rate': 5e-05, 'epoch': 3.49}

 70%|██████▉   | 17045/24420 [7:06:42<3:05:45,  1.51s/it]
 70%|██████▉   | 17046/24420 [7:06:44<3:05:42,  1.51s/it]
 70%|██████▉   | 17047/24420 [7:06:45<3:05:24,  1.51s/it]
 70%|██████▉   | 17048/24420 [7:06:47<3:05:19,  1.51s/it]
 70%|██████▉   | 17049/24420 [7:06:48<3:04:50,  1.50s/it]
 70%|██████▉   | 17050/24420 [7:06:50<3:05:11,  1.51s/it]
                                                         
{'loss': 0.0806, 'learning_rate': 5e-05, 'epoch': 3.49}

 70%|██████▉   | 17050/24420 [7:06:50<3:05:11,  1.51s/it]
 70%|██████▉   | 17051/24420 [7:06:51<3:05:25,  1.51s/it]
 70%|██████▉   | 17052/24420 [7:06:53<3:04:56,  1.51s/it]
 70%|██████▉   | 17053/24420 [7:06:54<3:04:50,  1.51s/it]
 70%|██████▉   | 17054/24420 [7:06:56<3:04:51,  1.51s/it]
 70%|██████▉   | 17055/24420 [7:06:57<3:05:38,  1.51s/it]
                                                         
{'loss': 0.0815, 'learning_rate': 5e-05, 'epoch': 3.49}

 70%|██████▉   | 17055/24420 [7:06:57<3:05:38,  1.51s/it]
 70%|██████▉   | 17056/24420 [7:06:59<3:05:42,  1.51s/it]
 70%|██████▉   | 17057/24420 [7:07:00<3:05:33,  1.51s/it]
 70%|██████▉   | 17058/24420 [7:07:02<3:04:52,  1.51s/it]
 70%|██████▉   | 17059/24420 [7:07:03<3:04:52,  1.51s/it]
 70%|██████▉   | 17060/24420 [7:07:05<3:04:59,  1.51s/it]
                                                         
{'loss': 0.0695, 'learning_rate': 5e-05, 'epoch': 3.49}

 70%|██████▉   | 17060/24420 [7:07:05<3:04:59,  1.51s/it]
 70%|██████▉   | 17061/24420 [7:07:06<3:04:57,  1.51s/it]
 70%|██████▉   | 17062/24420 [7:07:08<3:04:56,  1.51s/it]
 70%|██████▉   | 17063/24420 [7:07:09<3:05:10,  1.51s/it]
 70%|██████▉   | 17064/24420 [7:07:11<3:05:19,  1.51s/it]
 70%|██████▉   | 17065/24420 [7:07:12<3:05:23,  1.51s/it]
                                                         
{'loss': 0.0928, 'learning_rate': 5e-05, 'epoch': 3.49}

 70%|██████▉   | 17065/24420 [7:07:12<3:05:23,  1.51s/it]
 70%|██████▉   | 17066/24420 [7:07:14<3:05:42,  1.52s/it]
 70%|██████▉   | 17067/24420 [7:07:15<3:05:54,  1.52s/it]
 70%|██████▉   | 17068/24420 [7:07:17<3:05:50,  1.52s/it]
 70%|██████▉   | 17069/24420 [7:07:18<3:05:03,  1.51s/it]
 70%|██████▉   | 17070/24420 [7:07:20<3:04:42,  1.51s/it]
                                                         
{'loss': 0.0789, 'learning_rate': 5e-05, 'epoch': 3.5}

 70%|██████▉   | 17070/24420 [7:07:20<3:04:42,  1.51s/it]
 70%|██████▉   | 17071/24420 [7:07:21<3:04:19,  1.50s/it]
 70%|██████▉   | 17072/24420 [7:07:23<3:05:56,  1.52s/it]
 70%|██████▉   | 17073/24420 [7:07:25<3:05:38,  1.52s/it]
 70%|██████▉   | 17074/24420 [7:07:26<3:05:39,  1.52s/it]
 70%|██████▉   | 17075/24420 [7:07:28<3:05:29,  1.52s/it]
                                                         
{'loss': 0.0708, 'learning_rate': 5e-05, 'epoch': 3.5}

 70%|██████▉   | 17075/24420 [7:07:28<3:05:29,  1.52s/it]
 70%|██████▉   | 17076/24420 [7:07:29<3:05:12,  1.51s/it]
 70%|██████▉   | 17077/24420 [7:07:31<3:04:14,  1.51s/it]
 70%|██████▉   | 17078/24420 [7:07:32<3:04:39,  1.51s/it]
 70%|██████▉   | 17079/24420 [7:07:34<3:04:11,  1.51s/it]
 70%|██████▉   | 17080/24420 [7:07:35<3:04:39,  1.51s/it]
                                                         
{'loss': 0.0677, 'learning_rate': 5e-05, 'epoch': 3.5}

 70%|██████▉   | 17080/24420 [7:07:35<3:04:39,  1.51s/it]
 70%|██████▉   | 17081/24420 [7:07:37<3:04:49,  1.51s/it]
 70%|██████▉   | 17082/24420 [7:07:38<3:04:49,  1.51s/it]
 70%|██████▉   | 17083/24420 [7:07:40<3:04:49,  1.51s/it]
 70%|██████▉   | 17084/24420 [7:07:41<3:04:23,  1.51s/it]
 70%|██████▉   | 17085/24420 [7:07:43<3:03:50,  1.50s/it]
                                                         
{'loss': 0.0681, 'learning_rate': 5e-05, 'epoch': 3.5}

 70%|██████▉   | 17085/24420 [7:07:43<3:03:50,  1.50s/it]
 70%|██████▉   | 17086/24420 [7:07:44<3:03:36,  1.50s/it]
 70%|██████▉   | 17087/24420 [7:07:46<3:03:28,  1.50s/it]
 70%|██████▉   | 17088/24420 [7:07:47<3:03:58,  1.51s/it]
 70%|██████▉   | 17089/24420 [7:07:49<3:03:44,  1.50s/it]
 70%|██████▉   | 17090/24420 [7:07:50<3:03:43,  1.50s/it]
                                                         
{'loss': 0.0872, 'learning_rate': 5e-05, 'epoch': 3.5}

 70%|██████▉   | 17090/24420 [7:07:50<3:03:43,  1.50s/it]
 70%|██████▉   | 17091/24420 [7:07:52<3:03:52,  1.51s/it]
 70%|██████▉   | 17092/24420 [7:07:53<3:03:46,  1.50s/it]
 70%|██████▉   | 17093/24420 [7:07:55<3:03:11,  1.50s/it]
 70%|███████   | 17094/24420 [7:07:56<3:02:37,  1.50s/it]
 70%|███████   | 17095/24420 [7:07:58<3:02:03,  1.49s/it]
                                                         
{'loss': 0.0588, 'learning_rate': 5e-05, 'epoch': 3.5}

 70%|███████   | 17095/24420 [7:07:58<3:02:03,  1.49s/it]
 70%|███████   | 17096/24420 [7:07:59<3:02:03,  1.49s/it]
 70%|███████   | 17097/24420 [7:08:01<3:02:04,  1.49s/it]
 70%|███████   | 17098/24420 [7:08:02<3:01:45,  1.49s/it]
 70%|███████   | 17099/24420 [7:08:04<3:02:04,  1.49s/it]
 70%|███████   | 17100/24420 [7:08:05<3:02:11,  1.49s/it]
                                                         
{'loss': 0.0808, 'learning_rate': 5e-05, 'epoch': 3.5}

 70%|███████   | 17100/24420 [7:08:05<3:02:11,  1.49s/it]
 70%|███████   | 17101/24420 [7:08:07<3:04:03,  1.51s/it]
 70%|███████   | 17102/24420 [7:08:08<3:03:07,  1.50s/it]
 70%|███████   | 17103/24420 [7:08:10<3:02:14,  1.49s/it]
 70%|███████   | 17104/24420 [7:08:11<3:01:55,  1.49s/it]
 70%|███████   | 17105/24420 [7:08:13<3:01:34,  1.49s/it]
                                                         
{'loss': 0.0808, 'learning_rate': 5e-05, 'epoch': 3.5}

 70%|███████   | 17105/24420 [7:08:13<3:01:34,  1.49s/it]
 70%|███████   | 17106/24420 [7:08:14<3:01:31,  1.49s/it]
 70%|███████   | 17107/24420 [7:08:16<3:01:37,  1.49s/it]
 70%|███████   | 17108/24420 [7:08:17<3:01:47,  1.49s/it]
 70%|███████   | 17109/24420 [7:08:19<3:01:43,  1.49s/it]
 70%|███████   | 17110/24420 [7:08:20<3:01:56,  1.49s/it]
                                                         
{'loss': 0.0849, 'learning_rate': 5e-05, 'epoch': 3.5}

 70%|███████   | 17110/24420 [7:08:20<3:01:56,  1.49s/it]
 70%|███████   | 17111/24420 [7:08:22<3:02:04,  1.49s/it]
 70%|███████   | 17112/24420 [7:08:23<3:01:38,  1.49s/it]
 70%|███████   | 17113/24420 [7:08:24<3:01:43,  1.49s/it]
 70%|███████   | 17114/24420 [7:08:26<3:01:43,  1.49s/it]
 70%|███████   | 17115/24420 [7:08:27<3:01:30,  1.49s/it]
                                                         
{'loss': 0.0657, 'learning_rate': 5e-05, 'epoch': 3.5}

 70%|███████   | 17115/24420 [7:08:27<3:01:30,  1.49s/it]
 70%|███████   | 17116/24420 [7:08:29<3:01:25,  1.49s/it]
 70%|███████   | 17117/24420 [7:08:30<3:01:13,  1.49s/it]
 70%|███████   | 17118/24420 [7:08:32<3:00:57,  1.49s/it]
 70%|███████   | 17119/24420 [7:08:33<3:00:48,  1.49s/it]
 70%|███████   | 17120/24420 [7:08:35<3:00:48,  1.49s/it]
                                                         
{'loss': 0.0693, 'learning_rate': 5e-05, 'epoch': 3.51}

 70%|███████   | 17120/24420 [7:08:35<3:00:48,  1.49s/it]
 70%|███████   | 17121/24420 [7:08:36<3:01:09,  1.49s/it]
 70%|███████   | 17122/24420 [7:08:38<3:00:49,  1.49s/it]
 70%|███████   | 17123/24420 [7:08:39<3:01:06,  1.49s/it]
 70%|███████   | 17124/24420 [7:08:41<3:01:41,  1.49s/it]
 70%|███████   | 17125/24420 [7:08:42<3:01:26,  1.49s/it]
                                                         
{'loss': 0.0807, 'learning_rate': 5e-05, 'epoch': 3.51}

 70%|███████   | 17125/24420 [7:08:42<3:01:26,  1.49s/it]
 70%|███████   | 17126/24420 [7:08:44<3:01:30,  1.49s/it]
 70%|███████   | 17127/24420 [7:08:45<3:01:23,  1.49s/it]
 70%|███████   | 17128/24420 [7:08:47<3:00:55,  1.49s/it]
 70%|███████   | 17129/24420 [7:08:48<3:00:48,  1.49s/it]
 70%|███████   | 17130/24420 [7:08:50<3:00:36,  1.49s/it]
                                                         
{'loss': 0.0669, 'learning_rate': 5e-05, 'epoch': 3.51}

 70%|███████   | 17130/24420 [7:08:50<3:00:36,  1.49s/it]
 70%|███████   | 17131/24420 [7:08:51<3:00:39,  1.49s/it]
 70%|███████   | 17132/24420 [7:08:53<3:00:27,  1.49s/it]
 70%|███████   | 17133/24420 [7:08:54<3:00:32,  1.49s/it]
 70%|███████   | 17134/24420 [7:08:56<3:00:26,  1.49s/it]
 70%|███████   | 17135/24420 [7:08:57<3:00:15,  1.48s/it]
                                                         
{'loss': 0.0715, 'learning_rate': 5e-05, 'epoch': 3.51}

 70%|███████   | 17135/24420 [7:08:57<3:00:15,  1.48s/it]
 70%|███████   | 17136/24420 [7:08:59<3:00:48,  1.49s/it]
 70%|███████   | 17137/24420 [7:09:00<3:00:38,  1.49s/it]
 70%|███████   | 17138/24420 [7:09:02<3:00:23,  1.49s/it]
 70%|███████   | 17139/24420 [7:09:03<3:00:09,  1.48s/it]
 70%|███████   | 17140/24420 [7:09:05<3:00:17,  1.49s/it]
                                                         
{'loss': 0.0701, 'learning_rate': 5e-05, 'epoch': 3.51}

 70%|███████   | 17140/24420 [7:09:05<3:00:17,  1.49s/it]
 70%|███████   | 17141/24420 [7:09:06<3:00:38,  1.49s/it]
 70%|███████   | 17142/24420 [7:09:08<3:00:44,  1.49s/it]
 70%|███████   | 17143/24420 [7:09:09<3:00:52,  1.49s/it]
 70%|███████   | 17144/24420 [7:09:11<3:00:36,  1.49s/it]
 70%|███████   | 17145/24420 [7:09:12<3:00:57,  1.49s/it]
                                                         
{'loss': 0.0844, 'learning_rate': 5e-05, 'epoch': 3.51}

 70%|███████   | 17145/24420 [7:09:12<3:00:57,  1.49s/it]
 70%|███████   | 17146/24420 [7:09:14<3:00:44,  1.49s/it]
 70%|███████   | 17147/24420 [7:09:15<3:00:21,  1.49s/it]
 70%|███████   | 17148/24420 [7:09:17<3:00:06,  1.49s/it]
 70%|███████   | 17149/24420 [7:09:18<2:59:58,  1.49s/it]
 70%|███████   | 17150/24420 [7:09:20<3:00:01,  1.49s/it]
                                                         
{'loss': 0.0848, 'learning_rate': 5e-05, 'epoch': 3.51}

 70%|███████   | 17150/24420 [7:09:20<3:00:01,  1.49s/it]
 70%|███████   | 17151/24420 [7:09:21<3:00:00,  1.49s/it]
 70%|███████   | 17152/24420 [7:09:23<2:59:43,  1.48s/it]
 70%|███████   | 17153/24420 [7:09:24<2:59:44,  1.48s/it]
 70%|███████   | 17154/24420 [7:09:25<2:59:55,  1.49s/it]
 70%|███████   | 17155/24420 [7:09:27<3:00:32,  1.49s/it]
                                                         
{'loss': 0.0703, 'learning_rate': 5e-05, 'epoch': 3.51}

 70%|███████   | 17155/24420 [7:09:27<3:00:32,  1.49s/it]
 70%|███████   | 17156/24420 [7:09:28<3:00:13,  1.49s/it]
 70%|███████   | 17157/24420 [7:09:30<3:00:18,  1.49s/it]
 70%|███████   | 17158/24420 [7:09:31<3:00:07,  1.49s/it]
 70%|███████   | 17159/24420 [7:09:33<3:00:31,  1.49s/it]
 70%|███████   | 17160/24420 [7:09:34<3:00:07,  1.49s/it]
                                                         
{'loss': 0.0725, 'learning_rate': 5e-05, 'epoch': 3.51}

 70%|███████   | 17160/24420 [7:09:34<3:00:07,  1.49s/it]
 70%|███████   | 17161/24420 [7:09:36<3:00:12,  1.49s/it]
 70%|███████   | 17162/24420 [7:09:37<3:00:01,  1.49s/it]
 70%|███████   | 17163/24420 [7:09:39<2:59:49,  1.49s/it]
 70%|███████   | 17164/24420 [7:09:40<3:00:08,  1.49s/it]
 70%|███████   | 17165/24420 [7:09:42<2:59:57,  1.49s/it]
                                                         
{'loss': 0.0805, 'learning_rate': 5e-05, 'epoch': 3.51}

 70%|███████   | 17165/24420 [7:09:42<2:59:57,  1.49s/it]
 70%|███████   | 17166/24420 [7:09:43<2:59:55,  1.49s/it]
 70%|███████   | 17167/24420 [7:09:45<3:00:01,  1.49s/it]
 70%|███████   | 17168/24420 [7:09:46<3:00:22,  1.49s/it]
 70%|███████   | 17169/24420 [7:09:48<2:59:59,  1.49s/it]
 70%|███████   | 17170/24420 [7:09:49<2:59:54,  1.49s/it]
                                                         
{'loss': 0.0713, 'learning_rate': 5e-05, 'epoch': 3.52}

 70%|███████   | 17170/24420 [7:09:49<2:59:54,  1.49s/it]
 70%|███████   | 17171/24420 [7:09:51<2:59:48,  1.49s/it]
 70%|███████   | 17172/24420 [7:09:52<2:59:58,  1.49s/it]
 70%|███████   | 17173/24420 [7:09:54<2:59:52,  1.49s/it]
 70%|███████   | 17174/24420 [7:09:55<2:59:26,  1.49s/it]
 70%|███████   | 17175/24420 [7:09:57<2:59:57,  1.49s/it]
                                                         
{'loss': 0.074, 'learning_rate': 5e-05, 'epoch': 3.52}

 70%|███████   | 17175/24420 [7:09:57<2:59:57,  1.49s/it]
 70%|███████   | 17176/24420 [7:09:58<3:00:32,  1.50s/it]
 70%|███████   | 17177/24420 [7:10:00<3:00:21,  1.49s/it]
 70%|███████   | 17178/24420 [7:10:01<3:00:02,  1.49s/it]
 70%|███████   | 17179/24420 [7:10:03<3:00:02,  1.49s/it]
 70%|███████   | 17180/24420 [7:10:04<2:59:53,  1.49s/it]
                                                         
{'loss': 0.0917, 'learning_rate': 5e-05, 'epoch': 3.52}

 70%|███████   | 17180/24420 [7:10:04<2:59:53,  1.49s/it]
 70%|███████   | 17181/24420 [7:10:06<2:59:50,  1.49s/it]
 70%|███████   | 17182/24420 [7:10:07<2:59:31,  1.49s/it]
 70%|███████   | 17183/24420 [7:10:09<2:59:35,  1.49s/it]
 70%|███████   | 17184/24420 [7:10:10<2:59:39,  1.49s/it]
 70%|███████   | 17185/24420 [7:10:12<2:59:38,  1.49s/it]
                                                         
{'loss': 0.0671, 'learning_rate': 5e-05, 'epoch': 3.52}

 70%|███████   | 17185/24420 [7:10:12<2:59:38,  1.49s/it]
 70%|███████   | 17186/24420 [7:10:13<2:59:39,  1.49s/it]
 70%|███████   | 17187/24420 [7:10:15<2:59:49,  1.49s/it]
 70%|███████   | 17188/24420 [7:10:16<2:59:44,  1.49s/it]
 70%|███████   | 17189/24420 [7:10:18<2:59:31,  1.49s/it]
 70%|███████   | 17190/24420 [7:10:19<2:59:33,  1.49s/it]
                                                         
{'loss': 0.0777, 'learning_rate': 5e-05, 'epoch': 3.52}

 70%|███████   | 17190/24420 [7:10:19<2:59:33,  1.49s/it]
 70%|███████   | 17191/24420 [7:10:21<2:59:23,  1.49s/it]
 70%|███████   | 17192/24420 [7:10:22<2:59:51,  1.49s/it]
 70%|███████   | 17193/24420 [7:10:24<2:59:40,  1.49s/it]
 70%|███████   | 17194/24420 [7:10:25<2:59:44,  1.49s/it]
 70%|███████   | 17195/24420 [7:10:27<2:59:17,  1.49s/it]
                                                         
{'loss': 0.1049, 'learning_rate': 5e-05, 'epoch': 3.52}

 70%|███████   | 17195/24420 [7:10:27<2:59:17,  1.49s/it]
 70%|███████   | 17196/24420 [7:10:28<2:59:21,  1.49s/it]
 70%|███████   | 17197/24420 [7:10:30<2:59:13,  1.49s/it]
 70%|███████   | 17198/24420 [7:10:31<2:59:19,  1.49s/it]
 70%|███████   | 17199/24420 [7:10:33<2:59:25,  1.49s/it]
 70%|███████   | 17200/24420 [7:10:34<2:59:31,  1.49s/it]
                                                         
{'loss': 0.0866, 'learning_rate': 5e-05, 'epoch': 3.52}

 70%|███████   | 17200/24420 [7:10:34<2:59:31,  1.49s/it]
 70%|███████   | 17201/24420 [7:10:39<4:50:02,  2.41s/it]
 70%|███████   | 17202/24420 [7:10:40<4:16:30,  2.13s/it]
 70%|███████   | 17203/24420 [7:10:42<3:53:23,  1.94s/it]
 70%|███████   | 17204/24420 [7:10:43<3:36:52,  1.80s/it]
 70%|███████   | 17205/24420 [7:10:45<3:25:33,  1.71s/it]
                                                         
{'loss': 0.0737, 'learning_rate': 5e-05, 'epoch': 3.52}

 70%|███████   | 17205/24420 [7:10:45<3:25:33,  1.71s/it]
 70%|███████   | 17206/24420 [7:10:46<3:18:06,  1.65s/it]
 70%|███████   | 17207/24420 [7:10:48<3:13:28,  1.61s/it]
 70%|███████   | 17208/24420 [7:10:49<3:09:02,  1.57s/it]
 70%|███████   | 17209/24420 [7:10:51<3:06:23,  1.55s/it]
 70%|███████   | 17210/24420 [7:10:52<3:04:23,  1.53s/it]
                                                         
{'loss': 0.0673, 'learning_rate': 5e-05, 'epoch': 3.52}

 70%|███████   | 17210/24420 [7:10:52<3:04:23,  1.53s/it]
 70%|███████   | 17211/24420 [7:10:54<3:02:49,  1.52s/it]
 70%|███████   | 17212/24420 [7:10:55<3:01:23,  1.51s/it]
 70%|███████   | 17213/24420 [7:10:57<3:00:48,  1.51s/it]
 70%|███████   | 17214/24420 [7:10:58<3:00:29,  1.50s/it]
 70%|███████   | 17215/24420 [7:11:00<3:00:23,  1.50s/it]
                                                         
{'loss': 0.0733, 'learning_rate': 5e-05, 'epoch': 3.52}

 70%|███████   | 17215/24420 [7:11:00<3:00:23,  1.50s/it]
 70%|███████   | 17216/24420 [7:11:01<2:59:56,  1.50s/it]
 71%|███████   | 17217/24420 [7:11:02<2:59:35,  1.50s/it]
 71%|███████   | 17218/24420 [7:11:04<2:59:23,  1.49s/it]
 71%|███████   | 17219/24420 [7:11:05<2:59:08,  1.49s/it]
 71%|███████   | 17220/24420 [7:11:07<2:58:49,  1.49s/it]
                                                         
{'loss': 0.074, 'learning_rate': 5e-05, 'epoch': 3.53}

 71%|███████   | 17220/24420 [7:11:07<2:58:49,  1.49s/it]
 71%|███████   | 17221/24420 [7:11:08<2:58:25,  1.49s/it]
 71%|███████   | 17222/24420 [7:11:10<2:57:58,  1.48s/it]
 71%|███████   | 17223/24420 [7:11:11<2:58:26,  1.49s/it]
 71%|███████   | 17224/24420 [7:11:13<2:58:23,  1.49s/it]
 71%|███████   | 17225/24420 [7:11:14<2:58:11,  1.49s/it]
                                                         
{'loss': 0.0673, 'learning_rate': 5e-05, 'epoch': 3.53}

 71%|███████   | 17225/24420 [7:11:14<2:58:11,  1.49s/it]
 71%|███████   | 17226/24420 [7:11:16<2:58:21,  1.49s/it]
 71%|███████   | 17227/24420 [7:11:17<2:58:26,  1.49s/it]
 71%|███████   | 17228/24420 [7:11:19<2:58:15,  1.49s/it]
 71%|███████   | 17229/24420 [7:11:20<2:58:07,  1.49s/it]
 71%|███████   | 17230/24420 [7:11:22<2:58:03,  1.49s/it]
                                                         
{'loss': 0.103, 'learning_rate': 5e-05, 'epoch': 3.53}

 71%|███████   | 17230/24420 [7:11:22<2:58:03,  1.49s/it]
 71%|███████   | 17231/24420 [7:11:23<2:58:18,  1.49s/it]
 71%|███████   | 17232/24420 [7:11:25<2:58:26,  1.49s/it]
 71%|███████   | 17233/24420 [7:11:26<2:58:18,  1.49s/it]
 71%|███████   | 17234/24420 [7:11:28<2:58:08,  1.49s/it]
 71%|███████   | 17235/24420 [7:11:29<2:57:59,  1.49s/it]
                                                         
{'loss': 0.07, 'learning_rate': 5e-05, 'epoch': 3.53}

 71%|███████   | 17235/24420 [7:11:29<2:57:59,  1.49s/it]
 71%|███████   | 17236/24420 [7:11:31<2:58:03,  1.49s/it]
 71%|███████   | 17237/24420 [7:11:32<2:57:54,  1.49s/it]
 71%|███████   | 17238/24420 [7:11:34<2:58:04,  1.49s/it]
 71%|███████   | 17239/24420 [7:11:35<2:58:15,  1.49s/it]
 71%|███████   | 17240/24420 [7:11:37<2:58:24,  1.49s/it]
                                                         
{'loss': 0.0756, 'learning_rate': 5e-05, 'epoch': 3.53}

 71%|███████   | 17240/24420 [7:11:37<2:58:24,  1.49s/it]
 71%|███████   | 17241/24420 [7:11:38<2:58:47,  1.49s/it]
 71%|███████   | 17242/24420 [7:11:40<2:58:27,  1.49s/it]
 71%|███████   | 17243/24420 [7:11:41<2:58:34,  1.49s/it]
 71%|███████   | 17244/24420 [7:11:43<2:58:13,  1.49s/it]
 71%|███████   | 17245/24420 [7:11:44<2:57:57,  1.49s/it]
                                                         
{'loss': 0.0804, 'learning_rate': 5e-05, 'epoch': 3.53}

 71%|███████   | 17245/24420 [7:11:44<2:57:57,  1.49s/it]
 71%|███████   | 17246/24420 [7:11:46<2:57:36,  1.49s/it]
 71%|███████   | 17247/24420 [7:11:47<2:57:31,  1.48s/it]
 71%|███████   | 17248/24420 [7:11:49<2:57:34,  1.49s/it]
 71%|███████   | 17249/24420 [7:11:50<2:57:47,  1.49s/it]
 71%|███████   | 17250/24420 [7:11:52<2:57:46,  1.49s/it]
                                                         
{'loss': 0.087, 'learning_rate': 5e-05, 'epoch': 3.53}

 71%|███████   | 17250/24420 [7:11:52<2:57:46,  1.49s/it]
 71%|███████   | 17251/24420 [7:11:53<2:57:42,  1.49s/it]
 71%|███████   | 17252/24420 [7:11:55<2:57:57,  1.49s/it]
 71%|███████   | 17253/24420 [7:11:56<2:57:22,  1.48s/it]
 71%|███████   | 17254/24420 [7:11:58<2:57:19,  1.48s/it]
 71%|███████   | 17255/24420 [7:11:59<2:57:53,  1.49s/it]
                                                         
{'loss': 0.0768, 'learning_rate': 5e-05, 'epoch': 3.53}

 71%|███████   | 17255/24420 [7:11:59<2:57:53,  1.49s/it]
 71%|███████   | 17256/24420 [7:12:01<2:58:08,  1.49s/it]
 71%|███████   | 17257/24420 [7:12:02<2:58:04,  1.49s/it]
 71%|███████   | 17258/24420 [7:12:04<2:57:42,  1.49s/it]
 71%|███████   | 17259/24420 [7:12:05<2:57:40,  1.49s/it]
 71%|███████   | 17260/24420 [7:12:06<2:57:22,  1.49s/it]
                                                         
{'loss': 0.0719, 'learning_rate': 5e-05, 'epoch': 3.53}

 71%|███████   | 17260/24420 [7:12:06<2:57:22,  1.49s/it]
 71%|███████   | 17261/24420 [7:12:08<2:57:43,  1.49s/it]
 71%|███████   | 17262/24420 [7:12:09<2:57:50,  1.49s/it]
 71%|███████   | 17263/24420 [7:12:11<2:57:52,  1.49s/it]
 71%|███████   | 17264/24420 [7:12:12<2:57:41,  1.49s/it]
 71%|███████   | 17265/24420 [7:12:14<2:57:34,  1.49s/it]
                                                         
{'loss': 0.0863, 'learning_rate': 5e-05, 'epoch': 3.54}

 71%|███████   | 17265/24420 [7:12:14<2:57:34,  1.49s/it]
 71%|███████   | 17266/24420 [7:12:15<2:57:39,  1.49s/it]
 71%|███████   | 17267/24420 [7:12:17<2:57:19,  1.49s/it]
 71%|███████   | 17268/24420 [7:12:18<2:57:38,  1.49s/it]
 71%|███████   | 17269/24420 [7:12:20<2:57:17,  1.49s/it]
 71%|███████   | 17270/24420 [7:12:21<2:57:50,  1.49s/it]
                                                         
{'loss': 0.0791, 'learning_rate': 5e-05, 'epoch': 3.54}

 71%|███████   | 17270/24420 [7:12:21<2:57:50,  1.49s/it]
 71%|███████   | 17271/24420 [7:12:23<2:57:51,  1.49s/it]
 71%|███████   | 17272/24420 [7:12:24<2:57:49,  1.49s/it]
 71%|███████   | 17273/24420 [7:12:26<2:57:37,  1.49s/it]
 71%|███████   | 17274/24420 [7:12:27<2:57:40,  1.49s/it]
 71%|███████   | 17275/24420 [7:12:29<2:57:21,  1.49s/it]
                                                         
{'loss': 0.0822, 'learning_rate': 5e-05, 'epoch': 3.54}

 71%|███████   | 17275/24420 [7:12:29<2:57:21,  1.49s/it]
 71%|███████   | 17276/24420 [7:12:30<2:57:44,  1.49s/it]
 71%|███████   | 17277/24420 [7:12:32<2:57:23,  1.49s/it]
 71%|███████   | 17278/24420 [7:12:33<2:57:20,  1.49s/it]
 71%|███████   | 17279/24420 [7:12:35<2:57:43,  1.49s/it]
 71%|███████   | 17280/24420 [7:12:36<2:57:17,  1.49s/it]
                                                         
{'loss': 0.0734, 'learning_rate': 5e-05, 'epoch': 3.54}

 71%|███████   | 17280/24420 [7:12:36<2:57:17,  1.49s/it]
 71%|███████   | 17281/24420 [7:12:38<2:56:59,  1.49s/it]
 71%|███████   | 17282/24420 [7:12:39<2:56:58,  1.49s/it]
 71%|███████   | 17283/24420 [7:12:41<2:56:53,  1.49s/it]
 71%|███████   | 17284/24420 [7:12:42<2:57:22,  1.49s/it]
 71%|███████   | 17285/24420 [7:12:44<2:57:09,  1.49s/it]
                                                         
{'loss': 0.069, 'learning_rate': 5e-05, 'epoch': 3.54}

 71%|███████   | 17285/24420 [7:12:44<2:57:09,  1.49s/it]
 71%|███████   | 17286/24420 [7:12:45<2:57:09,  1.49s/it]
 71%|███████   | 17287/24420 [7:12:47<2:57:29,  1.49s/it]
 71%|███████   | 17288/24420 [7:12:48<2:57:24,  1.49s/it]
 71%|███████   | 17289/24420 [7:12:50<2:57:08,  1.49s/it]
 71%|███████   | 17290/24420 [7:12:51<2:57:20,  1.49s/it]
                                                         
{'loss': 0.0874, 'learning_rate': 5e-05, 'epoch': 3.54}

 71%|███████   | 17290/24420 [7:12:51<2:57:20,  1.49s/it]
 71%|███████   | 17291/24420 [7:12:53<2:57:13,  1.49s/it]
 71%|███████   | 17292/24420 [7:12:54<2:57:09,  1.49s/it]
 71%|███████   | 17293/24420 [7:12:56<2:58:36,  1.50s/it]
 71%|███████   | 17294/24420 [7:12:57<2:58:40,  1.50s/it]
 71%|███████   | 17295/24420 [7:12:59<2:58:44,  1.51s/it]
                                                         
{'loss': 0.0699, 'learning_rate': 5e-05, 'epoch': 3.54}

 71%|███████   | 17295/24420 [7:12:59<2:58:44,  1.51s/it]
 71%|███████   | 17296/24420 [7:13:00<2:58:13,  1.50s/it]
 71%|███████   | 17297/24420 [7:13:02<2:57:45,  1.50s/it]
 71%|███████   | 17298/24420 [7:13:03<2:57:08,  1.49s/it]
 71%|███████   | 17299/24420 [7:13:05<2:56:57,  1.49s/it]
 71%|███████   | 17300/24420 [7:13:06<2:57:22,  1.49s/it]
                                                         
{'loss': 0.0856, 'learning_rate': 5e-05, 'epoch': 3.54}

 71%|███████   | 17300/24420 [7:13:06<2:57:22,  1.49s/it]
 71%|███████   | 17301/24420 [7:13:08<2:57:49,  1.50s/it]
 71%|███████   | 17302/24420 [7:13:09<2:57:32,  1.50s/it]
 71%|███████   | 17303/24420 [7:13:11<2:57:15,  1.49s/it]
 71%|███████   | 17304/24420 [7:13:12<2:56:51,  1.49s/it]
 71%|███████   | 17305/24420 [7:13:14<2:56:59,  1.49s/it]
                                                         
{'loss': 0.0751, 'learning_rate': 5e-05, 'epoch': 3.54}

 71%|███████   | 17305/24420 [7:13:14<2:56:59,  1.49s/it]
 71%|███████   | 17306/24420 [7:13:15<2:56:48,  1.49s/it]
 71%|███████   | 17307/24420 [7:13:17<2:56:47,  1.49s/it]
 71%|███████   | 17308/24420 [7:13:18<2:56:52,  1.49s/it]
 71%|███████   | 17309/24420 [7:13:20<2:56:49,  1.49s/it]
 71%|███████   | 17310/24420 [7:13:21<2:57:05,  1.49s/it]
                                                         
{'loss': 0.0762, 'learning_rate': 5e-05, 'epoch': 3.54}

 71%|███████   | 17310/24420 [7:13:21<2:57:05,  1.49s/it]
 71%|███████   | 17311/24420 [7:13:23<2:56:55,  1.49s/it]
 71%|███████   | 17312/24420 [7:13:24<2:56:53,  1.49s/it]
 71%|███████   | 17313/24420 [7:13:26<2:56:43,  1.49s/it]
 71%|███████   | 17314/24420 [7:13:27<2:56:31,  1.49s/it]
 71%|███████   | 17315/24420 [7:13:29<2:56:42,  1.49s/it]
                                                         
{'loss': 0.0763, 'learning_rate': 5e-05, 'epoch': 3.55}

 71%|███████   | 17315/24420 [7:13:29<2:56:42,  1.49s/it]
 71%|███████   | 17316/24420 [7:13:30<2:56:53,  1.49s/it]
 71%|███████   | 17317/24420 [7:13:32<2:56:57,  1.49s/it]
 71%|███████   | 17318/24420 [7:13:33<2:56:52,  1.49s/it]
 71%|███████   | 17319/24420 [7:13:35<2:56:44,  1.49s/it]
 71%|███████   | 17320/24420 [7:13:36<2:56:58,  1.50s/it]
                                                         
{'loss': 0.0918, 'learning_rate': 5e-05, 'epoch': 3.55}

 71%|███████   | 17320/24420 [7:13:36<2:56:58,  1.50s/it]
 71%|███████   | 17321/24420 [7:13:38<2:57:21,  1.50s/it]
 71%|███████   | 17322/24420 [7:13:39<2:57:09,  1.50s/it]
 71%|███████   | 17323/24420 [7:13:41<2:56:54,  1.50s/it]
 71%|███████   | 17324/24420 [7:13:42<2:56:47,  1.49s/it]
 71%|███████   | 17325/24420 [7:13:44<2:56:42,  1.49s/it]
                                                         
{'loss': 0.078, 'learning_rate': 5e-05, 'epoch': 3.55}

 71%|███████   | 17325/24420 [7:13:44<2:56:42,  1.49s/it]
 71%|███████   | 17326/24420 [7:13:45<2:57:07,  1.50s/it]
 71%|███████   | 17327/24420 [7:13:47<2:56:56,  1.50s/it]
 71%|███████   | 17328/24420 [7:13:48<2:56:51,  1.50s/it]
 71%|███████   | 17329/24420 [7:13:50<2:56:54,  1.50s/it]
 71%|███████   | 17330/24420 [7:13:51<2:56:59,  1.50s/it]
                                                         
{'loss': 0.101, 'learning_rate': 5e-05, 'epoch': 3.55}

 71%|███████   | 17330/24420 [7:13:51<2:56:59,  1.50s/it]
 71%|███████   | 17331/24420 [7:13:53<2:57:10,  1.50s/it]
 71%|███████   | 17332/24420 [7:13:54<2:56:59,  1.50s/it]
 71%|███████   | 17333/24420 [7:13:56<2:56:54,  1.50s/it]
 71%|███████   | 17334/24420 [7:13:57<2:56:32,  1.49s/it]
 71%|███████   | 17335/24420 [7:13:59<2:56:31,  1.49s/it]
                                                         
{'loss': 0.0781, 'learning_rate': 5e-05, 'epoch': 3.55}

 71%|███████   | 17335/24420 [7:13:59<2:56:31,  1.49s/it]
 71%|███████   | 17336/24420 [7:14:00<2:56:44,  1.50s/it]
 71%|███████   | 17337/24420 [7:14:02<2:56:52,  1.50s/it]
 71%|███████   | 17338/24420 [7:14:03<2:56:29,  1.50s/it]
 71%|███████   | 17339/24420 [7:14:04<2:56:35,  1.50s/it]
 71%|███████   | 17340/24420 [7:14:06<2:56:50,  1.50s/it]
                                                         
{'loss': 0.0791, 'learning_rate': 5e-05, 'epoch': 3.55}

 71%|███████   | 17340/24420 [7:14:06<2:56:50,  1.50s/it]
 71%|███████   | 17341/24420 [7:14:07<2:57:00,  1.50s/it]
 71%|███████   | 17342/24420 [7:14:09<2:56:53,  1.50s/it]
 71%|███████   | 17343/24420 [7:14:10<2:57:00,  1.50s/it]
 71%|███████   | 17344/24420 [7:14:12<2:56:42,  1.50s/it]
 71%|███████   | 17345/24420 [7:14:13<2:56:31,  1.50s/it]
                                                         
{'loss': 0.077, 'learning_rate': 5e-05, 'epoch': 3.55}

 71%|███████   | 17345/24420 [7:14:13<2:56:31,  1.50s/it]
 71%|███████   | 17346/24420 [7:14:15<2:57:02,  1.50s/it]
 71%|███████   | 17347/24420 [7:14:16<2:56:46,  1.50s/it]
 71%|███████   | 17348/24420 [7:14:18<2:56:48,  1.50s/it]
 71%|███████   | 17349/24420 [7:14:19<2:56:20,  1.50s/it]
 71%|███████   | 17350/24420 [7:14:21<2:55:51,  1.49s/it]
                                                         
{'loss': 0.0843, 'learning_rate': 5e-05, 'epoch': 3.55}

 71%|███████   | 17350/24420 [7:14:21<2:55:51,  1.49s/it]
 71%|███████   | 17351/24420 [7:14:22<2:56:01,  1.49s/it]
 71%|███████   | 17352/24420 [7:14:24<2:55:54,  1.49s/it]
 71%|███████   | 17353/24420 [7:14:25<2:55:51,  1.49s/it]
 71%|███████   | 17354/24420 [7:14:27<2:55:52,  1.49s/it]
 71%|███████   | 17355/24420 [7:14:28<2:55:51,  1.49s/it]
                                                         
{'loss': 0.0722, 'learning_rate': 5e-05, 'epoch': 3.55}

 71%|███████   | 17355/24420 [7:14:28<2:55:51,  1.49s/it]
 71%|███████   | 17356/24420 [7:14:30<2:55:52,  1.49s/it]
 71%|███████   | 17357/24420 [7:14:31<2:55:46,  1.49s/it]
 71%|███████   | 17358/24420 [7:14:33<2:56:07,  1.50s/it]
 71%|███████   | 17359/24420 [7:14:34<2:56:01,  1.50s/it]
 71%|███████   | 17360/24420 [7:14:36<2:55:53,  1.49s/it]
                                                         
{'loss': 0.0716, 'learning_rate': 5e-05, 'epoch': 3.55}

 71%|███████   | 17360/24420 [7:14:36<2:55:53,  1.49s/it]
 71%|███████   | 17361/24420 [7:14:37<2:56:04,  1.50s/it]
 71%|███████   | 17362/24420 [7:14:39<2:55:51,  1.49s/it]
 71%|███████   | 17363/24420 [7:14:40<2:55:45,  1.49s/it]
 71%|███████   | 17364/24420 [7:14:42<2:55:35,  1.49s/it]
 71%|███████   | 17365/24420 [7:14:43<2:55:55,  1.50s/it]
                                                         
{'loss': 0.1103, 'learning_rate': 5e-05, 'epoch': 3.56}

 71%|███████   | 17365/24420 [7:14:43<2:55:55,  1.50s/it]
 71%|███████   | 17366/24420 [7:14:45<2:55:57,  1.50s/it]
 71%|███████   | 17367/24420 [7:14:46<2:56:04,  1.50s/it]
 71%|███████   | 17368/24420 [7:14:48<2:55:40,  1.49s/it]
 71%|███████   | 17369/24420 [7:14:49<2:56:04,  1.50s/it]
 71%|███████   | 17370/24420 [7:14:51<2:55:34,  1.49s/it]
                                                         
{'loss': 0.0758, 'learning_rate': 5e-05, 'epoch': 3.56}

 71%|███████   | 17370/24420 [7:14:51<2:55:34,  1.49s/it]
 71%|███████   | 17371/24420 [7:14:52<2:55:53,  1.50s/it]
 71%|███████   | 17372/24420 [7:14:54<2:55:32,  1.49s/it]
 71%|███████   | 17373/24420 [7:14:55<2:55:20,  1.49s/it]
 71%|███████   | 17374/24420 [7:14:57<2:55:28,  1.49s/it]
 71%|███████   | 17375/24420 [7:14:58<2:55:25,  1.49s/it]
                                                         
{'loss': 0.0928, 'learning_rate': 5e-05, 'epoch': 3.56}

 71%|███████   | 17375/24420 [7:14:58<2:55:25,  1.49s/it]
 71%|███████   | 17376/24420 [7:15:00<2:55:23,  1.49s/it]
 71%|███████   | 17377/24420 [7:15:01<2:55:16,  1.49s/it]
 71%|███████   | 17378/24420 [7:15:03<2:55:09,  1.49s/it]
 71%|███████   | 17379/24420 [7:15:04<2:55:06,  1.49s/it]
 71%|███████   | 17380/24420 [7:15:06<2:55:15,  1.49s/it]
                                                         
{'loss': 0.0745, 'learning_rate': 5e-05, 'epoch': 3.56}

 71%|███████   | 17380/24420 [7:15:06<2:55:15,  1.49s/it]
 71%|███████   | 17381/24420 [7:15:07<2:55:15,  1.49s/it]
 71%|███████   | 17382/24420 [7:15:09<2:55:15,  1.49s/it]
 71%|███████   | 17383/24420 [7:15:10<2:55:23,  1.50s/it]
 71%|███████   | 17384/24420 [7:15:12<2:55:20,  1.50s/it]
 71%|███████   | 17385/24420 [7:15:13<2:55:10,  1.49s/it]
                                                         
{'loss': 0.0775, 'learning_rate': 5e-05, 'epoch': 3.56}

 71%|███████   | 17385/24420 [7:15:13<2:55:10,  1.49s/it]
 71%|███████   | 17386/24420 [7:15:15<2:55:16,  1.50s/it]
 71%|███████   | 17387/24420 [7:15:16<2:55:09,  1.49s/it]
 71%|███████   | 17388/24420 [7:15:18<2:55:09,  1.49s/it]
 71%|███████   | 17389/24420 [7:15:19<2:54:57,  1.49s/it]
 71%|███████   | 17390/24420 [7:15:21<2:54:58,  1.49s/it]
                                                         
{'loss': 0.0796, 'learning_rate': 5e-05, 'epoch': 3.56}

 71%|███████   | 17390/24420 [7:15:21<2:54:58,  1.49s/it]
 71%|███████   | 17391/24420 [7:15:22<2:55:32,  1.50s/it]
 71%|███████   | 17392/24420 [7:15:24<2:56:05,  1.50s/it]
 71%|███████   | 17393/24420 [7:15:25<2:55:50,  1.50s/it]
 71%|███████   | 17394/24420 [7:15:27<2:55:35,  1.50s/it]
 71%|███████   | 17395/24420 [7:15:28<2:55:19,  1.50s/it]
                                                         
{'loss': 0.0796, 'learning_rate': 5e-05, 'epoch': 3.56}

 71%|███████   | 17395/24420 [7:15:28<2:55:19,  1.50s/it]
 71%|███████   | 17396/24420 [7:15:30<2:54:55,  1.49s/it]
 71%|███████   | 17397/24420 [7:15:31<2:54:34,  1.49s/it]
 71%|███████   | 17398/24420 [7:15:33<2:54:31,  1.49s/it]
 71%|███████   | 17399/24420 [7:15:34<2:54:14,  1.49s/it]
 71%|███████▏  | 17400/24420 [7:15:36<2:54:36,  1.49s/it]
                                                         
{'loss': 0.0854, 'learning_rate': 5e-05, 'epoch': 3.56}

 71%|███████▏  | 17400/24420 [7:15:36<2:54:36,  1.49s/it]
 71%|███████▏  | 17401/24420 [7:15:40<4:46:20,  2.45s/it]
 71%|███████▏  | 17402/24420 [7:15:42<4:12:44,  2.16s/it]
 71%|███████▏  | 17403/24420 [7:15:43<3:49:10,  1.96s/it]
 71%|███████▏  | 17404/24420 [7:15:45<3:32:59,  1.82s/it]
 71%|███████▏  | 17405/24420 [7:15:46<3:21:32,  1.72s/it]
                                                         
{'loss': 0.085, 'learning_rate': 5e-05, 'epoch': 3.56}

 71%|███████▏  | 17405/24420 [7:15:46<3:21:32,  1.72s/it]
 71%|███████▏  | 17406/24420 [7:15:48<3:13:41,  1.66s/it]
 71%|███████▏  | 17407/24420 [7:15:49<3:07:58,  1.61s/it]
 71%|███████▏  | 17408/24420 [7:15:51<3:04:11,  1.58s/it]
 71%|███████▏  | 17409/24420 [7:15:52<3:01:16,  1.55s/it]
 71%|███████▏  | 17410/24420 [7:15:54<2:59:15,  1.53s/it]
                                                         
{'loss': 0.083, 'learning_rate': 5e-05, 'epoch': 3.56}

 71%|███████▏  | 17410/24420 [7:15:54<2:59:15,  1.53s/it]
 71%|███████▏  | 17411/24420 [7:15:55<2:58:18,  1.53s/it]
 71%|███████▏  | 17412/24420 [7:15:57<2:56:56,  1.51s/it]
 71%|███████▏  | 17413/24420 [7:15:58<2:56:39,  1.51s/it]
 71%|███████▏  | 17414/24420 [7:16:00<2:56:07,  1.51s/it]
 71%|███████▏  | 17415/24420 [7:16:01<2:55:48,  1.51s/it]
                                                         
{'loss': 0.0905, 'learning_rate': 5e-05, 'epoch': 3.57}

 71%|███████▏  | 17415/24420 [7:16:01<2:55:48,  1.51s/it]
 71%|███████▏  | 17416/24420 [7:16:03<2:55:13,  1.50s/it]
 71%|███████▏  | 17417/24420 [7:16:04<2:55:17,  1.50s/it]
 71%|███████▏  | 17418/24420 [7:16:06<2:55:02,  1.50s/it]
 71%|███████▏  | 17419/24420 [7:16:07<2:54:45,  1.50s/it]
 71%|███████▏  | 17420/24420 [7:16:09<2:54:56,  1.50s/it]
                                                         
{'loss': 0.0782, 'learning_rate': 5e-05, 'epoch': 3.57}

 71%|███████▏  | 17420/24420 [7:16:09<2:54:56,  1.50s/it]
 71%|███████▏  | 17421/24420 [7:16:10<2:54:45,  1.50s/it]
 71%|███████▏  | 17422/24420 [7:16:12<2:54:22,  1.50s/it]
 71%|███████▏  | 17423/24420 [7:16:13<2:54:20,  1.49s/it]
 71%|███████▏  | 17424/24420 [7:16:15<2:54:01,  1.49s/it]
 71%|███████▏  | 17425/24420 [7:16:16<2:54:43,  1.50s/it]
                                                         
{'loss': 0.0661, 'learning_rate': 5e-05, 'epoch': 3.57}

 71%|███████▏  | 17425/24420 [7:16:16<2:54:43,  1.50s/it]
 71%|███████▏  | 17426/24420 [7:16:18<2:54:30,  1.50s/it]
 71%|███████▏  | 17427/24420 [7:16:19<2:54:22,  1.50s/it]
 71%|███████▏  | 17428/24420 [7:16:21<2:54:18,  1.50s/it]
 71%|███████▏  | 17429/24420 [7:16:22<2:54:04,  1.49s/it]
 71%|███████▏  | 17430/24420 [7:16:24<2:53:46,  1.49s/it]
                                                         
{'loss': 0.0776, 'learning_rate': 5e-05, 'epoch': 3.57}

 71%|███████▏  | 17430/24420 [7:16:24<2:53:46,  1.49s/it]
 71%|███████▏  | 17431/24420 [7:16:25<2:53:48,  1.49s/it]
 71%|███████▏  | 17432/24420 [7:16:27<2:53:52,  1.49s/it]
 71%|███████▏  | 17433/24420 [7:16:28<2:53:56,  1.49s/it]
 71%|███████▏  | 17434/24420 [7:16:30<2:54:09,  1.50s/it]
 71%|███████▏  | 17435/24420 [7:16:31<2:54:10,  1.50s/it]
                                                         
{'loss': 0.0675, 'learning_rate': 5e-05, 'epoch': 3.57}

 71%|███████▏  | 17435/24420 [7:16:31<2:54:10,  1.50s/it]
 71%|███████▏  | 17436/24420 [7:16:33<2:54:07,  1.50s/it]
 71%|███████▏  | 17437/24420 [7:16:34<2:54:04,  1.50s/it]
 71%|███████▏  | 17438/24420 [7:16:36<2:53:36,  1.49s/it]
 71%|███████▏  | 17439/24420 [7:16:37<2:53:31,  1.49s/it]
 71%|███████▏  | 17440/24420 [7:16:39<2:53:54,  1.49s/it]
                                                         
{'loss': 0.0895, 'learning_rate': 5e-05, 'epoch': 3.57}

 71%|███████▏  | 17440/24420 [7:16:39<2:53:54,  1.49s/it]
 71%|███████▏  | 17441/24420 [7:16:40<2:53:58,  1.50s/it]
 71%|███████▏  | 17442/24420 [7:16:42<2:53:57,  1.50s/it]
 71%|███████▏  | 17443/24420 [7:16:43<2:53:36,  1.49s/it]
 71%|███████▏  | 17444/24420 [7:16:45<2:53:32,  1.49s/it]
 71%|███████▏  | 17445/24420 [7:16:46<2:53:53,  1.50s/it]
                                                         
{'loss': 0.0882, 'learning_rate': 5e-05, 'epoch': 3.57}

 71%|███████▏  | 17445/24420 [7:16:46<2:53:53,  1.50s/it]
 71%|███████▏  | 17446/24420 [7:16:48<2:53:43,  1.49s/it]
 71%|███████▏  | 17447/24420 [7:16:49<2:53:49,  1.50s/it]
 71%|███████▏  | 17448/24420 [7:16:51<2:54:06,  1.50s/it]
 71%|███████▏  | 17449/24420 [7:16:52<2:53:43,  1.50s/it]
 71%|███████▏  | 17450/24420 [7:16:54<2:53:58,  1.50s/it]
                                                         
{'loss': 0.0945, 'learning_rate': 5e-05, 'epoch': 3.57}

 71%|███████▏  | 17450/24420 [7:16:54<2:53:58,  1.50s/it]
 71%|███████▏  | 17451/24420 [7:16:55<2:54:02,  1.50s/it]
 71%|███████▏  | 17452/24420 [7:16:57<2:53:59,  1.50s/it]
 71%|███████▏  | 17453/24420 [7:16:58<2:54:14,  1.50s/it]
 71%|███████▏  | 17454/24420 [7:17:00<2:54:18,  1.50s/it]
 71%|███████▏  | 17455/24420 [7:17:01<2:53:50,  1.50s/it]
                                                         
{'loss': 0.1066, 'learning_rate': 5e-05, 'epoch': 3.57}

 71%|███████▏  | 17455/24420 [7:17:01<2:53:50,  1.50s/it]
 71%|███████▏  | 17456/24420 [7:17:03<2:53:47,  1.50s/it]
 71%|███████▏  | 17457/24420 [7:17:04<2:53:56,  1.50s/it]
 71%|███████▏  | 17458/24420 [7:17:06<2:53:46,  1.50s/it]
 71%|███████▏  | 17459/24420 [7:17:07<2:53:37,  1.50s/it]
 71%|███████▏  | 17460/24420 [7:17:09<2:53:38,  1.50s/it]
                                                         
{'loss': 0.0759, 'learning_rate': 5e-05, 'epoch': 3.57}

 71%|███████▏  | 17460/24420 [7:17:09<2:53:38,  1.50s/it]
 72%|███████▏  | 17461/24420 [7:17:10<2:53:34,  1.50s/it]
 72%|███████▏  | 17462/24420 [7:17:12<2:53:18,  1.49s/it]
 72%|███████▏  | 17463/24420 [7:17:13<2:53:31,  1.50s/it]
 72%|███████▏  | 17464/24420 [7:17:15<2:53:21,  1.50s/it]
 72%|███████▏  | 17465/24420 [7:17:16<2:53:09,  1.49s/it]
                                                         
{'loss': 0.0781, 'learning_rate': 5e-05, 'epoch': 3.58}

 72%|███████▏  | 17465/24420 [7:17:16<2:53:09,  1.49s/it]
 72%|███████▏  | 17466/24420 [7:17:18<2:52:56,  1.49s/it]
 72%|███████▏  | 17467/24420 [7:17:19<2:52:58,  1.49s/it]
 72%|███████▏  | 17468/24420 [7:17:21<2:53:12,  1.49s/it]
 72%|███████▏  | 17469/24420 [7:17:22<2:53:08,  1.49s/it]
 72%|███████▏  | 17470/24420 [7:17:24<2:53:18,  1.50s/it]
                                                         
{'loss': 0.0876, 'learning_rate': 5e-05, 'epoch': 3.58}

 72%|███████▏  | 17470/24420 [7:17:24<2:53:18,  1.50s/it]
 72%|███████▏  | 17471/24420 [7:17:25<2:53:17,  1.50s/it]
 72%|███████▏  | 17472/24420 [7:17:27<2:52:55,  1.49s/it]
 72%|███████▏  | 17473/24420 [7:17:28<2:53:03,  1.49s/it]
 72%|███████▏  | 17474/24420 [7:17:30<2:53:05,  1.50s/it]
 72%|███████▏  | 17475/24420 [7:17:31<2:52:41,  1.49s/it]
                                                         
{'loss': 0.0876, 'learning_rate': 5e-05, 'epoch': 3.58}

 72%|███████▏  | 17475/24420 [7:17:31<2:52:41,  1.49s/it]
 72%|███████▏  | 17476/24420 [7:17:33<2:52:35,  1.49s/it]
 72%|███████▏  | 17477/24420 [7:17:34<2:52:36,  1.49s/it]
 72%|███████▏  | 17478/24420 [7:17:36<2:52:56,  1.49s/it]
 72%|███████▏  | 17479/24420 [7:17:37<2:53:05,  1.50s/it]
 72%|███████▏  | 17480/24420 [7:17:39<2:52:39,  1.49s/it]
                                                         
{'loss': 0.0747, 'learning_rate': 5e-05, 'epoch': 3.58}

 72%|███████▏  | 17480/24420 [7:17:39<2:52:39,  1.49s/it]
 72%|███████▏  | 17481/24420 [7:17:40<2:52:46,  1.49s/it]
 72%|███████▏  | 17482/24420 [7:17:42<2:52:48,  1.49s/it]
 72%|███████▏  | 17483/24420 [7:17:43<2:52:39,  1.49s/it]
 72%|███████▏  | 17484/24420 [7:17:45<2:52:23,  1.49s/it]
 72%|███████▏  | 17485/24420 [7:17:46<2:52:33,  1.49s/it]
                                                         
{'loss': 0.0733, 'learning_rate': 5e-05, 'epoch': 3.58}

 72%|███████▏  | 17485/24420 [7:17:46<2:52:33,  1.49s/it]
 72%|███████▏  | 17486/24420 [7:17:47<2:52:47,  1.50s/it]
 72%|███████▏  | 17487/24420 [7:17:49<2:52:49,  1.50s/it]
 72%|███████▏  | 17488/24420 [7:17:50<2:52:46,  1.50s/it]
 72%|███████▏  | 17489/24420 [7:17:52<2:53:03,  1.50s/it]
 72%|███████▏  | 17490/24420 [7:17:53<2:53:00,  1.50s/it]
                                                         
{'loss': 0.0878, 'learning_rate': 5e-05, 'epoch': 3.58}

 72%|███████▏  | 17490/24420 [7:17:53<2:53:00,  1.50s/it]
 72%|███████▏  | 17491/24420 [7:17:55<2:52:47,  1.50s/it]
 72%|███████▏  | 17492/24420 [7:17:56<2:52:20,  1.49s/it]
 72%|███████▏  | 17493/24420 [7:17:58<2:51:58,  1.49s/it]
 72%|███████▏  | 17494/24420 [7:17:59<2:52:18,  1.49s/it]
 72%|███████▏  | 17495/24420 [7:18:01<2:52:27,  1.49s/it]
                                                         
{'loss': 0.0733, 'learning_rate': 5e-05, 'epoch': 3.58}

 72%|███████▏  | 17495/24420 [7:18:01<2:52:27,  1.49s/it]
 72%|███████▏  | 17496/24420 [7:18:02<2:52:42,  1.50s/it]
 72%|███████▏  | 17497/24420 [7:18:04<2:52:30,  1.50s/it]
 72%|███████▏  | 17498/24420 [7:18:05<2:52:15,  1.49s/it]
 72%|███████▏  | 17499/24420 [7:18:07<2:52:16,  1.49s/it]
 72%|███████▏  | 17500/24420 [7:18:08<2:52:10,  1.49s/it]
                                                         
{'loss': 0.0678, 'learning_rate': 5e-05, 'epoch': 3.58}

 72%|███████▏  | 17500/24420 [7:18:08<2:52:10,  1.49s/it]
 72%|███████▏  | 17501/24420 [7:18:10<2:52:04,  1.49s/it]
 72%|███████▏  | 17502/24420 [7:18:11<2:52:23,  1.50s/it]
 72%|███████▏  | 17503/24420 [7:18:13<2:52:12,  1.49s/it]
 72%|███████▏  | 17504/24420 [7:18:14<2:52:19,  1.49s/it]
 72%|███████▏  | 17505/24420 [7:18:16<2:52:30,  1.50s/it]
                                                         
{'loss': 0.077, 'learning_rate': 5e-05, 'epoch': 3.58}

 72%|███████▏  | 17505/24420 [7:18:16<2:52:30,  1.50s/it]
 72%|███████▏  | 17506/24420 [7:18:17<2:53:03,  1.50s/it]
 72%|███████▏  | 17507/24420 [7:18:19<2:52:48,  1.50s/it]
 72%|███████▏  | 17508/24420 [7:18:20<2:52:33,  1.50s/it]
 72%|███████▏  | 17509/24420 [7:18:22<2:51:57,  1.49s/it]
 72%|███████▏  | 17510/24420 [7:18:23<2:51:44,  1.49s/it]
                                                         
{'loss': 0.0931, 'learning_rate': 5e-05, 'epoch': 3.59}

 72%|███████▏  | 17510/24420 [7:18:23<2:51:44,  1.49s/it]
 72%|███████▏  | 17511/24420 [7:18:25<2:51:32,  1.49s/it]
 72%|███████▏  | 17512/24420 [7:18:26<2:51:28,  1.49s/it]
 72%|███████▏  | 17513/24420 [7:18:28<2:51:16,  1.49s/it]
 72%|███████▏  | 17514/24420 [7:18:29<2:51:34,  1.49s/it]
 72%|███████▏  | 17515/24420 [7:18:31<2:51:27,  1.49s/it]
                                                         
{'loss': 0.0792, 'learning_rate': 5e-05, 'epoch': 3.59}

 72%|███████▏  | 17515/24420 [7:18:31<2:51:27,  1.49s/it]
 72%|███████▏  | 17516/24420 [7:18:32<2:51:39,  1.49s/it]
 72%|███████▏  | 17517/24420 [7:18:34<2:51:16,  1.49s/it]
 72%|███████▏  | 17518/24420 [7:18:35<2:51:23,  1.49s/it]
 72%|███████▏  | 17519/24420 [7:18:37<2:51:18,  1.49s/it]
 72%|███████▏  | 17520/24420 [7:18:38<2:51:18,  1.49s/it]
                                                         
{'loss': 0.0722, 'learning_rate': 5e-05, 'epoch': 3.59}

 72%|███████▏  | 17520/24420 [7:18:38<2:51:18,  1.49s/it]
 72%|███████▏  | 17521/24420 [7:18:40<2:50:44,  1.48s/it]
 72%|███████▏  | 17522/24420 [7:18:41<2:50:38,  1.48s/it]
 72%|███████▏  | 17523/24420 [7:18:43<2:50:31,  1.48s/it]
 72%|███████▏  | 17524/24420 [7:18:44<2:50:23,  1.48s/it]
 72%|███████▏  | 17525/24420 [7:18:46<2:50:48,  1.49s/it]
                                                         
{'loss': 0.0746, 'learning_rate': 5e-05, 'epoch': 3.59}

 72%|███████▏  | 17525/24420 [7:18:46<2:50:48,  1.49s/it]
 72%|███████▏  | 17526/24420 [7:18:47<2:50:31,  1.48s/it]
 72%|███████▏  | 17527/24420 [7:18:49<2:50:11,  1.48s/it]
 72%|███████▏  | 17528/24420 [7:18:50<2:49:57,  1.48s/it]
 72%|███████▏  | 17529/24420 [7:18:52<2:50:26,  1.48s/it]
 72%|███████▏  | 17530/24420 [7:18:53<2:50:35,  1.49s/it]
                                                         
{'loss': 0.07, 'learning_rate': 5e-05, 'epoch': 3.59}

 72%|███████▏  | 17530/24420 [7:18:53<2:50:35,  1.49s/it]
 72%|███████▏  | 17531/24420 [7:18:55<2:51:05,  1.49s/it]
 72%|███████▏  | 17532/24420 [7:18:56<2:50:50,  1.49s/it]
 72%|███████▏  | 17533/24420 [7:18:58<2:50:38,  1.49s/it]
 72%|███████▏  | 17534/24420 [7:18:59<2:50:37,  1.49s/it]
 72%|███████▏  | 17535/24420 [7:19:01<2:50:36,  1.49s/it]
                                                         
{'loss': 0.0712, 'learning_rate': 5e-05, 'epoch': 3.59}

 72%|███████▏  | 17535/24420 [7:19:01<2:50:36,  1.49s/it]
 72%|███████▏  | 17536/24420 [7:19:02<2:50:36,  1.49s/it]
 72%|███████▏  | 17537/24420 [7:19:04<2:50:56,  1.49s/it]
 72%|███████▏  | 17538/24420 [7:19:05<2:50:47,  1.49s/it]
 72%|███████▏  | 17539/24420 [7:19:06<2:50:36,  1.49s/it]
 72%|███████▏  | 17540/24420 [7:19:08<2:50:22,  1.49s/it]
                                                         
{'loss': 0.0849, 'learning_rate': 5e-05, 'epoch': 3.59}

 72%|███████▏  | 17540/24420 [7:19:08<2:50:22,  1.49s/it]
 72%|███████▏  | 17541/24420 [7:19:09<2:50:28,  1.49s/it]
 72%|███████▏  | 17542/24420 [7:19:11<2:50:13,  1.48s/it]
 72%|███████▏  | 17543/24420 [7:19:12<2:50:31,  1.49s/it]
 72%|███████▏  | 17544/24420 [7:19:14<2:50:32,  1.49s/it]
 72%|███████▏  | 17545/24420 [7:19:15<2:50:25,  1.49s/it]
                                                         
{'loss': 0.0913, 'learning_rate': 5e-05, 'epoch': 3.59}

 72%|███████▏  | 17545/24420 [7:19:15<2:50:25,  1.49s/it]
 72%|███████▏  | 17546/24420 [7:19:17<2:50:20,  1.49s/it]
 72%|███████▏  | 17547/24420 [7:19:18<2:50:13,  1.49s/it]
 72%|███████▏  | 17548/24420 [7:19:20<2:50:19,  1.49s/it]
 72%|███████▏  | 17549/24420 [7:19:21<2:51:00,  1.49s/it]
 72%|███████▏  | 17550/24420 [7:19:23<2:51:11,  1.50s/it]
                                                         
{'loss': 0.0858, 'learning_rate': 5e-05, 'epoch': 3.59}

 72%|███████▏  | 17550/24420 [7:19:23<2:51:11,  1.50s/it]
 72%|███████▏  | 17551/24420 [7:19:24<2:50:59,  1.49s/it]
 72%|███████▏  | 17552/24420 [7:19:26<2:51:22,  1.50s/it]
 72%|███████▏  | 17553/24420 [7:19:27<2:50:57,  1.49s/it]
 72%|███████▏  | 17554/24420 [7:19:29<2:51:08,  1.50s/it]
 72%|███████▏  | 17555/24420 [7:19:30<2:51:08,  1.50s/it]
                                                         
{'loss': 0.0752, 'learning_rate': 5e-05, 'epoch': 3.59}

 72%|███████▏  | 17555/24420 [7:19:30<2:51:08,  1.50s/it]
 72%|███████▏  | 17556/24420 [7:19:32<2:51:00,  1.49s/it]
 72%|███████▏  | 17557/24420 [7:19:33<2:50:43,  1.49s/it]
 72%|███████▏  | 17558/24420 [7:19:35<2:50:23,  1.49s/it]
 72%|███████▏  | 17559/24420 [7:19:36<2:50:07,  1.49s/it]
 72%|███████▏  | 17560/24420 [7:19:38<2:49:55,  1.49s/it]
                                                         
{'loss': 0.0811, 'learning_rate': 5e-05, 'epoch': 3.6}

 72%|███████▏  | 17560/24420 [7:19:38<2:49:55,  1.49s/it]
 72%|███████▏  | 17561/24420 [7:19:39<2:50:20,  1.49s/it]
 72%|███████▏  | 17562/24420 [7:19:41<2:50:14,  1.49s/it]
 72%|███████▏  | 17563/24420 [7:19:42<2:50:32,  1.49s/it]
 72%|███████▏  | 17564/24420 [7:19:44<2:50:14,  1.49s/it]
 72%|███████▏  | 17565/24420 [7:19:45<2:50:06,  1.49s/it]
                                                         
{'loss': 0.0835, 'learning_rate': 5e-05, 'epoch': 3.6}

 72%|███████▏  | 17565/24420 [7:19:45<2:50:06,  1.49s/it]
 72%|███████▏  | 17566/24420 [7:19:47<2:49:50,  1.49s/it]
 72%|███████▏  | 17567/24420 [7:19:48<2:50:07,  1.49s/it]
 72%|███████▏  | 17568/24420 [7:19:50<2:50:21,  1.49s/it]
 72%|███████▏  | 17569/24420 [7:19:51<2:50:07,  1.49s/it]
 72%|███████▏  | 17570/24420 [7:19:53<2:49:59,  1.49s/it]
                                                         
{'loss': 0.0772, 'learning_rate': 5e-05, 'epoch': 3.6}

 72%|███████▏  | 17570/24420 [7:19:53<2:49:59,  1.49s/it]
 72%|███████▏  | 17571/24420 [7:19:54<2:50:11,  1.49s/it]
 72%|███████▏  | 17572/24420 [7:19:56<2:49:59,  1.49s/it]
 72%|███████▏  | 17573/24420 [7:19:57<2:49:38,  1.49s/it]
 72%|███████▏  | 17574/24420 [7:19:59<2:49:39,  1.49s/it]
 72%|███████▏  | 17575/24420 [7:20:00<2:49:42,  1.49s/it]
                                                         
{'loss': 0.085, 'learning_rate': 5e-05, 'epoch': 3.6}

 72%|███████▏  | 17575/24420 [7:20:00<2:49:42,  1.49s/it]
 72%|███████▏  | 17576/24420 [7:20:02<2:49:32,  1.49s/it]
 72%|███████▏  | 17577/24420 [7:20:03<2:49:52,  1.49s/it]
 72%|███████▏  | 17578/24420 [7:20:05<2:49:37,  1.49s/it]
 72%|███████▏  | 17579/24420 [7:20:06<2:49:36,  1.49s/it]
 72%|███████▏  | 17580/24420 [7:20:08<2:49:20,  1.49s/it]
                                                         
{'loss': 0.0724, 'learning_rate': 5e-05, 'epoch': 3.6}

 72%|███████▏  | 17580/24420 [7:20:08<2:49:20,  1.49s/it]
 72%|███████▏  | 17581/24420 [7:20:09<2:49:16,  1.49s/it]
 72%|███████▏  | 17582/24420 [7:20:11<2:49:41,  1.49s/it]
 72%|███████▏  | 17583/24420 [7:20:12<2:49:22,  1.49s/it]
 72%|███████▏  | 17584/24420 [7:20:14<2:49:28,  1.49s/it]
 72%|███████▏  | 17585/24420 [7:20:15<2:49:25,  1.49s/it]
                                                         
{'loss': 0.0746, 'learning_rate': 5e-05, 'epoch': 3.6}

 72%|███████▏  | 17585/24420 [7:20:15<2:49:25,  1.49s/it]
 72%|███████▏  | 17586/24420 [7:20:16<2:49:33,  1.49s/it]
 72%|███████▏  | 17587/24420 [7:20:18<2:49:32,  1.49s/it]
 72%|███████▏  | 17588/24420 [7:20:19<2:49:16,  1.49s/it]
 72%|███████▏  | 17589/24420 [7:20:21<2:49:31,  1.49s/it]
 72%|███████▏  | 17590/24420 [7:20:22<2:49:33,  1.49s/it]
                                                         
{'loss': 0.0792, 'learning_rate': 5e-05, 'epoch': 3.6}

 72%|███████▏  | 17590/24420 [7:20:22<2:49:33,  1.49s/it]
 72%|███████▏  | 17591/24420 [7:20:24<2:49:24,  1.49s/it]
 72%|███████▏  | 17592/24420 [7:20:25<2:49:18,  1.49s/it]
 72%|███████▏  | 17593/24420 [7:20:27<2:49:05,  1.49s/it]
 72%|███████▏  | 17594/24420 [7:20:28<2:48:58,  1.49s/it]
 72%|███████▏  | 17595/24420 [7:20:30<2:48:50,  1.48s/it]
                                                         
{'loss': 0.0788, 'learning_rate': 5e-05, 'epoch': 3.6}

 72%|███████▏  | 17595/24420 [7:20:30<2:48:50,  1.48s/it]
 72%|███████▏  | 17596/24420 [7:20:31<2:48:41,  1.48s/it]
 72%|███████▏  | 17597/24420 [7:20:33<2:48:43,  1.48s/it]
 72%|███████▏  | 17598/24420 [7:20:34<2:48:29,  1.48s/it]
 72%|███████▏  | 17599/24420 [7:20:36<2:48:26,  1.48s/it]
 72%|███████▏  | 17600/24420 [7:20:37<2:48:30,  1.48s/it]
                                                         
{'loss': 0.0798, 'learning_rate': 5e-05, 'epoch': 3.6}

 72%|███████▏  | 17600/24420 [7:20:37<2:48:30,  1.48s/it]
 72%|███████▏  | 17601/24420 [7:20:42<4:35:16,  2.42s/it]
 72%|███████▏  | 17602/24420 [7:20:43<4:03:08,  2.14s/it]
 72%|███████▏  | 17603/24420 [7:20:45<3:41:02,  1.95s/it]
 72%|███████▏  | 17604/24420 [7:20:46<3:25:40,  1.81s/it]
 72%|███████▏  | 17605/24420 [7:20:48<3:14:28,  1.71s/it]
                                                         
{'loss': 0.0823, 'learning_rate': 5e-05, 'epoch': 3.6}

 72%|███████▏  | 17605/24420 [7:20:48<3:14:28,  1.71s/it]
 72%|███████▏  | 17606/24420 [7:20:49<3:06:37,  1.64s/it]
 72%|███████▏  | 17607/24420 [7:20:51<3:01:16,  1.60s/it]
 72%|███████▏  | 17608/24420 [7:20:52<2:57:17,  1.56s/it]
 72%|███████▏  | 17609/24420 [7:20:54<2:54:49,  1.54s/it]
 72%|███████▏  | 17610/24420 [7:20:55<2:52:40,  1.52s/it]
                                                         
{'loss': 0.0842, 'learning_rate': 5e-05, 'epoch': 3.61}

 72%|███████▏  | 17610/24420 [7:20:55<2:52:40,  1.52s/it]
 72%|███████▏  | 17611/24420 [7:20:57<2:51:21,  1.51s/it]
 72%|███████▏  | 17612/24420 [7:20:58<2:50:31,  1.50s/it]
 72%|███████▏  | 17613/24420 [7:21:00<2:49:46,  1.50s/it]
 72%|███████▏  | 17614/24420 [7:21:01<2:49:35,  1.50s/it]
 72%|███████▏  | 17615/24420 [7:21:03<2:49:20,  1.49s/it]
                                                         
{'loss': 0.0809, 'learning_rate': 5e-05, 'epoch': 3.61}

 72%|███████▏  | 17615/24420 [7:21:03<2:49:20,  1.49s/it]
 72%|███████▏  | 17616/24420 [7:21:04<2:48:55,  1.49s/it]
 72%|███████▏  | 17617/24420 [7:21:06<2:48:40,  1.49s/it]
 72%|███████▏  | 17618/24420 [7:21:07<2:48:23,  1.49s/it]
 72%|███████▏  | 17619/24420 [7:21:09<2:48:18,  1.48s/it]
 72%|███████▏  | 17620/24420 [7:21:10<2:48:04,  1.48s/it]
                                                         
{'loss': 0.0787, 'learning_rate': 5e-05, 'epoch': 3.61}

 72%|███████▏  | 17620/24420 [7:21:10<2:48:04,  1.48s/it]
 72%|███████▏  | 17621/24420 [7:21:12<2:47:57,  1.48s/it]
 72%|███████▏  | 17622/24420 [7:21:13<2:48:02,  1.48s/it]
 72%|███████▏  | 17623/24420 [7:21:15<2:48:07,  1.48s/it]
 72%|███████▏  | 17624/24420 [7:21:16<2:48:11,  1.48s/it]
 72%|███████▏  | 17625/24420 [7:21:18<2:48:02,  1.48s/it]
                                                         
{'loss': 0.0807, 'learning_rate': 5e-05, 'epoch': 3.61}

 72%|███████▏  | 17625/24420 [7:21:18<2:48:02,  1.48s/it]
 72%|███████▏  | 17626/24420 [7:21:19<2:48:22,  1.49s/it]
 72%|███████▏  | 17627/24420 [7:21:20<2:48:04,  1.48s/it]
 72%|███████▏  | 17628/24420 [7:21:22<2:48:10,  1.49s/it]
 72%|███████▏  | 17629/24420 [7:21:23<2:48:10,  1.49s/it]
 72%|███████▏  | 17630/24420 [7:21:25<2:48:10,  1.49s/it]
                                                         
{'loss': 0.0804, 'learning_rate': 5e-05, 'epoch': 3.61}

 72%|███████▏  | 17630/24420 [7:21:25<2:48:10,  1.49s/it]
 72%|███████▏  | 17631/24420 [7:21:26<2:48:05,  1.49s/it]
 72%|███████▏  | 17632/24420 [7:21:28<2:48:05,  1.49s/it]
 72%|███████▏  | 17633/24420 [7:21:29<2:47:53,  1.48s/it]
 72%|███████▏  | 17634/24420 [7:21:31<2:48:02,  1.49s/it]
 72%|███████▏  | 17635/24420 [7:21:32<2:47:58,  1.49s/it]
                                                         
{'loss': 0.0802, 'learning_rate': 5e-05, 'epoch': 3.61}

 72%|███████▏  | 17635/24420 [7:21:32<2:47:58,  1.49s/it]
 72%|███████▏  | 17636/24420 [7:21:34<2:47:52,  1.48s/it]
 72%|███████▏  | 17637/24420 [7:21:35<2:48:14,  1.49s/it]
 72%|███████▏  | 17638/24420 [7:21:37<2:47:58,  1.49s/it]
 72%|███████▏  | 17639/24420 [7:21:38<2:48:07,  1.49s/it]
 72%|███████▏  | 17640/24420 [7:21:40<2:48:00,  1.49s/it]
                                                         
{'loss': 0.0875, 'learning_rate': 5e-05, 'epoch': 3.61}

 72%|███████▏  | 17640/24420 [7:21:40<2:48:00,  1.49s/it]
 72%|███████▏  | 17641/24420 [7:21:41<2:47:59,  1.49s/it]
 72%|███████▏  | 17642/24420 [7:21:43<2:48:04,  1.49s/it]
 72%|███████▏  | 17643/24420 [7:21:44<2:47:52,  1.49s/it]
 72%|███████▏  | 17644/24420 [7:21:46<2:47:48,  1.49s/it]
 72%|███████▏  | 17645/24420 [7:21:47<2:48:00,  1.49s/it]
                                                         
{'loss': 0.0582, 'learning_rate': 5e-05, 'epoch': 3.61}

 72%|███████▏  | 17645/24420 [7:21:47<2:48:00,  1.49s/it]
 72%|███████▏  | 17646/24420 [7:21:49<2:47:54,  1.49s/it]
 72%|███████▏  | 17647/24420 [7:21:50<2:47:55,  1.49s/it]
 72%|███████▏  | 17648/24420 [7:21:52<2:47:46,  1.49s/it]
 72%|███████▏  | 17649/24420 [7:21:53<2:48:17,  1.49s/it]
 72%|███████▏  | 17650/24420 [7:21:55<2:47:57,  1.49s/it]
                                                         
{'loss': 0.0794, 'learning_rate': 5e-05, 'epoch': 3.61}

 72%|███████▏  | 17650/24420 [7:21:55<2:47:57,  1.49s/it]
 72%|███████▏  | 17651/24420 [7:21:56<2:47:36,  1.49s/it]
 72%|███████▏  | 17652/24420 [7:21:58<2:47:22,  1.48s/it]
 72%|███████▏  | 17653/24420 [7:21:59<2:47:26,  1.48s/it]
 72%|███████▏  | 17654/24420 [7:22:01<2:47:52,  1.49s/it]
 72%|███████▏  | 17655/24420 [7:22:02<2:47:49,  1.49s/it]
                                                         
{'loss': 0.0736, 'learning_rate': 5e-05, 'epoch': 3.61}

 72%|███████▏  | 17655/24420 [7:22:02<2:47:49,  1.49s/it]
 72%|███████▏  | 17656/24420 [7:22:04<2:47:42,  1.49s/it]
 72%|███████▏  | 17657/24420 [7:22:05<2:47:12,  1.48s/it]
 72%|███████▏  | 17658/24420 [7:22:07<2:47:03,  1.48s/it]
 72%|███████▏  | 17659/24420 [7:22:08<2:47:04,  1.48s/it]
 72%|███████▏  | 17660/24420 [7:22:10<2:46:59,  1.48s/it]
                                                         
{'loss': 0.0689, 'learning_rate': 5e-05, 'epoch': 3.62}

 72%|███████▏  | 17660/24420 [7:22:10<2:46:59,  1.48s/it]
 72%|███████▏  | 17661/24420 [7:22:11<2:47:00,  1.48s/it]
 72%|███████▏  | 17662/24420 [7:22:12<2:47:17,  1.49s/it]
 72%|███████▏  | 17663/24420 [7:22:14<2:47:21,  1.49s/it]
 72%|███████▏  | 17664/24420 [7:22:15<2:47:54,  1.49s/it]
 72%|███████▏  | 17665/24420 [7:22:17<2:47:52,  1.49s/it]
                                                         
{'loss': 0.073, 'learning_rate': 5e-05, 'epoch': 3.62}

 72%|███████▏  | 17665/24420 [7:22:17<2:47:52,  1.49s/it]
 72%|███████▏  | 17666/24420 [7:22:18<2:47:48,  1.49s/it]
 72%|███████▏  | 17667/24420 [7:22:20<2:47:29,  1.49s/it]
 72%|███████▏  | 17668/24420 [7:22:21<2:47:23,  1.49s/it]
 72%|███████▏  | 17669/24420 [7:22:23<2:47:24,  1.49s/it]
 72%|███████▏  | 17670/24420 [7:22:24<2:47:04,  1.49s/it]
                                                         
{'loss': 0.0789, 'learning_rate': 5e-05, 'epoch': 3.62}

 72%|███████▏  | 17670/24420 [7:22:24<2:47:04,  1.49s/it]
 72%|███████▏  | 17671/24420 [7:22:26<2:47:27,  1.49s/it]
 72%|███████▏  | 17672/24420 [7:22:27<2:47:07,  1.49s/it]
 72%|███████▏  | 17673/24420 [7:22:29<2:46:53,  1.48s/it]
 72%|███████▏  | 17674/24420 [7:22:30<2:46:37,  1.48s/it]
 72%|███████▏  | 17675/24420 [7:22:32<2:46:56,  1.49s/it]
                                                         
{'loss': 0.0825, 'learning_rate': 5e-05, 'epoch': 3.62}

 72%|███████▏  | 17675/24420 [7:22:32<2:46:56,  1.49s/it]
 72%|███████▏  | 17676/24420 [7:22:33<2:47:09,  1.49s/it]
 72%|███████▏  | 17677/24420 [7:22:35<2:47:27,  1.49s/it]
 72%|███████▏  | 17678/24420 [7:22:36<2:47:26,  1.49s/it]
 72%|███████▏  | 17679/24420 [7:22:38<2:47:56,  1.49s/it]
 72%|███████▏  | 17680/24420 [7:22:39<2:48:01,  1.50s/it]
                                                         
{'loss': 0.0811, 'learning_rate': 5e-05, 'epoch': 3.62}

 72%|███████▏  | 17680/24420 [7:22:39<2:48:01,  1.50s/it]
 72%|███████▏  | 17681/24420 [7:22:41<2:47:42,  1.49s/it]
 72%|███████▏  | 17682/24420 [7:22:42<2:47:28,  1.49s/it]
 72%|███████▏  | 17683/24420 [7:22:44<2:47:17,  1.49s/it]
 72%|███████▏  | 17684/24420 [7:22:45<2:47:09,  1.49s/it]
 72%|███████▏  | 17685/24420 [7:22:47<2:46:47,  1.49s/it]
                                                         
{'loss': 0.0801, 'learning_rate': 5e-05, 'epoch': 3.62}

 72%|███████▏  | 17685/24420 [7:22:47<2:46:47,  1.49s/it]
 72%|███████▏  | 17686/24420 [7:22:48<2:46:42,  1.49s/it]
 72%|███████▏  | 17687/24420 [7:22:50<2:46:50,  1.49s/it]
 72%|███████▏  | 17688/24420 [7:22:51<2:46:50,  1.49s/it]
 72%|███████▏  | 17689/24420 [7:22:53<2:46:51,  1.49s/it]
 72%|███████▏  | 17690/24420 [7:22:54<2:46:51,  1.49s/it]
                                                         
{'loss': 0.0817, 'learning_rate': 5e-05, 'epoch': 3.62}

 72%|███████▏  | 17690/24420 [7:22:54<2:46:51,  1.49s/it]
 72%|███████▏  | 17691/24420 [7:22:56<2:46:41,  1.49s/it]
 72%|███████▏  | 17692/24420 [7:22:57<2:46:32,  1.49s/it]
 72%|███████▏  | 17693/24420 [7:22:59<2:46:34,  1.49s/it]
 72%|███████▏  | 17694/24420 [7:23:00<2:46:28,  1.49s/it]
 72%|███████▏  | 17695/24420 [7:23:02<2:46:24,  1.48s/it]
                                                         
{'loss': 0.0781, 'learning_rate': 5e-05, 'epoch': 3.62}

 72%|███████▏  | 17695/24420 [7:23:02<2:46:24,  1.48s/it]
 72%|███████▏  | 17696/24420 [7:23:03<2:46:13,  1.48s/it]
 72%|███████▏  | 17697/24420 [7:23:05<2:46:13,  1.48s/it]
 72%|███████▏  | 17698/24420 [7:23:06<2:46:00,  1.48s/it]
 72%|███████▏  | 17699/24420 [7:23:08<2:46:44,  1.49s/it]
 72%|███████▏  | 17700/24420 [7:23:09<2:46:24,  1.49s/it]
                                                         
{'loss': 0.0944, 'learning_rate': 5e-05, 'epoch': 3.62}

 72%|███████▏  | 17700/24420 [7:23:09<2:46:24,  1.49s/it]
 72%|███████▏  | 17701/24420 [7:23:11<2:46:25,  1.49s/it]
 72%|███████▏  | 17702/24420 [7:23:12<2:46:20,  1.49s/it]
 72%|███████▏  | 17703/24420 [7:23:13<2:46:35,  1.49s/it]
 72%|███████▏  | 17704/24420 [7:23:15<2:46:42,  1.49s/it]
 73%|███████▎  | 17705/24420 [7:23:16<2:46:23,  1.49s/it]
                                                         
{'loss': 0.0678, 'learning_rate': 5e-05, 'epoch': 3.63}

 73%|███████▎  | 17705/24420 [7:23:16<2:46:23,  1.49s/it]
 73%|███████▎  | 17706/24420 [7:23:18<2:46:27,  1.49s/it]
 73%|███████▎  | 17707/24420 [7:23:19<2:46:34,  1.49s/it]
 73%|███████▎  | 17708/24420 [7:23:21<2:46:37,  1.49s/it]
 73%|███████▎  | 17709/24420 [7:23:22<2:46:29,  1.49s/it]
 73%|███████▎  | 17710/24420 [7:23:24<2:46:27,  1.49s/it]
                                                         
{'loss': 0.0851, 'learning_rate': 5e-05, 'epoch': 3.63}

 73%|███████▎  | 17710/24420 [7:23:24<2:46:27,  1.49s/it]
 73%|███████▎  | 17711/24420 [7:23:25<2:46:43,  1.49s/it]
 73%|███████▎  | 17712/24420 [7:23:27<2:46:29,  1.49s/it]
 73%|███████▎  | 17713/24420 [7:23:28<2:46:43,  1.49s/it]
 73%|███████▎  | 17714/24420 [7:23:30<2:46:40,  1.49s/it]
 73%|███████▎  | 17715/24420 [7:23:31<2:46:22,  1.49s/it]
                                                         
{'loss': 0.0747, 'learning_rate': 5e-05, 'epoch': 3.63}

 73%|███████▎  | 17715/24420 [7:23:31<2:46:22,  1.49s/it]
 73%|███████▎  | 17716/24420 [7:23:33<2:46:32,  1.49s/it]
 73%|███████▎  | 17717/24420 [7:23:34<2:46:20,  1.49s/it]
 73%|███████▎  | 17718/24420 [7:23:36<2:46:17,  1.49s/it]
 73%|███████▎  | 17719/24420 [7:23:37<2:46:08,  1.49s/it]
 73%|███████▎  | 17720/24420 [7:23:39<2:46:05,  1.49s/it]
                                                         
{'loss': 0.0768, 'learning_rate': 5e-05, 'epoch': 3.63}

 73%|███████▎  | 17720/24420 [7:23:39<2:46:05,  1.49s/it]
 73%|███████▎  | 17721/24420 [7:23:40<2:46:11,  1.49s/it]
 73%|███████▎  | 17722/24420 [7:23:42<2:46:07,  1.49s/it]
 73%|███████▎  | 17723/24420 [7:23:43<2:45:56,  1.49s/it]
 73%|███████▎  | 17724/24420 [7:23:45<2:45:48,  1.49s/it]
 73%|███████▎  | 17725/24420 [7:23:46<2:45:47,  1.49s/it]
                                                         
{'loss': 0.0841, 'learning_rate': 5e-05, 'epoch': 3.63}

 73%|███████▎  | 17725/24420 [7:23:46<2:45:47,  1.49s/it]
 73%|███████▎  | 17726/24420 [7:23:48<2:45:33,  1.48s/it]
 73%|███████▎  | 17727/24420 [7:23:49<2:45:29,  1.48s/it]
 73%|███████▎  | 17728/24420 [7:23:51<2:45:36,  1.48s/it]
 73%|███████▎  | 17729/24420 [7:23:52<2:45:40,  1.49s/it]
 73%|███████▎  | 17730/24420 [7:23:54<2:45:48,  1.49s/it]
                                                         
{'loss': 0.0671, 'learning_rate': 5e-05, 'epoch': 3.63}

 73%|███████▎  | 17730/24420 [7:23:54<2:45:48,  1.49s/it]
 73%|███████▎  | 17731/24420 [7:23:55<2:45:47,  1.49s/it]
 73%|███████▎  | 17732/24420 [7:23:57<2:46:08,  1.49s/it]
 73%|███████▎  | 17733/24420 [7:23:58<2:45:43,  1.49s/it]
 73%|███████▎  | 17734/24420 [7:24:00<2:45:36,  1.49s/it]
 73%|███████▎  | 17735/24420 [7:24:01<2:45:25,  1.48s/it]
                                                         
{'loss': 0.0818, 'learning_rate': 5e-05, 'epoch': 3.63}

 73%|███████▎  | 17735/24420 [7:24:01<2:45:25,  1.48s/it]
 73%|███████▎  | 17736/24420 [7:24:03<2:45:23,  1.48s/it]
 73%|███████▎  | 17737/24420 [7:24:04<2:45:41,  1.49s/it]
 73%|███████▎  | 17738/24420 [7:24:06<2:45:42,  1.49s/it]
 73%|███████▎  | 17739/24420 [7:24:07<2:45:51,  1.49s/it]
 73%|███████▎  | 17740/24420 [7:24:09<2:46:17,  1.49s/it]
                                                         
{'loss': 0.0845, 'learning_rate': 5e-05, 'epoch': 3.63}

 73%|███████▎  | 17740/24420 [7:24:09<2:46:17,  1.49s/it]
 73%|███████▎  | 17741/24420 [7:24:10<2:46:02,  1.49s/it]
 73%|███████▎  | 17742/24420 [7:24:12<2:45:58,  1.49s/it]
 73%|███████▎  | 17743/24420 [7:24:13<2:45:52,  1.49s/it]
 73%|███████▎  | 17744/24420 [7:24:15<2:45:33,  1.49s/it]
 73%|███████▎  | 17745/24420 [7:24:16<2:45:26,  1.49s/it]
                                                         
{'loss': 0.081, 'learning_rate': 5e-05, 'epoch': 3.63}

 73%|███████▎  | 17745/24420 [7:24:16<2:45:26,  1.49s/it]
 73%|███████▎  | 17746/24420 [7:24:17<2:45:33,  1.49s/it]
 73%|███████▎  | 17747/24420 [7:24:19<2:45:19,  1.49s/it]
 73%|███████▎  | 17748/24420 [7:24:20<2:45:39,  1.49s/it]
 73%|███████▎  | 17749/24420 [7:24:22<2:45:34,  1.49s/it]
 73%|███████▎  | 17750/24420 [7:24:23<2:45:14,  1.49s/it]
                                                         
{'loss': 0.0816, 'learning_rate': 5e-05, 'epoch': 3.63}

 73%|███████▎  | 17750/24420 [7:24:23<2:45:14,  1.49s/it]
 73%|███████▎  | 17751/24420 [7:24:25<2:45:19,  1.49s/it]
 73%|███████▎  | 17752/24420 [7:24:27<3:14:47,  1.75s/it]
 73%|███████▎  | 17753/24420 [7:24:29<3:05:40,  1.67s/it]
 73%|███████▎  | 17754/24420 [7:24:30<2:59:19,  1.61s/it]
 73%|███████▎  | 17755/24420 [7:24:32<2:55:16,  1.58s/it]
                                                         
{'loss': 0.0788, 'learning_rate': 5e-05, 'epoch': 3.64}

 73%|███████▎  | 17755/24420 [7:24:32<2:55:16,  1.58s/it]
 73%|███████▎  | 17756/24420 [7:24:33<2:52:40,  1.55s/it]
 73%|███████▎  | 17757/24420 [7:24:35<2:50:15,  1.53s/it]
 73%|███████▎  | 17758/24420 [7:24:36<2:48:56,  1.52s/it]
 73%|███████▎  | 17759/24420 [7:24:38<2:47:44,  1.51s/it]
 73%|███████▎  | 17760/24420 [7:24:39<2:47:15,  1.51s/it]
                                                         
{'loss': 0.0775, 'learning_rate': 5e-05, 'epoch': 3.64}

 73%|███████▎  | 17760/24420 [7:24:39<2:47:15,  1.51s/it]
 73%|███████▎  | 17761/24420 [7:24:41<2:46:35,  1.50s/it]
 73%|███████▎  | 17762/24420 [7:24:42<2:45:41,  1.49s/it]
 73%|███████▎  | 17763/24420 [7:24:44<2:45:19,  1.49s/it]
 73%|███████▎  | 17764/24420 [7:24:45<2:45:05,  1.49s/it]
 73%|███████▎  | 17765/24420 [7:24:47<2:45:12,  1.49s/it]
                                                         
{'loss': 0.0815, 'learning_rate': 5e-05, 'epoch': 3.64}

 73%|███████▎  | 17765/24420 [7:24:47<2:45:12,  1.49s/it]
 73%|███████▎  | 17766/24420 [7:24:48<2:45:05,  1.49s/it]
 73%|███████▎  | 17767/24420 [7:24:50<2:45:14,  1.49s/it]
 73%|███████▎  | 17768/24420 [7:24:51<2:45:32,  1.49s/it]
 73%|███████▎  | 17769/24420 [7:24:53<2:45:25,  1.49s/it]
 73%|███████▎  | 17770/24420 [7:24:54<2:45:03,  1.49s/it]
                                                         
{'loss': 0.0849, 'learning_rate': 5e-05, 'epoch': 3.64}

 73%|███████▎  | 17770/24420 [7:24:54<2:45:03,  1.49s/it]
 73%|███████▎  | 17771/24420 [7:24:56<2:44:57,  1.49s/it]
 73%|███████▎  | 17772/24420 [7:24:57<2:45:05,  1.49s/it]
 73%|███████▎  | 17773/24420 [7:24:59<2:44:59,  1.49s/it]
 73%|███████▎  | 17774/24420 [7:25:00<2:44:52,  1.49s/it]
 73%|███████▎  | 17775/24420 [7:25:02<2:44:41,  1.49s/it]
                                                         
{'loss': 0.0824, 'learning_rate': 5e-05, 'epoch': 3.64}

 73%|███████▎  | 17775/24420 [7:25:02<2:44:41,  1.49s/it]
 73%|███████▎  | 17776/24420 [7:25:03<2:44:43,  1.49s/it]
 73%|███████▎  | 17777/24420 [7:25:04<2:44:39,  1.49s/it]
 73%|███████▎  | 17778/24420 [7:25:06<2:44:29,  1.49s/it]
 73%|███████▎  | 17779/24420 [7:25:07<2:44:40,  1.49s/it]
 73%|███████▎  | 17780/24420 [7:25:09<2:44:44,  1.49s/it]
                                                         
{'loss': 0.0893, 'learning_rate': 5e-05, 'epoch': 3.64}

 73%|███████▎  | 17780/24420 [7:25:09<2:44:44,  1.49s/it]
 73%|███████▎  | 17781/24420 [7:25:10<2:44:32,  1.49s/it]
 73%|███████▎  | 17782/24420 [7:25:12<2:44:24,  1.49s/it]
 73%|███████▎  | 17783/24420 [7:25:13<2:44:33,  1.49s/it]
 73%|███████▎  | 17784/24420 [7:25:15<2:44:21,  1.49s/it]
 73%|███████▎  | 17785/24420 [7:25:16<2:44:32,  1.49s/it]
                                                         
{'loss': 0.0692, 'learning_rate': 5e-05, 'epoch': 3.64}

 73%|███████▎  | 17785/24420 [7:25:16<2:44:32,  1.49s/it]
 73%|███████▎  | 17786/24420 [7:25:18<2:44:32,  1.49s/it]
 73%|███████▎  | 17787/24420 [7:25:19<2:44:52,  1.49s/it]
 73%|███████▎  | 17788/24420 [7:25:21<2:44:58,  1.49s/it]
 73%|███████▎  | 17789/24420 [7:25:22<2:45:04,  1.49s/it]
 73%|███████▎  | 17790/24420 [7:25:24<2:44:44,  1.49s/it]
                                                         
{'loss': 0.0791, 'learning_rate': 5e-05, 'epoch': 3.64}

 73%|███████▎  | 17790/24420 [7:25:24<2:44:44,  1.49s/it]
 73%|███████▎  | 17791/24420 [7:25:25<2:44:24,  1.49s/it]
 73%|███████▎  | 17792/24420 [7:25:27<2:44:04,  1.49s/it]
 73%|███████▎  | 17793/24420 [7:25:28<2:43:57,  1.48s/it]
 73%|███████▎  | 17794/24420 [7:25:30<2:43:41,  1.48s/it]
 73%|███████▎  | 17795/24420 [7:25:31<2:44:01,  1.49s/it]
                                                         
{'loss': 0.0755, 'learning_rate': 5e-05, 'epoch': 3.64}

 73%|███████▎  | 17795/24420 [7:25:31<2:44:01,  1.49s/it]
 73%|███████▎  | 17796/24420 [7:25:33<2:44:08,  1.49s/it]
 73%|███████▎  | 17797/24420 [7:25:34<2:43:57,  1.49s/it]
 73%|███████▎  | 17798/24420 [7:25:36<2:44:02,  1.49s/it]
 73%|███████▎  | 17799/24420 [7:25:37<2:44:09,  1.49s/it]
 73%|███████▎  | 17800/24420 [7:25:39<2:44:18,  1.49s/it]
                                                         
{'loss': 0.0926, 'learning_rate': 5e-05, 'epoch': 3.64}

 73%|███████▎  | 17800/24420 [7:25:39<2:44:18,  1.49s/it]
 73%|███████▎  | 17801/24420 [7:25:43<4:27:06,  2.42s/it]
 73%|███████▎  | 17802/24420 [7:25:45<3:56:00,  2.14s/it]
 73%|███████▎  | 17803/24420 [7:25:46<3:34:30,  1.95s/it]
 73%|███████▎  | 17804/24420 [7:25:48<3:19:11,  1.81s/it]
 73%|███████▎  | 17805/24420 [7:25:49<3:08:36,  1.71s/it]
                                                         
{'loss': 0.0858, 'learning_rate': 5e-05, 'epoch': 3.65}

 73%|███████▎  | 17805/24420 [7:25:49<3:08:36,  1.71s/it]
 73%|███████▎  | 17806/24420 [7:25:51<3:01:09,  1.64s/it]
 73%|███████▎  | 17807/24420 [7:25:52<2:55:46,  1.59s/it]
 73%|███████▎  | 17808/24420 [7:25:54<2:52:03,  1.56s/it]
 73%|███████▎  | 17809/24420 [7:25:55<2:49:36,  1.54s/it]
 73%|███████▎  | 17810/24420 [7:25:57<2:48:15,  1.53s/it]
                                                         
{'loss': 0.0806, 'learning_rate': 5e-05, 'epoch': 3.65}

 73%|███████▎  | 17810/24420 [7:25:57<2:48:15,  1.53s/it]
 73%|███████▎  | 17811/24420 [7:25:58<2:46:54,  1.52s/it]
 73%|███████▎  | 17812/24420 [7:26:00<2:46:11,  1.51s/it]
 73%|███████▎  | 17813/24420 [7:26:01<2:45:32,  1.50s/it]
 73%|███████▎  | 17814/24420 [7:26:03<2:44:40,  1.50s/it]
 73%|███████▎  | 17815/24420 [7:26:04<2:44:12,  1.49s/it]
                                                         
{'loss': 0.0831, 'learning_rate': 5e-05, 'epoch': 3.65}

 73%|███████▎  | 17815/24420 [7:26:04<2:44:12,  1.49s/it]
 73%|███████▎  | 17816/24420 [7:26:06<2:43:52,  1.49s/it]
 73%|███████▎  | 17817/24420 [7:26:07<2:43:47,  1.49s/it]
 73%|███████▎  | 17818/24420 [7:26:09<2:43:35,  1.49s/it]
 73%|███████▎  | 17819/24420 [7:26:10<2:43:32,  1.49s/it]
 73%|███████▎  | 17820/24420 [7:26:12<2:43:20,  1.48s/it]
                                                         
{'loss': 0.0777, 'learning_rate': 5e-05, 'epoch': 3.65}

 73%|███████▎  | 17820/24420 [7:26:12<2:43:20,  1.48s/it]
 73%|███████▎  | 17821/24420 [7:26:13<2:43:15,  1.48s/it]
 73%|███████▎  | 17822/24420 [7:26:15<2:43:27,  1.49s/it]
 73%|███████▎  | 17823/24420 [7:26:16<2:43:24,  1.49s/it]
 73%|███████▎  | 17824/24420 [7:26:17<2:43:05,  1.48s/it]
 73%|███████▎  | 17825/24420 [7:26:19<2:42:57,  1.48s/it]
                                                         
{'loss': 0.0895, 'learning_rate': 5e-05, 'epoch': 3.65}

 73%|███████▎  | 17825/24420 [7:26:19<2:42:57,  1.48s/it]
 73%|███████▎  | 17826/24420 [7:26:20<2:43:05,  1.48s/it]
 73%|███████▎  | 17827/24420 [7:26:22<2:43:30,  1.49s/it]
 73%|███████▎  | 17828/24420 [7:26:23<2:43:28,  1.49s/it]
 73%|███████▎  | 17829/24420 [7:26:25<2:43:12,  1.49s/it]
 73%|███████▎  | 17830/24420 [7:26:26<2:42:50,  1.48s/it]
                                                         
{'loss': 0.0728, 'learning_rate': 5e-05, 'epoch': 3.65}

 73%|███████▎  | 17830/24420 [7:26:26<2:42:50,  1.48s/it]
 73%|███████▎  | 17831/24420 [7:26:28<2:42:53,  1.48s/it]
 73%|███████▎  | 17832/24420 [7:26:29<2:42:44,  1.48s/it]
 73%|███████▎  | 17833/24420 [7:26:31<2:42:38,  1.48s/it]
 73%|███████▎  | 17834/24420 [7:26:32<2:42:50,  1.48s/it]
 73%|███████▎  | 17835/24420 [7:26:34<2:42:49,  1.48s/it]
                                                         
{'loss': 0.0763, 'learning_rate': 5e-05, 'epoch': 3.65}

 73%|███████▎  | 17835/24420 [7:26:34<2:42:49,  1.48s/it]
 73%|███████▎  | 17836/24420 [7:26:35<2:42:45,  1.48s/it]
 73%|███████▎  | 17837/24420 [7:26:37<2:42:29,  1.48s/it]
 73%|███████▎  | 17838/24420 [7:26:38<2:42:38,  1.48s/it]
 73%|███████▎  | 17839/24420 [7:26:40<2:42:25,  1.48s/it]
 73%|███████▎  | 17840/24420 [7:26:41<2:42:24,  1.48s/it]
                                                         
{'loss': 0.0692, 'learning_rate': 5e-05, 'epoch': 3.65}

 73%|███████▎  | 17840/24420 [7:26:41<2:42:24,  1.48s/it]
 73%|███████▎  | 17841/24420 [7:26:43<2:42:30,  1.48s/it]
 73%|███████▎  | 17842/24420 [7:26:44<2:42:36,  1.48s/it]
 73%|███████▎  | 17843/24420 [7:26:46<2:43:03,  1.49s/it]
 73%|███████▎  | 17844/24420 [7:26:47<2:42:41,  1.48s/it]
 73%|███████▎  | 17845/24420 [7:26:49<2:42:41,  1.48s/it]
                                                         
{'loss': 0.0738, 'learning_rate': 5e-05, 'epoch': 3.65}

 73%|███████▎  | 17845/24420 [7:26:49<2:42:41,  1.48s/it]
 73%|███████▎  | 17846/24420 [7:26:50<2:42:41,  1.48s/it]
 73%|███████▎  | 17847/24420 [7:26:52<2:42:27,  1.48s/it]
 73%|███████▎  | 17848/24420 [7:26:53<2:42:22,  1.48s/it]
 73%|███████▎  | 17849/24420 [7:26:55<2:42:23,  1.48s/it]
 73%|███████▎  | 17850/24420 [7:26:56<2:42:59,  1.49s/it]
                                                         
{'loss': 0.079, 'learning_rate': 5e-05, 'epoch': 3.65}

 73%|███████▎  | 17850/24420 [7:26:56<2:42:59,  1.49s/it]
 73%|███████▎  | 17851/24420 [7:26:58<2:43:16,  1.49s/it]
 73%|███████▎  | 17852/24420 [7:26:59<2:43:16,  1.49s/it]
 73%|███████▎  | 17853/24420 [7:27:01<2:43:17,  1.49s/it]
 73%|███████▎  | 17854/24420 [7:27:02<2:43:08,  1.49s/it]
 73%|███████▎  | 17855/24420 [7:27:04<2:43:19,  1.49s/it]
                                                         
{'loss': 0.0896, 'learning_rate': 5e-05, 'epoch': 3.66}

 73%|███████▎  | 17855/24420 [7:27:04<2:43:19,  1.49s/it]
 73%|███████▎  | 17856/24420 [7:27:05<2:43:15,  1.49s/it]
 73%|███████▎  | 17857/24420 [7:27:07<2:42:45,  1.49s/it]
 73%|███████▎  | 17858/24420 [7:27:08<2:42:27,  1.49s/it]
 73%|███████▎  | 17859/24420 [7:27:09<2:42:16,  1.48s/it]
 73%|███████▎  | 17860/24420 [7:27:11<2:42:01,  1.48s/it]
                                                         
{'loss': 0.0781, 'learning_rate': 5e-05, 'epoch': 3.66}

 73%|███████▎  | 17860/24420 [7:27:11<2:42:01,  1.48s/it]
 73%|███████▎  | 17861/24420 [7:27:12<2:42:03,  1.48s/it]
 73%|███████▎  | 17862/24420 [7:27:14<2:42:15,  1.48s/it]
 73%|███████▎  | 17863/24420 [7:27:15<2:42:20,  1.49s/it]
 73%|███████▎  | 17864/24420 [7:27:17<2:42:09,  1.48s/it]
 73%|███████▎  | 17865/24420 [7:27:18<2:42:00,  1.48s/it]
                                                         
{'loss': 0.0835, 'learning_rate': 5e-05, 'epoch': 3.66}

 73%|███████▎  | 17865/24420 [7:27:18<2:42:00,  1.48s/it]
 73%|███████▎  | 17866/24420 [7:27:20<2:41:56,  1.48s/it]
 73%|███████▎  | 17867/24420 [7:27:21<2:42:03,  1.48s/it]
 73%|███████▎  | 17868/24420 [7:27:23<2:42:00,  1.48s/it]
 73%|███████▎  | 17869/24420 [7:27:24<2:42:00,  1.48s/it]
 73%|███████▎  | 17870/24420 [7:27:26<2:41:53,  1.48s/it]
                                                         
{'loss': 0.0747, 'learning_rate': 5e-05, 'epoch': 3.66}

 73%|███████▎  | 17870/24420 [7:27:26<2:41:53,  1.48s/it]
 73%|███████▎  | 17871/24420 [7:27:27<2:41:45,  1.48s/it]
 73%|███████▎  | 17872/24420 [7:27:29<2:41:47,  1.48s/it]
 73%|███████▎  | 17873/24420 [7:27:30<2:41:54,  1.48s/it]
 73%|███████▎  | 17874/24420 [7:27:32<2:42:18,  1.49s/it]
 73%|███████▎  | 17875/24420 [7:27:33<2:42:15,  1.49s/it]
                                                         
{'loss': 0.0844, 'learning_rate': 5e-05, 'epoch': 3.66}

 73%|███████▎  | 17875/24420 [7:27:33<2:42:15,  1.49s/it]
 73%|███████▎  | 17876/24420 [7:27:35<2:42:23,  1.49s/it]
 73%|███████▎  | 17877/24420 [7:27:36<2:42:11,  1.49s/it]
 73%|███████▎  | 17878/24420 [7:27:38<2:42:03,  1.49s/it]
 73%|███████▎  | 17879/24420 [7:27:39<2:41:56,  1.49s/it]
 73%|███████▎  | 17880/24420 [7:27:41<2:42:03,  1.49s/it]
                                                         
{'loss': 0.0612, 'learning_rate': 5e-05, 'epoch': 3.66}

 73%|███████▎  | 17880/24420 [7:27:41<2:42:03,  1.49s/it]
 73%|███████▎  | 17881/24420 [7:27:42<2:42:23,  1.49s/it]
 73%|███████▎  | 17882/24420 [7:27:44<2:42:21,  1.49s/it]
 73%|███████▎  | 17883/24420 [7:27:45<2:43:32,  1.50s/it]
 73%|███████▎  | 17884/24420 [7:27:47<2:43:12,  1.50s/it]
 73%|███████▎  | 17885/24420 [7:27:48<2:42:53,  1.50s/it]
                                                         
{'loss': 0.0811, 'learning_rate': 5e-05, 'epoch': 3.66}

 73%|███████▎  | 17885/24420 [7:27:48<2:42:53,  1.50s/it]
 73%|███████▎  | 17886/24420 [7:27:50<2:42:25,  1.49s/it]
 73%|███████▎  | 17887/24420 [7:27:51<2:42:29,  1.49s/it]
 73%|███████▎  | 17888/24420 [7:27:53<2:42:10,  1.49s/it]
 73%|███████▎  | 17889/24420 [7:27:54<2:42:12,  1.49s/it]
 73%|███████▎  | 17890/24420 [7:27:56<2:42:06,  1.49s/it]
                                                         
{'loss': 0.0867, 'learning_rate': 5e-05, 'epoch': 3.66}

 73%|███████▎  | 17890/24420 [7:27:56<2:42:06,  1.49s/it]
 73%|███████▎  | 17891/24420 [7:27:57<2:41:40,  1.49s/it]
 73%|███████▎  | 17892/24420 [7:27:59<2:41:59,  1.49s/it]
 73%|███████▎  | 17893/24420 [7:28:00<2:41:36,  1.49s/it]
 73%|███████▎  | 17894/24420 [7:28:02<2:41:42,  1.49s/it]
 73%|███████▎  | 17895/24420 [7:28:03<2:41:46,  1.49s/it]
                                                         
{'loss': 0.0753, 'learning_rate': 5e-05, 'epoch': 3.66}

 73%|███████▎  | 17895/24420 [7:28:03<2:41:46,  1.49s/it]
 73%|███████▎  | 17896/24420 [7:28:05<2:41:49,  1.49s/it]
 73%|███████▎  | 17897/24420 [7:28:06<2:41:35,  1.49s/it]
 73%|███████▎  | 17898/24420 [7:28:07<2:41:35,  1.49s/it]
 73%|███████▎  | 17899/24420 [7:28:09<2:41:30,  1.49s/it]
 73%|███████▎  | 17900/24420 [7:28:10<2:41:17,  1.48s/it]
                                                         
{'loss': 0.0737, 'learning_rate': 5e-05, 'epoch': 3.67}

 73%|███████▎  | 17900/24420 [7:28:10<2:41:17,  1.48s/it]
 73%|███████▎  | 17901/24420 [7:28:12<2:41:12,  1.48s/it]
 73%|███████▎  | 17902/24420 [7:28:13<2:41:14,  1.48s/it]
 73%|███████▎  | 17903/24420 [7:28:15<2:41:26,  1.49s/it]
 73%|███████▎  | 17904/24420 [7:28:16<2:41:35,  1.49s/it]
 73%|███████▎  | 17905/24420 [7:28:18<2:41:29,  1.49s/it]
                                                         
{'loss': 0.0781, 'learning_rate': 5e-05, 'epoch': 3.67}

 73%|███████▎  | 17905/24420 [7:28:18<2:41:29,  1.49s/it]
 73%|███████▎  | 17906/24420 [7:28:19<2:41:25,  1.49s/it]
 73%|███████▎  | 17907/24420 [7:28:21<2:41:13,  1.49s/it]
 73%|███████▎  | 17908/24420 [7:28:22<2:41:12,  1.49s/it]
 73%|███████▎  | 17909/24420 [7:28:24<2:41:28,  1.49s/it]
 73%|███████▎  | 17910/24420 [7:28:25<2:41:20,  1.49s/it]
                                                         
{'loss': 0.081, 'learning_rate': 5e-05, 'epoch': 3.67}

 73%|███████▎  | 17910/24420 [7:28:25<2:41:20,  1.49s/it]
 73%|███████▎  | 17911/24420 [7:28:27<2:41:24,  1.49s/it]
 73%|███████▎  | 17912/24420 [7:28:28<2:41:06,  1.49s/it]
 73%|███████▎  | 17913/24420 [7:28:30<2:40:59,  1.48s/it]
 73%|███████▎  | 17914/24420 [7:28:31<2:40:55,  1.48s/it]
 73%|███████▎  | 17915/24420 [7:28:33<2:41:07,  1.49s/it]
                                                         
{'loss': 0.076, 'learning_rate': 5e-05, 'epoch': 3.67}

 73%|███████▎  | 17915/24420 [7:28:33<2:41:07,  1.49s/it]
 73%|███████▎  | 17916/24420 [7:28:34<2:41:15,  1.49s/it]
 73%|███████▎  | 17917/24420 [7:28:36<2:41:15,  1.49s/it]
 73%|███████▎  | 17918/24420 [7:28:37<2:41:03,  1.49s/it]
 73%|███████▎  | 17919/24420 [7:28:39<2:41:11,  1.49s/it]
 73%|███████▎  | 17920/24420 [7:28:40<2:41:05,  1.49s/it]
                                                         
{'loss': 0.0825, 'learning_rate': 5e-05, 'epoch': 3.67}

 73%|███████▎  | 17920/24420 [7:28:40<2:41:05,  1.49s/it]
 73%|███████▎  | 17921/24420 [7:28:42<2:41:17,  1.49s/it]
 73%|███████▎  | 17922/24420 [7:28:43<2:41:14,  1.49s/it]
 73%|███████▎  | 17923/24420 [7:28:45<2:40:50,  1.49s/it]
 73%|███████▎  | 17924/24420 [7:28:46<2:41:00,  1.49s/it]
 73%|███████▎  | 17925/24420 [7:28:48<2:41:08,  1.49s/it]
                                                         
{'loss': 0.0829, 'learning_rate': 5e-05, 'epoch': 3.67}

 73%|███████▎  | 17925/24420 [7:28:48<2:41:08,  1.49s/it]
 73%|███████▎  | 17926/24420 [7:28:49<2:41:09,  1.49s/it]
 73%|███████▎  | 17927/24420 [7:28:51<2:42:01,  1.50s/it]
 73%|███████▎  | 17928/24420 [7:28:52<2:41:20,  1.49s/it]
 73%|███████▎  | 17929/24420 [7:28:54<2:41:16,  1.49s/it]
 73%|███████▎  | 17930/24420 [7:28:55<2:41:03,  1.49s/it]
                                                         
{'loss': 0.0769, 'learning_rate': 5e-05, 'epoch': 3.67}

 73%|███████▎  | 17930/24420 [7:28:55<2:41:03,  1.49s/it]
 73%|███████▎  | 17931/24420 [7:28:57<2:41:05,  1.49s/it]
 73%|███████▎  | 17932/24420 [7:28:58<2:41:01,  1.49s/it]
 73%|███████▎  | 17933/24420 [7:29:00<2:41:08,  1.49s/it]
 73%|███████▎  | 17934/24420 [7:29:01<2:40:52,  1.49s/it]
 73%|███████▎  | 17935/24420 [7:29:03<2:40:42,  1.49s/it]
                                                         
{'loss': 0.0764, 'learning_rate': 5e-05, 'epoch': 3.67}

 73%|███████▎  | 17935/24420 [7:29:03<2:40:42,  1.49s/it]
 73%|███████▎  | 17936/24420 [7:29:04<2:40:50,  1.49s/it]
 73%|███████▎  | 17937/24420 [7:29:05<2:40:47,  1.49s/it]
 73%|███████▎  | 17938/24420 [7:29:07<2:40:33,  1.49s/it]
 73%|███████▎  | 17939/24420 [7:29:08<2:41:07,  1.49s/it]
 73%|███████▎  | 17940/24420 [7:29:10<2:41:03,  1.49s/it]
                                                         
{'loss': 0.0693, 'learning_rate': 5e-05, 'epoch': 3.67}

 73%|███████▎  | 17940/24420 [7:29:10<2:41:03,  1.49s/it]
 73%|███████▎  | 17941/24420 [7:29:11<2:41:10,  1.49s/it]
 73%|███████▎  | 17942/24420 [7:29:13<2:41:07,  1.49s/it]
 73%|███████▎  | 17943/24420 [7:29:14<2:40:42,  1.49s/it]
 73%|███████▎  | 17944/24420 [7:29:16<2:40:33,  1.49s/it]
 73%|███████▎  | 17945/24420 [7:29:17<2:40:28,  1.49s/it]
                                                         
{'loss': 0.0793, 'learning_rate': 5e-05, 'epoch': 3.67}

 73%|███████▎  | 17945/24420 [7:29:17<2:40:28,  1.49s/it]
 73%|███████▎  | 17946/24420 [7:29:19<2:40:30,  1.49s/it]
 73%|███████▎  | 17947/24420 [7:29:20<2:40:37,  1.49s/it]
 73%|███████▎  | 17948/24420 [7:29:22<2:40:18,  1.49s/it]
 74%|███████▎  | 17949/24420 [7:29:23<2:40:38,  1.49s/it]
 74%|███████▎  | 17950/24420 [7:29:25<2:40:51,  1.49s/it]
                                                         
{'loss': 0.0794, 'learning_rate': 5e-05, 'epoch': 3.68}

 74%|███████▎  | 17950/24420 [7:29:25<2:40:51,  1.49s/it]
 74%|███████▎  | 17951/24420 [7:29:26<2:40:50,  1.49s/it]
 74%|███████▎  | 17952/24420 [7:29:28<2:40:34,  1.49s/it]
 74%|███████▎  | 17953/24420 [7:29:29<2:40:13,  1.49s/it]
 74%|███████▎  | 17954/24420 [7:29:31<2:40:16,  1.49s/it]
 74%|███████▎  | 17955/24420 [7:29:32<2:40:19,  1.49s/it]
                                                         
{'loss': 0.0798, 'learning_rate': 5e-05, 'epoch': 3.68}

 74%|███████▎  | 17955/24420 [7:29:32<2:40:19,  1.49s/it]
 74%|███████▎  | 17956/24420 [7:29:34<2:40:24,  1.49s/it]
 74%|███████▎  | 17957/24420 [7:29:35<2:40:23,  1.49s/it]
 74%|███████▎  | 17958/24420 [7:29:37<2:40:21,  1.49s/it]
 74%|███████▎  | 17959/24420 [7:29:38<2:40:12,  1.49s/it]
 74%|███████▎  | 17960/24420 [7:29:40<2:40:04,  1.49s/it]
                                                         
{'loss': 0.0858, 'learning_rate': 5e-05, 'epoch': 3.68}

 74%|███████▎  | 17960/24420 [7:29:40<2:40:04,  1.49s/it]
 74%|███████▎  | 17961/24420 [7:29:41<2:40:09,  1.49s/it]
 74%|███████▎  | 17962/24420 [7:29:43<2:39:50,  1.49s/it]
 74%|███████▎  | 17963/24420 [7:29:44<2:40:09,  1.49s/it]
 74%|███████▎  | 17964/24420 [7:29:46<2:39:52,  1.49s/it]
 74%|███████▎  | 17965/24420 [7:29:47<2:39:56,  1.49s/it]
                                                         
{'loss': 0.0784, 'learning_rate': 5e-05, 'epoch': 3.68}

 74%|███████▎  | 17965/24420 [7:29:47<2:39:56,  1.49s/it]
 74%|███████▎  | 17966/24420 [7:29:49<2:40:03,  1.49s/it]
 74%|███████▎  | 17967/24420 [7:29:50<2:39:49,  1.49s/it]
 74%|███████▎  | 17968/24420 [7:29:52<2:39:35,  1.48s/it]
 74%|███████▎  | 17969/24420 [7:29:53<2:39:28,  1.48s/it]
 74%|███████▎  | 17970/24420 [7:29:55<2:39:30,  1.48s/it]
                                                         
{'loss': 0.0721, 'learning_rate': 5e-05, 'epoch': 3.68}

 74%|███████▎  | 17970/24420 [7:29:55<2:39:30,  1.48s/it]
 74%|███████▎  | 17971/24420 [7:29:56<2:39:33,  1.48s/it]
 74%|███████▎  | 17972/24420 [7:29:58<2:39:34,  1.48s/it]
 74%|███████▎  | 17973/24420 [7:29:59<2:39:23,  1.48s/it]
 74%|███████▎  | 17974/24420 [7:30:01<2:39:38,  1.49s/it]
 74%|███████▎  | 17975/24420 [7:30:02<2:39:35,  1.49s/it]
                                                         
{'loss': 0.0724, 'learning_rate': 5e-05, 'epoch': 3.68}

 74%|███████▎  | 17975/24420 [7:30:02<2:39:35,  1.49s/it]
 74%|███████▎  | 17976/24420 [7:30:04<2:39:42,  1.49s/it]
 74%|███████▎  | 17977/24420 [7:30:05<2:39:40,  1.49s/it]
 74%|███████▎  | 17978/24420 [7:30:06<2:39:22,  1.48s/it]
 74%|███████▎  | 17979/24420 [7:30:08<2:39:31,  1.49s/it]
 74%|███████▎  | 17980/24420 [7:30:09<2:39:33,  1.49s/it]
                                                         
{'loss': 0.0609, 'learning_rate': 5e-05, 'epoch': 3.68}

 74%|███████▎  | 17980/24420 [7:30:09<2:39:33,  1.49s/it]
 74%|███████▎  | 17981/24420 [7:30:11<2:39:26,  1.49s/it]
 74%|███████▎  | 17982/24420 [7:30:12<2:39:19,  1.48s/it]
 74%|███████▎  | 17983/24420 [7:30:14<2:39:17,  1.48s/it]
 74%|███████▎  | 17984/24420 [7:30:15<2:39:16,  1.48s/it]
 74%|███████▎  | 17985/24420 [7:30:17<2:39:04,  1.48s/it]
                                                         
{'loss': 0.0669, 'learning_rate': 5e-05, 'epoch': 3.68}

 74%|███████▎  | 17985/24420 [7:30:17<2:39:04,  1.48s/it]
 74%|███████▎  | 17986/24420 [7:30:18<2:39:35,  1.49s/it]
 74%|███████▎  | 17987/24420 [7:30:20<2:39:41,  1.49s/it]
 74%|███████▎  | 17988/24420 [7:30:21<2:39:35,  1.49s/it]
 74%|███████▎  | 17989/24420 [7:30:23<2:39:54,  1.49s/it]
 74%|███████▎  | 17990/24420 [7:30:24<2:39:40,  1.49s/it]
                                                         
{'loss': 0.0801, 'learning_rate': 5e-05, 'epoch': 3.68}

 74%|███████▎  | 17990/24420 [7:30:24<2:39:40,  1.49s/it]
 74%|███████▎  | 17991/24420 [7:30:26<2:39:44,  1.49s/it]
 74%|███████▎  | 17992/24420 [7:30:27<2:39:32,  1.49s/it]
 74%|███████▎  | 17993/24420 [7:30:29<2:39:05,  1.49s/it]
 74%|███████▎  | 17994/24420 [7:30:30<2:38:55,  1.48s/it]
 74%|███████▎  | 17995/24420 [7:30:32<2:38:57,  1.48s/it]
                                                         
{'loss': 0.0799, 'learning_rate': 5e-05, 'epoch': 3.68}

 74%|███████▎  | 17995/24420 [7:30:32<2:38:57,  1.48s/it]
 74%|███████▎  | 17996/24420 [7:30:33<2:39:17,  1.49s/it]
 74%|███████▎  | 17997/24420 [7:30:35<2:39:10,  1.49s/it]
 74%|███████▎  | 17998/24420 [7:30:36<2:39:07,  1.49s/it]
 74%|███████▎  | 17999/24420 [7:30:38<2:39:24,  1.49s/it]
 74%|███████▎  | 18000/24420 [7:30:39<2:39:32,  1.49s/it]
                                                         
{'loss': 0.0827, 'learning_rate': 5e-05, 'epoch': 3.69}

 74%|███████▎  | 18000/24420 [7:30:39<2:39:32,  1.49s/it]
 74%|███████▎  | 18001/24420 [7:30:44<4:22:01,  2.45s/it]
 74%|███████▎  | 18002/24420 [7:30:45<3:51:07,  2.16s/it]
 74%|███████▎  | 18003/24420 [7:30:47<3:29:29,  1.96s/it]
 74%|███████▎  | 18004/24420 [7:30:48<3:14:26,  1.82s/it]
 74%|███████▎  | 18005/24420 [7:30:50<3:03:47,  1.72s/it]
                                                         
{'loss': 0.0766, 'learning_rate': 5e-05, 'epoch': 3.69}

 74%|███████▎  | 18005/24420 [7:30:50<3:03:47,  1.72s/it]
 74%|███████▎  | 18006/24420 [7:30:51<2:56:16,  1.65s/it]
 74%|███████▎  | 18007/24420 [7:30:53<2:51:09,  1.60s/it]
 74%|███████▎  | 18008/24420 [7:30:54<2:47:43,  1.57s/it]
 74%|███████▎  | 18009/24420 [7:30:56<2:44:59,  1.54s/it]
 74%|███████▍  | 18010/24420 [7:30:57<2:43:01,  1.53s/it]
                                                         
{'loss': 0.0957, 'learning_rate': 5e-05, 'epoch': 3.69}

 74%|███████▍  | 18010/24420 [7:30:57<2:43:01,  1.53s/it]
 74%|███████▍  | 18011/24420 [7:30:59<2:41:35,  1.51s/it]
 74%|███████▍  | 18012/24420 [7:31:00<2:40:42,  1.50s/it]
 74%|███████▍  | 18013/24420 [7:31:02<2:40:26,  1.50s/it]
 74%|███████▍  | 18014/24420 [7:31:03<2:39:45,  1.50s/it]
 74%|███████▍  | 18015/24420 [7:31:05<2:39:20,  1.49s/it]
                                                         
{'loss': 0.1024, 'learning_rate': 5e-05, 'epoch': 3.69}

 74%|███████▍  | 18015/24420 [7:31:05<2:39:20,  1.49s/it]
 74%|███████▍  | 18016/24420 [7:31:06<2:39:07,  1.49s/it]
 74%|███████▍  | 18017/24420 [7:31:08<2:39:06,  1.49s/it]
 74%|███████▍  | 18018/24420 [7:31:09<2:38:53,  1.49s/it]
 74%|███████▍  | 18019/24420 [7:31:11<2:38:42,  1.49s/it]
 74%|███████▍  | 18020/24420 [7:31:12<2:38:31,  1.49s/it]
                                                         
{'loss': 0.074, 'learning_rate': 5e-05, 'epoch': 3.69}

 74%|███████▍  | 18020/24420 [7:31:12<2:38:31,  1.49s/it]
 74%|███████▍  | 18021/24420 [7:31:14<2:38:41,  1.49s/it]
 74%|███████▍  | 18022/24420 [7:31:15<2:38:32,  1.49s/it]
 74%|███████▍  | 18023/24420 [7:31:17<2:38:29,  1.49s/it]
 74%|███████▍  | 18024/24420 [7:31:18<2:38:23,  1.49s/it]
 74%|███████▍  | 18025/24420 [7:31:20<2:38:11,  1.48s/it]
                                                         
{'loss': 0.087, 'learning_rate': 5e-05, 'epoch': 3.69}

 74%|███████▍  | 18025/24420 [7:31:20<2:38:11,  1.48s/it]
 74%|███████▍  | 18026/24420 [7:31:21<2:38:26,  1.49s/it]
 74%|███████▍  | 18027/24420 [7:31:23<2:38:22,  1.49s/it]
 74%|███████▍  | 18028/24420 [7:31:24<2:38:29,  1.49s/it]
 74%|███████▍  | 18029/24420 [7:31:26<2:38:26,  1.49s/it]
 74%|███████▍  | 18030/24420 [7:31:27<2:38:08,  1.48s/it]
                                                         
{'loss': 0.0767, 'learning_rate': 5e-05, 'epoch': 3.69}

 74%|███████▍  | 18030/24420 [7:31:27<2:38:08,  1.48s/it]
 74%|███████▍  | 18031/24420 [7:31:28<2:38:08,  1.49s/it]
 74%|███████▍  | 18032/24420 [7:31:30<2:38:09,  1.49s/it]
 74%|███████▍  | 18033/24420 [7:31:31<2:38:25,  1.49s/it]
 74%|███████▍  | 18034/24420 [7:31:33<2:38:18,  1.49s/it]
 74%|███████▍  | 18035/24420 [7:31:34<2:38:08,  1.49s/it]
                                                         
{'loss': 0.0754, 'learning_rate': 5e-05, 'epoch': 3.69}

 74%|███████▍  | 18035/24420 [7:31:34<2:38:08,  1.49s/it]
 74%|███████▍  | 18036/24420 [7:31:36<2:38:12,  1.49s/it]
 74%|███████▍  | 18037/24420 [7:31:37<2:38:21,  1.49s/it]
 74%|███████▍  | 18038/24420 [7:31:39<2:38:02,  1.49s/it]
 74%|███████▍  | 18039/24420 [7:31:40<2:37:43,  1.48s/it]
 74%|███████▍  | 18040/24420 [7:31:42<2:37:45,  1.48s/it]
                                                         
{'loss': 0.0745, 'learning_rate': 5e-05, 'epoch': 3.69}

 74%|███████▍  | 18040/24420 [7:31:42<2:37:45,  1.48s/it]
 74%|███████▍  | 18041/24420 [7:31:43<2:37:57,  1.49s/it]
 74%|███████▍  | 18042/24420 [7:31:45<2:38:20,  1.49s/it]
 74%|███████▍  | 18043/24420 [7:31:46<2:38:05,  1.49s/it]
 74%|███████▍  | 18044/24420 [7:31:48<2:38:04,  1.49s/it]
 74%|███████▍  | 18045/24420 [7:31:49<2:37:55,  1.49s/it]
                                                         
{'loss': 0.0798, 'learning_rate': 5e-05, 'epoch': 3.69}

 74%|███████▍  | 18045/24420 [7:31:49<2:37:55,  1.49s/it]
 74%|███████▍  | 18046/24420 [7:31:51<2:37:54,  1.49s/it]
 74%|███████▍  | 18047/24420 [7:31:52<2:38:04,  1.49s/it]
 74%|███████▍  | 18048/24420 [7:31:54<2:37:45,  1.49s/it]
 74%|███████▍  | 18049/24420 [7:31:55<2:37:45,  1.49s/it]
 74%|███████▍  | 18050/24420 [7:31:57<2:37:48,  1.49s/it]
                                                         
{'loss': 0.0882, 'learning_rate': 5e-05, 'epoch': 3.7}

 74%|███████▍  | 18050/24420 [7:31:57<2:37:48,  1.49s/it]
 74%|███████▍  | 18051/24420 [7:31:58<2:38:05,  1.49s/it]
 74%|███████▍  | 18052/24420 [7:32:00<2:38:00,  1.49s/it]
 74%|███████▍  | 18053/24420 [7:32:01<2:37:48,  1.49s/it]
 74%|███████▍  | 18054/24420 [7:32:03<2:37:44,  1.49s/it]
 74%|███████▍  | 18055/24420 [7:32:04<2:37:35,  1.49s/it]
                                                         
{'loss': 0.0789, 'learning_rate': 5e-05, 'epoch': 3.7}

 74%|███████▍  | 18055/24420 [7:32:04<2:37:35,  1.49s/it]
 74%|███████▍  | 18056/24420 [7:32:06<2:37:31,  1.49s/it]
 74%|███████▍  | 18057/24420 [7:32:07<2:38:16,  1.49s/it]
 74%|███████▍  | 18058/24420 [7:32:09<2:37:52,  1.49s/it]
 74%|███████▍  | 18059/24420 [7:32:10<2:37:50,  1.49s/it]
 74%|███████▍  | 18060/24420 [7:32:12<2:38:37,  1.50s/it]
                                                         
{'loss': 0.0727, 'learning_rate': 5e-05, 'epoch': 3.7}

 74%|███████▍  | 18060/24420 [7:32:12<2:38:37,  1.50s/it]
 74%|███████▍  | 18061/24420 [7:32:13<2:38:17,  1.49s/it]
 74%|███████▍  | 18062/24420 [7:32:15<2:37:58,  1.49s/it]
 74%|███████▍  | 18063/24420 [7:32:16<2:37:36,  1.49s/it]
 74%|███████▍  | 18064/24420 [7:32:18<2:37:45,  1.49s/it]
 74%|███████▍  | 18065/24420 [7:32:19<2:37:19,  1.49s/it]
                                                         
{'loss': 0.078, 'learning_rate': 5e-05, 'epoch': 3.7}

 74%|███████▍  | 18065/24420 [7:32:19<2:37:19,  1.49s/it]
 74%|███████▍  | 18066/24420 [7:32:21<2:37:10,  1.48s/it]
 74%|███████▍  | 18067/24420 [7:32:22<2:37:16,  1.49s/it]
 74%|███████▍  | 18068/24420 [7:32:24<2:37:27,  1.49s/it]
 74%|███████▍  | 18069/24420 [7:32:25<2:37:21,  1.49s/it]
 74%|███████▍  | 18070/24420 [7:32:27<2:37:30,  1.49s/it]
                                                         
{'loss': 0.0743, 'learning_rate': 5e-05, 'epoch': 3.7}

 74%|███████▍  | 18070/24420 [7:32:27<2:37:30,  1.49s/it]
 74%|███████▍  | 18071/24420 [7:32:28<2:37:40,  1.49s/it]
 74%|███████▍  | 18072/24420 [7:32:29<2:37:41,  1.49s/it]
 74%|███████▍  | 18073/24420 [7:32:31<2:37:42,  1.49s/it]
 74%|███████▍  | 18074/24420 [7:32:32<2:37:34,  1.49s/it]
 74%|███████▍  | 18075/24420 [7:32:34<2:37:29,  1.49s/it]
                                                         
{'loss': 0.0682, 'learning_rate': 5e-05, 'epoch': 3.7}

 74%|███████▍  | 18075/24420 [7:32:34<2:37:29,  1.49s/it]
 74%|███████▍  | 18076/24420 [7:32:35<2:37:13,  1.49s/it]
 74%|███████▍  | 18077/24420 [7:32:37<2:36:46,  1.48s/it]
 74%|███████▍  | 18078/24420 [7:32:38<2:36:48,  1.48s/it]
 74%|███████▍  | 18079/24420 [7:32:40<2:36:43,  1.48s/it]
 74%|███████▍  | 18080/24420 [7:32:41<2:36:31,  1.48s/it]
                                                         
{'loss': 0.072, 'learning_rate': 5e-05, 'epoch': 3.7}

 74%|███████▍  | 18080/24420 [7:32:41<2:36:31,  1.48s/it]
 74%|███████▍  | 18081/24420 [7:32:43<2:36:39,  1.48s/it]
 74%|███████▍  | 18082/24420 [7:32:44<2:36:47,  1.48s/it]
 74%|███████▍  | 18083/24420 [7:32:46<2:37:07,  1.49s/it]
 74%|███████▍  | 18084/24420 [7:32:47<2:36:55,  1.49s/it]
 74%|███████▍  | 18085/24420 [7:32:49<2:36:54,  1.49s/it]
                                                         
{'loss': 0.0798, 'learning_rate': 5e-05, 'epoch': 3.7}

 74%|███████▍  | 18085/24420 [7:32:49<2:36:54,  1.49s/it]
 74%|███████▍  | 18086/24420 [7:32:50<2:37:09,  1.49s/it]
 74%|███████▍  | 18087/24420 [7:32:52<2:37:18,  1.49s/it]
 74%|███████▍  | 18088/24420 [7:32:53<2:37:00,  1.49s/it]
 74%|███████▍  | 18089/24420 [7:32:55<2:37:07,  1.49s/it]
 74%|███████▍  | 18090/24420 [7:32:56<2:37:15,  1.49s/it]
                                                         
{'loss': 0.1017, 'learning_rate': 5e-05, 'epoch': 3.7}

 74%|███████▍  | 18090/24420 [7:32:56<2:37:15,  1.49s/it]
 74%|███████▍  | 18091/24420 [7:32:58<2:37:03,  1.49s/it]
 74%|███████▍  | 18092/24420 [7:32:59<2:36:54,  1.49s/it]
 74%|███████▍  | 18093/24420 [7:33:01<2:36:43,  1.49s/it]
 74%|███████▍  | 18094/24420 [7:33:02<2:36:59,  1.49s/it]
 74%|███████▍  | 18095/24420 [7:33:04<2:36:47,  1.49s/it]
                                                         
{'loss': 0.0976, 'learning_rate': 5e-05, 'epoch': 3.7}

 74%|███████▍  | 18095/24420 [7:33:04<2:36:47,  1.49s/it]
 74%|███████▍  | 18096/24420 [7:33:05<2:36:45,  1.49s/it]
 74%|███████▍  | 18097/24420 [7:33:07<2:36:41,  1.49s/it]
 74%|███████▍  | 18098/24420 [7:33:08<2:36:35,  1.49s/it]
 74%|███████▍  | 18099/24420 [7:33:10<2:36:30,  1.49s/it]
 74%|███████▍  | 18100/24420 [7:33:11<2:36:17,  1.48s/it]
                                                         
{'loss': 0.0752, 'learning_rate': 5e-05, 'epoch': 3.71}

 74%|███████▍  | 18100/24420 [7:33:11<2:36:17,  1.48s/it]
 74%|███████▍  | 18101/24420 [7:33:13<2:36:39,  1.49s/it]
 74%|███████▍  | 18102/24420 [7:33:14<2:36:38,  1.49s/it]
 74%|███████▍  | 18103/24420 [7:33:16<2:36:56,  1.49s/it]
 74%|███████▍  | 18104/24420 [7:33:17<2:36:45,  1.49s/it]
 74%|███████▍  | 18105/24420 [7:33:19<2:36:29,  1.49s/it]
                                                         
{'loss': 0.0856, 'learning_rate': 5e-05, 'epoch': 3.71}

 74%|███████▍  | 18105/24420 [7:33:19<2:36:29,  1.49s/it]
 74%|███████▍  | 18106/24420 [7:33:20<2:36:33,  1.49s/it]
 74%|███████▍  | 18107/24420 [7:33:22<2:36:20,  1.49s/it]
 74%|███████▍  | 18108/24420 [7:33:23<2:36:12,  1.48s/it]
 74%|███████▍  | 18109/24420 [7:33:25<2:36:36,  1.49s/it]
 74%|███████▍  | 18110/24420 [7:33:26<2:36:22,  1.49s/it]
                                                         
{'loss': 0.0707, 'learning_rate': 5e-05, 'epoch': 3.71}

 74%|███████▍  | 18110/24420 [7:33:26<2:36:22,  1.49s/it]
 74%|███████▍  | 18111/24420 [7:33:27<2:36:13,  1.49s/it]
 74%|███████▍  | 18112/24420 [7:33:29<2:36:09,  1.49s/it]
 74%|███████▍  | 18113/24420 [7:33:30<2:36:13,  1.49s/it]
 74%|███████▍  | 18114/24420 [7:33:32<2:36:05,  1.49s/it]
 74%|███████▍  | 18115/24420 [7:33:33<2:36:15,  1.49s/it]
                                                         
{'loss': 0.0821, 'learning_rate': 5e-05, 'epoch': 3.71}

 74%|███████▍  | 18115/24420 [7:33:33<2:36:15,  1.49s/it]
 74%|███████▍  | 18116/24420 [7:33:35<2:36:12,  1.49s/it]
 74%|███████▍  | 18117/24420 [7:33:36<2:36:06,  1.49s/it]
 74%|███████▍  | 18118/24420 [7:33:38<2:36:16,  1.49s/it]
 74%|███████▍  | 18119/24420 [7:33:39<2:36:22,  1.49s/it]
 74%|███████▍  | 18120/24420 [7:33:41<2:36:03,  1.49s/it]
                                                         
{'loss': 0.0731, 'learning_rate': 5e-05, 'epoch': 3.71}

 74%|███████▍  | 18120/24420 [7:33:41<2:36:03,  1.49s/it]
 74%|███████▍  | 18121/24420 [7:33:42<2:35:55,  1.49s/it]
 74%|███████▍  | 18122/24420 [7:33:44<2:35:47,  1.48s/it]
 74%|███████▍  | 18123/24420 [7:33:45<2:35:46,  1.48s/it]
 74%|███████▍  | 18124/24420 [7:33:47<2:36:00,  1.49s/it]
 74%|███████▍  | 18125/24420 [7:33:48<2:35:55,  1.49s/it]
                                                         
{'loss': 0.0821, 'learning_rate': 5e-05, 'epoch': 3.71}

 74%|███████▍  | 18125/24420 [7:33:48<2:35:55,  1.49s/it]
 74%|███████▍  | 18126/24420 [7:33:50<2:35:59,  1.49s/it]
 74%|███████▍  | 18127/24420 [7:33:51<2:36:05,  1.49s/it]
 74%|███████▍  | 18128/24420 [7:33:53<2:36:02,  1.49s/it]
 74%|███████▍  | 18129/24420 [7:33:54<2:35:41,  1.48s/it]
 74%|███████▍  | 18130/24420 [7:33:56<2:35:55,  1.49s/it]
                                                         
{'loss': 0.0813, 'learning_rate': 5e-05, 'epoch': 3.71}

 74%|███████▍  | 18130/24420 [7:33:56<2:35:55,  1.49s/it]
 74%|███████▍  | 18131/24420 [7:33:57<2:35:57,  1.49s/it]
 74%|███████▍  | 18132/24420 [7:33:59<2:36:06,  1.49s/it]
 74%|███████▍  | 18133/24420 [7:34:00<2:36:39,  1.50s/it]
 74%|███████▍  | 18134/24420 [7:34:02<2:36:23,  1.49s/it]
 74%|███████▍  | 18135/24420 [7:34:03<2:36:14,  1.49s/it]
                                                         
{'loss': 0.0848, 'learning_rate': 5e-05, 'epoch': 3.71}

 74%|███████▍  | 18135/24420 [7:34:03<2:36:14,  1.49s/it]
 74%|███████▍  | 18136/24420 [7:34:05<2:36:11,  1.49s/it]
 74%|███████▍  | 18137/24420 [7:34:06<2:35:54,  1.49s/it]
 74%|███████▍  | 18138/24420 [7:34:08<2:35:42,  1.49s/it]
 74%|███████▍  | 18139/24420 [7:34:09<2:35:32,  1.49s/it]
 74%|███████▍  | 18140/24420 [7:34:11<2:35:38,  1.49s/it]
                                                         
{'loss': 0.0795, 'learning_rate': 5e-05, 'epoch': 3.71}

 74%|███████▍  | 18140/24420 [7:34:11<2:35:38,  1.49s/it]
 74%|███████▍  | 18141/24420 [7:34:12<2:35:54,  1.49s/it]
 74%|███████▍  | 18142/24420 [7:34:14<2:35:49,  1.49s/it]
 74%|███████▍  | 18143/24420 [7:34:15<2:35:30,  1.49s/it]
 74%|███████▍  | 18144/24420 [7:34:17<2:35:24,  1.49s/it]
 74%|███████▍  | 18145/24420 [7:34:18<2:35:36,  1.49s/it]
                                                         
{'loss': 0.0679, 'learning_rate': 5e-05, 'epoch': 3.72}

 74%|███████▍  | 18145/24420 [7:34:18<2:35:36,  1.49s/it]
 74%|███████▍  | 18146/24420 [7:34:20<2:35:30,  1.49s/it]
 74%|███████▍  | 18147/24420 [7:34:21<2:35:16,  1.49s/it]
 74%|███████▍  | 18148/24420 [7:34:23<2:35:17,  1.49s/it]
 74%|███████▍  | 18149/24420 [7:34:24<2:35:26,  1.49s/it]
 74%|███████▍  | 18150/24420 [7:34:25<2:35:24,  1.49s/it]
                                                         
{'loss': 0.0714, 'learning_rate': 5e-05, 'epoch': 3.72}

 74%|███████▍  | 18150/24420 [7:34:25<2:35:24,  1.49s/it]
 74%|███████▍  | 18151/24420 [7:34:27<2:35:20,  1.49s/it]
 74%|███████▍  | 18152/24420 [7:34:28<2:35:02,  1.48s/it]
 74%|███████▍  | 18153/24420 [7:34:30<2:35:04,  1.48s/it]
 74%|███████▍  | 18154/24420 [7:34:31<2:35:11,  1.49s/it]
 74%|███████▍  | 18155/24420 [7:34:33<2:35:13,  1.49s/it]
                                                         
{'loss': 0.0671, 'learning_rate': 5e-05, 'epoch': 3.72}

 74%|███████▍  | 18155/24420 [7:34:33<2:35:13,  1.49s/it]
 74%|███████▍  | 18156/24420 [7:34:34<2:35:17,  1.49s/it]
 74%|███████▍  | 18157/24420 [7:34:36<2:35:15,  1.49s/it]
 74%|███████▍  | 18158/24420 [7:34:37<2:35:03,  1.49s/it]
 74%|███████▍  | 18159/24420 [7:34:39<2:35:39,  1.49s/it]
 74%|███████▍  | 18160/24420 [7:34:40<2:35:18,  1.49s/it]
                                                         
{'loss': 0.0907, 'learning_rate': 5e-05, 'epoch': 3.72}

 74%|███████▍  | 18160/24420 [7:34:40<2:35:18,  1.49s/it]
 74%|███████▍  | 18161/24420 [7:34:42<2:35:44,  1.49s/it]
 74%|███████▍  | 18162/24420 [7:34:43<2:35:44,  1.49s/it]
 74%|███████▍  | 18163/24420 [7:34:45<2:35:23,  1.49s/it]
 74%|███████▍  | 18164/24420 [7:34:46<2:35:36,  1.49s/it]
 74%|███████▍  | 18165/24420 [7:34:48<2:35:17,  1.49s/it]
                                                         
{'loss': 0.0766, 'learning_rate': 5e-05, 'epoch': 3.72}

 74%|███████▍  | 18165/24420 [7:34:48<2:35:17,  1.49s/it]
 74%|███████▍  | 18166/24420 [7:34:49<2:35:07,  1.49s/it]
 74%|███████▍  | 18167/24420 [7:34:51<2:35:09,  1.49s/it]
 74%|███████▍  | 18168/24420 [7:34:52<2:34:53,  1.49s/it]
 74%|███████▍  | 18169/24420 [7:34:54<2:35:07,  1.49s/it]
 74%|███████▍  | 18170/24420 [7:34:55<2:35:07,  1.49s/it]
                                                         
{'loss': 0.0901, 'learning_rate': 5e-05, 'epoch': 3.72}

 74%|███████▍  | 18170/24420 [7:34:55<2:35:07,  1.49s/it]
 74%|███████▍  | 18171/24420 [7:34:57<2:35:12,  1.49s/it]
 74%|███████▍  | 18172/24420 [7:34:58<2:34:59,  1.49s/it]
 74%|███████▍  | 18173/24420 [7:35:00<2:34:48,  1.49s/it]
 74%|███████▍  | 18174/24420 [7:35:01<2:34:34,  1.48s/it]
 74%|███████▍  | 18175/24420 [7:35:03<2:34:30,  1.48s/it]
                                                         
{'loss': 0.0905, 'learning_rate': 5e-05, 'epoch': 3.72}

 74%|███████▍  | 18175/24420 [7:35:03<2:34:30,  1.48s/it]
 74%|███████▍  | 18176/24420 [7:35:04<2:34:33,  1.49s/it]
 74%|███████▍  | 18177/24420 [7:35:06<2:34:35,  1.49s/it]
 74%|███████▍  | 18178/24420 [7:35:07<2:34:30,  1.49s/it]
 74%|███████▍  | 18179/24420 [7:35:09<2:34:31,  1.49s/it]
 74%|███████▍  | 18180/24420 [7:35:10<2:34:21,  1.48s/it]
                                                         
{'loss': 0.0674, 'learning_rate': 5e-05, 'epoch': 3.72}

 74%|███████▍  | 18180/24420 [7:35:10<2:34:21,  1.48s/it]
 74%|███████▍  | 18181/24420 [7:35:12<2:34:33,  1.49s/it]
 74%|███████▍  | 18182/24420 [7:35:13<2:34:38,  1.49s/it]
 74%|███████▍  | 18183/24420 [7:35:15<2:34:58,  1.49s/it]
 74%|███████▍  | 18184/24420 [7:35:16<2:35:02,  1.49s/it]
 74%|███████▍  | 18185/24420 [7:35:18<2:34:56,  1.49s/it]
                                                         
{'loss': 0.0705, 'learning_rate': 5e-05, 'epoch': 3.72}

 74%|███████▍  | 18185/24420 [7:35:18<2:34:56,  1.49s/it]
 74%|███████▍  | 18186/24420 [7:35:19<2:35:06,  1.49s/it]
 74%|███████▍  | 18187/24420 [7:35:21<2:35:00,  1.49s/it]
 74%|███████▍  | 18188/24420 [7:35:22<2:34:58,  1.49s/it]
 74%|███████▍  | 18189/24420 [7:35:24<2:34:21,  1.49s/it]
 74%|███████▍  | 18190/24420 [7:35:25<2:34:19,  1.49s/it]
                                                         
{'loss': 0.0908, 'learning_rate': 5e-05, 'epoch': 3.72}

 74%|███████▍  | 18190/24420 [7:35:25<2:34:19,  1.49s/it]
 74%|███████▍  | 18191/24420 [7:35:26<2:34:08,  1.48s/it]
 74%|███████▍  | 18192/24420 [7:35:28<2:34:09,  1.49s/it]
 75%|███████▍  | 18193/24420 [7:35:29<2:34:14,  1.49s/it]
 75%|███████▍  | 18194/24420 [7:35:31<2:34:39,  1.49s/it]
 75%|███████▍  | 18195/24420 [7:35:32<2:34:45,  1.49s/it]
                                                         
{'loss': 0.094, 'learning_rate': 5e-05, 'epoch': 3.73}

 75%|███████▍  | 18195/24420 [7:35:32<2:34:45,  1.49s/it]
 75%|███████▍  | 18196/24420 [7:35:34<2:34:50,  1.49s/it]
 75%|███████▍  | 18197/24420 [7:35:35<2:34:48,  1.49s/it]
 75%|███████▍  | 18198/24420 [7:35:37<2:34:29,  1.49s/it]
 75%|███████▍  | 18199/24420 [7:35:38<2:34:13,  1.49s/it]
 75%|███████▍  | 18200/24420 [7:35:40<2:33:57,  1.49s/it]
                                                         
{'loss': 0.0631, 'learning_rate': 5e-05, 'epoch': 3.73}

 75%|███████▍  | 18200/24420 [7:35:40<2:33:57,  1.49s/it]
 75%|███████▍  | 18201/24420 [7:35:45<4:13:02,  2.44s/it]
 75%|███████▍  | 18202/24420 [7:35:46<3:43:23,  2.16s/it]
 75%|███████▍  | 18203/24420 [7:35:48<3:22:28,  1.95s/it]
 75%|███████▍  | 18204/24420 [7:35:49<3:08:00,  1.81s/it]
 75%|███████▍  | 18205/24420 [7:35:51<2:58:12,  1.72s/it]
                                                         
{'loss': 0.0884, 'learning_rate': 5e-05, 'epoch': 3.73}

 75%|███████▍  | 18205/24420 [7:35:51<2:58:12,  1.72s/it]
 75%|███████▍  | 18206/24420 [7:35:52<2:51:05,  1.65s/it]
 75%|███████▍  | 18207/24420 [7:35:54<2:45:49,  1.60s/it]
 75%|███████▍  | 18208/24420 [7:35:55<2:42:12,  1.57s/it]
 75%|███████▍  | 18209/24420 [7:35:56<2:39:34,  1.54s/it]
 75%|███████▍  | 18210/24420 [7:35:58<2:37:48,  1.52s/it]
                                                         
{'loss': 0.0714, 'learning_rate': 5e-05, 'epoch': 3.73}

 75%|███████▍  | 18210/24420 [7:35:58<2:37:48,  1.52s/it]
 75%|███████▍  | 18211/24420 [7:35:59<2:36:45,  1.51s/it]
 75%|███████▍  | 18212/24420 [7:36:01<2:35:55,  1.51s/it]
 75%|███████▍  | 18213/24420 [7:36:02<2:35:22,  1.50s/it]
 75%|███████▍  | 18214/24420 [7:36:04<2:34:38,  1.50s/it]
 75%|███████▍  | 18215/24420 [7:36:05<2:34:39,  1.50s/it]
                                                         
{'loss': 0.0667, 'learning_rate': 5e-05, 'epoch': 3.73}

 75%|███████▍  | 18215/24420 [7:36:05<2:34:39,  1.50s/it]
 75%|███████▍  | 18216/24420 [7:36:07<2:34:40,  1.50s/it]
 75%|███████▍  | 18217/24420 [7:36:08<2:34:15,  1.49s/it]
 75%|███████▍  | 18218/24420 [7:36:10<2:34:01,  1.49s/it]
 75%|███████▍  | 18219/24420 [7:36:11<2:33:42,  1.49s/it]
 75%|███████▍  | 18220/24420 [7:36:13<2:33:48,  1.49s/it]
                                                         
{'loss': 0.0754, 'learning_rate': 5e-05, 'epoch': 3.73}

 75%|███████▍  | 18220/24420 [7:36:13<2:33:48,  1.49s/it]
 75%|███████▍  | 18221/24420 [7:36:14<2:33:52,  1.49s/it]
 75%|███████▍  | 18222/24420 [7:36:16<2:33:52,  1.49s/it]
 75%|███████▍  | 18223/24420 [7:36:17<2:33:29,  1.49s/it]
 75%|███████▍  | 18224/24420 [7:36:19<2:33:26,  1.49s/it]
 75%|███████▍  | 18225/24420 [7:36:20<2:33:36,  1.49s/it]
                                                         
{'loss': 0.0811, 'learning_rate': 5e-05, 'epoch': 3.73}

 75%|███████▍  | 18225/24420 [7:36:20<2:33:36,  1.49s/it]
 75%|███████▍  | 18226/24420 [7:36:22<2:34:20,  1.50s/it]
 75%|███████▍  | 18227/24420 [7:36:23<2:34:10,  1.49s/it]
 75%|███████▍  | 18228/24420 [7:36:25<2:33:56,  1.49s/it]
 75%|███████▍  | 18229/24420 [7:36:26<2:33:38,  1.49s/it]
 75%|███████▍  | 18230/24420 [7:36:28<2:33:21,  1.49s/it]
                                                         
{'loss': 0.0814, 'learning_rate': 5e-05, 'epoch': 3.73}

 75%|███████▍  | 18230/24420 [7:36:28<2:33:21,  1.49s/it]
 75%|███████▍  | 18231/24420 [7:36:29<2:33:19,  1.49s/it]
 75%|███████▍  | 18232/24420 [7:36:31<2:33:22,  1.49s/it]
 75%|███████▍  | 18233/24420 [7:36:32<2:33:51,  1.49s/it]
 75%|███████▍  | 18234/24420 [7:36:34<2:33:40,  1.49s/it]
 75%|███████▍  | 18235/24420 [7:36:35<2:33:41,  1.49s/it]
                                                         
{'loss': 0.0856, 'learning_rate': 5e-05, 'epoch': 3.73}

 75%|███████▍  | 18235/24420 [7:36:35<2:33:41,  1.49s/it]
 75%|███████▍  | 18236/24420 [7:36:37<2:33:52,  1.49s/it]
 75%|███████▍  | 18237/24420 [7:36:38<2:33:28,  1.49s/it]
 75%|███████▍  | 18238/24420 [7:36:40<2:35:54,  1.51s/it]
 75%|███████▍  | 18239/24420 [7:36:41<2:35:05,  1.51s/it]
 75%|███████▍  | 18240/24420 [7:36:43<2:34:33,  1.50s/it]
                                                         
{'loss': 0.0879, 'learning_rate': 5e-05, 'epoch': 3.73}

 75%|███████▍  | 18240/24420 [7:36:43<2:34:33,  1.50s/it]
 75%|███████▍  | 18241/24420 [7:36:44<2:34:07,  1.50s/it]
 75%|███████▍  | 18242/24420 [7:36:46<2:33:38,  1.49s/it]
 75%|███████▍  | 18243/24420 [7:36:47<2:33:21,  1.49s/it]
 75%|███████▍  | 18244/24420 [7:36:49<2:33:33,  1.49s/it]
 75%|███████▍  | 18245/24420 [7:36:50<2:33:28,  1.49s/it]
                                                         
{'loss': 0.0727, 'learning_rate': 5e-05, 'epoch': 3.74}

 75%|███████▍  | 18245/24420 [7:36:50<2:33:28,  1.49s/it]
 75%|███████▍  | 18246/24420 [7:36:52<2:33:18,  1.49s/it]
 75%|███████▍  | 18247/24420 [7:36:53<2:33:23,  1.49s/it]
 75%|███████▍  | 18248/24420 [7:36:55<2:33:22,  1.49s/it]
 75%|███████▍  | 18249/24420 [7:36:56<2:33:36,  1.49s/it]
 75%|███████▍  | 18250/24420 [7:36:58<2:33:21,  1.49s/it]
                                                         
{'loss': 0.0776, 'learning_rate': 5e-05, 'epoch': 3.74}

 75%|███████▍  | 18250/24420 [7:36:58<2:33:21,  1.49s/it]
 75%|███████▍  | 18251/24420 [7:36:59<2:33:09,  1.49s/it]
 75%|███████▍  | 18252/24420 [7:37:01<2:32:59,  1.49s/it]
 75%|███████▍  | 18253/24420 [7:37:02<2:32:59,  1.49s/it]
 75%|███████▍  | 18254/24420 [7:37:04<2:32:48,  1.49s/it]
 75%|███████▍  | 18255/24420 [7:37:05<2:32:58,  1.49s/it]
                                                         
{'loss': 0.0745, 'learning_rate': 5e-05, 'epoch': 3.74}

 75%|███████▍  | 18255/24420 [7:37:05<2:32:58,  1.49s/it]
 75%|███████▍  | 18256/24420 [7:37:07<2:33:12,  1.49s/it]
 75%|███████▍  | 18257/24420 [7:37:08<2:33:05,  1.49s/it]
 75%|███████▍  | 18258/24420 [7:37:10<2:33:02,  1.49s/it]
 75%|███████▍  | 18259/24420 [7:37:11<2:33:01,  1.49s/it]
 75%|███████▍  | 18260/24420 [7:37:13<2:32:57,  1.49s/it]
                                                         
{'loss': 0.0904, 'learning_rate': 5e-05, 'epoch': 3.74}

 75%|███████▍  | 18260/24420 [7:37:13<2:32:57,  1.49s/it]
 75%|███████▍  | 18261/24420 [7:37:14<2:32:57,  1.49s/it]
 75%|███████▍  | 18262/24420 [7:37:15<2:33:15,  1.49s/it]
 75%|███████▍  | 18263/24420 [7:37:17<2:32:56,  1.49s/it]
 75%|███████▍  | 18264/24420 [7:37:18<2:32:44,  1.49s/it]
 75%|███████▍  | 18265/24420 [7:37:20<2:32:34,  1.49s/it]
                                                         
{'loss': 0.0702, 'learning_rate': 5e-05, 'epoch': 3.74}

 75%|███████▍  | 18265/24420 [7:37:20<2:32:34,  1.49s/it]
 75%|███████▍  | 18266/24420 [7:37:21<2:32:20,  1.49s/it]
 75%|███████▍  | 18267/24420 [7:37:23<2:32:28,  1.49s/it]
 75%|███████▍  | 18268/24420 [7:37:24<2:32:29,  1.49s/it]
 75%|███████▍  | 18269/24420 [7:37:26<2:32:48,  1.49s/it]
 75%|███████▍  | 18270/24420 [7:37:27<2:32:34,  1.49s/it]
                                                         
{'loss': 0.0911, 'learning_rate': 5e-05, 'epoch': 3.74}

 75%|███████▍  | 18270/24420 [7:37:27<2:32:34,  1.49s/it]
 75%|███████▍  | 18271/24420 [7:37:29<2:32:29,  1.49s/it]
 75%|███████▍  | 18272/24420 [7:37:30<2:32:46,  1.49s/it]
 75%|███████▍  | 18273/24420 [7:37:32<2:32:25,  1.49s/it]
 75%|███████▍  | 18274/24420 [7:37:33<2:32:19,  1.49s/it]
 75%|███████▍  | 18275/24420 [7:37:35<2:32:23,  1.49s/it]
                                                         
{'loss': 0.068, 'learning_rate': 5e-05, 'epoch': 3.74}

 75%|███████▍  | 18275/24420 [7:37:35<2:32:23,  1.49s/it]
 75%|███████▍  | 18276/24420 [7:37:36<2:32:53,  1.49s/it]
 75%|███████▍  | 18277/24420 [7:37:38<2:32:33,  1.49s/it]
 75%|███████▍  | 18278/24420 [7:37:39<2:32:27,  1.49s/it]
 75%|███████▍  | 18279/24420 [7:37:41<2:32:14,  1.49s/it]
 75%|███████▍  | 18280/24420 [7:37:42<2:32:09,  1.49s/it]
                                                         
{'loss': 0.0637, 'learning_rate': 5e-05, 'epoch': 3.74}

 75%|███████▍  | 18280/24420 [7:37:42<2:32:09,  1.49s/it]
 75%|███████▍  | 18281/24420 [7:37:44<2:32:03,  1.49s/it]
 75%|███████▍  | 18282/24420 [7:37:45<2:31:51,  1.48s/it]
 75%|███████▍  | 18283/24420 [7:37:47<2:32:10,  1.49s/it]
 75%|███████▍  | 18284/24420 [7:37:48<2:31:51,  1.49s/it]
 75%|███████▍  | 18285/24420 [7:37:50<2:31:42,  1.48s/it]
                                                         
{'loss': 0.0821, 'learning_rate': 5e-05, 'epoch': 3.74}

 75%|███████▍  | 18285/24420 [7:37:50<2:31:42,  1.48s/it]
 75%|███████▍  | 18286/24420 [7:37:51<2:31:36,  1.48s/it]
 75%|███████▍  | 18287/24420 [7:37:53<2:31:37,  1.48s/it]
 75%|███████▍  | 18288/24420 [7:37:54<2:31:35,  1.48s/it]
 75%|███████▍  | 18289/24420 [7:37:56<2:32:00,  1.49s/it]
 75%|███████▍  | 18290/24420 [7:37:57<2:32:08,  1.49s/it]
                                                         
{'loss': 0.0828, 'learning_rate': 5e-05, 'epoch': 3.74}

 75%|███████▍  | 18290/24420 [7:37:57<2:32:08,  1.49s/it]
 75%|███████▍  | 18291/24420 [7:37:59<2:32:07,  1.49s/it]
 75%|███████▍  | 18292/24420 [7:38:00<2:31:58,  1.49s/it]
 75%|███████▍  | 18293/24420 [7:38:02<2:31:56,  1.49s/it]
 75%|███████▍  | 18294/24420 [7:38:03<2:32:01,  1.49s/it]
 75%|███████▍  | 18295/24420 [7:38:05<2:31:58,  1.49s/it]
                                                         
{'loss': 0.0953, 'learning_rate': 5e-05, 'epoch': 3.75}

 75%|███████▍  | 18295/24420 [7:38:05<2:31:58,  1.49s/it]
 75%|███████▍  | 18296/24420 [7:38:06<2:32:00,  1.49s/it]
 75%|███████▍  | 18297/24420 [7:38:08<2:32:09,  1.49s/it]
 75%|███████▍  | 18298/24420 [7:38:09<2:31:56,  1.49s/it]
 75%|███████▍  | 18299/24420 [7:38:11<2:31:48,  1.49s/it]
 75%|███████▍  | 18300/24420 [7:38:12<2:31:40,  1.49s/it]
                                                         
{'loss': 0.1118, 'learning_rate': 5e-05, 'epoch': 3.75}

 75%|███████▍  | 18300/24420 [7:38:12<2:31:40,  1.49s/it]
 75%|███████▍  | 18301/24420 [7:38:14<2:31:49,  1.49s/it]
 75%|███████▍  | 18302/24420 [7:38:15<2:31:47,  1.49s/it]
 75%|███████▍  | 18303/24420 [7:38:16<2:31:28,  1.49s/it]
 75%|███████▍  | 18304/24420 [7:38:18<2:31:42,  1.49s/it]
 75%|███████▍  | 18305/24420 [7:38:19<2:31:37,  1.49s/it]
                                                         
{'loss': 0.0811, 'learning_rate': 5e-05, 'epoch': 3.75}

 75%|███████▍  | 18305/24420 [7:38:19<2:31:37,  1.49s/it]
 75%|███████▍  | 18306/24420 [7:38:21<2:31:36,  1.49s/it]
 75%|███████▍  | 18307/24420 [7:38:22<2:31:43,  1.49s/it]
 75%|███████▍  | 18308/24420 [7:38:24<2:31:51,  1.49s/it]
 75%|███████▍  | 18309/24420 [7:38:25<2:31:50,  1.49s/it]
 75%|███████▍  | 18310/24420 [7:38:27<2:31:48,  1.49s/it]
                                                         
{'loss': 0.0707, 'learning_rate': 5e-05, 'epoch': 3.75}

 75%|███████▍  | 18310/24420 [7:38:27<2:31:48,  1.49s/it]
 75%|███████▍  | 18311/24420 [7:38:28<2:31:39,  1.49s/it]
 75%|███████▍  | 18312/24420 [7:38:30<2:31:35,  1.49s/it]
 75%|███████▍  | 18313/24420 [7:38:31<2:31:32,  1.49s/it]
 75%|███████▍  | 18314/24420 [7:38:33<2:31:31,  1.49s/it]
 75%|███████▌  | 18315/24420 [7:38:34<2:31:07,  1.49s/it]
                                                         
{'loss': 0.0835, 'learning_rate': 5e-05, 'epoch': 3.75}

 75%|███████▌  | 18315/24420 [7:38:34<2:31:07,  1.49s/it]
 75%|███████▌  | 18316/24420 [7:38:36<2:31:21,  1.49s/it]
 75%|███████▌  | 18317/24420 [7:38:37<2:31:17,  1.49s/it]
 75%|███████▌  | 18318/24420 [7:38:39<2:31:24,  1.49s/it]
 75%|███████▌  | 18319/24420 [7:38:40<2:31:28,  1.49s/it]
 75%|███████▌  | 18320/24420 [7:38:42<2:31:30,  1.49s/it]
                                                         
{'loss': 0.0713, 'learning_rate': 5e-05, 'epoch': 3.75}

 75%|███████▌  | 18320/24420 [7:38:42<2:31:30,  1.49s/it]
 75%|███████▌  | 18321/24420 [7:38:43<2:31:23,  1.49s/it]
 75%|███████▌  | 18322/24420 [7:38:45<2:31:08,  1.49s/it]
 75%|███████▌  | 18323/24420 [7:38:46<2:31:09,  1.49s/it]
 75%|███████▌  | 18324/24420 [7:38:48<2:31:00,  1.49s/it]
 75%|███████▌  | 18325/24420 [7:38:49<2:30:39,  1.48s/it]
                                                         
{'loss': 0.083, 'learning_rate': 5e-05, 'epoch': 3.75}

 75%|███████▌  | 18325/24420 [7:38:49<2:30:39,  1.48s/it]
 75%|███████▌  | 18326/24420 [7:38:51<2:30:37,  1.48s/it]
 75%|███████▌  | 18327/24420 [7:38:52<2:30:38,  1.48s/it]
 75%|███████▌  | 18328/24420 [7:38:54<2:30:39,  1.48s/it]
 75%|███████▌  | 18329/24420 [7:38:55<2:30:54,  1.49s/it]
 75%|███████▌  | 18330/24420 [7:38:57<2:30:40,  1.48s/it]
                                                         
{'loss': 0.0697, 'learning_rate': 5e-05, 'epoch': 3.75}

 75%|███████▌  | 18330/24420 [7:38:57<2:30:40,  1.48s/it]
 75%|███████▌  | 18331/24420 [7:38:58<2:30:59,  1.49s/it]
 75%|███████▌  | 18332/24420 [7:39:00<2:30:53,  1.49s/it]
 75%|███████▌  | 18333/24420 [7:39:01<2:30:40,  1.49s/it]
 75%|███████▌  | 18334/24420 [7:39:03<2:30:54,  1.49s/it]
 75%|███████▌  | 18335/24420 [7:39:04<2:30:47,  1.49s/it]
                                                         
{'loss': 0.063, 'learning_rate': 5e-05, 'epoch': 3.75}

 75%|███████▌  | 18335/24420 [7:39:04<2:30:47,  1.49s/it]
 75%|███████▌  | 18336/24420 [7:39:06<2:34:38,  1.53s/it]
 75%|███████▌  | 18337/24420 [7:39:07<2:33:30,  1.51s/it]
 75%|███████▌  | 18338/24420 [7:39:09<2:32:30,  1.50s/it]
 75%|███████▌  | 18339/24420 [7:39:10<2:32:08,  1.50s/it]
 75%|███████▌  | 18340/24420 [7:39:12<2:31:40,  1.50s/it]
                                                         
{'loss': 0.0809, 'learning_rate': 5e-05, 'epoch': 3.76}

 75%|███████▌  | 18340/24420 [7:39:12<2:31:40,  1.50s/it]
 75%|███████▌  | 18341/24420 [7:39:13<2:31:21,  1.49s/it]
 75%|███████▌  | 18342/24420 [7:39:15<2:30:57,  1.49s/it]
 75%|███████▌  | 18343/24420 [7:39:16<2:30:58,  1.49s/it]
 75%|███████▌  | 18344/24420 [7:39:18<2:30:48,  1.49s/it]
 75%|███████▌  | 18345/24420 [7:39:19<2:30:48,  1.49s/it]
                                                         
{'loss': 0.0804, 'learning_rate': 5e-05, 'epoch': 3.76}

 75%|███████▌  | 18345/24420 [7:39:19<2:30:48,  1.49s/it]
 75%|███████▌  | 18346/24420 [7:39:21<2:30:49,  1.49s/it]
 75%|███████▌  | 18347/24420 [7:39:22<2:30:37,  1.49s/it]
 75%|███████▌  | 18348/24420 [7:39:24<2:30:37,  1.49s/it]
 75%|███████▌  | 18349/24420 [7:39:25<2:30:35,  1.49s/it]
 75%|███████▌  | 18350/24420 [7:39:27<2:30:19,  1.49s/it]
                                                         
{'loss': 0.0709, 'learning_rate': 5e-05, 'epoch': 3.76}

 75%|███████▌  | 18350/24420 [7:39:27<2:30:19,  1.49s/it]
 75%|███████▌  | 18351/24420 [7:39:28<2:30:23,  1.49s/it]
 75%|███████▌  | 18352/24420 [7:39:29<2:30:37,  1.49s/it]
 75%|███████▌  | 18353/24420 [7:39:31<2:30:28,  1.49s/it]
 75%|███████▌  | 18354/24420 [7:39:32<2:31:14,  1.50s/it]
 75%|███████▌  | 18355/24420 [7:39:34<2:30:45,  1.49s/it]
                                                         
{'loss': 0.0729, 'learning_rate': 5e-05, 'epoch': 3.76}

 75%|███████▌  | 18355/24420 [7:39:34<2:30:45,  1.49s/it]
 75%|███████▌  | 18356/24420 [7:39:35<2:30:34,  1.49s/it]
 75%|███████▌  | 18357/24420 [7:39:37<2:30:23,  1.49s/it]
 75%|███████▌  | 18358/24420 [7:39:38<2:30:10,  1.49s/it]
 75%|███████▌  | 18359/24420 [7:39:40<2:30:06,  1.49s/it]
 75%|███████▌  | 18360/24420 [7:39:41<2:30:08,  1.49s/it]
                                                         
{'loss': 0.0814, 'learning_rate': 5e-05, 'epoch': 3.76}

 75%|███████▌  | 18360/24420 [7:39:41<2:30:08,  1.49s/it]
 75%|███████▌  | 18361/24420 [7:39:43<2:30:07,  1.49s/it]
 75%|███████▌  | 18362/24420 [7:39:44<2:30:04,  1.49s/it]
 75%|███████▌  | 18363/24420 [7:39:46<2:30:10,  1.49s/it]
 75%|███████▌  | 18364/24420 [7:39:47<2:30:10,  1.49s/it]
 75%|███████▌  | 18365/24420 [7:39:49<2:30:11,  1.49s/it]
                                                         
{'loss': 0.0741, 'learning_rate': 5e-05, 'epoch': 3.76}

 75%|███████▌  | 18365/24420 [7:39:49<2:30:11,  1.49s/it]
 75%|███████▌  | 18366/24420 [7:39:50<2:30:17,  1.49s/it]
 75%|███████▌  | 18367/24420 [7:39:52<2:29:54,  1.49s/it]
 75%|███████▌  | 18368/24420 [7:39:53<2:29:51,  1.49s/it]
 75%|███████▌  | 18369/24420 [7:39:55<2:29:57,  1.49s/it]
 75%|███████▌  | 18370/24420 [7:39:56<2:30:17,  1.49s/it]
                                                         
{'loss': 0.0765, 'learning_rate': 5e-05, 'epoch': 3.76}

 75%|███████▌  | 18370/24420 [7:39:56<2:30:17,  1.49s/it]
 75%|███████▌  | 18371/24420 [7:39:58<2:30:11,  1.49s/it]
 75%|███████▌  | 18372/24420 [7:39:59<2:30:09,  1.49s/it]
 75%|███████▌  | 18373/24420 [7:40:01<2:30:09,  1.49s/it]
 75%|███████▌  | 18374/24420 [7:40:02<2:30:07,  1.49s/it]
 75%|███████▌  | 18375/24420 [7:40:04<2:30:03,  1.49s/it]
                                                         
{'loss': 0.0717, 'learning_rate': 5e-05, 'epoch': 3.76}

 75%|███████▌  | 18375/24420 [7:40:04<2:30:03,  1.49s/it]
 75%|███████▌  | 18376/24420 [7:40:05<2:29:48,  1.49s/it]
 75%|███████▌  | 18377/24420 [7:40:07<2:29:51,  1.49s/it]
 75%|███████▌  | 18378/24420 [7:40:08<2:29:44,  1.49s/it]
 75%|███████▌  | 18379/24420 [7:40:10<2:29:32,  1.49s/it]
 75%|███████▌  | 18380/24420 [7:40:11<2:29:48,  1.49s/it]
                                                         
{'loss': 0.0686, 'learning_rate': 5e-05, 'epoch': 3.76}

 75%|███████▌  | 18380/24420 [7:40:11<2:29:48,  1.49s/it]
 75%|███████▌  | 18381/24420 [7:40:13<2:29:42,  1.49s/it]
 75%|███████▌  | 18382/24420 [7:40:14<2:29:37,  1.49s/it]
 75%|███████▌  | 18383/24420 [7:40:16<2:29:41,  1.49s/it]
 75%|███████▌  | 18384/24420 [7:40:17<2:30:03,  1.49s/it]
 75%|███████▌  | 18385/24420 [7:40:19<2:29:42,  1.49s/it]
                                                         
{'loss': 0.086, 'learning_rate': 5e-05, 'epoch': 3.76}

 75%|███████▌  | 18385/24420 [7:40:19<2:29:42,  1.49s/it]
 75%|███████▌  | 18386/24420 [7:40:20<2:29:34,  1.49s/it]
 75%|███████▌  | 18387/24420 [7:40:22<2:29:34,  1.49s/it]
 75%|███████▌  | 18388/24420 [7:40:23<2:29:45,  1.49s/it]
 75%|███████▌  | 18389/24420 [7:40:25<2:29:39,  1.49s/it]
 75%|███████▌  | 18390/24420 [7:40:26<2:29:39,  1.49s/it]
                                                         
{'loss': 0.0907, 'learning_rate': 5e-05, 'epoch': 3.77}

 75%|███████▌  | 18390/24420 [7:40:26<2:29:39,  1.49s/it]
 75%|███████▌  | 18391/24420 [7:40:28<2:29:33,  1.49s/it]
 75%|███████▌  | 18392/24420 [7:40:29<2:29:21,  1.49s/it]
 75%|███████▌  | 18393/24420 [7:40:31<2:29:17,  1.49s/it]
 75%|███████▌  | 18394/24420 [7:40:32<2:29:02,  1.48s/it]
 75%|███████▌  | 18395/24420 [7:40:33<2:29:03,  1.48s/it]
                                                         
{'loss': 0.0752, 'learning_rate': 5e-05, 'epoch': 3.77}

 75%|███████▌  | 18395/24420 [7:40:33<2:29:03,  1.48s/it]
 75%|███████▌  | 18396/24420 [7:40:35<2:29:01,  1.48s/it]
 75%|███████▌  | 18397/24420 [7:40:36<2:29:04,  1.49s/it]
 75%|███████▌  | 18398/24420 [7:40:38<2:29:07,  1.49s/it]
 75%|███████▌  | 18399/24420 [7:40:39<2:29:25,  1.49s/it]
 75%|███████▌  | 18400/24420 [7:40:41<2:29:17,  1.49s/it]
                                                         
{'loss': 0.0828, 'learning_rate': 5e-05, 'epoch': 3.77}

 75%|███████▌  | 18400/24420 [7:40:41<2:29:17,  1.49s/it]
 75%|███████▌  | 18401/24420 [7:40:46<4:04:53,  2.44s/it]
 75%|███████▌  | 18402/24420 [7:40:48<4:00:32,  2.40s/it]
 75%|███████▌  | 18403/24420 [7:40:49<3:33:09,  2.13s/it]
 75%|███████▌  | 18404/24420 [7:40:51<3:14:12,  1.94s/it]
 75%|███████▌  | 18405/24420 [7:40:52<3:00:39,  1.80s/it]
                                                         
{'loss': 0.078, 'learning_rate': 5e-05, 'epoch': 3.77}

 75%|███████▌  | 18405/24420 [7:40:52<3:00:39,  1.80s/it]
 75%|███████▌  | 18406/24420 [7:40:54<2:50:59,  1.71s/it]
 75%|███████▌  | 18407/24420 [7:40:55<2:44:43,  1.64s/it]
 75%|███████▌  | 18408/24420 [7:40:57<2:40:01,  1.60s/it]
 75%|███████▌  | 18409/24420 [7:40:58<2:36:39,  1.56s/it]
 75%|███████▌  | 18410/24420 [7:41:00<2:33:57,  1.54s/it]
                                                         
{'loss': 0.0766, 'learning_rate': 5e-05, 'epoch': 3.77}

 75%|███████▌  | 18410/24420 [7:41:00<2:33:57,  1.54s/it]
 75%|███████▌  | 18411/24420 [7:41:01<2:32:46,  1.53s/it]
 75%|███████▌  | 18412/24420 [7:41:03<2:31:39,  1.51s/it]
 75%|███████▌  | 18413/24420 [7:41:04<2:31:09,  1.51s/it]
 75%|███████▌  | 18414/24420 [7:41:06<2:30:21,  1.50s/it]
 75%|███████▌  | 18415/24420 [7:41:07<2:30:00,  1.50s/it]
                                                         
{'loss': 0.0794, 'learning_rate': 5e-05, 'epoch': 3.77}

 75%|███████▌  | 18415/24420 [7:41:07<2:30:00,  1.50s/it]
 75%|███████▌  | 18416/24420 [7:41:09<2:29:49,  1.50s/it]
 75%|███████▌  | 18417/24420 [7:41:10<2:29:31,  1.49s/it]
 75%|███████▌  | 18418/24420 [7:41:12<2:29:19,  1.49s/it]
 75%|███████▌  | 18419/24420 [7:41:13<2:29:17,  1.49s/it]
 75%|███████▌  | 18420/24420 [7:41:15<2:28:54,  1.49s/it]
                                                         
{'loss': 0.0796, 'learning_rate': 5e-05, 'epoch': 3.77}

 75%|███████▌  | 18420/24420 [7:41:15<2:28:54,  1.49s/it]
 75%|███████▌  | 18421/24420 [7:41:16<2:29:06,  1.49s/it]
 75%|███████▌  | 18422/24420 [7:41:18<2:29:09,  1.49s/it]
 75%|███████▌  | 18423/24420 [7:41:19<2:29:02,  1.49s/it]
 75%|███████▌  | 18424/24420 [7:41:21<2:28:56,  1.49s/it]
 75%|███████▌  | 18425/24420 [7:41:22<2:28:54,  1.49s/it]
                                                         
{'loss': 0.0885, 'learning_rate': 5e-05, 'epoch': 3.77}

 75%|███████▌  | 18425/24420 [7:41:22<2:28:54,  1.49s/it]
 75%|███████▌  | 18426/24420 [7:41:24<2:28:54,  1.49s/it]
 75%|███████▌  | 18427/24420 [7:41:25<2:28:41,  1.49s/it]
 75%|███████▌  | 18428/24420 [7:41:27<2:28:36,  1.49s/it]
 75%|███████▌  | 18429/24420 [7:41:28<2:28:21,  1.49s/it]
 75%|███████▌  | 18430/24420 [7:41:30<2:28:11,  1.48s/it]
                                                         
{'loss': 0.075, 'learning_rate': 5e-05, 'epoch': 3.77}

 75%|███████▌  | 18430/24420 [7:41:30<2:28:11,  1.48s/it]
 75%|███████▌  | 18431/24420 [7:41:31<2:28:14,  1.49s/it]
 75%|███████▌  | 18432/24420 [7:41:33<2:28:09,  1.48s/it]
 75%|███████▌  | 18433/24420 [7:41:34<2:28:21,  1.49s/it]
 75%|███████▌  | 18434/24420 [7:41:36<2:28:29,  1.49s/it]
 75%|███████▌  | 18435/24420 [7:41:37<2:28:19,  1.49s/it]
                                                         
{'loss': 0.0815, 'learning_rate': 5e-05, 'epoch': 3.77}

 75%|███████▌  | 18435/24420 [7:41:37<2:28:19,  1.49s/it]
 75%|███████▌  | 18436/24420 [7:41:38<2:28:28,  1.49s/it]
 75%|███████▌  | 18437/24420 [7:41:40<2:28:10,  1.49s/it]
 76%|███████▌  | 18438/24420 [7:41:41<2:28:03,  1.49s/it]
 76%|███████▌  | 18439/24420 [7:41:43<2:28:10,  1.49s/it]
 76%|███████▌  | 18440/24420 [7:41:44<2:28:01,  1.49s/it]
                                                         
{'loss': 0.0729, 'learning_rate': 5e-05, 'epoch': 3.78}

 76%|███████▌  | 18440/24420 [7:41:44<2:28:01,  1.49s/it]
 76%|███████▌  | 18441/24420 [7:41:46<2:28:20,  1.49s/it]
 76%|███████▌  | 18442/24420 [7:41:47<2:28:25,  1.49s/it]
 76%|███████▌  | 18443/24420 [7:41:49<2:28:14,  1.49s/it]
 76%|███████▌  | 18444/24420 [7:41:50<2:28:10,  1.49s/it]
 76%|███████▌  | 18445/24420 [7:41:52<2:28:06,  1.49s/it]
                                                         
{'loss': 0.0675, 'learning_rate': 5e-05, 'epoch': 3.78}

 76%|███████▌  | 18445/24420 [7:41:52<2:28:06,  1.49s/it]
 76%|███████▌  | 18446/24420 [7:41:53<2:28:25,  1.49s/it]
 76%|███████▌  | 18447/24420 [7:41:55<2:28:26,  1.49s/it]
 76%|███████▌  | 18448/24420 [7:41:56<2:28:15,  1.49s/it]
 76%|███████▌  | 18449/24420 [7:41:58<2:28:03,  1.49s/it]
 76%|███████▌  | 18450/24420 [7:41:59<2:28:18,  1.49s/it]
                                                         
{'loss': 0.0624, 'learning_rate': 5e-05, 'epoch': 3.78}

 76%|███████▌  | 18450/24420 [7:41:59<2:28:18,  1.49s/it]
 76%|███████▌  | 18451/24420 [7:42:01<2:28:12,  1.49s/it]
 76%|███████▌  | 18452/24420 [7:42:02<2:28:39,  1.49s/it]
 76%|███████▌  | 18453/24420 [7:42:04<2:28:11,  1.49s/it]
 76%|███████▌  | 18454/24420 [7:42:05<2:27:59,  1.49s/it]
 76%|███████▌  | 18455/24420 [7:42:07<2:27:42,  1.49s/it]
                                                         
{'loss': 0.0871, 'learning_rate': 5e-05, 'epoch': 3.78}

 76%|███████▌  | 18455/24420 [7:42:07<2:27:42,  1.49s/it]
 76%|███████▌  | 18456/24420 [7:42:08<2:27:40,  1.49s/it]
 76%|███████▌  | 18457/24420 [7:42:10<2:27:39,  1.49s/it]
 76%|███████▌  | 18458/24420 [7:42:11<2:27:39,  1.49s/it]
 76%|███████▌  | 18459/24420 [7:42:13<2:27:39,  1.49s/it]
 76%|███████▌  | 18460/24420 [7:42:14<2:27:25,  1.48s/it]
                                                         
{'loss': 0.0632, 'learning_rate': 5e-05, 'epoch': 3.78}

 76%|███████▌  | 18460/24420 [7:42:14<2:27:25,  1.48s/it]
 76%|███████▌  | 18461/24420 [7:42:16<2:27:30,  1.49s/it]
 76%|███████▌  | 18462/24420 [7:42:17<2:27:50,  1.49s/it]
 76%|███████▌  | 18463/24420 [7:42:19<2:27:47,  1.49s/it]
 76%|███████▌  | 18464/24420 [7:42:20<2:27:58,  1.49s/it]
 76%|███████▌  | 18465/24420 [7:42:22<2:27:54,  1.49s/it]
                                                         
{'loss': 0.0845, 'learning_rate': 5e-05, 'epoch': 3.78}

 76%|███████▌  | 18465/24420 [7:42:22<2:27:54,  1.49s/it]
 76%|███████▌  | 18466/24420 [7:42:23<2:27:51,  1.49s/it]
 76%|███████▌  | 18467/24420 [7:42:25<2:27:30,  1.49s/it]
 76%|███████▌  | 18468/24420 [7:42:26<2:27:06,  1.48s/it]
 76%|███████▌  | 18469/24420 [7:42:28<2:27:16,  1.48s/it]
 76%|███████▌  | 18470/24420 [7:42:29<2:29:02,  1.50s/it]
                                                         
{'loss': 0.0779, 'learning_rate': 5e-05, 'epoch': 3.78}

 76%|███████▌  | 18470/24420 [7:42:29<2:29:02,  1.50s/it]
 76%|███████▌  | 18471/24420 [7:42:31<2:28:38,  1.50s/it]
 76%|███████▌  | 18472/24420 [7:42:32<2:28:13,  1.50s/it]
 76%|███████▌  | 18473/24420 [7:42:34<2:27:45,  1.49s/it]
 76%|███████▌  | 18474/24420 [7:42:35<2:27:44,  1.49s/it]
 76%|███████▌  | 18475/24420 [7:42:37<2:27:28,  1.49s/it]
                                                         
{'loss': 0.0784, 'learning_rate': 5e-05, 'epoch': 3.78}

 76%|███████▌  | 18475/24420 [7:42:37<2:27:28,  1.49s/it]
 76%|███████▌  | 18476/24420 [7:42:38<2:27:44,  1.49s/it]
 76%|███████▌  | 18477/24420 [7:42:40<2:27:48,  1.49s/it]
 76%|███████▌  | 18478/24420 [7:42:41<2:27:36,  1.49s/it]
 76%|███████▌  | 18479/24420 [7:42:43<2:27:33,  1.49s/it]
 76%|███████▌  | 18480/24420 [7:42:44<2:27:23,  1.49s/it]
                                                         
{'loss': 0.0717, 'learning_rate': 5e-05, 'epoch': 3.78}

 76%|███████▌  | 18480/24420 [7:42:44<2:27:23,  1.49s/it]
 76%|███████▌  | 18481/24420 [7:42:46<2:27:35,  1.49s/it]
 76%|███████▌  | 18482/24420 [7:42:47<2:27:27,  1.49s/it]
 76%|███████▌  | 18483/24420 [7:42:48<2:27:00,  1.49s/it]
 76%|███████▌  | 18484/24420 [7:42:50<2:26:55,  1.49s/it]
 76%|███████▌  | 18485/24420 [7:42:51<2:27:01,  1.49s/it]
                                                         
{'loss': 0.0742, 'learning_rate': 5e-05, 'epoch': 3.78}

 76%|███████▌  | 18485/24420 [7:42:51<2:27:01,  1.49s/it]
 76%|███████▌  | 18486/24420 [7:42:53<2:27:17,  1.49s/it]
 76%|███████▌  | 18487/24420 [7:42:54<2:27:12,  1.49s/it]
 76%|███████▌  | 18488/24420 [7:42:56<2:27:13,  1.49s/it]
 76%|███████▌  | 18489/24420 [7:42:57<2:27:34,  1.49s/it]
 76%|███████▌  | 18490/24420 [7:42:59<2:27:23,  1.49s/it]
                                                         
{'loss': 0.0718, 'learning_rate': 5e-05, 'epoch': 3.79}

 76%|███████▌  | 18490/24420 [7:42:59<2:27:23,  1.49s/it]
 76%|███████▌  | 18491/24420 [7:43:00<2:27:04,  1.49s/it]
 76%|███████▌  | 18492/24420 [7:43:02<2:27:08,  1.49s/it]
 76%|███████▌  | 18493/24420 [7:43:03<2:27:16,  1.49s/it]
 76%|███████▌  | 18494/24420 [7:43:05<2:27:10,  1.49s/it]
 76%|███████▌  | 18495/24420 [7:43:06<2:27:04,  1.49s/it]
                                                         
{'loss': 0.0909, 'learning_rate': 5e-05, 'epoch': 3.79}

 76%|███████▌  | 18495/24420 [7:43:06<2:27:04,  1.49s/it]
 76%|███████▌  | 18496/24420 [7:43:08<2:27:19,  1.49s/it]
 76%|███████▌  | 18497/24420 [7:43:09<2:27:13,  1.49s/it]
 76%|███████▌  | 18498/24420 [7:43:11<2:27:24,  1.49s/it]
 76%|███████▌  | 18499/24420 [7:43:12<2:27:04,  1.49s/it]
 76%|███████▌  | 18500/24420 [7:43:14<2:26:54,  1.49s/it]
                                                         
{'loss': 0.0607, 'learning_rate': 5e-05, 'epoch': 3.79}

 76%|███████▌  | 18500/24420 [7:43:14<2:26:54,  1.49s/it]
 76%|███████▌  | 18501/24420 [7:43:15<2:26:51,  1.49s/it]
 76%|███████▌  | 18502/24420 [7:43:17<2:27:00,  1.49s/it]
 76%|███████▌  | 18503/24420 [7:43:18<2:27:32,  1.50s/it]
 76%|███████▌  | 18504/24420 [7:43:20<2:27:00,  1.49s/it]
 76%|███████▌  | 18505/24420 [7:43:21<2:26:57,  1.49s/it]
                                                         
{'loss': 0.0773, 'learning_rate': 5e-05, 'epoch': 3.79}

 76%|███████▌  | 18505/24420 [7:43:21<2:26:57,  1.49s/it]
 76%|███████▌  | 18506/24420 [7:43:23<2:27:01,  1.49s/it]
 76%|███████▌  | 18507/24420 [7:43:24<2:26:42,  1.49s/it]
 76%|███████▌  | 18508/24420 [7:43:26<2:26:50,  1.49s/it]
 76%|███████▌  | 18509/24420 [7:43:27<2:26:35,  1.49s/it]
 76%|███████▌  | 18510/24420 [7:43:29<2:26:43,  1.49s/it]
                                                         
{'loss': 0.0701, 'learning_rate': 5e-05, 'epoch': 3.79}

 76%|███████▌  | 18510/24420 [7:43:29<2:26:43,  1.49s/it]
 76%|███████▌  | 18511/24420 [7:43:30<2:26:46,  1.49s/it]
 76%|███████▌  | 18512/24420 [7:43:32<2:26:34,  1.49s/it]
 76%|███████▌  | 18513/24420 [7:43:33<2:26:28,  1.49s/it]
 76%|███████▌  | 18514/24420 [7:43:35<2:26:49,  1.49s/it]
 76%|███████▌  | 18515/24420 [7:43:36<2:26:53,  1.49s/it]
                                                         
{'loss': 0.0687, 'learning_rate': 5e-05, 'epoch': 3.79}

 76%|███████▌  | 18515/24420 [7:43:36<2:26:53,  1.49s/it]
 76%|███████▌  | 18516/24420 [7:43:38<2:27:00,  1.49s/it]
 76%|███████▌  | 18517/24420 [7:43:39<2:26:51,  1.49s/it]
 76%|███████▌  | 18518/24420 [7:43:41<2:26:35,  1.49s/it]
 76%|███████▌  | 18519/24420 [7:43:42<2:26:38,  1.49s/it]
 76%|███████▌  | 18520/24420 [7:43:44<2:26:46,  1.49s/it]
                                                         
{'loss': 0.0805, 'learning_rate': 5e-05, 'epoch': 3.79}

 76%|███████▌  | 18520/24420 [7:43:44<2:26:46,  1.49s/it]
 76%|███████▌  | 18521/24420 [7:43:45<2:26:36,  1.49s/it]
 76%|███████▌  | 18522/24420 [7:43:47<2:26:24,  1.49s/it]
 76%|███████▌  | 18523/24420 [7:43:48<2:26:32,  1.49s/it]
 76%|███████▌  | 18524/24420 [7:43:50<2:26:40,  1.49s/it]
 76%|███████▌  | 18525/24420 [7:43:51<2:26:18,  1.49s/it]
                                                         
{'loss': 0.083, 'learning_rate': 5e-05, 'epoch': 3.79}

 76%|███████▌  | 18525/24420 [7:43:51<2:26:18,  1.49s/it]
 76%|███████▌  | 18526/24420 [7:43:53<2:26:10,  1.49s/it]
 76%|███████▌  | 18527/24420 [7:43:54<2:26:14,  1.49s/it]
 76%|███████▌  | 18528/24420 [7:43:56<2:26:13,  1.49s/it]
 76%|███████▌  | 18529/24420 [7:43:57<2:26:12,  1.49s/it]
 76%|███████▌  | 18530/24420 [7:43:59<2:25:56,  1.49s/it]
                                                         
{'loss': 0.0789, 'learning_rate': 5e-05, 'epoch': 3.79}

 76%|███████▌  | 18530/24420 [7:43:59<2:25:56,  1.49s/it]
 76%|███████▌  | 18531/24420 [7:44:01<2:44:30,  1.68s/it]
 76%|███████▌  | 18532/24420 [7:44:02<2:39:12,  1.62s/it]
 76%|███████▌  | 18533/24420 [7:44:04<2:35:51,  1.59s/it]
 76%|███████▌  | 18534/24420 [7:44:05<2:32:58,  1.56s/it]
 76%|███████▌  | 18535/24420 [7:44:07<2:30:45,  1.54s/it]
                                                         
{'loss': 0.0817, 'learning_rate': 5e-05, 'epoch': 3.8}

 76%|███████▌  | 18535/24420 [7:44:07<2:30:45,  1.54s/it]
 76%|███████▌  | 18536/24420 [7:44:08<2:29:28,  1.52s/it]
 76%|███████▌  | 18537/24420 [7:44:10<2:28:12,  1.51s/it]
 76%|███████▌  | 18538/24420 [7:44:11<2:27:17,  1.50s/it]
 76%|███████▌  | 18539/24420 [7:44:13<2:26:59,  1.50s/it]
 76%|███████▌  | 18540/24420 [7:44:14<2:26:34,  1.50s/it]
                                                         
{'loss': 0.0891, 'learning_rate': 5e-05, 'epoch': 3.8}

 76%|███████▌  | 18540/24420 [7:44:14<2:26:34,  1.50s/it]
 76%|███████▌  | 18541/24420 [7:44:16<2:26:15,  1.49s/it]
 76%|███████▌  | 18542/24420 [7:44:17<2:26:12,  1.49s/it]
 76%|███████▌  | 18543/24420 [7:44:19<2:25:59,  1.49s/it]
 76%|███████▌  | 18544/24420 [7:44:20<2:26:51,  1.50s/it]
 76%|███████▌  | 18545/24420 [7:44:22<2:26:53,  1.50s/it]
                                                         
{'loss': 0.0838, 'learning_rate': 5e-05, 'epoch': 3.8}

 76%|███████▌  | 18545/24420 [7:44:22<2:26:53,  1.50s/it]
 76%|███████▌  | 18546/24420 [7:44:23<2:26:43,  1.50s/it]
 76%|███████▌  | 18547/24420 [7:44:25<2:26:27,  1.50s/it]
 76%|███████▌  | 18548/24420 [7:44:26<2:26:20,  1.50s/it]
 76%|███████▌  | 18549/24420 [7:44:28<2:26:08,  1.49s/it]
 76%|███████▌  | 18550/24420 [7:44:29<2:26:07,  1.49s/it]
                                                         
{'loss': 0.0693, 'learning_rate': 5e-05, 'epoch': 3.8}

 76%|███████▌  | 18550/24420 [7:44:29<2:26:07,  1.49s/it]
 76%|███████▌  | 18551/24420 [7:44:30<2:26:03,  1.49s/it]
 76%|███████▌  | 18552/24420 [7:44:32<2:25:41,  1.49s/it]
 76%|███████▌  | 18553/24420 [7:44:33<2:25:29,  1.49s/it]
 76%|███████▌  | 18554/24420 [7:44:35<2:25:14,  1.49s/it]
 76%|███████▌  | 18555/24420 [7:44:36<2:25:15,  1.49s/it]
                                                         
{'loss': 0.0682, 'learning_rate': 5e-05, 'epoch': 3.8}

 76%|███████▌  | 18555/24420 [7:44:36<2:25:15,  1.49s/it]
 76%|███████▌  | 18556/24420 [7:44:38<2:25:20,  1.49s/it]
 76%|███████▌  | 18557/24420 [7:44:39<2:25:38,  1.49s/it]
 76%|███████▌  | 18558/24420 [7:44:41<2:25:25,  1.49s/it]
 76%|███████▌  | 18559/24420 [7:44:42<2:25:31,  1.49s/it]
 76%|███████▌  | 18560/24420 [7:44:44<2:25:40,  1.49s/it]
                                                         
{'loss': 0.0751, 'learning_rate': 5e-05, 'epoch': 3.8}

 76%|███████▌  | 18560/24420 [7:44:44<2:25:40,  1.49s/it]
 76%|███████▌  | 18561/24420 [7:44:45<2:25:43,  1.49s/it]
 76%|███████▌  | 18562/24420 [7:44:47<2:25:35,  1.49s/it]
 76%|███████▌  | 18563/24420 [7:44:48<2:25:26,  1.49s/it]
 76%|███████▌  | 18564/24420 [7:44:50<2:25:18,  1.49s/it]
 76%|███████▌  | 18565/24420 [7:44:51<2:25:16,  1.49s/it]
                                                         
{'loss': 0.0751, 'learning_rate': 5e-05, 'epoch': 3.8}

 76%|███████▌  | 18565/24420 [7:44:51<2:25:16,  1.49s/it]
 76%|███████▌  | 18566/24420 [7:44:53<2:25:39,  1.49s/it]
 76%|███████▌  | 18567/24420 [7:44:54<2:25:42,  1.49s/it]
 76%|███████▌  | 18568/24420 [7:44:56<2:25:26,  1.49s/it]
 76%|███████▌  | 18569/24420 [7:44:57<2:25:13,  1.49s/it]
 76%|███████▌  | 18570/24420 [7:44:59<2:25:12,  1.49s/it]
                                                         
{'loss': 0.0843, 'learning_rate': 5e-05, 'epoch': 3.8}

 76%|███████▌  | 18570/24420 [7:44:59<2:25:12,  1.49s/it]
 76%|███████▌  | 18571/24420 [7:45:00<2:25:16,  1.49s/it]
 76%|███████▌  | 18572/24420 [7:45:02<2:25:21,  1.49s/it]
 76%|███████▌  | 18573/24420 [7:45:03<2:25:23,  1.49s/it]
 76%|███████▌  | 18574/24420 [7:45:05<2:25:31,  1.49s/it]
 76%|███████▌  | 18575/24420 [7:45:06<2:25:17,  1.49s/it]
                                                         
{'loss': 0.0655, 'learning_rate': 5e-05, 'epoch': 3.8}

 76%|███████▌  | 18575/24420 [7:45:06<2:25:17,  1.49s/it]
 76%|███████▌  | 18576/24420 [7:45:08<2:25:10,  1.49s/it]
 76%|███████▌  | 18577/24420 [7:45:09<2:24:50,  1.49s/it]
 76%|███████▌  | 18578/24420 [7:45:11<2:24:59,  1.49s/it]
 76%|███████▌  | 18579/24420 [7:45:12<2:24:51,  1.49s/it]
 76%|███████▌  | 18580/24420 [7:45:14<2:24:57,  1.49s/it]
                                                         
{'loss': 0.0731, 'learning_rate': 5e-05, 'epoch': 3.8}

 76%|███████▌  | 18580/24420 [7:45:14<2:24:57,  1.49s/it]
 76%|███████▌  | 18581/24420 [7:45:15<2:25:03,  1.49s/it]
 76%|███████▌  | 18582/24420 [7:45:17<2:25:08,  1.49s/it]
 76%|███████▌  | 18583/24420 [7:45:18<2:25:08,  1.49s/it]
 76%|███████▌  | 18584/24420 [7:45:20<2:25:09,  1.49s/it]
 76%|███████▌  | 18585/24420 [7:45:21<2:24:56,  1.49s/it]
                                                         
{'loss': 0.0865, 'learning_rate': 5e-05, 'epoch': 3.81}

 76%|███████▌  | 18585/24420 [7:45:21<2:24:56,  1.49s/it]
 76%|███████▌  | 18586/24420 [7:45:23<2:24:49,  1.49s/it]
 76%|███████▌  | 18587/24420 [7:45:24<2:24:48,  1.49s/it]
 76%|███████▌  | 18588/24420 [7:45:26<2:24:43,  1.49s/it]
 76%|███████▌  | 18589/24420 [7:45:27<2:24:50,  1.49s/it]
 76%|███████▌  | 18590/24420 [7:45:29<2:24:50,  1.49s/it]
                                                         
{'loss': 0.0764, 'learning_rate': 5e-05, 'epoch': 3.81}

 76%|███████▌  | 18590/24420 [7:45:29<2:24:50,  1.49s/it]
 76%|███████▌  | 18591/24420 [7:45:30<2:24:58,  1.49s/it]
 76%|███████▌  | 18592/24420 [7:45:32<2:25:21,  1.50s/it]
 76%|███████▌  | 18593/24420 [7:45:33<2:25:17,  1.50s/it]
 76%|███████▌  | 18594/24420 [7:45:35<2:25:10,  1.50s/it]
 76%|███████▌  | 18595/24420 [7:45:36<2:24:49,  1.49s/it]
                                                         
{'loss': 0.0757, 'learning_rate': 5e-05, 'epoch': 3.81}

 76%|███████▌  | 18595/24420 [7:45:36<2:24:49,  1.49s/it]
 76%|███████▌  | 18596/24420 [7:45:38<2:25:14,  1.50s/it]
 76%|███████▌  | 18597/24420 [7:45:39<2:25:02,  1.49s/it]
 76%|███████▌  | 18598/24420 [7:45:41<2:24:37,  1.49s/it]
 76%|███████▌  | 18599/24420 [7:45:42<2:24:39,  1.49s/it]
 76%|███████▌  | 18600/24420 [7:45:44<2:24:31,  1.49s/it]
                                                         
{'loss': 0.0877, 'learning_rate': 5e-05, 'epoch': 3.81}

 76%|███████▌  | 18600/24420 [7:45:44<2:24:31,  1.49s/it]
 76%|███████▌  | 18601/24420 [7:45:48<3:53:17,  2.41s/it]
 76%|███████▌  | 18602/24420 [7:45:50<3:26:41,  2.13s/it]
 76%|███████▌  | 18603/24420 [7:45:51<3:07:58,  1.94s/it]
 76%|███████▌  | 18604/24420 [7:45:53<2:54:32,  1.80s/it]
 76%|███████▌  | 18605/24420 [7:45:54<2:45:15,  1.71s/it]
                                                         
{'loss': 0.0746, 'learning_rate': 5e-05, 'epoch': 3.81}

 76%|███████▌  | 18605/24420 [7:45:54<2:45:15,  1.71s/it]
 76%|███████▌  | 18606/24420 [7:45:56<2:38:55,  1.64s/it]
 76%|███████▌  | 18607/24420 [7:45:57<2:34:45,  1.60s/it]
 76%|███████▌  | 18608/24420 [7:45:58<2:31:25,  1.56s/it]
 76%|███████▌  | 18609/24420 [7:46:00<2:29:14,  1.54s/it]
 76%|███████▌  | 18610/24420 [7:46:01<2:27:29,  1.52s/it]
                                                         
{'loss': 0.0919, 'learning_rate': 5e-05, 'epoch': 3.81}

 76%|███████▌  | 18610/24420 [7:46:01<2:27:29,  1.52s/it]
 76%|███████▌  | 18611/24420 [7:46:03<2:26:26,  1.51s/it]
 76%|███████▌  | 18612/24420 [7:46:04<2:25:30,  1.50s/it]
 76%|███████▌  | 18613/24420 [7:46:06<2:24:51,  1.50s/it]
 76%|███████▌  | 18614/24420 [7:46:07<2:24:59,  1.50s/it]
 76%|███████▌  | 18615/24420 [7:46:09<2:24:26,  1.49s/it]
                                                         
{'loss': 0.0862, 'learning_rate': 5e-05, 'epoch': 3.81}

 76%|███████▌  | 18615/24420 [7:46:09<2:24:26,  1.49s/it]
 76%|███████▌  | 18616/24420 [7:46:10<2:24:46,  1.50s/it]
 76%|███████▌  | 18617/24420 [7:46:12<2:24:54,  1.50s/it]
 76%|███████▌  | 18618/24420 [7:46:13<2:24:18,  1.49s/it]
 76%|███████▌  | 18619/24420 [7:46:15<2:23:57,  1.49s/it]
 76%|███████▌  | 18620/24420 [7:46:16<2:23:58,  1.49s/it]
                                                         
{'loss': 0.078, 'learning_rate': 5e-05, 'epoch': 3.81}

 76%|███████▌  | 18620/24420 [7:46:16<2:23:58,  1.49s/it]
 76%|███████▋  | 18621/24420 [7:46:18<2:23:46,  1.49s/it]
 76%|███████▋  | 18622/24420 [7:46:19<2:23:49,  1.49s/it]
 76%|███████▋  | 18623/24420 [7:46:21<2:23:52,  1.49s/it]
 76%|███████▋  | 18624/24420 [7:46:22<2:23:58,  1.49s/it]
 76%|███████▋  | 18625/24420 [7:46:24<2:23:46,  1.49s/it]
                                                         
{'loss': 0.0751, 'learning_rate': 5e-05, 'epoch': 3.81}

 76%|███████▋  | 18625/24420 [7:46:24<2:23:46,  1.49s/it]
 76%|███████▋  | 18626/24420 [7:46:25<2:23:43,  1.49s/it]
 76%|███████▋  | 18627/24420 [7:46:27<2:23:24,  1.49s/it]
 76%|███████▋  | 18628/24420 [7:46:28<2:23:37,  1.49s/it]
 76%|███████▋  | 18629/24420 [7:46:30<2:23:32,  1.49s/it]
 76%|███████▋  | 18630/24420 [7:46:31<2:23:53,  1.49s/it]
                                                         
{'loss': 0.0836, 'learning_rate': 5e-05, 'epoch': 3.81}

 76%|███████▋  | 18630/24420 [7:46:31<2:23:53,  1.49s/it]
 76%|███████▋  | 18631/24420 [7:46:33<2:24:38,  1.50s/it]
 76%|███████▋  | 18632/24420 [7:46:34<2:24:05,  1.49s/it]
 76%|███████▋  | 18633/24420 [7:46:36<2:24:02,  1.49s/it]
 76%|███████▋  | 18634/24420 [7:46:37<2:23:57,  1.49s/it]
 76%|███████▋  | 18635/24420 [7:46:39<2:23:53,  1.49s/it]
                                                         
{'loss': 0.0854, 'learning_rate': 5e-05, 'epoch': 3.82}

 76%|███████▋  | 18635/24420 [7:46:39<2:23:53,  1.49s/it]
 76%|███████▋  | 18636/24420 [7:46:40<2:23:49,  1.49s/it]
 76%|███████▋  | 18637/24420 [7:46:42<2:23:41,  1.49s/it]
 76%|███████▋  | 18638/24420 [7:46:43<2:23:33,  1.49s/it]
 76%|███████▋  | 18639/24420 [7:46:45<2:23:31,  1.49s/it]
 76%|███████▋  | 18640/24420 [7:46:46<2:23:42,  1.49s/it]
                                                         
{'loss': 0.0714, 'learning_rate': 5e-05, 'epoch': 3.82}

 76%|███████▋  | 18640/24420 [7:46:46<2:23:42,  1.49s/it]
 76%|███████▋  | 18641/24420 [7:46:48<2:23:51,  1.49s/it]
 76%|███████▋  | 18642/24420 [7:46:49<2:23:39,  1.49s/it]
 76%|███████▋  | 18643/24420 [7:46:51<2:23:33,  1.49s/it]
 76%|███████▋  | 18644/24420 [7:46:52<2:23:20,  1.49s/it]
 76%|███████▋  | 18645/24420 [7:46:54<2:23:10,  1.49s/it]
                                                         
{'loss': 0.0842, 'learning_rate': 5e-05, 'epoch': 3.82}

 76%|███████▋  | 18645/24420 [7:46:54<2:23:10,  1.49s/it]
 76%|███████▋  | 18646/24420 [7:46:55<2:23:12,  1.49s/it]
 76%|███████▋  | 18647/24420 [7:46:57<2:23:06,  1.49s/it]
 76%|███████▋  | 18648/24420 [7:46:58<2:23:03,  1.49s/it]
 76%|███████▋  | 18649/24420 [7:47:00<2:22:55,  1.49s/it]
 76%|███████▋  | 18650/24420 [7:47:01<2:23:01,  1.49s/it]
                                                         
{'loss': 0.0691, 'learning_rate': 5e-05, 'epoch': 3.82}

 76%|███████▋  | 18650/24420 [7:47:01<2:23:01,  1.49s/it]
 76%|███████▋  | 18651/24420 [7:47:03<2:22:59,  1.49s/it]
 76%|███████▋  | 18652/24420 [7:47:04<2:22:47,  1.49s/it]
 76%|███████▋  | 18653/24420 [7:47:05<2:22:41,  1.48s/it]
 76%|███████▋  | 18654/24420 [7:47:07<2:22:45,  1.49s/it]
 76%|███████▋  | 18655/24420 [7:47:08<2:23:06,  1.49s/it]
                                                         
{'loss': 0.0805, 'learning_rate': 5e-05, 'epoch': 3.82}

 76%|███████▋  | 18655/24420 [7:47:08<2:23:06,  1.49s/it]
 76%|███████▋  | 18656/24420 [7:47:10<2:23:02,  1.49s/it]
 76%|███████▋  | 18657/24420 [7:47:11<2:22:51,  1.49s/it]
 76%|███████▋  | 18658/24420 [7:47:13<2:23:03,  1.49s/it]
 76%|███████▋  | 18659/24420 [7:47:14<2:23:00,  1.49s/it]
 76%|███████▋  | 18660/24420 [7:47:16<2:22:57,  1.49s/it]
                                                         
{'loss': 0.0747, 'learning_rate': 5e-05, 'epoch': 3.82}

 76%|███████▋  | 18660/24420 [7:47:16<2:22:57,  1.49s/it]
 76%|███████▋  | 18661/24420 [7:47:17<2:22:58,  1.49s/it]
 76%|███████▋  | 18662/24420 [7:47:19<2:23:28,  1.50s/it]
 76%|███████▋  | 18663/24420 [7:47:20<2:23:06,  1.49s/it]
 76%|███████▋  | 18664/24420 [7:47:22<2:23:00,  1.49s/it]
 76%|███████▋  | 18665/24420 [7:47:23<2:22:49,  1.49s/it]
                                                         
{'loss': 0.0738, 'learning_rate': 5e-05, 'epoch': 3.82}

 76%|███████▋  | 18665/24420 [7:47:23<2:22:49,  1.49s/it]
 76%|███████▋  | 18666/24420 [7:47:25<2:22:55,  1.49s/it]
 76%|███████▋  | 18667/24420 [7:47:26<2:22:37,  1.49s/it]
 76%|███████▋  | 18668/24420 [7:47:28<2:22:40,  1.49s/it]
 76%|███████▋  | 18669/24420 [7:47:29<2:22:54,  1.49s/it]
 76%|███████▋  | 18670/24420 [7:47:31<2:23:05,  1.49s/it]
                                                         
{'loss': 0.0654, 'learning_rate': 5e-05, 'epoch': 3.82}

 76%|███████▋  | 18670/24420 [7:47:31<2:23:05,  1.49s/it]
 76%|███████▋  | 18671/24420 [7:47:32<2:23:25,  1.50s/it]
 76%|███████▋  | 18672/24420 [7:47:34<2:23:31,  1.50s/it]
 76%|███████▋  | 18673/24420 [7:47:35<2:23:29,  1.50s/it]
 76%|███████▋  | 18674/24420 [7:47:37<2:23:16,  1.50s/it]
 76%|███████▋  | 18675/24420 [7:47:38<2:23:21,  1.50s/it]
                                                         
{'loss': 0.0731, 'learning_rate': 5e-05, 'epoch': 3.82}

 76%|███████▋  | 18675/24420 [7:47:38<2:23:21,  1.50s/it]
 76%|███████▋  | 18676/24420 [7:47:40<2:23:15,  1.50s/it]
 76%|███████▋  | 18677/24420 [7:47:41<2:23:11,  1.50s/it]
 76%|███████▋  | 18678/24420 [7:47:43<2:23:04,  1.50s/it]
 76%|███████▋  | 18679/24420 [7:47:44<2:22:31,  1.49s/it]
 76%|███████▋  | 18680/24420 [7:47:46<2:22:28,  1.49s/it]
                                                         
{'loss': 0.0809, 'learning_rate': 5e-05, 'epoch': 3.82}

 76%|███████▋  | 18680/24420 [7:47:46<2:22:28,  1.49s/it]
 76%|███████▋  | 18681/24420 [7:47:47<2:22:36,  1.49s/it]
 77%|███████▋  | 18682/24420 [7:47:49<2:22:38,  1.49s/it]
 77%|███████▋  | 18683/24420 [7:47:50<2:22:30,  1.49s/it]
 77%|███████▋  | 18684/24420 [7:47:52<2:22:39,  1.49s/it]
 77%|███████▋  | 18685/24420 [7:47:53<2:22:25,  1.49s/it]
                                                         
{'loss': 0.0739, 'learning_rate': 5e-05, 'epoch': 3.83}

 77%|███████▋  | 18685/24420 [7:47:53<2:22:25,  1.49s/it]
 77%|███████▋  | 18686/24420 [7:47:55<2:22:27,  1.49s/it]
 77%|███████▋  | 18687/24420 [7:47:56<2:22:31,  1.49s/it]
 77%|███████▋  | 18688/24420 [7:47:58<2:22:29,  1.49s/it]
 77%|███████▋  | 18689/24420 [7:47:59<2:22:22,  1.49s/it]
 77%|███████▋  | 18690/24420 [7:48:01<2:22:30,  1.49s/it]
                                                         
{'loss': 0.0777, 'learning_rate': 5e-05, 'epoch': 3.83}

 77%|███████▋  | 18690/24420 [7:48:01<2:22:30,  1.49s/it]
 77%|███████▋  | 18691/24420 [7:48:02<2:22:33,  1.49s/it]
 77%|███████▋  | 18692/24420 [7:48:04<2:22:29,  1.49s/it]
 77%|███████▋  | 18693/24420 [7:48:05<2:22:24,  1.49s/it]
 77%|███████▋  | 18694/24420 [7:48:07<2:22:26,  1.49s/it]
 77%|███████▋  | 18695/24420 [7:48:08<2:22:25,  1.49s/it]
                                                         
{'loss': 0.0858, 'learning_rate': 5e-05, 'epoch': 3.83}

 77%|███████▋  | 18695/24420 [7:48:08<2:22:25,  1.49s/it]
 77%|███████▋  | 18696/24420 [7:48:10<2:22:17,  1.49s/it]
 77%|███████▋  | 18697/24420 [7:48:11<2:22:06,  1.49s/it]
 77%|███████▋  | 18698/24420 [7:48:13<2:22:20,  1.49s/it]
 77%|███████▋  | 18699/24420 [7:48:14<2:22:22,  1.49s/it]
 77%|███████▋  | 18700/24420 [7:48:16<2:22:17,  1.49s/it]
                                                         
{'loss': 0.0858, 'learning_rate': 5e-05, 'epoch': 3.83}

 77%|███████▋  | 18700/24420 [7:48:16<2:22:17,  1.49s/it]
 77%|███████▋  | 18701/24420 [7:48:17<2:21:59,  1.49s/it]
 77%|███████▋  | 18702/24420 [7:48:19<2:22:05,  1.49s/it]
 77%|███████▋  | 18703/24420 [7:48:20<2:22:13,  1.49s/it]
 77%|███████▋  | 18704/24420 [7:48:22<2:22:16,  1.49s/it]
 77%|███████▋  | 18705/24420 [7:48:23<2:22:03,  1.49s/it]
                                                         
{'loss': 0.0748, 'learning_rate': 5e-05, 'epoch': 3.83}

 77%|███████▋  | 18705/24420 [7:48:23<2:22:03,  1.49s/it]
 77%|███████▋  | 18706/24420 [7:48:25<2:21:50,  1.49s/it]
 77%|███████▋  | 18707/24420 [7:48:26<2:21:50,  1.49s/it]
 77%|███████▋  | 18708/24420 [7:48:28<2:21:43,  1.49s/it]
 77%|███████▋  | 18709/24420 [7:48:29<2:21:41,  1.49s/it]
 77%|███████▋  | 18710/24420 [7:48:31<2:21:41,  1.49s/it]
                                                         
{'loss': 0.0654, 'learning_rate': 5e-05, 'epoch': 3.83}

 77%|███████▋  | 18710/24420 [7:48:31<2:21:41,  1.49s/it]
 77%|███████▋  | 18711/24420 [7:48:32<2:21:50,  1.49s/it]
 77%|███████▋  | 18712/24420 [7:48:34<2:21:48,  1.49s/it]
 77%|███████▋  | 18713/24420 [7:48:35<2:21:52,  1.49s/it]
 77%|███████▋  | 18714/24420 [7:48:36<2:21:50,  1.49s/it]
 77%|███████▋  | 18715/24420 [7:48:38<2:22:01,  1.49s/it]
                                                         
{'loss': 0.0676, 'learning_rate': 5e-05, 'epoch': 3.83}

 77%|███████▋  | 18715/24420 [7:48:38<2:22:01,  1.49s/it]
 77%|███████▋  | 18716/24420 [7:48:39<2:21:45,  1.49s/it]
 77%|███████▋  | 18717/24420 [7:48:41<2:21:47,  1.49s/it]
 77%|███████▋  | 18718/24420 [7:48:42<2:21:33,  1.49s/it]
 77%|███████▋  | 18719/24420 [7:48:44<2:21:31,  1.49s/it]
 77%|███████▋  | 18720/24420 [7:48:45<2:21:35,  1.49s/it]
                                                         
{'loss': 0.0795, 'learning_rate': 5e-05, 'epoch': 3.83}

 77%|███████▋  | 18720/24420 [7:48:45<2:21:35,  1.49s/it]
 77%|███████▋  | 18721/24420 [7:48:47<2:23:37,  1.51s/it]
 77%|███████▋  | 18722/24420 [7:48:48<2:23:22,  1.51s/it]
 77%|███████▋  | 18723/24420 [7:48:50<2:23:02,  1.51s/it]
 77%|███████▋  | 18724/24420 [7:48:51<2:22:25,  1.50s/it]
 77%|███████▋  | 18725/24420 [7:48:53<2:22:13,  1.50s/it]
                                                         
{'loss': 0.0884, 'learning_rate': 5e-05, 'epoch': 3.83}

 77%|███████▋  | 18725/24420 [7:48:53<2:22:13,  1.50s/it]
 77%|███████▋  | 18726/24420 [7:48:54<2:21:53,  1.50s/it]
 77%|███████▋  | 18727/24420 [7:48:56<2:21:45,  1.49s/it]
 77%|███████▋  | 18728/24420 [7:48:57<2:21:41,  1.49s/it]
 77%|███████▋  | 18729/24420 [7:48:59<2:21:29,  1.49s/it]
 77%|███████▋  | 18730/24420 [7:49:00<2:21:31,  1.49s/it]
                                                         
{'loss': 0.0892, 'learning_rate': 5e-05, 'epoch': 3.83}

 77%|███████▋  | 18730/24420 [7:49:00<2:21:31,  1.49s/it]
 77%|███████▋  | 18731/24420 [7:49:02<2:21:48,  1.50s/it]
 77%|███████▋  | 18732/24420 [7:49:03<2:21:45,  1.50s/it]
 77%|███████▋  | 18733/24420 [7:49:05<2:22:02,  1.50s/it]
 77%|███████▋  | 18734/24420 [7:49:06<2:21:47,  1.50s/it]
 77%|███████▋  | 18735/24420 [7:49:08<2:21:24,  1.49s/it]
                                                         
{'loss': 0.0753, 'learning_rate': 5e-05, 'epoch': 3.84}

 77%|███████▋  | 18735/24420 [7:49:08<2:21:24,  1.49s/it]
 77%|███████▋  | 18736/24420 [7:49:09<2:21:32,  1.49s/it]
 77%|███████▋  | 18737/24420 [7:49:11<2:21:18,  1.49s/it]
 77%|███████▋  | 18738/24420 [7:49:12<2:21:09,  1.49s/it]
 77%|███████▋  | 18739/24420 [7:49:14<2:20:52,  1.49s/it]
 77%|███████▋  | 18740/24420 [7:49:15<2:20:39,  1.49s/it]
                                                         
{'loss': 0.0764, 'learning_rate': 5e-05, 'epoch': 3.84}

 77%|███████▋  | 18740/24420 [7:49:15<2:20:39,  1.49s/it]
 77%|███████▋  | 18741/24420 [7:49:17<2:21:01,  1.49s/it]
 77%|███████▋  | 18742/24420 [7:49:18<2:20:56,  1.49s/it]
 77%|███████▋  | 18743/24420 [7:49:20<2:21:00,  1.49s/it]
 77%|███████▋  | 18744/24420 [7:49:21<2:20:39,  1.49s/it]
 77%|███████▋  | 18745/24420 [7:49:23<2:20:49,  1.49s/it]
                                                         
{'loss': 0.0835, 'learning_rate': 5e-05, 'epoch': 3.84}

 77%|███████▋  | 18745/24420 [7:49:23<2:20:49,  1.49s/it]
 77%|███████▋  | 18746/24420 [7:49:24<2:22:33,  1.51s/it]
 77%|███████▋  | 18747/24420 [7:49:26<2:21:50,  1.50s/it]
 77%|███████▋  | 18748/24420 [7:49:27<2:21:30,  1.50s/it]
 77%|███████▋  | 18749/24420 [7:49:29<2:21:10,  1.49s/it]
 77%|███████▋  | 18750/24420 [7:49:30<2:20:57,  1.49s/it]
                                                         
{'loss': 0.0719, 'learning_rate': 5e-05, 'epoch': 3.84}

 77%|███████▋  | 18750/24420 [7:49:30<2:20:57,  1.49s/it]
 77%|███████▋  | 18751/24420 [7:49:32<2:20:59,  1.49s/it]
 77%|███████▋  | 18752/24420 [7:49:33<2:20:53,  1.49s/it]
 77%|███████▋  | 18753/24420 [7:49:35<2:20:47,  1.49s/it]
 77%|███████▋  | 18754/24420 [7:49:36<2:20:49,  1.49s/it]
 77%|███████▋  | 18755/24420 [7:49:38<2:20:41,  1.49s/it]
                                                         
{'loss': 0.0987, 'learning_rate': 5e-05, 'epoch': 3.84}

 77%|███████▋  | 18755/24420 [7:49:38<2:20:41,  1.49s/it]
 77%|███████▋  | 18756/24420 [7:49:39<2:20:48,  1.49s/it]
 77%|███████▋  | 18757/24420 [7:49:41<2:20:46,  1.49s/it]
 77%|███████▋  | 18758/24420 [7:49:42<2:20:41,  1.49s/it]
 77%|███████▋  | 18759/24420 [7:49:44<2:20:51,  1.49s/it]
 77%|███████▋  | 18760/24420 [7:49:45<2:20:58,  1.49s/it]
                                                         
{'loss': 0.0813, 'learning_rate': 5e-05, 'epoch': 3.84}

 77%|███████▋  | 18760/24420 [7:49:45<2:20:58,  1.49s/it]
 77%|███████▋  | 18761/24420 [7:49:47<2:20:39,  1.49s/it]
 77%|███████▋  | 18762/24420 [7:49:48<2:20:26,  1.49s/it]
 77%|███████▋  | 18763/24420 [7:49:50<2:20:23,  1.49s/it]
 77%|███████▋  | 18764/24420 [7:49:51<2:20:11,  1.49s/it]
 77%|███████▋  | 18765/24420 [7:49:53<2:20:10,  1.49s/it]
                                                         
{'loss': 0.0778, 'learning_rate': 5e-05, 'epoch': 3.84}

 77%|███████▋  | 18765/24420 [7:49:53<2:20:10,  1.49s/it]
 77%|███████▋  | 18766/24420 [7:49:54<2:20:41,  1.49s/it]
 77%|███████▋  | 18767/24420 [7:49:56<2:20:36,  1.49s/it]
 77%|███████▋  | 18768/24420 [7:49:57<2:20:24,  1.49s/it]
 77%|███████▋  | 18769/24420 [7:49:59<2:20:22,  1.49s/it]
 77%|███████▋  | 18770/24420 [7:50:00<2:20:18,  1.49s/it]
                                                         
{'loss': 0.0818, 'learning_rate': 5e-05, 'epoch': 3.84}

 77%|███████▋  | 18770/24420 [7:50:00<2:20:18,  1.49s/it]
 77%|███████▋  | 18771/24420 [7:50:02<2:20:04,  1.49s/it]
 77%|███████▋  | 18772/24420 [7:50:03<2:20:01,  1.49s/it]
 77%|███████▋  | 18773/24420 [7:50:05<2:20:05,  1.49s/it]
 77%|███████▋  | 18774/24420 [7:50:06<2:19:57,  1.49s/it]
 77%|███████▋  | 18775/24420 [7:50:08<2:20:03,  1.49s/it]
                                                         
{'loss': 0.0833, 'learning_rate': 5e-05, 'epoch': 3.84}

 77%|███████▋  | 18775/24420 [7:50:08<2:20:03,  1.49s/it]
 77%|███████▋  | 18776/24420 [7:50:09<2:20:04,  1.49s/it]
 77%|███████▋  | 18777/24420 [7:50:11<2:20:12,  1.49s/it]
 77%|███████▋  | 18778/24420 [7:50:12<2:19:58,  1.49s/it]
 77%|███████▋  | 18779/24420 [7:50:14<2:20:08,  1.49s/it]
 77%|███████▋  | 18780/24420 [7:50:15<2:20:03,  1.49s/it]
                                                         
{'loss': 0.0792, 'learning_rate': 5e-05, 'epoch': 3.85}

 77%|███████▋  | 18780/24420 [7:50:15<2:20:03,  1.49s/it]
 77%|███████▋  | 18781/24420 [7:50:16<2:20:10,  1.49s/it]
 77%|███████▋  | 18782/24420 [7:50:18<2:20:17,  1.49s/it]
 77%|███████▋  | 18783/24420 [7:50:19<2:20:02,  1.49s/it]
 77%|███████▋  | 18784/24420 [7:50:21<2:20:13,  1.49s/it]
 77%|███████▋  | 18785/24420 [7:50:22<2:20:19,  1.49s/it]
                                                         
{'loss': 0.0724, 'learning_rate': 5e-05, 'epoch': 3.85}

 77%|███████▋  | 18785/24420 [7:50:22<2:20:19,  1.49s/it]
 77%|███████▋  | 18786/24420 [7:50:24<2:20:15,  1.49s/it]
 77%|███████▋  | 18787/24420 [7:50:25<2:20:01,  1.49s/it]
 77%|███████▋  | 18788/24420 [7:50:27<2:19:44,  1.49s/it]
 77%|███████▋  | 18789/24420 [7:50:28<2:20:01,  1.49s/it]
 77%|███████▋  | 18790/24420 [7:50:30<2:20:22,  1.50s/it]
                                                         
{'loss': 0.0926, 'learning_rate': 5e-05, 'epoch': 3.85}

 77%|███████▋  | 18790/24420 [7:50:30<2:20:22,  1.50s/it]
 77%|███████▋  | 18791/24420 [7:50:31<2:20:00,  1.49s/it]
 77%|███████▋  | 18792/24420 [7:50:33<2:19:49,  1.49s/it]
 77%|███████▋  | 18793/24420 [7:50:34<2:20:10,  1.49s/it]
 77%|███████▋  | 18794/24420 [7:50:36<2:19:48,  1.49s/it]
 77%|███████▋  | 18795/24420 [7:50:37<2:19:47,  1.49s/it]
                                                         
{'loss': 0.0869, 'learning_rate': 5e-05, 'epoch': 3.85}

 77%|███████▋  | 18795/24420 [7:50:37<2:19:47,  1.49s/it]
 77%|███████▋  | 18796/24420 [7:50:39<2:19:47,  1.49s/it]
 77%|███████▋  | 18797/24420 [7:50:40<2:19:44,  1.49s/it]
 77%|███████▋  | 18798/24420 [7:50:42<2:19:46,  1.49s/it]
 77%|███████▋  | 18799/24420 [7:50:43<2:19:50,  1.49s/it]
 77%|███████▋  | 18800/24420 [7:50:45<2:19:31,  1.49s/it]
                                                         
{'loss': 0.0777, 'learning_rate': 5e-05, 'epoch': 3.85}

 77%|███████▋  | 18800/24420 [7:50:45<2:19:31,  1.49s/it]
 77%|███████▋  | 18801/24420 [7:50:50<3:49:05,  2.45s/it]
 77%|███████▋  | 18802/24420 [7:50:51<3:21:59,  2.16s/it]
 77%|███████▋  | 18803/24420 [7:50:52<3:03:06,  1.96s/it]
 77%|███████▋  | 18804/24420 [7:50:54<2:49:57,  1.82s/it]
 77%|███████▋  | 18805/24420 [7:50:55<2:40:30,  1.72s/it]
                                                         
{'loss': 0.0822, 'learning_rate': 5e-05, 'epoch': 3.85}

 77%|███████▋  | 18805/24420 [7:50:55<2:40:30,  1.72s/it]
 77%|███████▋  | 18806/24420 [7:50:57<2:34:03,  1.65s/it]
 77%|███████▋  | 18807/24420 [7:50:58<2:29:32,  1.60s/it]
 77%|███████▋  | 18808/24420 [7:51:00<2:26:39,  1.57s/it]
 77%|███████▋  | 18809/24420 [7:51:01<2:24:24,  1.54s/it]
 77%|███████▋  | 18810/24420 [7:51:03<2:22:32,  1.52s/it]
                                                         
{'loss': 0.0865, 'learning_rate': 5e-05, 'epoch': 3.85}

 77%|███████▋  | 18810/24420 [7:51:03<2:22:32,  1.52s/it]
 77%|███████▋  | 18811/24420 [7:51:04<2:21:13,  1.51s/it]
 77%|███████▋  | 18812/24420 [7:51:06<2:20:25,  1.50s/it]
 77%|███████▋  | 18813/24420 [7:51:07<2:19:49,  1.50s/it]
 77%|███████▋  | 18814/24420 [7:51:09<2:19:34,  1.49s/it]
 77%|███████▋  | 18815/24420 [7:51:10<2:19:21,  1.49s/it]
                                                         
{'loss': 0.0696, 'learning_rate': 5e-05, 'epoch': 3.85}

 77%|███████▋  | 18815/24420 [7:51:10<2:19:21,  1.49s/it]
 77%|███████▋  | 18816/24420 [7:51:12<2:19:14,  1.49s/it]
 77%|███████▋  | 18817/24420 [7:51:13<2:18:57,  1.49s/it]
 77%|███████▋  | 18818/24420 [7:51:15<2:19:02,  1.49s/it]
 77%|███████▋  | 18819/24420 [7:51:16<2:19:11,  1.49s/it]
 77%|███████▋  | 18820/24420 [7:51:18<2:19:19,  1.49s/it]
                                                         
{'loss': 0.075, 'learning_rate': 5e-05, 'epoch': 3.85}

 77%|███████▋  | 18820/24420 [7:51:18<2:19:19,  1.49s/it]
 77%|███████▋  | 18821/24420 [7:51:19<2:19:29,  1.49s/it]
 77%|███████▋  | 18822/24420 [7:51:21<2:19:09,  1.49s/it]
 77%|███████▋  | 18823/24420 [7:51:22<2:19:04,  1.49s/it]
 77%|███████▋  | 18824/24420 [7:51:24<2:18:45,  1.49s/it]
 77%|███████▋  | 18825/24420 [7:51:25<2:18:55,  1.49s/it]
                                                         
{'loss': 0.0867, 'learning_rate': 5e-05, 'epoch': 3.85}

 77%|███████▋  | 18825/24420 [7:51:25<2:18:55,  1.49s/it]
 77%|███████▋  | 18826/24420 [7:51:27<2:18:47,  1.49s/it]
 77%|███████▋  | 18827/24420 [7:51:28<2:18:36,  1.49s/it]
 77%|███████▋  | 18828/24420 [7:51:30<2:18:50,  1.49s/it]
 77%|███████▋  | 18829/24420 [7:51:31<2:18:46,  1.49s/it]
 77%|███████▋  | 18830/24420 [7:51:33<2:18:34,  1.49s/it]
                                                         
{'loss': 0.062, 'learning_rate': 5e-05, 'epoch': 3.86}

 77%|███████▋  | 18830/24420 [7:51:33<2:18:34,  1.49s/it]
 77%|███████▋  | 18831/24420 [7:51:34<2:18:53,  1.49s/it]
 77%|███████▋  | 18832/24420 [7:51:36<2:18:38,  1.49s/it]
 77%|███████▋  | 18833/24420 [7:51:37<2:18:26,  1.49s/it]
 77%|███████▋  | 18834/24420 [7:51:39<2:18:30,  1.49s/it]
 77%|███████▋  | 18835/24420 [7:51:40<2:18:27,  1.49s/it]
                                                         
{'loss': 0.0684, 'learning_rate': 5e-05, 'epoch': 3.86}

 77%|███████▋  | 18835/24420 [7:51:40<2:18:27,  1.49s/it]
 77%|███████▋  | 18836/24420 [7:51:42<2:18:28,  1.49s/it]
 77%|███████▋  | 18837/24420 [7:51:43<2:18:19,  1.49s/it]
 77%|███████▋  | 18838/24420 [7:51:45<2:18:36,  1.49s/it]
 77%|███████▋  | 18839/24420 [7:51:46<2:18:21,  1.49s/it]
 77%|███████▋  | 18840/24420 [7:51:48<2:18:26,  1.49s/it]
                                                         
{'loss': 0.0711, 'learning_rate': 5e-05, 'epoch': 3.86}

 77%|███████▋  | 18840/24420 [7:51:48<2:18:26,  1.49s/it]
 77%|███████▋  | 18841/24420 [7:51:49<2:18:24,  1.49s/it]
 77%|███████▋  | 18842/24420 [7:51:51<2:18:11,  1.49s/it]
 77%|███████▋  | 18843/24420 [7:51:52<2:18:04,  1.49s/it]
 77%|███████▋  | 18844/24420 [7:51:53<2:18:14,  1.49s/it]
 77%|███████▋  | 18845/24420 [7:51:55<2:18:10,  1.49s/it]
                                                         
{'loss': 0.0896, 'learning_rate': 5e-05, 'epoch': 3.86}

 77%|███████▋  | 18845/24420 [7:51:55<2:18:10,  1.49s/it]
 77%|███████▋  | 18846/24420 [7:51:56<2:18:07,  1.49s/it]
 77%|███████▋  | 18847/24420 [7:51:58<2:18:09,  1.49s/it]
 77%|███████▋  | 18848/24420 [7:51:59<2:18:04,  1.49s/it]
 77%|███████▋  | 18849/24420 [7:52:01<2:17:56,  1.49s/it]
 77%|███████▋  | 18850/24420 [7:52:02<2:17:51,  1.48s/it]
                                                         
{'loss': 0.0757, 'learning_rate': 5e-05, 'epoch': 3.86}

 77%|███████▋  | 18850/24420 [7:52:02<2:17:51,  1.48s/it]
 77%|███████▋  | 18851/24420 [7:52:04<2:17:57,  1.49s/it]
 77%|███████▋  | 18852/24420 [7:52:05<2:17:55,  1.49s/it]
 77%|███████▋  | 18853/24420 [7:52:07<2:17:51,  1.49s/it]
 77%|███████▋  | 18854/24420 [7:52:08<2:17:40,  1.48s/it]
 77%|███████▋  | 18855/24420 [7:52:10<2:18:01,  1.49s/it]
                                                         
{'loss': 0.0863, 'learning_rate': 5e-05, 'epoch': 3.86}

 77%|███████▋  | 18855/24420 [7:52:10<2:18:01,  1.49s/it]
 77%|███████▋  | 18856/24420 [7:52:11<2:18:07,  1.49s/it]
 77%|███████▋  | 18857/24420 [7:52:13<2:17:48,  1.49s/it]
 77%|███████▋  | 18858/24420 [7:52:14<2:17:52,  1.49s/it]
 77%|███████▋  | 18859/24420 [7:52:16<2:17:41,  1.49s/it]
 77%|███████▋  | 18860/24420 [7:52:17<2:17:53,  1.49s/it]
                                                         
{'loss': 0.0745, 'learning_rate': 5e-05, 'epoch': 3.86}

 77%|███████▋  | 18860/24420 [7:52:17<2:17:53,  1.49s/it]
 77%|███████▋  | 18861/24420 [7:52:19<2:17:50,  1.49s/it]
 77%|███████▋  | 18862/24420 [7:52:20<2:17:53,  1.49s/it]
 77%|███████▋  | 18863/24420 [7:52:22<2:17:51,  1.49s/it]
 77%|███████▋  | 18864/24420 [7:52:23<2:18:03,  1.49s/it]
 77%|███████▋  | 18865/24420 [7:52:25<2:17:42,  1.49s/it]
                                                         
{'loss': 0.1053, 'learning_rate': 5e-05, 'epoch': 3.86}

 77%|███████▋  | 18865/24420 [7:52:25<2:17:42,  1.49s/it]
 77%|███████▋  | 18866/24420 [7:52:26<2:17:36,  1.49s/it]
 77%|███████▋  | 18867/24420 [7:52:28<2:17:34,  1.49s/it]
 77%|███████▋  | 18868/24420 [7:52:29<2:17:38,  1.49s/it]
 77%|███████▋  | 18869/24420 [7:52:31<2:17:39,  1.49s/it]
 77%|███████▋  | 18870/24420 [7:52:32<2:17:36,  1.49s/it]
                                                         
{'loss': 0.0725, 'learning_rate': 5e-05, 'epoch': 3.86}

 77%|███████▋  | 18870/24420 [7:52:32<2:17:36,  1.49s/it]
 77%|███████▋  | 18871/24420 [7:52:34<2:17:59,  1.49s/it]
 77%|███████▋  | 18872/24420 [7:52:35<2:17:49,  1.49s/it]
 77%|███████▋  | 18873/24420 [7:52:37<2:17:43,  1.49s/it]
 77%|███████▋  | 18874/24420 [7:52:38<2:17:36,  1.49s/it]
 77%|███████▋  | 18875/24420 [7:52:40<2:17:29,  1.49s/it]
                                                         
{'loss': 0.076, 'learning_rate': 5e-05, 'epoch': 3.86}

 77%|███████▋  | 18875/24420 [7:52:40<2:17:29,  1.49s/it]
 77%|███████▋  | 18876/24420 [7:52:41<2:17:34,  1.49s/it]
 77%|███████▋  | 18877/24420 [7:52:43<2:17:30,  1.49s/it]
 77%|███████▋  | 18878/24420 [7:52:44<2:17:40,  1.49s/it]
 77%|███████▋  | 18879/24420 [7:52:46<2:17:49,  1.49s/it]
 77%|███████▋  | 18880/24420 [7:52:47<2:18:02,  1.50s/it]
                                                         
{'loss': 0.0976, 'learning_rate': 5e-05, 'epoch': 3.87}

 77%|███████▋  | 18880/24420 [7:52:47<2:18:02,  1.50s/it]
 77%|███████▋  | 18881/24420 [7:52:49<2:17:59,  1.49s/it]
 77%|███████▋  | 18882/24420 [7:52:50<2:17:55,  1.49s/it]
 77%|███████▋  | 18883/24420 [7:52:52<2:17:43,  1.49s/it]
 77%|███████▋  | 18884/24420 [7:52:53<2:17:38,  1.49s/it]
 77%|███████▋  | 18885/24420 [7:52:55<2:17:26,  1.49s/it]
                                                         
{'loss': 0.0709, 'learning_rate': 5e-05, 'epoch': 3.87}

 77%|███████▋  | 18885/24420 [7:52:55<2:17:26,  1.49s/it]
 77%|███████▋  | 18886/24420 [7:52:56<2:17:20,  1.49s/it]
 77%|███████▋  | 18887/24420 [7:52:57<2:17:04,  1.49s/it]
 77%|███████▋  | 18888/24420 [7:52:59<2:17:09,  1.49s/it]
 77%|███████▋  | 18889/24420 [7:53:00<2:16:54,  1.49s/it]
 77%|███████▋  | 18890/24420 [7:53:02<2:17:03,  1.49s/it]
                                                         
{'loss': 0.0763, 'learning_rate': 5e-05, 'epoch': 3.87}

 77%|███████▋  | 18890/24420 [7:53:02<2:17:03,  1.49s/it]
 77%|███████▋  | 18891/24420 [7:53:03<2:16:57,  1.49s/it]
 77%|███████▋  | 18892/24420 [7:53:05<2:16:49,  1.49s/it]
 77%|███████▋  | 18893/24420 [7:53:06<2:16:56,  1.49s/it]
 77%|███████▋  | 18894/24420 [7:53:08<2:17:10,  1.49s/it]
 77%|███████▋  | 18895/24420 [7:53:09<2:16:57,  1.49s/it]
                                                         
{'loss': 0.0818, 'learning_rate': 5e-05, 'epoch': 3.87}

 77%|███████▋  | 18895/24420 [7:53:09<2:16:57,  1.49s/it]
 77%|███████▋  | 18896/24420 [7:53:11<2:17:17,  1.49s/it]
 77%|███████▋  | 18897/24420 [7:53:12<2:17:01,  1.49s/it]
 77%|███████▋  | 18898/24420 [7:53:14<2:17:05,  1.49s/it]
 77%|███████▋  | 18899/24420 [7:53:15<2:17:05,  1.49s/it]
 77%|███████▋  | 18900/24420 [7:53:17<2:17:05,  1.49s/it]
                                                         
{'loss': 0.0818, 'learning_rate': 5e-05, 'epoch': 3.87}

 77%|███████▋  | 18900/24420 [7:53:17<2:17:05,  1.49s/it]
 77%|███████▋  | 18901/24420 [7:53:18<2:17:02,  1.49s/it]
 77%|███████▋  | 18902/24420 [7:53:20<2:17:13,  1.49s/it]
 77%|███████▋  | 18903/24420 [7:53:21<2:17:21,  1.49s/it]
 77%|███████▋  | 18904/24420 [7:53:23<2:17:03,  1.49s/it]
 77%|███████▋  | 18905/24420 [7:53:24<2:16:53,  1.49s/it]
                                                         
{'loss': 0.0805, 'learning_rate': 5e-05, 'epoch': 3.87}

 77%|███████▋  | 18905/24420 [7:53:24<2:16:53,  1.49s/it]
 77%|███████▋  | 18906/24420 [7:53:26<2:17:09,  1.49s/it]
 77%|███████▋  | 18907/24420 [7:53:27<2:16:57,  1.49s/it]
 77%|███████▋  | 18908/24420 [7:53:29<2:17:02,  1.49s/it]
 77%|███████▋  | 18909/24420 [7:53:30<2:17:10,  1.49s/it]
 77%|███████▋  | 18910/24420 [7:53:32<2:16:51,  1.49s/it]
                                                         
{'loss': 0.063, 'learning_rate': 5e-05, 'epoch': 3.87}

 77%|███████▋  | 18910/24420 [7:53:32<2:16:51,  1.49s/it]
 77%|███████▋  | 18911/24420 [7:53:33<2:17:02,  1.49s/it]
 77%|███████▋  | 18912/24420 [7:53:35<2:17:06,  1.49s/it]
 77%|███████▋  | 18913/24420 [7:53:36<2:16:49,  1.49s/it]
 77%|███████▋  | 18914/24420 [7:53:38<2:17:01,  1.49s/it]
 77%|███████▋  | 18915/24420 [7:53:39<2:16:57,  1.49s/it]
                                                         
{'loss': 0.1042, 'learning_rate': 5e-05, 'epoch': 3.87}

 77%|███████▋  | 18915/24420 [7:53:39<2:16:57,  1.49s/it]
 77%|███████▋  | 18916/24420 [7:53:41<2:16:54,  1.49s/it]
 77%|███████▋  | 18917/24420 [7:53:42<2:17:05,  1.49s/it]
 77%|███████▋  | 18918/24420 [7:53:44<2:17:12,  1.50s/it]
 77%|███████▋  | 18919/24420 [7:53:45<2:18:52,  1.51s/it]
 77%|███████▋  | 18920/24420 [7:53:47<2:17:44,  1.50s/it]
                                                         
{'loss': 0.0941, 'learning_rate': 5e-05, 'epoch': 3.87}

 77%|███████▋  | 18920/24420 [7:53:47<2:17:44,  1.50s/it]
 77%|███████▋  | 18921/24420 [7:53:48<2:17:38,  1.50s/it]
 77%|███████▋  | 18922/24420 [7:53:50<2:17:19,  1.50s/it]
 77%|███████▋  | 18923/24420 [7:53:51<2:16:59,  1.50s/it]
 77%|███████▋  | 18924/24420 [7:53:53<2:16:49,  1.49s/it]
 77%|███████▋  | 18925/24420 [7:53:54<2:16:35,  1.49s/it]
                                                         
{'loss': 0.0733, 'learning_rate': 5e-05, 'epoch': 3.87}

 77%|███████▋  | 18925/24420 [7:53:54<2:16:35,  1.49s/it]
 78%|███████▊  | 18926/24420 [7:53:56<2:18:05,  1.51s/it]
 78%|███████▊  | 18927/24420 [7:53:57<2:17:42,  1.50s/it]
 78%|███████▊  | 18928/24420 [7:53:59<2:17:07,  1.50s/it]
 78%|███████▊  | 18929/24420 [7:54:00<2:16:52,  1.50s/it]
 78%|███████▊  | 18930/24420 [7:54:02<2:16:36,  1.49s/it]
                                                         
{'loss': 0.0827, 'learning_rate': 5e-05, 'epoch': 3.88}

 78%|███████▊  | 18930/24420 [7:54:02<2:16:36,  1.49s/it]
 78%|███████▊  | 18931/24420 [7:54:03<2:16:21,  1.49s/it]
 78%|███████▊  | 18932/24420 [7:54:05<2:15:56,  1.49s/it]
 78%|███████▊  | 18933/24420 [7:54:06<2:15:58,  1.49s/it]
 78%|███████▊  | 18934/24420 [7:54:08<2:15:54,  1.49s/it]
 78%|███████▊  | 18935/24420 [7:54:09<2:15:56,  1.49s/it]
                                                         
{'loss': 0.0753, 'learning_rate': 5e-05, 'epoch': 3.88}

 78%|███████▊  | 18935/24420 [7:54:09<2:15:56,  1.49s/it]
 78%|███████▊  | 18936/24420 [7:54:11<2:15:54,  1.49s/it]
 78%|███████▊  | 18937/24420 [7:54:12<2:16:21,  1.49s/it]
 78%|███████▊  | 18938/24420 [7:54:14<2:16:18,  1.49s/it]
 78%|███████▊  | 18939/24420 [7:54:15<2:16:27,  1.49s/it]
 78%|███████▊  | 18940/24420 [7:54:17<2:16:23,  1.49s/it]
                                                         
{'loss': 0.0662, 'learning_rate': 5e-05, 'epoch': 3.88}

 78%|███████▊  | 18940/24420 [7:54:17<2:16:23,  1.49s/it]
 78%|███████▊  | 18941/24420 [7:54:18<2:16:05,  1.49s/it]
 78%|███████▊  | 18942/24420 [7:54:20<2:16:03,  1.49s/it]
 78%|███████▊  | 18943/24420 [7:54:21<2:16:08,  1.49s/it]
 78%|███████▊  | 18944/24420 [7:54:23<2:15:59,  1.49s/it]
 78%|███████▊  | 18945/24420 [7:54:24<2:16:02,  1.49s/it]
                                                         
{'loss': 0.0826, 'learning_rate': 5e-05, 'epoch': 3.88}

 78%|███████▊  | 18945/24420 [7:54:24<2:16:02,  1.49s/it]
 78%|███████▊  | 18946/24420 [7:54:26<2:16:16,  1.49s/it]
 78%|███████▊  | 18947/24420 [7:54:27<2:15:59,  1.49s/it]
 78%|███████▊  | 18948/24420 [7:54:29<2:15:45,  1.49s/it]
 78%|███████▊  | 18949/24420 [7:54:30<2:15:47,  1.49s/it]
 78%|███████▊  | 18950/24420 [7:54:31<2:15:36,  1.49s/it]
                                                         
{'loss': 0.0724, 'learning_rate': 5e-05, 'epoch': 3.88}

 78%|███████▊  | 18950/24420 [7:54:31<2:15:36,  1.49s/it]
 78%|███████▊  | 18951/24420 [7:54:33<2:15:30,  1.49s/it]
 78%|███████▊  | 18952/24420 [7:54:34<2:15:27,  1.49s/it]
 78%|███████▊  | 18953/24420 [7:54:36<2:15:20,  1.49s/it]
 78%|███████▊  | 18954/24420 [7:54:37<2:15:18,  1.49s/it]
 78%|███████▊  | 18955/24420 [7:54:39<2:15:33,  1.49s/it]
                                                         
{'loss': 0.0816, 'learning_rate': 5e-05, 'epoch': 3.88}

 78%|███████▊  | 18955/24420 [7:54:39<2:15:33,  1.49s/it]
 78%|███████▊  | 18956/24420 [7:54:40<2:15:25,  1.49s/it]
 78%|███████▊  | 18957/24420 [7:54:42<2:15:29,  1.49s/it]
 78%|███████▊  | 18958/24420 [7:54:43<2:15:27,  1.49s/it]
 78%|███████▊  | 18959/24420 [7:54:45<2:15:33,  1.49s/it]
 78%|███████▊  | 18960/24420 [7:54:46<2:15:44,  1.49s/it]
                                                         
{'loss': 0.0849, 'learning_rate': 5e-05, 'epoch': 3.88}

 78%|███████▊  | 18960/24420 [7:54:46<2:15:44,  1.49s/it]
 78%|███████▊  | 18961/24420 [7:54:48<2:15:36,  1.49s/it]
 78%|███████▊  | 18962/24420 [7:54:49<2:15:21,  1.49s/it]
 78%|███████▊  | 18963/24420 [7:54:51<2:14:59,  1.48s/it]
 78%|███████▊  | 18964/24420 [7:54:52<2:15:01,  1.48s/it]
 78%|███████▊  | 18965/24420 [7:54:54<2:14:46,  1.48s/it]
                                                         
{'loss': 0.0699, 'learning_rate': 5e-05, 'epoch': 3.88}

 78%|███████▊  | 18965/24420 [7:54:54<2:14:46,  1.48s/it]
 78%|███████▊  | 18966/24420 [7:54:55<2:14:47,  1.48s/it]
 78%|███████▊  | 18967/24420 [7:54:57<2:15:05,  1.49s/it]
 78%|███████▊  | 18968/24420 [7:54:58<2:15:07,  1.49s/it]
 78%|███████▊  | 18969/24420 [7:55:00<2:15:13,  1.49s/it]
 78%|███████▊  | 18970/24420 [7:55:01<2:15:02,  1.49s/it]
                                                         
{'loss': 0.0859, 'learning_rate': 5e-05, 'epoch': 3.88}

 78%|███████▊  | 18970/24420 [7:55:01<2:15:02,  1.49s/it]
 78%|███████▊  | 18971/24420 [7:55:03<2:15:31,  1.49s/it]
 78%|███████▊  | 18972/24420 [7:55:04<2:15:14,  1.49s/it]
 78%|███████▊  | 18973/24420 [7:55:06<2:15:07,  1.49s/it]
 78%|███████▊  | 18974/24420 [7:55:07<2:14:58,  1.49s/it]
 78%|███████▊  | 18975/24420 [7:55:09<2:15:00,  1.49s/it]
                                                         
{'loss': 0.0698, 'learning_rate': 5e-05, 'epoch': 3.89}

 78%|███████▊  | 18975/24420 [7:55:09<2:15:00,  1.49s/it]
 78%|███████▊  | 18976/24420 [7:55:10<2:15:15,  1.49s/it]
 78%|███████▊  | 18977/24420 [7:55:12<2:15:05,  1.49s/it]
 78%|███████▊  | 18978/24420 [7:55:13<2:14:50,  1.49s/it]
 78%|███████▊  | 18979/24420 [7:55:15<2:14:44,  1.49s/it]
 78%|███████▊  | 18980/24420 [7:55:16<2:14:57,  1.49s/it]
                                                         
{'loss': 0.0748, 'learning_rate': 5e-05, 'epoch': 3.89}

 78%|███████▊  | 18980/24420 [7:55:16<2:14:57,  1.49s/it]
 78%|███████▊  | 18981/24420 [7:55:18<2:14:56,  1.49s/it]
 78%|███████▊  | 18982/24420 [7:55:19<2:15:05,  1.49s/it]
 78%|███████▊  | 18983/24420 [7:55:21<2:14:59,  1.49s/it]
 78%|███████▊  | 18984/24420 [7:55:22<2:14:58,  1.49s/it]
 78%|███████▊  | 18985/24420 [7:55:24<2:15:07,  1.49s/it]
                                                         
{'loss': 0.0696, 'learning_rate': 5e-05, 'epoch': 3.89}

 78%|███████▊  | 18985/24420 [7:55:24<2:15:07,  1.49s/it]
 78%|███████▊  | 18986/24420 [7:55:25<2:15:03,  1.49s/it]
 78%|███████▊  | 18987/24420 [7:55:27<2:14:44,  1.49s/it]
 78%|███████▊  | 18988/24420 [7:55:28<2:14:46,  1.49s/it]
 78%|███████▊  | 18989/24420 [7:55:30<2:14:39,  1.49s/it]
 78%|███████▊  | 18990/24420 [7:55:31<2:14:41,  1.49s/it]
                                                         
{'loss': 0.0622, 'learning_rate': 5e-05, 'epoch': 3.89}

 78%|███████▊  | 18990/24420 [7:55:31<2:14:41,  1.49s/it]
 78%|███████▊  | 18991/24420 [7:55:32<2:14:45,  1.49s/it]
 78%|███████▊  | 18992/24420 [7:55:34<2:14:38,  1.49s/it]
 78%|███████▊  | 18993/24420 [7:55:35<2:14:34,  1.49s/it]
 78%|███████▊  | 18994/24420 [7:55:37<2:14:44,  1.49s/it]
 78%|███████▊  | 18995/24420 [7:55:38<2:14:30,  1.49s/it]
                                                         
{'loss': 0.0747, 'learning_rate': 5e-05, 'epoch': 3.89}

 78%|███████▊  | 18995/24420 [7:55:38<2:14:30,  1.49s/it]
 78%|███████▊  | 18996/24420 [7:55:40<2:14:31,  1.49s/it]
 78%|███████▊  | 18997/24420 [7:55:41<2:14:33,  1.49s/it]
 78%|███████▊  | 18998/24420 [7:55:43<2:14:34,  1.49s/it]
 78%|███████▊  | 18999/24420 [7:55:44<2:14:39,  1.49s/it]
 78%|███████▊  | 19000/24420 [7:55:46<2:14:49,  1.49s/it]
                                                         
{'loss': 0.0759, 'learning_rate': 5e-05, 'epoch': 3.89}

 78%|███████▊  | 19000/24420 [7:55:46<2:14:49,  1.49s/it]
 78%|███████▊  | 19001/24420 [7:55:51<3:39:32,  2.43s/it]
 78%|███████▊  | 19002/24420 [7:55:52<3:13:48,  2.15s/it]
 78%|███████▊  | 19003/24420 [7:55:54<2:56:07,  1.95s/it]
 78%|███████▊  | 19004/24420 [7:55:55<2:43:44,  1.81s/it]
 78%|███████▊  | 19005/24420 [7:55:56<2:34:48,  1.72s/it]
                                                         
{'loss': 0.0798, 'learning_rate': 5e-05, 'epoch': 3.89}

 78%|███████▊  | 19005/24420 [7:55:56<2:34:48,  1.72s/it]
 78%|███████▊  | 19006/24420 [7:55:58<2:28:46,  1.65s/it]
 78%|███████▊  | 19007/24420 [7:55:59<2:24:12,  1.60s/it]
 78%|███████▊  | 19008/24420 [7:56:01<2:21:12,  1.57s/it]
 78%|███████▊  | 19009/24420 [7:56:02<2:19:03,  1.54s/it]
 78%|███████▊  | 19010/24420 [7:56:04<2:18:03,  1.53s/it]
                                                         
{'loss': 0.0846, 'learning_rate': 5e-05, 'epoch': 3.89}

 78%|███████▊  | 19010/24420 [7:56:04<2:18:03,  1.53s/it]
 78%|███████▊  | 19011/24420 [7:56:05<2:17:00,  1.52s/it]
 78%|███████▊  | 19012/24420 [7:56:07<2:16:10,  1.51s/it]
 78%|███████▊  | 19013/24420 [7:56:08<2:15:37,  1.51s/it]
 78%|███████▊  | 19014/24420 [7:56:10<2:15:40,  1.51s/it]
 78%|███████▊  | 19015/24420 [7:56:11<2:15:07,  1.50s/it]
                                                         
{'loss': 0.0788, 'learning_rate': 5e-05, 'epoch': 3.89}

 78%|███████▊  | 19015/24420 [7:56:11<2:15:07,  1.50s/it]
 78%|███████▊  | 19016/24420 [7:56:13<2:14:47,  1.50s/it]
 78%|███████▊  | 19017/24420 [7:56:14<2:14:20,  1.49s/it]
 78%|███████▊  | 19018/24420 [7:56:16<2:14:17,  1.49s/it]
 78%|███████▊  | 19019/24420 [7:56:17<2:14:09,  1.49s/it]
 78%|███████▊  | 19020/24420 [7:56:19<2:13:47,  1.49s/it]
                                                         
{'loss': 0.0884, 'learning_rate': 5e-05, 'epoch': 3.89}

 78%|███████▊  | 19020/24420 [7:56:19<2:13:47,  1.49s/it]
 78%|███████▊  | 19021/24420 [7:56:20<2:13:43,  1.49s/it]
 78%|███████▊  | 19022/24420 [7:56:22<2:13:39,  1.49s/it]
 78%|███████▊  | 19023/24420 [7:56:23<2:13:35,  1.49s/it]
 78%|███████▊  | 19024/24420 [7:56:25<2:13:26,  1.48s/it]
 78%|███████▊  | 19025/24420 [7:56:26<2:13:23,  1.48s/it]
                                                         
{'loss': 0.0651, 'learning_rate': 5e-05, 'epoch': 3.9}

 78%|███████▊  | 19025/24420 [7:56:26<2:13:23,  1.48s/it]
 78%|███████▊  | 19026/24420 [7:56:28<2:13:24,  1.48s/it]
 78%|███████▊  | 19027/24420 [7:56:29<2:13:33,  1.49s/it]
 78%|███████▊  | 19028/24420 [7:56:31<2:13:25,  1.48s/it]
 78%|███████▊  | 19029/24420 [7:56:32<2:13:11,  1.48s/it]
 78%|███████▊  | 19030/24420 [7:56:34<2:13:08,  1.48s/it]
                                                         
{'loss': 0.0792, 'learning_rate': 5e-05, 'epoch': 3.9}

 78%|███████▊  | 19030/24420 [7:56:34<2:13:08,  1.48s/it]
 78%|███████▊  | 19031/24420 [7:56:35<2:13:10,  1.48s/it]
 78%|███████▊  | 19032/24420 [7:56:37<2:13:20,  1.48s/it]
 78%|███████▊  | 19033/24420 [7:56:38<2:13:11,  1.48s/it]
 78%|███████▊  | 19034/24420 [7:56:40<2:13:03,  1.48s/it]
 78%|███████▊  | 19035/24420 [7:56:41<2:13:07,  1.48s/it]
                                                         
{'loss': 0.061, 'learning_rate': 5e-05, 'epoch': 3.9}

 78%|███████▊  | 19035/24420 [7:56:41<2:13:07,  1.48s/it]
 78%|███████▊  | 19036/24420 [7:56:43<2:13:28,  1.49s/it]
 78%|███████▊  | 19037/24420 [7:56:44<2:13:16,  1.49s/it]
 78%|███████▊  | 19038/24420 [7:56:46<2:13:20,  1.49s/it]
 78%|███████▊  | 19039/24420 [7:56:47<2:13:20,  1.49s/it]
 78%|███████▊  | 19040/24420 [7:56:49<2:13:01,  1.48s/it]
                                                         
{'loss': 0.0827, 'learning_rate': 5e-05, 'epoch': 3.9}

 78%|███████▊  | 19040/24420 [7:56:49<2:13:01,  1.48s/it]
 78%|███████▊  | 19041/24420 [7:56:50<2:13:02,  1.48s/it]
 78%|███████▊  | 19042/24420 [7:56:51<2:12:54,  1.48s/it]
 78%|███████▊  | 19043/24420 [7:56:53<2:13:04,  1.48s/it]
 78%|███████▊  | 19044/24420 [7:56:55<2:21:34,  1.58s/it]
 78%|███████▊  | 19045/24420 [7:56:56<2:19:09,  1.55s/it]
                                                         
{'loss': 0.1031, 'learning_rate': 5e-05, 'epoch': 3.9}

 78%|███████▊  | 19045/24420 [7:56:56<2:19:09,  1.55s/it]
 78%|███████▊  | 19046/24420 [7:56:58<2:17:29,  1.54s/it]
 78%|███████▊  | 19047/24420 [7:56:59<2:16:08,  1.52s/it]
 78%|███████▊  | 19048/24420 [7:57:01<2:15:29,  1.51s/it]
 78%|███████▊  | 19049/24420 [7:57:02<2:14:44,  1.51s/it]
 78%|███████▊  | 19050/24420 [7:57:04<2:14:12,  1.50s/it]
                                                         
{'loss': 0.0828, 'learning_rate': 5e-05, 'epoch': 3.9}

 78%|███████▊  | 19050/24420 [7:57:04<2:14:12,  1.50s/it]
 78%|███████▊  | 19051/24420 [7:57:05<2:13:46,  1.49s/it]
 78%|███████▊  | 19052/24420 [7:57:07<2:13:44,  1.49s/it]
 78%|███████▊  | 19053/24420 [7:57:08<2:13:34,  1.49s/it]
 78%|███████▊  | 19054/24420 [7:57:10<2:13:35,  1.49s/it]
 78%|███████▊  | 19055/24420 [7:57:11<2:13:31,  1.49s/it]
                                                         
{'loss': 0.0834, 'learning_rate': 5e-05, 'epoch': 3.9}

 78%|███████▊  | 19055/24420 [7:57:11<2:13:31,  1.49s/it]
 78%|███████▊  | 19056/24420 [7:57:13<2:13:41,  1.50s/it]
 78%|███████▊  | 19057/24420 [7:57:14<2:13:25,  1.49s/it]
 78%|███████▊  | 19058/24420 [7:57:16<2:13:31,  1.49s/it]
 78%|███████▊  | 19059/24420 [7:57:17<2:13:17,  1.49s/it]
 78%|███████▊  | 19060/24420 [7:57:19<2:13:06,  1.49s/it]
                                                         
{'loss': 0.0826, 'learning_rate': 5e-05, 'epoch': 3.9}

 78%|███████▊  | 19060/24420 [7:57:19<2:13:06,  1.49s/it]
 78%|███████▊  | 19061/24420 [7:57:20<2:13:06,  1.49s/it]
 78%|███████▊  | 19062/24420 [7:57:22<2:12:53,  1.49s/it]
 78%|███████▊  | 19063/24420 [7:57:23<2:13:01,  1.49s/it]
 78%|███████▊  | 19064/24420 [7:57:25<2:12:51,  1.49s/it]
 78%|███████▊  | 19065/24420 [7:57:26<2:12:39,  1.49s/it]
                                                         
{'loss': 0.0776, 'learning_rate': 5e-05, 'epoch': 3.9}

 78%|███████▊  | 19065/24420 [7:57:26<2:12:39,  1.49s/it]
 78%|███████▊  | 19066/24420 [7:57:28<2:12:45,  1.49s/it]
 78%|███████▊  | 19067/24420 [7:57:29<2:13:13,  1.49s/it]
 78%|███████▊  | 19068/24420 [7:57:31<2:13:13,  1.49s/it]
 78%|███████▊  | 19069/24420 [7:57:32<2:12:54,  1.49s/it]
 78%|███████▊  | 19070/24420 [7:57:34<2:12:40,  1.49s/it]
                                                         
{'loss': 0.0708, 'learning_rate': 5e-05, 'epoch': 3.9}

 78%|███████▊  | 19070/24420 [7:57:34<2:12:40,  1.49s/it]
 78%|███████▊  | 19071/24420 [7:57:35<2:12:37,  1.49s/it]
 78%|███████▊  | 19072/24420 [7:57:36<2:12:39,  1.49s/it]
 78%|███████▊  | 19073/24420 [7:57:38<2:12:51,  1.49s/it]
 78%|███████▊  | 19074/24420 [7:57:39<2:12:36,  1.49s/it]
 78%|███████▊  | 19075/24420 [7:57:41<2:12:34,  1.49s/it]
                                                         
{'loss': 0.0857, 'learning_rate': 5e-05, 'epoch': 3.91}

 78%|███████▊  | 19075/24420 [7:57:41<2:12:34,  1.49s/it]
 78%|███████▊  | 19076/24420 [7:57:42<2:12:37,  1.49s/it]
 78%|███████▊  | 19077/24420 [7:57:44<2:12:53,  1.49s/it]
 78%|███████▊  | 19078/24420 [7:57:45<2:12:58,  1.49s/it]
 78%|███████▊  | 19079/24420 [7:57:47<2:12:56,  1.49s/it]
 78%|███████▊  | 19080/24420 [7:57:48<2:12:47,  1.49s/it]
                                                         
{'loss': 0.0753, 'learning_rate': 5e-05, 'epoch': 3.91}

 78%|███████▊  | 19080/24420 [7:57:48<2:12:47,  1.49s/it]
 78%|███████▊  | 19081/24420 [7:57:50<2:13:04,  1.50s/it]
 78%|███████▊  | 19082/24420 [7:57:51<2:13:00,  1.50s/it]
 78%|███████▊  | 19083/24420 [7:57:53<2:12:29,  1.49s/it]
 78%|███████▊  | 19084/24420 [7:57:54<2:12:17,  1.49s/it]
 78%|███████▊  | 19085/24420 [7:57:56<2:12:10,  1.49s/it]
                                                         
{'loss': 0.0801, 'learning_rate': 5e-05, 'epoch': 3.91}

 78%|███████▊  | 19085/24420 [7:57:56<2:12:10,  1.49s/it]
 78%|███████▊  | 19086/24420 [7:57:57<2:12:20,  1.49s/it]
 78%|███████▊  | 19087/24420 [7:57:59<2:12:08,  1.49s/it]
 78%|███████▊  | 19088/24420 [7:58:00<2:12:10,  1.49s/it]
 78%|███████▊  | 19089/24420 [7:58:02<2:12:09,  1.49s/it]
 78%|███████▊  | 19090/24420 [7:58:03<2:12:07,  1.49s/it]
                                                         
{'loss': 0.075, 'learning_rate': 5e-05, 'epoch': 3.91}

 78%|███████▊  | 19090/24420 [7:58:03<2:12:07,  1.49s/it]
 78%|███████▊  | 19091/24420 [7:58:05<2:12:25,  1.49s/it]
 78%|███████▊  | 19092/24420 [7:58:06<2:12:25,  1.49s/it]
 78%|███████▊  | 19093/24420 [7:58:08<2:11:57,  1.49s/it]
 78%|███████▊  | 19094/24420 [7:58:09<2:12:01,  1.49s/it]
 78%|███████▊  | 19095/24420 [7:58:11<2:11:57,  1.49s/it]
                                                         
{'loss': 0.0659, 'learning_rate': 5e-05, 'epoch': 3.91}

 78%|███████▊  | 19095/24420 [7:58:11<2:11:57,  1.49s/it]
 78%|███████▊  | 19096/24420 [7:58:12<2:12:29,  1.49s/it]
 78%|███████▊  | 19097/24420 [7:58:14<2:12:12,  1.49s/it]
 78%|███████▊  | 19098/24420 [7:58:15<2:12:01,  1.49s/it]
 78%|███████▊  | 19099/24420 [7:58:17<2:12:08,  1.49s/it]
 78%|███████▊  | 19100/24420 [7:58:18<2:12:03,  1.49s/it]
                                                         
{'loss': 0.0665, 'learning_rate': 5e-05, 'epoch': 3.91}

 78%|███████▊  | 19100/24420 [7:58:18<2:12:03,  1.49s/it]
 78%|███████▊  | 19101/24420 [7:58:20<2:12:20,  1.49s/it]
 78%|███████▊  | 19102/24420 [7:58:21<2:12:17,  1.49s/it]
 78%|███████▊  | 19103/24420 [7:58:23<2:12:18,  1.49s/it]
 78%|███████▊  | 19104/24420 [7:58:24<2:12:21,  1.49s/it]
 78%|███████▊  | 19105/24420 [7:58:26<2:12:07,  1.49s/it]
                                                         
{'loss': 0.088, 'learning_rate': 5e-05, 'epoch': 3.91}

 78%|███████▊  | 19105/24420 [7:58:26<2:12:07,  1.49s/it]
 78%|███████▊  | 19106/24420 [7:58:27<2:12:06,  1.49s/it]
 78%|███████▊  | 19107/24420 [7:58:29<2:11:56,  1.49s/it]
 78%|███████▊  | 19108/24420 [7:58:30<2:11:41,  1.49s/it]
 78%|███████▊  | 19109/24420 [7:58:32<2:11:43,  1.49s/it]
 78%|███████▊  | 19110/24420 [7:58:33<2:11:35,  1.49s/it]
                                                         
{'loss': 0.0834, 'learning_rate': 5e-05, 'epoch': 3.91}

 78%|███████▊  | 19110/24420 [7:58:33<2:11:35,  1.49s/it]
 78%|███████▊  | 19111/24420 [7:58:35<2:11:39,  1.49s/it]
 78%|███████▊  | 19112/24420 [7:58:36<2:11:22,  1.49s/it]
 78%|███████▊  | 19113/24420 [7:58:38<2:11:30,  1.49s/it]
 78%|███████▊  | 19114/24420 [7:58:39<2:11:59,  1.49s/it]
 78%|███████▊  | 19115/24420 [7:58:41<2:11:54,  1.49s/it]
                                                         
{'loss': 0.0668, 'learning_rate': 5e-05, 'epoch': 3.91}

 78%|███████▊  | 19115/24420 [7:58:41<2:11:54,  1.49s/it]
 78%|███████▊  | 19116/24420 [7:58:42<2:12:54,  1.50s/it]
 78%|███████▊  | 19117/24420 [7:58:44<2:12:27,  1.50s/it]
 78%|███████▊  | 19118/24420 [7:58:45<2:12:09,  1.50s/it]
 78%|███████▊  | 19119/24420 [7:58:47<2:12:01,  1.49s/it]
 78%|███████▊  | 19120/24420 [7:58:48<2:11:51,  1.49s/it]
                                                         
{'loss': 0.0821, 'learning_rate': 5e-05, 'epoch': 3.91}

 78%|███████▊  | 19120/24420 [7:58:48<2:11:51,  1.49s/it]
 78%|███████▊  | 19121/24420 [7:58:50<2:11:55,  1.49s/it]
 78%|███████▊  | 19122/24420 [7:58:51<2:11:42,  1.49s/it]
 78%|███████▊  | 19123/24420 [7:58:53<2:11:33,  1.49s/it]
 78%|███████▊  | 19124/24420 [7:58:54<2:11:27,  1.49s/it]
 78%|███████▊  | 19125/24420 [7:58:55<2:11:11,  1.49s/it]
                                                         
{'loss': 0.0782, 'learning_rate': 5e-05, 'epoch': 3.92}

 78%|███████▊  | 19125/24420 [7:58:55<2:11:11,  1.49s/it]
 78%|███████▊  | 19126/24420 [7:58:57<2:11:14,  1.49s/it]
 78%|███████▊  | 19127/24420 [7:58:58<2:11:11,  1.49s/it]
 78%|███████▊  | 19128/24420 [7:59:00<2:11:24,  1.49s/it]
 78%|███████▊  | 19129/24420 [7:59:01<2:11:27,  1.49s/it]
 78%|███████▊  | 19130/24420 [7:59:03<2:11:18,  1.49s/it]
                                                         
{'loss': 0.0782, 'learning_rate': 5e-05, 'epoch': 3.92}

 78%|███████▊  | 19130/24420 [7:59:03<2:11:18,  1.49s/it]
 78%|███████▊  | 19131/24420 [7:59:04<2:11:20,  1.49s/it]
 78%|███████▊  | 19132/24420 [7:59:06<2:11:17,  1.49s/it]
 78%|███████▊  | 19133/24420 [7:59:07<2:11:10,  1.49s/it]
 78%|███████▊  | 19134/24420 [7:59:09<2:11:26,  1.49s/it]
 78%|███████▊  | 19135/24420 [7:59:10<2:11:17,  1.49s/it]
                                                         
{'loss': 0.0904, 'learning_rate': 5e-05, 'epoch': 3.92}

 78%|███████▊  | 19135/24420 [7:59:10<2:11:17,  1.49s/it]
 78%|███████▊  | 19136/24420 [7:59:12<2:11:14,  1.49s/it]
 78%|███████▊  | 19137/24420 [7:59:13<2:11:39,  1.50s/it]
 78%|███████▊  | 19138/24420 [7:59:15<2:11:37,  1.50s/it]
 78%|███████▊  | 19139/24420 [7:59:16<2:11:51,  1.50s/it]
 78%|███████▊  | 19140/24420 [7:59:18<2:11:37,  1.50s/it]
                                                         
{'loss': 0.0728, 'learning_rate': 5e-05, 'epoch': 3.92}

 78%|███████▊  | 19140/24420 [7:59:18<2:11:37,  1.50s/it]
 78%|███████▊  | 19141/24420 [7:59:19<2:11:22,  1.49s/it]
 78%|███████▊  | 19142/24420 [7:59:21<2:11:09,  1.49s/it]
 78%|███████▊  | 19143/24420 [7:59:22<2:11:14,  1.49s/it]
 78%|███████▊  | 19144/24420 [7:59:24<2:11:09,  1.49s/it]
 78%|███████▊  | 19145/24420 [7:59:25<2:11:29,  1.50s/it]
                                                         
{'loss': 0.0842, 'learning_rate': 5e-05, 'epoch': 3.92}

 78%|███████▊  | 19145/24420 [7:59:25<2:11:29,  1.50s/it]
 78%|███████▊  | 19146/24420 [7:59:27<2:11:29,  1.50s/it]
 78%|███████▊  | 19147/24420 [7:59:28<2:11:47,  1.50s/it]
 78%|███████▊  | 19148/24420 [7:59:30<2:11:33,  1.50s/it]
 78%|███████▊  | 19149/24420 [7:59:31<2:11:36,  1.50s/it]
 78%|███████▊  | 19150/24420 [7:59:33<2:11:13,  1.49s/it]
                                                         
{'loss': 0.076, 'learning_rate': 5e-05, 'epoch': 3.92}

 78%|███████▊  | 19150/24420 [7:59:33<2:11:13,  1.49s/it]
 78%|███████▊  | 19151/24420 [7:59:34<2:11:19,  1.50s/it]
 78%|███████▊  | 19152/24420 [7:59:36<2:11:08,  1.49s/it]
 78%|███████▊  | 19153/24420 [7:59:37<2:11:12,  1.49s/it]
 78%|███████▊  | 19154/24420 [7:59:39<2:11:05,  1.49s/it]
 78%|███████▊  | 19155/24420 [7:59:40<2:11:01,  1.49s/it]
                                                         
{'loss': 0.0759, 'learning_rate': 5e-05, 'epoch': 3.92}

 78%|███████▊  | 19155/24420 [7:59:40<2:11:01,  1.49s/it]
 78%|███████▊  | 19156/24420 [7:59:42<2:11:08,  1.49s/it]
 78%|███████▊  | 19157/24420 [7:59:43<2:11:03,  1.49s/it]
 78%|███████▊  | 19158/24420 [7:59:45<2:10:43,  1.49s/it]
 78%|███████▊  | 19159/24420 [7:59:46<2:10:50,  1.49s/it]
 78%|███████▊  | 19160/24420 [7:59:48<2:11:12,  1.50s/it]
                                                         
{'loss': 0.0898, 'learning_rate': 5e-05, 'epoch': 3.92}

 78%|███████▊  | 19160/24420 [7:59:48<2:11:12,  1.50s/it]
 78%|███████▊  | 19161/24420 [7:59:49<2:10:55,  1.49s/it]
 78%|███████▊  | 19162/24420 [7:59:51<2:10:30,  1.49s/it]
 78%|███████▊  | 19163/24420 [7:59:52<2:10:26,  1.49s/it]
 78%|███████▊  | 19164/24420 [7:59:54<2:10:26,  1.49s/it]
 78%|███████▊  | 19165/24420 [7:59:55<2:10:37,  1.49s/it]
                                                         
{'loss': 0.0911, 'learning_rate': 5e-05, 'epoch': 3.92}

 78%|███████▊  | 19165/24420 [7:59:55<2:10:37,  1.49s/it]
 78%|███████▊  | 19166/24420 [7:59:57<2:10:27,  1.49s/it]
 78%|███████▊  | 19167/24420 [7:59:58<2:10:26,  1.49s/it]
 78%|███████▊  | 19168/24420 [8:00:00<2:10:16,  1.49s/it]
 78%|███████▊  | 19169/24420 [8:00:01<2:10:16,  1.49s/it]
 79%|███████▊  | 19170/24420 [8:00:03<2:10:02,  1.49s/it]
                                                         
{'loss': 0.0752, 'learning_rate': 5e-05, 'epoch': 3.93}

 79%|███████▊  | 19170/24420 [8:00:03<2:10:02,  1.49s/it]
 79%|███████▊  | 19171/24420 [8:00:04<2:09:53,  1.48s/it]
 79%|███████▊  | 19172/24420 [8:00:06<2:09:47,  1.48s/it]
 79%|███████▊  | 19173/24420 [8:00:07<2:10:02,  1.49s/it]
 79%|███████▊  | 19174/24420 [8:00:09<2:10:00,  1.49s/it]
 79%|███████▊  | 19175/24420 [8:00:10<2:09:58,  1.49s/it]
                                                         
{'loss': 0.0761, 'learning_rate': 5e-05, 'epoch': 3.93}

 79%|███████▊  | 19175/24420 [8:00:10<2:09:58,  1.49s/it]
 79%|███████▊  | 19176/24420 [8:00:12<2:10:00,  1.49s/it]
 79%|███████▊  | 19177/24420 [8:00:13<2:12:42,  1.52s/it]
 79%|███████▊  | 19178/24420 [8:00:15<2:11:59,  1.51s/it]
 79%|███████▊  | 19179/24420 [8:00:16<2:11:18,  1.50s/it]
 79%|███████▊  | 19180/24420 [8:00:18<2:10:55,  1.50s/it]
                                                         
{'loss': 0.0753, 'learning_rate': 5e-05, 'epoch': 3.93}

 79%|███████▊  | 19180/24420 [8:00:18<2:10:55,  1.50s/it]
 79%|███████▊  | 19181/24420 [8:00:19<2:10:45,  1.50s/it]
 79%|███████▊  | 19182/24420 [8:00:21<2:10:27,  1.49s/it]
 79%|███████▊  | 19183/24420 [8:00:22<2:10:47,  1.50s/it]
 79%|███████▊  | 19184/24420 [8:00:24<2:10:39,  1.50s/it]
 79%|███████▊  | 19185/24420 [8:00:25<2:10:34,  1.50s/it]
                                                         
{'loss': 0.0904, 'learning_rate': 5e-05, 'epoch': 3.93}

 79%|███████▊  | 19185/24420 [8:00:25<2:10:34,  1.50s/it]
 79%|███████▊  | 19186/24420 [8:00:27<2:10:23,  1.49s/it]
 79%|███████▊  | 19187/24420 [8:00:28<2:10:05,  1.49s/it]
 79%|███████▊  | 19188/24420 [8:00:30<2:09:46,  1.49s/it]
 79%|███████▊  | 19189/24420 [8:00:31<2:09:38,  1.49s/it]
 79%|███████▊  | 19190/24420 [8:00:33<2:09:27,  1.49s/it]
                                                         
{'loss': 0.064, 'learning_rate': 5e-05, 'epoch': 3.93}

 79%|███████▊  | 19190/24420 [8:00:33<2:09:27,  1.49s/it]
 79%|███████▊  | 19191/24420 [8:00:34<2:09:34,  1.49s/it]
 79%|███████▊  | 19192/24420 [8:00:35<2:09:27,  1.49s/it]
 79%|███████▊  | 19193/24420 [8:00:37<2:09:21,  1.48s/it]
 79%|███████▊  | 19194/24420 [8:00:38<2:09:20,  1.48s/it]
 79%|███████▊  | 19195/24420 [8:00:40<2:09:19,  1.49s/it]
                                                         
{'loss': 0.0864, 'learning_rate': 5e-05, 'epoch': 3.93}

 79%|███████▊  | 19195/24420 [8:00:40<2:09:19,  1.49s/it]
 79%|███████▊  | 19196/24420 [8:00:41<2:09:18,  1.49s/it]
 79%|███████▊  | 19197/24420 [8:00:43<2:09:09,  1.48s/it]
 79%|███████▊  | 19198/24420 [8:00:44<2:09:15,  1.49s/it]
 79%|███████▊  | 19199/24420 [8:00:46<2:09:23,  1.49s/it]
 79%|███████▊  | 19200/24420 [8:00:47<2:09:25,  1.49s/it]
                                                         
{'loss': 0.0658, 'learning_rate': 5e-05, 'epoch': 3.93}

 79%|███████▊  | 19200/24420 [8:00:47<2:09:25,  1.49s/it]
 79%|███████▊  | 19201/24420 [8:00:52<3:32:03,  2.44s/it]
 79%|███████▊  | 19202/24420 [8:00:54<3:07:21,  2.15s/it]
 79%|███████▊  | 19203/24420 [8:00:55<2:49:50,  1.95s/it]
 79%|███████▊  | 19204/24420 [8:00:56<2:37:36,  1.81s/it]
 79%|███████▊  | 19205/24420 [8:00:58<2:29:04,  1.72s/it]
                                                         
{'loss': 0.0729, 'learning_rate': 5e-05, 'epoch': 3.93}

 79%|███████▊  | 19205/24420 [8:00:58<2:29:04,  1.72s/it]
 79%|███████▊  | 19206/24420 [8:00:59<2:23:06,  1.65s/it]
 79%|███████▊  | 19207/24420 [8:01:01<2:19:04,  1.60s/it]
 79%|███████▊  | 19208/24420 [8:01:02<2:16:12,  1.57s/it]
 79%|███████▊  | 19209/24420 [8:01:04<2:13:48,  1.54s/it]
 79%|███████▊  | 19210/24420 [8:01:05<2:12:21,  1.52s/it]
                                                         
{'loss': 0.0775, 'learning_rate': 5e-05, 'epoch': 3.93}

 79%|███████▊  | 19210/24420 [8:01:05<2:12:21,  1.52s/it]
 79%|███████▊  | 19211/24420 [8:01:07<2:11:22,  1.51s/it]
 79%|███████▊  | 19212/24420 [8:01:08<2:10:34,  1.50s/it]
 79%|███████▊  | 19213/24420 [8:01:10<2:09:54,  1.50s/it]
 79%|███████▊  | 19214/24420 [8:01:11<2:09:47,  1.50s/it]
 79%|███████▊  | 19215/24420 [8:01:13<2:09:38,  1.49s/it]
                                                         
{'loss': 0.0899, 'learning_rate': 5e-05, 'epoch': 3.93}

 79%|███████▊  | 19215/24420 [8:01:13<2:09:38,  1.49s/it]
 79%|███████▊  | 19216/24420 [8:01:14<2:09:25,  1.49s/it]
 79%|███████▊  | 19217/24420 [8:01:16<2:09:03,  1.49s/it]
 79%|███████▊  | 19218/24420 [8:01:17<2:08:50,  1.49s/it]
 79%|███████▊  | 19219/24420 [8:01:19<2:08:58,  1.49s/it]
 79%|███████▊  | 19220/24420 [8:01:20<2:08:55,  1.49s/it]
                                                         
{'loss': 0.0687, 'learning_rate': 5e-05, 'epoch': 3.94}

 79%|███████▊  | 19220/24420 [8:01:20<2:08:55,  1.49s/it]
 79%|███████▊  | 19221/24420 [8:01:22<2:08:44,  1.49s/it]
 79%|███████▊  | 19222/24420 [8:01:23<2:08:48,  1.49s/it]
 79%|███████▊  | 19223/24420 [8:01:25<2:08:44,  1.49s/it]
 79%|███████▊  | 19224/24420 [8:01:26<2:08:47,  1.49s/it]
 79%|███████▊  | 19225/24420 [8:01:28<2:08:47,  1.49s/it]
                                                         
{'loss': 0.0907, 'learning_rate': 5e-05, 'epoch': 3.94}

 79%|███████▊  | 19225/24420 [8:01:28<2:08:47,  1.49s/it]
 79%|███████▊  | 19226/24420 [8:01:29<2:08:54,  1.49s/it]
 79%|███████▊  | 19227/24420 [8:01:31<2:08:47,  1.49s/it]
 79%|███████▊  | 19228/24420 [8:01:32<2:08:37,  1.49s/it]
 79%|███████▊  | 19229/24420 [8:01:34<2:08:48,  1.49s/it]
 79%|███████▊  | 19230/24420 [8:01:35<2:09:03,  1.49s/it]
                                                         
{'loss': 0.0801, 'learning_rate': 5e-05, 'epoch': 3.94}

 79%|███████▊  | 19230/24420 [8:01:35<2:09:03,  1.49s/it]
 79%|███████▉  | 19231/24420 [8:01:37<2:08:57,  1.49s/it]
 79%|███████▉  | 19232/24420 [8:01:38<2:08:32,  1.49s/it]
 79%|███████▉  | 19233/24420 [8:01:40<2:08:27,  1.49s/it]
 79%|███████▉  | 19234/24420 [8:01:41<2:08:23,  1.49s/it]
 79%|███████▉  | 19235/24420 [8:01:43<2:08:45,  1.49s/it]
                                                         
{'loss': 0.0633, 'learning_rate': 5e-05, 'epoch': 3.94}

 79%|███████▉  | 19235/24420 [8:01:43<2:08:45,  1.49s/it]
 79%|███████▉  | 19236/24420 [8:01:44<2:08:48,  1.49s/it]
 79%|███████▉  | 19237/24420 [8:01:46<2:08:37,  1.49s/it]
 79%|███████▉  | 19238/24420 [8:01:47<2:08:32,  1.49s/it]
 79%|███████▉  | 19239/24420 [8:01:49<2:08:29,  1.49s/it]
 79%|███████▉  | 19240/24420 [8:01:50<2:08:32,  1.49s/it]
                                                         
{'loss': 0.0711, 'learning_rate': 5e-05, 'epoch': 3.94}

 79%|███████▉  | 19240/24420 [8:01:50<2:08:32,  1.49s/it]
 79%|███████▉  | 19241/24420 [8:01:52<2:08:27,  1.49s/it]
 79%|███████▉  | 19242/24420 [8:01:53<2:08:25,  1.49s/it]
 79%|███████▉  | 19243/24420 [8:01:55<2:08:24,  1.49s/it]
 79%|███████▉  | 19244/24420 [8:01:56<2:08:11,  1.49s/it]
 79%|███████▉  | 19245/24420 [8:01:57<2:08:18,  1.49s/it]
                                                         
{'loss': 0.0771, 'learning_rate': 5e-05, 'epoch': 3.94}

 79%|███████▉  | 19245/24420 [8:01:57<2:08:18,  1.49s/it]
 79%|███████▉  | 19246/24420 [8:01:59<2:08:46,  1.49s/it]
 79%|███████▉  | 19247/24420 [8:02:00<2:08:36,  1.49s/it]
 79%|███████▉  | 19248/24420 [8:02:02<2:08:39,  1.49s/it]
 79%|███████▉  | 19249/24420 [8:02:03<2:08:38,  1.49s/it]
 79%|███████▉  | 19250/24420 [8:02:05<2:08:26,  1.49s/it]
                                                         
{'loss': 0.0658, 'learning_rate': 5e-05, 'epoch': 3.94}

 79%|███████▉  | 19250/24420 [8:02:05<2:08:26,  1.49s/it]
 79%|███████▉  | 19251/24420 [8:02:06<2:08:19,  1.49s/it]
 79%|███████▉  | 19252/24420 [8:02:08<2:08:06,  1.49s/it]
 79%|███████▉  | 19253/24420 [8:02:09<2:07:57,  1.49s/it]
 79%|███████▉  | 19254/24420 [8:02:11<2:08:07,  1.49s/it]
 79%|███████▉  | 19255/24420 [8:02:12<2:08:22,  1.49s/it]
                                                         
{'loss': 0.0833, 'learning_rate': 5e-05, 'epoch': 3.94}

 79%|███████▉  | 19255/24420 [8:02:12<2:08:22,  1.49s/it]
 79%|███████▉  | 19256/24420 [8:02:14<2:08:20,  1.49s/it]
 79%|███████▉  | 19257/24420 [8:02:15<2:08:09,  1.49s/it]
 79%|███████▉  | 19258/24420 [8:02:17<2:08:07,  1.49s/it]
 79%|███████▉  | 19259/24420 [8:02:18<2:07:59,  1.49s/it]
 79%|███████▉  | 19260/24420 [8:02:20<2:08:03,  1.49s/it]
                                                         
{'loss': 0.0794, 'learning_rate': 5e-05, 'epoch': 3.94}

 79%|███████▉  | 19260/24420 [8:02:20<2:08:03,  1.49s/it]
 79%|███████▉  | 19261/24420 [8:02:21<2:08:08,  1.49s/it]
 79%|███████▉  | 19262/24420 [8:02:23<2:07:55,  1.49s/it]
 79%|███████▉  | 19263/24420 [8:02:24<2:07:49,  1.49s/it]
 79%|███████▉  | 19264/24420 [8:02:26<2:07:51,  1.49s/it]
 79%|███████▉  | 19265/24420 [8:02:27<2:07:52,  1.49s/it]
                                                         
{'loss': 0.0764, 'learning_rate': 5e-05, 'epoch': 3.94}

 79%|███████▉  | 19265/24420 [8:02:27<2:07:52,  1.49s/it]
 79%|███████▉  | 19266/24420 [8:02:29<2:07:42,  1.49s/it]
 79%|███████▉  | 19267/24420 [8:02:30<2:07:51,  1.49s/it]
 79%|███████▉  | 19268/24420 [8:02:32<2:08:00,  1.49s/it]
 79%|███████▉  | 19269/24420 [8:02:33<2:08:02,  1.49s/it]
 79%|███████▉  | 19270/24420 [8:02:35<2:07:55,  1.49s/it]
                                                         
{'loss': 0.0743, 'learning_rate': 5e-05, 'epoch': 3.95}

 79%|███████▉  | 19270/24420 [8:02:35<2:07:55,  1.49s/it]
 79%|███████▉  | 19271/24420 [8:02:36<2:07:51,  1.49s/it]
 79%|███████▉  | 19272/24420 [8:02:38<2:07:42,  1.49s/it]
 79%|███████▉  | 19273/24420 [8:02:39<2:08:58,  1.50s/it]
 79%|███████▉  | 19274/24420 [8:02:41<2:08:36,  1.50s/it]
 79%|███████▉  | 19275/24420 [8:02:42<2:08:12,  1.50s/it]
                                                         
{'loss': 0.0837, 'learning_rate': 5e-05, 'epoch': 3.95}

 79%|███████▉  | 19275/24420 [8:02:42<2:08:12,  1.50s/it]
 79%|███████▉  | 19276/24420 [8:02:44<2:08:15,  1.50s/it]
 79%|███████▉  | 19277/24420 [8:02:45<2:07:43,  1.49s/it]
 79%|███████▉  | 19278/24420 [8:02:47<2:07:39,  1.49s/it]
 79%|███████▉  | 19279/24420 [8:02:48<2:07:43,  1.49s/it]
 79%|███████▉  | 19280/24420 [8:02:50<2:07:31,  1.49s/it]
                                                         
{'loss': 0.0726, 'learning_rate': 5e-05, 'epoch': 3.95}

 79%|███████▉  | 19280/24420 [8:02:50<2:07:31,  1.49s/it]
 79%|███████▉  | 19281/24420 [8:02:51<2:07:35,  1.49s/it]
 79%|███████▉  | 19282/24420 [8:02:53<2:07:55,  1.49s/it]
 79%|███████▉  | 19283/24420 [8:02:54<2:08:11,  1.50s/it]
 79%|███████▉  | 19284/24420 [8:02:56<2:07:57,  1.49s/it]
 79%|███████▉  | 19285/24420 [8:02:57<2:07:39,  1.49s/it]
                                                         
{'loss': 0.0809, 'learning_rate': 5e-05, 'epoch': 3.95}

 79%|███████▉  | 19285/24420 [8:02:57<2:07:39,  1.49s/it]
 79%|███████▉  | 19286/24420 [8:02:59<2:07:38,  1.49s/it]
 79%|███████▉  | 19287/24420 [8:03:00<2:07:43,  1.49s/it]
 79%|███████▉  | 19288/24420 [8:03:02<2:07:40,  1.49s/it]
 79%|███████▉  | 19289/24420 [8:03:03<2:07:26,  1.49s/it]
 79%|███████▉  | 19290/24420 [8:03:05<2:07:34,  1.49s/it]
                                                         
{'loss': 0.0734, 'learning_rate': 5e-05, 'epoch': 3.95}

 79%|███████▉  | 19290/24420 [8:03:05<2:07:34,  1.49s/it]
 79%|███████▉  | 19291/24420 [8:03:06<2:07:32,  1.49s/it]
 79%|███████▉  | 19292/24420 [8:03:08<2:07:31,  1.49s/it]
 79%|███████▉  | 19293/24420 [8:03:09<2:07:29,  1.49s/it]
 79%|███████▉  | 19294/24420 [8:03:11<2:07:32,  1.49s/it]
 79%|███████▉  | 19295/24420 [8:03:12<2:07:37,  1.49s/it]
                                                         
{'loss': 0.0705, 'learning_rate': 5e-05, 'epoch': 3.95}

 79%|███████▉  | 19295/24420 [8:03:12<2:07:37,  1.49s/it]
 79%|███████▉  | 19296/24420 [8:03:14<2:07:41,  1.50s/it]
 79%|███████▉  | 19297/24420 [8:03:15<2:07:42,  1.50s/it]
 79%|███████▉  | 19298/24420 [8:03:17<2:07:22,  1.49s/it]
 79%|███████▉  | 19299/24420 [8:03:18<2:07:14,  1.49s/it]
 79%|███████▉  | 19300/24420 [8:03:20<2:07:11,  1.49s/it]
                                                         
{'loss': 0.0786, 'learning_rate': 5e-05, 'epoch': 3.95}

 79%|███████▉  | 19300/24420 [8:03:20<2:07:11,  1.49s/it]
 79%|███████▉  | 19301/24420 [8:03:21<2:07:19,  1.49s/it]
 79%|███████▉  | 19302/24420 [8:03:22<2:06:57,  1.49s/it]
 79%|███████▉  | 19303/24420 [8:03:24<2:07:03,  1.49s/it]
 79%|███████▉  | 19304/24420 [8:03:25<2:06:57,  1.49s/it]
 79%|███████▉  | 19305/24420 [8:03:27<2:07:00,  1.49s/it]
                                                         
{'loss': 0.089, 'learning_rate': 5e-05, 'epoch': 3.95}

 79%|███████▉  | 19305/24420 [8:03:27<2:07:00,  1.49s/it]
 79%|███████▉  | 19306/24420 [8:03:28<2:07:01,  1.49s/it]
 79%|███████▉  | 19307/24420 [8:03:30<2:06:52,  1.49s/it]
 79%|███████▉  | 19308/24420 [8:03:31<2:06:53,  1.49s/it]
 79%|███████▉  | 19309/24420 [8:03:33<2:07:02,  1.49s/it]
 79%|███████▉  | 19310/24420 [8:03:34<2:06:47,  1.49s/it]
                                                         
{'loss': 0.0676, 'learning_rate': 5e-05, 'epoch': 3.95}

 79%|███████▉  | 19310/24420 [8:03:34<2:06:47,  1.49s/it]
 79%|███████▉  | 19311/24420 [8:03:36<2:06:47,  1.49s/it]
 79%|███████▉  | 19312/24420 [8:03:37<2:06:49,  1.49s/it]
 79%|███████▉  | 19313/24420 [8:03:39<2:06:52,  1.49s/it]
 79%|███████▉  | 19314/24420 [8:03:40<2:06:35,  1.49s/it]
 79%|███████▉  | 19315/24420 [8:03:42<2:06:28,  1.49s/it]
                                                         
{'loss': 0.0801, 'learning_rate': 5e-05, 'epoch': 3.95}

 79%|███████▉  | 19315/24420 [8:03:42<2:06:28,  1.49s/it]
 79%|███████▉  | 19316/24420 [8:03:43<2:06:30,  1.49s/it]
 79%|███████▉  | 19317/24420 [8:03:45<2:06:42,  1.49s/it]
 79%|███████▉  | 19318/24420 [8:03:46<2:06:49,  1.49s/it]
 79%|███████▉  | 19319/24420 [8:03:48<2:06:51,  1.49s/it]
 79%|███████▉  | 19320/24420 [8:03:49<2:06:51,  1.49s/it]
                                                         
{'loss': 0.0855, 'learning_rate': 5e-05, 'epoch': 3.96}

 79%|███████▉  | 19320/24420 [8:03:49<2:06:51,  1.49s/it]
 79%|███████▉  | 19321/24420 [8:03:51<2:06:48,  1.49s/it]
 79%|███████▉  | 19322/24420 [8:03:52<2:06:26,  1.49s/it]
 79%|███████▉  | 19323/24420 [8:03:54<2:06:22,  1.49s/it]
 79%|███████▉  | 19324/24420 [8:03:55<2:06:20,  1.49s/it]
 79%|███████▉  | 19325/24420 [8:03:57<2:06:18,  1.49s/it]
                                                         
{'loss': 0.0774, 'learning_rate': 5e-05, 'epoch': 3.96}

 79%|███████▉  | 19325/24420 [8:03:57<2:06:18,  1.49s/it]
 79%|███████▉  | 19326/24420 [8:03:58<2:09:26,  1.52s/it]
 79%|███████▉  | 19327/24420 [8:04:00<2:08:20,  1.51s/it]
 79%|███████▉  | 19328/24420 [8:04:01<2:07:37,  1.50s/it]
 79%|███████▉  | 19329/24420 [8:04:03<2:07:02,  1.50s/it]
 79%|███████▉  | 19330/24420 [8:04:04<2:06:53,  1.50s/it]
                                                         
{'loss': 0.0912, 'learning_rate': 5e-05, 'epoch': 3.96}

 79%|███████▉  | 19330/24420 [8:04:04<2:06:53,  1.50s/it]
 79%|███████▉  | 19331/24420 [8:04:06<2:06:41,  1.49s/it]
 79%|███████▉  | 19332/24420 [8:04:07<2:06:48,  1.50s/it]
 79%|███████▉  | 19333/24420 [8:04:09<2:06:20,  1.49s/it]
 79%|███████▉  | 19334/24420 [8:04:10<2:06:20,  1.49s/it]
 79%|███████▉  | 19335/24420 [8:04:12<2:06:14,  1.49s/it]
                                                         
{'loss': 0.0795, 'learning_rate': 5e-05, 'epoch': 3.96}

 79%|███████▉  | 19335/24420 [8:04:12<2:06:14,  1.49s/it]
 79%|███████▉  | 19336/24420 [8:04:13<2:06:36,  1.49s/it]
 79%|███████▉  | 19337/24420 [8:04:15<2:06:28,  1.49s/it]
 79%|███████▉  | 19338/24420 [8:04:16<2:06:20,  1.49s/it]
 79%|███████▉  | 19339/24420 [8:04:18<2:06:37,  1.50s/it]
 79%|███████▉  | 19340/24420 [8:04:19<2:06:20,  1.49s/it]
                                                         
{'loss': 0.0977, 'learning_rate': 5e-05, 'epoch': 3.96}

 79%|███████▉  | 19340/24420 [8:04:19<2:06:20,  1.49s/it]
 79%|███████▉  | 19341/24420 [8:04:21<2:06:20,  1.49s/it]
 79%|███████▉  | 19342/24420 [8:04:22<2:06:09,  1.49s/it]
 79%|███████▉  | 19343/24420 [8:04:24<2:06:03,  1.49s/it]
 79%|███████▉  | 19344/24420 [8:04:25<2:05:54,  1.49s/it]
 79%|███████▉  | 19345/24420 [8:04:27<2:05:50,  1.49s/it]
                                                         
{'loss': 0.0785, 'learning_rate': 5e-05, 'epoch': 3.96}

 79%|███████▉  | 19345/24420 [8:04:27<2:05:50,  1.49s/it]
 79%|███████▉  | 19346/24420 [8:04:28<2:05:46,  1.49s/it]
 79%|███████▉  | 19347/24420 [8:04:30<2:05:51,  1.49s/it]
 79%|███████▉  | 19348/24420 [8:04:31<2:05:51,  1.49s/it]
 79%|███████▉  | 19349/24420 [8:04:33<2:05:53,  1.49s/it]
 79%|███████▉  | 19350/24420 [8:04:34<2:05:54,  1.49s/it]
                                                         
{'loss': 0.0914, 'learning_rate': 5e-05, 'epoch': 3.96}

 79%|███████▉  | 19350/24420 [8:04:34<2:05:54,  1.49s/it]
 79%|███████▉  | 19351/24420 [8:04:36<2:05:51,  1.49s/it]
 79%|███████▉  | 19352/24420 [8:04:37<2:05:47,  1.49s/it]
 79%|███████▉  | 19353/24420 [8:04:39<2:05:38,  1.49s/it]
 79%|███████▉  | 19354/24420 [8:04:40<2:05:51,  1.49s/it]
 79%|███████▉  | 19355/24420 [8:04:42<2:05:51,  1.49s/it]
                                                         
{'loss': 0.0862, 'learning_rate': 5e-05, 'epoch': 3.96}

 79%|███████▉  | 19355/24420 [8:04:42<2:05:51,  1.49s/it]
 79%|███████▉  | 19356/24420 [8:04:43<2:05:43,  1.49s/it]
 79%|███████▉  | 19357/24420 [8:04:45<2:05:45,  1.49s/it]
 79%|███████▉  | 19358/24420 [8:04:46<2:05:35,  1.49s/it]
 79%|███████▉  | 19359/24420 [8:04:47<2:05:24,  1.49s/it]
 79%|███████▉  | 19360/24420 [8:04:49<2:05:54,  1.49s/it]
                                                         
{'loss': 0.0811, 'learning_rate': 5e-05, 'epoch': 3.96}

 79%|███████▉  | 19360/24420 [8:04:49<2:05:54,  1.49s/it]
 79%|███████▉  | 19361/24420 [8:04:50<2:05:42,  1.49s/it]
 79%|███████▉  | 19362/24420 [8:04:52<2:05:28,  1.49s/it]
 79%|███████▉  | 19363/24420 [8:04:53<2:05:44,  1.49s/it]
 79%|███████▉  | 19364/24420 [8:04:55<2:05:33,  1.49s/it]
 79%|███████▉  | 19365/24420 [8:04:56<2:05:48,  1.49s/it]
                                                         
{'loss': 0.0738, 'learning_rate': 5e-05, 'epoch': 3.96}

 79%|███████▉  | 19365/24420 [8:04:56<2:05:48,  1.49s/it]
 79%|███████▉  | 19366/24420 [8:04:58<2:05:38,  1.49s/it]
 79%|███████▉  | 19367/24420 [8:04:59<2:05:25,  1.49s/it]
 79%|███████▉  | 19368/24420 [8:05:01<2:05:31,  1.49s/it]
 79%|███████▉  | 19369/24420 [8:05:02<2:05:29,  1.49s/it]
 79%|███████▉  | 19370/24420 [8:05:04<2:05:19,  1.49s/it]
                                                         
{'loss': 0.0695, 'learning_rate': 5e-05, 'epoch': 3.97}

 79%|███████▉  | 19370/24420 [8:05:04<2:05:19,  1.49s/it]
 79%|███████▉  | 19371/24420 [8:05:05<2:05:27,  1.49s/it]
 79%|███████▉  | 19372/24420 [8:05:07<2:05:46,  1.49s/it]
 79%|███████▉  | 19373/24420 [8:05:08<2:05:29,  1.49s/it]
 79%|███████▉  | 19374/24420 [8:05:10<2:05:15,  1.49s/it]
 79%|███████▉  | 19375/24420 [8:05:11<2:05:32,  1.49s/it]
                                                         
{'loss': 0.0926, 'learning_rate': 5e-05, 'epoch': 3.97}

 79%|███████▉  | 19375/24420 [8:05:11<2:05:32,  1.49s/it]
 79%|███████▉  | 19376/24420 [8:05:13<2:05:27,  1.49s/it]
 79%|███████▉  | 19377/24420 [8:05:14<2:05:10,  1.49s/it]
 79%|███████▉  | 19378/24420 [8:05:16<2:05:07,  1.49s/it]
 79%|███████▉  | 19379/24420 [8:05:17<2:05:27,  1.49s/it]
 79%|███████▉  | 19380/24420 [8:05:19<2:05:10,  1.49s/it]
                                                         
{'loss': 0.0789, 'learning_rate': 5e-05, 'epoch': 3.97}

 79%|███████▉  | 19380/24420 [8:05:19<2:05:10,  1.49s/it]
 79%|███████▉  | 19381/24420 [8:05:20<2:05:33,  1.49s/it]
 79%|███████▉  | 19382/24420 [8:05:22<2:05:24,  1.49s/it]
 79%|███████▉  | 19383/24420 [8:05:23<2:05:08,  1.49s/it]
 79%|███████▉  | 19384/24420 [8:05:25<2:05:05,  1.49s/it]
 79%|███████▉  | 19385/24420 [8:05:26<2:04:52,  1.49s/it]
                                                         
{'loss': 0.0678, 'learning_rate': 5e-05, 'epoch': 3.97}

 79%|███████▉  | 19385/24420 [8:05:26<2:04:52,  1.49s/it]
 79%|███████▉  | 19386/24420 [8:05:28<2:04:59,  1.49s/it]
 79%|███████▉  | 19387/24420 [8:05:29<2:04:57,  1.49s/it]
 79%|███████▉  | 19388/24420 [8:05:31<2:04:58,  1.49s/it]
 79%|███████▉  | 19389/24420 [8:05:32<2:04:53,  1.49s/it]
 79%|███████▉  | 19390/24420 [8:05:34<2:04:45,  1.49s/it]
                                                         
{'loss': 0.0922, 'learning_rate': 5e-05, 'epoch': 3.97}

 79%|███████▉  | 19390/24420 [8:05:34<2:04:45,  1.49s/it]
 79%|███████▉  | 19391/24420 [8:05:35<2:05:05,  1.49s/it]
 79%|███████▉  | 19392/24420 [8:05:37<2:05:12,  1.49s/it]
 79%|███████▉  | 19393/24420 [8:05:38<2:04:50,  1.49s/it]
 79%|███████▉  | 19394/24420 [8:05:40<2:05:13,  1.49s/it]
 79%|███████▉  | 19395/24420 [8:05:41<2:05:23,  1.50s/it]
                                                         
{'loss': 0.0813, 'learning_rate': 5e-05, 'epoch': 3.97}

 79%|███████▉  | 19395/24420 [8:05:41<2:05:23,  1.50s/it]
 79%|███████▉  | 19396/24420 [8:05:43<2:05:15,  1.50s/it]
 79%|███████▉  | 19397/24420 [8:05:44<2:05:13,  1.50s/it]
 79%|███████▉  | 19398/24420 [8:05:46<2:05:03,  1.49s/it]
 79%|███████▉  | 19399/24420 [8:05:47<2:04:55,  1.49s/it]
 79%|███████▉  | 19400/24420 [8:05:49<2:04:39,  1.49s/it]
                                                         
{'loss': 0.0815, 'learning_rate': 5e-05, 'epoch': 3.97}

 79%|███████▉  | 19400/24420 [8:05:49<2:04:39,  1.49s/it]
 79%|███████▉  | 19401/24420 [8:05:53<3:22:29,  2.42s/it]
 79%|███████▉  | 19402/24420 [8:05:55<2:59:01,  2.14s/it]
 79%|███████▉  | 19403/24420 [8:05:56<2:42:26,  1.94s/it]
 79%|███████▉  | 19404/24420 [8:05:58<2:30:52,  1.80s/it]
 79%|███████▉  | 19405/24420 [8:05:59<2:23:05,  1.71s/it]
                                                         
{'loss': 0.0752, 'learning_rate': 5e-05, 'epoch': 3.97}

 79%|███████▉  | 19405/24420 [8:05:59<2:23:05,  1.71s/it]
 79%|███████▉  | 19406/24420 [8:06:01<2:17:33,  1.65s/it]
 79%|███████▉  | 19407/24420 [8:06:02<2:13:38,  1.60s/it]
 79%|███████▉  | 19408/24420 [8:06:04<2:10:44,  1.57s/it]
 79%|███████▉  | 19409/24420 [8:06:05<2:08:41,  1.54s/it]
 79%|███████▉  | 19410/24420 [8:06:07<2:07:10,  1.52s/it]
                                                         
{'loss': 0.0776, 'learning_rate': 5e-05, 'epoch': 3.97}

 79%|███████▉  | 19410/24420 [8:06:07<2:07:10,  1.52s/it]
 79%|███████▉  | 19411/24420 [8:06:08<2:06:21,  1.51s/it]
 79%|███████▉  | 19412/24420 [8:06:10<2:05:56,  1.51s/it]
 79%|███████▉  | 19413/24420 [8:06:11<2:05:24,  1.50s/it]
 80%|███████▉  | 19414/24420 [8:06:13<2:04:56,  1.50s/it]
 80%|███████▉  | 19415/24420 [8:06:14<2:04:34,  1.49s/it]
                                                         
{'loss': 0.0815, 'learning_rate': 5e-05, 'epoch': 3.98}

 80%|███████▉  | 19415/24420 [8:06:14<2:04:34,  1.49s/it]
 80%|███████▉  | 19416/24420 [8:06:16<2:04:21,  1.49s/it]
 80%|███████▉  | 19417/24420 [8:06:17<2:04:22,  1.49s/it]
 80%|███████▉  | 19418/24420 [8:06:19<2:04:19,  1.49s/it]
 80%|███████▉  | 19419/24420 [8:06:20<2:04:08,  1.49s/it]
 80%|███████▉  | 19420/24420 [8:06:22<2:04:10,  1.49s/it]
                                                         
{'loss': 0.0795, 'learning_rate': 5e-05, 'epoch': 3.98}

 80%|███████▉  | 19420/24420 [8:06:22<2:04:10,  1.49s/it]
 80%|███████▉  | 19421/24420 [8:06:23<2:04:26,  1.49s/it]
 80%|███████▉  | 19422/24420 [8:06:25<2:04:00,  1.49s/it]
 80%|███████▉  | 19423/24420 [8:06:26<2:03:53,  1.49s/it]
 80%|███████▉  | 19424/24420 [8:06:27<2:03:48,  1.49s/it]
 80%|███████▉  | 19425/24420 [8:06:29<2:03:38,  1.49s/it]
                                                         
{'loss': 0.0807, 'learning_rate': 5e-05, 'epoch': 3.98}

 80%|███████▉  | 19425/24420 [8:06:29<2:03:38,  1.49s/it]
 80%|███████▉  | 19426/24420 [8:06:30<2:03:43,  1.49s/it]
 80%|███████▉  | 19427/24420 [8:06:32<2:03:36,  1.49s/it]
 80%|███████▉  | 19428/24420 [8:06:33<2:03:31,  1.48s/it]
 80%|███████▉  | 19429/24420 [8:06:35<2:03:44,  1.49s/it]
 80%|███████▉  | 19430/24420 [8:06:36<2:03:49,  1.49s/it]
                                                         
{'loss': 0.0769, 'learning_rate': 5e-05, 'epoch': 3.98}

 80%|███████▉  | 19430/24420 [8:06:36<2:03:49,  1.49s/it]
 80%|███████▉  | 19431/24420 [8:06:38<2:03:55,  1.49s/it]
 80%|███████▉  | 19432/24420 [8:06:39<2:03:54,  1.49s/it]
 80%|███████▉  | 19433/24420 [8:06:41<2:03:54,  1.49s/it]
 80%|███████▉  | 19434/24420 [8:06:42<2:04:04,  1.49s/it]
 80%|███████▉  | 19435/24420 [8:06:44<2:03:46,  1.49s/it]
                                                         
{'loss': 0.0761, 'learning_rate': 5e-05, 'epoch': 3.98}

 80%|███████▉  | 19435/24420 [8:06:44<2:03:46,  1.49s/it]
 80%|███████▉  | 19436/24420 [8:06:45<2:03:51,  1.49s/it]
 80%|███████▉  | 19437/24420 [8:06:47<2:03:33,  1.49s/it]
 80%|███████▉  | 19438/24420 [8:06:48<2:03:24,  1.49s/it]
 80%|███████▉  | 19439/24420 [8:06:50<2:03:16,  1.48s/it]
 80%|███████▉  | 19440/24420 [8:06:51<2:03:27,  1.49s/it]
                                                         
{'loss': 0.0695, 'learning_rate': 5e-05, 'epoch': 3.98}

 80%|███████▉  | 19440/24420 [8:06:51<2:03:27,  1.49s/it]
 80%|███████▉  | 19441/24420 [8:06:53<2:03:30,  1.49s/it]
 80%|███████▉  | 19442/24420 [8:06:54<2:03:33,  1.49s/it]
 80%|███████▉  | 19443/24420 [8:06:56<2:03:42,  1.49s/it]
 80%|███████▉  | 19444/24420 [8:06:57<2:03:25,  1.49s/it]
 80%|███████▉  | 19445/24420 [8:06:59<2:03:47,  1.49s/it]
                                                         
{'loss': 0.0669, 'learning_rate': 5e-05, 'epoch': 3.98}

 80%|███████▉  | 19445/24420 [8:06:59<2:03:47,  1.49s/it]
 80%|███████▉  | 19446/24420 [8:07:00<2:03:40,  1.49s/it]
 80%|███████▉  | 19447/24420 [8:07:02<2:03:37,  1.49s/it]
 80%|███████▉  | 19448/24420 [8:07:03<2:03:30,  1.49s/it]
 80%|███████▉  | 19449/24420 [8:07:05<2:03:22,  1.49s/it]
 80%|███████▉  | 19450/24420 [8:07:06<2:03:09,  1.49s/it]
                                                         
{'loss': 0.0817, 'learning_rate': 5e-05, 'epoch': 3.98}

 80%|███████▉  | 19450/24420 [8:07:06<2:03:09,  1.49s/it]
 80%|███████▉  | 19451/24420 [8:07:08<2:03:16,  1.49s/it]
 80%|███████▉  | 19452/24420 [8:07:09<2:03:26,  1.49s/it]
 80%|███████▉  | 19453/24420 [8:07:11<2:03:15,  1.49s/it]
 80%|███████▉  | 19454/24420 [8:07:12<2:03:02,  1.49s/it]
 80%|███████▉  | 19455/24420 [8:07:14<2:02:49,  1.48s/it]
                                                         
{'loss': 0.0884, 'learning_rate': 5e-05, 'epoch': 3.98}

 80%|███████▉  | 19455/24420 [8:07:14<2:02:49,  1.48s/it]
 80%|███████▉  | 19456/24420 [8:07:15<2:02:48,  1.48s/it]
 80%|███████▉  | 19457/24420 [8:07:17<2:02:40,  1.48s/it]
 80%|███████▉  | 19458/24420 [8:07:18<2:02:32,  1.48s/it]
 80%|███████▉  | 19459/24420 [8:07:20<2:02:39,  1.48s/it]
 80%|███████▉  | 19460/24420 [8:07:21<2:02:43,  1.48s/it]
                                                         
{'loss': 0.0731, 'learning_rate': 5e-05, 'epoch': 3.98}

 80%|███████▉  | 19460/24420 [8:07:21<2:02:43,  1.48s/it]
 80%|███████▉  | 19461/24420 [8:07:23<2:02:43,  1.48s/it]
 80%|███████▉  | 19462/24420 [8:07:24<2:02:59,  1.49s/it]
 80%|███████▉  | 19463/24420 [8:07:26<2:03:14,  1.49s/it]
 80%|███████▉  | 19464/24420 [8:07:27<2:03:09,  1.49s/it]
 80%|███████▉  | 19465/24420 [8:07:28<2:02:59,  1.49s/it]
                                                         
{'loss': 0.0756, 'learning_rate': 5e-05, 'epoch': 3.99}

 80%|███████▉  | 19465/24420 [8:07:28<2:02:59,  1.49s/it]
 80%|███████▉  | 19466/24420 [8:07:30<2:03:05,  1.49s/it]
 80%|███████▉  | 19467/24420 [8:07:31<2:02:58,  1.49s/it]
 80%|███████▉  | 19468/24420 [8:07:33<2:02:43,  1.49s/it]
 80%|███████▉  | 19469/24420 [8:07:34<2:02:43,  1.49s/it]
 80%|███████▉  | 19470/24420 [8:07:36<2:02:36,  1.49s/it]
                                                         
{'loss': 0.0702, 'learning_rate': 5e-05, 'epoch': 3.99}

 80%|███████▉  | 19470/24420 [8:07:36<2:02:36,  1.49s/it]
 80%|███████▉  | 19471/24420 [8:07:37<2:02:52,  1.49s/it]
 80%|███████▉  | 19472/24420 [8:07:39<2:02:44,  1.49s/it]
 80%|███████▉  | 19473/24420 [8:07:40<2:02:49,  1.49s/it]
 80%|███████▉  | 19474/24420 [8:07:42<2:02:35,  1.49s/it]
 80%|███████▉  | 19475/24420 [8:07:43<2:02:57,  1.49s/it]
                                                         
{'loss': 0.0741, 'learning_rate': 5e-05, 'epoch': 3.99}

 80%|███████▉  | 19475/24420 [8:07:43<2:02:57,  1.49s/it]
 80%|███████▉  | 19476/24420 [8:07:45<2:02:51,  1.49s/it]
 80%|███████▉  | 19477/24420 [8:07:46<2:02:46,  1.49s/it]
 80%|███████▉  | 19478/24420 [8:07:48<2:02:43,  1.49s/it]
 80%|███████▉  | 19479/24420 [8:07:49<2:02:34,  1.49s/it]
 80%|███████▉  | 19480/24420 [8:07:51<2:02:34,  1.49s/it]
                                                         
{'loss': 0.0716, 'learning_rate': 5e-05, 'epoch': 3.99}

 80%|███████▉  | 19480/24420 [8:07:51<2:02:34,  1.49s/it]
 80%|███████▉  | 19481/24420 [8:07:52<2:02:58,  1.49s/it]
 80%|███████▉  | 19482/24420 [8:07:54<2:02:57,  1.49s/it]
 80%|███████▉  | 19483/24420 [8:07:55<2:02:42,  1.49s/it]
 80%|███████▉  | 19484/24420 [8:07:57<2:02:24,  1.49s/it]
 80%|███████▉  | 19485/24420 [8:07:58<2:02:26,  1.49s/it]
                                                         
{'loss': 0.0705, 'learning_rate': 5e-05, 'epoch': 3.99}

 80%|███████▉  | 19485/24420 [8:07:58<2:02:26,  1.49s/it]
 80%|███████▉  | 19486/24420 [8:08:00<2:02:34,  1.49s/it]
 80%|███████▉  | 19487/24420 [8:08:01<2:02:32,  1.49s/it]
 80%|███████▉  | 19488/24420 [8:08:03<2:02:23,  1.49s/it]
 80%|███████▉  | 19489/24420 [8:08:04<2:02:17,  1.49s/it]
 80%|███████▉  | 19490/24420 [8:08:06<2:02:09,  1.49s/it]
                                                         
{'loss': 0.0881, 'learning_rate': 5e-05, 'epoch': 3.99}

 80%|███████▉  | 19490/24420 [8:08:06<2:02:09,  1.49s/it]
 80%|███████▉  | 19491/24420 [8:08:07<2:02:08,  1.49s/it]
 80%|███████▉  | 19492/24420 [8:08:09<2:02:05,  1.49s/it]
 80%|███████▉  | 19493/24420 [8:08:10<2:02:02,  1.49s/it]
 80%|███████▉  | 19494/24420 [8:08:12<2:02:00,  1.49s/it]
 80%|███████▉  | 19495/24420 [8:08:13<2:02:07,  1.49s/it]
                                                         
{'loss': 0.0666, 'learning_rate': 5e-05, 'epoch': 3.99}

 80%|███████▉  | 19495/24420 [8:08:13<2:02:07,  1.49s/it]
 80%|███████▉  | 19496/24420 [8:08:15<2:02:12,  1.49s/it]
 80%|███████▉  | 19497/24420 [8:08:16<2:02:10,  1.49s/it]
 80%|███████▉  | 19498/24420 [8:08:18<2:02:11,  1.49s/it]
 80%|███████▉  | 19499/24420 [8:08:19<2:02:01,  1.49s/it]
 80%|███████▉  | 19500/24420 [8:08:21<2:02:03,  1.49s/it]
                                                         
{'loss': 0.071, 'learning_rate': 5e-05, 'epoch': 3.99}

 80%|███████▉  | 19500/24420 [8:08:21<2:02:03,  1.49s/it]
 80%|███████▉  | 19501/24420 [8:08:22<2:02:06,  1.49s/it]
 80%|███████▉  | 19502/24420 [8:08:24<2:02:13,  1.49s/it]
 80%|███████▉  | 19503/24420 [8:08:25<2:02:11,  1.49s/it]
 80%|███████▉  | 19504/24420 [8:08:27<2:02:12,  1.49s/it]
 80%|███████▉  | 19505/24420 [8:08:28<2:02:15,  1.49s/it]
                                                         
{'loss': 0.0766, 'learning_rate': 5e-05, 'epoch': 3.99}

 80%|███████▉  | 19505/24420 [8:08:28<2:02:15,  1.49s/it]
 80%|███████▉  | 19506/24420 [8:08:30<2:02:32,  1.50s/it]
 80%|███████▉  | 19507/24420 [8:08:31<2:02:27,  1.50s/it]
 80%|███████▉  | 19508/24420 [8:08:33<2:02:18,  1.49s/it]
 80%|███████▉  | 19509/24420 [8:08:34<2:02:02,  1.49s/it]
 80%|███████▉  | 19510/24420 [8:08:36<2:02:07,  1.49s/it]
                                                         
{'loss': 0.0783, 'learning_rate': 5e-05, 'epoch': 3.99}

 80%|███████▉  | 19510/24420 [8:08:36<2:02:07,  1.49s/it]
 80%|███████▉  | 19511/24420 [8:08:37<2:02:03,  1.49s/it]
 80%|███████▉  | 19512/24420 [8:08:39<2:01:57,  1.49s/it]
 80%|███████▉  | 19513/24420 [8:08:40<2:02:06,  1.49s/it]
 80%|███████▉  | 19514/24420 [8:08:42<2:02:09,  1.49s/it]
 80%|███████▉  | 19515/24420 [8:08:43<2:01:59,  1.49s/it]
                                                         
{'loss': 0.0806, 'learning_rate': 5e-05, 'epoch': 4.0}

 80%|███████▉  | 19515/24420 [8:08:43<2:01:59,  1.49s/it]
 80%|███████▉  | 19516/24420 [8:08:44<2:01:58,  1.49s/it]
 80%|███████▉  | 19517/24420 [8:08:46<2:02:09,  1.49s/it]
 80%|███████▉  | 19518/24420 [8:08:47<2:02:13,  1.50s/it]
 80%|███████▉  | 19519/24420 [8:08:49<2:02:20,  1.50s/it]
 80%|███████▉  | 19520/24420 [8:08:50<2:02:07,  1.50s/it]
                                                         
{'loss': 0.087, 'learning_rate': 5e-05, 'epoch': 4.0}

 80%|███████▉  | 19520/24420 [8:08:50<2:02:07,  1.50s/it]
 80%|███████▉  | 19521/24420 [8:08:52<2:01:53,  1.49s/it]
 80%|███████▉  | 19522/24420 [8:08:53<2:01:42,  1.49s/it]
 80%|███████▉  | 19523/24420 [8:08:55<2:01:31,  1.49s/it]
 80%|███████▉  | 19524/24420 [8:08:56<2:01:33,  1.49s/it]
 80%|███████▉  | 19525/24420 [8:08:58<2:01:56,  1.49s/it]
                                                         
{'loss': 0.0681, 'learning_rate': 5e-05, 'epoch': 4.0}

 80%|███████▉  | 19525/24420 [8:08:58<2:01:56,  1.49s/it]
 80%|███████▉  | 19526/24420 [8:08:59<2:01:43,  1.49s/it]
 80%|███████▉  | 19527/24420 [8:09:01<2:01:39,  1.49s/it]
 80%|███████▉  | 19528/24420 [8:09:02<2:01:56,  1.50s/it]
 80%|███████▉  | 19529/24420 [8:09:04<2:01:42,  1.49s/it]
 80%|███████▉  | 19530/24420 [8:09:05<2:01:42,  1.49s/it]
                                                         
{'loss': 0.0771, 'learning_rate': 5e-05, 'epoch': 4.0}

 80%|███████▉  | 19530/24420 [8:09:05<2:01:42,  1.49s/it]
 80%|███████▉  | 19531/24420 [8:09:07<2:01:39,  1.49s/it]
 80%|███████▉  | 19532/24420 [8:09:08<2:01:52,  1.50s/it]
 80%|███████▉  | 19533/24420 [8:09:10<2:01:37,  1.49s/it]
 80%|███████▉  | 19534/24420 [8:09:11<2:01:32,  1.49s/it]
 80%|███████▉  | 19535/24420 [8:09:13<2:01:10,  1.49s/it]
                                                         
{'loss': 0.0786, 'learning_rate': 5e-05, 'epoch': 4.0}

 80%|███████▉  | 19535/24420 [8:09:13<2:01:10,  1.49s/it]
 80%|████████  | 19536/24420 [8:09:14<2:01:23,  1.49s/it]
 80%|████████  | 19537/24420 [8:09:16<2:01:49,  1.50s/it]
 80%|████████  | 19538/24420 [8:09:17<2:01:47,  1.50s/it]
 80%|████████  | 19539/24420 [8:09:19<2:01:38,  1.50s/it]
 80%|████████  | 19540/24420 [8:09:20<2:01:44,  1.50s/it]
                                                         
{'loss': 0.0887, 'learning_rate': 5e-05, 'epoch': 4.0}

 80%|████████  | 19540/24420 [8:09:20<2:01:44,  1.50s/it]
 80%|████████  | 19541/24420 [8:09:22<2:01:39,  1.50s/it]
 80%|████████  | 19542/24420 [8:09:23<2:01:21,  1.49s/it]
 80%|████████  | 19543/24420 [8:09:25<2:01:25,  1.49s/it]
 80%|████████  | 19544/24420 [8:09:26<2:01:26,  1.49s/it]
 80%|████████  | 19545/24420 [8:09:28<2:01:16,  1.49s/it]
                                                         
{'loss': 0.0636, 'learning_rate': 5e-05, 'epoch': 4.0}

 80%|████████  | 19545/24420 [8:09:28<2:01:16,  1.49s/it]
 80%|████████  | 19546/24420 [8:09:29<2:01:16,  1.49s/it]
 80%|████████  | 19547/24420 [8:09:31<2:01:09,  1.49s/it]
 80%|████████  | 19548/24420 [8:09:32<2:01:14,  1.49s/it]
 80%|████████  | 19549/24420 [8:09:34<2:01:09,  1.49s/it]
 80%|████████  | 19550/24420 [8:09:35<2:01:11,  1.49s/it]
                                                         
{'loss': 0.0815, 'learning_rate': 5e-05, 'epoch': 4.0}

 80%|████████  | 19550/24420 [8:09:35<2:01:11,  1.49s/it]
 80%|████████  | 19551/24420 [8:09:37<2:00:56,  1.49s/it]
 80%|████████  | 19552/24420 [8:09:38<2:00:43,  1.49s/it]
 80%|████████  | 19553/24420 [8:09:40<2:00:53,  1.49s/it]
 80%|████████  | 19554/24420 [8:09:41<2:00:52,  1.49s/it]
 80%|████████  | 19555/24420 [8:09:43<2:00:51,  1.49s/it]
                                                         
{'loss': 0.0673, 'learning_rate': 5e-05, 'epoch': 4.0}

 80%|████████  | 19555/24420 [8:09:43<2:00:51,  1.49s/it]
 80%|████████  | 19556/24420 [8:09:44<2:00:46,  1.49s/it]
 80%|████████  | 19557/24420 [8:09:46<2:00:38,  1.49s/it]
 80%|████████  | 19558/24420 [8:09:47<2:00:37,  1.49s/it]
 80%|████████  | 19559/24420 [8:09:49<2:00:38,  1.49s/it]
 80%|████████  | 19560/24420 [8:09:50<2:00:50,  1.49s/it]
                                                         
{'loss': 0.072, 'learning_rate': 5e-05, 'epoch': 4.0}

 80%|████████  | 19560/24420 [8:09:50<2:00:50,  1.49s/it]
 80%|████████  | 19561/24420 [8:09:52<2:00:45,  1.49s/it]
 80%|████████  | 19562/24420 [8:09:53<2:00:36,  1.49s/it]
 80%|████████  | 19563/24420 [8:09:55<2:00:33,  1.49s/it]
 80%|████████  | 19564/24420 [8:09:56<2:00:27,  1.49s/it]
 80%|████████  | 19565/24420 [8:09:58<2:00:21,  1.49s/it]
                                                         
{'loss': 0.0691, 'learning_rate': 5e-05, 'epoch': 4.01}

 80%|████████  | 19565/24420 [8:09:58<2:00:21,  1.49s/it]
 80%|████████  | 19566/24420 [8:09:59<2:00:24,  1.49s/it]
 80%|████████  | 19567/24420 [8:10:01<2:00:19,  1.49s/it]
 80%|████████  | 19568/24420 [8:10:02<2:00:12,  1.49s/it]
 80%|████████  | 19569/24420 [8:10:04<2:00:04,  1.49s/it]
 80%|████████  | 19570/24420 [8:10:05<2:00:04,  1.49s/it]
                                                         
{'loss': 0.0722, 'learning_rate': 5e-05, 'epoch': 4.01}

 80%|████████  | 19570/24420 [8:10:05<2:00:04,  1.49s/it]
 80%|████████  | 19571/24420 [8:10:07<2:00:15,  1.49s/it]
 80%|████████  | 19572/24420 [8:10:08<2:00:25,  1.49s/it]
 80%|████████  | 19573/24420 [8:10:10<2:00:18,  1.49s/it]
 80%|████████  | 19574/24420 [8:10:11<2:00:19,  1.49s/it]
 80%|████████  | 19575/24420 [8:10:13<2:00:31,  1.49s/it]
                                                         
{'loss': 0.0872, 'learning_rate': 5e-05, 'epoch': 4.01}

 80%|████████  | 19575/24420 [8:10:13<2:00:31,  1.49s/it]
 80%|████████  | 19576/24420 [8:10:14<2:00:31,  1.49s/it]
 80%|████████  | 19577/24420 [8:10:15<2:00:16,  1.49s/it]
 80%|████████  | 19578/24420 [8:10:17<2:00:19,  1.49s/it]
 80%|████████  | 19579/24420 [8:10:18<2:00:10,  1.49s/it]
 80%|████████  | 19580/24420 [8:10:20<2:00:00,  1.49s/it]
                                                         
{'loss': 0.0735, 'learning_rate': 5e-05, 'epoch': 4.01}

 80%|████████  | 19580/24420 [8:10:20<2:00:00,  1.49s/it]
 80%|████████  | 19581/24420 [8:10:21<2:00:07,  1.49s/it]
 80%|████████  | 19582/24420 [8:10:23<2:00:21,  1.49s/it]
 80%|████████  | 19583/24420 [8:10:24<2:00:26,  1.49s/it]
 80%|████████  | 19584/24420 [8:10:26<2:00:15,  1.49s/it]
 80%|████████  | 19585/24420 [8:10:27<2:00:11,  1.49s/it]
                                                         
{'loss': 0.0695, 'learning_rate': 5e-05, 'epoch': 4.01}

 80%|████████  | 19585/24420 [8:10:27<2:00:11,  1.49s/it]
 80%|████████  | 19586/24420 [8:10:29<2:00:09,  1.49s/it]
 80%|████████  | 19587/24420 [8:10:30<1:59:58,  1.49s/it]
 80%|████████  | 19588/24420 [8:10:32<1:59:53,  1.49s/it]
 80%|████████  | 19589/24420 [8:10:33<1:59:56,  1.49s/it]
 80%|████████  | 19590/24420 [8:10:35<1:59:42,  1.49s/it]
                                                         
{'loss': 0.0708, 'learning_rate': 5e-05, 'epoch': 4.01}

 80%|████████  | 19590/24420 [8:10:35<1:59:42,  1.49s/it]
 80%|████████  | 19591/24420 [8:10:36<1:59:40,  1.49s/it]
 80%|████████  | 19592/24420 [8:10:38<1:59:40,  1.49s/it]
 80%|████████  | 19593/24420 [8:10:39<1:59:47,  1.49s/it]
 80%|████████  | 19594/24420 [8:10:41<1:59:56,  1.49s/it]
 80%|████████  | 19595/24420 [8:10:42<1:59:53,  1.49s/it]
                                                         
{'loss': 0.0818, 'learning_rate': 5e-05, 'epoch': 4.01}

 80%|████████  | 19595/24420 [8:10:42<1:59:53,  1.49s/it]
 80%|████████  | 19596/24420 [8:10:44<2:00:01,  1.49s/it]
 80%|████████  | 19597/24420 [8:10:45<2:00:08,  1.49s/it]
 80%|████████  | 19598/24420 [8:10:47<2:00:02,  1.49s/it]
 80%|████████  | 19599/24420 [8:10:48<1:59:54,  1.49s/it]
 80%|████████  | 19600/24420 [8:10:50<1:59:52,  1.49s/it]
                                                         
{'loss': 0.0754, 'learning_rate': 5e-05, 'epoch': 4.01}

 80%|████████  | 19600/24420 [8:10:50<1:59:52,  1.49s/it]
 80%|████████  | 19601/24420 [8:10:55<3:19:03,  2.48s/it]
 80%|████████  | 19602/24420 [8:10:56<2:55:07,  2.18s/it]
 80%|████████  | 19603/24420 [8:10:58<2:38:33,  1.98s/it]
 80%|████████  | 19604/24420 [8:10:59<2:26:42,  1.83s/it]
 80%|████████  | 19605/24420 [8:11:01<2:18:45,  1.73s/it]
                                                         
{'loss': 0.0767, 'learning_rate': 5e-05, 'epoch': 4.01}

 80%|████████  | 19605/24420 [8:11:01<2:18:45,  1.73s/it]
 80%|████████  | 19606/24420 [8:11:02<2:13:03,  1.66s/it]
 80%|████████  | 19607/24420 [8:11:03<2:08:59,  1.61s/it]
 80%|████████  | 19608/24420 [8:11:05<2:06:23,  1.58s/it]
 80%|████████  | 19609/24420 [8:11:06<2:04:21,  1.55s/it]
 80%|████████  | 19610/24420 [8:11:08<2:02:41,  1.53s/it]
                                                         
{'loss': 0.0668, 'learning_rate': 5e-05, 'epoch': 4.02}

 80%|████████  | 19610/24420 [8:11:08<2:02:41,  1.53s/it]
 80%|████████  | 19611/24420 [8:11:09<2:01:30,  1.52s/it]
 80%|████████  | 19612/24420 [8:11:11<2:00:44,  1.51s/it]
 80%|████████  | 19613/24420 [8:11:12<2:00:17,  1.50s/it]
 80%|████████  | 19614/24420 [8:11:14<1:59:53,  1.50s/it]
 80%|████████  | 19615/24420 [8:11:15<1:59:49,  1.50s/it]
                                                         
{'loss': 0.0627, 'learning_rate': 5e-05, 'epoch': 4.02}

 80%|████████  | 19615/24420 [8:11:15<1:59:49,  1.50s/it]
 80%|████████  | 19616/24420 [8:11:17<1:59:30,  1.49s/it]
 80%|████████  | 19617/24420 [8:11:18<1:59:20,  1.49s/it]
 80%|████████  | 19618/24420 [8:11:20<1:59:09,  1.49s/it]
 80%|████████  | 19619/24420 [8:11:21<1:59:03,  1.49s/it]
 80%|████████  | 19620/24420 [8:11:23<1:59:10,  1.49s/it]
                                                         
{'loss': 0.0784, 'learning_rate': 5e-05, 'epoch': 4.02}

 80%|████████  | 19620/24420 [8:11:23<1:59:10,  1.49s/it]
 80%|████████  | 19621/24420 [8:11:24<1:59:22,  1.49s/it]
 80%|████████  | 19622/24420 [8:11:26<1:59:07,  1.49s/it]
 80%|████████  | 19623/24420 [8:11:27<1:58:49,  1.49s/it]
 80%|████████  | 19624/24420 [8:11:29<1:59:14,  1.49s/it]
 80%|████████  | 19625/24420 [8:11:30<1:58:58,  1.49s/it]
                                                         
{'loss': 0.0826, 'learning_rate': 5e-05, 'epoch': 4.02}

 80%|████████  | 19625/24420 [8:11:30<1:58:58,  1.49s/it]
 80%|████████  | 19626/24420 [8:11:32<1:59:04,  1.49s/it]
 80%|████████  | 19627/24420 [8:11:33<1:59:09,  1.49s/it]
 80%|████████  | 19628/24420 [8:11:35<1:59:03,  1.49s/it]
 80%|████████  | 19629/24420 [8:11:36<1:58:52,  1.49s/it]
 80%|████████  | 19630/24420 [8:11:38<1:58:48,  1.49s/it]
                                                         
{'loss': 0.0855, 'learning_rate': 5e-05, 'epoch': 4.02}

 80%|████████  | 19630/24420 [8:11:38<1:58:48,  1.49s/it]
 80%|████████  | 19631/24420 [8:11:39<1:58:49,  1.49s/it]
 80%|████████  | 19632/24420 [8:11:41<1:58:39,  1.49s/it]
 80%|████████  | 19633/24420 [8:11:42<1:58:44,  1.49s/it]
 80%|████████  | 19634/24420 [8:11:44<1:58:46,  1.49s/it]
 80%|████████  | 19635/24420 [8:11:45<1:58:48,  1.49s/it]
                                                         
{'loss': 0.0766, 'learning_rate': 5e-05, 'epoch': 4.02}

 80%|████████  | 19635/24420 [8:11:45<1:58:48,  1.49s/it]
 80%|████████  | 19636/24420 [8:11:47<1:58:32,  1.49s/it]
 80%|████████  | 19637/24420 [8:11:48<1:58:55,  1.49s/it]
 80%|████████  | 19638/24420 [8:11:50<1:58:43,  1.49s/it]
 80%|████████  | 19639/24420 [8:11:51<1:58:37,  1.49s/it]
 80%|████████  | 19640/24420 [8:11:53<1:58:49,  1.49s/it]
                                                         
{'loss': 0.067, 'learning_rate': 5e-05, 'epoch': 4.02}

 80%|████████  | 19640/24420 [8:11:53<1:58:49,  1.49s/it]
 80%|████████  | 19641/24420 [8:11:54<1:58:45,  1.49s/it]
 80%|████████  | 19642/24420 [8:11:56<1:58:29,  1.49s/it]
 80%|████████  | 19643/24420 [8:11:57<1:58:22,  1.49s/it]
 80%|████████  | 19644/24420 [8:11:59<1:58:33,  1.49s/it]
 80%|████████  | 19645/24420 [8:12:00<1:58:31,  1.49s/it]
                                                         
{'loss': 0.0749, 'learning_rate': 5e-05, 'epoch': 4.02}

 80%|████████  | 19645/24420 [8:12:00<1:58:31,  1.49s/it]
 80%|████████  | 19646/24420 [8:12:02<1:58:30,  1.49s/it]
 80%|████████  | 19647/24420 [8:12:03<1:58:33,  1.49s/it]
 80%|████████  | 19648/24420 [8:12:05<1:58:19,  1.49s/it]
 80%|████████  | 19649/24420 [8:12:06<1:58:23,  1.49s/it]
 80%|████████  | 19650/24420 [8:12:08<1:58:20,  1.49s/it]
                                                         
{'loss': 0.0699, 'learning_rate': 5e-05, 'epoch': 4.02}

 80%|████████  | 19650/24420 [8:12:08<1:58:20,  1.49s/it]
 80%|████████  | 19651/24420 [8:12:09<1:58:28,  1.49s/it]
 80%|████████  | 19652/24420 [8:12:11<1:58:21,  1.49s/it]
 80%|████████  | 19653/24420 [8:12:12<1:58:26,  1.49s/it]
 80%|████████  | 19654/24420 [8:12:13<1:58:37,  1.49s/it]
 80%|████████  | 19655/24420 [8:12:15<1:58:23,  1.49s/it]
                                                         
{'loss': 0.0694, 'learning_rate': 5e-05, 'epoch': 4.02}

 80%|████████  | 19655/24420 [8:12:15<1:58:23,  1.49s/it]
 80%|████████  | 19656/24420 [8:12:16<1:58:27,  1.49s/it]
 80%|████████  | 19657/24420 [8:12:18<1:58:22,  1.49s/it]
 80%|████████  | 19658/24420 [8:12:19<1:58:33,  1.49s/it]
 81%|████████  | 19659/24420 [8:12:21<1:58:25,  1.49s/it]
 81%|████████  | 19660/24420 [8:12:22<1:58:11,  1.49s/it]
                                                         
{'loss': 0.0777, 'learning_rate': 5e-05, 'epoch': 4.03}

 81%|████████  | 19660/24420 [8:12:22<1:58:11,  1.49s/it]
 81%|████████  | 19661/24420 [8:12:24<1:58:12,  1.49s/it]
 81%|████████  | 19662/24420 [8:12:25<1:58:02,  1.49s/it]
 81%|████████  | 19663/24420 [8:12:27<1:58:10,  1.49s/it]
 81%|████████  | 19664/24420 [8:12:28<1:58:16,  1.49s/it]
 81%|████████  | 19665/24420 [8:12:30<1:58:14,  1.49s/it]
                                                         
{'loss': 0.0714, 'learning_rate': 5e-05, 'epoch': 4.03}

 81%|████████  | 19665/24420 [8:12:30<1:58:14,  1.49s/it]
 81%|████████  | 19666/24420 [8:12:31<1:58:18,  1.49s/it]
 81%|████████  | 19667/24420 [8:12:33<1:58:12,  1.49s/it]
 81%|████████  | 19668/24420 [8:12:34<1:58:04,  1.49s/it]
 81%|████████  | 19669/24420 [8:12:36<1:58:05,  1.49s/it]
 81%|████████  | 19670/24420 [8:12:37<1:57:53,  1.49s/it]
                                                         
{'loss': 0.0875, 'learning_rate': 5e-05, 'epoch': 4.03}

 81%|████████  | 19670/24420 [8:12:37<1:57:53,  1.49s/it]
 81%|████████  | 19671/24420 [8:12:39<1:58:18,  1.49s/it]
 81%|████████  | 19672/24420 [8:12:40<1:58:08,  1.49s/it]
 81%|████████  | 19673/24420 [8:12:42<1:58:04,  1.49s/it]
 81%|████████  | 19674/24420 [8:12:43<1:57:51,  1.49s/it]
 81%|████████  | 19675/24420 [8:12:45<1:57:52,  1.49s/it]
                                                         
{'loss': 0.0692, 'learning_rate': 5e-05, 'epoch': 4.03}

 81%|████████  | 19675/24420 [8:12:45<1:57:52,  1.49s/it]
 81%|████████  | 19676/24420 [8:12:46<1:57:47,  1.49s/it]
 81%|████████  | 19677/24420 [8:12:48<1:57:46,  1.49s/it]
 81%|████████  | 19678/24420 [8:12:49<1:57:54,  1.49s/it]
 81%|████████  | 19679/24420 [8:12:51<1:57:52,  1.49s/it]
 81%|████████  | 19680/24420 [8:12:52<1:57:38,  1.49s/it]
                                                         
{'loss': 0.0763, 'learning_rate': 5e-05, 'epoch': 4.03}

 81%|████████  | 19680/24420 [8:12:52<1:57:38,  1.49s/it]
 81%|████████  | 19681/24420 [8:12:54<1:57:39,  1.49s/it]
 81%|████████  | 19682/24420 [8:12:55<1:57:48,  1.49s/it]
 81%|████████  | 19683/24420 [8:12:57<1:57:29,  1.49s/it]
 81%|████████  | 19684/24420 [8:12:58<1:57:39,  1.49s/it]
 81%|████████  | 19685/24420 [8:13:00<1:57:31,  1.49s/it]
                                                         
{'loss': 0.0736, 'learning_rate': 5e-05, 'epoch': 4.03}

 81%|████████  | 19685/24420 [8:13:00<1:57:31,  1.49s/it]
 81%|████████  | 19686/24420 [8:13:01<1:57:30,  1.49s/it]
 81%|████████  | 19687/24420 [8:13:03<1:57:38,  1.49s/it]
 81%|████████  | 19688/24420 [8:13:04<1:57:51,  1.49s/it]
 81%|████████  | 19689/24420 [8:13:06<1:57:38,  1.49s/it]
 81%|████████  | 19690/24420 [8:13:07<1:57:42,  1.49s/it]
                                                         
{'loss': 0.0792, 'learning_rate': 5e-05, 'epoch': 4.03}

 81%|████████  | 19690/24420 [8:13:07<1:57:42,  1.49s/it]
 81%|████████  | 19691/24420 [8:13:09<1:57:44,  1.49s/it]
 81%|████████  | 19692/24420 [8:13:10<1:57:51,  1.50s/it]
 81%|████████  | 19693/24420 [8:13:12<1:57:28,  1.49s/it]
 81%|████████  | 19694/24420 [8:13:13<1:57:41,  1.49s/it]
 81%|████████  | 19695/24420 [8:13:15<1:57:28,  1.49s/it]
                                                         
{'loss': 0.0791, 'learning_rate': 5e-05, 'epoch': 4.03}

 81%|████████  | 19695/24420 [8:13:15<1:57:28,  1.49s/it]
 81%|████████  | 19696/24420 [8:13:16<1:57:19,  1.49s/it]
 81%|████████  | 19697/24420 [8:13:18<1:57:02,  1.49s/it]
 81%|████████  | 19698/24420 [8:13:19<1:57:08,  1.49s/it]
 81%|████████  | 19699/24420 [8:13:21<1:57:20,  1.49s/it]
 81%|████████  | 19700/24420 [8:13:22<1:57:09,  1.49s/it]
                                                         
{'loss': 0.0656, 'learning_rate': 5e-05, 'epoch': 4.03}

 81%|████████  | 19700/24420 [8:13:22<1:57:09,  1.49s/it]
 81%|████████  | 19701/24420 [8:13:24<1:57:13,  1.49s/it]
 81%|████████  | 19702/24420 [8:13:25<1:57:23,  1.49s/it]
 81%|████████  | 19703/24420 [8:13:27<1:57:13,  1.49s/it]
 81%|████████  | 19704/24420 [8:13:28<1:57:09,  1.49s/it]
 81%|████████  | 19705/24420 [8:13:30<1:57:06,  1.49s/it]
                                                         
{'loss': 0.0645, 'learning_rate': 5e-05, 'epoch': 4.03}

 81%|████████  | 19705/24420 [8:13:30<1:57:06,  1.49s/it]
 81%|████████  | 19706/24420 [8:13:31<1:57:06,  1.49s/it]
 81%|████████  | 19707/24420 [8:13:33<1:57:05,  1.49s/it]
 81%|████████  | 19708/24420 [8:13:34<1:56:51,  1.49s/it]
 81%|████████  | 19709/24420 [8:13:35<1:56:50,  1.49s/it]
 81%|████████  | 19710/24420 [8:13:37<1:56:37,  1.49s/it]
                                                         
{'loss': 0.0719, 'learning_rate': 5e-05, 'epoch': 4.04}

 81%|████████  | 19710/24420 [8:13:37<1:56:37,  1.49s/it]
 81%|████████  | 19711/24420 [8:13:38<1:56:37,  1.49s/it]
 81%|████████  | 19712/24420 [8:13:40<1:56:48,  1.49s/it]
 81%|████████  | 19713/24420 [8:13:41<1:57:02,  1.49s/it]
 81%|████████  | 19714/24420 [8:13:43<1:56:53,  1.49s/it]
 81%|████████  | 19715/24420 [8:13:44<1:56:48,  1.49s/it]
                                                         
{'loss': 0.0679, 'learning_rate': 5e-05, 'epoch': 4.04}

 81%|████████  | 19715/24420 [8:13:44<1:56:48,  1.49s/it]
 81%|████████  | 19716/24420 [8:13:46<1:56:54,  1.49s/it]
 81%|████████  | 19717/24420 [8:13:47<1:56:45,  1.49s/it]
 81%|████████  | 19718/24420 [8:13:49<1:56:38,  1.49s/it]
 81%|████████  | 19719/24420 [8:13:50<1:56:39,  1.49s/it]
 81%|████████  | 19720/24420 [8:13:52<1:56:36,  1.49s/it]
                                                         
{'loss': 0.0712, 'learning_rate': 5e-05, 'epoch': 4.04}

 81%|████████  | 19720/24420 [8:13:52<1:56:36,  1.49s/it]
 81%|████████  | 19721/24420 [8:13:53<1:56:39,  1.49s/it]
 81%|████████  | 19722/24420 [8:13:55<1:56:50,  1.49s/it]
 81%|████████  | 19723/24420 [8:13:56<1:56:39,  1.49s/it]
 81%|████████  | 19724/24420 [8:13:58<1:56:42,  1.49s/it]
 81%|████████  | 19725/24420 [8:13:59<1:56:39,  1.49s/it]
                                                         
{'loss': 0.0682, 'learning_rate': 5e-05, 'epoch': 4.04}

 81%|████████  | 19725/24420 [8:13:59<1:56:39,  1.49s/it]
 81%|████████  | 19726/24420 [8:14:01<1:56:32,  1.49s/it]
 81%|████████  | 19727/24420 [8:14:02<1:56:32,  1.49s/it]
 81%|████████  | 19728/24420 [8:14:04<1:56:26,  1.49s/it]
 81%|████████  | 19729/24420 [8:14:05<1:56:15,  1.49s/it]
 81%|████████  | 19730/24420 [8:14:07<1:56:11,  1.49s/it]
                                                         
{'loss': 0.0749, 'learning_rate': 5e-05, 'epoch': 4.04}

 81%|████████  | 19730/24420 [8:14:07<1:56:11,  1.49s/it]
 81%|████████  | 19731/24420 [8:14:08<1:56:24,  1.49s/it]
 81%|████████  | 19732/24420 [8:14:10<1:56:16,  1.49s/it]
 81%|████████  | 19733/24420 [8:14:11<1:56:19,  1.49s/it]
 81%|████████  | 19734/24420 [8:14:13<1:56:20,  1.49s/it]
 81%|████████  | 19735/24420 [8:14:14<1:56:26,  1.49s/it]
                                                         
{'loss': 0.0664, 'learning_rate': 5e-05, 'epoch': 4.04}

 81%|████████  | 19735/24420 [8:14:14<1:56:26,  1.49s/it]
 81%|████████  | 19736/24420 [8:14:16<1:56:16,  1.49s/it]
 81%|████████  | 19737/24420 [8:14:17<1:56:07,  1.49s/it]
 81%|████████  | 19738/24420 [8:14:19<1:55:57,  1.49s/it]
 81%|████████  | 19739/24420 [8:14:20<1:55:55,  1.49s/it]
 81%|████████  | 19740/24420 [8:14:22<1:56:09,  1.49s/it]
                                                         
{'loss': 0.067, 'learning_rate': 5e-05, 'epoch': 4.04}

 81%|████████  | 19740/24420 [8:14:22<1:56:09,  1.49s/it]
 81%|████████  | 19741/24420 [8:14:23<1:56:28,  1.49s/it]
 81%|████████  | 19742/24420 [8:14:25<1:56:14,  1.49s/it]
 81%|████████  | 19743/24420 [8:14:26<1:56:04,  1.49s/it]
 81%|████████  | 19744/24420 [8:14:28<1:56:06,  1.49s/it]
 81%|████████  | 19745/24420 [8:14:29<1:56:16,  1.49s/it]
                                                         
{'loss': 0.0692, 'learning_rate': 5e-05, 'epoch': 4.04}

 81%|████████  | 19745/24420 [8:14:29<1:56:16,  1.49s/it]
 81%|████████  | 19746/24420 [8:14:31<1:56:19,  1.49s/it]
 81%|████████  | 19747/24420 [8:14:32<1:56:03,  1.49s/it]
 81%|████████  | 19748/24420 [8:14:34<1:56:14,  1.49s/it]
 81%|████████  | 19749/24420 [8:14:35<1:56:14,  1.49s/it]
 81%|████████  | 19750/24420 [8:14:37<1:56:08,  1.49s/it]
                                                         
{'loss': 0.0684, 'learning_rate': 5e-05, 'epoch': 4.04}

 81%|████████  | 19750/24420 [8:14:37<1:56:08,  1.49s/it]
 81%|████████  | 19751/24420 [8:14:38<1:55:58,  1.49s/it]
 81%|████████  | 19752/24420 [8:14:40<1:56:12,  1.49s/it]
 81%|████████  | 19753/24420 [8:14:41<1:56:03,  1.49s/it]
 81%|████████  | 19754/24420 [8:14:43<1:56:23,  1.50s/it]
 81%|████████  | 19755/24420 [8:14:44<1:56:07,  1.49s/it]
                                                         
{'loss': 0.0665, 'learning_rate': 5e-05, 'epoch': 4.04}

 81%|████████  | 19755/24420 [8:14:44<1:56:07,  1.49s/it]
 81%|████████  | 19756/24420 [8:14:46<1:56:12,  1.49s/it]
 81%|████████  | 19757/24420 [8:14:47<1:56:01,  1.49s/it]
 81%|████████  | 19758/24420 [8:14:49<1:55:58,  1.49s/it]
 81%|████████  | 19759/24420 [8:14:50<1:56:12,  1.50s/it]
 81%|████████  | 19760/24420 [8:14:52<1:56:13,  1.50s/it]
                                                         
{'loss': 0.0828, 'learning_rate': 5e-05, 'epoch': 4.05}

 81%|████████  | 19760/24420 [8:14:52<1:56:13,  1.50s/it]
 81%|████████  | 19761/24420 [8:14:53<1:56:10,  1.50s/it]
 81%|████████  | 19762/24420 [8:14:55<1:56:03,  1.49s/it]
 81%|████████  | 19763/24420 [8:14:56<1:56:15,  1.50s/it]
 81%|████████  | 19764/24420 [8:14:58<1:55:49,  1.49s/it]
 81%|████████  | 19765/24420 [8:14:59<1:55:44,  1.49s/it]
                                                         
{'loss': 0.0737, 'learning_rate': 5e-05, 'epoch': 4.05}

 81%|████████  | 19765/24420 [8:14:59<1:55:44,  1.49s/it]
 81%|████████  | 19766/24420 [8:15:00<1:55:48,  1.49s/it]
 81%|████████  | 19767/24420 [8:15:02<1:55:53,  1.49s/it]
 81%|████████  | 19768/24420 [8:15:03<1:55:48,  1.49s/it]
 81%|████████  | 19769/24420 [8:15:05<1:55:41,  1.49s/it]
 81%|████████  | 19770/24420 [8:15:06<1:55:44,  1.49s/it]
                                                         
{'loss': 0.066, 'learning_rate': 5e-05, 'epoch': 4.05}

 81%|████████  | 19770/24420 [8:15:06<1:55:44,  1.49s/it]
 81%|████████  | 19771/24420 [8:15:08<1:55:45,  1.49s/it]
 81%|████████  | 19772/24420 [8:15:09<1:55:31,  1.49s/it]
 81%|████████  | 19773/24420 [8:15:11<1:55:16,  1.49s/it]
 81%|████████  | 19774/24420 [8:15:12<1:55:17,  1.49s/it]
 81%|████████  | 19775/24420 [8:15:14<1:55:41,  1.49s/it]
                                                         
{'loss': 0.0773, 'learning_rate': 5e-05, 'epoch': 4.05}

 81%|████████  | 19775/24420 [8:15:14<1:55:41,  1.49s/it]
 81%|████████  | 19776/24420 [8:15:15<1:55:37,  1.49s/it]
 81%|████████  | 19777/24420 [8:15:17<1:55:46,  1.50s/it]
 81%|████████  | 19778/24420 [8:15:18<1:55:32,  1.49s/it]
 81%|████████  | 19779/24420 [8:15:20<1:55:38,  1.49s/it]
 81%|████████  | 19780/24420 [8:15:21<1:55:28,  1.49s/it]
                                                         
{'loss': 0.0834, 'learning_rate': 5e-05, 'epoch': 4.05}

 81%|████████  | 19780/24420 [8:15:21<1:55:28,  1.49s/it]
 81%|████████  | 19781/24420 [8:15:23<1:55:24,  1.49s/it]
 81%|████████  | 19782/24420 [8:15:24<1:55:08,  1.49s/it]
 81%|████████  | 19783/24420 [8:15:26<1:55:40,  1.50s/it]
 81%|████████  | 19784/24420 [8:15:27<1:55:26,  1.49s/it]
 81%|████████  | 19785/24420 [8:15:29<1:55:26,  1.49s/it]
                                                         
{'loss': 0.0664, 'learning_rate': 5e-05, 'epoch': 4.05}

 81%|████████  | 19785/24420 [8:15:29<1:55:26,  1.49s/it]
 81%|████████  | 19786/24420 [8:15:30<1:55:28,  1.50s/it]
 81%|████████  | 19787/24420 [8:15:32<1:55:21,  1.49s/it]
 81%|████████  | 19788/24420 [8:15:33<1:55:35,  1.50s/it]
 81%|████████  | 19789/24420 [8:15:35<1:55:25,  1.50s/it]
 81%|████████  | 19790/24420 [8:15:36<1:55:17,  1.49s/it]
                                                         
{'loss': 0.0918, 'learning_rate': 5e-05, 'epoch': 4.05}

 81%|████████  | 19790/24420 [8:15:36<1:55:17,  1.49s/it]
 81%|████████  | 19791/24420 [8:15:38<1:55:15,  1.49s/it]
 81%|████████  | 19792/24420 [8:15:39<1:55:18,  1.49s/it]
 81%|████████  | 19793/24420 [8:15:41<1:55:08,  1.49s/it]
 81%|████████  | 19794/24420 [8:15:42<1:55:04,  1.49s/it]
 81%|████████  | 19795/24420 [8:15:44<1:54:46,  1.49s/it]
                                                         
{'loss': 0.0613, 'learning_rate': 5e-05, 'epoch': 4.05}

 81%|████████  | 19795/24420 [8:15:44<1:54:46,  1.49s/it]
 81%|████████  | 19796/24420 [8:15:45<1:54:49,  1.49s/it]
 81%|████████  | 19797/24420 [8:15:47<1:54:52,  1.49s/it]
 81%|████████  | 19798/24420 [8:15:48<1:54:48,  1.49s/it]
 81%|████████  | 19799/24420 [8:15:50<1:54:50,  1.49s/it]
 81%|████████  | 19800/24420 [8:15:51<1:55:00,  1.49s/it]
                                                         
{'loss': 0.0652, 'learning_rate': 5e-05, 'epoch': 4.05}

 81%|████████  | 19800/24420 [8:15:51<1:55:00,  1.49s/it]
 81%|████████  | 19801/24420 [8:15:57<3:36:44,  2.82s/it]
 81%|████████  | 19802/24420 [8:15:59<3:05:42,  2.41s/it]
 81%|████████  | 19803/24420 [8:16:00<2:44:32,  2.14s/it]
 81%|████████  | 19804/24420 [8:16:02<2:29:41,  1.95s/it]
 81%|████████  | 19805/24420 [8:16:03<2:19:11,  1.81s/it]
                                                         
{'loss': 0.0656, 'learning_rate': 5e-05, 'epoch': 4.06}

 81%|████████  | 19805/24420 [8:16:03<2:19:11,  1.81s/it]
 81%|████████  | 19806/24420 [8:16:05<2:11:46,  1.71s/it]
 81%|████████  | 19807/24420 [8:16:06<2:06:32,  1.65s/it]
 81%|████████  | 19808/24420 [8:16:08<2:03:14,  1.60s/it]
 81%|████████  | 19809/24420 [8:16:09<2:00:54,  1.57s/it]
 81%|████████  | 19810/24420 [8:16:11<2:01:38,  1.58s/it]
                                                         
{'loss': 0.1072, 'learning_rate': 5e-05, 'epoch': 4.06}

 81%|████████  | 19810/24420 [8:16:11<2:01:38,  1.58s/it]
 81%|████████  | 19811/24420 [8:16:12<1:59:48,  1.56s/it]
 81%|████████  | 19812/24420 [8:16:14<1:58:07,  1.54s/it]
 81%|████████  | 19813/24420 [8:16:15<1:57:30,  1.53s/it]
 81%|████████  | 19814/24420 [8:16:17<1:56:33,  1.52s/it]
 81%|████████  | 19815/24420 [8:16:18<1:56:01,  1.51s/it]
                                                         
{'loss': 0.0848, 'learning_rate': 5e-05, 'epoch': 4.06}

 81%|████████  | 19815/24420 [8:16:18<1:56:01,  1.51s/it]
 81%|████████  | 19816/24420 [8:16:20<1:55:36,  1.51s/it]
 81%|████████  | 19817/24420 [8:16:21<1:55:03,  1.50s/it]
 81%|████████  | 19818/24420 [8:16:23<1:54:54,  1.50s/it]
 81%|████████  | 19819/24420 [8:16:24<1:54:32,  1.49s/it]
 81%|████████  | 19820/24420 [8:16:26<1:54:19,  1.49s/it]
                                                         
{'loss': 0.0976, 'learning_rate': 5e-05, 'epoch': 4.06}

 81%|████████  | 19820/24420 [8:16:26<1:54:19,  1.49s/it]
 81%|████████  | 19821/24420 [8:16:27<1:54:13,  1.49s/it]
 81%|████████  | 19822/24420 [8:16:29<1:54:12,  1.49s/it]
 81%|████████  | 19823/24420 [8:16:30<1:54:10,  1.49s/it]
 81%|████████  | 19824/24420 [8:16:32<1:53:58,  1.49s/it]
 81%|████████  | 19825/24420 [8:16:33<1:53:47,  1.49s/it]
                                                         
{'loss': 0.0568, 'learning_rate': 5e-05, 'epoch': 4.06}

 81%|████████  | 19825/24420 [8:16:33<1:53:47,  1.49s/it]
 81%|████████  | 19826/24420 [8:16:35<1:53:55,  1.49s/it]
 81%|████████  | 19827/24420 [8:16:36<1:53:48,  1.49s/it]
 81%|████████  | 19828/24420 [8:16:38<1:53:36,  1.48s/it]
 81%|████████  | 19829/24420 [8:16:39<1:53:44,  1.49s/it]
 81%|████████  | 19830/24420 [8:16:41<1:53:49,  1.49s/it]
                                                         
{'loss': 0.081, 'learning_rate': 5e-05, 'epoch': 4.06}

 81%|████████  | 19830/24420 [8:16:41<1:53:49,  1.49s/it]
 81%|████████  | 19831/24420 [8:16:42<1:53:50,  1.49s/it]
 81%|████████  | 19832/24420 [8:16:43<1:53:35,  1.49s/it]
 81%|████████  | 19833/24420 [8:16:45<1:53:56,  1.49s/it]
 81%|████████  | 19834/24420 [8:16:46<1:54:00,  1.49s/it]
 81%|████████  | 19835/24420 [8:16:48<1:53:56,  1.49s/it]
                                                         
{'loss': 0.0659, 'learning_rate': 5e-05, 'epoch': 4.06}

 81%|████████  | 19835/24420 [8:16:48<1:53:56,  1.49s/it]
 81%|████████  | 19836/24420 [8:16:49<1:53:55,  1.49s/it]
 81%|████████  | 19837/24420 [8:16:51<1:53:54,  1.49s/it]
 81%|████████  | 19838/24420 [8:16:52<1:53:51,  1.49s/it]
 81%|████████  | 19839/24420 [8:16:54<1:53:43,  1.49s/it]
 81%|████████  | 19840/24420 [8:16:55<1:53:45,  1.49s/it]
                                                         
{'loss': 0.0667, 'learning_rate': 5e-05, 'epoch': 4.06}

 81%|████████  | 19840/24420 [8:16:55<1:53:45,  1.49s/it]
 81%|████████  | 19841/24420 [8:16:57<1:53:43,  1.49s/it]
 81%|████████▏ | 19842/24420 [8:16:58<1:53:41,  1.49s/it]
 81%|████████▏ | 19843/24420 [8:17:00<1:53:35,  1.49s/it]
 81%|████████▏ | 19844/24420 [8:17:01<1:53:31,  1.49s/it]
 81%|████████▏ | 19845/24420 [8:17:03<1:53:39,  1.49s/it]
                                                         
{'loss': 0.0674, 'learning_rate': 5e-05, 'epoch': 4.06}

 81%|████████▏ | 19845/24420 [8:17:03<1:53:39,  1.49s/it]
 81%|████████▏ | 19846/24420 [8:17:04<1:53:46,  1.49s/it]
 81%|████████▏ | 19847/24420 [8:17:06<1:54:02,  1.50s/it]
 81%|████████▏ | 19848/24420 [8:17:07<1:53:58,  1.50s/it]
 81%|████████▏ | 19849/24420 [8:17:09<1:53:37,  1.49s/it]
 81%|████████▏ | 19850/24420 [8:17:10<1:53:41,  1.49s/it]
                                                         
{'loss': 0.081, 'learning_rate': 5e-05, 'epoch': 4.06}

 81%|████████▏ | 19850/24420 [8:17:10<1:53:41,  1.49s/it]
 81%|████████▏ | 19851/24420 [8:17:12<1:53:44,  1.49s/it]
 81%|████████▏ | 19852/24420 [8:17:13<1:53:34,  1.49s/it]
 81%|████████▏ | 19853/24420 [8:17:15<1:53:13,  1.49s/it]
 81%|████████▏ | 19854/24420 [8:17:16<1:53:05,  1.49s/it]
 81%|████████▏ | 19855/24420 [8:17:18<1:52:53,  1.48s/it]
                                                         
{'loss': 0.0681, 'learning_rate': 5e-05, 'epoch': 4.07}

 81%|████████▏ | 19855/24420 [8:17:18<1:52:53,  1.48s/it]
 81%|████████▏ | 19856/24420 [8:17:19<1:53:01,  1.49s/it]
 81%|████████▏ | 19857/24420 [8:17:21<1:52:53,  1.48s/it]
 81%|████████▏ | 19858/24420 [8:17:22<1:52:49,  1.48s/it]
 81%|████████▏ | 19859/24420 [8:17:24<1:52:55,  1.49s/it]
 81%|████████▏ | 19860/24420 [8:17:25<1:53:04,  1.49s/it]
                                                         
{'loss': 0.0747, 'learning_rate': 5e-05, 'epoch': 4.07}

 81%|████████▏ | 19860/24420 [8:17:25<1:53:04,  1.49s/it]
 81%|████████▏ | 19861/24420 [8:17:27<1:53:18,  1.49s/it]
 81%|████████▏ | 19862/24420 [8:17:28<1:53:17,  1.49s/it]
 81%|████████▏ | 19863/24420 [8:17:30<1:53:12,  1.49s/it]
 81%|████████▏ | 19864/24420 [8:17:31<1:53:09,  1.49s/it]
 81%|████████▏ | 19865/24420 [8:17:33<1:53:04,  1.49s/it]
                                                         
{'loss': 0.0847, 'learning_rate': 5e-05, 'epoch': 4.07}

 81%|████████▏ | 19865/24420 [8:17:33<1:53:04,  1.49s/it]
 81%|████████▏ | 19866/24420 [8:17:34<1:53:02,  1.49s/it]
 81%|████████▏ | 19867/24420 [8:17:36<1:53:18,  1.49s/it]
 81%|████████▏ | 19868/24420 [8:17:37<1:53:09,  1.49s/it]
 81%|████████▏ | 19869/24420 [8:17:39<1:53:05,  1.49s/it]
 81%|████████▏ | 19870/24420 [8:17:40<1:52:55,  1.49s/it]
                                                         
{'loss': 0.069, 'learning_rate': 5e-05, 'epoch': 4.07}

 81%|████████▏ | 19870/24420 [8:17:40<1:52:55,  1.49s/it]
 81%|████████▏ | 19871/24420 [8:17:42<1:52:50,  1.49s/it]
 81%|████████▏ | 19872/24420 [8:17:43<1:52:51,  1.49s/it]
 81%|████████▏ | 19873/24420 [8:17:45<1:52:51,  1.49s/it]
 81%|████████▏ | 19874/24420 [8:17:46<1:52:48,  1.49s/it]
 81%|████████▏ | 19875/24420 [8:17:48<1:52:46,  1.49s/it]
                                                         
{'loss': 0.0751, 'learning_rate': 5e-05, 'epoch': 4.07}

 81%|████████▏ | 19875/24420 [8:17:48<1:52:46,  1.49s/it]
 81%|████████▏ | 19876/24420 [8:17:49<1:52:45,  1.49s/it]
 81%|████████▏ | 19877/24420 [8:17:51<1:52:46,  1.49s/it]
 81%|████████▏ | 19878/24420 [8:17:52<1:52:53,  1.49s/it]
 81%|████████▏ | 19879/24420 [8:17:54<1:52:48,  1.49s/it]
 81%|████████▏ | 19880/24420 [8:17:55<1:52:39,  1.49s/it]
                                                         
{'loss': 0.0613, 'learning_rate': 5e-05, 'epoch': 4.07}

 81%|████████▏ | 19880/24420 [8:17:55<1:52:39,  1.49s/it]
 81%|████████▏ | 19881/24420 [8:17:56<1:52:44,  1.49s/it]
 81%|████████▏ | 19882/24420 [8:17:58<1:52:41,  1.49s/it]
 81%|████████▏ | 19883/24420 [8:17:59<1:52:39,  1.49s/it]
 81%|████████▏ | 19884/24420 [8:18:01<1:52:36,  1.49s/it]
 81%|████████▏ | 19885/24420 [8:18:02<1:52:29,  1.49s/it]
                                                         
{'loss': 0.08, 'learning_rate': 5e-05, 'epoch': 4.07}

 81%|████████▏ | 19885/24420 [8:18:02<1:52:29,  1.49s/it]
 81%|████████▏ | 19886/24420 [8:18:04<1:52:25,  1.49s/it]
 81%|████████▏ | 19887/24420 [8:18:05<1:52:18,  1.49s/it]
 81%|████████▏ | 19888/24420 [8:18:07<1:52:25,  1.49s/it]
 81%|████████▏ | 19889/24420 [8:18:08<1:52:24,  1.49s/it]
 81%|████████▏ | 19890/24420 [8:18:10<1:52:19,  1.49s/it]
                                                         
{'loss': 0.0724, 'learning_rate': 5e-05, 'epoch': 4.07}

 81%|████████▏ | 19890/24420 [8:18:10<1:52:19,  1.49s/it]
 81%|████████▏ | 19891/24420 [8:18:11<1:52:19,  1.49s/it]
 81%|████████▏ | 19892/24420 [8:18:13<1:52:26,  1.49s/it]
 81%|████████▏ | 19893/24420 [8:18:14<1:52:44,  1.49s/it]
 81%|████████▏ | 19894/24420 [8:18:16<1:52:53,  1.50s/it]
 81%|████████▏ | 19895/24420 [8:18:17<1:52:58,  1.50s/it]
                                                         
{'loss': 0.0704, 'learning_rate': 5e-05, 'epoch': 4.07}

 81%|████████▏ | 19895/24420 [8:18:17<1:52:58,  1.50s/it]
 81%|████████▏ | 19896/24420 [8:18:19<1:52:39,  1.49s/it]
 81%|████████▏ | 19897/24420 [8:18:20<1:52:27,  1.49s/it]
 81%|████████▏ | 19898/24420 [8:18:22<1:52:35,  1.49s/it]
 81%|████████▏ | 19899/24420 [8:18:23<1:52:21,  1.49s/it]
 81%|████████▏ | 19900/24420 [8:18:25<1:52:03,  1.49s/it]
                                                         
{'loss': 0.0706, 'learning_rate': 5e-05, 'epoch': 4.07}

 81%|████████▏ | 19900/24420 [8:18:25<1:52:03,  1.49s/it]
 81%|████████▏ | 19901/24420 [8:18:26<1:51:56,  1.49s/it]
 81%|████████▏ | 19902/24420 [8:18:28<1:51:49,  1.49s/it]
 82%|████████▏ | 19903/24420 [8:18:29<1:51:46,  1.48s/it]
 82%|████████▏ | 19904/24420 [8:18:31<1:51:51,  1.49s/it]
 82%|████████▏ | 19905/24420 [8:18:32<1:51:55,  1.49s/it]
                                                         
{'loss': 0.0678, 'learning_rate': 5e-05, 'epoch': 4.08}

 82%|████████▏ | 19905/24420 [8:18:32<1:51:55,  1.49s/it]
 82%|████████▏ | 19906/24420 [8:18:34<1:52:02,  1.49s/it]
 82%|████████▏ | 19907/24420 [8:18:35<1:52:04,  1.49s/it]
 82%|████████▏ | 19908/24420 [8:18:37<1:51:55,  1.49s/it]
 82%|████████▏ | 19909/24420 [8:18:38<1:53:00,  1.50s/it]
 82%|████████▏ | 19910/24420 [8:18:40<1:52:46,  1.50s/it]
                                                         
{'loss': 0.0698, 'learning_rate': 5e-05, 'epoch': 4.08}

 82%|████████▏ | 19910/24420 [8:18:40<1:52:46,  1.50s/it]
 82%|████████▏ | 19911/24420 [8:18:41<1:52:34,  1.50s/it]
 82%|████████▏ | 19912/24420 [8:18:43<1:52:07,  1.49s/it]
 82%|████████▏ | 19913/24420 [8:18:44<1:52:00,  1.49s/it]
 82%|████████▏ | 19914/24420 [8:18:46<1:51:52,  1.49s/it]
 82%|████████▏ | 19915/24420 [8:18:47<1:51:45,  1.49s/it]
                                                         
{'loss': 0.0651, 'learning_rate': 5e-05, 'epoch': 4.08}

 82%|████████▏ | 19915/24420 [8:18:47<1:51:45,  1.49s/it]
 82%|████████▏ | 19916/24420 [8:18:49<1:51:50,  1.49s/it]
 82%|████████▏ | 19917/24420 [8:18:50<1:51:44,  1.49s/it]
 82%|████████▏ | 19918/24420 [8:18:52<1:51:52,  1.49s/it]
 82%|████████▏ | 19919/24420 [8:18:53<1:51:48,  1.49s/it]
 82%|████████▏ | 19920/24420 [8:18:55<1:51:45,  1.49s/it]
                                                         
{'loss': 0.0663, 'learning_rate': 5e-05, 'epoch': 4.08}

 82%|████████▏ | 19920/24420 [8:18:55<1:51:45,  1.49s/it]
 82%|████████▏ | 19921/24420 [8:18:56<1:52:07,  1.50s/it]
 82%|████████▏ | 19922/24420 [8:18:58<1:52:24,  1.50s/it]
 82%|████████▏ | 19923/24420 [8:18:59<1:52:07,  1.50s/it]
 82%|████████▏ | 19924/24420 [8:19:01<1:52:09,  1.50s/it]
 82%|████████▏ | 19925/24420 [8:19:02<1:51:46,  1.49s/it]
                                                         
{'loss': 0.089, 'learning_rate': 5e-05, 'epoch': 4.08}

 82%|████████▏ | 19925/24420 [8:19:02<1:51:46,  1.49s/it]
 82%|████████▏ | 19926/24420 [8:19:04<1:51:49,  1.49s/it]
 82%|████████▏ | 19927/24420 [8:19:05<1:51:37,  1.49s/it]
 82%|████████▏ | 19928/24420 [8:19:07<1:51:31,  1.49s/it]
 82%|████████▏ | 19929/24420 [8:19:08<1:51:34,  1.49s/it]
 82%|████████▏ | 19930/24420 [8:19:10<1:51:35,  1.49s/it]
                                                         
{'loss': 0.0743, 'learning_rate': 5e-05, 'epoch': 4.08}

 82%|████████▏ | 19930/24420 [8:19:10<1:51:35,  1.49s/it]
 82%|████████▏ | 19931/24420 [8:19:11<1:51:27,  1.49s/it]
 82%|████████▏ | 19932/24420 [8:19:13<1:51:23,  1.49s/it]
 82%|████████▏ | 19933/24420 [8:19:14<1:51:35,  1.49s/it]
 82%|████████▏ | 19934/24420 [8:19:16<1:51:27,  1.49s/it]
 82%|████████▏ | 19935/24420 [8:19:17<1:51:21,  1.49s/it]
                                                         
{'loss': 0.067, 'learning_rate': 5e-05, 'epoch': 4.08}

 82%|████████▏ | 19935/24420 [8:19:17<1:51:21,  1.49s/it]
 82%|████████▏ | 19936/24420 [8:19:19<1:51:20,  1.49s/it]
 82%|████████▏ | 19937/24420 [8:19:20<1:51:16,  1.49s/it]
 82%|████████▏ | 19938/24420 [8:19:21<1:51:20,  1.49s/it]
 82%|████████▏ | 19939/24420 [8:19:23<1:51:17,  1.49s/it]
 82%|████████▏ | 19940/24420 [8:19:24<1:51:11,  1.49s/it]
                                                         
{'loss': 0.0701, 'learning_rate': 5e-05, 'epoch': 4.08}

 82%|████████▏ | 19940/24420 [8:19:24<1:51:11,  1.49s/it]
 82%|████████▏ | 19941/24420 [8:19:26<1:51:08,  1.49s/it]
 82%|████████▏ | 19942/24420 [8:19:27<1:50:59,  1.49s/it]
 82%|████████▏ | 19943/24420 [8:19:29<1:50:57,  1.49s/it]
 82%|████████▏ | 19944/24420 [8:19:30<1:50:58,  1.49s/it]
 82%|████████▏ | 19945/24420 [8:19:32<1:51:03,  1.49s/it]
                                                         
{'loss': 0.0793, 'learning_rate': 5e-05, 'epoch': 4.08}

 82%|████████▏ | 19945/24420 [8:19:32<1:51:03,  1.49s/it]
 82%|████████▏ | 19946/24420 [8:19:33<1:51:03,  1.49s/it]
 82%|████████▏ | 19947/24420 [8:19:35<1:51:01,  1.49s/it]
 82%|████████▏ | 19948/24420 [8:19:36<1:50:54,  1.49s/it]
 82%|████████▏ | 19949/24420 [8:19:38<1:50:51,  1.49s/it]
 82%|████████▏ | 19950/24420 [8:19:39<1:50:51,  1.49s/it]
                                                         
{'loss': 0.0686, 'learning_rate': 5e-05, 'epoch': 4.08}

 82%|████████▏ | 19950/24420 [8:19:39<1:50:51,  1.49s/it]
 82%|████████▏ | 19951/24420 [8:19:41<1:50:52,  1.49s/it]
 82%|████████▏ | 19952/24420 [8:19:42<1:50:58,  1.49s/it]
 82%|████████▏ | 19953/24420 [8:19:44<1:51:12,  1.49s/it]
 82%|████████▏ | 19954/24420 [8:19:45<1:50:54,  1.49s/it]
 82%|████████▏ | 19955/24420 [8:19:47<1:50:38,  1.49s/it]
                                                         
{'loss': 0.0862, 'learning_rate': 5e-05, 'epoch': 4.09}

 82%|████████▏ | 19955/24420 [8:19:47<1:50:38,  1.49s/it]
 82%|████████▏ | 19956/24420 [8:19:48<1:50:43,  1.49s/it]
 82%|████████▏ | 19957/24420 [8:19:50<1:50:43,  1.49s/it]
 82%|████████▏ | 19958/24420 [8:19:51<1:50:47,  1.49s/it]
 82%|████████▏ | 19959/24420 [8:19:53<1:50:44,  1.49s/it]
 82%|████████▏ | 19960/24420 [8:19:54<1:50:48,  1.49s/it]
                                                         
{'loss': 0.0693, 'learning_rate': 5e-05, 'epoch': 4.09}

 82%|████████▏ | 19960/24420 [8:19:54<1:50:48,  1.49s/it]
 82%|████████▏ | 19961/24420 [8:19:56<1:50:42,  1.49s/it]
 82%|████████▏ | 19962/24420 [8:19:57<1:50:42,  1.49s/it]
 82%|████████▏ | 19963/24420 [8:19:59<1:50:38,  1.49s/it]
 82%|████████▏ | 19964/24420 [8:20:00<1:50:48,  1.49s/it]
 82%|████████▏ | 19965/24420 [8:20:02<1:50:44,  1.49s/it]
                                                         
{'loss': 0.0708, 'learning_rate': 5e-05, 'epoch': 4.09}

 82%|████████▏ | 19965/24420 [8:20:02<1:50:44,  1.49s/it]
 82%|████████▏ | 19966/24420 [8:20:03<1:51:02,  1.50s/it]
 82%|████████▏ | 19967/24420 [8:20:05<1:50:47,  1.49s/it]
 82%|████████▏ | 19968/24420 [8:20:06<1:50:35,  1.49s/it]
 82%|████████▏ | 19969/24420 [8:20:08<1:50:38,  1.49s/it]
 82%|████████▏ | 19970/24420 [8:20:09<1:50:30,  1.49s/it]
                                                         
{'loss': 0.0678, 'learning_rate': 5e-05, 'epoch': 4.09}

 82%|████████▏ | 19970/24420 [8:20:09<1:50:30,  1.49s/it]
 82%|████████▏ | 19971/24420 [8:20:11<1:50:33,  1.49s/it]
 82%|████████▏ | 19972/24420 [8:20:12<1:50:34,  1.49s/it]
 82%|████████▏ | 19973/24420 [8:20:14<1:50:28,  1.49s/it]
 82%|████████▏ | 19974/24420 [8:20:15<1:50:31,  1.49s/it]
 82%|████████▏ | 19975/24420 [8:20:17<1:50:28,  1.49s/it]
                                                         
{'loss': 0.0691, 'learning_rate': 5e-05, 'epoch': 4.09}

 82%|████████▏ | 19975/24420 [8:20:17<1:50:28,  1.49s/it]
 82%|████████▏ | 19976/24420 [8:20:18<1:50:33,  1.49s/it]
 82%|████████▏ | 19977/24420 [8:20:20<1:50:37,  1.49s/it]
 82%|████████▏ | 19978/24420 [8:20:21<1:50:25,  1.49s/it]
 82%|████████▏ | 19979/24420 [8:20:23<1:50:20,  1.49s/it]
 82%|████████▏ | 19980/24420 [8:20:24<1:50:09,  1.49s/it]
                                                         
{'loss': 0.055, 'learning_rate': 5e-05, 'epoch': 4.09}

 82%|████████▏ | 19980/24420 [8:20:24<1:50:09,  1.49s/it]
 82%|████████▏ | 19981/24420 [8:20:26<1:50:23,  1.49s/it]
 82%|████████▏ | 19982/24420 [8:20:27<1:50:23,  1.49s/it]
 82%|████████▏ | 19983/24420 [8:20:29<1:57:00,  1.58s/it]
 82%|████████▏ | 19984/24420 [8:20:30<1:54:43,  1.55s/it]
 82%|████████▏ | 19985/24420 [8:20:32<1:53:13,  1.53s/it]
                                                         
{'loss': 0.0711, 'learning_rate': 5e-05, 'epoch': 4.09}

 82%|████████▏ | 19985/24420 [8:20:32<1:53:13,  1.53s/it]
 82%|████████▏ | 19986/24420 [8:20:33<1:52:49,  1.53s/it]
 82%|████████▏ | 19987/24420 [8:20:35<1:52:30,  1.52s/it]
 82%|████████▏ | 19988/24420 [8:20:36<1:52:09,  1.52s/it]
 82%|████████▏ | 19989/24420 [8:20:38<1:51:42,  1.51s/it]
 82%|████████▏ | 19990/24420 [8:20:39<1:51:12,  1.51s/it]
                                                         
{'loss': 0.0788, 'learning_rate': 5e-05, 'epoch': 4.09}

 82%|████████▏ | 19990/24420 [8:20:39<1:51:12,  1.51s/it]
 82%|████████▏ | 19991/24420 [8:20:41<1:50:53,  1.50s/it]
 82%|████████▏ | 19992/24420 [8:20:42<1:50:32,  1.50s/it]
 82%|████████▏ | 19993/24420 [8:20:44<1:50:24,  1.50s/it]
 82%|████████▏ | 19994/24420 [8:20:45<1:50:04,  1.49s/it]
 82%|████████▏ | 19995/24420 [8:20:47<1:49:59,  1.49s/it]
                                                         
{'loss': 0.0701, 'learning_rate': 5e-05, 'epoch': 4.09}

 82%|████████▏ | 19995/24420 [8:20:47<1:49:59,  1.49s/it]
 82%|████████▏ | 19996/24420 [8:20:48<1:50:09,  1.49s/it]
 82%|████████▏ | 19997/24420 [8:20:50<1:49:59,  1.49s/it]
 82%|████████▏ | 19998/24420 [8:20:51<1:50:00,  1.49s/it]
 82%|████████▏ | 19999/24420 [8:20:53<1:49:59,  1.49s/it]
 82%|████████▏ | 20000/24420 [8:20:54<1:50:03,  1.49s/it]
                                                         
{'loss': 0.0719, 'learning_rate': 5e-05, 'epoch': 4.1}

 82%|████████▏ | 20000/24420 [8:20:54<1:50:03,  1.49s/it]
 82%|████████▏ | 20001/24420 [8:20:59<3:01:34,  2.47s/it]
 82%|████████▏ | 20002/24420 [8:21:00<2:39:46,  2.17s/it]
 82%|████████▏ | 20003/24420 [8:21:02<2:24:50,  1.97s/it]
 82%|████████▏ | 20004/24420 [8:21:03<2:14:13,  1.82s/it]
 82%|████████▏ | 20005/24420 [8:21:05<2:06:44,  1.72s/it]
                                                         
{'loss': 0.073, 'learning_rate': 5e-05, 'epoch': 4.1}

 82%|████████▏ | 20005/24420 [8:21:05<2:06:44,  1.72s/it]
 82%|████████▏ | 20006/24420 [8:21:06<2:01:32,  1.65s/it]
 82%|████████▏ | 20007/24420 [8:21:08<1:58:03,  1.61s/it]
 82%|████████▏ | 20008/24420 [8:21:09<1:55:39,  1.57s/it]
 82%|████████▏ | 20009/24420 [8:21:11<1:53:47,  1.55s/it]
 82%|████████▏ | 20010/24420 [8:21:12<1:52:40,  1.53s/it]
                                                         
{'loss': 0.0657, 'learning_rate': 5e-05, 'epoch': 4.1}

 82%|████████▏ | 20010/24420 [8:21:12<1:52:40,  1.53s/it]
 82%|████████▏ | 20011/24420 [8:21:14<1:51:55,  1.52s/it]
 82%|████████▏ | 20012/24420 [8:21:15<1:51:24,  1.52s/it]
 82%|████████▏ | 20013/24420 [8:21:17<1:50:49,  1.51s/it]
 82%|████████▏ | 20014/24420 [8:21:18<1:50:23,  1.50s/it]
 82%|████████▏ | 20015/24420 [8:21:20<1:50:13,  1.50s/it]
                                                         
{'loss': 0.0696, 'learning_rate': 5e-05, 'epoch': 4.1}

 82%|████████▏ | 20015/24420 [8:21:20<1:50:13,  1.50s/it]
 82%|████████▏ | 20016/24420 [8:21:21<1:50:13,  1.50s/it]
 82%|████████▏ | 20017/24420 [8:21:23<1:50:06,  1.50s/it]
 82%|████████▏ | 20018/24420 [8:21:24<1:49:45,  1.50s/it]
 82%|████████▏ | 20019/24420 [8:21:26<1:49:48,  1.50s/it]
 82%|████████▏ | 20020/24420 [8:21:27<1:49:29,  1.49s/it]
                                                         
{'loss': 0.0816, 'learning_rate': 5e-05, 'epoch': 4.1}

 82%|████████▏ | 20020/24420 [8:21:27<1:49:29,  1.49s/it]
 82%|████████▏ | 20021/24420 [8:21:29<1:49:37,  1.50s/it]
 82%|████████▏ | 20022/24420 [8:21:30<1:49:31,  1.49s/it]
 82%|████████▏ | 20023/24420 [8:21:32<1:49:15,  1.49s/it]
 82%|████████▏ | 20024/24420 [8:21:33<1:49:10,  1.49s/it]
 82%|████████▏ | 20025/24420 [8:21:35<1:49:17,  1.49s/it]
                                                         
{'loss': 0.0782, 'learning_rate': 5e-05, 'epoch': 4.1}

 82%|████████▏ | 20025/24420 [8:21:35<1:49:17,  1.49s/it]
 82%|████████▏ | 20026/24420 [8:21:36<1:49:18,  1.49s/it]
 82%|████████▏ | 20027/24420 [8:21:38<1:49:13,  1.49s/it]
 82%|████████▏ | 20028/24420 [8:21:39<1:49:15,  1.49s/it]
 82%|████████▏ | 20029/24420 [8:21:41<1:49:15,  1.49s/it]
 82%|████████▏ | 20030/24420 [8:21:42<1:49:10,  1.49s/it]
                                                         
{'loss': 0.0766, 'learning_rate': 5e-05, 'epoch': 4.1}

 82%|████████▏ | 20030/24420 [8:21:42<1:49:10,  1.49s/it]
 82%|████████▏ | 20031/24420 [8:21:44<1:49:10,  1.49s/it]
 82%|████████▏ | 20032/24420 [8:21:45<1:49:02,  1.49s/it]
 82%|████████▏ | 20033/24420 [8:21:47<1:49:04,  1.49s/it]
 82%|████████▏ | 20034/24420 [8:21:48<1:48:55,  1.49s/it]
 82%|████████▏ | 20035/24420 [8:21:50<1:48:49,  1.49s/it]
                                                         
{'loss': 0.0755, 'learning_rate': 5e-05, 'epoch': 4.1}

 82%|████████▏ | 20035/24420 [8:21:50<1:48:49,  1.49s/it]
 82%|████████▏ | 20036/24420 [8:21:51<1:48:48,  1.49s/it]
 82%|████████▏ | 20037/24420 [8:21:53<1:48:51,  1.49s/it]
 82%|████████▏ | 20038/24420 [8:21:54<1:48:46,  1.49s/it]
 82%|████████▏ | 20039/24420 [8:21:56<1:48:52,  1.49s/it]
 82%|████████▏ | 20040/24420 [8:21:57<1:48:53,  1.49s/it]
                                                         
{'loss': 0.0782, 'learning_rate': 5e-05, 'epoch': 4.1}

 82%|████████▏ | 20040/24420 [8:21:57<1:48:53,  1.49s/it]
 82%|████████▏ | 20041/24420 [8:21:59<1:48:43,  1.49s/it]
 82%|████████▏ | 20042/24420 [8:22:00<1:48:41,  1.49s/it]
 82%|████████▏ | 20043/24420 [8:22:02<1:48:44,  1.49s/it]
 82%|████████▏ | 20044/24420 [8:22:03<1:48:39,  1.49s/it]
 82%|████████▏ | 20045/24420 [8:22:05<1:48:37,  1.49s/it]
                                                         
{'loss': 0.0646, 'learning_rate': 5e-05, 'epoch': 4.1}

 82%|████████▏ | 20045/24420 [8:22:05<1:48:37,  1.49s/it]
 82%|████████▏ | 20046/24420 [8:22:06<1:48:57,  1.49s/it]
 82%|████████▏ | 20047/24420 [8:22:08<1:48:54,  1.49s/it]
 82%|████████▏ | 20048/24420 [8:22:09<1:48:43,  1.49s/it]
 82%|████████▏ | 20049/24420 [8:22:11<1:48:22,  1.49s/it]
 82%|████████▏ | 20050/24420 [8:22:12<1:48:29,  1.49s/it]
                                                         
{'loss': 0.0616, 'learning_rate': 5e-05, 'epoch': 4.11}

 82%|████████▏ | 20050/24420 [8:22:12<1:48:29,  1.49s/it]
 82%|████████▏ | 20051/24420 [8:22:14<1:48:43,  1.49s/it]
 82%|████████▏ | 20052/24420 [8:22:15<1:48:43,  1.49s/it]
 82%|████████▏ | 20053/24420 [8:22:17<1:48:44,  1.49s/it]
 82%|████████▏ | 20054/24420 [8:22:18<1:48:46,  1.49s/it]
 82%|████████▏ | 20055/24420 [8:22:20<1:48:36,  1.49s/it]
                                                         
{'loss': 0.0747, 'learning_rate': 5e-05, 'epoch': 4.11}

 82%|████████▏ | 20055/24420 [8:22:20<1:48:36,  1.49s/it]
 82%|████████▏ | 20056/24420 [8:22:21<1:48:30,  1.49s/it]
 82%|████████▏ | 20057/24420 [8:22:23<1:48:32,  1.49s/it]
 82%|████████▏ | 20058/24420 [8:22:24<1:48:15,  1.49s/it]
 82%|████████▏ | 20059/24420 [8:22:26<1:48:18,  1.49s/it]
 82%|████████▏ | 20060/24420 [8:22:27<1:48:05,  1.49s/it]
                                                         
{'loss': 0.0778, 'learning_rate': 5e-05, 'epoch': 4.11}

 82%|████████▏ | 20060/24420 [8:22:27<1:48:05,  1.49s/it]
 82%|████████▏ | 20061/24420 [8:22:29<1:48:18,  1.49s/it]
 82%|████████▏ | 20062/24420 [8:22:30<1:48:27,  1.49s/it]
 82%|████████▏ | 20063/24420 [8:22:31<1:48:21,  1.49s/it]
 82%|████████▏ | 20064/24420 [8:22:33<1:48:34,  1.50s/it]
 82%|████████▏ | 20065/24420 [8:22:34<1:48:26,  1.49s/it]
                                                         
{'loss': 0.0736, 'learning_rate': 5e-05, 'epoch': 4.11}

 82%|████████▏ | 20065/24420 [8:22:34<1:48:26,  1.49s/it]
 82%|████████▏ | 20066/24420 [8:22:36<1:48:14,  1.49s/it]
 82%|████████▏ | 20067/24420 [8:22:37<1:48:15,  1.49s/it]
 82%|████████▏ | 20068/24420 [8:22:39<1:48:15,  1.49s/it]
 82%|████████▏ | 20069/24420 [8:22:40<1:48:02,  1.49s/it]
 82%|████████▏ | 20070/24420 [8:22:42<1:47:56,  1.49s/it]
                                                         
{'loss': 0.0552, 'learning_rate': 5e-05, 'epoch': 4.11}

 82%|████████▏ | 20070/24420 [8:22:42<1:47:56,  1.49s/it]
 82%|████████▏ | 20071/24420 [8:22:43<1:47:57,  1.49s/it]
 82%|████████▏ | 20072/24420 [8:22:45<1:47:58,  1.49s/it]
 82%|████████▏ | 20073/24420 [8:22:46<1:48:15,  1.49s/it]
 82%|████████▏ | 20074/24420 [8:22:48<1:48:14,  1.49s/it]
 82%|████████▏ | 20075/24420 [8:22:49<1:48:05,  1.49s/it]
                                                         
{'loss': 0.0809, 'learning_rate': 5e-05, 'epoch': 4.11}

 82%|████████▏ | 20075/24420 [8:22:49<1:48:05,  1.49s/it]
 82%|████████▏ | 20076/24420 [8:22:51<1:48:18,  1.50s/it]
 82%|████████▏ | 20077/24420 [8:22:52<1:48:07,  1.49s/it]
 82%|████████▏ | 20078/24420 [8:22:54<1:48:07,  1.49s/it]
 82%|████████▏ | 20079/24420 [8:22:55<1:47:53,  1.49s/it]
 82%|████████▏ | 20080/24420 [8:22:57<1:47:59,  1.49s/it]
                                                         
{'loss': 0.0641, 'learning_rate': 5e-05, 'epoch': 4.11}

 82%|████████▏ | 20080/24420 [8:22:57<1:47:59,  1.49s/it]
 82%|████████▏ | 20081/24420 [8:22:58<1:47:51,  1.49s/it]
 82%|████████▏ | 20082/24420 [8:23:00<1:47:42,  1.49s/it]
 82%|████████▏ | 20083/24420 [8:23:01<1:47:52,  1.49s/it]
 82%|████████▏ | 20084/24420 [8:23:03<1:48:08,  1.50s/it]
 82%|████████▏ | 20085/24420 [8:23:04<1:48:05,  1.50s/it]
                                                         
{'loss': 0.0775, 'learning_rate': 5e-05, 'epoch': 4.11}

 82%|████████▏ | 20085/24420 [8:23:04<1:48:05,  1.50s/it]
 82%|████████▏ | 20086/24420 [8:23:06<1:47:57,  1.49s/it]
 82%|████████▏ | 20087/24420 [8:23:07<1:48:06,  1.50s/it]
 82%|████████▏ | 20088/24420 [8:23:09<1:48:02,  1.50s/it]
 82%|████████▏ | 20089/24420 [8:23:10<1:48:02,  1.50s/it]
 82%|████████▏ | 20090/24420 [8:23:12<1:47:50,  1.49s/it]
                                                         
{'loss': 0.0741, 'learning_rate': 5e-05, 'epoch': 4.11}

 82%|████████▏ | 20090/24420 [8:23:12<1:47:50,  1.49s/it]
 82%|████████▏ | 20091/24420 [8:23:13<1:47:47,  1.49s/it]
 82%|████████▏ | 20092/24420 [8:23:15<1:47:30,  1.49s/it]
 82%|████████▏ | 20093/24420 [8:23:16<1:47:33,  1.49s/it]
 82%|████████▏ | 20094/24420 [8:23:18<1:47:21,  1.49s/it]
 82%|████████▏ | 20095/24420 [8:23:19<1:47:26,  1.49s/it]
                                                         
{'loss': 0.0798, 'learning_rate': 5e-05, 'epoch': 4.11}

 82%|████████▏ | 20095/24420 [8:23:19<1:47:26,  1.49s/it]
 82%|████████▏ | 20096/24420 [8:23:21<1:47:27,  1.49s/it]
 82%|████████▏ | 20097/24420 [8:23:22<1:47:25,  1.49s/it]
 82%|████████▏ | 20098/24420 [8:23:24<1:47:19,  1.49s/it]
 82%|████████▏ | 20099/24420 [8:23:25<1:47:12,  1.49s/it]
 82%|████████▏ | 20100/24420 [8:23:27<1:47:07,  1.49s/it]
                                                         
{'loss': 0.0766, 'learning_rate': 5e-05, 'epoch': 4.12}

 82%|████████▏ | 20100/24420 [8:23:27<1:47:07,  1.49s/it]
 82%|████████▏ | 20101/24420 [8:23:28<1:47:18,  1.49s/it]
 82%|████████▏ | 20102/24420 [8:23:30<1:47:15,  1.49s/it]
 82%|████████▏ | 20103/24420 [8:23:31<1:47:05,  1.49s/it]
 82%|████████▏ | 20104/24420 [8:23:33<1:47:03,  1.49s/it]
 82%|████████▏ | 20105/24420 [8:23:34<1:47:08,  1.49s/it]
                                                         
{'loss': 0.0697, 'learning_rate': 5e-05, 'epoch': 4.12}

 82%|████████▏ | 20105/24420 [8:23:34<1:47:08,  1.49s/it]
 82%|████████▏ | 20106/24420 [8:23:36<1:47:12,  1.49s/it]
 82%|████████▏ | 20107/24420 [8:23:37<1:47:16,  1.49s/it]
 82%|████████▏ | 20108/24420 [8:23:39<1:47:15,  1.49s/it]
 82%|████████▏ | 20109/24420 [8:23:40<1:46:55,  1.49s/it]
 82%|████████▏ | 20110/24420 [8:23:42<1:47:01,  1.49s/it]
                                                         
{'loss': 0.0781, 'learning_rate': 5e-05, 'epoch': 4.12}

 82%|████████▏ | 20110/24420 [8:23:42<1:47:01,  1.49s/it]
 82%|████████▏ | 20111/24420 [8:23:43<1:47:08,  1.49s/it]
 82%|████████▏ | 20112/24420 [8:23:45<1:47:06,  1.49s/it]
 82%|████████▏ | 20113/24420 [8:23:46<1:47:00,  1.49s/it]
 82%|████████▏ | 20114/24420 [8:23:48<1:47:07,  1.49s/it]
 82%|████████▏ | 20115/24420 [8:23:49<1:47:16,  1.50s/it]
                                                         
{'loss': 0.0781, 'learning_rate': 5e-05, 'epoch': 4.12}

 82%|████████▏ | 20115/24420 [8:23:49<1:47:16,  1.50s/it]
 82%|████████▏ | 20116/24420 [8:23:51<1:47:37,  1.50s/it]
 82%|████████▏ | 20117/24420 [8:23:52<1:47:11,  1.49s/it]
 82%|████████▏ | 20118/24420 [8:23:54<1:47:14,  1.50s/it]
 82%|████████▏ | 20119/24420 [8:23:55<1:47:06,  1.49s/it]
 82%|████████▏ | 20120/24420 [8:23:57<1:47:05,  1.49s/it]
                                                         
{'loss': 0.0821, 'learning_rate': 5e-05, 'epoch': 4.12}

 82%|████████▏ | 20120/24420 [8:23:57<1:47:05,  1.49s/it]
 82%|████████▏ | 20121/24420 [8:23:58<1:46:56,  1.49s/it]
 82%|████████▏ | 20122/24420 [8:24:00<1:46:53,  1.49s/it]
 82%|████████▏ | 20123/24420 [8:24:01<1:46:54,  1.49s/it]
 82%|████████▏ | 20124/24420 [8:24:03<1:46:48,  1.49s/it]
 82%|████████▏ | 20125/24420 [8:24:04<1:46:39,  1.49s/it]
                                                         
{'loss': 0.0638, 'learning_rate': 5e-05, 'epoch': 4.12}

 82%|████████▏ | 20125/24420 [8:24:04<1:46:39,  1.49s/it]
 82%|████████▏ | 20126/24420 [8:24:06<1:46:37,  1.49s/it]
 82%|████████▏ | 20127/24420 [8:24:07<1:46:30,  1.49s/it]
 82%|████████▏ | 20128/24420 [8:24:08<1:46:25,  1.49s/it]
 82%|████████▏ | 20129/24420 [8:24:10<1:46:19,  1.49s/it]
 82%|████████▏ | 20130/24420 [8:24:11<1:46:24,  1.49s/it]
                                                         
{'loss': 0.0812, 'learning_rate': 5e-05, 'epoch': 4.12}

 82%|████████▏ | 20130/24420 [8:24:11<1:46:24,  1.49s/it]
 82%|████████▏ | 20131/24420 [8:24:13<1:46:25,  1.49s/it]
 82%|████████▏ | 20132/24420 [8:24:14<1:46:32,  1.49s/it]
 82%|████████▏ | 20133/24420 [8:24:16<1:46:35,  1.49s/it]
 82%|████████▏ | 20134/24420 [8:24:17<1:46:29,  1.49s/it]
 82%|████████▏ | 20135/24420 [8:24:19<1:46:19,  1.49s/it]
                                                         
{'loss': 0.0812, 'learning_rate': 5e-05, 'epoch': 4.12}

 82%|████████▏ | 20135/24420 [8:24:19<1:46:19,  1.49s/it]
 82%|████████▏ | 20136/24420 [8:24:20<1:46:21,  1.49s/it]
 82%|████████▏ | 20137/24420 [8:24:22<1:46:24,  1.49s/it]
 82%|████████▏ | 20138/24420 [8:24:23<1:46:19,  1.49s/it]
 82%|████████▏ | 20139/24420 [8:24:25<1:46:25,  1.49s/it]
 82%|████████▏ | 20140/24420 [8:24:26<1:46:21,  1.49s/it]
                                                         
{'loss': 0.0774, 'learning_rate': 5e-05, 'epoch': 4.12}

 82%|████████▏ | 20140/24420 [8:24:26<1:46:21,  1.49s/it]
 82%|████████▏ | 20141/24420 [8:24:28<1:46:16,  1.49s/it]
 82%|████████▏ | 20142/24420 [8:24:29<1:46:10,  1.49s/it]
 82%|████████▏ | 20143/24420 [8:24:31<1:45:56,  1.49s/it]
 82%|████████▏ | 20144/24420 [8:24:32<1:45:56,  1.49s/it]
 82%|████████▏ | 20145/24420 [8:24:34<1:45:52,  1.49s/it]
                                                         
{'loss': 0.0708, 'learning_rate': 5e-05, 'epoch': 4.12}

 82%|████████▏ | 20145/24420 [8:24:34<1:45:52,  1.49s/it]
 82%|████████▏ | 20146/24420 [8:24:35<1:46:12,  1.49s/it]
 83%|████████▎ | 20147/24420 [8:24:37<1:46:03,  1.49s/it]
 83%|████████▎ | 20148/24420 [8:24:38<1:46:03,  1.49s/it]
 83%|████████▎ | 20149/24420 [8:24:40<1:45:56,  1.49s/it]
 83%|████████▎ | 20150/24420 [8:24:41<1:45:55,  1.49s/it]
                                                         
{'loss': 0.0733, 'learning_rate': 5e-05, 'epoch': 4.13}

 83%|████████▎ | 20150/24420 [8:24:41<1:45:55,  1.49s/it]
 83%|████████▎ | 20151/24420 [8:24:43<1:45:52,  1.49s/it]
 83%|████████▎ | 20152/24420 [8:24:44<1:45:46,  1.49s/it]
 83%|████████▎ | 20153/24420 [8:24:46<1:45:52,  1.49s/it]
 83%|████████▎ | 20154/24420 [8:24:47<1:45:47,  1.49s/it]
 83%|████████▎ | 20155/24420 [8:24:49<1:45:34,  1.49s/it]
                                                         
{'loss': 0.065, 'learning_rate': 5e-05, 'epoch': 4.13}

 83%|████████▎ | 20155/24420 [8:24:49<1:45:34,  1.49s/it]
 83%|████████▎ | 20156/24420 [8:24:50<1:45:49,  1.49s/it]
 83%|████████▎ | 20157/24420 [8:24:52<1:45:48,  1.49s/it]
 83%|████████▎ | 20158/24420 [8:24:53<1:45:52,  1.49s/it]
 83%|████████▎ | 20159/24420 [8:24:55<1:45:49,  1.49s/it]
 83%|████████▎ | 20160/24420 [8:24:56<1:46:06,  1.49s/it]
                                                         
{'loss': 0.0858, 'learning_rate': 5e-05, 'epoch': 4.13}

 83%|████████▎ | 20160/24420 [8:24:56<1:46:06,  1.49s/it]
 83%|████████▎ | 20161/24420 [8:24:58<1:45:56,  1.49s/it]
 83%|████████▎ | 20162/24420 [8:24:59<1:46:02,  1.49s/it]
 83%|████████▎ | 20163/24420 [8:25:01<1:45:57,  1.49s/it]
 83%|████████▎ | 20164/24420 [8:25:02<1:45:49,  1.49s/it]
 83%|████████▎ | 20165/24420 [8:25:04<1:45:57,  1.49s/it]
                                                         
{'loss': 0.0881, 'learning_rate': 5e-05, 'epoch': 4.13}

 83%|████████▎ | 20165/24420 [8:25:04<1:45:57,  1.49s/it]
 83%|████████▎ | 20166/24420 [8:25:05<1:45:44,  1.49s/it]
 83%|████████▎ | 20167/24420 [8:25:07<1:45:37,  1.49s/it]
 83%|████████▎ | 20168/24420 [8:25:08<1:45:24,  1.49s/it]
 83%|████████▎ | 20169/24420 [8:25:10<1:45:23,  1.49s/it]
 83%|████████▎ | 20170/24420 [8:25:11<1:45:18,  1.49s/it]
                                                         
{'loss': 0.0751, 'learning_rate': 5e-05, 'epoch': 4.13}

 83%|████████▎ | 20170/24420 [8:25:11<1:45:18,  1.49s/it]
 83%|████████▎ | 20171/24420 [8:25:13<1:45:29,  1.49s/it]
 83%|████████▎ | 20172/24420 [8:25:14<1:45:21,  1.49s/it]
 83%|████████▎ | 20173/24420 [8:25:16<1:45:21,  1.49s/it]
 83%|████████▎ | 20174/24420 [8:25:17<1:45:58,  1.50s/it]
 83%|████████▎ | 20175/24420 [8:25:19<1:45:52,  1.50s/it]
                                                         
{'loss': 0.0854, 'learning_rate': 5e-05, 'epoch': 4.13}

 83%|████████▎ | 20175/24420 [8:25:19<1:45:52,  1.50s/it]
 83%|████████▎ | 20176/24420 [8:25:20<1:46:02,  1.50s/it]
 83%|████████▎ | 20177/24420 [8:25:22<1:45:45,  1.50s/it]
 83%|████████▎ | 20178/24420 [8:25:23<1:45:29,  1.49s/it]
 83%|████████▎ | 20179/24420 [8:25:24<1:45:20,  1.49s/it]
 83%|████████▎ | 20180/24420 [8:25:26<1:45:19,  1.49s/it]
                                                         
{'loss': 0.0745, 'learning_rate': 5e-05, 'epoch': 4.13}

 83%|████████▎ | 20180/24420 [8:25:26<1:45:19,  1.49s/it]
 83%|████████▎ | 20181/24420 [8:25:27<1:45:22,  1.49s/it]
 83%|████████▎ | 20182/24420 [8:25:29<1:45:27,  1.49s/it]
 83%|████████▎ | 20183/24420 [8:25:30<1:45:30,  1.49s/it]
 83%|████████▎ | 20184/24420 [8:25:32<1:45:19,  1.49s/it]
 83%|████████▎ | 20185/24420 [8:25:33<1:45:18,  1.49s/it]
                                                         
{'loss': 0.0767, 'learning_rate': 5e-05, 'epoch': 4.13}

 83%|████████▎ | 20185/24420 [8:25:33<1:45:18,  1.49s/it]
 83%|████████▎ | 20186/24420 [8:25:35<1:45:23,  1.49s/it]
 83%|████████▎ | 20187/24420 [8:25:36<1:45:18,  1.49s/it]
 83%|████████▎ | 20188/24420 [8:25:38<1:45:15,  1.49s/it]
 83%|████████▎ | 20189/24420 [8:25:39<1:45:11,  1.49s/it]
 83%|████████▎ | 20190/24420 [8:25:41<1:45:06,  1.49s/it]
                                                         
{'loss': 0.0683, 'learning_rate': 5e-05, 'epoch': 4.13}

 83%|████████▎ | 20190/24420 [8:25:41<1:45:06,  1.49s/it]
 83%|████████▎ | 20191/24420 [8:25:42<1:45:07,  1.49s/it]
 83%|████████▎ | 20192/24420 [8:25:44<1:45:11,  1.49s/it]
 83%|████████▎ | 20193/24420 [8:25:45<1:45:09,  1.49s/it]
 83%|████████▎ | 20194/24420 [8:25:47<1:45:10,  1.49s/it]
 83%|████████▎ | 20195/24420 [8:25:48<1:45:06,  1.49s/it]
                                                         
{'loss': 0.0909, 'learning_rate': 5e-05, 'epoch': 4.13}

 83%|████████▎ | 20195/24420 [8:25:48<1:45:06,  1.49s/it]
 83%|████████▎ | 20196/24420 [8:25:50<1:45:06,  1.49s/it]
 83%|████████▎ | 20197/24420 [8:25:51<1:44:55,  1.49s/it]
 83%|████████▎ | 20198/24420 [8:25:53<1:44:54,  1.49s/it]
 83%|████████▎ | 20199/24420 [8:25:54<1:44:46,  1.49s/it]
 83%|████████▎ | 20200/24420 [8:25:56<1:44:49,  1.49s/it]
                                                         
{'loss': 0.0707, 'learning_rate': 5e-05, 'epoch': 4.14}

 83%|████████▎ | 20200/24420 [8:25:56<1:44:49,  1.49s/it]
 83%|████████▎ | 20201/24420 [8:26:01<2:52:34,  2.45s/it]
 83%|████████▎ | 20202/24420 [8:26:02<2:32:09,  2.16s/it]
 83%|████████▎ | 20203/24420 [8:26:03<2:17:53,  1.96s/it]
 83%|████████▎ | 20204/24420 [8:26:05<2:07:49,  1.82s/it]
 83%|████████▎ | 20205/24420 [8:26:06<2:00:53,  1.72s/it]
                                                         
{'loss': 0.0789, 'learning_rate': 5e-05, 'epoch': 4.14}

 83%|████████▎ | 20205/24420 [8:26:06<2:00:53,  1.72s/it]
 83%|████████▎ | 20206/24420 [8:26:08<1:55:57,  1.65s/it]
 83%|████████▎ | 20207/24420 [8:26:09<1:52:33,  1.60s/it]
 83%|████████▎ | 20208/24420 [8:26:11<1:50:03,  1.57s/it]
 83%|████████▎ | 20209/24420 [8:26:12<1:48:17,  1.54s/it]
 83%|████████▎ | 20210/24420 [8:26:14<1:47:33,  1.53s/it]
                                                         
{'loss': 0.0828, 'learning_rate': 5e-05, 'epoch': 4.14}

 83%|████████▎ | 20210/24420 [8:26:14<1:47:33,  1.53s/it]
 83%|████████▎ | 20211/24420 [8:26:15<1:46:34,  1.52s/it]
 83%|████████▎ | 20212/24420 [8:26:17<1:45:48,  1.51s/it]
 83%|████████▎ | 20213/24420 [8:26:18<1:45:14,  1.50s/it]
 83%|████████▎ | 20214/24420 [8:26:20<1:44:44,  1.49s/it]
 83%|████████▎ | 20215/24420 [8:26:21<1:44:42,  1.49s/it]
                                                         
{'loss': 0.0638, 'learning_rate': 5e-05, 'epoch': 4.14}

 83%|████████▎ | 20215/24420 [8:26:21<1:44:42,  1.49s/it]
 83%|████████▎ | 20216/24420 [8:26:23<1:44:41,  1.49s/it]
 83%|████████▎ | 20217/24420 [8:26:24<1:44:26,  1.49s/it]
 83%|████████▎ | 20218/24420 [8:26:26<1:44:19,  1.49s/it]
 83%|████████▎ | 20219/24420 [8:26:27<1:44:12,  1.49s/it]
 83%|████████▎ | 20220/24420 [8:26:29<1:44:22,  1.49s/it]
                                                         
{'loss': 0.0712, 'learning_rate': 5e-05, 'epoch': 4.14}

 83%|████████▎ | 20220/24420 [8:26:29<1:44:22,  1.49s/it]
 83%|████████▎ | 20221/24420 [8:26:30<1:44:18,  1.49s/it]
 83%|████████▎ | 20222/24420 [8:26:32<1:44:16,  1.49s/it]
 83%|████████▎ | 20223/24420 [8:26:33<1:44:12,  1.49s/it]
 83%|████████▎ | 20224/24420 [8:26:35<1:43:58,  1.49s/it]
 83%|████████▎ | 20225/24420 [8:26:36<1:44:06,  1.49s/it]
                                                         
{'loss': 0.0668, 'learning_rate': 5e-05, 'epoch': 4.14}

 83%|████████▎ | 20225/24420 [8:26:36<1:44:06,  1.49s/it]
 83%|████████▎ | 20226/24420 [8:26:38<1:44:05,  1.49s/it]
 83%|████████▎ | 20227/24420 [8:26:39<1:44:06,  1.49s/it]
 83%|████████▎ | 20228/24420 [8:26:41<1:44:02,  1.49s/it]
 83%|████████▎ | 20229/24420 [8:26:42<1:44:13,  1.49s/it]
 83%|████████▎ | 20230/24420 [8:26:44<1:44:08,  1.49s/it]
                                                         
{'loss': 0.085, 'learning_rate': 5e-05, 'epoch': 4.14}

 83%|████████▎ | 20230/24420 [8:26:44<1:44:08,  1.49s/it]
 83%|████████▎ | 20231/24420 [8:26:45<1:44:19,  1.49s/it]
 83%|████████▎ | 20232/24420 [8:26:47<1:44:14,  1.49s/it]
 83%|████████▎ | 20233/24420 [8:26:48<1:44:10,  1.49s/it]
 83%|████████▎ | 20234/24420 [8:26:50<1:44:01,  1.49s/it]
 83%|████████▎ | 20235/24420 [8:26:51<1:43:52,  1.49s/it]
                                                         
{'loss': 0.0683, 'learning_rate': 5e-05, 'epoch': 4.14}

 83%|████████▎ | 20235/24420 [8:26:51<1:43:52,  1.49s/it]
 83%|████████▎ | 20236/24420 [8:26:53<1:43:40,  1.49s/it]
 83%|████████▎ | 20237/24420 [8:26:54<1:43:46,  1.49s/it]
 83%|████████▎ | 20238/24420 [8:26:56<1:43:47,  1.49s/it]
 83%|████████▎ | 20239/24420 [8:26:57<1:43:46,  1.49s/it]
 83%|████████▎ | 20240/24420 [8:26:59<1:43:44,  1.49s/it]
                                                         
{'loss': 0.0713, 'learning_rate': 5e-05, 'epoch': 4.14}

 83%|████████▎ | 20240/24420 [8:26:59<1:43:44,  1.49s/it]
 83%|████████▎ | 20241/24420 [8:27:00<1:43:42,  1.49s/it]
 83%|████████▎ | 20242/24420 [8:27:02<1:43:36,  1.49s/it]
 83%|████████▎ | 20243/24420 [8:27:03<1:43:33,  1.49s/it]
 83%|████████▎ | 20244/24420 [8:27:05<1:43:27,  1.49s/it]
 83%|████████▎ | 20245/24420 [8:27:06<1:43:20,  1.49s/it]
                                                         
{'loss': 0.0696, 'learning_rate': 5e-05, 'epoch': 4.15}

 83%|████████▎ | 20245/24420 [8:27:06<1:43:20,  1.49s/it]
 83%|████████▎ | 20246/24420 [8:27:08<1:43:28,  1.49s/it]
 83%|████████▎ | 20247/24420 [8:27:09<1:43:21,  1.49s/it]
 83%|████████▎ | 20248/24420 [8:27:10<1:43:29,  1.49s/it]
 83%|████████▎ | 20249/24420 [8:27:12<1:43:50,  1.49s/it]
 83%|████████▎ | 20250/24420 [8:27:13<1:43:37,  1.49s/it]
                                                         
{'loss': 0.0752, 'learning_rate': 5e-05, 'epoch': 4.15}

 83%|████████▎ | 20250/24420 [8:27:13<1:43:37,  1.49s/it]
 83%|████████▎ | 20251/24420 [8:27:15<1:43:40,  1.49s/it]
 83%|████████▎ | 20252/24420 [8:27:16<1:43:38,  1.49s/it]
 83%|████████▎ | 20253/24420 [8:27:18<1:43:35,  1.49s/it]
 83%|████████▎ | 20254/24420 [8:27:19<1:43:25,  1.49s/it]
 83%|████████▎ | 20255/24420 [8:27:21<1:43:14,  1.49s/it]
                                                         
{'loss': 0.072, 'learning_rate': 5e-05, 'epoch': 4.15}

 83%|████████▎ | 20255/24420 [8:27:21<1:43:14,  1.49s/it]
 83%|████████▎ | 20256/24420 [8:27:22<1:43:12,  1.49s/it]
 83%|████████▎ | 20257/24420 [8:27:24<1:43:22,  1.49s/it]
 83%|████████▎ | 20258/24420 [8:27:25<1:43:17,  1.49s/it]
 83%|████████▎ | 20259/24420 [8:27:27<1:43:33,  1.49s/it]
 83%|████████▎ | 20260/24420 [8:27:28<1:43:28,  1.49s/it]
                                                         
{'loss': 0.0771, 'learning_rate': 5e-05, 'epoch': 4.15}

 83%|████████▎ | 20260/24420 [8:27:28<1:43:28,  1.49s/it]
 83%|████████▎ | 20261/24420 [8:27:30<1:43:23,  1.49s/it]
 83%|████████▎ | 20262/24420 [8:27:31<1:43:30,  1.49s/it]
 83%|████████▎ | 20263/24420 [8:27:33<1:43:45,  1.50s/it]
 83%|████████▎ | 20264/24420 [8:27:34<1:43:33,  1.50s/it]
 83%|████████▎ | 20265/24420 [8:27:36<1:43:19,  1.49s/it]
                                                         
{'loss': 0.0571, 'learning_rate': 5e-05, 'epoch': 4.15}

 83%|████████▎ | 20265/24420 [8:27:36<1:43:19,  1.49s/it]
 83%|████████▎ | 20266/24420 [8:27:37<1:43:25,  1.49s/it]
 83%|████████▎ | 20267/24420 [8:27:39<1:43:17,  1.49s/it]
 83%|████████▎ | 20268/24420 [8:27:40<1:43:13,  1.49s/it]
 83%|████████▎ | 20269/24420 [8:27:42<1:43:04,  1.49s/it]
 83%|████████▎ | 20270/24420 [8:27:43<1:43:06,  1.49s/it]
                                                         
{'loss': 0.0721, 'learning_rate': 5e-05, 'epoch': 4.15}

 83%|████████▎ | 20270/24420 [8:27:43<1:43:06,  1.49s/it]
 83%|████████▎ | 20271/24420 [8:27:45<1:43:03,  1.49s/it]
 83%|████████▎ | 20272/24420 [8:27:46<1:43:00,  1.49s/it]
 83%|████████▎ | 20273/24420 [8:27:48<1:42:50,  1.49s/it]
 83%|████████▎ | 20274/24420 [8:27:49<1:42:58,  1.49s/it]
 83%|████████▎ | 20275/24420 [8:27:51<1:42:57,  1.49s/it]
                                                         
{'loss': 0.0828, 'learning_rate': 5e-05, 'epoch': 4.15}

 83%|████████▎ | 20275/24420 [8:27:51<1:42:57,  1.49s/it]
 83%|████████▎ | 20276/24420 [8:27:52<1:43:00,  1.49s/it]
 83%|████████▎ | 20277/24420 [8:27:54<1:42:58,  1.49s/it]
 83%|████████▎ | 20278/24420 [8:27:55<1:43:04,  1.49s/it]
 83%|████████▎ | 20279/24420 [8:27:57<1:43:05,  1.49s/it]
 83%|████████▎ | 20280/24420 [8:27:58<1:42:56,  1.49s/it]
                                                         
{'loss': 0.0619, 'learning_rate': 5e-05, 'epoch': 4.15}

 83%|████████▎ | 20280/24420 [8:27:58<1:42:56,  1.49s/it]
 83%|████████▎ | 20281/24420 [8:28:00<1:42:52,  1.49s/it]
 83%|████████▎ | 20282/24420 [8:28:01<1:42:43,  1.49s/it]
 83%|████████▎ | 20283/24420 [8:28:03<1:43:15,  1.50s/it]
 83%|████████▎ | 20284/24420 [8:28:04<1:43:05,  1.50s/it]
 83%|████████▎ | 20285/24420 [8:28:06<1:43:04,  1.50s/it]
                                                         
{'loss': 0.0684, 'learning_rate': 5e-05, 'epoch': 4.15}

 83%|████████▎ | 20285/24420 [8:28:06<1:43:04,  1.50s/it]
 83%|████████▎ | 20286/24420 [8:28:07<1:43:07,  1.50s/it]
 83%|████████▎ | 20287/24420 [8:28:09<1:43:15,  1.50s/it]
 83%|████████▎ | 20288/24420 [8:28:10<1:42:55,  1.49s/it]
 83%|████████▎ | 20289/24420 [8:28:12<1:42:42,  1.49s/it]
 83%|████████▎ | 20290/24420 [8:28:13<1:42:40,  1.49s/it]
                                                         
{'loss': 0.07, 'learning_rate': 5e-05, 'epoch': 4.15}

 83%|████████▎ | 20290/24420 [8:28:13<1:42:40,  1.49s/it]
 83%|████████▎ | 20291/24420 [8:28:15<1:42:40,  1.49s/it]
 83%|████████▎ | 20292/24420 [8:28:16<1:42:33,  1.49s/it]
 83%|████████▎ | 20293/24420 [8:28:18<1:42:27,  1.49s/it]
 83%|████████▎ | 20294/24420 [8:28:19<1:42:32,  1.49s/it]
 83%|████████▎ | 20295/24420 [8:28:21<1:42:34,  1.49s/it]
                                                         
{'loss': 0.0869, 'learning_rate': 5e-05, 'epoch': 4.16}

 83%|████████▎ | 20295/24420 [8:28:21<1:42:34,  1.49s/it]
 83%|████████▎ | 20296/24420 [8:28:22<1:42:36,  1.49s/it]
 83%|████████▎ | 20297/24420 [8:28:24<1:42:32,  1.49s/it]
 83%|████████▎ | 20298/24420 [8:28:25<1:42:25,  1.49s/it]
 83%|████████▎ | 20299/24420 [8:28:27<1:42:44,  1.50s/it]
 83%|████████▎ | 20300/24420 [8:28:28<1:42:35,  1.49s/it]
                                                         
{'loss': 0.0664, 'learning_rate': 5e-05, 'epoch': 4.16}

 83%|████████▎ | 20300/24420 [8:28:28<1:42:35,  1.49s/it]
 83%|████████▎ | 20301/24420 [8:28:30<1:42:29,  1.49s/it]
 83%|████████▎ | 20302/24420 [8:28:31<1:42:28,  1.49s/it]
 83%|████████▎ | 20303/24420 [8:28:33<1:42:17,  1.49s/it]
 83%|████████▎ | 20304/24420 [8:28:34<1:42:22,  1.49s/it]
 83%|████████▎ | 20305/24420 [8:28:36<1:42:04,  1.49s/it]
                                                         
{'loss': 0.0769, 'learning_rate': 5e-05, 'epoch': 4.16}

 83%|████████▎ | 20305/24420 [8:28:36<1:42:04,  1.49s/it]
 83%|████████▎ | 20306/24420 [8:28:37<1:42:03,  1.49s/it]
 83%|████████▎ | 20307/24420 [8:28:39<1:42:07,  1.49s/it]
 83%|████████▎ | 20308/24420 [8:28:40<1:42:20,  1.49s/it]
 83%|████████▎ | 20309/24420 [8:28:42<1:42:09,  1.49s/it]
 83%|████████▎ | 20310/24420 [8:28:43<1:42:15,  1.49s/it]
                                                         
{'loss': 0.0769, 'learning_rate': 5e-05, 'epoch': 4.16}

 83%|████████▎ | 20310/24420 [8:28:43<1:42:15,  1.49s/it]
 83%|████████▎ | 20311/24420 [8:28:45<1:42:25,  1.50s/it]
 83%|████████▎ | 20312/24420 [8:28:46<1:42:11,  1.49s/it]
 83%|████████▎ | 20313/24420 [8:28:47<1:42:11,  1.49s/it]
 83%|████████▎ | 20314/24420 [8:28:49<1:42:01,  1.49s/it]
 83%|████████▎ | 20315/24420 [8:28:50<1:42:14,  1.49s/it]
                                                         
{'loss': 0.0711, 'learning_rate': 5e-05, 'epoch': 4.16}

 83%|████████▎ | 20315/24420 [8:28:50<1:42:14,  1.49s/it]
 83%|████████▎ | 20316/24420 [8:28:52<1:42:15,  1.50s/it]
 83%|████████▎ | 20317/24420 [8:28:53<1:42:15,  1.50s/it]
 83%|████████▎ | 20318/24420 [8:28:55<1:42:18,  1.50s/it]
 83%|████████▎ | 20319/24420 [8:28:56<1:42:16,  1.50s/it]
 83%|████████▎ | 20320/24420 [8:28:58<1:42:08,  1.49s/it]
                                                         
{'loss': 0.0702, 'learning_rate': 5e-05, 'epoch': 4.16}

 83%|████████▎ | 20320/24420 [8:28:58<1:42:08,  1.49s/it]
 83%|████████▎ | 20321/24420 [8:28:59<1:42:02,  1.49s/it]
 83%|████████▎ | 20322/24420 [8:29:01<1:41:51,  1.49s/it]
 83%|████████▎ | 20323/24420 [8:29:02<1:42:27,  1.50s/it]
 83%|████████▎ | 20324/24420 [8:29:04<1:42:18,  1.50s/it]
 83%|████████▎ | 20325/24420 [8:29:05<1:42:21,  1.50s/it]
                                                         
{'loss': 0.0904, 'learning_rate': 5e-05, 'epoch': 4.16}

 83%|████████▎ | 20325/24420 [8:29:05<1:42:21,  1.50s/it]
 83%|████████▎ | 20326/24420 [8:29:07<1:42:36,  1.50s/it]
 83%|████████▎ | 20327/24420 [8:29:08<1:42:27,  1.50s/it]
 83%|████████▎ | 20328/24420 [8:29:10<1:42:18,  1.50s/it]
 83%|████████▎ | 20329/24420 [8:29:11<1:42:07,  1.50s/it]
 83%|████████▎ | 20330/24420 [8:29:13<1:42:05,  1.50s/it]
                                                         
{'loss': 0.0856, 'learning_rate': 5e-05, 'epoch': 4.16}

 83%|████████▎ | 20330/24420 [8:29:13<1:42:05,  1.50s/it]
 83%|████████▎ | 20331/24420 [8:29:14<1:42:07,  1.50s/it]
 83%|████████▎ | 20332/24420 [8:29:16<1:42:02,  1.50s/it]
 83%|████████▎ | 20333/24420 [8:29:17<1:41:51,  1.50s/it]
 83%|████████▎ | 20334/24420 [8:29:19<1:41:48,  1.50s/it]
 83%|████████▎ | 20335/24420 [8:29:20<1:41:50,  1.50s/it]
                                                         
{'loss': 0.0722, 'learning_rate': 5e-05, 'epoch': 4.16}

 83%|████████▎ | 20335/24420 [8:29:20<1:41:50,  1.50s/it]
 83%|████████▎ | 20336/24420 [8:29:22<1:41:49,  1.50s/it]
 83%|████████▎ | 20337/24420 [8:29:23<1:41:40,  1.49s/it]
 83%|████████▎ | 20338/24420 [8:29:25<1:41:34,  1.49s/it]
 83%|████████▎ | 20339/24420 [8:29:26<1:41:24,  1.49s/it]
 83%|████████▎ | 20340/24420 [8:29:28<1:41:30,  1.49s/it]
                                                         
{'loss': 0.0676, 'learning_rate': 5e-05, 'epoch': 4.16}

 83%|████████▎ | 20340/24420 [8:29:28<1:41:30,  1.49s/it]
 83%|████████▎ | 20341/24420 [8:29:29<1:41:45,  1.50s/it]
 83%|████████▎ | 20342/24420 [8:29:31<1:41:40,  1.50s/it]
 83%|████████▎ | 20343/24420 [8:29:32<1:41:50,  1.50s/it]
 83%|████████▎ | 20344/24420 [8:29:34<1:41:55,  1.50s/it]
 83%|████████▎ | 20345/24420 [8:29:35<1:41:46,  1.50s/it]
                                                         
{'loss': 0.0722, 'learning_rate': 5e-05, 'epoch': 4.17}

 83%|████████▎ | 20345/24420 [8:29:35<1:41:46,  1.50s/it]
 83%|████████▎ | 20346/24420 [8:29:37<1:41:33,  1.50s/it]
 83%|████████▎ | 20347/24420 [8:29:38<1:41:43,  1.50s/it]
 83%|████████▎ | 20348/24420 [8:29:40<1:41:38,  1.50s/it]
 83%|████████▎ | 20349/24420 [8:29:41<1:41:29,  1.50s/it]
 83%|████████▎ | 20350/24420 [8:29:43<1:41:27,  1.50s/it]
                                                         
{'loss': 0.078, 'learning_rate': 5e-05, 'epoch': 4.17}

 83%|████████▎ | 20350/24420 [8:29:43<1:41:27,  1.50s/it]
 83%|████████▎ | 20351/24420 [8:29:44<1:41:57,  1.50s/it]
 83%|████████▎ | 20352/24420 [8:29:46<1:41:42,  1.50s/it]
 83%|████████▎ | 20353/24420 [8:29:47<1:41:29,  1.50s/it]
 83%|████████▎ | 20354/24420 [8:29:49<1:41:12,  1.49s/it]
 83%|████████▎ | 20355/24420 [8:29:50<1:40:56,  1.49s/it]
                                                         
{'loss': 0.0604, 'learning_rate': 5e-05, 'epoch': 4.17}

 83%|████████▎ | 20355/24420 [8:29:50<1:40:56,  1.49s/it]
 83%|████████▎ | 20356/24420 [8:29:52<1:41:13,  1.49s/it]
 83%|████████▎ | 20357/24420 [8:29:53<1:41:06,  1.49s/it]
 83%|████████▎ | 20358/24420 [8:29:55<1:41:18,  1.50s/it]
 83%|████████▎ | 20359/24420 [8:29:56<1:41:10,  1.49s/it]
 83%|████████▎ | 20360/24420 [8:29:58<1:41:07,  1.49s/it]
                                                         
{'loss': 0.06, 'learning_rate': 5e-05, 'epoch': 4.17}

 83%|████████▎ | 20360/24420 [8:29:58<1:41:07,  1.49s/it]
 83%|████████▎ | 20361/24420 [8:29:59<1:41:20,  1.50s/it]
 83%|████████▎ | 20362/24420 [8:30:01<1:41:12,  1.50s/it]
 83%|████████▎ | 20363/24420 [8:30:02<1:41:00,  1.49s/it]
 83%|████████▎ | 20364/24420 [8:30:04<1:40:58,  1.49s/it]
 83%|████████▎ | 20365/24420 [8:30:05<1:41:05,  1.50s/it]
                                                         
{'loss': 0.0921, 'learning_rate': 5e-05, 'epoch': 4.17}

 83%|████████▎ | 20365/24420 [8:30:05<1:41:05,  1.50s/it]
 83%|████████▎ | 20366/24420 [8:30:07<1:41:02,  1.50s/it]
 83%|████████▎ | 20367/24420 [8:30:08<1:41:03,  1.50s/it]
 83%|████████▎ | 20368/24420 [8:30:10<1:40:56,  1.49s/it]
 83%|████████▎ | 20369/24420 [8:30:11<1:40:47,  1.49s/it]
 83%|████████▎ | 20370/24420 [8:30:13<1:40:58,  1.50s/it]
                                                         
{'loss': 0.0672, 'learning_rate': 5e-05, 'epoch': 4.17}

 83%|████████▎ | 20370/24420 [8:30:13<1:40:58,  1.50s/it]
 83%|████████▎ | 20371/24420 [8:30:14<1:41:12,  1.50s/it]
 83%|████████▎ | 20372/24420 [8:30:16<1:40:56,  1.50s/it]
 83%|████████▎ | 20373/24420 [8:30:17<1:42:38,  1.52s/it]
 83%|████████▎ | 20374/24420 [8:30:19<1:42:03,  1.51s/it]
 83%|████████▎ | 20375/24420 [8:30:20<1:41:39,  1.51s/it]
                                                         
{'loss': 0.0755, 'learning_rate': 5e-05, 'epoch': 4.17}

 83%|████████▎ | 20375/24420 [8:30:20<1:41:39,  1.51s/it]
 83%|████████▎ | 20376/24420 [8:30:22<1:41:25,  1.50s/it]
 83%|████████▎ | 20377/24420 [8:30:23<1:41:06,  1.50s/it]
 83%|████████▎ | 20378/24420 [8:30:25<1:40:59,  1.50s/it]
 83%|████████▎ | 20379/24420 [8:30:26<1:40:47,  1.50s/it]
 83%|████████▎ | 20380/24420 [8:30:28<1:40:50,  1.50s/it]
                                                         
{'loss': 0.0789, 'learning_rate': 5e-05, 'epoch': 4.17}

 83%|████████▎ | 20380/24420 [8:30:28<1:40:50,  1.50s/it]
 83%|████████▎ | 20381/24420 [8:30:29<1:40:44,  1.50s/it]
 83%|████████▎ | 20382/24420 [8:30:31<1:40:25,  1.49s/it]
 83%|████████▎ | 20383/24420 [8:30:32<1:40:22,  1.49s/it]
 83%|████████▎ | 20384/24420 [8:30:34<1:40:18,  1.49s/it]
 83%|████████▎ | 20385/24420 [8:30:35<1:40:18,  1.49s/it]
                                                         
{'loss': 0.0713, 'learning_rate': 5e-05, 'epoch': 4.17}

 83%|████████▎ | 20385/24420 [8:30:35<1:40:18,  1.49s/it]
 83%|████████▎ | 20386/24420 [8:30:37<1:40:17,  1.49s/it]
 83%|████████▎ | 20387/24420 [8:30:38<1:40:16,  1.49s/it]
 83%|████████▎ | 20388/24420 [8:30:40<1:40:21,  1.49s/it]
 83%|████████▎ | 20389/24420 [8:30:41<1:40:25,  1.49s/it]
 83%|████████▎ | 20390/24420 [8:30:43<1:40:10,  1.49s/it]
                                                         
{'loss': 0.0785, 'learning_rate': 5e-05, 'epoch': 4.17}

 83%|████████▎ | 20390/24420 [8:30:43<1:40:10,  1.49s/it]
 84%|████████▎ | 20391/24420 [8:30:44<1:40:14,  1.49s/it]
 84%|████████▎ | 20392/24420 [8:30:46<1:40:15,  1.49s/it]
 84%|████████▎ | 20393/24420 [8:30:47<1:40:29,  1.50s/it]
 84%|████████▎ | 20394/24420 [8:30:49<1:40:34,  1.50s/it]
 84%|████████▎ | 20395/24420 [8:30:50<1:40:33,  1.50s/it]
                                                         
{'loss': 0.0669, 'learning_rate': 5e-05, 'epoch': 4.18}

 84%|████████▎ | 20395/24420 [8:30:50<1:40:33,  1.50s/it]
 84%|████████▎ | 20396/24420 [8:30:52<1:40:20,  1.50s/it]
 84%|████████▎ | 20397/24420 [8:30:53<1:40:31,  1.50s/it]
 84%|████████▎ | 20398/24420 [8:30:55<1:40:34,  1.50s/it]
 84%|████████▎ | 20399/24420 [8:30:56<1:40:31,  1.50s/it]
 84%|████████▎ | 20400/24420 [8:30:58<1:40:25,  1.50s/it]
                                                         
{'loss': 0.0731, 'learning_rate': 5e-05, 'epoch': 4.18}

 84%|████████▎ | 20400/24420 [8:30:58<1:40:25,  1.50s/it]
 84%|████████▎ | 20401/24420 [8:31:02<2:44:23,  2.45s/it]
 84%|████████▎ | 20402/24420 [8:31:04<2:25:08,  2.17s/it]
 84%|████████▎ | 20403/24420 [8:31:05<2:11:33,  1.96s/it]
 84%|████████▎ | 20404/24420 [8:31:07<2:02:05,  1.82s/it]
 84%|████████▎ | 20405/24420 [8:31:08<1:55:16,  1.72s/it]
                                                         
{'loss': 0.0776, 'learning_rate': 5e-05, 'epoch': 4.18}

 84%|████████▎ | 20405/24420 [8:31:08<1:55:16,  1.72s/it]
 84%|████████▎ | 20406/24420 [8:31:10<1:50:29,  1.65s/it]
 84%|████████▎ | 20407/24420 [8:31:11<1:47:03,  1.60s/it]
 84%|████████▎ | 20408/24420 [8:31:13<1:44:58,  1.57s/it]
 84%|████████▎ | 20409/24420 [8:31:14<1:43:30,  1.55s/it]
 84%|████████▎ | 20410/24420 [8:31:16<1:42:16,  1.53s/it]
                                                         
{'loss': 0.0914, 'learning_rate': 5e-05, 'epoch': 4.18}

 84%|████████▎ | 20410/24420 [8:31:16<1:42:16,  1.53s/it]
 84%|████████▎ | 20411/24420 [8:31:17<1:41:27,  1.52s/it]
 84%|████████▎ | 20412/24420 [8:31:19<1:40:57,  1.51s/it]
 84%|████████▎ | 20413/24420 [8:31:20<1:40:25,  1.50s/it]
 84%|████████▎ | 20414/24420 [8:31:22<1:40:18,  1.50s/it]
 84%|████████▎ | 20415/24420 [8:31:23<1:40:06,  1.50s/it]
                                                         
{'loss': 0.0772, 'learning_rate': 5e-05, 'epoch': 4.18}

 84%|████████▎ | 20415/24420 [8:31:23<1:40:06,  1.50s/it]
 84%|████████▎ | 20416/24420 [8:31:25<1:40:08,  1.50s/it]
 84%|████████▎ | 20417/24420 [8:31:26<1:39:57,  1.50s/it]
 84%|████████▎ | 20418/24420 [8:31:28<1:39:49,  1.50s/it]
 84%|████████▎ | 20419/24420 [8:31:29<1:39:49,  1.50s/it]
 84%|████████▎ | 20420/24420 [8:31:31<1:39:44,  1.50s/it]
                                                         
{'loss': 0.0794, 'learning_rate': 5e-05, 'epoch': 4.18}

 84%|████████▎ | 20420/24420 [8:31:31<1:39:44,  1.50s/it]
 84%|████████▎ | 20421/24420 [8:31:32<1:39:27,  1.49s/it]
 84%|████████▎ | 20422/24420 [8:31:34<1:39:39,  1.50s/it]
 84%|████████▎ | 20423/24420 [8:31:35<1:39:43,  1.50s/it]
 84%|████████▎ | 20424/24420 [8:31:37<1:39:39,  1.50s/it]
 84%|████████▎ | 20425/24420 [8:31:38<1:39:31,  1.49s/it]
                                                         
{'loss': 0.0681, 'learning_rate': 5e-05, 'epoch': 4.18}

 84%|████████▎ | 20425/24420 [8:31:38<1:39:31,  1.49s/it]
 84%|████████▎ | 20426/24420 [8:31:40<1:39:27,  1.49s/it]
 84%|████████▎ | 20427/24420 [8:31:41<1:39:18,  1.49s/it]
 84%|████████▎ | 20428/24420 [8:31:43<1:39:34,  1.50s/it]
 84%|████████▎ | 20429/24420 [8:31:44<1:39:23,  1.49s/it]
 84%|████████▎ | 20430/24420 [8:31:46<1:39:30,  1.50s/it]
                                                         
{'loss': 0.0663, 'learning_rate': 5e-05, 'epoch': 4.18}

 84%|████████▎ | 20430/24420 [8:31:46<1:39:30,  1.50s/it]
 84%|████████▎ | 20431/24420 [8:31:47<1:39:27,  1.50s/it]
 84%|████████▎ | 20432/24420 [8:31:49<1:39:21,  1.49s/it]
 84%|████████▎ | 20433/24420 [8:31:50<1:39:12,  1.49s/it]
 84%|████████▎ | 20434/24420 [8:31:52<1:38:58,  1.49s/it]
 84%|████████▎ | 20435/24420 [8:31:53<1:39:19,  1.50s/it]
                                                         
{'loss': 0.0711, 'learning_rate': 5e-05, 'epoch': 4.18}

 84%|████████▎ | 20435/24420 [8:31:53<1:39:19,  1.50s/it]
 84%|████████▎ | 20436/24420 [8:31:55<1:39:26,  1.50s/it]
 84%|████████▎ | 20437/24420 [8:31:56<1:39:27,  1.50s/it]
 84%|████████▎ | 20438/24420 [8:31:58<1:39:13,  1.50s/it]
 84%|████████▎ | 20439/24420 [8:31:59<1:39:15,  1.50s/it]
 84%|████████▎ | 20440/24420 [8:32:01<1:39:16,  1.50s/it]
                                                         
{'loss': 0.0792, 'learning_rate': 5e-05, 'epoch': 4.19}

 84%|████████▎ | 20440/24420 [8:32:01<1:39:16,  1.50s/it]
 84%|████████▎ | 20441/24420 [8:32:02<1:39:04,  1.49s/it]
 84%|████████▎ | 20442/24420 [8:32:04<1:39:12,  1.50s/it]
 84%|████████▎ | 20443/24420 [8:32:05<1:38:58,  1.49s/it]
 84%|████████▎ | 20444/24420 [8:32:07<1:39:02,  1.49s/it]
 84%|████████▎ | 20445/24420 [8:32:08<1:38:57,  1.49s/it]
                                                         
{'loss': 0.0713, 'learning_rate': 5e-05, 'epoch': 4.19}

 84%|████████▎ | 20445/24420 [8:32:08<1:38:57,  1.49s/it]
 84%|████████▎ | 20446/24420 [8:32:10<1:38:57,  1.49s/it]
 84%|████████▎ | 20447/24420 [8:32:11<1:38:42,  1.49s/it]
 84%|████████▎ | 20448/24420 [8:32:13<1:38:39,  1.49s/it]
 84%|████████▎ | 20449/24420 [8:32:14<1:38:50,  1.49s/it]
 84%|████████▎ | 20450/24420 [8:32:16<1:38:42,  1.49s/it]
                                                         
{'loss': 0.0729, 'learning_rate': 5e-05, 'epoch': 4.19}

 84%|████████▎ | 20450/24420 [8:32:16<1:38:42,  1.49s/it]
 84%|████████▎ | 20451/24420 [8:32:17<1:38:41,  1.49s/it]
 84%|████████▍ | 20452/24420 [8:32:19<1:38:39,  1.49s/it]
 84%|████████▍ | 20453/24420 [8:32:20<1:38:34,  1.49s/it]
 84%|████████▍ | 20454/24420 [8:32:22<1:38:25,  1.49s/it]
 84%|████████▍ | 20455/24420 [8:32:23<1:38:32,  1.49s/it]
                                                         
{'loss': 0.0708, 'learning_rate': 5e-05, 'epoch': 4.19}

 84%|████████▍ | 20455/24420 [8:32:23<1:38:32,  1.49s/it]
 84%|████████▍ | 20456/24420 [8:32:25<1:38:36,  1.49s/it]
 84%|████████▍ | 20457/24420 [8:32:26<1:38:31,  1.49s/it]
 84%|████████▍ | 20458/24420 [8:32:28<1:38:27,  1.49s/it]
 84%|████████▍ | 20459/24420 [8:32:29<1:38:39,  1.49s/it]
 84%|████████▍ | 20460/24420 [8:32:31<1:38:33,  1.49s/it]
                                                         
{'loss': 0.0638, 'learning_rate': 5e-05, 'epoch': 4.19}

 84%|████████▍ | 20460/24420 [8:32:31<1:38:33,  1.49s/it]
 84%|████████▍ | 20461/24420 [8:32:32<1:38:26,  1.49s/it]
 84%|████████▍ | 20462/24420 [8:32:34<1:38:45,  1.50s/it]
 84%|████████▍ | 20463/24420 [8:32:35<1:38:44,  1.50s/it]
 84%|████████▍ | 20464/24420 [8:32:37<1:38:53,  1.50s/it]
 84%|████████▍ | 20465/24420 [8:32:38<1:38:34,  1.50s/it]
                                                         
{'loss': 0.0734, 'learning_rate': 5e-05, 'epoch': 4.19}

 84%|████████▍ | 20465/24420 [8:32:38<1:38:34,  1.50s/it]
 84%|████████▍ | 20466/24420 [8:32:40<1:38:38,  1.50s/it]
 84%|████████▍ | 20467/24420 [8:32:41<1:38:24,  1.49s/it]
 84%|████████▍ | 20468/24420 [8:32:42<1:38:12,  1.49s/it]
 84%|████████▍ | 20469/24420 [8:32:44<1:38:11,  1.49s/it]
 84%|████████▍ | 20470/24420 [8:32:45<1:38:02,  1.49s/it]
                                                         
{'loss': 0.0644, 'learning_rate': 5e-05, 'epoch': 4.19}

 84%|████████▍ | 20470/24420 [8:32:45<1:38:02,  1.49s/it]
 84%|████████▍ | 20471/24420 [8:32:47<1:37:56,  1.49s/it]
 84%|████████▍ | 20472/24420 [8:32:48<1:38:03,  1.49s/it]
 84%|████████▍ | 20473/24420 [8:32:50<1:38:05,  1.49s/it]
 84%|████████▍ | 20474/24420 [8:32:51<1:37:55,  1.49s/it]
 84%|████████▍ | 20475/24420 [8:32:53<1:37:53,  1.49s/it]
                                                         
{'loss': 0.0716, 'learning_rate': 5e-05, 'epoch': 4.19}

 84%|████████▍ | 20475/24420 [8:32:53<1:37:53,  1.49s/it]
 84%|████████▍ | 20476/24420 [8:32:54<1:37:56,  1.49s/it]
 84%|████████▍ | 20477/24420 [8:32:56<1:37:52,  1.49s/it]
 84%|████████▍ | 20478/24420 [8:32:57<1:37:54,  1.49s/it]
 84%|████████▍ | 20479/24420 [8:32:59<1:38:03,  1.49s/it]
 84%|████████▍ | 20480/24420 [8:33:00<1:38:08,  1.49s/it]
                                                         
{'loss': 0.0712, 'learning_rate': 5e-05, 'epoch': 4.19}

 84%|████████▍ | 20480/24420 [8:33:00<1:38:08,  1.49s/it]
 84%|████████▍ | 20481/24420 [8:33:02<1:37:53,  1.49s/it]
 84%|████████▍ | 20482/24420 [8:33:03<1:37:51,  1.49s/it]
 84%|████████▍ | 20483/24420 [8:33:05<1:37:50,  1.49s/it]
 84%|████████▍ | 20484/24420 [8:33:06<1:37:47,  1.49s/it]
 84%|████████▍ | 20485/24420 [8:33:08<1:37:48,  1.49s/it]
                                                         
{'loss': 0.0644, 'learning_rate': 5e-05, 'epoch': 4.19}

 84%|████████▍ | 20485/24420 [8:33:08<1:37:48,  1.49s/it]
 84%|████████▍ | 20486/24420 [8:33:09<1:37:49,  1.49s/it]
 84%|████████▍ | 20487/24420 [8:33:11<1:37:54,  1.49s/it]
 84%|████████▍ | 20488/24420 [8:33:12<1:37:49,  1.49s/it]
 84%|████████▍ | 20489/24420 [8:33:14<1:37:48,  1.49s/it]
 84%|████████▍ | 20490/24420 [8:33:15<1:37:31,  1.49s/it]
                                                         
{'loss': 0.0746, 'learning_rate': 5e-05, 'epoch': 4.2}

 84%|████████▍ | 20490/24420 [8:33:15<1:37:31,  1.49s/it]
 84%|████████▍ | 20491/24420 [8:33:17<1:37:36,  1.49s/it]
 84%|████████▍ | 20492/24420 [8:33:18<1:37:44,  1.49s/it]
 84%|████████▍ | 20493/24420 [8:33:20<1:37:42,  1.49s/it]
 84%|████████▍ | 20494/24420 [8:33:21<1:37:35,  1.49s/it]
 84%|████████▍ | 20495/24420 [8:33:23<1:37:34,  1.49s/it]
                                                         
{'loss': 0.0752, 'learning_rate': 5e-05, 'epoch': 4.2}

 84%|████████▍ | 20495/24420 [8:33:23<1:37:34,  1.49s/it]
 84%|████████▍ | 20496/24420 [8:33:24<1:37:40,  1.49s/it]
 84%|████████▍ | 20497/24420 [8:33:26<1:37:33,  1.49s/it]
 84%|████████▍ | 20498/24420 [8:33:27<1:37:37,  1.49s/it]
 84%|████████▍ | 20499/24420 [8:33:29<1:37:32,  1.49s/it]
 84%|████████▍ | 20500/24420 [8:33:30<1:37:28,  1.49s/it]
                                                         
{'loss': 0.0816, 'learning_rate': 5e-05, 'epoch': 4.2}

 84%|████████▍ | 20500/24420 [8:33:30<1:37:28,  1.49s/it]
 84%|████████▍ | 20501/24420 [8:33:32<1:37:20,  1.49s/it]
 84%|████████▍ | 20502/24420 [8:33:33<1:37:23,  1.49s/it]
 84%|████████▍ | 20503/24420 [8:33:35<1:37:35,  1.49s/it]
 84%|████████▍ | 20504/24420 [8:33:36<1:37:47,  1.50s/it]
 84%|████████▍ | 20505/24420 [8:33:38<1:37:32,  1.49s/it]
                                                         
{'loss': 0.0739, 'learning_rate': 5e-05, 'epoch': 4.2}

 84%|████████▍ | 20505/24420 [8:33:38<1:37:32,  1.49s/it]
 84%|████████▍ | 20506/24420 [8:33:39<1:37:32,  1.50s/it]
 84%|████████▍ | 20507/24420 [8:33:41<1:37:28,  1.49s/it]
 84%|████████▍ | 20508/24420 [8:33:42<1:37:29,  1.50s/it]
 84%|████████▍ | 20509/24420 [8:33:44<1:38:04,  1.50s/it]
 84%|████████▍ | 20510/24420 [8:33:45<1:37:37,  1.50s/it]
                                                         
{'loss': 0.0742, 'learning_rate': 5e-05, 'epoch': 4.2}

 84%|████████▍ | 20510/24420 [8:33:45<1:37:37,  1.50s/it]
 84%|████████▍ | 20511/24420 [8:33:47<1:37:39,  1.50s/it]
 84%|████████▍ | 20512/24420 [8:33:48<1:37:32,  1.50s/it]
 84%|████████▍ | 20513/24420 [8:33:50<1:37:59,  1.50s/it]
 84%|████████▍ | 20514/24420 [8:33:51<1:37:47,  1.50s/it]
 84%|████████▍ | 20515/24420 [8:33:53<1:37:43,  1.50s/it]
                                                         
{'loss': 0.0791, 'learning_rate': 5e-05, 'epoch': 4.2}

 84%|████████▍ | 20515/24420 [8:33:53<1:37:43,  1.50s/it]
 84%|████████▍ | 20516/24420 [8:33:54<1:38:50,  1.52s/it]
 84%|████████▍ | 20517/24420 [8:33:56<1:38:28,  1.51s/it]
 84%|████████▍ | 20518/24420 [8:33:57<1:37:58,  1.51s/it]
 84%|████████▍ | 20519/24420 [8:33:59<1:37:34,  1.50s/it]
 84%|████████▍ | 20520/24420 [8:34:00<1:37:30,  1.50s/it]
                                                         
{'loss': 0.0746, 'learning_rate': 5e-05, 'epoch': 4.2}

 84%|████████▍ | 20520/24420 [8:34:00<1:37:30,  1.50s/it]
 84%|████████▍ | 20521/24420 [8:34:02<1:37:15,  1.50s/it]
 84%|████████▍ | 20522/24420 [8:34:03<1:37:05,  1.49s/it]
 84%|████████▍ | 20523/24420 [8:34:05<1:36:50,  1.49s/it]
 84%|████████▍ | 20524/24420 [8:34:06<1:37:01,  1.49s/it]
 84%|████████▍ | 20525/24420 [8:34:08<1:37:03,  1.50s/it]
                                                         
{'loss': 0.0581, 'learning_rate': 5e-05, 'epoch': 4.2}

 84%|████████▍ | 20525/24420 [8:34:08<1:37:03,  1.50s/it]
 84%|████████▍ | 20526/24420 [8:34:09<1:37:05,  1.50s/it]
 84%|████████▍ | 20527/24420 [8:34:11<1:36:54,  1.49s/it]
 84%|████████▍ | 20528/24420 [8:34:12<1:36:58,  1.50s/it]
 84%|████████▍ | 20529/24420 [8:34:14<1:36:57,  1.50s/it]
 84%|████████▍ | 20530/24420 [8:34:15<1:36:50,  1.49s/it]
                                                         
{'loss': 0.0742, 'learning_rate': 5e-05, 'epoch': 4.2}

 84%|████████▍ | 20530/24420 [8:34:15<1:36:50,  1.49s/it]
 84%|████████▍ | 20531/24420 [8:34:17<1:36:56,  1.50s/it]
 84%|████████▍ | 20532/24420 [8:34:18<1:36:48,  1.49s/it]
 84%|████████▍ | 20533/24420 [8:34:20<1:38:04,  1.51s/it]
 84%|████████▍ | 20534/24420 [8:34:21<1:37:33,  1.51s/it]
 84%|████████▍ | 20535/24420 [8:34:23<1:37:01,  1.50s/it]
                                                         
{'loss': 0.0632, 'learning_rate': 5e-05, 'epoch': 4.2}

 84%|████████▍ | 20535/24420 [8:34:23<1:37:01,  1.50s/it]
 84%|████████▍ | 20536/24420 [8:34:24<1:37:07,  1.50s/it]
 84%|████████▍ | 20537/24420 [8:34:26<1:37:03,  1.50s/it]
 84%|████████▍ | 20538/24420 [8:34:27<1:36:51,  1.50s/it]
 84%|████████▍ | 20539/24420 [8:34:29<1:36:36,  1.49s/it]
 84%|████████▍ | 20540/24420 [8:34:30<1:36:29,  1.49s/it]
                                                         
{'loss': 0.0918, 'learning_rate': 5e-05, 'epoch': 4.21}

 84%|████████▍ | 20540/24420 [8:34:30<1:36:29,  1.49s/it]
 84%|████████▍ | 20541/24420 [8:34:32<1:36:26,  1.49s/it]
 84%|████████▍ | 20542/24420 [8:34:33<1:36:31,  1.49s/it]
 84%|████████▍ | 20543/24420 [8:34:35<1:36:19,  1.49s/it]
 84%|████████▍ | 20544/24420 [8:34:36<1:36:04,  1.49s/it]
 84%|████████▍ | 20545/24420 [8:34:38<1:36:12,  1.49s/it]
                                                         
{'loss': 0.0654, 'learning_rate': 5e-05, 'epoch': 4.21}

 84%|████████▍ | 20545/24420 [8:34:38<1:36:12,  1.49s/it]
 84%|████████▍ | 20546/24420 [8:34:39<1:35:57,  1.49s/it]
 84%|████████▍ | 20547/24420 [8:34:41<1:36:00,  1.49s/it]
 84%|████████▍ | 20548/24420 [8:34:42<1:36:05,  1.49s/it]
 84%|████████▍ | 20549/24420 [8:34:44<1:36:05,  1.49s/it]
 84%|████████▍ | 20550/24420 [8:34:45<1:36:11,  1.49s/it]
                                                         
{'loss': 0.0697, 'learning_rate': 5e-05, 'epoch': 4.21}

 84%|████████▍ | 20550/24420 [8:34:45<1:36:11,  1.49s/it]
 84%|████████▍ | 20551/24420 [8:34:47<1:36:19,  1.49s/it]
 84%|████████▍ | 20552/24420 [8:34:48<1:36:17,  1.49s/it]
 84%|████████▍ | 20553/24420 [8:34:50<1:36:17,  1.49s/it]
 84%|████████▍ | 20554/24420 [8:34:51<1:36:14,  1.49s/it]
 84%|████████▍ | 20555/24420 [8:34:53<1:36:05,  1.49s/it]
                                                         
{'loss': 0.0649, 'learning_rate': 5e-05, 'epoch': 4.21}

 84%|████████▍ | 20555/24420 [8:34:53<1:36:05,  1.49s/it]
 84%|████████▍ | 20556/24420 [8:34:54<1:35:58,  1.49s/it]
 84%|████████▍ | 20557/24420 [8:34:55<1:35:48,  1.49s/it]
 84%|████████▍ | 20558/24420 [8:34:57<1:35:44,  1.49s/it]
 84%|████████▍ | 20559/24420 [8:34:58<1:35:40,  1.49s/it]
 84%|████████▍ | 20560/24420 [8:35:00<1:35:49,  1.49s/it]
                                                         
{'loss': 0.0646, 'learning_rate': 5e-05, 'epoch': 4.21}

 84%|████████▍ | 20560/24420 [8:35:00<1:35:49,  1.49s/it]
 84%|████████▍ | 20561/24420 [8:35:01<1:35:45,  1.49s/it]
 84%|████████▍ | 20562/24420 [8:35:03<1:35:44,  1.49s/it]
 84%|████████▍ | 20563/24420 [8:35:04<1:35:42,  1.49s/it]
 84%|████████▍ | 20564/24420 [8:35:06<1:35:51,  1.49s/it]
 84%|████████▍ | 20565/24420 [8:35:07<1:35:40,  1.49s/it]
                                                         
{'loss': 0.0771, 'learning_rate': 5e-05, 'epoch': 4.21}

 84%|████████▍ | 20565/24420 [8:35:07<1:35:40,  1.49s/it]
 84%|████████▍ | 20566/24420 [8:35:09<1:35:42,  1.49s/it]
 84%|████████▍ | 20567/24420 [8:35:10<1:35:43,  1.49s/it]
 84%|████████▍ | 20568/24420 [8:35:12<1:35:40,  1.49s/it]
 84%|████████▍ | 20569/24420 [8:35:13<1:35:37,  1.49s/it]
 84%|████████▍ | 20570/24420 [8:35:15<1:35:38,  1.49s/it]
                                                         
{'loss': 0.067, 'learning_rate': 5e-05, 'epoch': 4.21}

 84%|████████▍ | 20570/24420 [8:35:15<1:35:38,  1.49s/it]
 84%|████████▍ | 20571/24420 [8:35:16<1:35:37,  1.49s/it]
 84%|████████▍ | 20572/24420 [8:35:18<1:35:48,  1.49s/it]
 84%|████████▍ | 20573/24420 [8:35:19<1:35:56,  1.50s/it]
 84%|████████▍ | 20574/24420 [8:35:21<1:35:52,  1.50s/it]
 84%|████████▍ | 20575/24420 [8:35:22<1:35:44,  1.49s/it]
                                                         
{'loss': 0.0674, 'learning_rate': 5e-05, 'epoch': 4.21}

 84%|████████▍ | 20575/24420 [8:35:22<1:35:44,  1.49s/it]
 84%|████████▍ | 20576/24420 [8:35:24<1:35:35,  1.49s/it]
 84%|████████▍ | 20577/24420 [8:35:25<1:35:34,  1.49s/it]
 84%|████████▍ | 20578/24420 [8:35:27<1:35:27,  1.49s/it]
 84%|████████▍ | 20579/24420 [8:35:28<1:35:35,  1.49s/it]
 84%|████████▍ | 20580/24420 [8:35:30<1:35:23,  1.49s/it]
                                                         
{'loss': 0.0781, 'learning_rate': 5e-05, 'epoch': 4.21}

 84%|████████▍ | 20580/24420 [8:35:30<1:35:23,  1.49s/it]
 84%|████████▍ | 20581/24420 [8:35:31<1:35:19,  1.49s/it]
 84%|████████▍ | 20582/24420 [8:35:33<1:35:19,  1.49s/it]
 84%|████████▍ | 20583/24420 [8:35:34<1:35:20,  1.49s/it]
 84%|████████▍ | 20584/24420 [8:35:36<1:35:12,  1.49s/it]
 84%|████████▍ | 20585/24420 [8:35:37<1:35:20,  1.49s/it]
                                                         
{'loss': 0.0842, 'learning_rate': 5e-05, 'epoch': 4.21}

 84%|████████▍ | 20585/24420 [8:35:37<1:35:20,  1.49s/it]
 84%|████████▍ | 20586/24420 [8:35:39<1:35:30,  1.49s/it]
 84%|████████▍ | 20587/24420 [8:35:40<1:35:23,  1.49s/it]
 84%|████████▍ | 20588/24420 [8:35:42<1:35:23,  1.49s/it]
 84%|████████▍ | 20589/24420 [8:35:43<1:35:16,  1.49s/it]
 84%|████████▍ | 20590/24420 [8:35:45<1:35:13,  1.49s/it]
                                                         
{'loss': 0.0811, 'learning_rate': 5e-05, 'epoch': 4.22}

 84%|████████▍ | 20590/24420 [8:35:45<1:35:13,  1.49s/it]
 84%|████████▍ | 20591/24420 [8:35:46<1:35:09,  1.49s/it]
 84%|████████▍ | 20592/24420 [8:35:48<1:35:07,  1.49s/it]
 84%|████████▍ | 20593/24420 [8:35:49<1:35:03,  1.49s/it]
 84%|████████▍ | 20594/24420 [8:35:51<1:35:01,  1.49s/it]
 84%|████████▍ | 20595/24420 [8:35:52<1:35:01,  1.49s/it]
                                                         
{'loss': 0.0723, 'learning_rate': 5e-05, 'epoch': 4.22}

 84%|████████▍ | 20595/24420 [8:35:52<1:35:01,  1.49s/it]
 84%|████████▍ | 20596/24420 [8:35:54<1:35:11,  1.49s/it]
 84%|████████▍ | 20597/24420 [8:35:55<1:35:09,  1.49s/it]
 84%|████████▍ | 20598/24420 [8:35:57<1:35:14,  1.50s/it]
 84%|████████▍ | 20599/24420 [8:35:58<1:35:06,  1.49s/it]
 84%|████████▍ | 20600/24420 [8:36:00<1:35:00,  1.49s/it]
                                                         
{'loss': 0.0666, 'learning_rate': 5e-05, 'epoch': 4.22}

 84%|████████▍ | 20600/24420 [8:36:00<1:35:00,  1.49s/it]
 84%|████████▍ | 20601/24420 [8:36:04<2:37:05,  2.47s/it]
 84%|████████▍ | 20602/24420 [8:36:06<2:18:51,  2.18s/it]
 84%|████████▍ | 20603/24420 [8:36:07<2:05:33,  1.97s/it]
 84%|████████▍ | 20604/24420 [8:36:09<1:56:09,  1.83s/it]
 84%|████████▍ | 20605/24420 [8:36:10<1:49:52,  1.73s/it]
                                                         
{'loss': 0.0732, 'learning_rate': 5e-05, 'epoch': 4.22}

 84%|████████▍ | 20605/24420 [8:36:10<1:49:52,  1.73s/it]
 84%|████████▍ | 20606/24420 [8:36:12<1:45:27,  1.66s/it]
 84%|████████▍ | 20607/24420 [8:36:13<1:42:11,  1.61s/it]
 84%|████████▍ | 20608/24420 [8:36:15<1:39:52,  1.57s/it]
 84%|████████▍ | 20609/24420 [8:36:16<1:38:24,  1.55s/it]
 84%|████████▍ | 20610/24420 [8:36:18<1:37:16,  1.53s/it]
                                                         
{'loss': 0.061, 'learning_rate': 5e-05, 'epoch': 4.22}

 84%|████████▍ | 20610/24420 [8:36:18<1:37:16,  1.53s/it]
 84%|████████▍ | 20611/24420 [8:36:19<1:36:47,  1.52s/it]
 84%|████████▍ | 20612/24420 [8:36:21<1:36:02,  1.51s/it]
 84%|████████▍ | 20613/24420 [8:36:22<1:35:28,  1.50s/it]
 84%|████████▍ | 20614/24420 [8:36:24<1:34:57,  1.50s/it]
 84%|████████▍ | 20615/24420 [8:36:25<1:34:41,  1.49s/it]
                                                         
{'loss': 0.09, 'learning_rate': 5e-05, 'epoch': 4.22}

 84%|████████▍ | 20615/24420 [8:36:25<1:34:41,  1.49s/it]
 84%|████████▍ | 20616/24420 [8:36:27<1:34:37,  1.49s/it]
 84%|████████▍ | 20617/24420 [8:36:28<1:34:27,  1.49s/it]
 84%|████████▍ | 20618/24420 [8:36:30<1:34:35,  1.49s/it]
 84%|████████▍ | 20619/24420 [8:36:31<1:34:26,  1.49s/it]
 84%|████████▍ | 20620/24420 [8:36:33<1:34:27,  1.49s/it]
                                                         
{'loss': 0.0633, 'learning_rate': 5e-05, 'epoch': 4.22}

 84%|████████▍ | 20620/24420 [8:36:33<1:34:27,  1.49s/it]
 84%|████████▍ | 20621/24420 [8:36:34<1:34:35,  1.49s/it]
 84%|████████▍ | 20622/24420 [8:36:36<1:34:28,  1.49s/it]
 84%|████████▍ | 20623/24420 [8:36:37<1:34:15,  1.49s/it]
 84%|████████▍ | 20624/24420 [8:36:39<1:34:04,  1.49s/it]
 84%|████████▍ | 20625/24420 [8:36:40<1:34:01,  1.49s/it]
                                                         
{'loss': 0.0669, 'learning_rate': 5e-05, 'epoch': 4.22}

 84%|████████▍ | 20625/24420 [8:36:40<1:34:01,  1.49s/it]
 84%|████████▍ | 20626/24420 [8:36:42<1:34:16,  1.49s/it]
 84%|████████▍ | 20627/24420 [8:36:43<1:34:16,  1.49s/it]
 84%|████████▍ | 20628/24420 [8:36:45<1:34:19,  1.49s/it]
 84%|████████▍ | 20629/24420 [8:36:46<1:34:13,  1.49s/it]
 84%|████████▍ | 20630/24420 [8:36:48<1:34:11,  1.49s/it]
                                                         
{'loss': 0.0822, 'learning_rate': 5e-05, 'epoch': 4.22}

 84%|████████▍ | 20630/24420 [8:36:48<1:34:11,  1.49s/it]
 84%|████████▍ | 20631/24420 [8:36:49<1:34:06,  1.49s/it]
 84%|████████▍ | 20632/24420 [8:36:51<1:34:20,  1.49s/it]
 84%|████████▍ | 20633/24420 [8:36:52<1:34:12,  1.49s/it]
 84%|████████▍ | 20634/24420 [8:36:54<1:34:22,  1.50s/it]
 85%|████████▍ | 20635/24420 [8:36:55<1:34:17,  1.49s/it]
                                                         
{'loss': 0.0718, 'learning_rate': 5e-05, 'epoch': 4.23}

 85%|████████▍ | 20635/24420 [8:36:55<1:34:17,  1.49s/it]
 85%|████████▍ | 20636/24420 [8:36:57<1:34:12,  1.49s/it]
 85%|████████▍ | 20637/24420 [8:36:58<1:34:16,  1.50s/it]
 85%|████████▍ | 20638/24420 [8:37:00<1:34:19,  1.50s/it]
 85%|████████▍ | 20639/24420 [8:37:01<1:34:20,  1.50s/it]
 85%|████████▍ | 20640/24420 [8:37:03<1:34:07,  1.49s/it]
                                                         
{'loss': 0.072, 'learning_rate': 5e-05, 'epoch': 4.23}

 85%|████████▍ | 20640/24420 [8:37:03<1:34:07,  1.49s/it]
 85%|████████▍ | 20641/24420 [8:37:04<1:34:08,  1.49s/it]
 85%|████████▍ | 20642/24420 [8:37:06<1:34:01,  1.49s/it]
 85%|████████▍ | 20643/24420 [8:37:07<1:33:53,  1.49s/it]
 85%|████████▍ | 20644/24420 [8:37:09<1:33:43,  1.49s/it]
 85%|████████▍ | 20645/24420 [8:37:10<1:33:45,  1.49s/it]
                                                         
{'loss': 0.0791, 'learning_rate': 5e-05, 'epoch': 4.23}

 85%|████████▍ | 20645/24420 [8:37:10<1:33:45,  1.49s/it]
 85%|████████▍ | 20646/24420 [8:37:12<1:33:35,  1.49s/it]
 85%|████████▍ | 20647/24420 [8:37:13<1:33:40,  1.49s/it]
 85%|████████▍ | 20648/24420 [8:37:14<1:33:50,  1.49s/it]
 85%|████████▍ | 20649/24420 [8:37:16<1:33:55,  1.49s/it]
 85%|████████▍ | 20650/24420 [8:37:17<1:33:37,  1.49s/it]
                                                         
{'loss': 0.0687, 'learning_rate': 5e-05, 'epoch': 4.23}

 85%|████████▍ | 20650/24420 [8:37:17<1:33:37,  1.49s/it]
 85%|████████▍ | 20651/24420 [8:37:19<1:33:53,  1.49s/it]
 85%|████████▍ | 20652/24420 [8:37:20<1:33:40,  1.49s/it]
 85%|████████▍ | 20653/24420 [8:37:22<1:33:42,  1.49s/it]
 85%|████████▍ | 20654/24420 [8:37:23<1:33:43,  1.49s/it]
 85%|████████▍ | 20655/24420 [8:37:25<1:33:39,  1.49s/it]
                                                         
{'loss': 0.0782, 'learning_rate': 5e-05, 'epoch': 4.23}

 85%|████████▍ | 20655/24420 [8:37:25<1:33:39,  1.49s/it]
 85%|████████▍ | 20656/24420 [8:37:26<1:33:36,  1.49s/it]
 85%|████████▍ | 20657/24420 [8:37:28<1:33:46,  1.50s/it]
 85%|████████▍ | 20658/24420 [8:37:29<1:33:46,  1.50s/it]
 85%|████████▍ | 20659/24420 [8:37:31<1:33:39,  1.49s/it]
 85%|████████▍ | 20660/24420 [8:37:32<1:33:35,  1.49s/it]
                                                         
{'loss': 0.0709, 'learning_rate': 5e-05, 'epoch': 4.23}

 85%|████████▍ | 20660/24420 [8:37:32<1:33:35,  1.49s/it]
 85%|████████▍ | 20661/24420 [8:37:34<1:33:37,  1.49s/it]
 85%|████████▍ | 20662/24420 [8:37:35<1:33:41,  1.50s/it]
 85%|████████▍ | 20663/24420 [8:37:37<1:33:44,  1.50s/it]
 85%|████████▍ | 20664/24420 [8:37:38<1:33:26,  1.49s/it]
 85%|████████▍ | 20665/24420 [8:37:40<1:33:23,  1.49s/it]
                                                         
{'loss': 0.0772, 'learning_rate': 5e-05, 'epoch': 4.23}

 85%|████████▍ | 20665/24420 [8:37:40<1:33:23,  1.49s/it]
 85%|████████▍ | 20666/24420 [8:37:41<1:33:25,  1.49s/it]
 85%|████████▍ | 20667/24420 [8:37:43<1:33:21,  1.49s/it]
 85%|████████▍ | 20668/24420 [8:37:44<1:33:12,  1.49s/it]
 85%|████████▍ | 20669/24420 [8:37:46<1:33:03,  1.49s/it]
 85%|████████▍ | 20670/24420 [8:37:47<1:33:02,  1.49s/it]
                                                         
{'loss': 0.069, 'learning_rate': 5e-05, 'epoch': 4.23}

 85%|████████▍ | 20670/24420 [8:37:47<1:33:02,  1.49s/it]
 85%|████████▍ | 20671/24420 [8:37:49<1:33:13,  1.49s/it]
 85%|████████▍ | 20672/24420 [8:37:50<1:33:10,  1.49s/it]
 85%|████████▍ | 20673/24420 [8:37:52<1:33:03,  1.49s/it]
 85%|████████▍ | 20674/24420 [8:37:53<1:33:09,  1.49s/it]
 85%|████████▍ | 20675/24420 [8:37:55<1:33:14,  1.49s/it]
                                                         
{'loss': 0.0793, 'learning_rate': 5e-05, 'epoch': 4.23}

 85%|████████▍ | 20675/24420 [8:37:55<1:33:14,  1.49s/it]
 85%|████████▍ | 20676/24420 [8:37:56<1:33:11,  1.49s/it]
 85%|████████▍ | 20677/24420 [8:37:58<1:33:06,  1.49s/it]
 85%|████████▍ | 20678/24420 [8:37:59<1:33:01,  1.49s/it]
 85%|████████▍ | 20679/24420 [8:38:01<1:33:04,  1.49s/it]
 85%|████████▍ | 20680/24420 [8:38:02<1:32:54,  1.49s/it]
                                                         
{'loss': 0.072, 'learning_rate': 5e-05, 'epoch': 4.23}

 85%|████████▍ | 20680/24420 [8:38:02<1:32:54,  1.49s/it]
 85%|████████▍ | 20681/24420 [8:38:04<1:32:53,  1.49s/it]
 85%|████████▍ | 20682/24420 [8:38:05<1:33:03,  1.49s/it]
 85%|████████▍ | 20683/24420 [8:38:07<1:32:56,  1.49s/it]
 85%|████████▍ | 20684/24420 [8:38:08<1:32:43,  1.49s/it]
 85%|████████▍ | 20685/24420 [8:38:10<1:32:47,  1.49s/it]
                                                         
{'loss': 0.0674, 'learning_rate': 5e-05, 'epoch': 4.24}

 85%|████████▍ | 20685/24420 [8:38:10<1:32:47,  1.49s/it]
 85%|████████▍ | 20686/24420 [8:38:11<1:32:59,  1.49s/it]
 85%|████████▍ | 20687/24420 [8:38:13<1:33:03,  1.50s/it]
 85%|████████▍ | 20688/24420 [8:38:14<1:33:07,  1.50s/it]
 85%|████████▍ | 20689/24420 [8:38:16<1:32:52,  1.49s/it]
 85%|████████▍ | 20690/24420 [8:38:17<1:32:55,  1.49s/it]
                                                         
{'loss': 0.077, 'learning_rate': 5e-05, 'epoch': 4.24}

 85%|████████▍ | 20690/24420 [8:38:17<1:32:55,  1.49s/it]
 85%|████████▍ | 20691/24420 [8:38:19<1:32:48,  1.49s/it]
 85%|████████▍ | 20692/24420 [8:38:20<1:32:45,  1.49s/it]
 85%|████████▍ | 20693/24420 [8:38:22<1:32:42,  1.49s/it]
 85%|████████▍ | 20694/24420 [8:38:23<1:32:46,  1.49s/it]
 85%|████████▍ | 20695/24420 [8:38:25<1:32:28,  1.49s/it]
                                                         
{'loss': 0.071, 'learning_rate': 5e-05, 'epoch': 4.24}

 85%|████████▍ | 20695/24420 [8:38:25<1:32:28,  1.49s/it]
 85%|████████▍ | 20696/24420 [8:38:26<1:32:40,  1.49s/it]
 85%|████████▍ | 20697/24420 [8:38:28<1:32:33,  1.49s/it]
 85%|████████▍ | 20698/24420 [8:38:29<1:32:32,  1.49s/it]
 85%|████████▍ | 20699/24420 [8:38:31<1:32:18,  1.49s/it]
 85%|████████▍ | 20700/24420 [8:38:32<1:32:25,  1.49s/it]
                                                         
{'loss': 0.0839, 'learning_rate': 5e-05, 'epoch': 4.24}

 85%|████████▍ | 20700/24420 [8:38:32<1:32:25,  1.49s/it]
 85%|████████▍ | 20701/24420 [8:38:34<1:32:39,  1.49s/it]
 85%|████████▍ | 20702/24420 [8:38:35<1:32:33,  1.49s/it]
 85%|████████▍ | 20703/24420 [8:38:37<1:32:34,  1.49s/it]
 85%|████████▍ | 20704/24420 [8:38:38<1:32:30,  1.49s/it]
 85%|████████▍ | 20705/24420 [8:38:40<1:32:40,  1.50s/it]
                                                         
{'loss': 0.0676, 'learning_rate': 5e-05, 'epoch': 4.24}

 85%|████████▍ | 20705/24420 [8:38:40<1:32:40,  1.50s/it]
 85%|████████▍ | 20706/24420 [8:38:41<1:32:24,  1.49s/it]
 85%|████████▍ | 20707/24420 [8:38:43<1:32:36,  1.50s/it]
 85%|████████▍ | 20708/24420 [8:38:44<1:32:50,  1.50s/it]
 85%|████████▍ | 20709/24420 [8:38:46<1:32:42,  1.50s/it]
 85%|████████▍ | 20710/24420 [8:38:47<1:32:48,  1.50s/it]
                                                         
{'loss': 0.0709, 'learning_rate': 5e-05, 'epoch': 4.24}

 85%|████████▍ | 20710/24420 [8:38:47<1:32:48,  1.50s/it]
 85%|████████▍ | 20711/24420 [8:38:49<1:32:51,  1.50s/it]
 85%|████████▍ | 20712/24420 [8:38:50<1:32:56,  1.50s/it]
 85%|████████▍ | 20713/24420 [8:38:52<1:32:40,  1.50s/it]
 85%|████████▍ | 20714/24420 [8:38:53<1:32:34,  1.50s/it]
 85%|████████▍ | 20715/24420 [8:38:55<1:32:34,  1.50s/it]
                                                         
{'loss': 0.0769, 'learning_rate': 5e-05, 'epoch': 4.24}

 85%|████████▍ | 20715/24420 [8:38:55<1:32:34,  1.50s/it]
 85%|████████▍ | 20716/24420 [8:38:56<1:32:28,  1.50s/it]
 85%|████████▍ | 20717/24420 [8:38:58<1:32:10,  1.49s/it]
 85%|████████▍ | 20718/24420 [8:38:59<1:32:07,  1.49s/it]
 85%|████████▍ | 20719/24420 [8:39:01<1:32:11,  1.49s/it]
 85%|████████▍ | 20720/24420 [8:39:02<1:32:13,  1.50s/it]
                                                         
{'loss': 0.0757, 'learning_rate': 5e-05, 'epoch': 4.24}

 85%|████████▍ | 20720/24420 [8:39:02<1:32:13,  1.50s/it]
 85%|████████▍ | 20721/24420 [8:39:04<1:32:16,  1.50s/it]
 85%|████████▍ | 20722/24420 [8:39:05<1:32:04,  1.49s/it]
 85%|████████▍ | 20723/24420 [8:39:07<1:31:48,  1.49s/it]
 85%|████████▍ | 20724/24420 [8:39:08<1:31:44,  1.49s/it]
 85%|████████▍ | 20725/24420 [8:39:09<1:31:31,  1.49s/it]
                                                         
{'loss': 0.0758, 'learning_rate': 5e-05, 'epoch': 4.24}

 85%|████████▍ | 20725/24420 [8:39:09<1:31:31,  1.49s/it]
 85%|████████▍ | 20726/24420 [8:39:11<1:31:27,  1.49s/it]
 85%|████████▍ | 20727/24420 [8:39:12<1:31:32,  1.49s/it]
 85%|████████▍ | 20728/24420 [8:39:14<1:31:30,  1.49s/it]
 85%|████████▍ | 20729/24420 [8:39:15<1:31:24,  1.49s/it]
 85%|████████▍ | 20730/24420 [8:39:17<1:31:33,  1.49s/it]
                                                         
{'loss': 0.0757, 'learning_rate': 5e-05, 'epoch': 4.24}

 85%|████████▍ | 20730/24420 [8:39:17<1:31:33,  1.49s/it]
 85%|████████▍ | 20731/24420 [8:39:18<1:31:30,  1.49s/it]
 85%|████████▍ | 20732/24420 [8:39:20<1:31:31,  1.49s/it]
 85%|████████▍ | 20733/24420 [8:39:21<1:31:27,  1.49s/it]
 85%|████████▍ | 20734/24420 [8:39:23<1:31:26,  1.49s/it]
 85%|████████▍ | 20735/24420 [8:39:24<1:31:30,  1.49s/it]
                                                         
{'loss': 0.0649, 'learning_rate': 5e-05, 'epoch': 4.25}

 85%|████████▍ | 20735/24420 [8:39:24<1:31:30,  1.49s/it]
 85%|████████▍ | 20736/24420 [8:39:26<1:31:28,  1.49s/it]
 85%|████████▍ | 20737/24420 [8:39:27<1:31:12,  1.49s/it]
 85%|████████▍ | 20738/24420 [8:39:29<1:31:18,  1.49s/it]
 85%|████████▍ | 20739/24420 [8:39:30<1:31:27,  1.49s/it]
 85%|████████▍ | 20740/24420 [8:39:32<1:31:18,  1.49s/it]
                                                         
{'loss': 0.0832, 'learning_rate': 5e-05, 'epoch': 4.25}

 85%|████████▍ | 20740/24420 [8:39:32<1:31:18,  1.49s/it]
 85%|████████▍ | 20741/24420 [8:39:33<1:31:18,  1.49s/it]
 85%|████████▍ | 20742/24420 [8:39:35<1:31:26,  1.49s/it]
 85%|████████▍ | 20743/24420 [8:39:36<1:31:22,  1.49s/it]
 85%|████████▍ | 20744/24420 [8:39:38<1:31:17,  1.49s/it]
 85%|████████▍ | 20745/24420 [8:39:39<1:31:35,  1.50s/it]
                                                         
{'loss': 0.0839, 'learning_rate': 5e-05, 'epoch': 4.25}

 85%|████████▍ | 20745/24420 [8:39:39<1:31:35,  1.50s/it]
 85%|████████▍ | 20746/24420 [8:39:41<1:31:25,  1.49s/it]
 85%|████████▍ | 20747/24420 [8:39:42<1:31:22,  1.49s/it]
 85%|████████▍ | 20748/24420 [8:39:44<1:31:21,  1.49s/it]
 85%|████████▍ | 20749/24420 [8:39:45<1:31:20,  1.49s/it]
 85%|████████▍ | 20750/24420 [8:39:47<1:31:13,  1.49s/it]
                                                         
{'loss': 0.0708, 'learning_rate': 5e-05, 'epoch': 4.25}

 85%|████████▍ | 20750/24420 [8:39:47<1:31:13,  1.49s/it]
 85%|████████▍ | 20751/24420 [8:39:48<1:31:18,  1.49s/it]
 85%|████████▍ | 20752/24420 [8:39:50<1:31:11,  1.49s/it]
 85%|████████▍ | 20753/24420 [8:39:51<1:31:03,  1.49s/it]
 85%|████████▍ | 20754/24420 [8:39:53<1:30:58,  1.49s/it]
 85%|████████▍ | 20755/24420 [8:39:54<1:30:48,  1.49s/it]
                                                         
{'loss': 0.0566, 'learning_rate': 5e-05, 'epoch': 4.25}

 85%|████████▍ | 20755/24420 [8:39:54<1:30:48,  1.49s/it]
 85%|████████▍ | 20756/24420 [8:39:56<1:30:58,  1.49s/it]
 85%|████████▌ | 20757/24420 [8:39:57<1:30:49,  1.49s/it]
 85%|████████▌ | 20758/24420 [8:39:59<1:30:58,  1.49s/it]
 85%|████████▌ | 20759/24420 [8:40:00<1:30:58,  1.49s/it]
 85%|████████▌ | 20760/24420 [8:40:02<1:30:58,  1.49s/it]
                                                         
{'loss': 0.0705, 'learning_rate': 5e-05, 'epoch': 4.25}

 85%|████████▌ | 20760/24420 [8:40:02<1:30:58,  1.49s/it]
 85%|████████▌ | 20761/24420 [8:40:03<1:31:01,  1.49s/it]
 85%|████████▌ | 20762/24420 [8:40:05<1:31:05,  1.49s/it]
 85%|████████▌ | 20763/24420 [8:40:06<1:31:08,  1.50s/it]
 85%|████████▌ | 20764/24420 [8:40:08<1:31:03,  1.49s/it]
 85%|████████▌ | 20765/24420 [8:40:09<1:31:04,  1.50s/it]
                                                         
{'loss': 0.0789, 'learning_rate': 5e-05, 'epoch': 4.25}

 85%|████████▌ | 20765/24420 [8:40:09<1:31:04,  1.50s/it]
 85%|████████▌ | 20766/24420 [8:40:11<1:30:55,  1.49s/it]
 85%|████████▌ | 20767/24420 [8:40:12<1:30:49,  1.49s/it]
 85%|████████▌ | 20768/24420 [8:40:14<1:30:48,  1.49s/it]
 85%|████████▌ | 20769/24420 [8:40:15<1:30:43,  1.49s/it]
 85%|████████▌ | 20770/24420 [8:40:17<1:30:43,  1.49s/it]
                                                         
{'loss': 0.0787, 'learning_rate': 5e-05, 'epoch': 4.25}

 85%|████████▌ | 20770/24420 [8:40:17<1:30:43,  1.49s/it]
 85%|████████▌ | 20771/24420 [8:40:18<1:30:37,  1.49s/it]
 85%|████████▌ | 20772/24420 [8:40:20<1:30:46,  1.49s/it]
 85%|████████▌ | 20773/24420 [8:40:21<1:30:39,  1.49s/it]
 85%|████████▌ | 20774/24420 [8:40:23<1:30:34,  1.49s/it]
 85%|████████▌ | 20775/24420 [8:40:24<1:30:32,  1.49s/it]
                                                         
{'loss': 0.0903, 'learning_rate': 5e-05, 'epoch': 4.25}

 85%|████████▌ | 20775/24420 [8:40:24<1:30:32,  1.49s/it]
 85%|████████▌ | 20776/24420 [8:40:26<1:30:26,  1.49s/it]
 85%|████████▌ | 20777/24420 [8:40:27<1:30:16,  1.49s/it]
 85%|████████▌ | 20778/24420 [8:40:28<1:30:10,  1.49s/it]
 85%|████████▌ | 20779/24420 [8:40:30<1:30:10,  1.49s/it]
 85%|████████▌ | 20780/24420 [8:40:31<1:30:14,  1.49s/it]
                                                         
{'loss': 0.0693, 'learning_rate': 5e-05, 'epoch': 4.25}

 85%|████████▌ | 20780/24420 [8:40:31<1:30:14,  1.49s/it]
 85%|████████▌ | 20781/24420 [8:40:33<1:30:16,  1.49s/it]
 85%|████████▌ | 20782/24420 [8:40:34<1:30:28,  1.49s/it]
 85%|████████▌ | 20783/24420 [8:40:36<1:30:31,  1.49s/it]
 85%|████████▌ | 20784/24420 [8:40:37<1:30:22,  1.49s/it]
 85%|████████▌ | 20785/24420 [8:40:39<1:30:13,  1.49s/it]
                                                         
{'loss': 0.0694, 'learning_rate': 5e-05, 'epoch': 4.26}

 85%|████████▌ | 20785/24420 [8:40:39<1:30:13,  1.49s/it]
 85%|████████▌ | 20786/24420 [8:40:40<1:30:41,  1.50s/it]
 85%|████████▌ | 20787/24420 [8:40:42<1:30:33,  1.50s/it]
 85%|████████▌ | 20788/24420 [8:40:43<1:30:33,  1.50s/it]
 85%|████████▌ | 20789/24420 [8:40:45<1:30:25,  1.49s/it]
 85%|████████▌ | 20790/24420 [8:40:46<1:30:17,  1.49s/it]
                                                         
{'loss': 0.0851, 'learning_rate': 5e-05, 'epoch': 4.26}

 85%|████████▌ | 20790/24420 [8:40:46<1:30:17,  1.49s/it]
 85%|████████▌ | 20791/24420 [8:40:48<1:30:16,  1.49s/it]
 85%|████████▌ | 20792/24420 [8:40:49<1:30:18,  1.49s/it]
 85%|████████▌ | 20793/24420 [8:40:51<1:30:15,  1.49s/it]
 85%|████████▌ | 20794/24420 [8:40:52<1:30:23,  1.50s/it]
 85%|████████▌ | 20795/24420 [8:40:54<1:30:13,  1.49s/it]
                                                         
{'loss': 0.0849, 'learning_rate': 5e-05, 'epoch': 4.26}

 85%|████████▌ | 20795/24420 [8:40:54<1:30:13,  1.49s/it]
 85%|████████▌ | 20796/24420 [8:40:55<1:30:04,  1.49s/it]
 85%|████████▌ | 20797/24420 [8:40:57<1:30:04,  1.49s/it]
 85%|████████▌ | 20798/24420 [8:40:58<1:30:43,  1.50s/it]
 85%|████████▌ | 20799/24420 [8:41:00<1:30:32,  1.50s/it]
 85%|████████▌ | 20800/24420 [8:41:01<1:30:15,  1.50s/it]
                                                         
{'loss': 0.0799, 'learning_rate': 5e-05, 'epoch': 4.26}

 85%|████████▌ | 20800/24420 [8:41:01<1:30:15,  1.50s/it]
 85%|████████▌ | 20801/24420 [8:41:08<3:01:29,  3.01s/it]
 85%|████████▌ | 20802/24420 [8:41:09<2:33:51,  2.55s/it]
 85%|████████▌ | 20803/24420 [8:41:11<2:14:26,  2.23s/it]
 85%|████████▌ | 20804/24420 [8:41:12<2:00:55,  2.01s/it]
 85%|████████▌ | 20805/24420 [8:41:14<1:51:46,  1.86s/it]
                                                         
{'loss': 0.0752, 'learning_rate': 5e-05, 'epoch': 4.26}

 85%|████████▌ | 20805/24420 [8:41:14<1:51:46,  1.86s/it]
 85%|████████▌ | 20806/24420 [8:41:15<1:45:22,  1.75s/it]
 85%|████████▌ | 20807/24420 [8:41:17<1:40:34,  1.67s/it]
 85%|████████▌ | 20808/24420 [8:41:18<1:37:13,  1.62s/it]
 85%|████████▌ | 20809/24420 [8:41:20<1:34:57,  1.58s/it]
 85%|████████▌ | 20810/24420 [8:41:21<1:33:23,  1.55s/it]
                                                         
{'loss': 0.0801, 'learning_rate': 5e-05, 'epoch': 4.26}

 85%|████████▌ | 20810/24420 [8:41:21<1:33:23,  1.55s/it]
 85%|████████▌ | 20811/24420 [8:41:23<1:32:13,  1.53s/it]
 85%|████████▌ | 20812/24420 [8:41:24<1:31:30,  1.52s/it]
 85%|████████▌ | 20813/24420 [8:41:26<1:31:03,  1.51s/it]
 85%|████████▌ | 20814/24420 [8:41:27<1:30:29,  1.51s/it]
 85%|████████▌ | 20815/24420 [8:41:29<1:30:16,  1.50s/it]
                                                         
{'loss': 0.082, 'learning_rate': 5e-05, 'epoch': 4.26}

 85%|████████▌ | 20815/24420 [8:41:29<1:30:16,  1.50s/it]
 85%|████████▌ | 20816/24420 [8:41:30<1:30:10,  1.50s/it]
 85%|████████▌ | 20817/24420 [8:41:32<1:29:58,  1.50s/it]
 85%|████████▌ | 20818/24420 [8:41:33<1:29:52,  1.50s/it]
 85%|████████▌ | 20819/24420 [8:41:35<1:29:36,  1.49s/it]
 85%|████████▌ | 20820/24420 [8:41:36<1:29:38,  1.49s/it]
                                                         
{'loss': 0.0643, 'learning_rate': 5e-05, 'epoch': 4.26}

 85%|████████▌ | 20820/24420 [8:41:36<1:29:38,  1.49s/it]
 85%|████████▌ | 20821/24420 [8:41:38<1:29:34,  1.49s/it]
 85%|████████▌ | 20822/24420 [8:41:39<1:29:26,  1.49s/it]
 85%|████████▌ | 20823/24420 [8:41:41<1:29:22,  1.49s/it]
 85%|████████▌ | 20824/24420 [8:41:42<1:29:42,  1.50s/it]
 85%|████████▌ | 20825/24420 [8:41:44<1:29:37,  1.50s/it]
                                                         
{'loss': 0.067, 'learning_rate': 5e-05, 'epoch': 4.26}

 85%|████████▌ | 20825/24420 [8:41:44<1:29:37,  1.50s/it]
 85%|████████▌ | 20826/24420 [8:41:45<1:29:49,  1.50s/it]
 85%|████████▌ | 20827/24420 [8:41:47<1:29:42,  1.50s/it]
 85%|████████▌ | 20828/24420 [8:41:48<1:29:29,  1.49s/it]
 85%|████████▌ | 20829/24420 [8:41:50<1:29:35,  1.50s/it]
 85%|████████▌ | 20830/24420 [8:41:51<1:29:24,  1.49s/it]
                                                         
{'loss': 0.0632, 'learning_rate': 5e-05, 'epoch': 4.26}

 85%|████████▌ | 20830/24420 [8:41:51<1:29:24,  1.49s/it]
 85%|████████▌ | 20831/24420 [8:41:53<1:29:21,  1.49s/it]
 85%|████████▌ | 20832/24420 [8:41:54<1:29:20,  1.49s/it]
 85%|████████▌ | 20833/24420 [8:41:56<1:29:30,  1.50s/it]
 85%|████████▌ | 20834/24420 [8:41:57<1:29:23,  1.50s/it]
 85%|████████▌ | 20835/24420 [8:41:59<1:29:27,  1.50s/it]
                                                         
{'loss': 0.0829, 'learning_rate': 5e-05, 'epoch': 4.27}

 85%|████████▌ | 20835/24420 [8:41:59<1:29:27,  1.50s/it]
 85%|████████▌ | 20836/24420 [8:42:00<1:29:22,  1.50s/it]
 85%|████████▌ | 20837/24420 [8:42:02<1:29:22,  1.50s/it]
 85%|████████▌ | 20838/24420 [8:42:03<1:29:21,  1.50s/it]
 85%|████████▌ | 20839/24420 [8:42:05<1:29:17,  1.50s/it]
 85%|████████▌ | 20840/24420 [8:42:06<1:29:18,  1.50s/it]
                                                         
{'loss': 0.0806, 'learning_rate': 5e-05, 'epoch': 4.27}

 85%|████████▌ | 20840/24420 [8:42:06<1:29:18,  1.50s/it]
 85%|████████▌ | 20841/24420 [8:42:08<1:29:14,  1.50s/it]
 85%|████████▌ | 20842/24420 [8:42:09<1:29:05,  1.49s/it]
 85%|████████▌ | 20843/24420 [8:42:11<1:29:04,  1.49s/it]
 85%|████████▌ | 20844/24420 [8:42:12<1:28:58,  1.49s/it]
 85%|████████▌ | 20845/24420 [8:42:14<1:28:54,  1.49s/it]
                                                         
{'loss': 0.0783, 'learning_rate': 5e-05, 'epoch': 4.27}

 85%|████████▌ | 20845/24420 [8:42:14<1:28:54,  1.49s/it]
 85%|████████▌ | 20846/24420 [8:42:15<1:28:51,  1.49s/it]
 85%|████████▌ | 20847/24420 [8:42:17<1:28:48,  1.49s/it]
 85%|████████▌ | 20848/24420 [8:42:18<1:28:41,  1.49s/it]
 85%|████████▌ | 20849/24420 [8:42:20<1:28:39,  1.49s/it]
 85%|████████▌ | 20850/24420 [8:42:21<1:28:38,  1.49s/it]
                                                         
{'loss': 0.0742, 'learning_rate': 5e-05, 'epoch': 4.27}

 85%|████████▌ | 20850/24420 [8:42:21<1:28:38,  1.49s/it]
 85%|████████▌ | 20851/24420 [8:42:23<1:28:58,  1.50s/it]
 85%|████████▌ | 20852/24420 [8:42:24<1:28:51,  1.49s/it]
 85%|████████▌ | 20853/24420 [8:42:26<1:28:44,  1.49s/it]
 85%|████████▌ | 20854/24420 [8:42:27<1:28:41,  1.49s/it]
 85%|████████▌ | 20855/24420 [8:42:29<1:28:38,  1.49s/it]
                                                         
{'loss': 0.077, 'learning_rate': 5e-05, 'epoch': 4.27}

 85%|████████▌ | 20855/24420 [8:42:29<1:28:38,  1.49s/it]
 85%|████████▌ | 20856/24420 [8:42:30<1:29:00,  1.50s/it]
 85%|████████▌ | 20857/24420 [8:42:32<1:28:49,  1.50s/it]
 85%|████████▌ | 20858/24420 [8:42:33<1:28:36,  1.49s/it]
 85%|████████▌ | 20859/24420 [8:42:35<1:28:42,  1.49s/it]
 85%|████████▌ | 20860/24420 [8:42:36<1:28:44,  1.50s/it]
                                                         
{'loss': 0.0887, 'learning_rate': 5e-05, 'epoch': 4.27}

 85%|████████▌ | 20860/24420 [8:42:36<1:28:44,  1.50s/it]
 85%|████████▌ | 20861/24420 [8:42:38<1:29:37,  1.51s/it]
 85%|████████▌ | 20862/24420 [8:42:39<1:29:23,  1.51s/it]
 85%|████████▌ | 20863/24420 [8:42:41<1:29:02,  1.50s/it]
 85%|████████▌ | 20864/24420 [8:42:42<1:28:55,  1.50s/it]
 85%|████████▌ | 20865/24420 [8:42:44<1:28:44,  1.50s/it]
                                                         
{'loss': 0.0697, 'learning_rate': 5e-05, 'epoch': 4.27}

 85%|████████▌ | 20865/24420 [8:42:44<1:28:44,  1.50s/it]
 85%|████████▌ | 20866/24420 [8:42:45<1:28:36,  1.50s/it]
 85%|████████▌ | 20867/24420 [8:42:47<1:28:28,  1.49s/it]
 85%|████████▌ | 20868/24420 [8:42:48<1:28:33,  1.50s/it]
 85%|████████▌ | 20869/24420 [8:42:50<1:28:36,  1.50s/it]
 85%|████████▌ | 20870/24420 [8:42:51<1:28:47,  1.50s/it]
                                                         
{'loss': 0.0724, 'learning_rate': 5e-05, 'epoch': 4.27}

 85%|████████▌ | 20870/24420 [8:42:51<1:28:47,  1.50s/it]
 85%|████████▌ | 20871/24420 [8:42:53<1:28:30,  1.50s/it]
 85%|████████▌ | 20872/24420 [8:42:54<1:28:26,  1.50s/it]
 85%|████████▌ | 20873/24420 [8:42:56<1:28:21,  1.49s/it]
 85%|████████▌ | 20874/24420 [8:42:57<1:28:21,  1.50s/it]
 85%|████████▌ | 20875/24420 [8:42:58<1:28:22,  1.50s/it]
                                                         
{'loss': 0.0677, 'learning_rate': 5e-05, 'epoch': 4.27}

 85%|████████▌ | 20875/24420 [8:42:58<1:28:22,  1.50s/it]
 85%|████████▌ | 20876/24420 [8:43:00<1:28:24,  1.50s/it]
 85%|████████▌ | 20877/24420 [8:43:01<1:28:23,  1.50s/it]
 85%|████████▌ | 20878/24420 [8:43:03<1:28:20,  1.50s/it]
 85%|████████▌ | 20879/24420 [8:43:04<1:28:23,  1.50s/it]
 86%|████████▌ | 20880/24420 [8:43:06<1:28:11,  1.49s/it]
                                                         
{'loss': 0.0762, 'learning_rate': 5e-05, 'epoch': 4.28}

 86%|████████▌ | 20880/24420 [8:43:06<1:28:11,  1.49s/it]
 86%|████████▌ | 20881/24420 [8:43:08<1:29:06,  1.51s/it]
 86%|████████▌ | 20882/24420 [8:43:09<1:28:46,  1.51s/it]
 86%|████████▌ | 20883/24420 [8:43:11<1:28:23,  1.50s/it]
 86%|████████▌ | 20884/24420 [8:43:12<1:28:09,  1.50s/it]
 86%|████████▌ | 20885/24420 [8:43:13<1:28:02,  1.49s/it]
                                                         
{'loss': 0.0671, 'learning_rate': 5e-05, 'epoch': 4.28}

 86%|████████▌ | 20885/24420 [8:43:13<1:28:02,  1.49s/it]
 86%|████████▌ | 20886/24420 [8:43:15<1:28:14,  1.50s/it]
 86%|████████▌ | 20887/24420 [8:43:16<1:28:05,  1.50s/it]
 86%|████████▌ | 20888/24420 [8:43:18<1:28:06,  1.50s/it]
 86%|████████▌ | 20889/24420 [8:43:19<1:27:54,  1.49s/it]
 86%|████████▌ | 20890/24420 [8:43:21<1:27:49,  1.49s/it]
                                                         
{'loss': 0.0708, 'learning_rate': 5e-05, 'epoch': 4.28}

 86%|████████▌ | 20890/24420 [8:43:21<1:27:49,  1.49s/it]
 86%|████████▌ | 20891/24420 [8:43:22<1:27:49,  1.49s/it]
 86%|████████▌ | 20892/24420 [8:43:24<1:27:42,  1.49s/it]
 86%|████████▌ | 20893/24420 [8:43:25<1:27:38,  1.49s/it]
 86%|████████▌ | 20894/24420 [8:43:27<1:27:38,  1.49s/it]
 86%|████████▌ | 20895/24420 [8:43:28<1:27:36,  1.49s/it]
                                                         
{'loss': 0.0924, 'learning_rate': 5e-05, 'epoch': 4.28}

 86%|████████▌ | 20895/24420 [8:43:28<1:27:36,  1.49s/it]
 86%|████████▌ | 20896/24420 [8:43:30<1:27:33,  1.49s/it]
 86%|████████▌ | 20897/24420 [8:43:31<1:27:38,  1.49s/it]
 86%|████████▌ | 20898/24420 [8:43:33<1:27:46,  1.50s/it]
 86%|████████▌ | 20899/24420 [8:43:34<1:27:39,  1.49s/it]
 86%|████████▌ | 20900/24420 [8:43:36<1:27:41,  1.49s/it]
                                                         
{'loss': 0.0777, 'learning_rate': 5e-05, 'epoch': 4.28}

 86%|████████▌ | 20900/24420 [8:43:36<1:27:41,  1.49s/it]
 86%|████████▌ | 20901/24420 [8:43:37<1:27:50,  1.50s/it]
 86%|████████▌ | 20902/24420 [8:43:39<1:28:05,  1.50s/it]
 86%|████████▌ | 20903/24420 [8:43:40<1:27:55,  1.50s/it]
 86%|████████▌ | 20904/24420 [8:43:42<1:27:55,  1.50s/it]
 86%|████████▌ | 20905/24420 [8:43:43<1:27:51,  1.50s/it]
                                                         
{'loss': 0.0803, 'learning_rate': 5e-05, 'epoch': 4.28}

 86%|████████▌ | 20905/24420 [8:43:43<1:27:51,  1.50s/it]
 86%|████████▌ | 20906/24420 [8:43:45<1:27:48,  1.50s/it]
 86%|████████▌ | 20907/24420 [8:43:46<1:27:42,  1.50s/it]
 86%|████████▌ | 20908/24420 [8:43:48<1:27:42,  1.50s/it]
 86%|████████▌ | 20909/24420 [8:43:49<1:27:25,  1.49s/it]
 86%|████████▌ | 20910/24420 [8:43:51<1:27:23,  1.49s/it]
                                                         
{'loss': 0.0813, 'learning_rate': 5e-05, 'epoch': 4.28}

 86%|████████▌ | 20910/24420 [8:43:51<1:27:23,  1.49s/it]
 86%|████████▌ | 20911/24420 [8:43:52<1:27:20,  1.49s/it]
 86%|████████▌ | 20912/24420 [8:43:54<1:27:13,  1.49s/it]
 86%|████████▌ | 20913/24420 [8:43:55<1:27:12,  1.49s/it]
 86%|████████▌ | 20914/24420 [8:43:57<1:27:15,  1.49s/it]
 86%|████████▌ | 20915/24420 [8:43:58<1:27:14,  1.49s/it]
                                                         
{'loss': 0.091, 'learning_rate': 5e-05, 'epoch': 4.28}

 86%|████████▌ | 20915/24420 [8:43:58<1:27:14,  1.49s/it]
 86%|████████▌ | 20916/24420 [8:44:00<1:27:17,  1.49s/it]
 86%|████████▌ | 20917/24420 [8:44:01<1:27:06,  1.49s/it]
 86%|████████▌ | 20918/24420 [8:44:03<1:27:03,  1.49s/it]
 86%|████████▌ | 20919/24420 [8:44:04<1:27:12,  1.49s/it]
 86%|████████▌ | 20920/24420 [8:44:06<1:27:08,  1.49s/it]
                                                         
{'loss': 0.0803, 'learning_rate': 5e-05, 'epoch': 4.28}

 86%|████████▌ | 20920/24420 [8:44:06<1:27:08,  1.49s/it]
 86%|████████▌ | 20921/24420 [8:44:07<1:26:59,  1.49s/it]
 86%|████████▌ | 20922/24420 [8:44:09<1:26:58,  1.49s/it]
 86%|████████▌ | 20923/24420 [8:44:10<1:27:00,  1.49s/it]
 86%|████████▌ | 20924/24420 [8:44:12<1:26:52,  1.49s/it]
 86%|████████▌ | 20925/24420 [8:44:13<1:26:47,  1.49s/it]
                                                         
{'loss': 0.071, 'learning_rate': 5e-05, 'epoch': 4.28}

 86%|████████▌ | 20925/24420 [8:44:13<1:26:47,  1.49s/it]
 86%|████████▌ | 20926/24420 [8:44:15<1:26:49,  1.49s/it]
 86%|████████▌ | 20927/24420 [8:44:16<1:26:42,  1.49s/it]
 86%|████████▌ | 20928/24420 [8:44:18<1:26:54,  1.49s/it]
 86%|████████▌ | 20929/24420 [8:44:19<1:27:06,  1.50s/it]
 86%|████████▌ | 20930/24420 [8:44:21<1:26:59,  1.50s/it]
                                                         
{'loss': 0.0715, 'learning_rate': 5e-05, 'epoch': 4.29}

 86%|████████▌ | 20930/24420 [8:44:21<1:26:59,  1.50s/it]
 86%|████████▌ | 20931/24420 [8:44:22<1:27:05,  1.50s/it]
 86%|████████▌ | 20932/24420 [8:44:24<1:26:58,  1.50s/it]
 86%|████████▌ | 20933/24420 [8:44:25<1:26:50,  1.49s/it]
 86%|████████▌ | 20934/24420 [8:44:27<1:26:37,  1.49s/it]
 86%|████████▌ | 20935/24420 [8:44:28<1:26:29,  1.49s/it]
                                                         
{'loss': 0.0804, 'learning_rate': 5e-05, 'epoch': 4.29}

 86%|████████▌ | 20935/24420 [8:44:28<1:26:29,  1.49s/it]
 86%|████████▌ | 20936/24420 [8:44:30<1:26:34,  1.49s/it]
 86%|████████▌ | 20937/24420 [8:44:31<1:26:24,  1.49s/it]
 86%|████████▌ | 20938/24420 [8:44:33<1:26:22,  1.49s/it]
 86%|████████▌ | 20939/24420 [8:44:34<1:26:20,  1.49s/it]
 86%|████████▌ | 20940/24420 [8:44:36<1:26:14,  1.49s/it]
                                                         
{'loss': 0.0851, 'learning_rate': 5e-05, 'epoch': 4.29}

 86%|████████▌ | 20940/24420 [8:44:36<1:26:14,  1.49s/it]
 86%|████████▌ | 20941/24420 [8:44:37<1:26:11,  1.49s/it]
 86%|████████▌ | 20942/24420 [8:44:39<1:26:18,  1.49s/it]
 86%|████████▌ | 20943/24420 [8:44:40<1:26:26,  1.49s/it]
 86%|████████▌ | 20944/24420 [8:44:42<1:26:19,  1.49s/it]
 86%|████████▌ | 20945/24420 [8:44:43<1:26:16,  1.49s/it]
                                                         
{'loss': 0.0673, 'learning_rate': 5e-05, 'epoch': 4.29}

 86%|████████▌ | 20945/24420 [8:44:43<1:26:16,  1.49s/it]
 86%|████████▌ | 20946/24420 [8:44:45<1:26:22,  1.49s/it]
 86%|████████▌ | 20947/24420 [8:44:46<1:26:32,  1.50s/it]
 86%|████████▌ | 20948/24420 [8:44:48<1:26:26,  1.49s/it]
 86%|████████▌ | 20949/24420 [8:44:49<1:26:27,  1.49s/it]
 86%|████████▌ | 20950/24420 [8:44:51<1:26:37,  1.50s/it]
                                                         
{'loss': 0.0716, 'learning_rate': 5e-05, 'epoch': 4.29}

 86%|████████▌ | 20950/24420 [8:44:51<1:26:37,  1.50s/it]
 86%|████████▌ | 20951/24420 [8:44:52<1:26:34,  1.50s/it]
 86%|████████▌ | 20952/24420 [8:44:54<1:26:20,  1.49s/it]
 86%|████████▌ | 20953/24420 [8:44:55<1:26:07,  1.49s/it]
 86%|████████▌ | 20954/24420 [8:44:57<1:26:07,  1.49s/it]
 86%|████████▌ | 20955/24420 [8:44:58<1:26:16,  1.49s/it]
                                                         
{'loss': 0.0706, 'learning_rate': 5e-05, 'epoch': 4.29}

 86%|████████▌ | 20955/24420 [8:44:58<1:26:16,  1.49s/it]
 86%|████████▌ | 20956/24420 [8:45:00<1:26:25,  1.50s/it]
 86%|████████▌ | 20957/24420 [8:45:01<1:26:25,  1.50s/it]
 86%|████████▌ | 20958/24420 [8:45:03<1:26:32,  1.50s/it]
 86%|████████▌ | 20959/24420 [8:45:04<1:26:20,  1.50s/it]
 86%|████████▌ | 20960/24420 [8:45:06<1:26:21,  1.50s/it]
                                                         
{'loss': 0.0649, 'learning_rate': 5e-05, 'epoch': 4.29}

 86%|████████▌ | 20960/24420 [8:45:06<1:26:21,  1.50s/it]
 86%|████████▌ | 20961/24420 [8:45:07<1:26:09,  1.49s/it]
 86%|████████▌ | 20962/24420 [8:45:08<1:25:59,  1.49s/it]
 86%|████████▌ | 20963/24420 [8:45:10<1:25:59,  1.49s/it]
 86%|████████▌ | 20964/24420 [8:45:11<1:25:51,  1.49s/it]
 86%|████████▌ | 20965/24420 [8:45:13<1:25:50,  1.49s/it]
                                                         
{'loss': 0.0651, 'learning_rate': 5e-05, 'epoch': 4.29}

 86%|████████▌ | 20965/24420 [8:45:13<1:25:50,  1.49s/it]
 86%|████████▌ | 20966/24420 [8:45:14<1:25:47,  1.49s/it]
 86%|████████▌ | 20967/24420 [8:45:16<1:25:52,  1.49s/it]
 86%|████████▌ | 20968/24420 [8:45:17<1:25:47,  1.49s/it]
 86%|████████▌ | 20969/24420 [8:45:19<1:25:45,  1.49s/it]
 86%|████████▌ | 20970/24420 [8:45:20<1:25:46,  1.49s/it]
                                                         
{'loss': 0.0757, 'learning_rate': 5e-05, 'epoch': 4.29}

 86%|████████▌ | 20970/24420 [8:45:20<1:25:46,  1.49s/it]
 86%|████████▌ | 20971/24420 [8:45:22<1:25:42,  1.49s/it]
 86%|████████▌ | 20972/24420 [8:45:23<1:25:28,  1.49s/it]
 86%|████████▌ | 20973/24420 [8:45:25<1:25:26,  1.49s/it]
 86%|████████▌ | 20974/24420 [8:45:26<1:25:28,  1.49s/it]
 86%|████████▌ | 20975/24420 [8:45:28<1:25:30,  1.49s/it]
                                                         
{'loss': 0.0626, 'learning_rate': 5e-05, 'epoch': 4.29}

 86%|████████▌ | 20975/24420 [8:45:28<1:25:30,  1.49s/it]
 86%|████████▌ | 20976/24420 [8:45:29<1:25:42,  1.49s/it]
 86%|████████▌ | 20977/24420 [8:45:31<1:25:54,  1.50s/it]
 86%|████████▌ | 20978/24420 [8:45:32<1:25:56,  1.50s/it]
 86%|████████▌ | 20979/24420 [8:45:34<1:25:49,  1.50s/it]
 86%|████████▌ | 20980/24420 [8:45:35<1:25:46,  1.50s/it]
                                                         
{'loss': 0.0687, 'learning_rate': 5e-05, 'epoch': 4.3}

 86%|████████▌ | 20980/24420 [8:45:35<1:25:46,  1.50s/it]
 86%|████████▌ | 20981/24420 [8:45:37<1:25:46,  1.50s/it]
 86%|████████▌ | 20982/24420 [8:45:38<1:25:34,  1.49s/it]
 86%|████████▌ | 20983/24420 [8:45:40<1:25:32,  1.49s/it]
 86%|████████▌ | 20984/24420 [8:45:41<1:25:21,  1.49s/it]
 86%|████████▌ | 20985/24420 [8:45:43<1:25:57,  1.50s/it]
                                                         
{'loss': 0.0579, 'learning_rate': 5e-05, 'epoch': 4.3}

 86%|████████▌ | 20985/24420 [8:45:43<1:25:57,  1.50s/it]
 86%|████████▌ | 20986/24420 [8:45:44<1:25:41,  1.50s/it]
 86%|████████▌ | 20987/24420 [8:45:46<1:25:37,  1.50s/it]
 86%|████████▌ | 20988/24420 [8:45:47<1:25:33,  1.50s/it]
 86%|████████▌ | 20989/24420 [8:45:49<1:25:33,  1.50s/it]
 86%|████████▌ | 20990/24420 [8:45:50<1:25:25,  1.49s/it]
                                                         
{'loss': 0.0986, 'learning_rate': 5e-05, 'epoch': 4.3}

 86%|████████▌ | 20990/24420 [8:45:50<1:25:25,  1.49s/it]
 86%|████████▌ | 20991/24420 [8:45:52<1:25:35,  1.50s/it]
 86%|████████▌ | 20992/24420 [8:45:53<1:25:28,  1.50s/it]
 86%|████████▌ | 20993/24420 [8:45:55<1:25:28,  1.50s/it]
 86%|████████▌ | 20994/24420 [8:45:56<1:25:12,  1.49s/it]
 86%|████████▌ | 20995/24420 [8:45:58<1:24:59,  1.49s/it]
                                                         
{'loss': 0.0759, 'learning_rate': 5e-05, 'epoch': 4.3}

 86%|████████▌ | 20995/24420 [8:45:58<1:24:59,  1.49s/it]
 86%|████████▌ | 20996/24420 [8:45:59<1:25:01,  1.49s/it]
 86%|████████▌ | 20997/24420 [8:46:01<1:24:59,  1.49s/it]
 86%|████████▌ | 20998/24420 [8:46:02<1:24:59,  1.49s/it]
 86%|████████▌ | 20999/24420 [8:46:04<1:25:09,  1.49s/it]
 86%|████████▌ | 21000/24420 [8:46:05<1:25:10,  1.49s/it]
                                                         
{'loss': 0.0772, 'learning_rate': 5e-05, 'epoch': 4.3}

 86%|████████▌ | 21000/24420 [8:46:05<1:25:10,  1.49s/it]
 86%|████████▌ | 21001/24420 [8:46:10<2:17:49,  2.42s/it]
 86%|████████▌ | 21002/24420 [8:46:11<2:01:49,  2.14s/it]
 86%|████████▌ | 21003/24420 [8:46:13<1:50:41,  1.94s/it]
 86%|████████▌ | 21004/24420 [8:46:14<1:42:53,  1.81s/it]
 86%|████████▌ | 21005/24420 [8:46:16<1:37:28,  1.71s/it]
                                                         
{'loss': 0.0704, 'learning_rate': 5e-05, 'epoch': 4.3}

 86%|████████▌ | 21005/24420 [8:46:16<1:37:28,  1.71s/it]
 86%|████████▌ | 21006/24420 [8:46:17<1:33:55,  1.65s/it]
 86%|████████▌ | 21007/24420 [8:46:19<1:31:08,  1.60s/it]
 86%|████████▌ | 21008/24420 [8:46:20<1:29:14,  1.57s/it]
 86%|████████▌ | 21009/24420 [8:46:22<1:28:02,  1.55s/it]
 86%|████████▌ | 21010/24420 [8:46:23<1:26:55,  1.53s/it]
                                                         
{'loss': 0.0815, 'learning_rate': 5e-05, 'epoch': 4.3}

 86%|████████▌ | 21010/24420 [8:46:23<1:26:55,  1.53s/it]
 86%|████████▌ | 21011/24420 [8:46:25<1:26:26,  1.52s/it]
 86%|████████▌ | 21012/24420 [8:46:26<1:25:50,  1.51s/it]
 86%|████████▌ | 21013/24420 [8:46:28<1:25:30,  1.51s/it]
 86%|████████▌ | 21014/24420 [8:46:29<1:25:09,  1.50s/it]
 86%|████████▌ | 21015/24420 [8:46:31<1:24:52,  1.50s/it]
                                                         
{'loss': 0.0704, 'learning_rate': 5e-05, 'epoch': 4.3}

 86%|████████▌ | 21015/24420 [8:46:31<1:24:52,  1.50s/it]
 86%|████████▌ | 21016/24420 [8:46:32<1:24:47,  1.49s/it]
 86%|████████▌ | 21017/24420 [8:46:34<1:24:46,  1.49s/it]
 86%|████████▌ | 21018/24420 [8:46:35<1:24:38,  1.49s/it]
 86%|████████▌ | 21019/24420 [8:46:37<1:24:34,  1.49s/it]
 86%|████████▌ | 21020/24420 [8:46:38<1:24:25,  1.49s/it]
                                                         
{'loss': 0.0794, 'learning_rate': 5e-05, 'epoch': 4.3}

 86%|████████▌ | 21020/24420 [8:46:38<1:24:25,  1.49s/it]
 86%|████████▌ | 21021/24420 [8:46:40<1:24:31,  1.49s/it]
 86%|████████▌ | 21022/24420 [8:46:41<1:24:25,  1.49s/it]
 86%|████████▌ | 21023/24420 [8:46:43<1:24:39,  1.50s/it]
 86%|████████▌ | 21024/24420 [8:46:44<1:24:47,  1.50s/it]
 86%|████████▌ | 21025/24420 [8:46:46<1:24:37,  1.50s/it]
                                                         
{'loss': 0.0843, 'learning_rate': 5e-05, 'epoch': 4.3}

 86%|████████▌ | 21025/24420 [8:46:46<1:24:37,  1.50s/it]
 86%|████████▌ | 21026/24420 [8:46:47<1:24:53,  1.50s/it]
 86%|████████▌ | 21027/24420 [8:46:49<1:24:32,  1.50s/it]
 86%|████████▌ | 21028/24420 [8:46:50<1:24:40,  1.50s/it]
 86%|████████▌ | 21029/24420 [8:46:52<1:24:23,  1.49s/it]
 86%|████████▌ | 21030/24420 [8:46:53<1:24:21,  1.49s/it]
                                                         
{'loss': 0.0674, 'learning_rate': 5e-05, 'epoch': 4.31}

 86%|████████▌ | 21030/24420 [8:46:53<1:24:21,  1.49s/it]
 86%|████████▌ | 21031/24420 [8:46:55<1:24:21,  1.49s/it]
 86%|████████▌ | 21032/24420 [8:46:56<1:24:17,  1.49s/it]
 86%|████████▌ | 21033/24420 [8:46:58<1:24:13,  1.49s/it]
 86%|████████▌ | 21034/24420 [8:46:59<1:24:05,  1.49s/it]
 86%|████████▌ | 21035/24420 [8:47:01<1:24:02,  1.49s/it]
                                                         
{'loss': 0.0836, 'learning_rate': 5e-05, 'epoch': 4.31}

 86%|████████▌ | 21035/24420 [8:47:01<1:24:02,  1.49s/it]
 86%|████████▌ | 21036/24420 [8:47:02<1:24:02,  1.49s/it]
 86%|████████▌ | 21037/24420 [8:47:04<1:24:04,  1.49s/it]
 86%|████████▌ | 21038/24420 [8:47:05<1:24:08,  1.49s/it]
 86%|████████▌ | 21039/24420 [8:47:07<1:24:00,  1.49s/it]
 86%|████████▌ | 21040/24420 [8:47:08<1:23:59,  1.49s/it]
                                                         
{'loss': 0.0616, 'learning_rate': 5e-05, 'epoch': 4.31}

 86%|████████▌ | 21040/24420 [8:47:08<1:23:59,  1.49s/it]
 86%|████████▌ | 21041/24420 [8:47:10<1:24:08,  1.49s/it]
 86%|████████▌ | 21042/24420 [8:47:11<1:24:00,  1.49s/it]
 86%|████████▌ | 21043/24420 [8:47:12<1:23:50,  1.49s/it]
 86%|████████▌ | 21044/24420 [8:47:14<1:23:44,  1.49s/it]
 86%|████████▌ | 21045/24420 [8:47:15<1:24:01,  1.49s/it]
                                                         
{'loss': 0.0754, 'learning_rate': 5e-05, 'epoch': 4.31}

 86%|████████▌ | 21045/24420 [8:47:15<1:24:01,  1.49s/it]
 86%|████████▌ | 21046/24420 [8:47:17<1:23:57,  1.49s/it]
 86%|████████▌ | 21047/24420 [8:47:18<1:24:07,  1.50s/it]
 86%|████████▌ | 21048/24420 [8:47:20<1:24:01,  1.50s/it]
 86%|████████▌ | 21049/24420 [8:47:21<1:24:29,  1.50s/it]
 86%|████████▌ | 21050/24420 [8:47:23<1:24:14,  1.50s/it]
                                                         
{'loss': 0.0692, 'learning_rate': 5e-05, 'epoch': 4.31}

 86%|████████▌ | 21050/24420 [8:47:23<1:24:14,  1.50s/it]
 86%|████████▌ | 21051/24420 [8:47:24<1:24:03,  1.50s/it]
 86%|████████▌ | 21052/24420 [8:47:26<1:24:00,  1.50s/it]
 86%|████████▌ | 21053/24420 [8:47:27<1:23:53,  1.50s/it]
 86%|████████▌ | 21054/24420 [8:47:29<1:23:49,  1.49s/it]
 86%|████████▌ | 21055/24420 [8:47:30<1:23:32,  1.49s/it]
                                                         
{'loss': 0.0768, 'learning_rate': 5e-05, 'epoch': 4.31}

 86%|████████▌ | 21055/24420 [8:47:30<1:23:32,  1.49s/it]
 86%|████████▌ | 21056/24420 [8:47:32<1:23:34,  1.49s/it]
 86%|████████▌ | 21057/24420 [8:47:33<1:23:34,  1.49s/it]
 86%|████████▌ | 21058/24420 [8:47:35<1:23:31,  1.49s/it]
 86%|████████▌ | 21059/24420 [8:47:36<1:23:31,  1.49s/it]
 86%|████████▌ | 21060/24420 [8:47:38<1:23:25,  1.49s/it]
                                                         
{'loss': 0.0631, 'learning_rate': 5e-05, 'epoch': 4.31}

 86%|████████▌ | 21060/24420 [8:47:38<1:23:25,  1.49s/it]
 86%|████████▌ | 21061/24420 [8:47:39<1:23:26,  1.49s/it]
 86%|████████▌ | 21062/24420 [8:47:41<1:23:20,  1.49s/it]
 86%|████████▋ | 21063/24420 [8:47:42<1:23:22,  1.49s/it]
 86%|████████▋ | 21064/24420 [8:47:44<1:23:24,  1.49s/it]
 86%|████████▋ | 21065/24420 [8:47:45<1:23:26,  1.49s/it]
                                                         
{'loss': 0.0881, 'learning_rate': 5e-05, 'epoch': 4.31}

 86%|████████▋ | 21065/24420 [8:47:45<1:23:26,  1.49s/it]
 86%|████████▋ | 21066/24420 [8:47:47<1:23:20,  1.49s/it]
 86%|████████▋ | 21067/24420 [8:47:48<1:23:22,  1.49s/it]
 86%|████████▋ | 21068/24420 [8:47:50<1:23:29,  1.49s/it]
 86%|████████▋ | 21069/24420 [8:47:51<1:23:28,  1.49s/it]
 86%|████████▋ | 21070/24420 [8:47:53<1:23:19,  1.49s/it]
                                                         
{'loss': 0.0798, 'learning_rate': 5e-05, 'epoch': 4.31}

 86%|████████▋ | 21070/24420 [8:47:53<1:23:19,  1.49s/it]
 86%|████████▋ | 21071/24420 [8:47:54<1:23:29,  1.50s/it]
 86%|████████▋ | 21072/24420 [8:47:56<1:23:21,  1.49s/it]
 86%|████████▋ | 21073/24420 [8:47:57<1:23:18,  1.49s/it]
 86%|████████▋ | 21074/24420 [8:47:59<1:23:15,  1.49s/it]
 86%|████████▋ | 21075/24420 [8:48:00<1:23:28,  1.50s/it]
                                                         
{'loss': 0.0802, 'learning_rate': 5e-05, 'epoch': 4.32}

 86%|████████▋ | 21075/24420 [8:48:00<1:23:28,  1.50s/it]
 86%|████████▋ | 21076/24420 [8:48:02<1:23:19,  1.49s/it]
 86%|████████▋ | 21077/24420 [8:48:03<1:23:12,  1.49s/it]
 86%|████████▋ | 21078/24420 [8:48:05<1:23:09,  1.49s/it]
 86%|████████▋ | 21079/24420 [8:48:06<1:23:08,  1.49s/it]
 86%|████████▋ | 21080/24420 [8:48:08<1:22:59,  1.49s/it]
                                                         
{'loss': 0.0611, 'learning_rate': 5e-05, 'epoch': 4.32}

 86%|████████▋ | 21080/24420 [8:48:08<1:22:59,  1.49s/it]
 86%|████████▋ | 21081/24420 [8:48:09<1:23:08,  1.49s/it]
 86%|████████▋ | 21082/24420 [8:48:11<1:23:03,  1.49s/it]
 86%|████████▋ | 21083/24420 [8:48:12<1:23:07,  1.49s/it]
 86%|████████▋ | 21084/24420 [8:48:14<1:23:12,  1.50s/it]
 86%|████████▋ | 21085/24420 [8:48:15<1:23:08,  1.50s/it]
                                                         
{'loss': 0.0618, 'learning_rate': 5e-05, 'epoch': 4.32}

 86%|████████▋ | 21085/24420 [8:48:15<1:23:08,  1.50s/it]
 86%|████████▋ | 21086/24420 [8:48:17<1:23:07,  1.50s/it]
 86%|████████▋ | 21087/24420 [8:48:18<1:23:01,  1.49s/it]
 86%|████████▋ | 21088/24420 [8:48:20<1:23:06,  1.50s/it]
 86%|████████▋ | 21089/24420 [8:48:21<1:23:02,  1.50s/it]
 86%|████████▋ | 21090/24420 [8:48:23<1:23:05,  1.50s/it]
                                                         
{'loss': 0.069, 'learning_rate': 5e-05, 'epoch': 4.32}

 86%|████████▋ | 21090/24420 [8:48:23<1:23:05,  1.50s/it]
 86%|████████▋ | 21091/24420 [8:48:24<1:22:56,  1.49s/it]
 86%|████████▋ | 21092/24420 [8:48:26<1:23:02,  1.50s/it]
 86%|████████▋ | 21093/24420 [8:48:27<1:23:05,  1.50s/it]
 86%|████████▋ | 21094/24420 [8:48:29<1:22:58,  1.50s/it]
 86%|████████▋ | 21095/24420 [8:48:30<1:22:54,  1.50s/it]
                                                         
{'loss': 0.0731, 'learning_rate': 5e-05, 'epoch': 4.32}

 86%|████████▋ | 21095/24420 [8:48:30<1:22:54,  1.50s/it]
 86%|████████▋ | 21096/24420 [8:48:32<1:22:49,  1.50s/it]
 86%|████████▋ | 21097/24420 [8:48:33<1:22:42,  1.49s/it]
 86%|████████▋ | 21098/24420 [8:48:35<1:22:38,  1.49s/it]
 86%|████████▋ | 21099/24420 [8:48:36<1:22:35,  1.49s/it]
 86%|████████▋ | 21100/24420 [8:48:38<1:22:48,  1.50s/it]
                                                         
{'loss': 0.0656, 'learning_rate': 5e-05, 'epoch': 4.32}

 86%|████████▋ | 21100/24420 [8:48:38<1:22:48,  1.50s/it]
 86%|████████▋ | 21101/24420 [8:48:39<1:22:51,  1.50s/it]
 86%|████████▋ | 21102/24420 [8:48:41<1:22:54,  1.50s/it]
 86%|████████▋ | 21103/24420 [8:48:42<1:22:44,  1.50s/it]
 86%|████████▋ | 21104/24420 [8:48:44<1:22:39,  1.50s/it]
 86%|████████▋ | 21105/24420 [8:48:45<1:22:28,  1.49s/it]
                                                         
{'loss': 0.0661, 'learning_rate': 5e-05, 'epoch': 4.32}

 86%|████████▋ | 21105/24420 [8:48:45<1:22:28,  1.49s/it]
 86%|████████▋ | 21106/24420 [8:48:47<1:22:25,  1.49s/it]
 86%|████████▋ | 21107/24420 [8:48:48<1:22:28,  1.49s/it]
 86%|████████▋ | 21108/24420 [8:48:50<1:22:27,  1.49s/it]
 86%|████████▋ | 21109/24420 [8:48:51<1:22:16,  1.49s/it]
 86%|████████▋ | 21110/24420 [8:48:53<1:22:11,  1.49s/it]
                                                         
{'loss': 0.0721, 'learning_rate': 5e-05, 'epoch': 4.32}

 86%|████████▋ | 21110/24420 [8:48:53<1:22:11,  1.49s/it]
 86%|████████▋ | 21111/24420 [8:48:54<1:22:13,  1.49s/it]
 86%|████████▋ | 21112/24420 [8:48:56<1:22:14,  1.49s/it]
 86%|████████▋ | 21113/24420 [8:48:57<1:22:07,  1.49s/it]
 86%|████████▋ | 21114/24420 [8:48:59<1:22:10,  1.49s/it]
 86%|████████▋ | 21115/24420 [8:49:00<1:22:02,  1.49s/it]
                                                         
{'loss': 0.0724, 'learning_rate': 5e-05, 'epoch': 4.32}

 86%|████████▋ | 21115/24420 [8:49:00<1:22:02,  1.49s/it]
 86%|████████▋ | 21116/24420 [8:49:02<1:22:15,  1.49s/it]
 86%|████████▋ | 21117/24420 [8:49:03<1:28:41,  1.61s/it]
 86%|████████▋ | 21118/24420 [8:49:05<1:26:40,  1.57s/it]
 86%|████████▋ | 21119/24420 [8:49:06<1:25:27,  1.55s/it]
 86%|████████▋ | 21120/24420 [8:49:08<1:24:28,  1.54s/it]
                                                         
{'loss': 0.0793, 'learning_rate': 5e-05, 'epoch': 4.32}

 86%|████████▋ | 21120/24420 [8:49:08<1:24:28,  1.54s/it]
 86%|████████▋ | 21121/24420 [8:49:09<1:23:47,  1.52s/it]
 86%|████████▋ | 21122/24420 [8:49:11<1:23:07,  1.51s/it]
 86%|████████▋ | 21123/24420 [8:49:12<1:22:47,  1.51s/it]
 87%|████████▋ | 21124/24420 [8:49:14<1:22:25,  1.50s/it]
 87%|████████▋ | 21125/24420 [8:49:15<1:22:27,  1.50s/it]
                                                         
{'loss': 0.0718, 'learning_rate': 5e-05, 'epoch': 4.33}

 87%|████████▋ | 21125/24420 [8:49:15<1:22:27,  1.50s/it]
 87%|████████▋ | 21126/24420 [8:49:17<1:22:14,  1.50s/it]
 87%|████████▋ | 21127/24420 [8:49:18<1:22:07,  1.50s/it]
 87%|████████▋ | 21128/24420 [8:49:20<1:22:08,  1.50s/it]
 87%|████████▋ | 21129/24420 [8:49:21<1:21:52,  1.49s/it]
 87%|████████▋ | 21130/24420 [8:49:23<1:21:56,  1.49s/it]
                                                         
{'loss': 0.0778, 'learning_rate': 5e-05, 'epoch': 4.33}

 87%|████████▋ | 21130/24420 [8:49:23<1:21:56,  1.49s/it]
 87%|████████▋ | 21131/24420 [8:49:24<1:21:51,  1.49s/it]
 87%|████████▋ | 21132/24420 [8:49:26<1:21:45,  1.49s/it]
 87%|████████▋ | 21133/24420 [8:49:27<1:21:47,  1.49s/it]
 87%|████████▋ | 21134/24420 [8:49:29<1:21:40,  1.49s/it]
 87%|████████▋ | 21135/24420 [8:49:30<1:21:35,  1.49s/it]
                                                         
{'loss': 0.0781, 'learning_rate': 5e-05, 'epoch': 4.33}

 87%|████████▋ | 21135/24420 [8:49:30<1:21:35,  1.49s/it]
 87%|████████▋ | 21136/24420 [8:49:32<1:21:35,  1.49s/it]
 87%|████████▋ | 21137/24420 [8:49:33<1:21:33,  1.49s/it]
 87%|████████▋ | 21138/24420 [8:49:35<1:21:44,  1.49s/it]
 87%|████████▋ | 21139/24420 [8:49:36<1:21:36,  1.49s/it]
 87%|████████▋ | 21140/24420 [8:49:38<1:21:29,  1.49s/it]
                                                         
{'loss': 0.0661, 'learning_rate': 5e-05, 'epoch': 4.33}

 87%|████████▋ | 21140/24420 [8:49:38<1:21:29,  1.49s/it]
 87%|████████▋ | 21141/24420 [8:49:39<1:21:43,  1.50s/it]
 87%|████████▋ | 21142/24420 [8:49:41<1:21:40,  1.50s/it]
 87%|████████▋ | 21143/24420 [8:49:42<1:21:29,  1.49s/it]
 87%|████████▋ | 21144/24420 [8:49:44<1:21:24,  1.49s/it]
 87%|████████▋ | 21145/24420 [8:49:45<1:21:24,  1.49s/it]
                                                         
{'loss': 0.0733, 'learning_rate': 5e-05, 'epoch': 4.33}

 87%|████████▋ | 21145/24420 [8:49:45<1:21:24,  1.49s/it]
 87%|████████▋ | 21146/24420 [8:49:47<1:21:23,  1.49s/it]
 87%|████████▋ | 21147/24420 [8:49:48<1:21:34,  1.50s/it]
 87%|████████▋ | 21148/24420 [8:49:50<1:21:37,  1.50s/it]
 87%|████████▋ | 21149/24420 [8:49:51<1:21:26,  1.49s/it]
 87%|████████▋ | 21150/24420 [8:49:53<1:21:20,  1.49s/it]
                                                         
{'loss': 0.0672, 'learning_rate': 5e-05, 'epoch': 4.33}

 87%|████████▋ | 21150/24420 [8:49:53<1:21:20,  1.49s/it]
 87%|████████▋ | 21151/24420 [8:49:54<1:21:27,  1.50s/it]
 87%|████████▋ | 21152/24420 [8:49:56<1:21:19,  1.49s/it]
 87%|████████▋ | 21153/24420 [8:49:57<1:21:18,  1.49s/it]
 87%|████████▋ | 21154/24420 [8:49:59<1:21:18,  1.49s/it]
 87%|████████▋ | 21155/24420 [8:50:00<1:21:06,  1.49s/it]
                                                         
{'loss': 0.0722, 'learning_rate': 5e-05, 'epoch': 4.33}

 87%|████████▋ | 21155/24420 [8:50:00<1:21:06,  1.49s/it]
 87%|████████▋ | 21156/24420 [8:50:02<1:21:17,  1.49s/it]
 87%|████████▋ | 21157/24420 [8:50:03<1:21:06,  1.49s/it]
 87%|████████▋ | 21158/24420 [8:50:05<1:21:15,  1.49s/it]
 87%|████████▋ | 21159/24420 [8:50:06<1:21:04,  1.49s/it]
 87%|████████▋ | 21160/24420 [8:50:08<1:21:00,  1.49s/it]
                                                         
{'loss': 0.0675, 'learning_rate': 5e-05, 'epoch': 4.33}

 87%|████████▋ | 21160/24420 [8:50:08<1:21:00,  1.49s/it]
 87%|████████▋ | 21161/24420 [8:50:09<1:21:01,  1.49s/it]
 87%|████████▋ | 21162/24420 [8:50:11<1:20:57,  1.49s/it]
 87%|████████▋ | 21163/24420 [8:50:12<1:20:52,  1.49s/it]
 87%|████████▋ | 21164/24420 [8:50:14<1:20:56,  1.49s/it]
 87%|████████▋ | 21165/24420 [8:50:15<1:20:54,  1.49s/it]
                                                         
{'loss': 0.0929, 'learning_rate': 5e-05, 'epoch': 4.33}

 87%|████████▋ | 21165/24420 [8:50:15<1:20:54,  1.49s/it]
 87%|████████▋ | 21166/24420 [8:50:17<1:20:57,  1.49s/it]
 87%|████████▋ | 21167/24420 [8:50:18<1:20:55,  1.49s/it]
 87%|████████▋ | 21168/24420 [8:50:20<1:20:45,  1.49s/it]
 87%|████████▋ | 21169/24420 [8:50:21<1:20:38,  1.49s/it]
 87%|████████▋ | 21170/24420 [8:50:23<1:20:38,  1.49s/it]
                                                         
{'loss': 0.061, 'learning_rate': 5e-05, 'epoch': 4.33}

 87%|████████▋ | 21170/24420 [8:50:23<1:20:38,  1.49s/it]
 87%|████████▋ | 21171/24420 [8:50:24<1:20:37,  1.49s/it]
 87%|████████▋ | 21172/24420 [8:50:25<1:20:34,  1.49s/it]
 87%|████████▋ | 21173/24420 [8:50:27<1:20:35,  1.49s/it]
 87%|████████▋ | 21174/24420 [8:50:28<1:20:37,  1.49s/it]
 87%|████████▋ | 21175/24420 [8:50:30<1:20:42,  1.49s/it]
                                                         
{'loss': 0.0776, 'learning_rate': 5e-05, 'epoch': 4.34}

 87%|████████▋ | 21175/24420 [8:50:30<1:20:42,  1.49s/it]
 87%|████████▋ | 21176/24420 [8:50:31<1:20:29,  1.49s/it]
 87%|████████▋ | 21177/24420 [8:50:33<1:20:30,  1.49s/it]
 87%|████████▋ | 21178/24420 [8:50:34<1:20:31,  1.49s/it]
 87%|████████▋ | 21179/24420 [8:50:36<1:20:21,  1.49s/it]
 87%|████████▋ | 21180/24420 [8:50:37<1:20:22,  1.49s/it]
                                                         
{'loss': 0.0706, 'learning_rate': 5e-05, 'epoch': 4.34}

 87%|████████▋ | 21180/24420 [8:50:37<1:20:22,  1.49s/it]
 87%|████████▋ | 21181/24420 [8:50:39<1:20:23,  1.49s/it]
 87%|████████▋ | 21182/24420 [8:50:40<1:20:27,  1.49s/it]
 87%|████████▋ | 21183/24420 [8:50:42<1:20:16,  1.49s/it]
 87%|████████▋ | 21184/24420 [8:50:43<1:20:10,  1.49s/it]
 87%|████████▋ | 21185/24420 [8:50:45<1:20:12,  1.49s/it]
                                                         
{'loss': 0.0751, 'learning_rate': 5e-05, 'epoch': 4.34}

 87%|████████▋ | 21185/24420 [8:50:45<1:20:12,  1.49s/it]
 87%|████████▋ | 21186/24420 [8:50:46<1:20:16,  1.49s/it]
 87%|████████▋ | 21187/24420 [8:50:48<1:20:15,  1.49s/it]
 87%|████████▋ | 21188/24420 [8:50:49<1:20:11,  1.49s/it]
 87%|████████▋ | 21189/24420 [8:50:51<1:20:04,  1.49s/it]
 87%|████████▋ | 21190/24420 [8:50:52<1:20:06,  1.49s/it]
                                                         
{'loss': 0.0691, 'learning_rate': 5e-05, 'epoch': 4.34}

 87%|████████▋ | 21190/24420 [8:50:52<1:20:06,  1.49s/it]
 87%|████████▋ | 21191/24420 [8:50:54<1:20:07,  1.49s/it]
 87%|████████▋ | 21192/24420 [8:50:55<1:20:04,  1.49s/it]
 87%|████████▋ | 21193/24420 [8:50:57<1:20:08,  1.49s/it]
 87%|████████▋ | 21194/24420 [8:50:58<1:20:07,  1.49s/it]
 87%|████████▋ | 21195/24420 [8:51:00<1:20:09,  1.49s/it]
                                                         
{'loss': 0.07, 'learning_rate': 5e-05, 'epoch': 4.34}

 87%|████████▋ | 21195/24420 [8:51:00<1:20:09,  1.49s/it]
 87%|████████▋ | 21196/24420 [8:51:01<1:20:09,  1.49s/it]
 87%|████████▋ | 21197/24420 [8:51:03<1:20:06,  1.49s/it]
 87%|████████▋ | 21198/24420 [8:51:04<1:20:03,  1.49s/it]
 87%|████████▋ | 21199/24420 [8:51:06<1:19:52,  1.49s/it]
 87%|████████▋ | 21200/24420 [8:51:07<1:19:50,  1.49s/it]
                                                         
{'loss': 0.0809, 'learning_rate': 5e-05, 'epoch': 4.34}

 87%|████████▋ | 21200/24420 [8:51:07<1:19:50,  1.49s/it]
 87%|████████▋ | 21201/24420 [8:51:12<2:10:23,  2.43s/it]
 87%|████████▋ | 21202/24420 [8:51:13<1:55:09,  2.15s/it]
 87%|████████▋ | 21203/24420 [8:51:15<1:44:29,  1.95s/it]
 87%|████████▋ | 21204/24420 [8:51:16<1:36:57,  1.81s/it]
 87%|████████▋ | 21205/24420 [8:51:18<1:31:39,  1.71s/it]
                                                         
{'loss': 0.0636, 'learning_rate': 5e-05, 'epoch': 4.34}

 87%|████████▋ | 21205/24420 [8:51:18<1:31:39,  1.71s/it]
 87%|████████▋ | 21206/24420 [8:51:19<1:28:03,  1.64s/it]
 87%|████████▋ | 21207/24420 [8:51:21<1:25:23,  1.59s/it]
 87%|████████▋ | 21208/24420 [8:51:22<1:23:31,  1.56s/it]
 87%|████████▋ | 21209/24420 [8:51:24<1:22:16,  1.54s/it]
 87%|████████▋ | 21210/24420 [8:51:25<1:21:23,  1.52s/it]
                                                         
{'loss': 0.0706, 'learning_rate': 5e-05, 'epoch': 4.34}

 87%|████████▋ | 21210/24420 [8:51:25<1:21:23,  1.52s/it]
 87%|████████▋ | 21211/24420 [8:51:27<1:20:46,  1.51s/it]
 87%|████████▋ | 21212/24420 [8:51:28<1:20:21,  1.50s/it]
 87%|████████▋ | 21213/24420 [8:51:30<1:20:14,  1.50s/it]
 87%|████████▋ | 21214/24420 [8:51:31<1:20:00,  1.50s/it]
 87%|████████▋ | 21215/24420 [8:51:33<1:19:58,  1.50s/it]
                                                         
{'loss': 0.0758, 'learning_rate': 5e-05, 'epoch': 4.34}

 87%|████████▋ | 21215/24420 [8:51:33<1:19:58,  1.50s/it]
 87%|████████▋ | 21216/24420 [8:51:34<1:19:51,  1.50s/it]
 87%|████████▋ | 21217/24420 [8:51:36<1:19:46,  1.49s/it]
 87%|████████▋ | 21218/24420 [8:51:37<1:19:43,  1.49s/it]
 87%|████████▋ | 21219/24420 [8:51:39<1:19:36,  1.49s/it]
 87%|████████▋ | 21220/24420 [8:51:40<1:19:34,  1.49s/it]
                                                         
{'loss': 0.069, 'learning_rate': 5e-05, 'epoch': 4.34}

 87%|████████▋ | 21220/24420 [8:51:40<1:19:34,  1.49s/it]
 87%|████████▋ | 21221/24420 [8:51:42<1:19:32,  1.49s/it]
 87%|████████▋ | 21222/24420 [8:51:43<1:19:29,  1.49s/it]
 87%|████████▋ | 21223/24420 [8:51:45<1:19:27,  1.49s/it]
 87%|████████▋ | 21224/24420 [8:51:46<1:19:25,  1.49s/it]
 87%|████████▋ | 21225/24420 [8:51:48<1:19:13,  1.49s/it]
                                                         
{'loss': 0.0656, 'learning_rate': 5e-05, 'epoch': 4.35}

 87%|████████▋ | 21225/24420 [8:51:48<1:19:13,  1.49s/it]
 87%|████████▋ | 21226/24420 [8:51:49<1:19:17,  1.49s/it]
 87%|████████▋ | 21227/24420 [8:51:51<1:19:18,  1.49s/it]
 87%|████████▋ | 21228/24420 [8:51:52<1:24:44,  1.59s/it]
 87%|████████▋ | 21229/24420 [8:51:54<1:22:57,  1.56s/it]
 87%|████████▋ | 21230/24420 [8:51:55<1:21:56,  1.54s/it]
                                                         
{'loss': 0.0943, 'learning_rate': 5e-05, 'epoch': 4.35}

 87%|████████▋ | 21230/24420 [8:51:55<1:21:56,  1.54s/it]
 87%|████████▋ | 21231/24420 [8:51:57<1:21:07,  1.53s/it]
 87%|████████▋ | 21232/24420 [8:51:58<1:20:19,  1.51s/it]
 87%|████████▋ | 21233/24420 [8:52:00<1:19:58,  1.51s/it]
 87%|████████▋ | 21234/24420 [8:52:01<1:19:39,  1.50s/it]
 87%|████████▋ | 21235/24420 [8:52:03<1:19:29,  1.50s/it]
                                                         
{'loss': 0.0715, 'learning_rate': 5e-05, 'epoch': 4.35}

 87%|████████▋ | 21235/24420 [8:52:03<1:19:29,  1.50s/it]
 87%|████████▋ | 21236/24420 [8:52:04<1:19:23,  1.50s/it]
 87%|████████▋ | 21237/24420 [8:52:06<1:19:14,  1.49s/it]
 87%|████████▋ | 21238/24420 [8:52:07<1:19:06,  1.49s/it]
 87%|████████▋ | 21239/24420 [8:52:09<1:18:55,  1.49s/it]
 87%|████████▋ | 21240/24420 [8:52:10<1:18:50,  1.49s/it]
                                                         
{'loss': 0.0832, 'learning_rate': 5e-05, 'epoch': 4.35}

 87%|████████▋ | 21240/24420 [8:52:10<1:18:50,  1.49s/it]
 87%|████████▋ | 21241/24420 [8:52:12<1:18:55,  1.49s/it]
 87%|████████▋ | 21242/24420 [8:52:13<1:18:51,  1.49s/it]
 87%|████████▋ | 21243/24420 [8:52:15<1:18:58,  1.49s/it]
 87%|████████▋ | 21244/24420 [8:52:16<1:19:00,  1.49s/it]
 87%|████████▋ | 21245/24420 [8:52:18<1:18:54,  1.49s/it]
                                                         
{'loss': 0.0676, 'learning_rate': 5e-05, 'epoch': 4.35}

 87%|████████▋ | 21245/24420 [8:52:18<1:18:54,  1.49s/it]
 87%|████████▋ | 21246/24420 [8:52:19<1:18:47,  1.49s/it]
 87%|████████▋ | 21247/24420 [8:52:21<1:18:49,  1.49s/it]
 87%|████████▋ | 21248/24420 [8:52:22<1:18:49,  1.49s/it]
 87%|████████▋ | 21249/24420 [8:52:24<1:18:52,  1.49s/it]
 87%|████████▋ | 21250/24420 [8:52:25<1:18:45,  1.49s/it]
                                                         
{'loss': 0.0716, 'learning_rate': 5e-05, 'epoch': 4.35}

 87%|████████▋ | 21250/24420 [8:52:25<1:18:45,  1.49s/it]
 87%|████████▋ | 21251/24420 [8:52:27<1:18:39,  1.49s/it]
 87%|████████▋ | 21252/24420 [8:52:28<1:18:47,  1.49s/it]
 87%|████████▋ | 21253/24420 [8:52:30<1:18:47,  1.49s/it]
 87%|████████▋ | 21254/24420 [8:52:31<1:18:43,  1.49s/it]
 87%|████████▋ | 21255/24420 [8:52:33<1:18:39,  1.49s/it]
                                                         
{'loss': 0.0681, 'learning_rate': 5e-05, 'epoch': 4.35}

 87%|████████▋ | 21255/24420 [8:52:33<1:18:39,  1.49s/it]
 87%|████████▋ | 21256/24420 [8:52:34<1:18:48,  1.49s/it]
 87%|████████▋ | 21257/24420 [8:52:36<1:18:38,  1.49s/it]
 87%|████████▋ | 21258/24420 [8:52:37<1:18:37,  1.49s/it]
 87%|████████▋ | 21259/24420 [8:52:39<1:18:31,  1.49s/it]
 87%|████████▋ | 21260/24420 [8:52:40<1:18:35,  1.49s/it]
                                                         
{'loss': 0.076, 'learning_rate': 5e-05, 'epoch': 4.35}

 87%|████████▋ | 21260/24420 [8:52:40<1:18:35,  1.49s/it]
 87%|████████▋ | 21261/24420 [8:52:42<1:18:35,  1.49s/it]
 87%|████████▋ | 21262/24420 [8:52:43<1:18:35,  1.49s/it]
 87%|████████▋ | 21263/24420 [8:52:45<1:18:23,  1.49s/it]
 87%|████████▋ | 21264/24420 [8:52:46<1:18:18,  1.49s/it]
 87%|████████▋ | 21265/24420 [8:52:47<1:18:23,  1.49s/it]
                                                         
{'loss': 0.0773, 'learning_rate': 5e-05, 'epoch': 4.35}

 87%|████████▋ | 21265/24420 [8:52:47<1:18:23,  1.49s/it]
 87%|████████▋ | 21266/24420 [8:52:49<1:18:21,  1.49s/it]
 87%|████████▋ | 21267/24420 [8:52:50<1:18:12,  1.49s/it]
 87%|████████▋ | 21268/24420 [8:52:52<1:18:15,  1.49s/it]
 87%|████████▋ | 21269/24420 [8:52:53<1:18:14,  1.49s/it]
 87%|████████▋ | 21270/24420 [8:52:55<1:18:17,  1.49s/it]
                                                         
{'loss': 0.0794, 'learning_rate': 5e-05, 'epoch': 4.36}

 87%|████████▋ | 21270/24420 [8:52:55<1:18:17,  1.49s/it]
 87%|████████▋ | 21271/24420 [8:52:56<1:18:10,  1.49s/it]
 87%|████████▋ | 21272/24420 [8:52:58<1:18:11,  1.49s/it]
 87%|████████▋ | 21273/24420 [8:52:59<1:18:18,  1.49s/it]
 87%|████████▋ | 21274/24420 [8:53:01<1:18:16,  1.49s/it]
 87%|████████▋ | 21275/24420 [8:53:02<1:18:12,  1.49s/it]
                                                         
{'loss': 0.0845, 'learning_rate': 5e-05, 'epoch': 4.36}

 87%|████████▋ | 21275/24420 [8:53:02<1:18:12,  1.49s/it]
 87%|████████▋ | 21276/24420 [8:53:04<1:18:19,  1.49s/it]
 87%|████████▋ | 21277/24420 [8:53:05<1:18:29,  1.50s/it]
 87%|████████▋ | 21278/24420 [8:53:07<1:18:17,  1.49s/it]
 87%|████████▋ | 21279/24420 [8:53:08<1:18:07,  1.49s/it]
 87%|████████▋ | 21280/24420 [8:53:10<1:18:03,  1.49s/it]
                                                         
{'loss': 0.065, 'learning_rate': 5e-05, 'epoch': 4.36}

 87%|████████▋ | 21280/24420 [8:53:10<1:18:03,  1.49s/it]
 87%|████████▋ | 21281/24420 [8:53:11<1:17:57,  1.49s/it]
 87%|████████▋ | 21282/24420 [8:53:13<1:17:58,  1.49s/it]
 87%|████████▋ | 21283/24420 [8:53:14<1:18:11,  1.50s/it]
 87%|████████▋ | 21284/24420 [8:53:16<1:18:00,  1.49s/it]
 87%|████████▋ | 21285/24420 [8:53:17<1:18:03,  1.49s/it]
                                                         
{'loss': 0.0868, 'learning_rate': 5e-05, 'epoch': 4.36}

 87%|████████▋ | 21285/24420 [8:53:17<1:18:03,  1.49s/it]
 87%|████████▋ | 21286/24420 [8:53:19<1:18:00,  1.49s/it]
 87%|████████▋ | 21287/24420 [8:53:20<1:18:04,  1.50s/it]
 87%|████████▋ | 21288/24420 [8:53:22<1:18:01,  1.49s/it]
 87%|████████▋ | 21289/24420 [8:53:23<1:17:53,  1.49s/it]
 87%|████████▋ | 21290/24420 [8:53:25<1:17:46,  1.49s/it]
                                                         
{'loss': 0.0829, 'learning_rate': 5e-05, 'epoch': 4.36}

 87%|████████▋ | 21290/24420 [8:53:25<1:17:46,  1.49s/it]
 87%|████████▋ | 21291/24420 [8:53:26<1:17:53,  1.49s/it]
 87%|████████▋ | 21292/24420 [8:53:28<1:17:56,  1.50s/it]
 87%|████████▋ | 21293/24420 [8:53:29<1:17:52,  1.49s/it]
 87%|████████▋ | 21294/24420 [8:53:31<1:17:48,  1.49s/it]
 87%|████████▋ | 21295/24420 [8:53:32<1:17:39,  1.49s/it]
                                                         
{'loss': 0.0813, 'learning_rate': 5e-05, 'epoch': 4.36}

 87%|████████▋ | 21295/24420 [8:53:32<1:17:39,  1.49s/it]
 87%|████████▋ | 21296/24420 [8:53:34<1:17:37,  1.49s/it]
 87%|████████▋ | 21297/24420 [8:53:35<1:17:39,  1.49s/it]
 87%|████████▋ | 21298/24420 [8:53:37<1:17:40,  1.49s/it]
 87%|████████▋ | 21299/24420 [8:53:38<1:17:29,  1.49s/it]
 87%|████████▋ | 21300/24420 [8:53:40<1:17:33,  1.49s/it]
                                                         
{'loss': 0.0792, 'learning_rate': 5e-05, 'epoch': 4.36}

 87%|████████▋ | 21300/24420 [8:53:40<1:17:33,  1.49s/it]
 87%|████████▋ | 21301/24420 [8:53:41<1:17:24,  1.49s/it]
 87%|████████▋ | 21302/24420 [8:53:43<1:17:23,  1.49s/it]
 87%|████████▋ | 21303/24420 [8:53:44<1:17:21,  1.49s/it]
 87%|████████▋ | 21304/24420 [8:53:46<1:17:18,  1.49s/it]
 87%|████████▋ | 21305/24420 [8:53:47<1:17:14,  1.49s/it]
                                                         
{'loss': 0.0577, 'learning_rate': 5e-05, 'epoch': 4.36}

 87%|████████▋ | 21305/24420 [8:53:47<1:17:14,  1.49s/it]
 87%|████████▋ | 21306/24420 [8:53:49<1:17:07,  1.49s/it]
 87%|████████▋ | 21307/24420 [8:53:50<1:18:28,  1.51s/it]
 87%|████████▋ | 21308/24420 [8:53:52<1:18:04,  1.51s/it]
 87%|████████▋ | 21309/24420 [8:53:53<1:17:44,  1.50s/it]
 87%|████████▋ | 21310/24420 [8:53:55<1:17:31,  1.50s/it]
                                                         
{'loss': 0.0728, 'learning_rate': 5e-05, 'epoch': 4.36}

 87%|████████▋ | 21310/24420 [8:53:55<1:17:31,  1.50s/it]
 87%|████████▋ | 21311/24420 [8:53:56<1:17:23,  1.49s/it]
 87%|████████▋ | 21312/24420 [8:53:58<1:17:12,  1.49s/it]
 87%|████████▋ | 21313/24420 [8:53:59<1:17:09,  1.49s/it]
 87%|████████▋ | 21314/24420 [8:54:01<1:17:05,  1.49s/it]
 87%|████████▋ | 21315/24420 [8:54:02<1:17:08,  1.49s/it]
                                                         
{'loss': 0.0841, 'learning_rate': 5e-05, 'epoch': 4.36}

 87%|████████▋ | 21315/24420 [8:54:02<1:17:08,  1.49s/it]
 87%|████████▋ | 21316/24420 [8:54:04<1:17:13,  1.49s/it]
 87%|████████▋ | 21317/24420 [8:54:05<1:17:04,  1.49s/it]
 87%|████████▋ | 21318/24420 [8:54:07<1:17:02,  1.49s/it]
 87%|████████▋ | 21319/24420 [8:54:08<1:17:02,  1.49s/it]
 87%|████████▋ | 21320/24420 [8:54:10<1:17:09,  1.49s/it]
                                                         
{'loss': 0.0726, 'learning_rate': 5e-05, 'epoch': 4.37}

 87%|████████▋ | 21320/24420 [8:54:10<1:17:09,  1.49s/it]
 87%|████████▋ | 21321/24420 [8:54:11<1:17:06,  1.49s/it]
 87%|████████▋ | 21322/24420 [8:54:13<1:17:19,  1.50s/it]
 87%|████████▋ | 21323/24420 [8:54:14<1:17:08,  1.49s/it]
 87%|████████▋ | 21324/24420 [8:54:16<1:16:59,  1.49s/it]
 87%|████████▋ | 21325/24420 [8:54:17<1:16:49,  1.49s/it]
                                                         
{'loss': 0.0799, 'learning_rate': 5e-05, 'epoch': 4.37}

 87%|████████▋ | 21325/24420 [8:54:17<1:16:49,  1.49s/it]
 87%|████████▋ | 21326/24420 [8:54:19<1:16:48,  1.49s/it]
 87%|████████▋ | 21327/24420 [8:54:20<1:16:45,  1.49s/it]
 87%|████████▋ | 21328/24420 [8:54:22<1:16:43,  1.49s/it]
 87%|████████▋ | 21329/24420 [8:54:23<1:16:42,  1.49s/it]
 87%|████████▋ | 21330/24420 [8:54:24<1:16:34,  1.49s/it]
                                                         
{'loss': 0.0712, 'learning_rate': 5e-05, 'epoch': 4.37}

 87%|████████▋ | 21330/24420 [8:54:24<1:16:34,  1.49s/it]
 87%|████████▋ | 21331/24420 [8:54:26<1:16:38,  1.49s/it]
 87%|████████▋ | 21332/24420 [8:54:27<1:16:27,  1.49s/it]
 87%|████████▋ | 21333/24420 [8:54:29<1:16:35,  1.49s/it]
 87%|████████▋ | 21334/24420 [8:54:30<1:16:30,  1.49s/it]
 87%|████████▋ | 21335/24420 [8:54:32<1:16:27,  1.49s/it]
                                                         
{'loss': 0.0739, 'learning_rate': 5e-05, 'epoch': 4.37}

 87%|████████▋ | 21335/24420 [8:54:32<1:16:27,  1.49s/it]
 87%|████████▋ | 21336/24420 [8:54:33<1:16:27,  1.49s/it]
 87%|████████▋ | 21337/24420 [8:54:35<1:16:32,  1.49s/it]
 87%|████████▋ | 21338/24420 [8:54:36<1:16:27,  1.49s/it]
 87%|████████▋ | 21339/24420 [8:54:38<1:16:32,  1.49s/it]
 87%|████████▋ | 21340/24420 [8:54:39<1:16:31,  1.49s/it]
                                                         
{'loss': 0.0734, 'learning_rate': 5e-05, 'epoch': 4.37}

 87%|████████▋ | 21340/24420 [8:54:39<1:16:31,  1.49s/it]
 87%|████████▋ | 21341/24420 [8:54:41<1:16:32,  1.49s/it]
 87%|████████▋ | 21342/24420 [8:54:42<1:16:22,  1.49s/it]
 87%|████████▋ | 21343/24420 [8:54:44<1:16:22,  1.49s/it]
 87%|████████▋ | 21344/24420 [8:54:45<1:16:35,  1.49s/it]
 87%|████████▋ | 21345/24420 [8:54:47<1:16:36,  1.49s/it]
                                                         
{'loss': 0.1013, 'learning_rate': 5e-05, 'epoch': 4.37}

 87%|████████▋ | 21345/24420 [8:54:47<1:16:36,  1.49s/it]
 87%|████████▋ | 21346/24420 [8:54:48<1:16:29,  1.49s/it]
 87%|████████▋ | 21347/24420 [8:54:50<1:16:29,  1.49s/it]
 87%|████████▋ | 21348/24420 [8:54:51<1:16:25,  1.49s/it]
 87%|████████▋ | 21349/24420 [8:54:53<1:16:16,  1.49s/it]
 87%|████████▋ | 21350/24420 [8:54:54<1:16:19,  1.49s/it]
                                                         
{'loss': 0.0708, 'learning_rate': 5e-05, 'epoch': 4.37}

 87%|████████▋ | 21350/24420 [8:54:54<1:16:19,  1.49s/it]
 87%|████████▋ | 21351/24420 [8:54:56<1:16:23,  1.49s/it]
 87%|████████▋ | 21352/24420 [8:54:57<1:16:37,  1.50s/it]
 87%|████████▋ | 21353/24420 [8:54:59<1:16:27,  1.50s/it]
 87%|████████▋ | 21354/24420 [8:55:00<1:16:14,  1.49s/it]
 87%|████████▋ | 21355/24420 [8:55:02<1:16:20,  1.49s/it]
                                                         
{'loss': 0.0665, 'learning_rate': 5e-05, 'epoch': 4.37}

 87%|████████▋ | 21355/24420 [8:55:02<1:16:20,  1.49s/it]
 87%|████████▋ | 21356/24420 [8:55:03<1:16:29,  1.50s/it]
 87%|████████▋ | 21357/24420 [8:55:05<1:16:13,  1.49s/it]
 87%|████████▋ | 21358/24420 [8:55:06<1:16:06,  1.49s/it]
 87%|████████▋ | 21359/24420 [8:55:08<1:16:11,  1.49s/it]
 87%|████████▋ | 21360/24420 [8:55:09<1:16:08,  1.49s/it]
                                                         
{'loss': 0.0865, 'learning_rate': 5e-05, 'epoch': 4.37}

 87%|████████▋ | 21360/24420 [8:55:09<1:16:08,  1.49s/it]
 87%|████████▋ | 21361/24420 [8:55:11<1:16:01,  1.49s/it]
 87%|████████▋ | 21362/24420 [8:55:12<1:16:05,  1.49s/it]
 87%|████████▋ | 21363/24420 [8:55:14<1:15:53,  1.49s/it]
 87%|████████▋ | 21364/24420 [8:55:15<1:15:47,  1.49s/it]
 87%|████████▋ | 21365/24420 [8:55:17<1:15:49,  1.49s/it]
                                                         
{'loss': 0.0683, 'learning_rate': 5e-05, 'epoch': 4.37}

 87%|████████▋ | 21365/24420 [8:55:17<1:15:49,  1.49s/it]
 87%|████████▋ | 21366/24420 [8:55:18<1:15:44,  1.49s/it]
 87%|████████▋ | 21367/24420 [8:55:20<1:15:45,  1.49s/it]
 88%|████████▊ | 21368/24420 [8:55:21<1:15:45,  1.49s/it]
 88%|████████▊ | 21369/24420 [8:55:23<1:15:38,  1.49s/it]
 88%|████████▊ | 21370/24420 [8:55:24<1:15:44,  1.49s/it]
                                                         
{'loss': 0.0715, 'learning_rate': 5e-05, 'epoch': 4.38}

 88%|████████▊ | 21370/24420 [8:55:24<1:15:44,  1.49s/it]
 88%|████████▊ | 21371/24420 [8:55:26<1:15:40,  1.49s/it]
 88%|████████▊ | 21372/24420 [8:55:27<1:15:41,  1.49s/it]
 88%|████████▊ | 21373/24420 [8:55:29<1:15:37,  1.49s/it]
 88%|████████▊ | 21374/24420 [8:55:30<1:15:40,  1.49s/it]
 88%|████████▊ | 21375/24420 [8:55:32<1:15:36,  1.49s/it]
                                                         
{'loss': 0.073, 'learning_rate': 5e-05, 'epoch': 4.38}

 88%|████████▊ | 21375/24420 [8:55:32<1:15:36,  1.49s/it]
 88%|████████▊ | 21376/24420 [8:55:33<1:15:44,  1.49s/it]
 88%|████████▊ | 21377/24420 [8:55:35<1:15:39,  1.49s/it]
 88%|████████▊ | 21378/24420 [8:55:36<1:15:41,  1.49s/it]
 88%|████████▊ | 21379/24420 [8:55:38<1:15:32,  1.49s/it]
 88%|████████▊ | 21380/24420 [8:55:39<1:15:33,  1.49s/it]
                                                         
{'loss': 0.0808, 'learning_rate': 5e-05, 'epoch': 4.38}

 88%|████████▊ | 21380/24420 [8:55:39<1:15:33,  1.49s/it]
 88%|████████▊ | 21381/24420 [8:55:41<1:15:29,  1.49s/it]
 88%|████████▊ | 21382/24420 [8:55:42<1:15:25,  1.49s/it]
 88%|████████▊ | 21383/24420 [8:55:44<1:15:19,  1.49s/it]
 88%|████████▊ | 21384/24420 [8:55:45<1:15:24,  1.49s/it]
 88%|████████▊ | 21385/24420 [8:55:46<1:15:31,  1.49s/it]
                                                         
{'loss': 0.067, 'learning_rate': 5e-05, 'epoch': 4.38}

 88%|████████▊ | 21385/24420 [8:55:47<1:15:31,  1.49s/it]
 88%|████████▊ | 21386/24420 [8:55:48<1:15:29,  1.49s/it]
 88%|████████▊ | 21387/24420 [8:55:49<1:15:19,  1.49s/it]
 88%|████████▊ | 21388/24420 [8:55:51<1:15:14,  1.49s/it]
 88%|████████▊ | 21389/24420 [8:55:52<1:15:17,  1.49s/it]
 88%|████████▊ | 21390/24420 [8:55:54<1:15:15,  1.49s/it]
                                                         
{'loss': 0.0746, 'learning_rate': 5e-05, 'epoch': 4.38}

 88%|████████▊ | 21390/24420 [8:55:54<1:15:15,  1.49s/it]
 88%|████████▊ | 21391/24420 [8:55:55<1:15:13,  1.49s/it]
 88%|████████▊ | 21392/24420 [8:55:57<1:15:05,  1.49s/it]
 88%|████████▊ | 21393/24420 [8:55:58<1:15:00,  1.49s/it]
 88%|████████▊ | 21394/24420 [8:56:00<1:15:02,  1.49s/it]
 88%|████████▊ | 21395/24420 [8:56:01<1:15:00,  1.49s/it]
                                                         
{'loss': 0.0792, 'learning_rate': 5e-05, 'epoch': 4.38}

 88%|████████▊ | 21395/24420 [8:56:01<1:15:00,  1.49s/it]
 88%|████████▊ | 21396/24420 [8:56:03<1:15:14,  1.49s/it]
 88%|████████▊ | 21397/24420 [8:56:04<1:15:03,  1.49s/it]
 88%|████████▊ | 21398/24420 [8:56:06<1:15:06,  1.49s/it]
 88%|████████▊ | 21399/24420 [8:56:07<1:15:05,  1.49s/it]
 88%|████████▊ | 21400/24420 [8:56:09<1:14:57,  1.49s/it]
                                                         
{'loss': 0.0752, 'learning_rate': 5e-05, 'epoch': 4.38}

 88%|████████▊ | 21400/24420 [8:56:09<1:14:57,  1.49s/it]
 88%|████████▊ | 21401/24420 [8:56:13<2:01:36,  2.42s/it]
 88%|████████▊ | 21402/24420 [8:56:15<1:47:25,  2.14s/it]
 88%|████████▊ | 21403/24420 [8:56:16<1:37:25,  1.94s/it]
 88%|████████▊ | 21404/24420 [8:56:18<1:30:36,  1.80s/it]
 88%|████████▊ | 21405/24420 [8:56:19<1:25:54,  1.71s/it]
                                                         
{'loss': 0.0678, 'learning_rate': 5e-05, 'epoch': 4.38}

 88%|████████▊ | 21405/24420 [8:56:19<1:25:54,  1.71s/it]
 88%|████████▊ | 21406/24420 [8:56:21<1:22:37,  1.64s/it]
 88%|████████▊ | 21407/24420 [8:56:22<1:20:13,  1.60s/it]
 88%|████████▊ | 21408/24420 [8:56:24<1:18:35,  1.57s/it]
 88%|████████▊ | 21409/24420 [8:56:25<1:17:20,  1.54s/it]
 88%|████████▊ | 21410/24420 [8:56:27<1:16:34,  1.53s/it]
                                                         
{'loss': 0.071, 'learning_rate': 5e-05, 'epoch': 4.38}

 88%|████████▊ | 21410/24420 [8:56:27<1:16:34,  1.53s/it]
 88%|████████▊ | 21411/24420 [8:56:28<1:15:56,  1.51s/it]
 88%|████████▊ | 21412/24420 [8:56:30<1:15:41,  1.51s/it]
 88%|████████▊ | 21413/24420 [8:56:31<1:15:20,  1.50s/it]
 88%|████████▊ | 21414/24420 [8:56:33<1:15:05,  1.50s/it]
 88%|████████▊ | 21415/24420 [8:56:34<1:14:55,  1.50s/it]
                                                         
{'loss': 0.0855, 'learning_rate': 5e-05, 'epoch': 4.38}

 88%|████████▊ | 21415/24420 [8:56:34<1:14:55,  1.50s/it]
 88%|████████▊ | 21416/24420 [8:56:36<1:14:50,  1.49s/it]
 88%|████████▊ | 21417/24420 [8:56:37<1:14:40,  1.49s/it]
 88%|████████▊ | 21418/24420 [8:56:39<1:14:43,  1.49s/it]
 88%|████████▊ | 21419/24420 [8:56:40<1:14:34,  1.49s/it]
 88%|████████▊ | 21420/24420 [8:56:42<1:14:39,  1.49s/it]
                                                         
{'loss': 0.0763, 'learning_rate': 5e-05, 'epoch': 4.39}

 88%|████████▊ | 21420/24420 [8:56:42<1:14:39,  1.49s/it]
 88%|████████▊ | 21421/24420 [8:56:43<1:14:28,  1.49s/it]
 88%|████████▊ | 21422/24420 [8:56:45<1:14:23,  1.49s/it]
 88%|████████▊ | 21423/24420 [8:56:46<1:14:19,  1.49s/it]
 88%|████████▊ | 21424/24420 [8:56:48<1:14:25,  1.49s/it]
 88%|████████▊ | 21425/24420 [8:56:49<1:14:23,  1.49s/it]
                                                         
{'loss': 0.0691, 'learning_rate': 5e-05, 'epoch': 4.39}

 88%|████████▊ | 21425/24420 [8:56:49<1:14:23,  1.49s/it]
 88%|████████▊ | 21426/24420 [8:56:51<1:14:27,  1.49s/it]
 88%|████████▊ | 21427/24420 [8:56:52<1:14:23,  1.49s/it]
 88%|████████▊ | 21428/24420 [8:56:54<1:14:27,  1.49s/it]
 88%|████████▊ | 21429/24420 [8:56:55<1:14:21,  1.49s/it]
 88%|████████▊ | 21430/24420 [8:56:57<1:14:08,  1.49s/it]
                                                         
{'loss': 0.0715, 'learning_rate': 5e-05, 'epoch': 4.39}

 88%|████████▊ | 21430/24420 [8:56:57<1:14:08,  1.49s/it]
 88%|████████▊ | 21431/24420 [8:56:58<1:14:11,  1.49s/it]
 88%|████████▊ | 21432/24420 [8:57:00<1:14:07,  1.49s/it]
 88%|████████▊ | 21433/24420 [8:57:01<1:14:24,  1.49s/it]
 88%|████████▊ | 21434/24420 [8:57:03<1:14:19,  1.49s/it]
 88%|████████▊ | 21435/24420 [8:57:04<1:14:15,  1.49s/it]
                                                         
{'loss': 0.0716, 'learning_rate': 5e-05, 'epoch': 4.39}

 88%|████████▊ | 21435/24420 [8:57:04<1:14:15,  1.49s/it]
 88%|████████▊ | 21436/24420 [8:57:06<1:14:12,  1.49s/it]
 88%|████████▊ | 21437/24420 [8:57:07<1:14:26,  1.50s/it]
 88%|████████▊ | 21438/24420 [8:57:09<1:14:16,  1.49s/it]
 88%|████████▊ | 21439/24420 [8:57:10<1:14:18,  1.50s/it]
 88%|████████▊ | 21440/24420 [8:57:12<1:14:05,  1.49s/it]
                                                         
{'loss': 0.0739, 'learning_rate': 5e-05, 'epoch': 4.39}

 88%|████████▊ | 21440/24420 [8:57:12<1:14:05,  1.49s/it]
 88%|████████▊ | 21441/24420 [8:57:13<1:14:04,  1.49s/it]
 88%|████████▊ | 21442/24420 [8:57:15<1:14:00,  1.49s/it]
 88%|████████▊ | 21443/24420 [8:57:16<1:13:56,  1.49s/it]
 88%|████████▊ | 21444/24420 [8:57:17<1:13:48,  1.49s/it]
 88%|████████▊ | 21445/24420 [8:57:19<1:13:51,  1.49s/it]
                                                         
{'loss': 0.07, 'learning_rate': 5e-05, 'epoch': 4.39}

 88%|████████▊ | 21445/24420 [8:57:19<1:13:51,  1.49s/it]
 88%|████████▊ | 21446/24420 [8:57:20<1:13:53,  1.49s/it]
 88%|████████▊ | 21447/24420 [8:57:22<1:13:42,  1.49s/it]
 88%|████████▊ | 21448/24420 [8:57:23<1:13:42,  1.49s/it]
 88%|████████▊ | 21449/24420 [8:57:25<1:13:44,  1.49s/it]
 88%|████████▊ | 21450/24420 [8:57:26<1:13:45,  1.49s/it]
                                                         
{'loss': 0.0748, 'learning_rate': 5e-05, 'epoch': 4.39}

 88%|████████▊ | 21450/24420 [8:57:26<1:13:45,  1.49s/it]
 88%|████████▊ | 21451/24420 [8:57:28<1:13:43,  1.49s/it]
 88%|████████▊ | 21452/24420 [8:57:29<1:13:54,  1.49s/it]
 88%|████████▊ | 21453/24420 [8:57:31<1:13:44,  1.49s/it]
 88%|████████▊ | 21454/24420 [8:57:32<1:13:38,  1.49s/it]
 88%|████████▊ | 21455/24420 [8:57:34<1:13:34,  1.49s/it]
                                                         
{'loss': 0.0651, 'learning_rate': 5e-05, 'epoch': 4.39}

 88%|████████▊ | 21455/24420 [8:57:34<1:13:34,  1.49s/it]
 88%|████████▊ | 21456/24420 [8:57:35<1:13:32,  1.49s/it]
 88%|████████▊ | 21457/24420 [8:57:37<1:13:28,  1.49s/it]
 88%|████████▊ | 21458/24420 [8:57:38<1:13:39,  1.49s/it]
 88%|████████▊ | 21459/24420 [8:57:40<1:13:38,  1.49s/it]
 88%|████████▊ | 21460/24420 [8:57:41<1:13:37,  1.49s/it]
                                                         
{'loss': 0.0746, 'learning_rate': 5e-05, 'epoch': 4.39}

 88%|████████▊ | 21460/24420 [8:57:41<1:13:37,  1.49s/it]
 88%|████████▊ | 21461/24420 [8:57:43<1:13:37,  1.49s/it]
 88%|████████▊ | 21462/24420 [8:57:44<1:13:35,  1.49s/it]
 88%|████████▊ | 21463/24420 [8:57:46<1:13:32,  1.49s/it]
 88%|████████▊ | 21464/24420 [8:57:47<1:13:41,  1.50s/it]
 88%|████████▊ | 21465/24420 [8:57:49<1:13:33,  1.49s/it]
                                                         
{'loss': 0.0911, 'learning_rate': 5e-05, 'epoch': 4.39}

 88%|████████▊ | 21465/24420 [8:57:49<1:13:33,  1.49s/it]
 88%|████████▊ | 21466/24420 [8:57:50<1:13:24,  1.49s/it]
 88%|████████▊ | 21467/24420 [8:57:52<1:13:19,  1.49s/it]
 88%|████████▊ | 21468/24420 [8:57:53<1:13:16,  1.49s/it]
 88%|████████▊ | 21469/24420 [8:57:55<1:13:19,  1.49s/it]
 88%|████████▊ | 21470/24420 [8:57:56<1:13:10,  1.49s/it]
                                                         
{'loss': 0.0798, 'learning_rate': 5e-05, 'epoch': 4.4}

 88%|████████▊ | 21470/24420 [8:57:56<1:13:10,  1.49s/it]
 88%|████████▊ | 21471/24420 [8:57:58<1:13:20,  1.49s/it]
 88%|████████▊ | 21472/24420 [8:57:59<1:13:19,  1.49s/it]
 88%|████████▊ | 21473/24420 [8:58:01<1:13:25,  1.50s/it]
 88%|████████▊ | 21474/24420 [8:58:02<1:13:13,  1.49s/it]
 88%|████████▊ | 21475/24420 [8:58:04<1:13:19,  1.49s/it]
                                                         
{'loss': 0.0856, 'learning_rate': 5e-05, 'epoch': 4.4}

 88%|████████▊ | 21475/24420 [8:58:04<1:13:19,  1.49s/it]
 88%|████████▊ | 21476/24420 [8:58:05<1:13:17,  1.49s/it]
 88%|████████▊ | 21477/24420 [8:58:07<1:13:10,  1.49s/it]
 88%|████████▊ | 21478/24420 [8:58:08<1:13:05,  1.49s/it]
 88%|████████▊ | 21479/24420 [8:58:10<1:13:06,  1.49s/it]
 88%|████████▊ | 21480/24420 [8:58:11<1:13:07,  1.49s/it]
                                                         
{'loss': 0.0781, 'learning_rate': 5e-05, 'epoch': 4.4}

 88%|████████▊ | 21480/24420 [8:58:11<1:13:07,  1.49s/it]
 88%|████████▊ | 21481/24420 [8:58:13<1:13:11,  1.49s/it]
 88%|████████▊ | 21482/24420 [8:58:14<1:13:15,  1.50s/it]
 88%|████████▊ | 21483/24420 [8:58:16<1:13:12,  1.50s/it]
 88%|████████▊ | 21484/24420 [8:58:17<1:13:07,  1.49s/it]
 88%|████████▊ | 21485/24420 [8:58:19<1:12:58,  1.49s/it]
                                                         
{'loss': 0.0636, 'learning_rate': 5e-05, 'epoch': 4.4}

 88%|████████▊ | 21485/24420 [8:58:19<1:12:58,  1.49s/it]
 88%|████████▊ | 21486/24420 [8:58:20<1:12:59,  1.49s/it]
 88%|████████▊ | 21487/24420 [8:58:22<1:12:55,  1.49s/it]
 88%|████████▊ | 21488/24420 [8:58:23<1:12:50,  1.49s/it]
 88%|████████▊ | 21489/24420 [8:58:25<1:12:46,  1.49s/it]
 88%|████████▊ | 21490/24420 [8:58:26<1:12:35,  1.49s/it]
                                                         
{'loss': 0.0831, 'learning_rate': 5e-05, 'epoch': 4.4}

 88%|████████▊ | 21490/24420 [8:58:26<1:12:35,  1.49s/it]
 88%|████████▊ | 21491/24420 [8:58:28<1:12:34,  1.49s/it]
 88%|████████▊ | 21492/24420 [8:58:29<1:12:38,  1.49s/it]
 88%|████████▊ | 21493/24420 [8:58:31<1:12:37,  1.49s/it]
 88%|████████▊ | 21494/24420 [8:58:32<1:12:32,  1.49s/it]
 88%|████████▊ | 21495/24420 [8:58:34<1:12:24,  1.49s/it]
                                                         
{'loss': 0.0925, 'learning_rate': 5e-05, 'epoch': 4.4}

 88%|████████▊ | 21495/24420 [8:58:34<1:12:24,  1.49s/it]
 88%|████████▊ | 21496/24420 [8:58:35<1:12:28,  1.49s/it]
 88%|████████▊ | 21497/24420 [8:58:37<1:12:37,  1.49s/it]
 88%|████████▊ | 21498/24420 [8:58:38<1:12:35,  1.49s/it]
 88%|████████▊ | 21499/24420 [8:58:40<1:12:37,  1.49s/it]
 88%|████████▊ | 21500/24420 [8:58:41<1:12:32,  1.49s/it]
                                                         
{'loss': 0.0705, 'learning_rate': 5e-05, 'epoch': 4.4}

 88%|████████▊ | 21500/24420 [8:58:41<1:12:32,  1.49s/it]
 88%|████████▊ | 21501/24420 [8:58:42<1:12:43,  1.49s/it]
 88%|████████▊ | 21502/24420 [8:58:44<1:12:41,  1.49s/it]
 88%|████████▊ | 21503/24420 [8:58:45<1:12:32,  1.49s/it]
 88%|████████▊ | 21504/24420 [8:58:47<1:12:34,  1.49s/it]
 88%|████████▊ | 21505/24420 [8:58:48<1:12:26,  1.49s/it]
                                                         
{'loss': 0.0755, 'learning_rate': 5e-05, 'epoch': 4.4}

 88%|████████▊ | 21505/24420 [8:58:48<1:12:26,  1.49s/it]
 88%|████████▊ | 21506/24420 [8:58:50<1:12:26,  1.49s/it]
 88%|████████▊ | 21507/24420 [8:58:51<1:12:24,  1.49s/it]
 88%|████████▊ | 21508/24420 [8:58:53<1:12:28,  1.49s/it]
 88%|████████▊ | 21509/24420 [8:58:54<1:12:29,  1.49s/it]
 88%|████████▊ | 21510/24420 [8:58:56<1:12:17,  1.49s/it]
                                                         
{'loss': 0.0865, 'learning_rate': 5e-05, 'epoch': 4.4}

 88%|████████▊ | 21510/24420 [8:58:56<1:12:17,  1.49s/it]
 88%|████████▊ | 21511/24420 [8:58:57<1:12:22,  1.49s/it]
 88%|████████▊ | 21512/24420 [8:58:59<1:12:19,  1.49s/it]
 88%|████████▊ | 21513/24420 [8:59:00<1:12:14,  1.49s/it]
 88%|████████▊ | 21514/24420 [8:59:02<1:12:12,  1.49s/it]
 88%|████████▊ | 21515/24420 [8:59:03<1:12:07,  1.49s/it]
                                                         
{'loss': 0.0852, 'learning_rate': 5e-05, 'epoch': 4.41}

 88%|████████▊ | 21515/24420 [8:59:03<1:12:07,  1.49s/it]
 88%|████████▊ | 21516/24420 [8:59:05<1:12:07,  1.49s/it]
 88%|████████▊ | 21517/24420 [8:59:06<1:12:01,  1.49s/it]
 88%|████████▊ | 21518/24420 [8:59:08<1:12:00,  1.49s/it]
 88%|████████▊ | 21519/24420 [8:59:09<1:11:49,  1.49s/it]
 88%|████████▊ | 21520/24420 [8:59:11<1:11:52,  1.49s/it]
                                                         
{'loss': 0.0597, 'learning_rate': 5e-05, 'epoch': 4.41}

 88%|████████▊ | 21520/24420 [8:59:11<1:11:52,  1.49s/it]
 88%|████████▊ | 21521/24420 [8:59:12<1:11:56,  1.49s/it]
 88%|████████▊ | 21522/24420 [8:59:14<1:11:54,  1.49s/it]
 88%|████████▊ | 21523/24420 [8:59:15<1:11:55,  1.49s/it]
 88%|████████▊ | 21524/24420 [8:59:17<1:11:52,  1.49s/it]
 88%|████████▊ | 21525/24420 [8:59:18<1:11:49,  1.49s/it]
                                                         
{'loss': 0.0795, 'learning_rate': 5e-05, 'epoch': 4.41}

 88%|████████▊ | 21525/24420 [8:59:18<1:11:49,  1.49s/it]
 88%|████████▊ | 21526/24420 [8:59:20<1:11:53,  1.49s/it]
 88%|████████▊ | 21527/24420 [8:59:21<1:11:47,  1.49s/it]
 88%|████████▊ | 21528/24420 [8:59:23<1:11:35,  1.49s/it]
 88%|████████▊ | 21529/24420 [8:59:24<1:11:41,  1.49s/it]
 88%|████████▊ | 21530/24420 [8:59:26<1:11:35,  1.49s/it]
                                                         
{'loss': 0.0703, 'learning_rate': 5e-05, 'epoch': 4.41}

 88%|████████▊ | 21530/24420 [8:59:26<1:11:35,  1.49s/it]
 88%|████████▊ | 21531/24420 [8:59:27<1:11:38,  1.49s/it]
 88%|████████▊ | 21532/24420 [8:59:29<1:11:38,  1.49s/it]
 88%|████████▊ | 21533/24420 [8:59:30<1:11:36,  1.49s/it]
 88%|████████▊ | 21534/24420 [8:59:32<1:11:31,  1.49s/it]
 88%|████████▊ | 21535/24420 [8:59:33<1:11:27,  1.49s/it]
                                                         
{'loss': 0.0724, 'learning_rate': 5e-05, 'epoch': 4.41}

 88%|████████▊ | 21535/24420 [8:59:33<1:11:27,  1.49s/it]
 88%|████████▊ | 21536/24420 [8:59:35<1:11:30,  1.49s/it]
 88%|████████▊ | 21537/24420 [8:59:36<1:11:38,  1.49s/it]
 88%|████████▊ | 21538/24420 [8:59:38<1:11:40,  1.49s/it]
 88%|████████▊ | 21539/24420 [8:59:39<1:11:36,  1.49s/it]
 88%|████████▊ | 21540/24420 [8:59:41<1:11:49,  1.50s/it]
                                                         
{'loss': 0.0768, 'learning_rate': 5e-05, 'epoch': 4.41}

 88%|████████▊ | 21540/24420 [8:59:41<1:11:49,  1.50s/it]
 88%|████████▊ | 21541/24420 [8:59:42<1:11:48,  1.50s/it]
 88%|████████▊ | 21542/24420 [8:59:44<1:11:36,  1.49s/it]
 88%|████████▊ | 21543/24420 [8:59:45<1:11:29,  1.49s/it]
 88%|████████▊ | 21544/24420 [8:59:47<1:11:26,  1.49s/it]
 88%|████████▊ | 21545/24420 [8:59:48<1:11:14,  1.49s/it]
                                                         
{'loss': 0.0717, 'learning_rate': 5e-05, 'epoch': 4.41}

 88%|████████▊ | 21545/24420 [8:59:48<1:11:14,  1.49s/it]
 88%|████████▊ | 21546/24420 [8:59:50<1:11:15,  1.49s/it]
 88%|████████▊ | 21547/24420 [8:59:51<1:11:08,  1.49s/it]
 88%|████████▊ | 21548/24420 [8:59:52<1:11:04,  1.48s/it]
 88%|████████▊ | 21549/24420 [8:59:54<1:11:09,  1.49s/it]
 88%|████████▊ | 21550/24420 [8:59:55<1:11:20,  1.49s/it]
                                                         
{'loss': 0.0717, 'learning_rate': 5e-05, 'epoch': 4.41}

 88%|████████▊ | 21550/24420 [8:59:55<1:11:20,  1.49s/it]
 88%|████████▊ | 21551/24420 [8:59:57<1:11:22,  1.49s/it]
 88%|████████▊ | 21552/24420 [8:59:58<1:11:16,  1.49s/it]
 88%|████████▊ | 21553/24420 [9:00:00<1:11:15,  1.49s/it]
 88%|████████▊ | 21554/24420 [9:00:01<1:11:13,  1.49s/it]
 88%|████████▊ | 21555/24420 [9:00:03<1:11:21,  1.49s/it]
                                                         
{'loss': 0.0798, 'learning_rate': 5e-05, 'epoch': 4.41}

 88%|████████▊ | 21555/24420 [9:00:03<1:11:21,  1.49s/it]
 88%|████████▊ | 21556/24420 [9:00:04<1:11:08,  1.49s/it]
 88%|████████▊ | 21557/24420 [9:00:06<1:11:04,  1.49s/it]
 88%|████████▊ | 21558/24420 [9:00:07<1:11:03,  1.49s/it]
 88%|████████▊ | 21559/24420 [9:00:09<1:10:52,  1.49s/it]
 88%|████████▊ | 21560/24420 [9:00:10<1:10:50,  1.49s/it]
                                                         
{'loss': 0.0744, 'learning_rate': 5e-05, 'epoch': 4.41}

 88%|████████▊ | 21560/24420 [9:00:10<1:10:50,  1.49s/it]
 88%|████████▊ | 21561/24420 [9:00:12<1:11:00,  1.49s/it]
 88%|████████▊ | 21562/24420 [9:00:13<1:10:54,  1.49s/it]
 88%|████████▊ | 21563/24420 [9:00:15<1:10:47,  1.49s/it]
 88%|████████▊ | 21564/24420 [9:00:16<1:10:45,  1.49s/it]
 88%|████████▊ | 21565/24420 [9:00:18<1:10:48,  1.49s/it]
                                                         
{'loss': 0.0685, 'learning_rate': 5e-05, 'epoch': 4.42}

 88%|████████▊ | 21565/24420 [9:00:18<1:10:48,  1.49s/it]
 88%|████████▊ | 21566/24420 [9:00:19<1:11:01,  1.49s/it]
 88%|████████▊ | 21567/24420 [9:00:21<1:10:46,  1.49s/it]
 88%|████████▊ | 21568/24420 [9:00:22<1:10:43,  1.49s/it]
 88%|████████▊ | 21569/24420 [9:00:24<1:10:51,  1.49s/it]
 88%|████████▊ | 21570/24420 [9:00:25<1:10:41,  1.49s/it]
                                                         
{'loss': 0.0875, 'learning_rate': 5e-05, 'epoch': 4.42}

 88%|████████▊ | 21570/24420 [9:00:25<1:10:41,  1.49s/it]
 88%|████████▊ | 21571/24420 [9:00:27<1:10:40,  1.49s/it]
 88%|████████▊ | 21572/24420 [9:00:28<1:10:37,  1.49s/it]
 88%|████████▊ | 21573/24420 [9:00:30<1:10:33,  1.49s/it]
 88%|████████▊ | 21574/24420 [9:00:31<1:10:26,  1.48s/it]
 88%|████████▊ | 21575/24420 [9:00:33<1:10:29,  1.49s/it]
                                                         
{'loss': 0.0774, 'learning_rate': 5e-05, 'epoch': 4.42}

 88%|████████▊ | 21575/24420 [9:00:33<1:10:29,  1.49s/it]
 88%|████████▊ | 21576/24420 [9:00:34<1:10:29,  1.49s/it]
 88%|████████▊ | 21577/24420 [9:00:36<1:10:23,  1.49s/it]
 88%|████████▊ | 21578/24420 [9:00:37<1:10:26,  1.49s/it]
 88%|████████▊ | 21579/24420 [9:00:39<1:10:27,  1.49s/it]
 88%|████████▊ | 21580/24420 [9:00:40<1:10:34,  1.49s/it]
                                                         
{'loss': 0.0671, 'learning_rate': 5e-05, 'epoch': 4.42}

 88%|████████▊ | 21580/24420 [9:00:40<1:10:34,  1.49s/it]
 88%|████████▊ | 21581/24420 [9:00:42<1:10:30,  1.49s/it]
 88%|████████▊ | 21582/24420 [9:00:43<1:10:29,  1.49s/it]
 88%|████████▊ | 21583/24420 [9:00:45<1:10:18,  1.49s/it]
 88%|████████▊ | 21584/24420 [9:00:46<1:10:19,  1.49s/it]
 88%|████████▊ | 21585/24420 [9:00:48<1:10:25,  1.49s/it]
                                                         
{'loss': 0.0826, 'learning_rate': 5e-05, 'epoch': 4.42}

 88%|████████▊ | 21585/24420 [9:00:48<1:10:25,  1.49s/it]
 88%|████████▊ | 21586/24420 [9:00:49<1:10:28,  1.49s/it]
 88%|████████▊ | 21587/24420 [9:00:51<1:10:32,  1.49s/it]
 88%|████████▊ | 21588/24420 [9:00:52<1:10:35,  1.50s/it]
 88%|████████▊ | 21589/24420 [9:00:54<1:10:34,  1.50s/it]
 88%|████████▊ | 21590/24420 [9:00:55<1:10:27,  1.49s/it]
                                                         
{'loss': 0.0862, 'learning_rate': 5e-05, 'epoch': 4.42}

 88%|████████▊ | 21590/24420 [9:00:55<1:10:27,  1.49s/it]
 88%|████████▊ | 21591/24420 [9:00:57<1:10:32,  1.50s/it]
 88%|████████▊ | 21592/24420 [9:00:58<1:10:25,  1.49s/it]
 88%|████████▊ | 21593/24420 [9:01:00<1:11:05,  1.51s/it]
 88%|████████▊ | 21594/24420 [9:01:01<1:10:47,  1.50s/it]
 88%|████████▊ | 21595/24420 [9:01:03<1:10:24,  1.50s/it]
                                                         
{'loss': 0.081, 'learning_rate': 5e-05, 'epoch': 4.42}

 88%|████████▊ | 21595/24420 [9:01:03<1:10:24,  1.50s/it]
 88%|████████▊ | 21596/24420 [9:01:04<1:10:22,  1.50s/it]
 88%|████████▊ | 21597/24420 [9:01:06<1:10:19,  1.49s/it]
 88%|████████▊ | 21598/24420 [9:01:07<1:10:18,  1.49s/it]
 88%|████████▊ | 21599/24420 [9:01:09<1:10:22,  1.50s/it]
 88%|████████▊ | 21600/24420 [9:01:10<1:10:18,  1.50s/it]
                                                         
{'loss': 0.0709, 'learning_rate': 5e-05, 'epoch': 4.42}

 88%|████████▊ | 21600/24420 [9:01:10<1:10:18,  1.50s/it]
 88%|████████▊ | 21601/24420 [9:01:16<2:19:22,  2.97s/it]
 88%|████████▊ | 21602/24420 [9:01:18<1:58:21,  2.52s/it]
 88%|████████▊ | 21603/24420 [9:01:19<1:43:40,  2.21s/it]
 88%|████████▊ | 21604/24420 [9:01:21<1:33:31,  1.99s/it]
 88%|████████▊ | 21605/24420 [9:01:22<1:26:26,  1.84s/it]
                                                         
{'loss': 0.0663, 'learning_rate': 5e-05, 'epoch': 4.42}

 88%|████████▊ | 21605/24420 [9:01:22<1:26:26,  1.84s/it]
 88%|████████▊ | 21606/24420 [9:01:24<1:21:23,  1.74s/it]
 88%|████████▊ | 21607/24420 [9:01:25<1:17:55,  1.66s/it]
 88%|████████▊ | 21608/24420 [9:01:27<1:15:32,  1.61s/it]
 88%|████████▊ | 21609/24420 [9:01:28<1:13:45,  1.57s/it]
 88%|████████▊ | 21610/24420 [9:01:30<1:12:33,  1.55s/it]
                                                         
{'loss': 0.0732, 'learning_rate': 5e-05, 'epoch': 4.42}

 88%|████████▊ | 21610/24420 [9:01:30<1:12:33,  1.55s/it]
 88%|████████▊ | 21611/24420 [9:01:31<1:11:50,  1.53s/it]
 89%|████████▊ | 21612/24420 [9:01:33<1:11:10,  1.52s/it]
 89%|████████▊ | 21613/24420 [9:01:34<1:10:39,  1.51s/it]
 89%|████████▊ | 21614/24420 [9:01:36<1:10:17,  1.50s/it]
 89%|████████▊ | 21615/24420 [9:01:37<1:10:01,  1.50s/it]
                                                         
{'loss': 0.0791, 'learning_rate': 5e-05, 'epoch': 4.43}

 89%|████████▊ | 21615/24420 [9:01:37<1:10:01,  1.50s/it]
 89%|████████▊ | 21616/24420 [9:01:39<1:09:59,  1.50s/it]
 89%|████████▊ | 21617/24420 [9:01:40<1:09:48,  1.49s/it]
 89%|████████▊ | 21618/24420 [9:01:42<1:09:47,  1.49s/it]
 89%|████████▊ | 21619/24420 [9:01:43<1:09:37,  1.49s/it]
 89%|████████▊ | 21620/24420 [9:01:45<1:09:33,  1.49s/it]
                                                         
{'loss': 0.0794, 'learning_rate': 5e-05, 'epoch': 4.43}

 89%|████████▊ | 21620/24420 [9:01:45<1:09:33,  1.49s/it]
 89%|████████▊ | 21621/24420 [9:01:46<1:09:42,  1.49s/it]
 89%|████████▊ | 21622/24420 [9:01:48<1:09:36,  1.49s/it]
 89%|████████▊ | 21623/24420 [9:01:49<1:09:33,  1.49s/it]
 89%|████████▊ | 21624/24420 [9:01:51<1:09:23,  1.49s/it]
 89%|████████▊ | 21625/24420 [9:01:52<1:09:22,  1.49s/it]
                                                         
{'loss': 0.0792, 'learning_rate': 5e-05, 'epoch': 4.43}

 89%|████████▊ | 21625/24420 [9:01:52<1:09:22,  1.49s/it]
 89%|████████▊ | 21626/24420 [9:01:54<1:09:22,  1.49s/it]
 89%|████████▊ | 21627/24420 [9:01:55<1:09:17,  1.49s/it]
 89%|████████▊ | 21628/24420 [9:01:57<1:09:21,  1.49s/it]
 89%|████████▊ | 21629/24420 [9:01:58<1:09:25,  1.49s/it]
 89%|████████▊ | 21630/24420 [9:02:00<1:09:24,  1.49s/it]
                                                         
{'loss': 0.0708, 'learning_rate': 5e-05, 'epoch': 4.43}

 89%|████████▊ | 21630/24420 [9:02:00<1:09:24,  1.49s/it]
 89%|████████▊ | 21631/24420 [9:02:01<1:09:28,  1.49s/it]
 89%|████████▊ | 21632/24420 [9:02:03<1:09:23,  1.49s/it]
 89%|████████▊ | 21633/24420 [9:02:04<1:09:15,  1.49s/it]
 89%|████████▊ | 21634/24420 [9:02:06<1:09:07,  1.49s/it]
 89%|████████▊ | 21635/24420 [9:02:07<1:09:01,  1.49s/it]
                                                         
{'loss': 0.0617, 'learning_rate': 5e-05, 'epoch': 4.43}

 89%|████████▊ | 21635/24420 [9:02:07<1:09:01,  1.49s/it]
 89%|████████▊ | 21636/24420 [9:02:09<1:09:07,  1.49s/it]
 89%|████████▊ | 21637/24420 [9:02:10<1:09:04,  1.49s/it]
 89%|████████▊ | 21638/24420 [9:02:12<1:09:03,  1.49s/it]
 89%|████████▊ | 21639/24420 [9:02:13<1:09:03,  1.49s/it]
 89%|████████▊ | 21640/24420 [9:02:15<1:08:58,  1.49s/it]
                                                         
{'loss': 0.0863, 'learning_rate': 5e-05, 'epoch': 4.43}

 89%|████████▊ | 21640/24420 [9:02:15<1:08:58,  1.49s/it]
 89%|████████▊ | 21641/24420 [9:02:16<1:08:56,  1.49s/it]
 89%|████████▊ | 21642/24420 [9:02:18<1:09:05,  1.49s/it]
 89%|████████▊ | 21643/24420 [9:02:19<1:09:03,  1.49s/it]
 89%|████████▊ | 21644/24420 [9:02:21<1:08:54,  1.49s/it]
 89%|████████▊ | 21645/24420 [9:02:22<1:08:58,  1.49s/it]
                                                         
{'loss': 0.0765, 'learning_rate': 5e-05, 'epoch': 4.43}

 89%|████████▊ | 21645/24420 [9:02:22<1:08:58,  1.49s/it]
 89%|████████▊ | 21646/24420 [9:02:24<1:08:54,  1.49s/it]
 89%|████████▊ | 21647/24420 [9:02:25<1:08:50,  1.49s/it]
 89%|████████▊ | 21648/24420 [9:02:26<1:08:47,  1.49s/it]
 89%|████████▊ | 21649/24420 [9:02:28<1:08:40,  1.49s/it]
 89%|████████▊ | 21650/24420 [9:02:29<1:08:38,  1.49s/it]
                                                         
{'loss': 0.0806, 'learning_rate': 5e-05, 'epoch': 4.43}

 89%|████████▊ | 21650/24420 [9:02:29<1:08:38,  1.49s/it]
 89%|████████▊ | 21651/24420 [9:02:31<1:08:43,  1.49s/it]
 89%|████████▊ | 21652/24420 [9:02:32<1:08:40,  1.49s/it]
 89%|████████▊ | 21653/24420 [9:02:34<1:08:39,  1.49s/it]
 89%|████████▊ | 21654/24420 [9:02:35<1:08:39,  1.49s/it]
 89%|████████▊ | 21655/24420 [9:02:37<1:08:44,  1.49s/it]
                                                         
{'loss': 0.0841, 'learning_rate': 5e-05, 'epoch': 4.43}

 89%|████████▊ | 21655/24420 [9:02:37<1:08:44,  1.49s/it]
 89%|████████▊ | 21656/24420 [9:02:38<1:08:42,  1.49s/it]
 89%|████████▊ | 21657/24420 [9:02:40<1:08:51,  1.50s/it]
 89%|████████▊ | 21658/24420 [9:02:41<1:08:41,  1.49s/it]
 89%|████████▊ | 21659/24420 [9:02:43<1:08:36,  1.49s/it]
 89%|████████▊ | 21660/24420 [9:02:44<1:08:31,  1.49s/it]
                                                         
{'loss': 0.0686, 'learning_rate': 5e-05, 'epoch': 4.43}

 89%|████████▊ | 21660/24420 [9:02:44<1:08:31,  1.49s/it]
 89%|████████▊ | 21661/24420 [9:02:46<1:08:33,  1.49s/it]
 89%|████████▊ | 21662/24420 [9:02:47<1:08:25,  1.49s/it]
 89%|████████▊ | 21663/24420 [9:02:49<1:08:24,  1.49s/it]
 89%|████████▊ | 21664/24420 [9:02:50<1:08:20,  1.49s/it]
 89%|████████▊ | 21665/24420 [9:02:52<1:08:14,  1.49s/it]
                                                         
{'loss': 0.1009, 'learning_rate': 5e-05, 'epoch': 4.44}

 89%|████████▊ | 21665/24420 [9:02:52<1:08:14,  1.49s/it]
 89%|████████▊ | 21666/24420 [9:02:53<1:08:21,  1.49s/it]
 89%|████████▊ | 21667/24420 [9:02:55<1:08:17,  1.49s/it]
 89%|████████▊ | 21668/24420 [9:02:56<1:08:28,  1.49s/it]
 89%|████████▊ | 21669/24420 [9:02:58<1:08:32,  1.49s/it]
 89%|████████▊ | 21670/24420 [9:02:59<1:08:36,  1.50s/it]
                                                         
{'loss': 0.0659, 'learning_rate': 5e-05, 'epoch': 4.44}

 89%|████████▊ | 21670/24420 [9:02:59<1:08:36,  1.50s/it]
 89%|████████▊ | 21671/24420 [9:03:01<1:08:24,  1.49s/it]
 89%|████████▊ | 21672/24420 [9:03:02<1:08:16,  1.49s/it]
 89%|████████▉ | 21673/24420 [9:03:04<1:08:07,  1.49s/it]
 89%|████████▉ | 21674/24420 [9:03:05<1:08:06,  1.49s/it]
 89%|████████▉ | 21675/24420 [9:03:07<1:08:05,  1.49s/it]
                                                         
{'loss': 0.0733, 'learning_rate': 5e-05, 'epoch': 4.44}

 89%|████████▉ | 21675/24420 [9:03:07<1:08:05,  1.49s/it]
 89%|████████▉ | 21676/24420 [9:03:08<1:08:09,  1.49s/it]
 89%|████████▉ | 21677/24420 [9:03:10<1:08:05,  1.49s/it]
 89%|████████▉ | 21678/24420 [9:03:11<1:08:09,  1.49s/it]
 89%|████████▉ | 21679/24420 [9:03:13<1:08:01,  1.49s/it]
 89%|████████▉ | 21680/24420 [9:03:14<1:07:58,  1.49s/it]
                                                         
{'loss': 0.0604, 'learning_rate': 5e-05, 'epoch': 4.44}

 89%|████████▉ | 21680/24420 [9:03:14<1:07:58,  1.49s/it]
 89%|████████▉ | 21681/24420 [9:03:16<1:07:49,  1.49s/it]
 89%|████████▉ | 21682/24420 [9:03:17<1:07:56,  1.49s/it]
 89%|████████▉ | 21683/24420 [9:03:19<1:07:53,  1.49s/it]
 89%|████████▉ | 21684/24420 [9:03:20<1:07:59,  1.49s/it]
 89%|████████▉ | 21685/24420 [9:03:22<1:07:57,  1.49s/it]
                                                         
{'loss': 0.07, 'learning_rate': 5e-05, 'epoch': 4.44}

 89%|████████▉ | 21685/24420 [9:03:22<1:07:57,  1.49s/it]
 89%|████████▉ | 21686/24420 [9:03:23<1:07:48,  1.49s/it]
 89%|████████▉ | 21687/24420 [9:03:25<1:07:51,  1.49s/it]
 89%|████████▉ | 21688/24420 [9:03:26<1:07:52,  1.49s/it]
 89%|████████▉ | 21689/24420 [9:03:28<1:07:44,  1.49s/it]
 89%|████████▉ | 21690/24420 [9:03:29<1:07:43,  1.49s/it]
                                                         
{'loss': 0.0675, 'learning_rate': 5e-05, 'epoch': 4.44}

 89%|████████▉ | 21690/24420 [9:03:29<1:07:43,  1.49s/it]
 89%|████████▉ | 21691/24420 [9:03:31<1:07:47,  1.49s/it]
 89%|████████▉ | 21692/24420 [9:03:32<1:07:37,  1.49s/it]
 89%|████████▉ | 21693/24420 [9:03:34<1:07:39,  1.49s/it]
 89%|████████▉ | 21694/24420 [9:03:35<1:07:41,  1.49s/it]
 89%|████████▉ | 21695/24420 [9:03:36<1:07:38,  1.49s/it]
                                                         
{'loss': 0.0658, 'learning_rate': 5e-05, 'epoch': 4.44}

 89%|████████▉ | 21695/24420 [9:03:36<1:07:38,  1.49s/it]
 89%|████████▉ | 21696/24420 [9:03:38<1:07:39,  1.49s/it]
 89%|████████▉ | 21697/24420 [9:03:39<1:07:35,  1.49s/it]
 89%|████████▉ | 21698/24420 [9:03:41<1:07:38,  1.49s/it]
 89%|████████▉ | 21699/24420 [9:03:42<1:07:34,  1.49s/it]
 89%|████████▉ | 21700/24420 [9:03:44<1:07:39,  1.49s/it]
                                                         
{'loss': 0.0808, 'learning_rate': 5e-05, 'epoch': 4.44}

 89%|████████▉ | 21700/24420 [9:03:44<1:07:39,  1.49s/it]
 89%|████████▉ | 21701/24420 [9:03:45<1:07:33,  1.49s/it]
 89%|████████▉ | 21702/24420 [9:03:47<1:07:28,  1.49s/it]
 89%|████████▉ | 21703/24420 [9:03:48<1:07:23,  1.49s/it]
 89%|████████▉ | 21704/24420 [9:03:50<1:07:27,  1.49s/it]
 89%|████████▉ | 21705/24420 [9:03:51<1:07:27,  1.49s/it]
                                                         
{'loss': 0.0661, 'learning_rate': 5e-05, 'epoch': 4.44}

 89%|████████▉ | 21705/24420 [9:03:51<1:07:27,  1.49s/it]
 89%|████████▉ | 21706/24420 [9:03:53<1:07:24,  1.49s/it]
 89%|████████▉ | 21707/24420 [9:03:54<1:07:23,  1.49s/it]
 89%|████████▉ | 21708/24420 [9:03:56<1:07:18,  1.49s/it]
 89%|████████▉ | 21709/24420 [9:03:57<1:07:14,  1.49s/it]
 89%|████████▉ | 21710/24420 [9:03:59<1:07:10,  1.49s/it]
                                                         
{'loss': 0.0624, 'learning_rate': 5e-05, 'epoch': 4.45}

 89%|████████▉ | 21710/24420 [9:03:59<1:07:10,  1.49s/it]
 89%|████████▉ | 21711/24420 [9:04:00<1:07:08,  1.49s/it]
 89%|████████▉ | 21712/24420 [9:04:02<1:07:36,  1.50s/it]
 89%|████████▉ | 21713/24420 [9:04:03<1:07:27,  1.50s/it]
 89%|████████▉ | 21714/24420 [9:04:05<1:07:20,  1.49s/it]
 89%|████████▉ | 21715/24420 [9:04:06<1:07:14,  1.49s/it]
                                                         
{'loss': 0.0802, 'learning_rate': 5e-05, 'epoch': 4.45}

 89%|████████▉ | 21715/24420 [9:04:06<1:07:14,  1.49s/it]
 89%|████████▉ | 21716/24420 [9:04:08<1:07:10,  1.49s/it]
 89%|████████▉ | 21717/24420 [9:04:09<1:07:15,  1.49s/it]
 89%|████████▉ | 21718/24420 [9:04:11<1:07:22,  1.50s/it]
 89%|████████▉ | 21719/24420 [9:04:12<1:07:14,  1.49s/it]
 89%|████████▉ | 21720/24420 [9:04:14<1:07:08,  1.49s/it]
                                                         
{'loss': 0.0756, 'learning_rate': 5e-05, 'epoch': 4.45}

 89%|████████▉ | 21720/24420 [9:04:14<1:07:08,  1.49s/it]
 89%|████████▉ | 21721/24420 [9:04:15<1:07:05,  1.49s/it]
 89%|████████▉ | 21722/24420 [9:04:17<1:07:08,  1.49s/it]
 89%|████████▉ | 21723/24420 [9:04:18<1:07:06,  1.49s/it]
 89%|████████▉ | 21724/24420 [9:04:20<1:07:05,  1.49s/it]
 89%|████████▉ | 21725/24420 [9:04:21<1:07:07,  1.49s/it]
                                                         
{'loss': 0.0744, 'learning_rate': 5e-05, 'epoch': 4.45}

 89%|████████▉ | 21725/24420 [9:04:21<1:07:07,  1.49s/it]
 89%|████████▉ | 21726/24420 [9:04:23<1:07:03,  1.49s/it]
 89%|████████▉ | 21727/24420 [9:04:24<1:07:02,  1.49s/it]
 89%|████████▉ | 21728/24420 [9:04:26<1:07:00,  1.49s/it]
 89%|████████▉ | 21729/24420 [9:04:27<1:06:54,  1.49s/it]
 89%|████████▉ | 21730/24420 [9:04:29<1:06:53,  1.49s/it]
                                                         
{'loss': 0.08, 'learning_rate': 5e-05, 'epoch': 4.45}

 89%|████████▉ | 21730/24420 [9:04:29<1:06:53,  1.49s/it]
 89%|████████▉ | 21731/24420 [9:04:30<1:06:50,  1.49s/it]
 89%|████████▉ | 21732/24420 [9:04:32<1:06:53,  1.49s/it]
 89%|████████▉ | 21733/24420 [9:04:33<1:06:46,  1.49s/it]
 89%|████████▉ | 21734/24420 [9:04:35<1:06:50,  1.49s/it]
 89%|████████▉ | 21735/24420 [9:04:36<1:06:47,  1.49s/it]
                                                         
{'loss': 0.065, 'learning_rate': 5e-05, 'epoch': 4.45}

 89%|████████▉ | 21735/24420 [9:04:36<1:06:47,  1.49s/it]
 89%|████████▉ | 21736/24420 [9:04:38<1:06:43,  1.49s/it]
 89%|████████▉ | 21737/24420 [9:04:39<1:06:39,  1.49s/it]
 89%|████████▉ | 21738/24420 [9:04:41<1:06:35,  1.49s/it]
 89%|████████▉ | 21739/24420 [9:04:42<1:06:30,  1.49s/it]
 89%|████████▉ | 21740/24420 [9:04:44<1:06:27,  1.49s/it]
                                                         
{'loss': 0.0747, 'learning_rate': 5e-05, 'epoch': 4.45}

 89%|████████▉ | 21740/24420 [9:04:44<1:06:27,  1.49s/it]
 89%|████████▉ | 21741/24420 [9:04:45<1:06:39,  1.49s/it]
 89%|████████▉ | 21742/24420 [9:04:47<1:06:33,  1.49s/it]
 89%|████████▉ | 21743/24420 [9:04:48<1:06:45,  1.50s/it]
 89%|████████▉ | 21744/24420 [9:04:50<1:06:38,  1.49s/it]
 89%|████████▉ | 21745/24420 [9:04:51<1:06:31,  1.49s/it]
                                                         
{'loss': 0.0705, 'learning_rate': 5e-05, 'epoch': 4.45}

 89%|████████▉ | 21745/24420 [9:04:51<1:06:31,  1.49s/it]
 89%|████████▉ | 21746/24420 [9:04:53<1:06:38,  1.50s/it]
 89%|████████▉ | 21747/24420 [9:04:54<1:06:36,  1.50s/it]
 89%|████████▉ | 21748/24420 [9:04:56<1:06:23,  1.49s/it]
 89%|████████▉ | 21749/24420 [9:04:57<1:06:13,  1.49s/it]
 89%|████████▉ | 21750/24420 [9:04:59<1:06:13,  1.49s/it]
                                                         
{'loss': 0.0631, 'learning_rate': 5e-05, 'epoch': 4.45}

 89%|████████▉ | 21750/24420 [9:04:59<1:06:13,  1.49s/it]
 89%|████████▉ | 21751/24420 [9:05:00<1:06:15,  1.49s/it]
 89%|████████▉ | 21752/24420 [9:05:02<1:06:07,  1.49s/it]
 89%|████████▉ | 21753/24420 [9:05:03<1:06:11,  1.49s/it]
 89%|████████▉ | 21754/24420 [9:05:04<1:06:07,  1.49s/it]
 89%|████████▉ | 21755/24420 [9:05:06<1:06:05,  1.49s/it]
                                                         
{'loss': 0.0705, 'learning_rate': 5e-05, 'epoch': 4.45}

 89%|████████▉ | 21755/24420 [9:05:06<1:06:05,  1.49s/it]
 89%|████████▉ | 21756/24420 [9:05:07<1:06:11,  1.49s/it]
 89%|████████▉ | 21757/24420 [9:05:09<1:06:04,  1.49s/it]
 89%|████████▉ | 21758/24420 [9:05:10<1:05:54,  1.49s/it]
 89%|████████▉ | 21759/24420 [9:05:12<1:05:55,  1.49s/it]
 89%|████████▉ | 21760/24420 [9:05:13<1:05:55,  1.49s/it]
                                                         
{'loss': 0.0674, 'learning_rate': 5e-05, 'epoch': 4.46}

 89%|████████▉ | 21760/24420 [9:05:13<1:05:55,  1.49s/it]
 89%|████████▉ | 21761/24420 [9:05:15<1:06:01,  1.49s/it]
 89%|████████▉ | 21762/24420 [9:05:16<1:05:53,  1.49s/it]
 89%|████████▉ | 21763/24420 [9:05:18<1:05:55,  1.49s/it]
 89%|████████▉ | 21764/24420 [9:05:19<1:05:53,  1.49s/it]
 89%|████████▉ | 21765/24420 [9:05:21<1:05:51,  1.49s/it]
                                                         
{'loss': 0.0811, 'learning_rate': 5e-05, 'epoch': 4.46}

 89%|████████▉ | 21765/24420 [9:05:21<1:05:51,  1.49s/it]
 89%|████████▉ | 21766/24420 [9:05:22<1:05:48,  1.49s/it]
 89%|████████▉ | 21767/24420 [9:05:24<1:05:44,  1.49s/it]
 89%|████████▉ | 21768/24420 [9:05:25<1:05:48,  1.49s/it]
 89%|████████▉ | 21769/24420 [9:05:27<1:05:47,  1.49s/it]
 89%|████████▉ | 21770/24420 [9:05:28<1:05:44,  1.49s/it]
                                                         
{'loss': 0.0774, 'learning_rate': 5e-05, 'epoch': 4.46}

 89%|████████▉ | 21770/24420 [9:05:28<1:05:44,  1.49s/it]
 89%|████████▉ | 21771/24420 [9:05:30<1:05:38,  1.49s/it]
 89%|████████▉ | 21772/24420 [9:05:31<1:05:32,  1.49s/it]
 89%|████████▉ | 21773/24420 [9:05:33<1:06:06,  1.50s/it]
 89%|████████▉ | 21774/24420 [9:05:34<1:06:01,  1.50s/it]
 89%|████████▉ | 21775/24420 [9:05:36<1:05:53,  1.49s/it]
                                                         
{'loss': 0.072, 'learning_rate': 5e-05, 'epoch': 4.46}

 89%|████████▉ | 21775/24420 [9:05:36<1:05:53,  1.49s/it]
 89%|████████▉ | 21776/24420 [9:05:37<1:05:46,  1.49s/it]
 89%|████████▉ | 21777/24420 [9:05:39<1:05:38,  1.49s/it]
 89%|████████▉ | 21778/24420 [9:05:40<1:05:44,  1.49s/it]
 89%|████████▉ | 21779/24420 [9:05:42<1:05:38,  1.49s/it]
 89%|████████▉ | 21780/24420 [9:05:43<1:05:30,  1.49s/it]
                                                         
{'loss': 0.076, 'learning_rate': 5e-05, 'epoch': 4.46}

 89%|████████▉ | 21780/24420 [9:05:43<1:05:30,  1.49s/it]
 89%|████████▉ | 21781/24420 [9:05:45<1:05:35,  1.49s/it]
 89%|████████▉ | 21782/24420 [9:05:46<1:05:31,  1.49s/it]
 89%|████████▉ | 21783/24420 [9:05:48<1:05:32,  1.49s/it]
 89%|████████▉ | 21784/24420 [9:05:49<1:05:33,  1.49s/it]
 89%|████████▉ | 21785/24420 [9:05:51<1:05:25,  1.49s/it]
                                                         
{'loss': 0.0773, 'learning_rate': 5e-05, 'epoch': 4.46}

 89%|████████▉ | 21785/24420 [9:05:51<1:05:25,  1.49s/it]
 89%|████████▉ | 21786/24420 [9:05:52<1:05:25,  1.49s/it]
 89%|████████▉ | 21787/24420 [9:05:54<1:05:23,  1.49s/it]
 89%|████████▉ | 21788/24420 [9:05:55<1:05:18,  1.49s/it]
 89%|████████▉ | 21789/24420 [9:05:57<1:05:17,  1.49s/it]
 89%|████████▉ | 21790/24420 [9:05:58<1:05:18,  1.49s/it]
                                                         
{'loss': 0.0657, 'learning_rate': 5e-05, 'epoch': 4.46}

 89%|████████▉ | 21790/24420 [9:05:58<1:05:18,  1.49s/it]
 89%|████████▉ | 21791/24420 [9:06:00<1:05:20,  1.49s/it]
 89%|████████▉ | 21792/24420 [9:06:01<1:05:13,  1.49s/it]
 89%|████████▉ | 21793/24420 [9:06:03<1:05:06,  1.49s/it]
 89%|████████▉ | 21794/24420 [9:06:04<1:05:02,  1.49s/it]
 89%|████████▉ | 21795/24420 [9:06:06<1:05:04,  1.49s/it]
                                                         
{'loss': 0.0645, 'learning_rate': 5e-05, 'epoch': 4.46}

 89%|████████▉ | 21795/24420 [9:06:06<1:05:04,  1.49s/it]
 89%|████████▉ | 21796/24420 [9:06:07<1:05:04,  1.49s/it]
 89%|████████▉ | 21797/24420 [9:06:09<1:05:02,  1.49s/it]
 89%|████████▉ | 21798/24420 [9:06:10<1:05:00,  1.49s/it]
 89%|████████▉ | 21799/24420 [9:06:12<1:04:53,  1.49s/it]
 89%|████████▉ | 21800/24420 [9:06:13<1:04:54,  1.49s/it]
                                                         
{'loss': 0.0753, 'learning_rate': 5e-05, 'epoch': 4.46}

 89%|████████▉ | 21800/24420 [9:06:13<1:04:54,  1.49s/it]
 89%|████████▉ | 21801/24420 [9:06:18<1:47:49,  2.47s/it]
 89%|████████▉ | 21802/24420 [9:06:19<1:35:03,  2.18s/it]
 89%|████████▉ | 21803/24420 [9:06:21<1:26:05,  1.97s/it]
 89%|████████▉ | 21804/24420 [9:06:22<1:19:48,  1.83s/it]
 89%|████████▉ | 21805/24420 [9:06:24<1:15:14,  1.73s/it]
                                                         
{'loss': 0.0764, 'learning_rate': 5e-05, 'epoch': 4.46}

 89%|████████▉ | 21805/24420 [9:06:24<1:15:14,  1.73s/it]
 89%|████████▉ | 21806/24420 [9:06:25<1:11:59,  1.65s/it]
 89%|████████▉ | 21807/24420 [9:06:27<1:09:43,  1.60s/it]
 89%|████████▉ | 21808/24420 [9:06:28<1:08:09,  1.57s/it]
 89%|████████▉ | 21809/24420 [9:06:30<1:07:18,  1.55s/it]
 89%|████████▉ | 21810/24420 [9:06:31<1:06:38,  1.53s/it]
                                                         
{'loss': 0.0719, 'learning_rate': 5e-05, 'epoch': 4.47}

 89%|████████▉ | 21810/24420 [9:06:31<1:06:38,  1.53s/it]
 89%|████████▉ | 21811/24420 [9:06:33<1:06:00,  1.52s/it]
 89%|████████▉ | 21812/24420 [9:06:34<1:05:43,  1.51s/it]
 89%|████████▉ | 21813/24420 [9:06:36<1:05:27,  1.51s/it]
 89%|████████▉ | 21814/24420 [9:06:37<1:05:12,  1.50s/it]
 89%|████████▉ | 21815/24420 [9:06:39<1:04:56,  1.50s/it]
                                                         
{'loss': 0.0707, 'learning_rate': 5e-05, 'epoch': 4.47}

 89%|████████▉ | 21815/24420 [9:06:39<1:04:56,  1.50s/it]
 89%|████████▉ | 21816/24420 [9:06:40<1:04:54,  1.50s/it]
 89%|████████▉ | 21817/24420 [9:06:42<1:14:04,  1.71s/it]
 89%|████████▉ | 21818/24420 [9:06:44<1:11:11,  1.64s/it]
 89%|████████▉ | 21819/24420 [9:06:45<1:09:07,  1.59s/it]
 89%|████████▉ | 21820/24420 [9:06:47<1:07:52,  1.57s/it]
                                                         
{'loss': 0.0614, 'learning_rate': 5e-05, 'epoch': 4.47}

 89%|████████▉ | 21820/24420 [9:06:47<1:07:52,  1.57s/it]
 89%|████████▉ | 21821/24420 [9:06:48<1:06:52,  1.54s/it]
 89%|████████▉ | 21822/24420 [9:06:50<1:06:04,  1.53s/it]
 89%|████████▉ | 21823/24420 [9:06:51<1:05:26,  1.51s/it]
 89%|████████▉ | 21824/24420 [9:06:53<1:05:05,  1.50s/it]
 89%|████████▉ | 21825/24420 [9:06:54<1:04:46,  1.50s/it]
                                                         
{'loss': 0.077, 'learning_rate': 5e-05, 'epoch': 4.47}

 89%|████████▉ | 21825/24420 [9:06:54<1:04:46,  1.50s/it]
 89%|████████▉ | 21826/24420 [9:06:56<1:04:36,  1.49s/it]
 89%|████████▉ | 21827/24420 [9:06:57<1:04:40,  1.50s/it]
 89%|████████▉ | 21828/24420 [9:06:59<1:04:27,  1.49s/it]
 89%|████████▉ | 21829/24420 [9:07:00<1:04:19,  1.49s/it]
 89%|████████▉ | 21830/24420 [9:07:02<1:04:21,  1.49s/it]
                                                         
{'loss': 0.0896, 'learning_rate': 5e-05, 'epoch': 4.47}

 89%|████████▉ | 21830/24420 [9:07:02<1:04:21,  1.49s/it]
 89%|████████▉ | 21831/24420 [9:07:03<1:04:14,  1.49s/it]
 89%|████████▉ | 21832/24420 [9:07:05<1:04:10,  1.49s/it]
 89%|████████▉ | 21833/24420 [9:07:06<1:04:17,  1.49s/it]
 89%|████████▉ | 21834/24420 [9:07:08<1:04:17,  1.49s/it]
 89%|████████▉ | 21835/24420 [9:07:09<1:04:10,  1.49s/it]
                                                         
{'loss': 0.0801, 'learning_rate': 5e-05, 'epoch': 4.47}

 89%|████████▉ | 21835/24420 [9:07:09<1:04:10,  1.49s/it]
 89%|████████▉ | 21836/24420 [9:07:11<1:04:12,  1.49s/it]
 89%|████████▉ | 21837/24420 [9:07:12<1:04:14,  1.49s/it]
 89%|████████▉ | 21838/24420 [9:07:14<1:04:14,  1.49s/it]
 89%|████████▉ | 21839/24420 [9:07:15<1:04:06,  1.49s/it]
 89%|████████▉ | 21840/24420 [9:07:17<1:04:01,  1.49s/it]
                                                         
{'loss': 0.0695, 'learning_rate': 5e-05, 'epoch': 4.47}

 89%|████████▉ | 21840/24420 [9:07:17<1:04:01,  1.49s/it]
 89%|████████▉ | 21841/24420 [9:07:18<1:04:00,  1.49s/it]
 89%|████████▉ | 21842/24420 [9:07:20<1:04:01,  1.49s/it]
 89%|████████▉ | 21843/24420 [9:07:21<1:03:50,  1.49s/it]
 89%|████████▉ | 21844/24420 [9:07:23<1:03:48,  1.49s/it]
 89%|████████▉ | 21845/24420 [9:07:24<1:03:43,  1.48s/it]
                                                         
{'loss': 0.0724, 'learning_rate': 5e-05, 'epoch': 4.47}

 89%|████████▉ | 21845/24420 [9:07:24<1:03:43,  1.48s/it]
 89%|████████▉ | 21846/24420 [9:07:25<1:03:44,  1.49s/it]
 89%|████████▉ | 21847/24420 [9:07:27<1:03:48,  1.49s/it]
 89%|████████▉ | 21848/24420 [9:07:28<1:03:44,  1.49s/it]
 89%|████████▉ | 21849/24420 [9:07:30<1:03:44,  1.49s/it]
 89%|████████▉ | 21850/24420 [9:07:31<1:03:47,  1.49s/it]
                                                         
{'loss': 0.0705, 'learning_rate': 5e-05, 'epoch': 4.47}

 89%|████████▉ | 21850/24420 [9:07:31<1:03:47,  1.49s/it]
 89%|████████▉ | 21851/24420 [9:07:33<1:03:47,  1.49s/it]
 89%|████████▉ | 21852/24420 [9:07:34<1:03:52,  1.49s/it]
 89%|████████▉ | 21853/24420 [9:07:36<1:03:50,  1.49s/it]
 89%|████████▉ | 21854/24420 [9:07:37<1:03:41,  1.49s/it]
 89%|████████▉ | 21855/24420 [9:07:39<1:03:44,  1.49s/it]
                                                         
{'loss': 0.0726, 'learning_rate': 5e-05, 'epoch': 4.47}

 89%|████████▉ | 21855/24420 [9:07:39<1:03:44,  1.49s/it]
 90%|████████▉ | 21856/24420 [9:07:40<1:03:43,  1.49s/it]
 90%|████████▉ | 21857/24420 [9:07:42<1:03:37,  1.49s/it]
 90%|████████▉ | 21858/24420 [9:07:43<1:03:32,  1.49s/it]
 90%|████████▉ | 21859/24420 [9:07:45<1:03:26,  1.49s/it]
 90%|████████▉ | 21860/24420 [9:07:46<1:03:32,  1.49s/it]
                                                         
{'loss': 0.0703, 'learning_rate': 5e-05, 'epoch': 4.48}

 90%|████████▉ | 21860/24420 [9:07:46<1:03:32,  1.49s/it]
 90%|████████▉ | 21861/24420 [9:07:48<1:03:44,  1.49s/it]
 90%|████████▉ | 21862/24420 [9:07:49<1:03:45,  1.50s/it]
 90%|████████▉ | 21863/24420 [9:07:51<1:03:36,  1.49s/it]
 90%|████████▉ | 21864/24420 [9:07:52<1:03:36,  1.49s/it]
 90%|████████▉ | 21865/24420 [9:07:54<1:03:43,  1.50s/it]
                                                         
{'loss': 0.1047, 'learning_rate': 5e-05, 'epoch': 4.48}

 90%|████████▉ | 21865/24420 [9:07:54<1:03:43,  1.50s/it]
 90%|████████▉ | 21866/24420 [9:07:55<1:03:41,  1.50s/it]
 90%|████████▉ | 21867/24420 [9:07:57<1:03:33,  1.49s/it]
 90%|████████▉ | 21868/24420 [9:07:58<1:03:26,  1.49s/it]
 90%|████████▉ | 21869/24420 [9:08:00<1:03:32,  1.49s/it]
 90%|████████▉ | 21870/24420 [9:08:01<1:03:28,  1.49s/it]
                                                         
{'loss': 0.0807, 'learning_rate': 5e-05, 'epoch': 4.48}

 90%|████████▉ | 21870/24420 [9:08:01<1:03:28,  1.49s/it]
 90%|████████▉ | 21871/24420 [9:08:03<1:03:31,  1.50s/it]
 90%|████████▉ | 21872/24420 [9:08:04<1:03:23,  1.49s/it]
 90%|████████▉ | 21873/24420 [9:08:06<1:03:18,  1.49s/it]
 90%|████████▉ | 21874/24420 [9:08:07<1:03:15,  1.49s/it]
 90%|████████▉ | 21875/24420 [9:08:09<1:03:05,  1.49s/it]
                                                         
{'loss': 0.0703, 'learning_rate': 5e-05, 'epoch': 4.48}

 90%|████████▉ | 21875/24420 [9:08:09<1:03:05,  1.49s/it]
 90%|████████▉ | 21876/24420 [9:08:10<1:03:09,  1.49s/it]
 90%|████████▉ | 21877/24420 [9:08:12<1:03:08,  1.49s/it]
 90%|████████▉ | 21878/24420 [9:08:13<1:03:09,  1.49s/it]
 90%|████████▉ | 21879/24420 [9:08:15<1:03:12,  1.49s/it]
 90%|████████▉ | 21880/24420 [9:08:16<1:03:09,  1.49s/it]
                                                         
{'loss': 0.0787, 'learning_rate': 5e-05, 'epoch': 4.48}

 90%|████████▉ | 21880/24420 [9:08:16<1:03:09,  1.49s/it]
 90%|████████▉ | 21881/24420 [9:08:18<1:03:09,  1.49s/it]
 90%|████████▉ | 21882/24420 [9:08:19<1:03:09,  1.49s/it]
 90%|████████▉ | 21883/24420 [9:08:21<1:03:02,  1.49s/it]
 90%|████████▉ | 21884/24420 [9:08:22<1:03:01,  1.49s/it]
 90%|████████▉ | 21885/24420 [9:08:24<1:03:00,  1.49s/it]
                                                         
{'loss': 0.07, 'learning_rate': 5e-05, 'epoch': 4.48}

 90%|████████▉ | 21885/24420 [9:08:24<1:03:00,  1.49s/it]
 90%|████████▉ | 21886/24420 [9:08:25<1:02:55,  1.49s/it]
 90%|████████▉ | 21887/24420 [9:08:27<1:02:50,  1.49s/it]
 90%|████████▉ | 21888/24420 [9:08:28<1:02:43,  1.49s/it]
 90%|████████▉ | 21889/24420 [9:08:30<1:02:42,  1.49s/it]
 90%|████████▉ | 21890/24420 [9:08:31<1:02:38,  1.49s/it]
                                                         
{'loss': 0.0708, 'learning_rate': 5e-05, 'epoch': 4.48}

 90%|████████▉ | 21890/24420 [9:08:31<1:02:38,  1.49s/it]
 90%|████████▉ | 21891/24420 [9:08:33<1:02:45,  1.49s/it]
 90%|████████▉ | 21892/24420 [9:08:34<1:02:54,  1.49s/it]
 90%|████████▉ | 21893/24420 [9:08:36<1:02:46,  1.49s/it]
 90%|████████▉ | 21894/24420 [9:08:37<1:02:44,  1.49s/it]
 90%|████████▉ | 21895/24420 [9:08:39<1:02:41,  1.49s/it]
                                                         
{'loss': 0.0692, 'learning_rate': 5e-05, 'epoch': 4.48}

 90%|████████▉ | 21895/24420 [9:08:39<1:02:41,  1.49s/it]
 90%|████████▉ | 21896/24420 [9:08:40<1:02:39,  1.49s/it]
 90%|████████▉ | 21897/24420 [9:08:42<1:02:37,  1.49s/it]
 90%|████████▉ | 21898/24420 [9:08:43<1:02:40,  1.49s/it]
 90%|████████▉ | 21899/24420 [9:08:45<1:02:37,  1.49s/it]
 90%|████████▉ | 21900/24420 [9:08:46<1:02:30,  1.49s/it]
                                                         
{'loss': 0.0655, 'learning_rate': 5e-05, 'epoch': 4.48}

 90%|████████▉ | 21900/24420 [9:08:46<1:02:30,  1.49s/it]
 90%|████████▉ | 21901/24420 [9:08:47<1:02:28,  1.49s/it]
 90%|████████▉ | 21902/24420 [9:08:49<1:02:32,  1.49s/it]
 90%|████████▉ | 21903/24420 [9:08:50<1:02:25,  1.49s/it]
 90%|████████▉ | 21904/24420 [9:08:52<1:02:20,  1.49s/it]
 90%|████████▉ | 21905/24420 [9:08:53<1:02:15,  1.49s/it]
                                                         
{'loss': 0.0711, 'learning_rate': 5e-05, 'epoch': 4.49}

 90%|████████▉ | 21905/24420 [9:08:53<1:02:15,  1.49s/it]
 90%|████████▉ | 21906/24420 [9:08:55<1:02:26,  1.49s/it]
 90%|████████▉ | 21907/24420 [9:08:56<1:02:27,  1.49s/it]
 90%|████████▉ | 21908/24420 [9:08:58<1:02:35,  1.50s/it]
 90%|████████▉ | 21909/24420 [9:08:59<1:02:26,  1.49s/it]
 90%|████████▉ | 21910/24420 [9:09:01<1:02:36,  1.50s/it]
                                                         
{'loss': 0.0697, 'learning_rate': 5e-05, 'epoch': 4.49}

 90%|████████▉ | 21910/24420 [9:09:01<1:02:36,  1.50s/it]
 90%|████████▉ | 21911/24420 [9:09:02<1:02:32,  1.50s/it]
 90%|████████▉ | 21912/24420 [9:09:04<1:02:27,  1.49s/it]
 90%|████████▉ | 21913/24420 [9:09:05<1:02:22,  1.49s/it]
 90%|████████▉ | 21914/24420 [9:09:07<1:02:19,  1.49s/it]
 90%|████████▉ | 21915/24420 [9:09:08<1:02:21,  1.49s/it]
                                                         
{'loss': 0.0821, 'learning_rate': 5e-05, 'epoch': 4.49}

 90%|████████▉ | 21915/24420 [9:09:08<1:02:21,  1.49s/it]
 90%|████████▉ | 21916/24420 [9:09:10<1:02:52,  1.51s/it]
 90%|████████▉ | 21917/24420 [9:09:11<1:02:39,  1.50s/it]
 90%|████████▉ | 21918/24420 [9:09:13<1:02:32,  1.50s/it]
 90%|████████▉ | 21919/24420 [9:09:14<1:02:24,  1.50s/it]
 90%|████████▉ | 21920/24420 [9:09:16<1:02:16,  1.49s/it]
                                                         
{'loss': 0.0818, 'learning_rate': 5e-05, 'epoch': 4.49}

 90%|████████▉ | 21920/24420 [9:09:16<1:02:16,  1.49s/it]
 90%|████████▉ | 21921/24420 [9:09:17<1:02:18,  1.50s/it]
 90%|████████▉ | 21922/24420 [9:09:19<1:02:07,  1.49s/it]
 90%|████████▉ | 21923/24420 [9:09:20<1:01:57,  1.49s/it]
 90%|████████▉ | 21924/24420 [9:09:22<1:01:58,  1.49s/it]
 90%|████████▉ | 21925/24420 [9:09:23<1:02:02,  1.49s/it]
                                                         
{'loss': 0.076, 'learning_rate': 5e-05, 'epoch': 4.49}

 90%|████████▉ | 21925/24420 [9:09:23<1:02:02,  1.49s/it]
 90%|████████▉ | 21926/24420 [9:09:25<1:02:35,  1.51s/it]
 90%|████████▉ | 21927/24420 [9:09:26<1:02:19,  1.50s/it]
 90%|████████▉ | 21928/24420 [9:09:28<1:02:09,  1.50s/it]
 90%|████████▉ | 21929/24420 [9:09:29<1:02:11,  1.50s/it]
 90%|████████▉ | 21930/24420 [9:09:31<1:01:59,  1.49s/it]
                                                         
{'loss': 0.0846, 'learning_rate': 5e-05, 'epoch': 4.49}

 90%|████████▉ | 21930/24420 [9:09:31<1:01:59,  1.49s/it]
 90%|████████▉ | 21931/24420 [9:09:32<1:01:52,  1.49s/it]
 90%|████████▉ | 21932/24420 [9:09:34<1:01:53,  1.49s/it]
 90%|████████▉ | 21933/24420 [9:09:35<1:01:53,  1.49s/it]
 90%|████████▉ | 21934/24420 [9:09:37<1:01:54,  1.49s/it]
 90%|████████▉ | 21935/24420 [9:09:38<1:01:53,  1.49s/it]
                                                         
{'loss': 0.086, 'learning_rate': 5e-05, 'epoch': 4.49}

 90%|████████▉ | 21935/24420 [9:09:38<1:01:53,  1.49s/it]
 90%|████████▉ | 21936/24420 [9:09:40<1:01:50,  1.49s/it]
 90%|████████▉ | 21937/24420 [9:09:41<1:01:44,  1.49s/it]
 90%|████████▉ | 21938/24420 [9:09:43<1:01:46,  1.49s/it]
 90%|████████▉ | 21939/24420 [9:09:44<1:01:44,  1.49s/it]
 90%|████████▉ | 21940/24420 [9:09:46<1:01:35,  1.49s/it]
                                                         
{'loss': 0.0723, 'learning_rate': 5e-05, 'epoch': 4.49}

 90%|████████▉ | 21940/24420 [9:09:46<1:01:35,  1.49s/it]
 90%|████████▉ | 21941/24420 [9:09:47<1:01:34,  1.49s/it]
 90%|████████▉ | 21942/24420 [9:09:49<1:01:31,  1.49s/it]
 90%|████████▉ | 21943/24420 [9:09:50<1:01:35,  1.49s/it]
 90%|████████▉ | 21944/24420 [9:09:52<1:01:29,  1.49s/it]
 90%|████████▉ | 21945/24420 [9:09:53<1:01:25,  1.49s/it]
                                                         
{'loss': 0.0833, 'learning_rate': 5e-05, 'epoch': 4.49}

 90%|████████▉ | 21945/24420 [9:09:53<1:01:25,  1.49s/it]
 90%|████████▉ | 21946/24420 [9:09:55<1:01:26,  1.49s/it]
 90%|████████▉ | 21947/24420 [9:09:56<1:01:24,  1.49s/it]
 90%|████████▉ | 21948/24420 [9:09:58<1:01:21,  1.49s/it]
 90%|████████▉ | 21949/24420 [9:09:59<1:01:15,  1.49s/it]
 90%|████████▉ | 21950/24420 [9:10:01<1:01:20,  1.49s/it]
                                                         
{'loss': 0.0733, 'learning_rate': 5e-05, 'epoch': 4.49}

 90%|████████▉ | 21950/24420 [9:10:01<1:01:20,  1.49s/it]
 90%|████████▉ | 21951/24420 [9:10:02<1:01:14,  1.49s/it]
 90%|████████▉ | 21952/24420 [9:10:04<1:01:14,  1.49s/it]
 90%|████████▉ | 21953/24420 [9:10:05<1:01:11,  1.49s/it]
 90%|████████▉ | 21954/24420 [9:10:07<1:01:11,  1.49s/it]
 90%|████████▉ | 21955/24420 [9:10:08<1:01:05,  1.49s/it]
                                                         
{'loss': 0.0722, 'learning_rate': 5e-05, 'epoch': 4.5}

 90%|████████▉ | 21955/24420 [9:10:08<1:01:05,  1.49s/it]
 90%|████████▉ | 21956/24420 [9:10:10<1:01:13,  1.49s/it]
 90%|████████▉ | 21957/24420 [9:10:11<1:01:07,  1.49s/it]
 90%|████████▉ | 21958/24420 [9:10:13<1:01:10,  1.49s/it]
 90%|████████▉ | 21959/24420 [9:10:14<1:01:05,  1.49s/it]
 90%|████████▉ | 21960/24420 [9:10:16<1:01:02,  1.49s/it]
                                                         
{'loss': 0.0788, 'learning_rate': 5e-05, 'epoch': 4.5}

 90%|████████▉ | 21960/24420 [9:10:16<1:01:02,  1.49s/it]
 90%|████████▉ | 21961/24420 [9:10:17<1:01:07,  1.49s/it]
 90%|████████▉ | 21962/24420 [9:10:19<1:00:59,  1.49s/it]
 90%|████████▉ | 21963/24420 [9:10:20<1:00:52,  1.49s/it]
 90%|████████▉ | 21964/24420 [9:10:21<1:01:01,  1.49s/it]
 90%|████████▉ | 21965/24420 [9:10:23<1:01:03,  1.49s/it]
                                                         
{'loss': 0.0732, 'learning_rate': 5e-05, 'epoch': 4.5}

 90%|████████▉ | 21965/24420 [9:10:23<1:01:03,  1.49s/it]
 90%|████████▉ | 21966/24420 [9:10:24<1:01:05,  1.49s/it]
 90%|████████▉ | 21967/24420 [9:10:26<1:00:59,  1.49s/it]
 90%|████████▉ | 21968/24420 [9:10:27<1:00:53,  1.49s/it]
 90%|████████▉ | 21969/24420 [9:10:29<1:00:56,  1.49s/it]
 90%|████████▉ | 21970/24420 [9:10:30<1:00:46,  1.49s/it]
                                                         
{'loss': 0.0696, 'learning_rate': 5e-05, 'epoch': 4.5}

 90%|████████▉ | 21970/24420 [9:10:30<1:00:46,  1.49s/it]
 90%|████████▉ | 21971/24420 [9:10:32<1:00:47,  1.49s/it]
 90%|████████▉ | 21972/24420 [9:10:33<1:00:37,  1.49s/it]
 90%|████████▉ | 21973/24420 [9:10:35<1:00:37,  1.49s/it]
 90%|████████▉ | 21974/24420 [9:10:36<1:00:36,  1.49s/it]
 90%|████████▉ | 21975/24420 [9:10:38<1:00:39,  1.49s/it]
                                                         
{'loss': 0.0754, 'learning_rate': 5e-05, 'epoch': 4.5}

 90%|████████▉ | 21975/24420 [9:10:38<1:00:39,  1.49s/it]
 90%|████████▉ | 21976/24420 [9:10:39<1:00:35,  1.49s/it]
 90%|████████▉ | 21977/24420 [9:10:41<1:01:03,  1.50s/it]
 90%|█████████ | 21978/24420 [9:10:42<1:00:58,  1.50s/it]
 90%|█████████ | 21979/24420 [9:10:44<1:00:53,  1.50s/it]
 90%|█████████ | 21980/24420 [9:10:45<1:00:45,  1.49s/it]
                                                         
{'loss': 0.0756, 'learning_rate': 5e-05, 'epoch': 4.5}

 90%|█████████ | 21980/24420 [9:10:45<1:00:45,  1.49s/it]
 90%|█████████ | 21981/24420 [9:10:47<1:00:43,  1.49s/it]
 90%|█████████ | 21982/24420 [9:10:48<1:00:43,  1.49s/it]
 90%|█████████ | 21983/24420 [9:10:50<1:00:33,  1.49s/it]
 90%|█████████ | 21984/24420 [9:10:51<1:00:19,  1.49s/it]
 90%|█████████ | 21985/24420 [9:10:53<1:00:28,  1.49s/it]
                                                         
{'loss': 0.0721, 'learning_rate': 5e-05, 'epoch': 4.5}

 90%|█████████ | 21985/24420 [9:10:53<1:00:28,  1.49s/it]
 90%|█████████ | 21986/24420 [9:10:54<1:00:19,  1.49s/it]
 90%|█████████ | 21987/24420 [9:10:56<1:00:18,  1.49s/it]
 90%|█████████ | 21988/24420 [9:10:57<1:00:17,  1.49s/it]
 90%|█████████ | 21989/24420 [9:10:59<1:00:20,  1.49s/it]
 90%|█████████ | 21990/24420 [9:11:00<1:00:20,  1.49s/it]
                                                         
{'loss': 0.0702, 'learning_rate': 5e-05, 'epoch': 4.5}

 90%|█████████ | 21990/24420 [9:11:00<1:00:20,  1.49s/it]
 90%|█████████ | 21991/24420 [9:11:02<1:00:20,  1.49s/it]
 90%|█████████ | 21992/24420 [9:11:03<1:00:13,  1.49s/it]
 90%|█████████ | 21993/24420 [9:11:05<1:00:06,  1.49s/it]
 90%|█████████ | 21994/24420 [9:11:06<1:00:01,  1.48s/it]
 90%|█████████ | 21995/24420 [9:11:08<1:00:02,  1.49s/it]
                                                         
{'loss': 0.0634, 'learning_rate': 5e-05, 'epoch': 4.5}

 90%|█████████ | 21995/24420 [9:11:08<1:00:02,  1.49s/it]
 90%|█████████ | 21996/24420 [9:11:09<1:00:12,  1.49s/it]
 90%|█████████ | 21997/24420 [9:11:11<1:00:08,  1.49s/it]
 90%|█████████ | 21998/24420 [9:11:12<1:00:05,  1.49s/it]
 90%|█████████ | 21999/24420 [9:11:14<1:00:07,  1.49s/it]
 90%|█████████ | 22000/24420 [9:11:15<1:00:00,  1.49s/it]
                                                         
{'loss': 0.0752, 'learning_rate': 5e-05, 'epoch': 4.5}

 90%|█████████ | 22000/24420 [9:11:15<1:00:00,  1.49s/it]
 90%|█████████ | 22001/24420 [9:11:20<1:40:35,  2.49s/it]
 90%|█████████ | 22002/24420 [9:11:21<1:28:20,  2.19s/it]
 90%|█████████ | 22003/24420 [9:11:23<1:19:51,  1.98s/it]
 90%|█████████ | 22004/24420 [9:11:24<1:13:46,  1.83s/it]
 90%|█████████ | 22005/24420 [9:11:26<1:09:35,  1.73s/it]
                                                         
{'loss': 0.0752, 'learning_rate': 5e-05, 'epoch': 4.51}

 90%|█████████ | 22005/24420 [9:11:26<1:09:35,  1.73s/it]
 90%|█████████ | 22006/24420 [9:11:27<1:06:40,  1.66s/it]
 90%|█████████ | 22007/24420 [9:11:29<1:04:34,  1.61s/it]
 90%|█████████ | 22008/24420 [9:11:30<1:03:06,  1.57s/it]
 90%|█████████ | 22009/24420 [9:11:32<1:02:02,  1.54s/it]
 90%|█████████ | 22010/24420 [9:11:33<1:01:24,  1.53s/it]
                                                         
{'loss': 0.0722, 'learning_rate': 5e-05, 'epoch': 4.51}

 90%|█████████ | 22010/24420 [9:11:33<1:01:24,  1.53s/it]
 90%|█████████ | 22011/24420 [9:11:35<1:01:01,  1.52s/it]
 90%|█████████ | 22012/24420 [9:11:36<1:00:35,  1.51s/it]
 90%|█████████ | 22013/24420 [9:11:38<1:00:14,  1.50s/it]
 90%|█████████ | 22014/24420 [9:11:39<1:01:22,  1.53s/it]
 90%|█████████ | 22015/24420 [9:11:41<1:00:48,  1.52s/it]
                                                         
{'loss': 0.0827, 'learning_rate': 5e-05, 'epoch': 4.51}

 90%|█████████ | 22015/24420 [9:11:41<1:00:48,  1.52s/it]
 90%|█████████ | 22016/24420 [9:11:42<1:00:26,  1.51s/it]
 90%|█████████ | 22017/24420 [9:11:44<1:00:09,  1.50s/it]
 90%|█████████ | 22018/24420 [9:11:45<59:59,  1.50s/it]  
 90%|█████████ | 22019/24420 [9:11:47<59:50,  1.50s/it]
 90%|█████████ | 22020/24420 [9:11:48<59:46,  1.49s/it]
                                                       
{'loss': 0.0699, 'learning_rate': 5e-05, 'epoch': 4.51}

 90%|█████████ | 22020/24420 [9:11:48<59:46,  1.49s/it]
 90%|█████████ | 22021/24420 [9:11:50<59:43,  1.49s/it]
 90%|█████████ | 22022/24420 [9:11:51<59:36,  1.49s/it]
 90%|█████████ | 22023/24420 [9:11:53<59:35,  1.49s/it]
 90%|█████████ | 22024/24420 [9:11:54<59:30,  1.49s/it]
 90%|█████████ | 22025/24420 [9:11:56<59:28,  1.49s/it]
                                                       
{'loss': 0.078, 'learning_rate': 5e-05, 'epoch': 4.51}

 90%|█████████ | 22025/24420 [9:11:56<59:28,  1.49s/it]
 90%|█████████ | 22026/24420 [9:11:57<59:22,  1.49s/it]
 90%|█████████ | 22027/24420 [9:11:59<59:20,  1.49s/it]
 90%|█████████ | 22028/24420 [9:12:00<59:24,  1.49s/it]
 90%|█████████ | 22029/24420 [9:12:02<59:23,  1.49s/it]
 90%|█████████ | 22030/24420 [9:12:03<59:28,  1.49s/it]
                                                       
{'loss': 0.0698, 'learning_rate': 5e-05, 'epoch': 4.51}

 90%|█████████ | 22030/24420 [9:12:03<59:28,  1.49s/it]
 90%|█████████ | 22031/24420 [9:12:05<59:31,  1.50s/it]
 90%|█████████ | 22032/24420 [9:12:06<59:28,  1.49s/it]
 90%|█████████ | 22033/24420 [9:12:08<59:25,  1.49s/it]
 90%|█████████ | 22034/24420 [9:12:09<59:17,  1.49s/it]
 90%|█████████ | 22035/24420 [9:12:11<59:07,  1.49s/it]
                                                       
{'loss': 0.0706, 'learning_rate': 5e-05, 'epoch': 4.51}

 90%|█████████ | 22035/24420 [9:12:11<59:07,  1.49s/it]
 90%|█████████ | 22036/24420 [9:12:12<59:12,  1.49s/it]
 90%|█████████ | 22037/24420 [9:12:14<59:10,  1.49s/it]
 90%|█████████ | 22038/24420 [9:12:15<59:05,  1.49s/it]
 90%|█████████ | 22039/24420 [9:12:17<59:07,  1.49s/it]
 90%|█████████ | 22040/24420 [9:12:18<59:11,  1.49s/it]
                                                       
{'loss': 0.0735, 'learning_rate': 5e-05, 'epoch': 4.51}

 90%|█████████ | 22040/24420 [9:12:18<59:11,  1.49s/it]
 90%|█████████ | 22041/24420 [9:12:20<59:11,  1.49s/it]
 90%|█████████ | 22042/24420 [9:12:21<59:17,  1.50s/it]
 90%|█████████ | 22043/24420 [9:12:23<59:19,  1.50s/it]
 90%|█████████ | 22044/24420 [9:12:24<59:08,  1.49s/it]
 90%|█████████ | 22045/24420 [9:12:26<59:06,  1.49s/it]
                                                       
{'loss': 0.0679, 'learning_rate': 5e-05, 'epoch': 4.51}

 90%|█████████ | 22045/24420 [9:12:26<59:06,  1.49s/it]
 90%|█████████ | 22046/24420 [9:12:27<59:16,  1.50s/it]
 90%|█████████ | 22047/24420 [9:12:29<59:16,  1.50s/it]
 90%|█████████ | 22048/24420 [9:12:30<59:08,  1.50s/it]
 90%|█████████ | 22049/24420 [9:12:32<59:06,  1.50s/it]
 90%|█████████ | 22050/24420 [9:12:33<58:59,  1.49s/it]
                                                       
{'loss': 0.0884, 'learning_rate': 5e-05, 'epoch': 4.51}

 90%|█████████ | 22050/24420 [9:12:33<58:59,  1.49s/it]
 90%|█████████ | 22051/24420 [9:12:35<58:56,  1.49s/it]
 90%|█████████ | 22052/24420 [9:12:36<58:47,  1.49s/it]
 90%|█████████ | 22053/24420 [9:12:38<58:43,  1.49s/it]
 90%|█████████ | 22054/24420 [9:12:39<58:49,  1.49s/it]
 90%|█████████ | 22055/24420 [9:12:41<58:47,  1.49s/it]
                                                       
{'loss': 0.0799, 'learning_rate': 5e-05, 'epoch': 4.52}

 90%|█████████ | 22055/24420 [9:12:41<58:47,  1.49s/it]
 90%|█████████ | 22056/24420 [9:12:42<58:44,  1.49s/it]
 90%|█████████ | 22057/24420 [9:12:44<58:37,  1.49s/it]
 90%|█████████ | 22058/24420 [9:12:45<58:38,  1.49s/it]
 90%|█████████ | 22059/24420 [9:12:47<58:32,  1.49s/it]
 90%|█████████ | 22060/24420 [9:12:48<58:47,  1.49s/it]
                                                       
{'loss': 0.0781, 'learning_rate': 5e-05, 'epoch': 4.52}

 90%|█████████ | 22060/24420 [9:12:48<58:47,  1.49s/it]
 90%|█████████ | 22061/24420 [9:12:50<58:41,  1.49s/it]
 90%|█████████ | 22062/24420 [9:12:51<58:40,  1.49s/it]
 90%|█████████ | 22063/24420 [9:12:53<58:33,  1.49s/it]
 90%|█████████ | 22064/24420 [9:12:54<58:20,  1.49s/it]
 90%|█████████ | 22065/24420 [9:12:55<58:16,  1.48s/it]
                                                       
{'loss': 0.0659, 'learning_rate': 5e-05, 'epoch': 4.52}

 90%|█████████ | 22065/24420 [9:12:55<58:16,  1.48s/it]
 90%|█████████ | 22066/24420 [9:12:57<58:22,  1.49s/it]
 90%|█████████ | 22067/24420 [9:12:58<58:25,  1.49s/it]
 90%|█████████ | 22068/24420 [9:13:00<58:24,  1.49s/it]
 90%|█████████ | 22069/24420 [9:13:01<58:21,  1.49s/it]
 90%|█████████ | 22070/24420 [9:13:03<58:32,  1.49s/it]
                                                       
{'loss': 0.0868, 'learning_rate': 5e-05, 'epoch': 4.52}

 90%|█████████ | 22070/24420 [9:13:03<58:32,  1.49s/it]
 90%|█████████ | 22071/24420 [9:13:04<58:30,  1.49s/it]
 90%|█████████ | 22072/24420 [9:13:06<58:23,  1.49s/it]
 90%|█████████ | 22073/24420 [9:13:07<58:21,  1.49s/it]
 90%|█████████ | 22074/24420 [9:13:09<58:19,  1.49s/it]
 90%|█████████ | 22075/24420 [9:13:10<58:12,  1.49s/it]
                                                       
{'loss': 0.085, 'learning_rate': 5e-05, 'epoch': 4.52}

 90%|█████████ | 22075/24420 [9:13:10<58:12,  1.49s/it]
 90%|█████████ | 22076/24420 [9:13:12<58:19,  1.49s/it]
 90%|█████████ | 22077/24420 [9:13:13<58:15,  1.49s/it]
 90%|█████████ | 22078/24420 [9:13:15<58:15,  1.49s/it]
 90%|█████████ | 22079/24420 [9:13:16<58:21,  1.50s/it]
 90%|█████████ | 22080/24420 [9:13:18<58:14,  1.49s/it]
                                                       
{'loss': 0.0698, 'learning_rate': 5e-05, 'epoch': 4.52}

 90%|█████████ | 22080/24420 [9:13:18<58:14,  1.49s/it]
 90%|█████████ | 22081/24420 [9:13:19<58:12,  1.49s/it]
 90%|█████████ | 22082/24420 [9:13:21<58:07,  1.49s/it]
 90%|█████████ | 22083/24420 [9:13:22<58:00,  1.49s/it]
 90%|█████████ | 22084/24420 [9:13:24<57:54,  1.49s/it]
 90%|█████████ | 22085/24420 [9:13:25<57:54,  1.49s/it]
                                                       
{'loss': 0.0781, 'learning_rate': 5e-05, 'epoch': 4.52}

 90%|█████████ | 22085/24420 [9:13:25<57:54,  1.49s/it]
 90%|█████████ | 22086/24420 [9:13:27<57:59,  1.49s/it]
 90%|█████████ | 22087/24420 [9:13:28<57:51,  1.49s/it]
 90%|█████████ | 22088/24420 [9:13:30<57:49,  1.49s/it]
 90%|█████████ | 22089/24420 [9:13:31<57:47,  1.49s/it]
 90%|█████████ | 22090/24420 [9:13:33<57:57,  1.49s/it]
                                                       
{'loss': 0.0803, 'learning_rate': 5e-05, 'epoch': 4.52}

 90%|█████████ | 22090/24420 [9:13:33<57:57,  1.49s/it]
 90%|█████████ | 22091/24420 [9:13:34<57:54,  1.49s/it]
 90%|█████████ | 22092/24420 [9:13:36<57:50,  1.49s/it]
 90%|█████████ | 22093/24420 [9:13:37<57:45,  1.49s/it]
 90%|█████████ | 22094/24420 [9:13:39<57:50,  1.49s/it]
 90%|█████████ | 22095/24420 [9:13:40<57:47,  1.49s/it]
                                                       
{'loss': 0.0921, 'learning_rate': 5e-05, 'epoch': 4.52}

 90%|█████████ | 22095/24420 [9:13:40<57:47,  1.49s/it]
 90%|█████████ | 22096/24420 [9:13:42<57:44,  1.49s/it]
 90%|█████████ | 22097/24420 [9:13:43<57:38,  1.49s/it]
 90%|█████████ | 22098/24420 [9:13:45<57:42,  1.49s/it]
 90%|█████████ | 22099/24420 [9:13:46<57:42,  1.49s/it]
 90%|█████████ | 22100/24420 [9:13:48<57:44,  1.49s/it]
                                                       
{'loss': 0.0875, 'learning_rate': 5e-05, 'epoch': 4.52}

 90%|█████████ | 22100/24420 [9:13:48<57:44,  1.49s/it]
 91%|█████████ | 22101/24420 [9:13:49<57:38,  1.49s/it]
 91%|█████████ | 22102/24420 [9:13:51<57:40,  1.49s/it]
 91%|█████████ | 22103/24420 [9:13:52<57:36,  1.49s/it]
 91%|█████████ | 22104/24420 [9:13:54<57:26,  1.49s/it]
 91%|█████████ | 22105/24420 [9:13:55<57:24,  1.49s/it]
                                                       
{'loss': 0.0606, 'learning_rate': 5e-05, 'epoch': 4.53}

 91%|█████████ | 22105/24420 [9:13:55<57:24,  1.49s/it]
 91%|█████████ | 22106/24420 [9:13:57<57:18,  1.49s/it]
 91%|█████████ | 22107/24420 [9:13:58<57:24,  1.49s/it]
 91%|█████████ | 22108/24420 [9:14:00<57:23,  1.49s/it]
 91%|█████████ | 22109/24420 [9:14:01<57:18,  1.49s/it]
 91%|█████████ | 22110/24420 [9:14:03<57:21,  1.49s/it]
                                                       
{'loss': 0.086, 'learning_rate': 5e-05, 'epoch': 4.53}

 91%|█████████ | 22110/24420 [9:14:03<57:21,  1.49s/it]
 91%|█████████ | 22111/24420 [9:14:04<57:15,  1.49s/it]
 91%|█████████ | 22112/24420 [9:14:06<57:15,  1.49s/it]
 91%|█████████ | 22113/24420 [9:14:07<57:17,  1.49s/it]
 91%|█████████ | 22114/24420 [9:14:09<57:18,  1.49s/it]
 91%|█████████ | 22115/24420 [9:14:10<57:19,  1.49s/it]
                                                       
{'loss': 0.0661, 'learning_rate': 5e-05, 'epoch': 4.53}

 91%|█████████ | 22115/24420 [9:14:10<57:19,  1.49s/it]
 91%|█████████ | 22116/24420 [9:14:12<57:17,  1.49s/it]
 91%|█████████ | 22117/24420 [9:14:13<57:08,  1.49s/it]
 91%|█████████ | 22118/24420 [9:14:14<57:03,  1.49s/it]
 91%|█████████ | 22119/24420 [9:14:16<56:58,  1.49s/it]
 91%|█████████ | 22120/24420 [9:14:17<57:00,  1.49s/it]
                                                       
{'loss': 0.0702, 'learning_rate': 5e-05, 'epoch': 4.53}

 91%|█████████ | 22120/24420 [9:14:17<57:00,  1.49s/it]
 91%|█████████ | 22121/24420 [9:14:19<57:04,  1.49s/it]
 91%|█████████ | 22122/24420 [9:14:20<56:58,  1.49s/it]
 91%|█████████ | 22123/24420 [9:14:22<57:01,  1.49s/it]
 91%|█████████ | 22124/24420 [9:14:23<56:57,  1.49s/it]
 91%|█████████ | 22125/24420 [9:14:25<57:03,  1.49s/it]
                                                       
{'loss': 0.073, 'learning_rate': 5e-05, 'epoch': 4.53}

 91%|█████████ | 22125/24420 [9:14:25<57:03,  1.49s/it]
 91%|█████████ | 22126/24420 [9:14:26<57:05,  1.49s/it]
 91%|█████████ | 22127/24420 [9:14:28<56:57,  1.49s/it]
 91%|█████████ | 22128/24420 [9:14:29<56:52,  1.49s/it]
 91%|█████████ | 22129/24420 [9:14:31<56:45,  1.49s/it]
 91%|█████████ | 22130/24420 [9:14:32<56:43,  1.49s/it]
                                                       
{'loss': 0.0608, 'learning_rate': 5e-05, 'epoch': 4.53}

 91%|█████████ | 22130/24420 [9:14:32<56:43,  1.49s/it]
 91%|█████████ | 22131/24420 [9:14:34<56:45,  1.49s/it]
 91%|█████████ | 22132/24420 [9:14:35<56:42,  1.49s/it]
 91%|█████████ | 22133/24420 [9:14:37<56:45,  1.49s/it]
 91%|█████████ | 22134/24420 [9:14:38<56:56,  1.49s/it]
 91%|█████████ | 22135/24420 [9:14:40<57:11,  1.50s/it]
                                                       
{'loss': 0.0666, 'learning_rate': 5e-05, 'epoch': 4.53}

 91%|█████████ | 22135/24420 [9:14:40<57:11,  1.50s/it]
 91%|█████████ | 22136/24420 [9:14:41<56:59,  1.50s/it]
 91%|█████████ | 22137/24420 [9:14:43<56:54,  1.50s/it]
 91%|█████████ | 22138/24420 [9:14:44<56:47,  1.49s/it]
 91%|█████████ | 22139/24420 [9:14:46<56:43,  1.49s/it]
 91%|█████████ | 22140/24420 [9:14:47<56:35,  1.49s/it]
                                                       
{'loss': 0.0667, 'learning_rate': 5e-05, 'epoch': 4.53}

 91%|█████████ | 22140/24420 [9:14:47<56:35,  1.49s/it]
 91%|█████████ | 22141/24420 [9:14:49<56:33,  1.49s/it]
 91%|█████████ | 22142/24420 [9:14:50<56:25,  1.49s/it]
 91%|█████████ | 22143/24420 [9:14:52<56:26,  1.49s/it]
 91%|█████████ | 22144/24420 [9:14:53<56:24,  1.49s/it]
 91%|█████████ | 22145/24420 [9:14:55<56:25,  1.49s/it]
                                                       
{'loss': 0.0756, 'learning_rate': 5e-05, 'epoch': 4.53}

 91%|█████████ | 22145/24420 [9:14:55<56:25,  1.49s/it]
 91%|█████████ | 22146/24420 [9:14:56<56:21,  1.49s/it]
 91%|█████████ | 22147/24420 [9:14:58<56:25,  1.49s/it]
 91%|█████████ | 22148/24420 [9:14:59<56:18,  1.49s/it]
 91%|█████████ | 22149/24420 [9:15:01<56:18,  1.49s/it]
 91%|█████████ | 22150/24420 [9:15:02<56:11,  1.49s/it]
                                                       
{'loss': 0.0652, 'learning_rate': 5e-05, 'epoch': 4.54}

 91%|█████████ | 22150/24420 [9:15:02<56:11,  1.49s/it]
 91%|█████████ | 22151/24420 [9:15:04<56:21,  1.49s/it]
 91%|█████████ | 22152/24420 [9:15:05<56:22,  1.49s/it]
 91%|█████████ | 22153/24420 [9:15:07<56:19,  1.49s/it]
 91%|█████████ | 22154/24420 [9:15:08<56:18,  1.49s/it]
 91%|█████████ | 22155/24420 [9:15:10<56:22,  1.49s/it]
                                                       
{'loss': 0.0742, 'learning_rate': 5e-05, 'epoch': 4.54}

 91%|█████████ | 22155/24420 [9:15:10<56:22,  1.49s/it]
 91%|█████████ | 22156/24420 [9:15:11<56:30,  1.50s/it]
 91%|█████████ | 22157/24420 [9:15:13<56:22,  1.49s/it]
 91%|█████████ | 22158/24420 [9:15:14<56:22,  1.50s/it]
 91%|█████████ | 22159/24420 [9:15:16<56:12,  1.49s/it]
 91%|█████████ | 22160/24420 [9:15:17<56:07,  1.49s/it]
                                                       
{'loss': 0.0812, 'learning_rate': 5e-05, 'epoch': 4.54}

 91%|█████████ | 22160/24420 [9:15:17<56:07,  1.49s/it]
 91%|█████████ | 22161/24420 [9:15:19<56:02,  1.49s/it]
 91%|█████████ | 22162/24420 [9:15:20<56:08,  1.49s/it]
 91%|█████████ | 22163/24420 [9:15:22<56:04,  1.49s/it]
 91%|█████████ | 22164/24420 [9:15:23<56:00,  1.49s/it]
 91%|█████████ | 22165/24420 [9:15:25<56:01,  1.49s/it]
                                                       
{'loss': 0.0691, 'learning_rate': 5e-05, 'epoch': 4.54}

 91%|█████████ | 22165/24420 [9:15:25<56:01,  1.49s/it]
 91%|█████████ | 22166/24420 [9:15:26<56:08,  1.49s/it]
 91%|█████████ | 22167/24420 [9:15:28<56:09,  1.50s/it]
 91%|█████████ | 22168/24420 [9:15:29<56:07,  1.50s/it]
 91%|█████████ | 22169/24420 [9:15:31<56:03,  1.49s/it]
 91%|█████████ | 22170/24420 [9:15:32<56:01,  1.49s/it]
                                                       
{'loss': 0.0631, 'learning_rate': 5e-05, 'epoch': 4.54}

 91%|█████████ | 22170/24420 [9:15:32<56:01,  1.49s/it]
 91%|█████████ | 22171/24420 [9:15:33<56:01,  1.49s/it]
 91%|█████████ | 22172/24420 [9:15:35<55:58,  1.49s/it]
 91%|█████████ | 22173/24420 [9:15:36<55:55,  1.49s/it]
 91%|█████████ | 22174/24420 [9:15:38<55:53,  1.49s/it]
 91%|█████████ | 22175/24420 [9:15:39<55:47,  1.49s/it]
                                                       
{'loss': 0.0774, 'learning_rate': 5e-05, 'epoch': 4.54}

 91%|█████████ | 22175/24420 [9:15:39<55:47,  1.49s/it]
 91%|█████████ | 22176/24420 [9:15:41<55:45,  1.49s/it]
 91%|█████████ | 22177/24420 [9:15:42<55:51,  1.49s/it]
 91%|█████████ | 22178/24420 [9:15:44<55:42,  1.49s/it]
 91%|█████████ | 22179/24420 [9:15:45<55:40,  1.49s/it]
 91%|█████████ | 22180/24420 [9:15:47<55:34,  1.49s/it]
                                                       
{'loss': 0.0673, 'learning_rate': 5e-05, 'epoch': 4.54}

 91%|█████████ | 22180/24420 [9:15:47<55:34,  1.49s/it]
 91%|█████████ | 22181/24420 [9:15:48<55:39,  1.49s/it]
 91%|█████████ | 22182/24420 [9:15:50<55:34,  1.49s/it]
 91%|█████████ | 22183/24420 [9:15:51<55:33,  1.49s/it]
 91%|█████████ | 22184/24420 [9:15:53<55:38,  1.49s/it]
 91%|█████████ | 22185/24420 [9:15:54<55:30,  1.49s/it]
                                                       
{'loss': 0.0856, 'learning_rate': 5e-05, 'epoch': 4.54}

 91%|█████████ | 22185/24420 [9:15:54<55:30,  1.49s/it]
 91%|█████████ | 22186/24420 [9:15:56<55:31,  1.49s/it]
 91%|█████████ | 22187/24420 [9:15:57<55:27,  1.49s/it]
 91%|█████████ | 22188/24420 [9:15:59<55:20,  1.49s/it]
 91%|█████████ | 22189/24420 [9:16:00<55:23,  1.49s/it]
 91%|█████████ | 22190/24420 [9:16:02<55:22,  1.49s/it]
                                                       
{'loss': 0.0671, 'learning_rate': 5e-05, 'epoch': 4.54}

 91%|█████████ | 22190/24420 [9:16:02<55:22,  1.49s/it]
 91%|█████████ | 22191/24420 [9:16:03<55:19,  1.49s/it]
 91%|█████████ | 22192/24420 [9:16:05<55:14,  1.49s/it]
 91%|█████████ | 22193/24420 [9:16:06<55:15,  1.49s/it]
 91%|█████████ | 22194/24420 [9:16:08<55:10,  1.49s/it]
 91%|█████████ | 22195/24420 [9:16:09<55:06,  1.49s/it]
                                                       
{'loss': 0.0649, 'learning_rate': 5e-05, 'epoch': 4.54}

 91%|█████████ | 22195/24420 [9:16:09<55:06,  1.49s/it]
 91%|█████████ | 22196/24420 [9:16:11<55:07,  1.49s/it]
 91%|█████████ | 22197/24420 [9:16:12<55:06,  1.49s/it]
 91%|█████████ | 22198/24420 [9:16:14<55:02,  1.49s/it]
 91%|█████████ | 22199/24420 [9:16:15<55:07,  1.49s/it]
 91%|█████████ | 22200/24420 [9:16:17<54:58,  1.49s/it]
                                                       
{'loss': 0.0713, 'learning_rate': 5e-05, 'epoch': 4.55}

 91%|█████████ | 22200/24420 [9:16:17<54:58,  1.49s/it]
 91%|█████████ | 22201/24420 [9:16:21<1:29:10,  2.41s/it]
 91%|█████████ | 22202/24420 [9:16:23<1:18:58,  2.14s/it]
 91%|█████████ | 22203/24420 [9:16:24<1:11:42,  1.94s/it]
 91%|█████████ | 22204/24420 [9:16:26<1:06:43,  1.81s/it]
 91%|█████████ | 22205/24420 [9:16:27<1:03:14,  1.71s/it]
                                                         
{'loss': 0.0698, 'learning_rate': 5e-05, 'epoch': 4.55}

 91%|█████████ | 22205/24420 [9:16:27<1:03:14,  1.71s/it]
 91%|█████████ | 22206/24420 [9:16:29<1:00:52,  1.65s/it]
 91%|█████████ | 22207/24420 [9:16:30<59:08,  1.60s/it]  
 91%|█████████ | 22208/24420 [9:16:32<57:55,  1.57s/it]
 91%|█████████ | 22209/24420 [9:16:33<57:01,  1.55s/it]
 91%|█████████ | 22210/24420 [9:16:35<56:16,  1.53s/it]
                                                       
{'loss': 0.084, 'learning_rate': 5e-05, 'epoch': 4.55}

 91%|█████████ | 22210/24420 [9:16:35<56:16,  1.53s/it]
 91%|█████████ | 22211/24420 [9:16:36<55:48,  1.52s/it]
 91%|█████████ | 22212/24420 [9:16:38<55:31,  1.51s/it]
 91%|█████████ | 22213/24420 [9:16:39<55:18,  1.50s/it]
 91%|█████████ | 22214/24420 [9:16:41<55:07,  1.50s/it]
 91%|█████████ | 22215/24420 [9:16:42<54:57,  1.50s/it]
                                                       
{'loss': 0.0752, 'learning_rate': 5e-05, 'epoch': 4.55}

 91%|█████████ | 22215/24420 [9:16:42<54:57,  1.50s/it]
 91%|█████████ | 22216/24420 [9:16:44<54:52,  1.49s/it]
 91%|█████████ | 22217/24420 [9:16:45<54:45,  1.49s/it]
 91%|█████████ | 22218/24420 [9:16:47<54:45,  1.49s/it]
 91%|█████████ | 22219/24420 [9:16:48<54:44,  1.49s/it]
 91%|█████████ | 22220/24420 [9:16:50<54:39,  1.49s/it]
                                                       
{'loss': 0.077, 'learning_rate': 5e-05, 'epoch': 4.55}

 91%|█████████ | 22220/24420 [9:16:50<54:39,  1.49s/it]
 91%|█████████ | 22221/24420 [9:16:51<54:34,  1.49s/it]
 91%|█████████ | 22222/24420 [9:16:53<54:31,  1.49s/it]
 91%|█████████ | 22223/24420 [9:16:54<54:29,  1.49s/it]
 91%|█████████ | 22224/24420 [9:16:56<54:26,  1.49s/it]
 91%|█████████ | 22225/24420 [9:16:57<54:24,  1.49s/it]
                                                       
{'loss': 0.0612, 'learning_rate': 5e-05, 'epoch': 4.55}

 91%|█████████ | 22225/24420 [9:16:57<54:24,  1.49s/it]
 91%|█████████ | 22226/24420 [9:16:59<54:21,  1.49s/it]
 91%|█████████ | 22227/24420 [9:17:00<54:21,  1.49s/it]
 91%|█████████ | 22228/24420 [9:17:01<54:22,  1.49s/it]
 91%|█████████ | 22229/24420 [9:17:03<54:22,  1.49s/it]
 91%|█████████ | 22230/24420 [9:17:04<54:24,  1.49s/it]
                                                       
{'loss': 0.0699, 'learning_rate': 5e-05, 'epoch': 4.55}

 91%|█████████ | 22230/24420 [9:17:04<54:24,  1.49s/it]
 91%|█████████ | 22231/24420 [9:17:06<54:25,  1.49s/it]
 91%|█████████ | 22232/24420 [9:17:07<54:18,  1.49s/it]
 91%|█████████ | 22233/24420 [9:17:09<54:14,  1.49s/it]
 91%|█████████ | 22234/24420 [9:17:10<54:09,  1.49s/it]
 91%|█████████ | 22235/24420 [9:17:12<54:04,  1.48s/it]
                                                       
{'loss': 0.0769, 'learning_rate': 5e-05, 'epoch': 4.55}

 91%|█████████ | 22235/24420 [9:17:12<54:04,  1.48s/it]
 91%|█████████ | 22236/24420 [9:17:13<54:03,  1.49s/it]
 91%|█████████ | 22237/24420 [9:17:15<53:58,  1.48s/it]
 91%|█████████ | 22238/24420 [9:17:16<53:58,  1.48s/it]
 91%|█████████ | 22239/24420 [9:17:18<54:01,  1.49s/it]
 91%|█████████ | 22240/24420 [9:17:19<54:00,  1.49s/it]
                                                       
{'loss': 0.0742, 'learning_rate': 5e-05, 'epoch': 4.55}

 91%|█████████ | 22240/24420 [9:17:19<54:00,  1.49s/it]
 91%|█████████ | 22241/24420 [9:17:21<54:01,  1.49s/it]
 91%|█████████ | 22242/24420 [9:17:22<53:59,  1.49s/it]
 91%|█████████ | 22243/24420 [9:17:24<54:00,  1.49s/it]
 91%|█████████ | 22244/24420 [9:17:25<53:56,  1.49s/it]
 91%|█████████ | 22245/24420 [9:17:27<53:51,  1.49s/it]
                                                       
{'loss': 0.0816, 'learning_rate': 5e-05, 'epoch': 4.55}

 91%|█████████ | 22245/24420 [9:17:27<53:51,  1.49s/it]
 91%|█████████ | 22246/24420 [9:17:28<53:57,  1.49s/it]
 91%|█████████ | 22247/24420 [9:17:30<53:49,  1.49s/it]
 91%|█████████ | 22248/24420 [9:17:31<53:46,  1.49s/it]
 91%|█████████ | 22249/24420 [9:17:33<53:49,  1.49s/it]
 91%|█████████ | 22250/24420 [9:17:34<53:41,  1.48s/it]
                                                       
{'loss': 0.0699, 'learning_rate': 5e-05, 'epoch': 4.56}

 91%|█████████ | 22250/24420 [9:17:34<53:41,  1.48s/it]
 91%|█████████ | 22251/24420 [9:17:36<53:47,  1.49s/it]
 91%|█████████ | 22252/24420 [9:17:37<53:50,  1.49s/it]
 91%|█████████ | 22253/24420 [9:17:39<53:49,  1.49s/it]
 91%|█████████ | 22254/24420 [9:17:40<53:46,  1.49s/it]
 91%|█████████ | 22255/24420 [9:17:42<53:41,  1.49s/it]
                                                       
{'loss': 0.0613, 'learning_rate': 5e-05, 'epoch': 4.56}

 91%|█████████ | 22255/24420 [9:17:42<53:41,  1.49s/it]
 91%|█████████ | 22256/24420 [9:17:43<53:48,  1.49s/it]
 91%|█████████ | 22257/24420 [9:17:45<53:41,  1.49s/it]
 91%|█████████ | 22258/24420 [9:17:46<53:38,  1.49s/it]
 91%|█████████ | 22259/24420 [9:17:48<53:44,  1.49s/it]
 91%|█████████ | 22260/24420 [9:17:49<53:41,  1.49s/it]
                                                       
{'loss': 0.0761, 'learning_rate': 5e-05, 'epoch': 4.56}

 91%|█████████ | 22260/24420 [9:17:49<53:41,  1.49s/it]
 91%|█████████ | 22261/24420 [9:17:51<53:43,  1.49s/it]
 91%|█████████ | 22262/24420 [9:17:52<53:35,  1.49s/it]
 91%|█████████ | 22263/24420 [9:17:54<53:34,  1.49s/it]
 91%|█████████ | 22264/24420 [9:17:55<53:35,  1.49s/it]
 91%|█████████ | 22265/24420 [9:17:57<53:38,  1.49s/it]
                                                       
{'loss': 0.069, 'learning_rate': 5e-05, 'epoch': 4.56}

 91%|█████████ | 22265/24420 [9:17:57<53:38,  1.49s/it]
 91%|█████████ | 22266/24420 [9:17:58<53:36,  1.49s/it]
 91%|█████████ | 22267/24420 [9:18:00<53:38,  1.49s/it]
 91%|█████████ | 22268/24420 [9:18:01<53:38,  1.50s/it]
 91%|█████████ | 22269/24420 [9:18:03<53:33,  1.49s/it]
 91%|█████████ | 22270/24420 [9:18:04<53:30,  1.49s/it]
                                                       
{'loss': 0.0661, 'learning_rate': 5e-05, 'epoch': 4.56}

 91%|█████████ | 22270/24420 [9:18:04<53:30,  1.49s/it]
 91%|█████████ | 22271/24420 [9:18:06<53:25,  1.49s/it]
 91%|█████████ | 22272/24420 [9:18:07<53:21,  1.49s/it]
 91%|█████████ | 22273/24420 [9:18:09<53:22,  1.49s/it]
 91%|█████████ | 22274/24420 [9:18:10<53:17,  1.49s/it]
 91%|█████████ | 22275/24420 [9:18:11<53:11,  1.49s/it]
                                                       
{'loss': 0.0762, 'learning_rate': 5e-05, 'epoch': 4.56}

 91%|█████████ | 22275/24420 [9:18:11<53:11,  1.49s/it]
 91%|█████████ | 22276/24420 [9:18:13<53:11,  1.49s/it]
 91%|█████████ | 22277/24420 [9:18:14<53:10,  1.49s/it]
 91%|█████████ | 22278/24420 [9:18:16<53:12,  1.49s/it]
 91%|█████████ | 22279/24420 [9:18:17<53:12,  1.49s/it]
 91%|█████████ | 22280/24420 [9:18:19<53:15,  1.49s/it]
                                                       
{'loss': 0.0691, 'learning_rate': 5e-05, 'epoch': 4.56}

 91%|█████████ | 22280/24420 [9:18:19<53:15,  1.49s/it]
 91%|█████████ | 22281/24420 [9:18:20<53:12,  1.49s/it]
 91%|█████████ | 22282/24420 [9:18:22<53:14,  1.49s/it]
 91%|█████████ | 22283/24420 [9:18:23<53:10,  1.49s/it]
 91%|█████████▏| 22284/24420 [9:18:25<53:04,  1.49s/it]
 91%|█████████▏| 22285/24420 [9:18:26<52:57,  1.49s/it]
                                                       
{'loss': 0.0808, 'learning_rate': 5e-05, 'epoch': 4.56}

 91%|█████████▏| 22285/24420 [9:18:26<52:57,  1.49s/it]
 91%|█████████▏| 22286/24420 [9:18:28<53:00,  1.49s/it]
 91%|█████████▏| 22287/24420 [9:18:29<53:02,  1.49s/it]
 91%|█████████▏| 22288/24420 [9:18:31<52:57,  1.49s/it]
 91%|█████████▏| 22289/24420 [9:18:32<52:53,  1.49s/it]
 91%|█████████▏| 22290/24420 [9:18:34<52:51,  1.49s/it]
                                                       
{'loss': 0.0755, 'learning_rate': 5e-05, 'epoch': 4.56}

 91%|█████████▏| 22290/24420 [9:18:34<52:51,  1.49s/it]
 91%|█████████▏| 22291/24420 [9:18:35<52:51,  1.49s/it]
 91%|█████████▏| 22292/24420 [9:18:37<52:45,  1.49s/it]
 91%|█████████▏| 22293/24420 [9:18:38<52:45,  1.49s/it]
 91%|█████████▏| 22294/24420 [9:18:40<52:42,  1.49s/it]
 91%|█████████▏| 22295/24420 [9:18:41<52:39,  1.49s/it]
                                                       
{'loss': 0.0721, 'learning_rate': 5e-05, 'epoch': 4.56}

 91%|█████████▏| 22295/24420 [9:18:41<52:39,  1.49s/it]
 91%|█████████▏| 22296/24420 [9:18:43<52:37,  1.49s/it]
 91%|█████████▏| 22297/24420 [9:18:44<52:35,  1.49s/it]
 91%|█████████▏| 22298/24420 [9:18:46<52:36,  1.49s/it]
 91%|█████████▏| 22299/24420 [9:18:47<52:39,  1.49s/it]
 91%|█████████▏| 22300/24420 [9:18:49<52:30,  1.49s/it]
                                                       
{'loss': 0.0809, 'learning_rate': 5e-05, 'epoch': 4.57}

 91%|█████████▏| 22300/24420 [9:18:49<52:30,  1.49s/it]
 91%|█████████▏| 22301/24420 [9:18:50<52:36,  1.49s/it]
 91%|█████████▏| 22302/24420 [9:18:52<52:32,  1.49s/it]
 91%|█████████▏| 22303/24420 [9:18:53<52:27,  1.49s/it]
 91%|█████████▏| 22304/24420 [9:18:55<52:24,  1.49s/it]
 91%|█████████▏| 22305/24420 [9:18:56<52:30,  1.49s/it]
                                                       
{'loss': 0.0631, 'learning_rate': 5e-05, 'epoch': 4.57}

 91%|█████████▏| 22305/24420 [9:18:56<52:30,  1.49s/it]
 91%|█████████▏| 22306/24420 [9:18:58<52:30,  1.49s/it]
 91%|█████████▏| 22307/24420 [9:18:59<52:23,  1.49s/it]
 91%|█████████▏| 22308/24420 [9:19:01<52:28,  1.49s/it]
 91%|█████████▏| 22309/24420 [9:19:02<52:28,  1.49s/it]
 91%|█████████▏| 22310/24420 [9:19:04<52:23,  1.49s/it]
                                                       
{'loss': 0.0758, 'learning_rate': 5e-05, 'epoch': 4.57}

 91%|█████████▏| 22310/24420 [9:19:04<52:23,  1.49s/it]
 91%|█████████▏| 22311/24420 [9:19:05<52:18,  1.49s/it]
 91%|█████████▏| 22312/24420 [9:19:07<52:10,  1.49s/it]
 91%|█████████▏| 22313/24420 [9:19:08<52:12,  1.49s/it]
 91%|█████████▏| 22314/24420 [9:19:10<52:14,  1.49s/it]
 91%|█████████▏| 22315/24420 [9:19:11<52:17,  1.49s/it]
                                                       
{'loss': 0.0621, 'learning_rate': 5e-05, 'epoch': 4.57}

 91%|█████████▏| 22315/24420 [9:19:11<52:17,  1.49s/it]
 91%|█████████▏| 22316/24420 [9:19:13<52:15,  1.49s/it]
 91%|█████████▏| 22317/24420 [9:19:14<52:13,  1.49s/it]
 91%|█████████▏| 22318/24420 [9:19:16<52:11,  1.49s/it]
 91%|█████████▏| 22319/24420 [9:19:17<52:05,  1.49s/it]
 91%|█████████▏| 22320/24420 [9:19:18<52:04,  1.49s/it]
                                                       
{'loss': 0.0592, 'learning_rate': 5e-05, 'epoch': 4.57}

 91%|█████████▏| 22320/24420 [9:19:18<52:04,  1.49s/it]
 91%|█████████▏| 22321/24420 [9:19:20<52:09,  1.49s/it]
 91%|█████████▏| 22322/24420 [9:19:21<52:06,  1.49s/it]
 91%|█████████▏| 22323/24420 [9:19:23<52:06,  1.49s/it]
 91%|█████████▏| 22324/24420 [9:19:24<52:09,  1.49s/it]
 91%|█████████▏| 22325/24420 [9:19:26<52:07,  1.49s/it]
                                                       
{'loss': 0.0699, 'learning_rate': 5e-05, 'epoch': 4.57}

 91%|█████████▏| 22325/24420 [9:19:26<52:07,  1.49s/it]
 91%|█████████▏| 22326/24420 [9:19:27<52:04,  1.49s/it]
 91%|█████████▏| 22327/24420 [9:19:29<52:00,  1.49s/it]
 91%|█████████▏| 22328/24420 [9:19:30<51:57,  1.49s/it]
 91%|█████████▏| 22329/24420 [9:19:32<51:58,  1.49s/it]
 91%|█████████▏| 22330/24420 [9:19:33<51:59,  1.49s/it]
                                                       
{'loss': 0.069, 'learning_rate': 5e-05, 'epoch': 4.57}

 91%|█████████▏| 22330/24420 [9:19:33<51:59,  1.49s/it]
 91%|█████████▏| 22331/24420 [9:19:35<51:51,  1.49s/it]
 91%|█████████▏| 22332/24420 [9:19:36<51:52,  1.49s/it]
 91%|█████████▏| 22333/24420 [9:19:38<51:56,  1.49s/it]
 91%|█████████▏| 22334/24420 [9:19:39<51:49,  1.49s/it]
 91%|█████████▏| 22335/24420 [9:19:41<51:44,  1.49s/it]
                                                       
{'loss': 0.0689, 'learning_rate': 5e-05, 'epoch': 4.57}

 91%|█████████▏| 22335/24420 [9:19:41<51:44,  1.49s/it]
 91%|█████████▏| 22336/24420 [9:19:42<51:38,  1.49s/it]
 91%|█████████▏| 22337/24420 [9:19:44<51:36,  1.49s/it]
 91%|█████████▏| 22338/24420 [9:19:45<51:41,  1.49s/it]
 91%|█████████▏| 22339/24420 [9:19:47<51:39,  1.49s/it]
 91%|█████████▏| 22340/24420 [9:19:48<51:33,  1.49s/it]
                                                       
{'loss': 0.0671, 'learning_rate': 5e-05, 'epoch': 4.57}

 91%|█████████▏| 22340/24420 [9:19:48<51:33,  1.49s/it]
 91%|█████████▏| 22341/24420 [9:19:50<51:32,  1.49s/it]
 91%|█████████▏| 22342/24420 [9:19:51<51:27,  1.49s/it]
 91%|█████████▏| 22343/24420 [9:19:53<51:26,  1.49s/it]
 91%|█████████▏| 22344/24420 [9:19:54<51:28,  1.49s/it]
 92%|█████████▏| 22345/24420 [9:19:56<51:23,  1.49s/it]
                                                       
{'loss': 0.0701, 'learning_rate': 5e-05, 'epoch': 4.58}

 92%|█████████▏| 22345/24420 [9:19:56<51:23,  1.49s/it]
 92%|█████████▏| 22346/24420 [9:19:57<51:23,  1.49s/it]
 92%|█████████▏| 22347/24420 [9:19:59<51:26,  1.49s/it]
 92%|█████████▏| 22348/24420 [9:20:00<51:23,  1.49s/it]
 92%|█████████▏| 22349/24420 [9:20:02<51:22,  1.49s/it]
 92%|█████████▏| 22350/24420 [9:20:03<51:23,  1.49s/it]
                                                       
{'loss': 0.0788, 'learning_rate': 5e-05, 'epoch': 4.58}

 92%|█████████▏| 22350/24420 [9:20:03<51:23,  1.49s/it]
 92%|█████████▏| 22351/24420 [9:20:05<51:22,  1.49s/it]
 92%|█████████▏| 22352/24420 [9:20:06<51:20,  1.49s/it]
 92%|█████████▏| 22353/24420 [9:20:08<51:15,  1.49s/it]
 92%|█████████▏| 22354/24420 [9:20:09<51:22,  1.49s/it]
 92%|█████████▏| 22355/24420 [9:20:11<51:21,  1.49s/it]
                                                       
{'loss': 0.069, 'learning_rate': 5e-05, 'epoch': 4.58}

 92%|█████████▏| 22355/24420 [9:20:11<51:21,  1.49s/it]
 92%|█████████▏| 22356/24420 [9:20:12<51:21,  1.49s/it]
 92%|█████████▏| 22357/24420 [9:20:14<51:16,  1.49s/it]
 92%|█████████▏| 22358/24420 [9:20:15<51:15,  1.49s/it]
 92%|█████████▏| 22359/24420 [9:20:17<51:20,  1.49s/it]
 92%|█████████▏| 22360/24420 [9:20:18<51:19,  1.49s/it]
                                                       
{'loss': 0.0695, 'learning_rate': 5e-05, 'epoch': 4.58}

 92%|█████████▏| 22360/24420 [9:20:18<51:19,  1.49s/it]
 92%|█████████▏| 22361/24420 [9:20:20<51:21,  1.50s/it]
 92%|█████████▏| 22362/24420 [9:20:21<51:19,  1.50s/it]
 92%|█████████▏| 22363/24420 [9:20:23<51:11,  1.49s/it]
 92%|█████████▏| 22364/24420 [9:20:24<51:10,  1.49s/it]
 92%|█████████▏| 22365/24420 [9:20:26<51:04,  1.49s/it]
                                                       
{'loss': 0.0665, 'learning_rate': 5e-05, 'epoch': 4.58}

 92%|█████████▏| 22365/24420 [9:20:26<51:04,  1.49s/it]
 92%|█████████▏| 22366/24420 [9:20:27<51:04,  1.49s/it]
 92%|█████████▏| 22367/24420 [9:20:29<50:57,  1.49s/it]
 92%|█████████▏| 22368/24420 [9:20:30<50:56,  1.49s/it]
 92%|█████████▏| 22369/24420 [9:20:32<50:52,  1.49s/it]
 92%|█████████▏| 22370/24420 [9:20:33<50:56,  1.49s/it]
                                                       
{'loss': 0.0634, 'learning_rate': 5e-05, 'epoch': 4.58}

 92%|█████████▏| 22370/24420 [9:20:33<50:56,  1.49s/it]
 92%|█████████▏| 22371/24420 [9:20:35<50:53,  1.49s/it]
 92%|█████████▏| 22372/24420 [9:20:36<50:55,  1.49s/it]
 92%|█████████▏| 22373/24420 [9:20:37<50:49,  1.49s/it]
 92%|█████████▏| 22374/24420 [9:20:39<50:50,  1.49s/it]
 92%|█████████▏| 22375/24420 [9:20:40<50:46,  1.49s/it]
                                                       
{'loss': 0.0644, 'learning_rate': 5e-05, 'epoch': 4.58}

 92%|█████████▏| 22375/24420 [9:20:40<50:46,  1.49s/it]
 92%|█████████▏| 22376/24420 [9:20:42<50:43,  1.49s/it]
 92%|█████████▏| 22377/24420 [9:20:43<50:44,  1.49s/it]
 92%|█████████▏| 22378/24420 [9:20:45<50:39,  1.49s/it]
 92%|█████████▏| 22379/24420 [9:20:46<50:37,  1.49s/it]
 92%|█████████▏| 22380/24420 [9:20:48<50:42,  1.49s/it]
                                                       
{'loss': 0.0773, 'learning_rate': 5e-05, 'epoch': 4.58}

 92%|█████████▏| 22380/24420 [9:20:48<50:42,  1.49s/it]
 92%|█████████▏| 22381/24420 [9:20:49<50:36,  1.49s/it]
 92%|█████████▏| 22382/24420 [9:20:51<50:34,  1.49s/it]
 92%|█████████▏| 22383/24420 [9:20:52<50:28,  1.49s/it]
 92%|█████████▏| 22384/24420 [9:20:54<50:27,  1.49s/it]
 92%|█████████▏| 22385/24420 [9:20:55<50:23,  1.49s/it]
                                                       
{'loss': 0.0706, 'learning_rate': 5e-05, 'epoch': 4.58}

 92%|█████████▏| 22385/24420 [9:20:55<50:23,  1.49s/it]
 92%|█████████▏| 22386/24420 [9:20:57<50:20,  1.49s/it]
 92%|█████████▏| 22387/24420 [9:20:58<50:16,  1.48s/it]
 92%|█████████▏| 22388/24420 [9:21:00<50:23,  1.49s/it]
 92%|█████████▏| 22389/24420 [9:21:01<50:20,  1.49s/it]
 92%|█████████▏| 22390/24420 [9:21:03<50:22,  1.49s/it]
                                                       
{'loss': 0.0619, 'learning_rate': 5e-05, 'epoch': 4.58}

 92%|█████████▏| 22390/24420 [9:21:03<50:22,  1.49s/it]
 92%|█████████▏| 22391/24420 [9:21:04<50:21,  1.49s/it]
 92%|█████████▏| 22392/24420 [9:21:06<50:16,  1.49s/it]
 92%|█████████▏| 22393/24420 [9:21:07<50:16,  1.49s/it]
 92%|█████████▏| 22394/24420 [9:21:09<50:17,  1.49s/it]
 92%|█████████▏| 22395/24420 [9:21:10<50:07,  1.49s/it]
                                                       
{'loss': 0.0657, 'learning_rate': 5e-05, 'epoch': 4.59}

 92%|█████████▏| 22395/24420 [9:21:10<50:07,  1.49s/it]
 92%|█████████▏| 22396/24420 [9:21:12<50:09,  1.49s/it]
 92%|█████████▏| 22397/24420 [9:21:13<50:05,  1.49s/it]
 92%|█████████▏| 22398/24420 [9:21:15<50:01,  1.48s/it]
 92%|█████████▏| 22399/24420 [9:21:16<49:57,  1.48s/it]
 92%|█████████▏| 22400/24420 [9:21:18<49:58,  1.48s/it]
                                                       
{'loss': 0.0724, 'learning_rate': 5e-05, 'epoch': 4.59}

 92%|█████████▏| 22400/24420 [9:21:18<49:58,  1.48s/it]
 92%|█████████▏| 22401/24420 [9:21:22<1:21:54,  2.43s/it]
 92%|█████████▏| 22402/24420 [9:21:24<1:12:28,  2.15s/it]
 92%|█████████▏| 22403/24420 [9:21:25<1:05:46,  1.96s/it]
 92%|█████████▏| 22404/24420 [9:21:27<1:00:56,  1.81s/it]
 92%|█████████▏| 22405/24420 [9:21:28<57:37,  1.72s/it]  
                                                       
{'loss': 0.0693, 'learning_rate': 5e-05, 'epoch': 4.59}

 92%|█████████▏| 22405/24420 [9:21:28<57:37,  1.72s/it]
 92%|█████████▏| 22406/24420 [9:21:30<55:28,  1.65s/it]
 92%|█████████▏| 22407/24420 [9:21:31<53:46,  1.60s/it]
 92%|█████████▏| 22408/24420 [9:21:33<52:35,  1.57s/it]
 92%|█████████▏| 22409/24420 [9:21:34<51:43,  1.54s/it]
 92%|█████████▏| 22410/24420 [9:21:36<51:11,  1.53s/it]
                                                       
{'loss': 0.0809, 'learning_rate': 5e-05, 'epoch': 4.59}

 92%|█████████▏| 22410/24420 [9:21:36<51:11,  1.53s/it]
 92%|█████████▏| 22411/24420 [9:21:37<50:42,  1.51s/it]
 92%|█████████▏| 22412/24420 [9:21:39<50:23,  1.51s/it]
 92%|█████████▏| 22413/24420 [9:21:40<50:05,  1.50s/it]
 92%|█████████▏| 22414/24420 [9:21:42<49:58,  1.49s/it]
 92%|█████████▏| 22415/24420 [9:21:43<49:54,  1.49s/it]
                                                       
{'loss': 0.0721, 'learning_rate': 5e-05, 'epoch': 4.59}

 92%|█████████▏| 22415/24420 [9:21:43<49:54,  1.49s/it]
 92%|█████████▏| 22416/24420 [9:21:45<49:47,  1.49s/it]
 92%|█████████▏| 22417/24420 [9:21:46<49:40,  1.49s/it]
 92%|█████████▏| 22418/24420 [9:21:48<49:34,  1.49s/it]
 92%|█████████▏| 22419/24420 [9:21:49<49:36,  1.49s/it]
 92%|█████████▏| 22420/24420 [9:21:51<49:32,  1.49s/it]
                                                       
{'loss': 0.0609, 'learning_rate': 5e-05, 'epoch': 4.59}

 92%|█████████▏| 22420/24420 [9:21:51<49:32,  1.49s/it]
 92%|█████████▏| 22421/24420 [9:21:52<49:35,  1.49s/it]
 92%|█████████▏| 22422/24420 [9:21:54<49:34,  1.49s/it]
 92%|█████████▏| 22423/24420 [9:21:55<49:37,  1.49s/it]
 92%|█████████▏| 22424/24420 [9:21:57<49:45,  1.50s/it]
 92%|█████████▏| 22425/24420 [9:21:58<49:37,  1.49s/it]
                                                       
{'loss': 0.0853, 'learning_rate': 5e-05, 'epoch': 4.59}

 92%|█████████▏| 22425/24420 [9:21:58<49:37,  1.49s/it]
 92%|█████████▏| 22426/24420 [9:22:00<49:34,  1.49s/it]
 92%|█████████▏| 22427/24420 [9:22:01<49:35,  1.49s/it]
 92%|█████████▏| 22428/24420 [9:22:03<49:40,  1.50s/it]
 92%|█████████▏| 22429/24420 [9:22:04<49:31,  1.49s/it]
 92%|█████████▏| 22430/24420 [9:22:05<49:32,  1.49s/it]
                                                       
{'loss': 0.085, 'learning_rate': 5e-05, 'epoch': 4.59}

 92%|█████████▏| 22430/24420 [9:22:05<49:32,  1.49s/it]
 92%|█████████▏| 22431/24420 [9:22:07<49:27,  1.49s/it]
 92%|█████████▏| 22432/24420 [9:22:08<49:21,  1.49s/it]
 92%|█████████▏| 22433/24420 [9:22:10<49:24,  1.49s/it]
 92%|█████████▏| 22434/24420 [9:22:11<49:21,  1.49s/it]
 92%|█████████▏| 22435/24420 [9:22:13<49:11,  1.49s/it]
                                                       
{'loss': 0.0725, 'learning_rate': 5e-05, 'epoch': 4.59}

 92%|█████████▏| 22435/24420 [9:22:13<49:11,  1.49s/it]
 92%|█████████▏| 22436/24420 [9:22:14<49:08,  1.49s/it]
 92%|█████████▏| 22437/24420 [9:22:16<49:05,  1.49s/it]
 92%|█████████▏| 22438/24420 [9:22:17<49:05,  1.49s/it]
 92%|█████████▏| 22439/24420 [9:22:19<49:02,  1.49s/it]
 92%|█████████▏| 22440/24420 [9:22:20<49:02,  1.49s/it]
                                                       
{'loss': 0.0842, 'learning_rate': 5e-05, 'epoch': 4.59}

 92%|█████████▏| 22440/24420 [9:22:20<49:02,  1.49s/it]
 92%|█████████▏| 22441/24420 [9:22:22<49:08,  1.49s/it]
 92%|█████████▏| 22442/24420 [9:22:23<49:15,  1.49s/it]
 92%|█████████▏| 22443/24420 [9:22:25<49:06,  1.49s/it]
 92%|█████████▏| 22444/24420 [9:22:26<49:07,  1.49s/it]
 92%|█████████▏| 22445/24420 [9:22:28<49:00,  1.49s/it]
                                                       
{'loss': 0.069, 'learning_rate': 5e-05, 'epoch': 4.6}

 92%|█████████▏| 22445/24420 [9:22:28<49:00,  1.49s/it]
 92%|█████████▏| 22446/24420 [9:22:29<48:57,  1.49s/it]
 92%|█████████▏| 22447/24420 [9:22:31<48:56,  1.49s/it]
 92%|█████████▏| 22448/24420 [9:22:32<48:53,  1.49s/it]
 92%|█████████▏| 22449/24420 [9:22:34<48:52,  1.49s/it]
 92%|█████████▏| 22450/24420 [9:22:35<48:52,  1.49s/it]
                                                       
{'loss': 0.0689, 'learning_rate': 5e-05, 'epoch': 4.6}

 92%|█████████▏| 22450/24420 [9:22:35<48:52,  1.49s/it]
 92%|█████████▏| 22451/24420 [9:22:37<48:53,  1.49s/it]
 92%|█████████▏| 22452/24420 [9:22:38<48:49,  1.49s/it]
 92%|█████████▏| 22453/24420 [9:22:40<48:47,  1.49s/it]
 92%|█████████▏| 22454/24420 [9:22:41<48:49,  1.49s/it]
 92%|█████████▏| 22455/24420 [9:22:43<48:46,  1.49s/it]
                                                       
{'loss': 0.0761, 'learning_rate': 5e-05, 'epoch': 4.6}

 92%|█████████▏| 22455/24420 [9:22:43<48:46,  1.49s/it]
 92%|█████████▏| 22456/24420 [9:22:44<48:46,  1.49s/it]
 92%|█████████▏| 22457/24420 [9:22:46<48:51,  1.49s/it]
 92%|█████████▏| 22458/24420 [9:22:47<48:48,  1.49s/it]
 92%|█████████▏| 22459/24420 [9:22:49<48:46,  1.49s/it]
 92%|█████████▏| 22460/24420 [9:22:50<48:40,  1.49s/it]
                                                       
{'loss': 0.0815, 'learning_rate': 5e-05, 'epoch': 4.6}

 92%|█████████▏| 22460/24420 [9:22:50<48:40,  1.49s/it]
 92%|█████████▏| 22461/24420 [9:22:52<48:37,  1.49s/it]
 92%|█████████▏| 22462/24420 [9:22:53<48:38,  1.49s/it]
 92%|█████████▏| 22463/24420 [9:22:55<48:33,  1.49s/it]
 92%|█████████▏| 22464/24420 [9:22:56<48:30,  1.49s/it]
 92%|█████████▏| 22465/24420 [9:22:58<48:29,  1.49s/it]
                                                       
{'loss': 0.0641, 'learning_rate': 5e-05, 'epoch': 4.6}

 92%|█████████▏| 22465/24420 [9:22:58<48:29,  1.49s/it]
 92%|█████████▏| 22466/24420 [9:22:59<48:26,  1.49s/it]
 92%|█████████▏| 22467/24420 [9:23:01<48:24,  1.49s/it]
 92%|█████████▏| 22468/24420 [9:23:02<48:27,  1.49s/it]
 92%|█████████▏| 22469/24420 [9:23:04<48:29,  1.49s/it]
 92%|█████████▏| 22470/24420 [9:23:05<48:23,  1.49s/it]
                                                       
{'loss': 0.0674, 'learning_rate': 5e-05, 'epoch': 4.6}

 92%|█████████▏| 22470/24420 [9:23:05<48:23,  1.49s/it]
 92%|█████████▏| 22471/24420 [9:23:07<48:30,  1.49s/it]
 92%|█████████▏| 22472/24420 [9:23:08<48:31,  1.49s/it]
 92%|█████████▏| 22473/24420 [9:23:10<48:30,  1.49s/it]
 92%|█████████▏| 22474/24420 [9:23:11<48:29,  1.50s/it]
 92%|█████████▏| 22475/24420 [9:23:13<48:20,  1.49s/it]
                                                       
{'loss': 0.0785, 'learning_rate': 5e-05, 'epoch': 4.6}

 92%|█████████▏| 22475/24420 [9:23:13<48:20,  1.49s/it]
 92%|█████████▏| 22476/24420 [9:23:14<48:14,  1.49s/it]
 92%|█████████▏| 22477/24420 [9:23:16<48:13,  1.49s/it]
 92%|█████████▏| 22478/24420 [9:23:17<48:15,  1.49s/it]
 92%|█████████▏| 22479/24420 [9:23:18<48:14,  1.49s/it]
 92%|█████████▏| 22480/24420 [9:23:20<48:15,  1.49s/it]
                                                       
{'loss': 0.0748, 'learning_rate': 5e-05, 'epoch': 4.6}

 92%|█████████▏| 22480/24420 [9:23:20<48:15,  1.49s/it]
 92%|█████████▏| 22481/24420 [9:23:21<48:15,  1.49s/it]
 92%|█████████▏| 22482/24420 [9:23:23<48:11,  1.49s/it]
 92%|█████████▏| 22483/24420 [9:23:24<48:06,  1.49s/it]
 92%|█████████▏| 22484/24420 [9:23:26<48:05,  1.49s/it]
 92%|█████████▏| 22485/24420 [9:23:27<47:59,  1.49s/it]
                                                       
{'loss': 0.0649, 'learning_rate': 5e-05, 'epoch': 4.6}

 92%|█████████▏| 22485/24420 [9:23:27<47:59,  1.49s/it]
 92%|█████████▏| 22486/24420 [9:23:29<47:59,  1.49s/it]
 92%|█████████▏| 22487/24420 [9:23:30<47:54,  1.49s/it]
 92%|█████████▏| 22488/24420 [9:23:32<48:00,  1.49s/it]
 92%|█████████▏| 22489/24420 [9:23:33<47:55,  1.49s/it]
 92%|█████████▏| 22490/24420 [9:23:35<47:51,  1.49s/it]
                                                       
{'loss': 0.0717, 'learning_rate': 5e-05, 'epoch': 4.6}

 92%|█████████▏| 22490/24420 [9:23:35<47:51,  1.49s/it]
 92%|█████████▏| 22491/24420 [9:23:36<47:47,  1.49s/it]
 92%|█████████▏| 22492/24420 [9:23:38<47:46,  1.49s/it]
 92%|█████████▏| 22493/24420 [9:23:39<47:48,  1.49s/it]
 92%|█████████▏| 22494/24420 [9:23:41<47:49,  1.49s/it]
 92%|█████████▏| 22495/24420 [9:23:42<47:58,  1.50s/it]
                                                       
{'loss': 0.068, 'learning_rate': 5e-05, 'epoch': 4.61}

 92%|█████████▏| 22495/24420 [9:23:42<47:58,  1.50s/it]
 92%|█████████▏| 22496/24420 [9:23:44<47:55,  1.49s/it]
 92%|█████████▏| 22497/24420 [9:23:45<47:55,  1.50s/it]
 92%|█████████▏| 22498/24420 [9:23:47<47:53,  1.50s/it]
 92%|█████████▏| 22499/24420 [9:23:48<47:43,  1.49s/it]
 92%|█████████▏| 22500/24420 [9:23:50<47:37,  1.49s/it]
                                                       
{'loss': 0.0733, 'learning_rate': 5e-05, 'epoch': 4.61}

 92%|█████████▏| 22500/24420 [9:23:50<47:37,  1.49s/it]
 92%|█████████▏| 22501/24420 [9:23:51<47:32,  1.49s/it]
 92%|█████████▏| 22502/24420 [9:23:53<47:29,  1.49s/it]
 92%|█████████▏| 22503/24420 [9:23:54<47:33,  1.49s/it]
 92%|█████████▏| 22504/24420 [9:23:56<47:32,  1.49s/it]
 92%|█████████▏| 22505/24420 [9:23:57<47:28,  1.49s/it]
                                                       
{'loss': 0.0749, 'learning_rate': 5e-05, 'epoch': 4.61}

 92%|█████████▏| 22505/24420 [9:23:57<47:28,  1.49s/it]
 92%|█████████▏| 22506/24420 [9:23:59<47:29,  1.49s/it]
 92%|█████████▏| 22507/24420 [9:24:00<47:25,  1.49s/it]
 92%|█████████▏| 22508/24420 [9:24:02<47:24,  1.49s/it]
 92%|█████████▏| 22509/24420 [9:24:03<47:21,  1.49s/it]
 92%|█████████▏| 22510/24420 [9:24:05<47:24,  1.49s/it]
                                                       
{'loss': 0.0735, 'learning_rate': 5e-05, 'epoch': 4.61}

 92%|█████████▏| 22510/24420 [9:24:05<47:24,  1.49s/it]
 92%|█████████▏| 22511/24420 [9:24:06<47:27,  1.49s/it]
 92%|█████████▏| 22512/24420 [9:24:08<47:27,  1.49s/it]
 92%|█████████▏| 22513/24420 [9:24:09<47:29,  1.49s/it]
 92%|█████████▏| 22514/24420 [9:24:11<47:33,  1.50s/it]
 92%|█████████▏| 22515/24420 [9:24:12<47:25,  1.49s/it]
                                                       
{'loss': 0.0727, 'learning_rate': 5e-05, 'epoch': 4.61}

 92%|█████████▏| 22515/24420 [9:24:12<47:25,  1.49s/it]
 92%|█████████▏| 22516/24420 [9:24:14<47:17,  1.49s/it]
 92%|█████████▏| 22517/24420 [9:24:15<47:10,  1.49s/it]
 92%|█████████▏| 22518/24420 [9:24:17<47:11,  1.49s/it]
 92%|█████████▏| 22519/24420 [9:24:18<47:15,  1.49s/it]
 92%|█████████▏| 22520/24420 [9:24:20<47:09,  1.49s/it]
                                                       
{'loss': 0.0773, 'learning_rate': 5e-05, 'epoch': 4.61}

 92%|█████████▏| 22520/24420 [9:24:20<47:09,  1.49s/it]
 92%|█████████▏| 22521/24420 [9:24:21<47:08,  1.49s/it]
 92%|█████████▏| 22522/24420 [9:24:23<47:06,  1.49s/it]
 92%|█████████▏| 22523/24420 [9:24:24<47:07,  1.49s/it]
 92%|█████████▏| 22524/24420 [9:24:26<47:06,  1.49s/it]
 92%|█████████▏| 22525/24420 [9:24:27<47:03,  1.49s/it]
                                                       
{'loss': 0.0728, 'learning_rate': 5e-05, 'epoch': 4.61}

 92%|█████████▏| 22525/24420 [9:24:27<47:03,  1.49s/it]
 92%|█████████▏| 22526/24420 [9:24:29<46:59,  1.49s/it]
 92%|█████████▏| 22527/24420 [9:24:30<46:54,  1.49s/it]
 92%|█████████▏| 22528/24420 [9:24:31<46:55,  1.49s/it]
 92%|█████████▏| 22529/24420 [9:24:33<46:50,  1.49s/it]
 92%|█████████▏| 22530/24420 [9:24:34<46:56,  1.49s/it]
                                                       
{'loss': 0.0619, 'learning_rate': 5e-05, 'epoch': 4.61}

 92%|█████████▏| 22530/24420 [9:24:34<46:56,  1.49s/it]
 92%|█████████▏| 22531/24420 [9:24:36<46:55,  1.49s/it]
 92%|█████████▏| 22532/24420 [9:24:37<46:53,  1.49s/it]
 92%|█████████▏| 22533/24420 [9:24:39<46:51,  1.49s/it]
 92%|█████████▏| 22534/24420 [9:24:40<46:50,  1.49s/it]
 92%|█████████▏| 22535/24420 [9:24:42<46:52,  1.49s/it]
                                                       
{'loss': 0.073, 'learning_rate': 5e-05, 'epoch': 4.61}

 92%|█████████▏| 22535/24420 [9:24:42<46:52,  1.49s/it]
 92%|█████████▏| 22536/24420 [9:24:43<46:49,  1.49s/it]
 92%|█████████▏| 22537/24420 [9:24:45<46:47,  1.49s/it]
 92%|█████████▏| 22538/24420 [9:24:46<46:48,  1.49s/it]
 92%|█████████▏| 22539/24420 [9:24:48<46:49,  1.49s/it]
 92%|█████████▏| 22540/24420 [9:24:49<46:42,  1.49s/it]
                                                       
{'loss': 0.0682, 'learning_rate': 5e-05, 'epoch': 4.62}

 92%|█████████▏| 22540/24420 [9:24:49<46:42,  1.49s/it]
 92%|█████████▏| 22541/24420 [9:24:51<46:42,  1.49s/it]
 92%|█████████▏| 22542/24420 [9:24:52<46:36,  1.49s/it]
 92%|█████████▏| 22543/24420 [9:24:54<46:35,  1.49s/it]
 92%|█████████▏| 22544/24420 [9:24:55<46:32,  1.49s/it]
 92%|█████████▏| 22545/24420 [9:24:57<46:27,  1.49s/it]
                                                       
{'loss': 0.0664, 'learning_rate': 5e-05, 'epoch': 4.62}

 92%|█████████▏| 22545/24420 [9:24:57<46:27,  1.49s/it]
 92%|█████████▏| 22546/24420 [9:24:58<46:25,  1.49s/it]
 92%|█████████▏| 22547/24420 [9:25:00<46:25,  1.49s/it]
 92%|█████████▏| 22548/24420 [9:25:01<46:28,  1.49s/it]
 92%|█████████▏| 22549/24420 [9:25:03<46:25,  1.49s/it]
 92%|█████████▏| 22550/24420 [9:25:04<46:29,  1.49s/it]
                                                       
{'loss': 0.0785, 'learning_rate': 5e-05, 'epoch': 4.62}

 92%|█████████▏| 22550/24420 [9:25:04<46:29,  1.49s/it]
 92%|█████████▏| 22551/24420 [9:25:06<46:32,  1.49s/it]
 92%|█████████▏| 22552/24420 [9:25:07<46:27,  1.49s/it]
 92%|█████████▏| 22553/24420 [9:25:09<46:18,  1.49s/it]
 92%|█████████▏| 22554/24420 [9:25:10<46:19,  1.49s/it]
 92%|█████████▏| 22555/24420 [9:25:12<46:21,  1.49s/it]
                                                       
{'loss': 0.0682, 'learning_rate': 5e-05, 'epoch': 4.62}

 92%|█████████▏| 22555/24420 [9:25:12<46:21,  1.49s/it]
 92%|█████████▏| 22556/24420 [9:25:13<46:19,  1.49s/it]
 92%|█████████▏| 22557/24420 [9:25:15<46:17,  1.49s/it]
 92%|█████████▏| 22558/24420 [9:25:16<46:25,  1.50s/it]
 92%|█████████▏| 22559/24420 [9:25:18<46:20,  1.49s/it]
 92%|█████████▏| 22560/24420 [9:25:19<46:12,  1.49s/it]
                                                       
{'loss': 0.087, 'learning_rate': 5e-05, 'epoch': 4.62}

 92%|█████████▏| 22560/24420 [9:25:19<46:12,  1.49s/it]
 92%|█████████▏| 22561/24420 [9:25:21<46:09,  1.49s/it]
 92%|█████████▏| 22562/24420 [9:25:22<46:06,  1.49s/it]
 92%|█████████▏| 22563/24420 [9:25:24<46:06,  1.49s/it]
 92%|█████████▏| 22564/24420 [9:25:25<46:03,  1.49s/it]
 92%|█████████▏| 22565/24420 [9:25:27<46:05,  1.49s/it]
                                                       
{'loss': 0.0695, 'learning_rate': 5e-05, 'epoch': 4.62}

 92%|█████████▏| 22565/24420 [9:25:27<46:05,  1.49s/it]
 92%|█████████▏| 22566/24420 [9:25:28<46:03,  1.49s/it]
 92%|█████████▏| 22567/24420 [9:25:30<45:58,  1.49s/it]
 92%|█████████▏| 22568/24420 [9:25:31<45:56,  1.49s/it]
 92%|█████████▏| 22569/24420 [9:25:33<45:50,  1.49s/it]
 92%|█████████▏| 22570/24420 [9:25:34<45:55,  1.49s/it]
                                                       
{'loss': 0.0717, 'learning_rate': 5e-05, 'epoch': 4.62}

 92%|█████████▏| 22570/24420 [9:25:34<45:55,  1.49s/it]
 92%|█████████▏| 22571/24420 [9:25:36<45:52,  1.49s/it]
 92%|█████████▏| 22572/24420 [9:25:37<45:49,  1.49s/it]
 92%|█████████▏| 22573/24420 [9:25:39<45:44,  1.49s/it]
 92%|█████████▏| 22574/24420 [9:25:40<45:46,  1.49s/it]
 92%|█████████▏| 22575/24420 [9:25:42<45:43,  1.49s/it]
                                                       
{'loss': 0.0647, 'learning_rate': 5e-05, 'epoch': 4.62}

 92%|█████████▏| 22575/24420 [9:25:42<45:43,  1.49s/it]
 92%|█████████▏| 22576/24420 [9:25:43<45:43,  1.49s/it]
 92%|█████████▏| 22577/24420 [9:25:44<45:44,  1.49s/it]
 92%|█████████▏| 22578/24420 [9:25:46<45:38,  1.49s/it]
 92%|█████████▏| 22579/24420 [9:25:47<45:36,  1.49s/it]
 92%|█████████▏| 22580/24420 [9:25:49<45:37,  1.49s/it]
                                                       
{'loss': 0.078, 'learning_rate': 5e-05, 'epoch': 4.62}

 92%|█████████▏| 22580/24420 [9:25:49<45:37,  1.49s/it]
 92%|█████████▏| 22581/24420 [9:25:50<45:33,  1.49s/it]
 92%|█████████▏| 22582/24420 [9:25:52<45:29,  1.48s/it]
 92%|█████████▏| 22583/24420 [9:25:53<45:34,  1.49s/it]
 92%|█████████▏| 22584/24420 [9:25:55<45:29,  1.49s/it]
 92%|█████████▏| 22585/24420 [9:25:56<45:25,  1.49s/it]
                                                       
{'loss': 0.0729, 'learning_rate': 5e-05, 'epoch': 4.62}

 92%|█████████▏| 22585/24420 [9:25:56<45:25,  1.49s/it]
 92%|█████████▏| 22586/24420 [9:25:58<45:29,  1.49s/it]
 92%|█████████▏| 22587/24420 [9:25:59<45:26,  1.49s/it]
 92%|█████████▏| 22588/24420 [9:26:01<45:23,  1.49s/it]
 93%|█████████▎| 22589/24420 [9:26:02<45:22,  1.49s/it]
 93%|█████████▎| 22590/24420 [9:26:04<45:23,  1.49s/it]
                                                       
{'loss': 0.0781, 'learning_rate': 5e-05, 'epoch': 4.63}

 93%|█████████▎| 22590/24420 [9:26:04<45:23,  1.49s/it]
 93%|█████████▎| 22591/24420 [9:26:05<45:28,  1.49s/it]
 93%|█████████▎| 22592/24420 [9:26:07<45:22,  1.49s/it]
 93%|█████████▎| 22593/24420 [9:26:08<45:22,  1.49s/it]
 93%|█████████▎| 22594/24420 [9:26:10<45:17,  1.49s/it]
 93%|█████████▎| 22595/24420 [9:26:11<45:11,  1.49s/it]
                                                       
{'loss': 0.0747, 'learning_rate': 5e-05, 'epoch': 4.63}

 93%|█████████▎| 22595/24420 [9:26:11<45:11,  1.49s/it]
 93%|█████████▎| 22596/24420 [9:26:13<45:12,  1.49s/it]
 93%|█████████▎| 22597/24420 [9:26:14<45:15,  1.49s/it]
 93%|█████████▎| 22598/24420 [9:26:16<45:15,  1.49s/it]
 93%|█████████▎| 22599/24420 [9:26:17<45:10,  1.49s/it]
 93%|█████████▎| 22600/24420 [9:26:19<45:12,  1.49s/it]
                                                       
{'loss': 0.0739, 'learning_rate': 5e-05, 'epoch': 4.63}

 93%|█████████▎| 22600/24420 [9:26:19<45:12,  1.49s/it]
 93%|█████████▎| 22601/24420 [9:26:23<1:13:26,  2.42s/it]
 93%|█████████▎| 22602/24420 [9:26:25<1:04:50,  2.14s/it]
 93%|█████████▎| 22603/24420 [9:26:26<58:59,  1.95s/it]  
 93%|█████████▎| 22604/24420 [9:26:28<54:51,  1.81s/it]
 93%|█████████▎| 22605/24420 [9:26:29<51:50,  1.71s/it]
                                                       
{'loss': 0.0744, 'learning_rate': 5e-05, 'epoch': 4.63}

 93%|█████████▎| 22605/24420 [9:26:29<51:50,  1.71s/it]
 93%|█████████▎| 22606/24420 [9:26:31<49:48,  1.65s/it]
 93%|█████████▎| 22607/24420 [9:26:32<48:20,  1.60s/it]
 93%|█████████▎| 22608/24420 [9:26:34<47:15,  1.57s/it]
 93%|█████████▎| 22609/24420 [9:26:35<46:35,  1.54s/it]
 93%|█████████▎| 22610/24420 [9:26:37<46:01,  1.53s/it]
                                                       
{'loss': 0.0762, 'learning_rate': 5e-05, 'epoch': 4.63}

 93%|█████████▎| 22610/24420 [9:26:37<46:01,  1.53s/it]
 93%|█████████▎| 22611/24420 [9:26:38<45:50,  1.52s/it]
 93%|█████████▎| 22612/24420 [9:26:40<45:30,  1.51s/it]
 93%|█████████▎| 22613/24420 [9:26:41<45:18,  1.50s/it]
 93%|█████████▎| 22614/24420 [9:26:43<45:05,  1.50s/it]
 93%|█████████▎| 22615/24420 [9:26:44<44:56,  1.49s/it]
                                                       
{'loss': 0.083, 'learning_rate': 5e-05, 'epoch': 4.63}

 93%|█████████▎| 22615/24420 [9:26:44<44:56,  1.49s/it]
 93%|█████████▎| 22616/24420 [9:26:46<44:52,  1.49s/it]
 93%|█████████▎| 22617/24420 [9:26:47<44:45,  1.49s/it]
 93%|█████████▎| 22618/24420 [9:26:49<44:42,  1.49s/it]
 93%|█████████▎| 22619/24420 [9:26:50<44:35,  1.49s/it]
 93%|█████████▎| 22620/24420 [9:26:52<44:43,  1.49s/it]
                                                       
{'loss': 0.0698, 'learning_rate': 5e-05, 'epoch': 4.63}

 93%|█████████▎| 22620/24420 [9:26:52<44:43,  1.49s/it]
 93%|█████████▎| 22621/24420 [9:26:53<44:43,  1.49s/it]
 93%|█████████▎| 22622/24420 [9:26:55<44:37,  1.49s/it]
 93%|█████████▎| 22623/24420 [9:26:56<44:32,  1.49s/it]
 93%|█████████▎| 22624/24420 [9:26:58<44:32,  1.49s/it]
 93%|█████████▎| 22625/24420 [9:26:59<44:30,  1.49s/it]
                                                       
{'loss': 0.0793, 'learning_rate': 5e-05, 'epoch': 4.63}

 93%|█████████▎| 22625/24420 [9:26:59<44:30,  1.49s/it]
 93%|█████████▎| 22626/24420 [9:27:01<44:34,  1.49s/it]
 93%|█████████▎| 22627/24420 [9:27:02<44:30,  1.49s/it]
 93%|█████████▎| 22628/24420 [9:27:04<44:27,  1.49s/it]
 93%|█████████▎| 22629/24420 [9:27:05<44:29,  1.49s/it]
 93%|█████████▎| 22630/24420 [9:27:06<44:26,  1.49s/it]
                                                       
{'loss': 0.0715, 'learning_rate': 5e-05, 'epoch': 4.63}

 93%|█████████▎| 22630/24420 [9:27:07<44:26,  1.49s/it]
 93%|█████████▎| 22631/24420 [9:27:08<44:20,  1.49s/it]
 93%|█████████▎| 22632/24420 [9:27:09<44:19,  1.49s/it]
 93%|█████████▎| 22633/24420 [9:27:11<44:16,  1.49s/it]
 93%|█████████▎| 22634/24420 [9:27:12<44:21,  1.49s/it]
 93%|█████████▎| 22635/24420 [9:27:14<44:19,  1.49s/it]
                                                       
{'loss': 0.0646, 'learning_rate': 5e-05, 'epoch': 4.63}

 93%|█████████▎| 22635/24420 [9:27:14<44:19,  1.49s/it]
 93%|█████████▎| 22636/24420 [9:27:15<44:29,  1.50s/it]
 93%|█████████▎| 22637/24420 [9:27:17<44:28,  1.50s/it]
 93%|█████████▎| 22638/24420 [9:27:18<44:20,  1.49s/it]
 93%|█████████▎| 22639/24420 [9:27:20<44:15,  1.49s/it]
 93%|█████████▎| 22640/24420 [9:27:21<44:12,  1.49s/it]
                                                       
{'loss': 0.0597, 'learning_rate': 5e-05, 'epoch': 4.64}

 93%|█████████▎| 22640/24420 [9:27:21<44:12,  1.49s/it]
 93%|█████████▎| 22641/24420 [9:27:23<44:17,  1.49s/it]
 93%|█████████▎| 22642/24420 [9:27:24<44:20,  1.50s/it]
 93%|█████████▎| 22643/24420 [9:27:26<44:16,  1.49s/it]
 93%|█████████▎| 22644/24420 [9:27:27<44:16,  1.50s/it]
 93%|█████████▎| 22645/24420 [9:27:29<44:10,  1.49s/it]
                                                       
{'loss': 0.0876, 'learning_rate': 5e-05, 'epoch': 4.64}

 93%|█████████▎| 22645/24420 [9:27:29<44:10,  1.49s/it]
 93%|█████████▎| 22646/24420 [9:27:30<44:03,  1.49s/it]
 93%|█████████▎| 22647/24420 [9:27:32<44:02,  1.49s/it]
 93%|█████████▎| 22648/24420 [9:27:33<44:01,  1.49s/it]
 93%|█████████▎| 22649/24420 [9:27:35<44:01,  1.49s/it]
 93%|█████████▎| 22650/24420 [9:27:36<43:57,  1.49s/it]
                                                       
{'loss': 0.0905, 'learning_rate': 5e-05, 'epoch': 4.64}

 93%|█████████▎| 22650/24420 [9:27:36<43:57,  1.49s/it]
 93%|█████████▎| 22651/24420 [9:27:38<43:48,  1.49s/it]
 93%|█████████▎| 22652/24420 [9:27:39<43:48,  1.49s/it]
 93%|█████████▎| 22653/24420 [9:27:41<43:48,  1.49s/it]
 93%|█████████▎| 22654/24420 [9:27:42<43:47,  1.49s/it]
 93%|█████████▎| 22655/24420 [9:27:44<43:45,  1.49s/it]
                                                       
{'loss': 0.0701, 'learning_rate': 5e-05, 'epoch': 4.64}

 93%|█████████▎| 22655/24420 [9:27:44<43:45,  1.49s/it]
 93%|█████████▎| 22656/24420 [9:27:45<43:45,  1.49s/it]
 93%|█████████▎| 22657/24420 [9:27:47<43:40,  1.49s/it]
 93%|█████████▎| 22658/24420 [9:27:48<43:39,  1.49s/it]
 93%|█████████▎| 22659/24420 [9:27:50<43:38,  1.49s/it]
 93%|█████████▎| 22660/24420 [9:27:51<43:34,  1.49s/it]
                                                       
{'loss': 0.0772, 'learning_rate': 5e-05, 'epoch': 4.64}

 93%|█████████▎| 22660/24420 [9:27:51<43:34,  1.49s/it]
 93%|█████████▎| 22661/24420 [9:27:53<43:36,  1.49s/it]
 93%|█████████▎| 22662/24420 [9:27:54<43:32,  1.49s/it]
 93%|█████████▎| 22663/24420 [9:27:56<43:29,  1.49s/it]
 93%|█████████▎| 22664/24420 [9:27:57<43:27,  1.49s/it]
 93%|█████████▎| 22665/24420 [9:27:59<43:23,  1.48s/it]
                                                       
{'loss': 0.0716, 'learning_rate': 5e-05, 'epoch': 4.64}

 93%|█████████▎| 22665/24420 [9:27:59<43:23,  1.48s/it]
 93%|█████████▎| 22666/24420 [9:28:00<43:25,  1.49s/it]
 93%|█████████▎| 22667/24420 [9:28:02<43:25,  1.49s/it]
 93%|█████████▎| 22668/24420 [9:28:03<43:33,  1.49s/it]
 93%|█████████▎| 22669/24420 [9:28:05<43:34,  1.49s/it]
 93%|█████████▎| 22670/24420 [9:28:06<43:29,  1.49s/it]
                                                       
{'loss': 0.088, 'learning_rate': 5e-05, 'epoch': 4.64}

 93%|█████████▎| 22670/24420 [9:28:06<43:29,  1.49s/it]
 93%|█████████▎| 22671/24420 [9:28:08<43:26,  1.49s/it]
 93%|█████████▎| 22672/24420 [9:28:09<43:22,  1.49s/it]
 93%|█████████▎| 22673/24420 [9:28:11<43:18,  1.49s/it]
 93%|█████████▎| 22674/24420 [9:28:12<43:23,  1.49s/it]
 93%|█████████▎| 22675/24420 [9:28:14<43:22,  1.49s/it]
                                                       
{'loss': 0.0651, 'learning_rate': 5e-05, 'epoch': 4.64}

 93%|█████████▎| 22675/24420 [9:28:14<43:22,  1.49s/it]
 93%|█████████▎| 22676/24420 [9:28:15<43:19,  1.49s/it]
 93%|█████████▎| 22677/24420 [9:28:17<43:17,  1.49s/it]
 93%|█████████▎| 22678/24420 [9:28:18<43:12,  1.49s/it]
 93%|█████████▎| 22679/24420 [9:28:20<43:24,  1.50s/it]
 93%|█████████▎| 22680/24420 [9:28:21<43:18,  1.49s/it]
                                                       
{'loss': 0.0706, 'learning_rate': 5e-05, 'epoch': 4.64}

 93%|█████████▎| 22680/24420 [9:28:21<43:18,  1.49s/it]
 93%|█████████▎| 22681/24420 [9:28:22<43:14,  1.49s/it]
 93%|█████████▎| 22682/24420 [9:28:24<43:18,  1.49s/it]
 93%|█████████▎| 22683/24420 [9:28:25<43:11,  1.49s/it]
 93%|█████████▎| 22684/24420 [9:28:27<43:04,  1.49s/it]
 93%|█████████▎| 22685/24420 [9:28:28<43:08,  1.49s/it]
                                                       
{'loss': 0.0773, 'learning_rate': 5e-05, 'epoch': 4.64}

 93%|█████████▎| 22685/24420 [9:28:28<43:08,  1.49s/it]
 93%|█████████▎| 22686/24420 [9:28:30<43:04,  1.49s/it]
 93%|█████████▎| 22687/24420 [9:28:31<43:01,  1.49s/it]
 93%|█████████▎| 22688/24420 [9:28:33<43:00,  1.49s/it]
 93%|█████████▎| 22689/24420 [9:28:34<43:01,  1.49s/it]
 93%|█████████▎| 22690/24420 [9:28:36<42:57,  1.49s/it]
                                                       
{'loss': 0.0732, 'learning_rate': 5e-05, 'epoch': 4.65}

 93%|█████████▎| 22690/24420 [9:28:36<42:57,  1.49s/it]
 93%|█████████▎| 22691/24420 [9:28:37<43:02,  1.49s/it]
 93%|█████████▎| 22692/24420 [9:28:39<42:58,  1.49s/it]
 93%|█████████▎| 22693/24420 [9:28:40<42:58,  1.49s/it]
 93%|█████████▎| 22694/24420 [9:28:42<42:56,  1.49s/it]
 93%|█████████▎| 22695/24420 [9:28:43<42:51,  1.49s/it]
                                                       
{'loss': 0.0683, 'learning_rate': 5e-05, 'epoch': 4.65}

 93%|█████████▎| 22695/24420 [9:28:43<42:51,  1.49s/it]
 93%|█████████▎| 22696/24420 [9:28:45<42:52,  1.49s/it]
 93%|█████████▎| 22697/24420 [9:28:46<42:48,  1.49s/it]
 93%|█████████▎| 22698/24420 [9:28:48<42:39,  1.49s/it]
 93%|█████████▎| 22699/24420 [9:28:49<42:42,  1.49s/it]
 93%|█████████▎| 22700/24420 [9:28:51<42:42,  1.49s/it]
                                                       
{'loss': 0.0839, 'learning_rate': 5e-05, 'epoch': 4.65}

 93%|█████████▎| 22700/24420 [9:28:51<42:42,  1.49s/it]
 93%|█████████▎| 22701/24420 [9:28:52<42:40,  1.49s/it]
 93%|█████████▎| 22702/24420 [9:28:54<42:38,  1.49s/it]
 93%|█████████▎| 22703/24420 [9:28:55<42:33,  1.49s/it]
 93%|█████████▎| 22704/24420 [9:28:57<42:31,  1.49s/it]
 93%|█████████▎| 22705/24420 [9:28:58<42:33,  1.49s/it]
                                                       
{'loss': 0.0719, 'learning_rate': 5e-05, 'epoch': 4.65}

 93%|█████████▎| 22705/24420 [9:28:58<42:33,  1.49s/it]
 93%|█████████▎| 22706/24420 [9:29:00<42:28,  1.49s/it]
 93%|█████████▎| 22707/24420 [9:29:01<42:22,  1.48s/it]
 93%|█████████▎| 22708/24420 [9:29:03<42:20,  1.48s/it]
 93%|█████████▎| 22709/24420 [9:29:04<42:17,  1.48s/it]
 93%|█████████▎| 22710/24420 [9:29:06<42:18,  1.48s/it]
                                                       
{'loss': 0.076, 'learning_rate': 5e-05, 'epoch': 4.65}

 93%|█████████▎| 22710/24420 [9:29:06<42:18,  1.48s/it]
 93%|█████████▎| 22711/24420 [9:29:07<42:18,  1.49s/it]
 93%|█████████▎| 22712/24420 [9:29:09<42:15,  1.48s/it]
 93%|█████████▎| 22713/24420 [9:29:10<42:17,  1.49s/it]
 93%|█████████▎| 22714/24420 [9:29:12<42:16,  1.49s/it]
 93%|█████████▎| 22715/24420 [9:29:13<42:15,  1.49s/it]
                                                       
{'loss': 0.0624, 'learning_rate': 5e-05, 'epoch': 4.65}

 93%|█████████▎| 22715/24420 [9:29:13<42:15,  1.49s/it]
 93%|█████████▎| 22716/24420 [9:29:15<42:17,  1.49s/it]
 93%|█████████▎| 22717/24420 [9:29:16<42:21,  1.49s/it]
 93%|█████████▎| 22718/24420 [9:29:18<42:21,  1.49s/it]
 93%|█████████▎| 22719/24420 [9:29:19<42:14,  1.49s/it]
 93%|█████████▎| 22720/24420 [9:29:21<42:09,  1.49s/it]
                                                       
{'loss': 0.0678, 'learning_rate': 5e-05, 'epoch': 4.65}

 93%|█████████▎| 22720/24420 [9:29:21<42:09,  1.49s/it]
 93%|█████████▎| 22721/24420 [9:29:22<42:11,  1.49s/it]
 93%|█████████▎| 22722/24420 [9:29:24<42:11,  1.49s/it]
 93%|█████████▎| 22723/24420 [9:29:25<42:09,  1.49s/it]
 93%|█████████▎| 22724/24420 [9:29:27<42:05,  1.49s/it]
 93%|█████████▎| 22725/24420 [9:29:28<42:04,  1.49s/it]
                                                       
{'loss': 0.0838, 'learning_rate': 5e-05, 'epoch': 4.65}

 93%|█████████▎| 22725/24420 [9:29:28<42:04,  1.49s/it]
 93%|█████████▎| 22726/24420 [9:29:29<41:58,  1.49s/it]
 93%|█████████▎| 22727/24420 [9:29:31<41:58,  1.49s/it]
 93%|█████████▎| 22728/24420 [9:29:32<41:54,  1.49s/it]
 93%|█████████▎| 22729/24420 [9:29:34<41:57,  1.49s/it]
 93%|█████████▎| 22730/24420 [9:29:35<41:55,  1.49s/it]
                                                       
{'loss': 0.0651, 'learning_rate': 5e-05, 'epoch': 4.65}

 93%|█████████▎| 22730/24420 [9:29:35<41:55,  1.49s/it]
 93%|█████████▎| 22731/24420 [9:29:37<41:58,  1.49s/it]
 93%|█████████▎| 22732/24420 [9:29:38<41:54,  1.49s/it]
 93%|█████████▎| 22733/24420 [9:29:40<41:54,  1.49s/it]
 93%|█████████▎| 22734/24420 [9:29:41<41:49,  1.49s/it]
 93%|█████████▎| 22735/24420 [9:29:43<41:53,  1.49s/it]
                                                       
{'loss': 0.1055, 'learning_rate': 5e-05, 'epoch': 4.65}

 93%|█████████▎| 22735/24420 [9:29:43<41:53,  1.49s/it]
 93%|█████████▎| 22736/24420 [9:29:44<41:48,  1.49s/it]
 93%|█████████▎| 22737/24420 [9:29:46<41:49,  1.49s/it]
 93%|█████████▎| 22738/24420 [9:29:47<41:44,  1.49s/it]
 93%|█████████▎| 22739/24420 [9:29:49<41:41,  1.49s/it]
 93%|█████████▎| 22740/24420 [9:29:50<41:36,  1.49s/it]
                                                       
{'loss': 0.0641, 'learning_rate': 5e-05, 'epoch': 4.66}

 93%|█████████▎| 22740/24420 [9:29:50<41:36,  1.49s/it]
 93%|█████████▎| 22741/24420 [9:29:52<41:39,  1.49s/it]
 93%|█████████▎| 22742/24420 [9:29:53<41:40,  1.49s/it]
 93%|█████████▎| 22743/24420 [9:29:55<41:36,  1.49s/it]
 93%|█████████▎| 22744/24420 [9:29:56<41:50,  1.50s/it]
 93%|█████████▎| 22745/24420 [9:29:58<41:44,  1.50s/it]
                                                       
{'loss': 0.0709, 'learning_rate': 5e-05, 'epoch': 4.66}

 93%|█████████▎| 22745/24420 [9:29:58<41:44,  1.50s/it]
 93%|█████████▎| 22746/24420 [9:29:59<41:41,  1.49s/it]
 93%|█████████▎| 22747/24420 [9:30:01<41:35,  1.49s/it]
 93%|█████████▎| 22748/24420 [9:30:02<41:32,  1.49s/it]
 93%|█████████▎| 22749/24420 [9:30:04<41:27,  1.49s/it]
 93%|█████████▎| 22750/24420 [9:30:05<41:29,  1.49s/it]
                                                       
{'loss': 0.0893, 'learning_rate': 5e-05, 'epoch': 4.66}

 93%|█████████▎| 22750/24420 [9:30:05<41:29,  1.49s/it]
 93%|█████████▎| 22751/24420 [9:30:07<41:26,  1.49s/it]
 93%|█████████▎| 22752/24420 [9:30:08<41:23,  1.49s/it]
 93%|█████████▎| 22753/24420 [9:30:10<41:18,  1.49s/it]
 93%|█████████▎| 22754/24420 [9:30:11<41:19,  1.49s/it]
 93%|█████████▎| 22755/24420 [9:30:13<41:17,  1.49s/it]
                                                       
{'loss': 0.0872, 'learning_rate': 5e-05, 'epoch': 4.66}

 93%|█████████▎| 22755/24420 [9:30:13<41:17,  1.49s/it]
 93%|█████████▎| 22756/24420 [9:30:14<41:15,  1.49s/it]
 93%|█████████▎| 22757/24420 [9:30:16<41:10,  1.49s/it]
 93%|█████████▎| 22758/24420 [9:30:17<41:08,  1.49s/it]
 93%|█████████▎| 22759/24420 [9:30:19<41:06,  1.49s/it]
 93%|█████████▎| 22760/24420 [9:30:20<41:06,  1.49s/it]
                                                       
{'loss': 0.0685, 'learning_rate': 5e-05, 'epoch': 4.66}

 93%|█████████▎| 22760/24420 [9:30:20<41:06,  1.49s/it]
 93%|█████████▎| 22761/24420 [9:30:22<41:03,  1.48s/it]
 93%|█████████▎| 22762/24420 [9:30:23<41:00,  1.48s/it]
 93%|█████████▎| 22763/24420 [9:30:25<40:56,  1.48s/it]
 93%|█████████▎| 22764/24420 [9:30:26<40:57,  1.48s/it]
 93%|█████████▎| 22765/24420 [9:30:28<40:54,  1.48s/it]
                                                       
{'loss': 0.0713, 'learning_rate': 5e-05, 'epoch': 4.66}

 93%|█████████▎| 22765/24420 [9:30:28<40:54,  1.48s/it]
 93%|█████████▎| 22766/24420 [9:30:29<40:56,  1.49s/it]
 93%|█████████▎| 22767/24420 [9:30:31<40:56,  1.49s/it]
 93%|█████████▎| 22768/24420 [9:30:32<40:50,  1.48s/it]
 93%|█████████▎| 22769/24420 [9:30:33<40:50,  1.48s/it]
 93%|█████████▎| 22770/24420 [9:30:35<40:51,  1.49s/it]
                                                       
{'loss': 0.0846, 'learning_rate': 5e-05, 'epoch': 4.66}

 93%|█████████▎| 22770/24420 [9:30:35<40:51,  1.49s/it]
 93%|█████████▎| 22771/24420 [9:30:36<40:50,  1.49s/it]
 93%|█████████▎| 22772/24420 [9:30:38<40:49,  1.49s/it]
 93%|█████████▎| 22773/24420 [9:30:39<40:47,  1.49s/it]
 93%|█████████▎| 22774/24420 [9:30:41<40:50,  1.49s/it]
 93%|█████████▎| 22775/24420 [9:30:42<40:49,  1.49s/it]
                                                       
{'loss': 0.0902, 'learning_rate': 5e-05, 'epoch': 4.66}

 93%|█████████▎| 22775/24420 [9:30:42<40:49,  1.49s/it]
 93%|█████████▎| 22776/24420 [9:30:44<40:48,  1.49s/it]
 93%|█████████▎| 22777/24420 [9:30:45<40:40,  1.49s/it]
 93%|█████████▎| 22778/24420 [9:30:47<40:35,  1.48s/it]
 93%|█████████▎| 22779/24420 [9:30:48<40:32,  1.48s/it]
 93%|█████████▎| 22780/24420 [9:30:50<40:33,  1.48s/it]
                                                       
{'loss': 0.0632, 'learning_rate': 5e-05, 'epoch': 4.66}

 93%|█████████▎| 22780/24420 [9:30:50<40:33,  1.48s/it]
 93%|█████████▎| 22781/24420 [9:30:51<40:33,  1.48s/it]
 93%|█████████▎| 22782/24420 [9:30:53<40:36,  1.49s/it]
 93%|█████████▎| 22783/24420 [9:30:54<40:35,  1.49s/it]
 93%|█████████▎| 22784/24420 [9:30:56<40:33,  1.49s/it]
 93%|█████████▎| 22785/24420 [9:30:57<40:28,  1.49s/it]
                                                       
{'loss': 0.0617, 'learning_rate': 5e-05, 'epoch': 4.67}

 93%|█████████▎| 22785/24420 [9:30:57<40:28,  1.49s/it]
 93%|█████████▎| 22786/24420 [9:30:59<40:27,  1.49s/it]
 93%|█████████▎| 22787/24420 [9:31:00<40:24,  1.48s/it]
 93%|█████████▎| 22788/24420 [9:31:02<40:22,  1.48s/it]
 93%|█████████▎| 22789/24420 [9:31:03<40:20,  1.48s/it]
 93%|█████████▎| 22790/24420 [9:31:05<40:17,  1.48s/it]
                                                       
{'loss': 0.0644, 'learning_rate': 5e-05, 'epoch': 4.67}

 93%|█████████▎| 22790/24420 [9:31:05<40:17,  1.48s/it]
 93%|█████████▎| 22791/24420 [9:31:06<40:16,  1.48s/it]
 93%|█████████▎| 22792/24420 [9:31:08<40:21,  1.49s/it]
 93%|█████████▎| 22793/24420 [9:31:09<40:21,  1.49s/it]
 93%|█████████▎| 22794/24420 [9:31:11<40:23,  1.49s/it]
 93%|█████████▎| 22795/24420 [9:31:12<40:22,  1.49s/it]
                                                       
{'loss': 0.0807, 'learning_rate': 5e-05, 'epoch': 4.67}

 93%|█████████▎| 22795/24420 [9:31:12<40:22,  1.49s/it]
 93%|█████████▎| 22796/24420 [9:31:14<40:16,  1.49s/it]
 93%|█████████▎| 22797/24420 [9:31:15<40:15,  1.49s/it]
 93%|█████████▎| 22798/24420 [9:31:17<40:15,  1.49s/it]
 93%|█████████▎| 22799/24420 [9:31:18<40:17,  1.49s/it]
 93%|█████████▎| 22800/24420 [9:31:20<40:15,  1.49s/it]
                                                       
{'loss': 0.069, 'learning_rate': 5e-05, 'epoch': 4.67}

 93%|█████████▎| 22800/24420 [9:31:20<40:15,  1.49s/it]
 93%|█████████▎| 22801/24420 [9:31:24<1:05:40,  2.43s/it]
 93%|█████████▎| 22802/24420 [9:31:26<57:57,  2.15s/it]  
 93%|█████████▎| 22803/24420 [9:31:27<52:32,  1.95s/it]
 93%|█████████▎| 22804/24420 [9:31:29<48:57,  1.82s/it]
 93%|█████████▎| 22805/24420 [9:31:30<46:12,  1.72s/it]
                                                       
{'loss': 0.0927, 'learning_rate': 5e-05, 'epoch': 4.67}

 93%|█████████▎| 22805/24420 [9:31:30<46:12,  1.72s/it]
 93%|█████████▎| 22806/24420 [9:31:32<44:19,  1.65s/it]
 93%|█████████▎| 22807/24420 [9:31:33<43:00,  1.60s/it]
 93%|█████████▎| 22808/24420 [9:31:35<42:00,  1.56s/it]
 93%|█████████▎| 22809/24420 [9:31:36<41:27,  1.54s/it]
 93%|█████████▎| 22810/24420 [9:31:38<41:00,  1.53s/it]
                                                       
{'loss': 0.0776, 'learning_rate': 5e-05, 'epoch': 4.67}

 93%|█████████▎| 22810/24420 [9:31:38<41:00,  1.53s/it]
 93%|█████████▎| 22811/24420 [9:31:39<40:43,  1.52s/it]
 93%|█████████▎| 22812/24420 [9:31:41<40:25,  1.51s/it]
 93%|█████████▎| 22813/24420 [9:31:42<40:13,  1.50s/it]
 93%|█████████▎| 22814/24420 [9:31:44<40:03,  1.50s/it]
 93%|█████████▎| 22815/24420 [9:31:45<39:53,  1.49s/it]
                                                       
{'loss': 0.073, 'learning_rate': 5e-05, 'epoch': 4.67}

 93%|█████████▎| 22815/24420 [9:31:45<39:53,  1.49s/it]
 93%|█████████▎| 22816/24420 [9:31:47<39:51,  1.49s/it]
 93%|█████████▎| 22817/24420 [9:31:48<39:48,  1.49s/it]
 93%|█████████▎| 22818/24420 [9:31:50<39:45,  1.49s/it]
 93%|█████████▎| 22819/24420 [9:31:51<39:42,  1.49s/it]
 93%|█████████▎| 22820/24420 [9:31:52<39:43,  1.49s/it]
                                                       
{'loss': 0.0876, 'learning_rate': 5e-05, 'epoch': 4.67}

 93%|█████████▎| 22820/24420 [9:31:52<39:43,  1.49s/it]
 93%|█████████▎| 22821/24420 [9:31:54<39:42,  1.49s/it]
 93%|█████████▎| 22822/24420 [9:31:55<39:40,  1.49s/it]
 93%|█████████▎| 22823/24420 [9:31:57<39:48,  1.50s/it]
 93%|█████████▎| 22824/24420 [9:31:58<39:44,  1.49s/it]
 93%|█████████▎| 22825/24420 [9:32:00<39:44,  1.50s/it]
                                                       
{'loss': 0.0633, 'learning_rate': 5e-05, 'epoch': 4.67}

 93%|█████████▎| 22825/24420 [9:32:00<39:44,  1.50s/it]
 93%|█████████▎| 22826/24420 [9:32:01<39:41,  1.49s/it]
 93%|█████████▎| 22827/24420 [9:32:03<39:38,  1.49s/it]
 93%|█████████▎| 22828/24420 [9:32:04<39:31,  1.49s/it]
 93%|█████████▎| 22829/24420 [9:32:06<39:31,  1.49s/it]
 93%|█████████▎| 22830/24420 [9:32:07<39:31,  1.49s/it]
                                                       
{'loss': 0.0716, 'learning_rate': 5e-05, 'epoch': 4.67}

 93%|█████████▎| 22830/24420 [9:32:07<39:31,  1.49s/it]
 93%|█████████▎| 22831/24420 [9:32:09<39:26,  1.49s/it]
 93%|█████████▎| 22832/24420 [9:32:10<39:23,  1.49s/it]
 94%|█████████▎| 22833/24420 [9:32:12<39:23,  1.49s/it]
 94%|█████████▎| 22834/24420 [9:32:13<39:23,  1.49s/it]
 94%|█████████▎| 22835/24420 [9:32:15<39:24,  1.49s/it]
                                                       
{'loss': 0.0889, 'learning_rate': 5e-05, 'epoch': 4.68}

 94%|█████████▎| 22835/24420 [9:32:15<39:24,  1.49s/it]
 94%|█████████▎| 22836/24420 [9:32:16<39:24,  1.49s/it]
 94%|█████████▎| 22837/24420 [9:32:18<39:23,  1.49s/it]
 94%|█████████▎| 22838/24420 [9:32:19<39:22,  1.49s/it]
 94%|█████████▎| 22839/24420 [9:32:21<39:16,  1.49s/it]
 94%|█████████▎| 22840/24420 [9:32:22<39:13,  1.49s/it]
                                                       
{'loss': 0.0771, 'learning_rate': 5e-05, 'epoch': 4.68}

 94%|█████████▎| 22840/24420 [9:32:22<39:13,  1.49s/it]
 94%|█████████▎| 22841/24420 [9:32:24<39:15,  1.49s/it]
 94%|█████████▎| 22842/24420 [9:32:25<39:14,  1.49s/it]
 94%|█████████▎| 22843/24420 [9:32:27<39:08,  1.49s/it]
 94%|█████████▎| 22844/24420 [9:32:28<39:10,  1.49s/it]
 94%|█████████▎| 22845/24420 [9:32:30<39:11,  1.49s/it]
                                                       
{'loss': 0.0698, 'learning_rate': 5e-05, 'epoch': 4.68}

 94%|█████████▎| 22845/24420 [9:32:30<39:11,  1.49s/it]
 94%|█████████▎| 22846/24420 [9:32:31<39:07,  1.49s/it]
 94%|█████████▎| 22847/24420 [9:32:33<39:11,  1.49s/it]
 94%|█████████▎| 22848/24420 [9:32:34<39:08,  1.49s/it]
 94%|█████████▎| 22849/24420 [9:32:36<39:04,  1.49s/it]
 94%|█████████▎| 22850/24420 [9:32:37<39:03,  1.49s/it]
                                                       
{'loss': 0.082, 'learning_rate': 5e-05, 'epoch': 4.68}

 94%|█████████▎| 22850/24420 [9:32:37<39:03,  1.49s/it]
 94%|█████████▎| 22851/24420 [9:32:39<38:59,  1.49s/it]
 94%|█████████▎| 22852/24420 [9:32:40<38:55,  1.49s/it]
 94%|█████████▎| 22853/24420 [9:32:42<38:58,  1.49s/it]
 94%|█████████▎| 22854/24420 [9:32:43<39:00,  1.49s/it]
 94%|█████████▎| 22855/24420 [9:32:45<39:00,  1.50s/it]
                                                       
{'loss': 0.0769, 'learning_rate': 5e-05, 'epoch': 4.68}

 94%|█████████▎| 22855/24420 [9:32:45<39:00,  1.50s/it]
 94%|█████████▎| 22856/24420 [9:32:46<38:54,  1.49s/it]
 94%|█████████▎| 22857/24420 [9:32:48<38:54,  1.49s/it]
 94%|█████████▎| 22858/24420 [9:32:49<38:50,  1.49s/it]
 94%|█████████▎| 22859/24420 [9:32:51<38:49,  1.49s/it]
 94%|█████████▎| 22860/24420 [9:32:52<38:47,  1.49s/it]
                                                       
{'loss': 0.0727, 'learning_rate': 5e-05, 'epoch': 4.68}

 94%|█████████▎| 22860/24420 [9:32:52<38:47,  1.49s/it]
 94%|█████████▎| 22861/24420 [9:32:54<38:50,  1.49s/it]
 94%|█████████▎| 22862/24420 [9:32:55<38:47,  1.49s/it]
 94%|█████████▎| 22863/24420 [9:32:57<38:39,  1.49s/it]
 94%|█████████▎| 22864/24420 [9:32:58<38:35,  1.49s/it]
 94%|█████████▎| 22865/24420 [9:33:00<38:32,  1.49s/it]
                                                       
{'loss': 0.0696, 'learning_rate': 5e-05, 'epoch': 4.68}

 94%|█████████▎| 22865/24420 [9:33:00<38:32,  1.49s/it]
 94%|█████████▎| 22866/24420 [9:33:01<38:36,  1.49s/it]
 94%|█████████▎| 22867/24420 [9:33:03<38:31,  1.49s/it]
 94%|█████████▎| 22868/24420 [9:33:04<38:30,  1.49s/it]
 94%|█████████▎| 22869/24420 [9:33:06<38:27,  1.49s/it]
 94%|█████████▎| 22870/24420 [9:33:07<38:30,  1.49s/it]
                                                       
{'loss': 0.0886, 'learning_rate': 5e-05, 'epoch': 4.68}

 94%|█████████▎| 22870/24420 [9:33:07<38:30,  1.49s/it]
 94%|█████████▎| 22871/24420 [9:33:09<38:23,  1.49s/it]
 94%|█████████▎| 22872/24420 [9:33:10<38:22,  1.49s/it]
 94%|█████████▎| 22873/24420 [9:33:12<38:16,  1.48s/it]
 94%|█████████▎| 22874/24420 [9:33:13<38:15,  1.49s/it]
 94%|█████████▎| 22875/24420 [9:33:14<38:15,  1.49s/it]
                                                       
{'loss': 0.0654, 'learning_rate': 5e-05, 'epoch': 4.68}

 94%|█████████▎| 22875/24420 [9:33:14<38:15,  1.49s/it]
 94%|█████████▎| 22876/24420 [9:33:16<38:18,  1.49s/it]
 94%|█████████▎| 22877/24420 [9:33:17<38:21,  1.49s/it]
 94%|█████████▎| 22878/24420 [9:33:19<38:13,  1.49s/it]
 94%|█████████▎| 22879/24420 [9:33:20<38:14,  1.49s/it]
 94%|█████████▎| 22880/24420 [9:33:22<38:15,  1.49s/it]
                                                       
{'loss': 0.0772, 'learning_rate': 5e-05, 'epoch': 4.68}

 94%|█████████▎| 22880/24420 [9:33:22<38:15,  1.49s/it]
 94%|█████████▎| 22881/24420 [9:33:23<38:13,  1.49s/it]
 94%|█████████▎| 22882/24420 [9:33:25<38:08,  1.49s/it]
 94%|█████████▎| 22883/24420 [9:33:26<38:15,  1.49s/it]
 94%|█████████▎| 22884/24420 [9:33:28<38:11,  1.49s/it]
 94%|█████████▎| 22885/24420 [9:33:29<38:12,  1.49s/it]
                                                       
{'loss': 0.0795, 'learning_rate': 5e-05, 'epoch': 4.69}

 94%|█████████▎| 22885/24420 [9:33:29<38:12,  1.49s/it]
 94%|█████████▎| 22886/24420 [9:33:31<38:09,  1.49s/it]
 94%|█████████▎| 22887/24420 [9:33:32<38:02,  1.49s/it]
 94%|█████████▎| 22888/24420 [9:33:34<38:02,  1.49s/it]
 94%|█████████▎| 22889/24420 [9:33:35<38:05,  1.49s/it]
 94%|█████████▎| 22890/24420 [9:33:37<38:04,  1.49s/it]
                                                       
{'loss': 0.0631, 'learning_rate': 5e-05, 'epoch': 4.69}

 94%|█████████▎| 22890/24420 [9:33:37<38:04,  1.49s/it]
 94%|█████████▎| 22891/24420 [9:33:38<38:12,  1.50s/it]
 94%|█████████▎| 22892/24420 [9:33:40<38:06,  1.50s/it]
 94%|█████████▎| 22893/24420 [9:33:41<37:59,  1.49s/it]
 94%|█████████▍| 22894/24420 [9:33:43<37:59,  1.49s/it]
 94%|█████████▍| 22895/24420 [9:33:44<37:51,  1.49s/it]
                                                       
{'loss': 0.0686, 'learning_rate': 5e-05, 'epoch': 4.69}

 94%|█████████▍| 22895/24420 [9:33:44<37:51,  1.49s/it]
 94%|█████████▍| 22896/24420 [9:33:46<37:51,  1.49s/it]
 94%|█████████▍| 22897/24420 [9:33:47<37:44,  1.49s/it]
 94%|█████████▍| 22898/24420 [9:33:49<37:41,  1.49s/it]
 94%|█████████▍| 22899/24420 [9:33:50<37:41,  1.49s/it]
 94%|█████████▍| 22900/24420 [9:33:52<37:41,  1.49s/it]
                                                       
{'loss': 0.0603, 'learning_rate': 5e-05, 'epoch': 4.69}

 94%|█████████▍| 22900/24420 [9:33:52<37:41,  1.49s/it]
 94%|█████████▍| 22901/24420 [9:33:53<37:40,  1.49s/it]
 94%|█████████▍| 22902/24420 [9:33:55<37:36,  1.49s/it]
 94%|█████████▍| 22903/24420 [9:33:56<37:36,  1.49s/it]
 94%|█████████▍| 22904/24420 [9:33:58<37:34,  1.49s/it]
 94%|█████████▍| 22905/24420 [9:33:59<37:31,  1.49s/it]
                                                       
{'loss': 0.0725, 'learning_rate': 5e-05, 'epoch': 4.69}

 94%|█████████▍| 22905/24420 [9:33:59<37:31,  1.49s/it]
 94%|█████████▍| 22906/24420 [9:34:01<37:27,  1.48s/it]
 94%|█████████▍| 22907/24420 [9:34:02<37:23,  1.48s/it]
 94%|█████████▍| 22908/24420 [9:34:04<37:20,  1.48s/it]
 94%|█████████▍| 22909/24420 [9:34:05<37:19,  1.48s/it]
 94%|█████████▍| 22910/24420 [9:34:07<37:17,  1.48s/it]
                                                       
{'loss': 0.0596, 'learning_rate': 5e-05, 'epoch': 4.69}

 94%|█████████▍| 22910/24420 [9:34:07<37:17,  1.48s/it]
 94%|█████████▍| 22911/24420 [9:34:08<37:20,  1.48s/it]
 94%|█████████▍| 22912/24420 [9:34:10<37:23,  1.49s/it]
 94%|█████████▍| 22913/24420 [9:34:11<37:21,  1.49s/it]
 94%|█████████▍| 22914/24420 [9:34:13<37:24,  1.49s/it]
 94%|█████████▍| 22915/24420 [9:34:14<37:23,  1.49s/it]
                                                       
{'loss': 0.0715, 'learning_rate': 5e-05, 'epoch': 4.69}

 94%|█████████▍| 22915/24420 [9:34:14<37:23,  1.49s/it]
 94%|█████████▍| 22916/24420 [9:34:16<37:25,  1.49s/it]
 94%|█████████▍| 22917/24420 [9:34:17<37:22,  1.49s/it]
 94%|█████████▍| 22918/24420 [9:34:19<37:16,  1.49s/it]
 94%|█████████▍| 22919/24420 [9:34:20<37:15,  1.49s/it]
 94%|█████████▍| 22920/24420 [9:34:22<37:22,  1.49s/it]
                                                       
{'loss': 0.079, 'learning_rate': 5e-05, 'epoch': 4.69}

 94%|█████████▍| 22920/24420 [9:34:22<37:22,  1.49s/it]
 94%|█████████▍| 22921/24420 [9:34:23<38:03,  1.52s/it]
 94%|█████████▍| 22922/24420 [9:34:25<37:44,  1.51s/it]
 94%|█████████▍| 22923/24420 [9:34:26<37:35,  1.51s/it]
 94%|█████████▍| 22924/24420 [9:34:28<37:28,  1.50s/it]
 94%|█████████▍| 22925/24420 [9:34:29<37:18,  1.50s/it]
                                                       
{'loss': 0.0666, 'learning_rate': 5e-05, 'epoch': 4.69}

 94%|█████████▍| 22925/24420 [9:34:29<37:18,  1.50s/it]
 94%|█████████▍| 22926/24420 [9:34:31<37:16,  1.50s/it]
 94%|█████████▍| 22927/24420 [9:34:32<37:08,  1.49s/it]
 94%|█████████▍| 22928/24420 [9:34:34<37:00,  1.49s/it]
 94%|█████████▍| 22929/24420 [9:34:35<36:55,  1.49s/it]
 94%|█████████▍| 22930/24420 [9:34:37<36:56,  1.49s/it]
                                                       
{'loss': 0.0736, 'learning_rate': 5e-05, 'epoch': 4.69}

 94%|█████████▍| 22930/24420 [9:34:37<36:56,  1.49s/it]
 94%|█████████▍| 22931/24420 [9:34:38<36:57,  1.49s/it]
 94%|█████████▍| 22932/24420 [9:34:39<36:51,  1.49s/it]
 94%|█████████▍| 22933/24420 [9:34:41<36:52,  1.49s/it]
 94%|█████████▍| 22934/24420 [9:34:42<36:52,  1.49s/it]
 94%|█████████▍| 22935/24420 [9:34:44<36:51,  1.49s/it]
                                                       
{'loss': 0.0704, 'learning_rate': 5e-05, 'epoch': 4.7}

 94%|█████████▍| 22935/24420 [9:34:44<36:51,  1.49s/it]
 94%|█████████▍| 22936/24420 [9:34:45<36:47,  1.49s/it]
 94%|█████████▍| 22937/24420 [9:34:47<36:42,  1.49s/it]
 94%|█████████▍| 22938/24420 [9:34:48<36:41,  1.49s/it]
 94%|█████████▍| 22939/24420 [9:34:50<36:37,  1.48s/it]
 94%|█████████▍| 22940/24420 [9:34:51<36:36,  1.48s/it]
                                                       
{'loss': 0.065, 'learning_rate': 5e-05, 'epoch': 4.7}

 94%|█████████▍| 22940/24420 [9:34:51<36:36,  1.48s/it]
 94%|█████████▍| 22941/24420 [9:34:53<36:39,  1.49s/it]
 94%|█████████▍| 22942/24420 [9:34:54<36:35,  1.49s/it]
 94%|█████████▍| 22943/24420 [9:34:56<36:32,  1.48s/it]
 94%|█████████▍| 22944/24420 [9:34:57<36:30,  1.48s/it]
 94%|█████████▍| 22945/24420 [9:34:59<36:28,  1.48s/it]
                                                       
{'loss': 0.0676, 'learning_rate': 5e-05, 'epoch': 4.7}

 94%|█████████▍| 22945/24420 [9:34:59<36:28,  1.48s/it]
 94%|█████████▍| 22946/24420 [9:35:00<36:32,  1.49s/it]
 94%|█████████▍| 22947/24420 [9:35:02<36:27,  1.49s/it]
 94%|█████████▍| 22948/24420 [9:35:03<36:29,  1.49s/it]
 94%|█████████▍| 22949/24420 [9:35:05<36:32,  1.49s/it]
 94%|█████████▍| 22950/24420 [9:35:06<36:28,  1.49s/it]
                                                       
{'loss': 0.0611, 'learning_rate': 5e-05, 'epoch': 4.7}

 94%|█████████▍| 22950/24420 [9:35:06<36:28,  1.49s/it]
 94%|█████████▍| 22951/24420 [9:35:08<36:28,  1.49s/it]
 94%|█████████▍| 22952/24420 [9:35:09<36:21,  1.49s/it]
 94%|█████████▍| 22953/24420 [9:35:11<36:20,  1.49s/it]
 94%|█████████▍| 22954/24420 [9:35:12<36:15,  1.48s/it]
 94%|█████████▍| 22955/24420 [9:35:14<36:13,  1.48s/it]
                                                       
{'loss': 0.0651, 'learning_rate': 5e-05, 'epoch': 4.7}

 94%|█████████▍| 22955/24420 [9:35:14<36:13,  1.48s/it]
 94%|█████████▍| 22956/24420 [9:35:15<36:15,  1.49s/it]
 94%|█████████▍| 22957/24420 [9:35:17<36:13,  1.49s/it]
 94%|█████████▍| 22958/24420 [9:35:18<36:11,  1.49s/it]
 94%|█████████▍| 22959/24420 [9:35:20<36:09,  1.48s/it]
 94%|█████████▍| 22960/24420 [9:35:21<36:08,  1.49s/it]
                                                       
{'loss': 0.0716, 'learning_rate': 5e-05, 'epoch': 4.7}

 94%|█████████▍| 22960/24420 [9:35:21<36:08,  1.49s/it]
 94%|█████████▍| 22961/24420 [9:35:23<36:06,  1.49s/it]
 94%|█████████▍| 22962/24420 [9:35:24<36:07,  1.49s/it]
 94%|█████████▍| 22963/24420 [9:35:26<36:15,  1.49s/it]
 94%|█████████▍| 22964/24420 [9:35:27<36:14,  1.49s/it]
 94%|█████████▍| 22965/24420 [9:35:29<36:10,  1.49s/it]
                                                       
{'loss': 0.0987, 'learning_rate': 5e-05, 'epoch': 4.7}

 94%|█████████▍| 22965/24420 [9:35:29<36:10,  1.49s/it]
 94%|█████████▍| 22966/24420 [9:35:30<36:06,  1.49s/it]
 94%|█████████▍| 22967/24420 [9:35:32<36:05,  1.49s/it]
 94%|█████████▍| 22968/24420 [9:35:33<36:01,  1.49s/it]
 94%|█████████▍| 22969/24420 [9:35:35<35:58,  1.49s/it]
 94%|█████████▍| 22970/24420 [9:35:36<35:56,  1.49s/it]
                                                       
{'loss': 0.0832, 'learning_rate': 5e-05, 'epoch': 4.7}

 94%|█████████▍| 22970/24420 [9:35:36<35:56,  1.49s/it]
 94%|█████████▍| 22971/24420 [9:35:37<36:03,  1.49s/it]
 94%|█████████▍| 22972/24420 [9:35:39<36:00,  1.49s/it]
 94%|█████████▍| 22973/24420 [9:35:40<35:56,  1.49s/it]
 94%|█████████▍| 22974/24420 [9:35:42<35:55,  1.49s/it]
 94%|█████████▍| 22975/24420 [9:35:43<35:52,  1.49s/it]
                                                       
{'loss': 0.0715, 'learning_rate': 5e-05, 'epoch': 4.7}

 94%|█████████▍| 22975/24420 [9:35:43<35:52,  1.49s/it]
 94%|█████████▍| 22976/24420 [9:35:45<35:51,  1.49s/it]
 94%|█████████▍| 22977/24420 [9:35:46<35:51,  1.49s/it]
 94%|█████████▍| 22978/24420 [9:35:48<35:47,  1.49s/it]
 94%|█████████▍| 22979/24420 [9:35:49<35:44,  1.49s/it]
 94%|█████████▍| 22980/24420 [9:35:51<35:42,  1.49s/it]
                                                       
{'loss': 0.0816, 'learning_rate': 5e-05, 'epoch': 4.71}

 94%|█████████▍| 22980/24420 [9:35:51<35:42,  1.49s/it]
 94%|█████████▍| 22981/24420 [9:35:52<35:39,  1.49s/it]
 94%|█████████▍| 22982/24420 [9:35:54<35:35,  1.49s/it]
 94%|█████████▍| 22983/24420 [9:35:55<35:35,  1.49s/it]
 94%|█████████▍| 22984/24420 [9:35:57<35:30,  1.48s/it]
 94%|█████████▍| 22985/24420 [9:35:58<35:33,  1.49s/it]
                                                       
{'loss': 0.0817, 'learning_rate': 5e-05, 'epoch': 4.71}

 94%|█████████▍| 22985/24420 [9:35:58<35:33,  1.49s/it]
 94%|█████████▍| 22986/24420 [9:36:00<35:35,  1.49s/it]
 94%|█████████▍| 22987/24420 [9:36:01<35:29,  1.49s/it]
 94%|█████████▍| 22988/24420 [9:36:03<35:26,  1.49s/it]
 94%|█████████▍| 22989/24420 [9:36:04<35:24,  1.48s/it]
 94%|█████████▍| 22990/24420 [9:36:06<35:22,  1.48s/it]
                                                       
{'loss': 0.0692, 'learning_rate': 5e-05, 'epoch': 4.71}

 94%|█████████▍| 22990/24420 [9:36:06<35:22,  1.48s/it]
 94%|█████████▍| 22991/24420 [9:36:07<35:19,  1.48s/it]
 94%|█████████▍| 22992/24420 [9:36:09<35:21,  1.49s/it]
 94%|█████████▍| 22993/24420 [9:36:10<35:15,  1.48s/it]
 94%|█████████▍| 22994/24420 [9:36:12<35:17,  1.49s/it]
 94%|█████████▍| 22995/24420 [9:36:13<35:18,  1.49s/it]
                                                       
{'loss': 0.0767, 'learning_rate': 5e-05, 'epoch': 4.71}

 94%|█████████▍| 22995/24420 [9:36:13<35:18,  1.49s/it]
 94%|█████████▍| 22996/24420 [9:36:15<35:20,  1.49s/it]
 94%|█████████▍| 22997/24420 [9:36:16<35:16,  1.49s/it]
 94%|█████████▍| 22998/24420 [9:36:18<35:16,  1.49s/it]
 94%|█████████▍| 22999/24420 [9:36:19<35:16,  1.49s/it]
 94%|█████████▍| 23000/24420 [9:36:21<35:16,  1.49s/it]
                                                       
{'loss': 0.0971, 'learning_rate': 5e-05, 'epoch': 4.71}

 94%|█████████▍| 23000/24420 [9:36:21<35:16,  1.49s/it]
 94%|█████████▍| 23001/24420 [9:36:25<57:51,  2.45s/it]
 94%|█████████▍| 23002/24420 [9:36:27<51:00,  2.16s/it]
 94%|█████████▍| 23003/24420 [9:36:28<46:10,  1.96s/it]
 94%|█████████▍| 23004/24420 [9:36:30<42:58,  1.82s/it]
 94%|█████████▍| 23005/24420 [9:36:31<40:44,  1.73s/it]
                                                       
{'loss': 0.0796, 'learning_rate': 5e-05, 'epoch': 4.71}

 94%|█████████▍| 23005/24420 [9:36:31<40:44,  1.73s/it]
 94%|█████████▍| 23006/24420 [9:36:33<39:01,  1.66s/it]
 94%|█████████▍| 23007/24420 [9:36:34<37:48,  1.61s/it]
 94%|█████████▍| 23008/24420 [9:36:36<36:54,  1.57s/it]
 94%|█████████▍| 23009/24420 [9:36:37<36:12,  1.54s/it]
 94%|█████████▍| 23010/24420 [9:36:39<35:47,  1.52s/it]
                                                       
{'loss': 0.0699, 'learning_rate': 5e-05, 'epoch': 4.71}

 94%|█████████▍| 23010/24420 [9:36:39<35:47,  1.52s/it]
 94%|█████████▍| 23011/24420 [9:36:40<35:29,  1.51s/it]
 94%|█████████▍| 23012/24420 [9:36:42<35:16,  1.50s/it]
 94%|█████████▍| 23013/24420 [9:36:43<35:06,  1.50s/it]
 94%|█████████▍| 23014/24420 [9:36:45<35:02,  1.50s/it]
 94%|█████████▍| 23015/24420 [9:36:46<34:58,  1.49s/it]
                                                       
{'loss': 0.0711, 'learning_rate': 5e-05, 'epoch': 4.71}

 94%|█████████▍| 23015/24420 [9:36:46<34:58,  1.49s/it]
 94%|█████████▍| 23016/24420 [9:36:48<34:51,  1.49s/it]
 94%|█████████▍| 23017/24420 [9:36:49<34:49,  1.49s/it]
 94%|█████████▍| 23018/24420 [9:36:51<34:46,  1.49s/it]
 94%|█████████▍| 23019/24420 [9:36:52<34:42,  1.49s/it]
 94%|█████████▍| 23020/24420 [9:36:54<34:40,  1.49s/it]
                                                       
{'loss': 0.0796, 'learning_rate': 5e-05, 'epoch': 4.71}

 94%|█████████▍| 23020/24420 [9:36:54<34:40,  1.49s/it]
 94%|█████████▍| 23021/24420 [9:36:55<34:43,  1.49s/it]
 94%|█████████▍| 23022/24420 [9:36:57<34:38,  1.49s/it]
 94%|█████████▍| 23023/24420 [9:36:58<34:37,  1.49s/it]
 94%|█████████▍| 23024/24420 [9:37:00<34:38,  1.49s/it]
 94%|█████████▍| 23025/24420 [9:37:01<34:36,  1.49s/it]
                                                       
{'loss': 0.0657, 'learning_rate': 5e-05, 'epoch': 4.71}

 94%|█████████▍| 23025/24420 [9:37:01<34:36,  1.49s/it]
 94%|█████████▍| 23026/24420 [9:37:02<34:35,  1.49s/it]
 94%|█████████▍| 23027/24420 [9:37:04<34:31,  1.49s/it]
 94%|█████████▍| 23028/24420 [9:37:05<34:28,  1.49s/it]
 94%|█████████▍| 23029/24420 [9:37:07<34:29,  1.49s/it]
 94%|█████████▍| 23030/24420 [9:37:08<34:30,  1.49s/it]
                                                       
{'loss': 0.061, 'learning_rate': 5e-05, 'epoch': 4.72}

 94%|█████████▍| 23030/24420 [9:37:08<34:30,  1.49s/it]
 94%|█████████▍| 23031/24420 [9:37:10<34:25,  1.49s/it]
 94%|█████████▍| 23032/24420 [9:37:11<34:22,  1.49s/it]
 94%|█████████▍| 23033/24420 [9:37:13<34:27,  1.49s/it]
 94%|█████████▍| 23034/24420 [9:37:14<34:27,  1.49s/it]
 94%|█████████▍| 23035/24420 [9:37:16<34:22,  1.49s/it]
                                                       
{'loss': 0.0764, 'learning_rate': 5e-05, 'epoch': 4.72}

 94%|█████████▍| 23035/24420 [9:37:16<34:22,  1.49s/it]
 94%|█████████▍| 23036/24420 [9:37:17<34:21,  1.49s/it]
 94%|█████████▍| 23037/24420 [9:37:19<34:15,  1.49s/it]
 94%|█████████▍| 23038/24420 [9:37:20<34:14,  1.49s/it]
 94%|█████████▍| 23039/24420 [9:37:22<34:07,  1.48s/it]
 94%|█████████▍| 23040/24420 [9:37:23<34:08,  1.48s/it]
                                                       
{'loss': 0.0816, 'learning_rate': 5e-05, 'epoch': 4.72}

 94%|█████████▍| 23040/24420 [9:37:23<34:08,  1.48s/it]
 94%|█████████▍| 23041/24420 [9:37:25<34:10,  1.49s/it]
 94%|█████████▍| 23042/24420 [9:37:26<34:09,  1.49s/it]
 94%|█████████▍| 23043/24420 [9:37:28<34:09,  1.49s/it]
 94%|█████████▍| 23044/24420 [9:37:29<34:05,  1.49s/it]
 94%|█████████▍| 23045/24420 [9:37:31<34:03,  1.49s/it]
                                                       
{'loss': 0.0703, 'learning_rate': 5e-05, 'epoch': 4.72}

 94%|█████████▍| 23045/24420 [9:37:31<34:03,  1.49s/it]
 94%|█████████▍| 23046/24420 [9:37:32<34:03,  1.49s/it]
 94%|█████████▍| 23047/24420 [9:37:34<33:59,  1.49s/it]
 94%|█████████▍| 23048/24420 [9:37:35<33:59,  1.49s/it]
 94%|█████████▍| 23049/24420 [9:37:37<33:59,  1.49s/it]
 94%|█████████▍| 23050/24420 [9:37:38<33:56,  1.49s/it]
                                                       
{'loss': 0.0591, 'learning_rate': 5e-05, 'epoch': 4.72}

 94%|█████████▍| 23050/24420 [9:37:38<33:56,  1.49s/it]
 94%|█████████▍| 23051/24420 [9:37:40<33:57,  1.49s/it]
 94%|█████████▍| 23052/24420 [9:37:41<33:55,  1.49s/it]
 94%|█████████▍| 23053/24420 [9:37:43<33:53,  1.49s/it]
 94%|█████████▍| 23054/24420 [9:37:44<33:50,  1.49s/it]
 94%|█████████▍| 23055/24420 [9:37:46<33:47,  1.49s/it]
                                                       
{'loss': 0.0729, 'learning_rate': 5e-05, 'epoch': 4.72}

 94%|█████████▍| 23055/24420 [9:37:46<33:47,  1.49s/it]
 94%|█████████▍| 23056/24420 [9:37:47<33:46,  1.49s/it]
 94%|█████████▍| 23057/24420 [9:37:49<33:54,  1.49s/it]
 94%|█████████▍| 23058/24420 [9:37:50<33:51,  1.49s/it]
 94%|█████████▍| 23059/24420 [9:37:52<33:50,  1.49s/it]
 94%|█████████▍| 23060/24420 [9:37:53<33:50,  1.49s/it]
                                                       
{'loss': 0.0776, 'learning_rate': 5e-05, 'epoch': 4.72}

 94%|█████████▍| 23060/24420 [9:37:53<33:50,  1.49s/it]
 94%|█████████▍| 23061/24420 [9:37:55<33:44,  1.49s/it]
 94%|█████████▍| 23062/24420 [9:37:56<33:41,  1.49s/it]
 94%|█████████▍| 23063/24420 [9:37:58<33:38,  1.49s/it]
 94%|█████████▍| 23064/24420 [9:37:59<33:39,  1.49s/it]
 94%|█████████▍| 23065/24420 [9:38:01<33:35,  1.49s/it]
                                                       
{'loss': 0.0626, 'learning_rate': 5e-05, 'epoch': 4.72}

 94%|█████████▍| 23065/24420 [9:38:01<33:35,  1.49s/it]
 94%|█████████▍| 23066/24420 [9:38:02<33:37,  1.49s/it]
 94%|█████████▍| 23067/24420 [9:38:03<33:33,  1.49s/it]
 94%|█████████▍| 23068/24420 [9:38:05<33:29,  1.49s/it]
 94%|█████████▍| 23069/24420 [9:38:06<33:28,  1.49s/it]
 94%|█████████▍| 23070/24420 [9:38:08<33:31,  1.49s/it]
                                                       
{'loss': 0.0663, 'learning_rate': 5e-05, 'epoch': 4.72}

 94%|█████████▍| 23070/24420 [9:38:08<33:31,  1.49s/it]
 94%|█████████▍| 23071/24420 [9:38:09<33:33,  1.49s/it]
 94%|█████████▍| 23072/24420 [9:38:11<33:31,  1.49s/it]
 94%|█████████▍| 23073/24420 [9:38:12<33:27,  1.49s/it]
 94%|█████████▍| 23074/24420 [9:38:14<33:23,  1.49s/it]
 94%|█████████▍| 23075/24420 [9:38:15<33:22,  1.49s/it]
                                                       
{'loss': 0.0688, 'learning_rate': 5e-05, 'epoch': 4.72}

 94%|█████████▍| 23075/24420 [9:38:15<33:22,  1.49s/it]
 94%|█████████▍| 23076/24420 [9:38:17<33:22,  1.49s/it]
 95%|█████████▍| 23077/24420 [9:38:18<33:17,  1.49s/it]
 95%|█████████▍| 23078/24420 [9:38:20<33:16,  1.49s/it]
 95%|█████████▍| 23079/24420 [9:38:21<33:14,  1.49s/it]
 95%|█████████▍| 23080/24420 [9:38:23<33:20,  1.49s/it]
                                                       
{'loss': 0.0664, 'learning_rate': 5e-05, 'epoch': 4.73}

 95%|█████████▍| 23080/24420 [9:38:23<33:20,  1.49s/it]
 95%|█████████▍| 23081/24420 [9:38:24<33:16,  1.49s/it]
 95%|█████████▍| 23082/24420 [9:38:26<33:16,  1.49s/it]
 95%|█████████▍| 23083/24420 [9:38:27<33:13,  1.49s/it]
 95%|█████████▍| 23084/24420 [9:38:29<33:11,  1.49s/it]
 95%|█████████▍| 23085/24420 [9:38:30<33:08,  1.49s/it]
                                                       
{'loss': 0.0659, 'learning_rate': 5e-05, 'epoch': 4.73}

 95%|█████████▍| 23085/24420 [9:38:30<33:08,  1.49s/it]
 95%|█████████▍| 23086/24420 [9:38:32<33:05,  1.49s/it]
 95%|█████████▍| 23087/24420 [9:38:33<33:08,  1.49s/it]
 95%|█████████▍| 23088/24420 [9:38:35<33:08,  1.49s/it]
 95%|█████████▍| 23089/24420 [9:38:36<33:08,  1.49s/it]
 95%|█████████▍| 23090/24420 [9:38:38<33:05,  1.49s/it]
                                                       
{'loss': 0.0661, 'learning_rate': 5e-05, 'epoch': 4.73}

 95%|█████████▍| 23090/24420 [9:38:38<33:05,  1.49s/it]
 95%|█████████▍| 23091/24420 [9:38:39<33:03,  1.49s/it]
 95%|█████████▍| 23092/24420 [9:38:41<33:05,  1.50s/it]
 95%|█████████▍| 23093/24420 [9:38:42<33:04,  1.50s/it]
 95%|█████████▍| 23094/24420 [9:38:44<33:03,  1.50s/it]
 95%|█████████▍| 23095/24420 [9:38:45<33:08,  1.50s/it]
                                                       
{'loss': 0.0951, 'learning_rate': 5e-05, 'epoch': 4.73}

 95%|█████████▍| 23095/24420 [9:38:45<33:08,  1.50s/it]
 95%|█████████▍| 23096/24420 [9:38:47<33:04,  1.50s/it]
 95%|█████████▍| 23097/24420 [9:38:48<33:00,  1.50s/it]
 95%|█████████▍| 23098/24420 [9:38:50<32:56,  1.50s/it]
 95%|█████████▍| 23099/24420 [9:38:51<32:58,  1.50s/it]
 95%|█████████▍| 23100/24420 [9:38:53<32:51,  1.49s/it]
                                                       
{'loss': 0.0828, 'learning_rate': 5e-05, 'epoch': 4.73}

 95%|█████████▍| 23100/24420 [9:38:53<32:51,  1.49s/it]
 95%|█████████▍| 23101/24420 [9:38:54<32:50,  1.49s/it]
 95%|█████████▍| 23102/24420 [9:38:56<32:46,  1.49s/it]
 95%|█████████▍| 23103/24420 [9:38:57<32:45,  1.49s/it]
 95%|█████████▍| 23104/24420 [9:38:59<32:41,  1.49s/it]
 95%|█████████▍| 23105/24420 [9:39:00<32:39,  1.49s/it]
                                                       
{'loss': 0.0831, 'learning_rate': 5e-05, 'epoch': 4.73}

 95%|█████████▍| 23105/24420 [9:39:00<32:39,  1.49s/it]
 95%|█████████▍| 23106/24420 [9:39:02<32:38,  1.49s/it]
 95%|█████████▍| 23107/24420 [9:39:03<32:36,  1.49s/it]
 95%|█████████▍| 23108/24420 [9:39:05<32:33,  1.49s/it]
 95%|█████████▍| 23109/24420 [9:39:06<32:31,  1.49s/it]
 95%|█████████▍| 23110/24420 [9:39:08<32:30,  1.49s/it]
                                                       
{'loss': 0.0753, 'learning_rate': 5e-05, 'epoch': 4.73}

 95%|█████████▍| 23110/24420 [9:39:08<32:30,  1.49s/it]
 95%|█████████▍| 23111/24420 [9:39:09<32:28,  1.49s/it]
 95%|█████████▍| 23112/24420 [9:39:11<32:24,  1.49s/it]
 95%|█████████▍| 23113/24420 [9:39:12<32:21,  1.49s/it]
 95%|█████████▍| 23114/24420 [9:39:14<32:27,  1.49s/it]
 95%|█████████▍| 23115/24420 [9:39:15<32:22,  1.49s/it]
                                                       
{'loss': 0.0804, 'learning_rate': 5e-05, 'epoch': 4.73}

 95%|█████████▍| 23115/24420 [9:39:15<32:22,  1.49s/it]
 95%|█████████▍| 23116/24420 [9:39:17<32:24,  1.49s/it]
 95%|█████████▍| 23117/24420 [9:39:18<32:17,  1.49s/it]
 95%|█████████▍| 23118/24420 [9:39:20<32:17,  1.49s/it]
 95%|█████████▍| 23119/24420 [9:39:21<32:15,  1.49s/it]
 95%|█████████▍| 23120/24420 [9:39:23<32:19,  1.49s/it]
                                                       
{'loss': 0.0795, 'learning_rate': 5e-05, 'epoch': 4.73}

 95%|█████████▍| 23120/24420 [9:39:23<32:19,  1.49s/it]
 95%|█████████▍| 23121/24420 [9:39:24<32:16,  1.49s/it]
 95%|█████████▍| 23122/24420 [9:39:26<32:13,  1.49s/it]
 95%|█████████▍| 23123/24420 [9:39:27<32:10,  1.49s/it]
 95%|█████████▍| 23124/24420 [9:39:28<32:09,  1.49s/it]
 95%|█████████▍| 23125/24420 [9:39:30<32:08,  1.49s/it]
                                                       
{'loss': 0.0724, 'learning_rate': 5e-05, 'epoch': 4.73}

 95%|█████████▍| 23125/24420 [9:39:30<32:08,  1.49s/it]
 95%|█████████▍| 23126/24420 [9:39:31<32:06,  1.49s/it]
 95%|█████████▍| 23127/24420 [9:39:33<32:04,  1.49s/it]
 95%|█████████▍| 23128/24420 [9:39:34<32:05,  1.49s/it]
 95%|█████████▍| 23129/24420 [9:39:36<32:01,  1.49s/it]
 95%|█████████▍| 23130/24420 [9:39:37<32:02,  1.49s/it]
                                                       
{'loss': 0.09, 'learning_rate': 5e-05, 'epoch': 4.74}

 95%|█████████▍| 23130/24420 [9:39:37<32:02,  1.49s/it]
 95%|█████████▍| 23131/24420 [9:39:39<32:05,  1.49s/it]
 95%|█████████▍| 23132/24420 [9:39:40<32:00,  1.49s/it]
 95%|█████████▍| 23133/24420 [9:39:42<31:57,  1.49s/it]
 95%|█████████▍| 23134/24420 [9:39:43<31:54,  1.49s/it]
 95%|█████████▍| 23135/24420 [9:39:45<31:51,  1.49s/it]
                                                       
{'loss': 0.0843, 'learning_rate': 5e-05, 'epoch': 4.74}

 95%|█████████▍| 23135/24420 [9:39:45<31:51,  1.49s/it]
 95%|█████████▍| 23136/24420 [9:39:46<31:51,  1.49s/it]
 95%|█████████▍| 23137/24420 [9:39:48<31:45,  1.49s/it]
 95%|█████████▍| 23138/24420 [9:39:49<31:45,  1.49s/it]
 95%|█████████▍| 23139/24420 [9:39:51<31:49,  1.49s/it]
 95%|█████████▍| 23140/24420 [9:39:52<31:45,  1.49s/it]
                                                       
{'loss': 0.0739, 'learning_rate': 5e-05, 'epoch': 4.74}

 95%|█████████▍| 23140/24420 [9:39:52<31:45,  1.49s/it]
 95%|█████████▍| 23141/24420 [9:39:54<31:43,  1.49s/it]
 95%|█████████▍| 23142/24420 [9:39:55<31:38,  1.49s/it]
 95%|█████████▍| 23143/24420 [9:39:57<31:37,  1.49s/it]
 95%|█████████▍| 23144/24420 [9:39:58<31:35,  1.49s/it]
 95%|█████████▍| 23145/24420 [9:40:00<31:38,  1.49s/it]
                                                       
{'loss': 0.0753, 'learning_rate': 5e-05, 'epoch': 4.74}

 95%|█████████▍| 23145/24420 [9:40:00<31:38,  1.49s/it]
 95%|█████████▍| 23146/24420 [9:40:01<31:34,  1.49s/it]
 95%|█████████▍| 23147/24420 [9:40:03<31:32,  1.49s/it]
 95%|█████████▍| 23148/24420 [9:40:04<31:28,  1.48s/it]
 95%|█████████▍| 23149/24420 [9:40:06<31:30,  1.49s/it]
 95%|█████████▍| 23150/24420 [9:40:07<31:27,  1.49s/it]
                                                       
{'loss': 0.0616, 'learning_rate': 5e-05, 'epoch': 4.74}

 95%|█████████▍| 23150/24420 [9:40:07<31:27,  1.49s/it]
 95%|█████████▍| 23151/24420 [9:40:09<31:26,  1.49s/it]
 95%|█████████▍| 23152/24420 [9:40:10<31:27,  1.49s/it]
 95%|█████████▍| 23153/24420 [9:40:12<31:26,  1.49s/it]
 95%|█████████▍| 23154/24420 [9:40:13<31:26,  1.49s/it]
 95%|█████████▍| 23155/24420 [9:40:15<31:27,  1.49s/it]
                                                       
{'loss': 0.0701, 'learning_rate': 5e-05, 'epoch': 4.74}

 95%|█████████▍| 23155/24420 [9:40:15<31:27,  1.49s/it]
 95%|█████████▍| 23156/24420 [9:40:16<31:25,  1.49s/it]
 95%|█████████▍| 23157/24420 [9:40:18<31:25,  1.49s/it]
 95%|█████████▍| 23158/24420 [9:40:19<31:26,  1.49s/it]
 95%|█████████▍| 23159/24420 [9:40:21<31:21,  1.49s/it]
 95%|█████████▍| 23160/24420 [9:40:22<31:15,  1.49s/it]
                                                       
{'loss': 0.0752, 'learning_rate': 5e-05, 'epoch': 4.74}

 95%|█████████▍| 23160/24420 [9:40:22<31:15,  1.49s/it]
 95%|█████████▍| 23161/24420 [9:40:24<31:15,  1.49s/it]
 95%|█████████▍| 23162/24420 [9:40:25<31:14,  1.49s/it]
 95%|█████████▍| 23163/24420 [9:40:27<31:09,  1.49s/it]
 95%|█████████▍| 23164/24420 [9:40:28<31:07,  1.49s/it]
 95%|█████████▍| 23165/24420 [9:40:30<31:08,  1.49s/it]
                                                       
{'loss': 0.0758, 'learning_rate': 5e-05, 'epoch': 4.74}

 95%|█████████▍| 23165/24420 [9:40:30<31:08,  1.49s/it]
 95%|█████████▍| 23166/24420 [9:40:31<31:07,  1.49s/it]
 95%|█████████▍| 23167/24420 [9:40:33<31:06,  1.49s/it]
 95%|█████████▍| 23168/24420 [9:40:34<31:06,  1.49s/it]
 95%|█████████▍| 23169/24420 [9:40:35<31:02,  1.49s/it]
 95%|█████████▍| 23170/24420 [9:40:37<30:56,  1.49s/it]
                                                       
{'loss': 0.0715, 'learning_rate': 5e-05, 'epoch': 4.74}

 95%|█████████▍| 23170/24420 [9:40:37<30:56,  1.49s/it]
 95%|█████████▍| 23171/24420 [9:40:38<30:59,  1.49s/it]
 95%|█████████▍| 23172/24420 [9:40:40<31:02,  1.49s/it]
 95%|█████████▍| 23173/24420 [9:40:41<31:01,  1.49s/it]
 95%|█████████▍| 23174/24420 [9:40:43<30:58,  1.49s/it]
 95%|█████████▍| 23175/24420 [9:40:44<30:55,  1.49s/it]
                                                       
{'loss': 0.0774, 'learning_rate': 5e-05, 'epoch': 4.75}

 95%|█████████▍| 23175/24420 [9:40:44<30:55,  1.49s/it]
 95%|█████████▍| 23176/24420 [9:40:46<30:56,  1.49s/it]
 95%|█████████▍| 23177/24420 [9:40:47<30:56,  1.49s/it]
 95%|█████████▍| 23178/24420 [9:40:49<30:54,  1.49s/it]
 95%|█████████▍| 23179/24420 [9:40:50<30:52,  1.49s/it]
 95%|█████████▍| 23180/24420 [9:40:52<30:47,  1.49s/it]
                                                       
{'loss': 0.0832, 'learning_rate': 5e-05, 'epoch': 4.75}

 95%|█████████▍| 23180/24420 [9:40:52<30:47,  1.49s/it]
 95%|█████████▍| 23181/24420 [9:40:53<30:49,  1.49s/it]
 95%|█████████▍| 23182/24420 [9:40:55<30:43,  1.49s/it]
 95%|█████████▍| 23183/24420 [9:40:56<30:42,  1.49s/it]
 95%|█████████▍| 23184/24420 [9:40:58<30:39,  1.49s/it]
 95%|█████████▍| 23185/24420 [9:40:59<30:36,  1.49s/it]
                                                       
{'loss': 0.0756, 'learning_rate': 5e-05, 'epoch': 4.75}

 95%|█████████▍| 23185/24420 [9:40:59<30:36,  1.49s/it]
 95%|█████████▍| 23186/24420 [9:41:01<30:37,  1.49s/it]
 95%|█████████▍| 23187/24420 [9:41:02<30:35,  1.49s/it]
 95%|█████████▍| 23188/24420 [9:41:04<30:35,  1.49s/it]
 95%|█████████▍| 23189/24420 [9:41:05<30:34,  1.49s/it]
 95%|█████████▍| 23190/24420 [9:41:07<30:32,  1.49s/it]
                                                       
{'loss': 0.0693, 'learning_rate': 5e-05, 'epoch': 4.75}

 95%|█████████▍| 23190/24420 [9:41:07<30:32,  1.49s/it]
 95%|█████████▍| 23191/24420 [9:41:08<30:28,  1.49s/it]
 95%|█████████▍| 23192/24420 [9:41:10<30:23,  1.49s/it]
 95%|█████████▍| 23193/24420 [9:41:11<30:22,  1.49s/it]
 95%|█████████▍| 23194/24420 [9:41:13<30:23,  1.49s/it]
 95%|█████████▍| 23195/24420 [9:41:14<30:23,  1.49s/it]
                                                       
{'loss': 0.0712, 'learning_rate': 5e-05, 'epoch': 4.75}

 95%|█████████▍| 23195/24420 [9:41:14<30:23,  1.49s/it]
 95%|█████████▍| 23196/24420 [9:41:16<30:22,  1.49s/it]
 95%|█████████▍| 23197/24420 [9:41:17<30:24,  1.49s/it]
 95%|█████████▍| 23198/24420 [9:41:19<30:20,  1.49s/it]
 95%|█████████▌| 23199/24420 [9:41:20<30:16,  1.49s/it]
 95%|█████████▌| 23200/24420 [9:41:22<30:17,  1.49s/it]
                                                       
{'loss': 0.0713, 'learning_rate': 5e-05, 'epoch': 4.75}

 95%|█████████▌| 23200/24420 [9:41:22<30:17,  1.49s/it]
 95%|█████████▌| 23201/24420 [9:41:26<49:43,  2.45s/it]
 95%|█████████▌| 23202/24420 [9:41:28<43:50,  2.16s/it]
 95%|█████████▌| 23203/24420 [9:41:29<39:43,  1.96s/it]
 95%|█████████▌| 23204/24420 [9:41:31<36:49,  1.82s/it]
 95%|█████████▌| 23205/24420 [9:41:32<34:54,  1.72s/it]
                                                       
{'loss': 0.0841, 'learning_rate': 5e-05, 'epoch': 4.75}

 95%|█████████▌| 23205/24420 [9:41:32<34:54,  1.72s/it]
 95%|█████████▌| 23206/24420 [9:41:34<33:26,  1.65s/it]
 95%|█████████▌| 23207/24420 [9:41:35<32:31,  1.61s/it]
 95%|█████████▌| 23208/24420 [9:41:37<31:50,  1.58s/it]
 95%|█████████▌| 23209/24420 [9:41:38<31:19,  1.55s/it]
 95%|█████████▌| 23210/24420 [9:41:40<31:00,  1.54s/it]
                                                       
{'loss': 0.0825, 'learning_rate': 5e-05, 'epoch': 4.75}

 95%|█████████▌| 23210/24420 [9:41:40<31:00,  1.54s/it]
 95%|█████████▌| 23211/24420 [9:41:41<30:40,  1.52s/it]
 95%|█████████▌| 23212/24420 [9:41:43<30:26,  1.51s/it]
 95%|█████████▌| 23213/24420 [9:41:44<30:18,  1.51s/it]
 95%|█████████▌| 23214/24420 [9:41:46<30:07,  1.50s/it]
 95%|█████████▌| 23215/24420 [9:41:47<30:01,  1.49s/it]
                                                       
{'loss': 0.0745, 'learning_rate': 5e-05, 'epoch': 4.75}

 95%|█████████▌| 23215/24420 [9:41:47<30:01,  1.49s/it]
 95%|█████████▌| 23216/24420 [9:41:49<30:01,  1.50s/it]
 95%|█████████▌| 23217/24420 [9:41:50<29:54,  1.49s/it]
 95%|█████████▌| 23218/24420 [9:41:52<29:52,  1.49s/it]
 95%|█████████▌| 23219/24420 [9:41:53<29:50,  1.49s/it]
 95%|█████████▌| 23220/24420 [9:41:55<29:52,  1.49s/it]
                                                       
{'loss': 0.0726, 'learning_rate': 5e-05, 'epoch': 4.75}

 95%|█████████▌| 23220/24420 [9:41:55<29:52,  1.49s/it]
 95%|█████████▌| 23221/24420 [9:41:56<29:51,  1.49s/it]
 95%|█████████▌| 23222/24420 [9:41:58<29:47,  1.49s/it]
 95%|█████████▌| 23223/24420 [9:41:59<29:43,  1.49s/it]
 95%|█████████▌| 23224/24420 [9:42:01<29:40,  1.49s/it]
 95%|█████████▌| 23225/24420 [9:42:02<29:34,  1.49s/it]
                                                       
{'loss': 0.0787, 'learning_rate': 5e-05, 'epoch': 4.76}

 95%|█████████▌| 23225/24420 [9:42:02<29:34,  1.49s/it]
 95%|█████████▌| 23226/24420 [9:42:04<29:35,  1.49s/it]
 95%|█████████▌| 23227/24420 [9:42:05<29:38,  1.49s/it]
 95%|█████████▌| 23228/24420 [9:42:07<29:36,  1.49s/it]
 95%|█████████▌| 23229/24420 [9:42:08<29:35,  1.49s/it]
 95%|█████████▌| 23230/24420 [9:42:10<29:32,  1.49s/it]
                                                       
{'loss': 0.0757, 'learning_rate': 5e-05, 'epoch': 4.76}

 95%|█████████▌| 23230/24420 [9:42:10<29:32,  1.49s/it]
 95%|█████████▌| 23231/24420 [9:42:11<29:32,  1.49s/it]
 95%|█████████▌| 23232/24420 [9:42:13<29:28,  1.49s/it]
 95%|█████████▌| 23233/24420 [9:42:14<29:27,  1.49s/it]
 95%|█████████▌| 23234/24420 [9:42:16<29:23,  1.49s/it]
 95%|█████████▌| 23235/24420 [9:42:17<29:24,  1.49s/it]
                                                       
{'loss': 0.0674, 'learning_rate': 5e-05, 'epoch': 4.76}

 95%|█████████▌| 23235/24420 [9:42:17<29:24,  1.49s/it]
 95%|█████████▌| 23236/24420 [9:42:19<29:23,  1.49s/it]
 95%|█████████▌| 23237/24420 [9:42:20<29:22,  1.49s/it]
 95%|█████████▌| 23238/24420 [9:42:22<29:18,  1.49s/it]
 95%|█████████▌| 23239/24420 [9:42:23<29:14,  1.49s/it]
 95%|█████████▌| 23240/24420 [9:42:24<29:12,  1.49s/it]
                                                       
{'loss': 0.0662, 'learning_rate': 5e-05, 'epoch': 4.76}

 95%|█████████▌| 23240/24420 [9:42:24<29:12,  1.49s/it]
 95%|█████████▌| 23241/24420 [9:42:26<29:16,  1.49s/it]
 95%|█████████▌| 23242/24420 [9:42:27<29:16,  1.49s/it]
 95%|█████████▌| 23243/24420 [9:42:29<29:16,  1.49s/it]
 95%|█████████▌| 23244/24420 [9:42:30<29:12,  1.49s/it]
 95%|█████████▌| 23245/24420 [9:42:32<29:09,  1.49s/it]
                                                       
{'loss': 0.0693, 'learning_rate': 5e-05, 'epoch': 4.76}

 95%|█████████▌| 23245/24420 [9:42:32<29:09,  1.49s/it]
 95%|█████████▌| 23246/24420 [9:42:33<29:13,  1.49s/it]
 95%|█████████▌| 23247/24420 [9:42:35<29:10,  1.49s/it]
 95%|█████████▌| 23248/24420 [9:42:36<29:07,  1.49s/it]
 95%|█████████▌| 23249/24420 [9:42:38<29:06,  1.49s/it]
 95%|█████████▌| 23250/24420 [9:42:39<29:02,  1.49s/it]
                                                       
{'loss': 0.0627, 'learning_rate': 5e-05, 'epoch': 4.76}

 95%|█████████▌| 23250/24420 [9:42:39<29:02,  1.49s/it]
 95%|█████████▌| 23251/24420 [9:42:41<29:00,  1.49s/it]
 95%|█████████▌| 23252/24420 [9:42:42<29:00,  1.49s/it]
 95%|█████████▌| 23253/24420 [9:42:44<29:00,  1.49s/it]
 95%|█████████▌| 23254/24420 [9:42:45<29:00,  1.49s/it]
 95%|█████████▌| 23255/24420 [9:42:47<28:58,  1.49s/it]
                                                       
{'loss': 0.0884, 'learning_rate': 5e-05, 'epoch': 4.76}

 95%|█████████▌| 23255/24420 [9:42:47<28:58,  1.49s/it]
 95%|█████████▌| 23256/24420 [9:42:48<28:52,  1.49s/it]
 95%|█████████▌| 23257/24420 [9:42:50<28:53,  1.49s/it]
 95%|█████████▌| 23258/24420 [9:42:51<28:51,  1.49s/it]
 95%|█████████▌| 23259/24420 [9:42:53<28:49,  1.49s/it]
 95%|█████████▌| 23260/24420 [9:42:54<28:45,  1.49s/it]
                                                       
{'loss': 0.0638, 'learning_rate': 5e-05, 'epoch': 4.76}

 95%|█████████▌| 23260/24420 [9:42:54<28:45,  1.49s/it]
 95%|█████████▌| 23261/24420 [9:42:56<28:48,  1.49s/it]
 95%|█████████▌| 23262/24420 [9:42:57<28:46,  1.49s/it]
 95%|█████████▌| 23263/24420 [9:42:59<28:46,  1.49s/it]
 95%|█████████▌| 23264/24420 [9:43:00<28:44,  1.49s/it]
 95%|█████████▌| 23265/24420 [9:43:02<28:41,  1.49s/it]
                                                       
{'loss': 0.0894, 'learning_rate': 5e-05, 'epoch': 4.76}

 95%|█████████▌| 23265/24420 [9:43:02<28:41,  1.49s/it]
 95%|█████████▌| 23266/24420 [9:43:03<28:37,  1.49s/it]
 95%|█████████▌| 23267/24420 [9:43:05<28:38,  1.49s/it]
 95%|█████████▌| 23268/24420 [9:43:06<28:35,  1.49s/it]
 95%|█████████▌| 23269/24420 [9:43:08<28:31,  1.49s/it]
 95%|█████████▌| 23270/24420 [9:43:09<28:29,  1.49s/it]
                                                       
{'loss': 0.0711, 'learning_rate': 5e-05, 'epoch': 4.76}

 95%|█████████▌| 23270/24420 [9:43:09<28:29,  1.49s/it]
 95%|█████████▌| 23271/24420 [9:43:11<28:29,  1.49s/it]
 95%|█████████▌| 23272/24420 [9:43:12<28:28,  1.49s/it]
 95%|█████████▌| 23273/24420 [9:43:14<28:29,  1.49s/it]
 95%|█████████▌| 23274/24420 [9:43:15<28:28,  1.49s/it]
 95%|█████████▌| 23275/24420 [9:43:17<28:25,  1.49s/it]
                                                       
{'loss': 0.0705, 'learning_rate': 5e-05, 'epoch': 4.77}

 95%|█████████▌| 23275/24420 [9:43:17<28:25,  1.49s/it]
 95%|█████████▌| 23276/24420 [9:43:18<28:23,  1.49s/it]
 95%|█████████▌| 23277/24420 [9:43:20<28:19,  1.49s/it]
 95%|█████████▌| 23278/24420 [9:43:21<28:19,  1.49s/it]
 95%|█████████▌| 23279/24420 [9:43:23<28:18,  1.49s/it]
 95%|█████████▌| 23280/24420 [9:43:24<28:18,  1.49s/it]
                                                       
{'loss': 0.0877, 'learning_rate': 5e-05, 'epoch': 4.77}

 95%|█████████▌| 23280/24420 [9:43:24<28:18,  1.49s/it]
 95%|█████████▌| 23281/24420 [9:43:26<28:15,  1.49s/it]
 95%|█████████▌| 23282/24420 [9:43:27<28:09,  1.49s/it]
 95%|█████████▌| 23283/24420 [9:43:29<28:09,  1.49s/it]
 95%|█████████▌| 23284/24420 [9:43:30<28:07,  1.49s/it]
 95%|█████████▌| 23285/24420 [9:43:32<28:08,  1.49s/it]
                                                       
{'loss': 0.0747, 'learning_rate': 5e-05, 'epoch': 4.77}

 95%|█████████▌| 23285/24420 [9:43:32<28:08,  1.49s/it]
 95%|█████████▌| 23286/24420 [9:43:33<28:07,  1.49s/it]
 95%|█████████▌| 23287/24420 [9:43:34<28:03,  1.49s/it]
 95%|█████████▌| 23288/24420 [9:43:36<28:02,  1.49s/it]
 95%|█████████▌| 23289/24420 [9:43:37<28:01,  1.49s/it]
 95%|█████████▌| 23290/24420 [9:43:39<27:57,  1.48s/it]
                                                       
{'loss': 0.0687, 'learning_rate': 5e-05, 'epoch': 4.77}

 95%|█████████▌| 23290/24420 [9:43:39<27:57,  1.48s/it]
 95%|█████████▌| 23291/24420 [9:43:40<28:01,  1.49s/it]
 95%|█████████▌| 23292/24420 [9:43:42<27:57,  1.49s/it]
 95%|█████████▌| 23293/24420 [9:43:43<27:53,  1.49s/it]
 95%|█████████▌| 23294/24420 [9:43:45<27:53,  1.49s/it]
 95%|█████████▌| 23295/24420 [9:43:46<28:36,  1.53s/it]
                                                       
{'loss': 0.0755, 'learning_rate': 5e-05, 'epoch': 4.77}

 95%|█████████▌| 23295/24420 [9:43:47<28:36,  1.53s/it]
 95%|█████████▌| 23296/24420 [9:43:48<28:21,  1.51s/it]
 95%|█████████▌| 23297/24420 [9:43:49<28:09,  1.50s/it]
 95%|█████████▌| 23298/24420 [9:43:51<28:03,  1.50s/it]
 95%|█████████▌| 23299/24420 [9:43:52<27:58,  1.50s/it]
 95%|█████████▌| 23300/24420 [9:43:54<27:55,  1.50s/it]
                                                       
{'loss': 0.0877, 'learning_rate': 5e-05, 'epoch': 4.77}

 95%|█████████▌| 23300/24420 [9:43:54<27:55,  1.50s/it]
 95%|█████████▌| 23301/24420 [9:43:55<27:54,  1.50s/it]
 95%|█████████▌| 23302/24420 [9:43:57<27:51,  1.49s/it]
 95%|█████████▌| 23303/24420 [9:43:58<27:44,  1.49s/it]
 95%|█████████▌| 23304/24420 [9:44:00<27:43,  1.49s/it]
 95%|█████████▌| 23305/24420 [9:44:01<27:41,  1.49s/it]
                                                       
{'loss': 0.0807, 'learning_rate': 5e-05, 'epoch': 4.77}

 95%|█████████▌| 23305/24420 [9:44:01<27:41,  1.49s/it]
 95%|█████████▌| 23306/24420 [9:44:03<27:39,  1.49s/it]
 95%|█████████▌| 23307/24420 [9:44:04<27:33,  1.49s/it]
 95%|█████████▌| 23308/24420 [9:44:06<27:31,  1.49s/it]
 95%|█████████▌| 23309/24420 [9:44:07<27:29,  1.48s/it]
 95%|█████████▌| 23310/24420 [9:44:09<27:27,  1.48s/it]
                                                       
{'loss': 0.0703, 'learning_rate': 5e-05, 'epoch': 4.77}

 95%|█████████▌| 23310/24420 [9:44:09<27:27,  1.48s/it]
 95%|█████████▌| 23311/24420 [9:44:10<27:25,  1.48s/it]
 95%|█████████▌| 23312/24420 [9:44:12<27:25,  1.48s/it]
 95%|█████████▌| 23313/24420 [9:44:13<27:25,  1.49s/it]
 95%|█████████▌| 23314/24420 [9:44:15<27:24,  1.49s/it]
 95%|█████████▌| 23315/24420 [9:44:16<27:22,  1.49s/it]
                                                       
{'loss': 0.087, 'learning_rate': 5e-05, 'epoch': 4.77}

 95%|█████████▌| 23315/24420 [9:44:16<27:22,  1.49s/it]
 95%|█████████▌| 23316/24420 [9:44:18<27:20,  1.49s/it]
 95%|█████████▌| 23317/24420 [9:44:19<27:19,  1.49s/it]
 95%|█████████▌| 23318/24420 [9:44:21<27:16,  1.49s/it]
 95%|█████████▌| 23319/24420 [9:44:22<27:13,  1.48s/it]
 95%|█████████▌| 23320/24420 [9:44:24<27:12,  1.48s/it]
                                                       
{'loss': 0.0623, 'learning_rate': 5e-05, 'epoch': 4.77}

 95%|█████████▌| 23320/24420 [9:44:24<27:12,  1.48s/it]
 95%|█████████▌| 23321/24420 [9:44:25<27:13,  1.49s/it]
 96%|█████████▌| 23322/24420 [9:44:27<27:09,  1.48s/it]
 96%|█████████▌| 23323/24420 [9:44:28<27:09,  1.49s/it]
 96%|█████████▌| 23324/24420 [9:44:30<27:06,  1.48s/it]
 96%|█████████▌| 23325/24420 [9:44:31<27:05,  1.48s/it]
                                                       
{'loss': 0.0749, 'learning_rate': 5e-05, 'epoch': 4.78}

 96%|█████████▌| 23325/24420 [9:44:31<27:05,  1.48s/it]
 96%|█████████▌| 23326/24420 [9:44:33<27:03,  1.48s/it]
 96%|█████████▌| 23327/24420 [9:44:34<27:01,  1.48s/it]
 96%|█████████▌| 23328/24420 [9:44:36<27:01,  1.49s/it]
 96%|█████████▌| 23329/24420 [9:44:37<26:58,  1.48s/it]
 96%|█████████▌| 23330/24420 [9:44:39<26:58,  1.48s/it]
                                                       
{'loss': 0.0739, 'learning_rate': 5e-05, 'epoch': 4.78}

 96%|█████████▌| 23330/24420 [9:44:39<26:58,  1.48s/it]
 96%|█████████▌| 23331/24420 [9:44:40<26:56,  1.48s/it]
 96%|█████████▌| 23332/24420 [9:44:41<26:52,  1.48s/it]
 96%|█████████▌| 23333/24420 [9:44:43<26:51,  1.48s/it]
 96%|█████████▌| 23334/24420 [9:44:44<26:51,  1.48s/it]
 96%|█████████▌| 23335/24420 [9:44:46<26:50,  1.48s/it]
                                                       
{'loss': 0.0677, 'learning_rate': 5e-05, 'epoch': 4.78}

 96%|█████████▌| 23335/24420 [9:44:46<26:50,  1.48s/it]
 96%|█████████▌| 23336/24420 [9:44:47<26:50,  1.49s/it]
 96%|█████████▌| 23337/24420 [9:44:49<26:49,  1.49s/it]
 96%|█████████▌| 23338/24420 [9:44:50<26:48,  1.49s/it]
 96%|█████████▌| 23339/24420 [9:44:52<26:47,  1.49s/it]
 96%|█████████▌| 23340/24420 [9:44:53<26:47,  1.49s/it]
                                                       
{'loss': 0.0653, 'learning_rate': 5e-05, 'epoch': 4.78}

 96%|█████████▌| 23340/24420 [9:44:53<26:47,  1.49s/it]
 96%|█████████▌| 23341/24420 [9:44:55<26:48,  1.49s/it]
 96%|█████████▌| 23342/24420 [9:44:56<26:43,  1.49s/it]
 96%|█████████▌| 23343/24420 [9:44:58<26:46,  1.49s/it]
 96%|█████████▌| 23344/24420 [9:44:59<26:45,  1.49s/it]
 96%|█████████▌| 23345/24420 [9:45:01<26:40,  1.49s/it]
                                                       
{'loss': 0.0774, 'learning_rate': 5e-05, 'epoch': 4.78}

 96%|█████████▌| 23345/24420 [9:45:01<26:40,  1.49s/it]
 96%|█████████▌| 23346/24420 [9:45:02<26:38,  1.49s/it]
 96%|█████████▌| 23347/24420 [9:45:04<26:38,  1.49s/it]
 96%|█████████▌| 23348/24420 [9:45:05<26:35,  1.49s/it]
 96%|█████████▌| 23349/24420 [9:45:07<26:35,  1.49s/it]
 96%|█████████▌| 23350/24420 [9:45:08<26:33,  1.49s/it]
                                                       
{'loss': 0.0667, 'learning_rate': 5e-05, 'epoch': 4.78}

 96%|█████████▌| 23350/24420 [9:45:08<26:33,  1.49s/it]
 96%|█████████▌| 23351/24420 [9:45:10<26:38,  1.50s/it]
 96%|█████████▌| 23352/24420 [9:45:11<26:35,  1.49s/it]
 96%|█████████▌| 23353/24420 [9:45:13<26:33,  1.49s/it]
 96%|█████████▌| 23354/24420 [9:45:14<26:29,  1.49s/it]
 96%|█████████▌| 23355/24420 [9:45:16<26:28,  1.49s/it]
                                                       
{'loss': 0.076, 'learning_rate': 5e-05, 'epoch': 4.78}

 96%|█████████▌| 23355/24420 [9:45:16<26:28,  1.49s/it]
 96%|█████████▌| 23356/24420 [9:45:17<26:25,  1.49s/it]
 96%|█████████▌| 23357/24420 [9:45:19<26:25,  1.49s/it]
 96%|█████████▌| 23358/24420 [9:45:20<26:22,  1.49s/it]
 96%|█████████▌| 23359/24420 [9:45:22<26:18,  1.49s/it]
 96%|█████████▌| 23360/24420 [9:45:23<26:15,  1.49s/it]
                                                       
{'loss': 0.0811, 'learning_rate': 5e-05, 'epoch': 4.78}

 96%|█████████▌| 23360/24420 [9:45:23<26:15,  1.49s/it]
 96%|█████████▌| 23361/24420 [9:45:25<26:14,  1.49s/it]
 96%|█████████▌| 23362/24420 [9:45:26<26:12,  1.49s/it]
 96%|█████████▌| 23363/24420 [9:45:28<26:08,  1.48s/it]
 96%|█████████▌| 23364/24420 [9:45:29<26:06,  1.48s/it]
 96%|█████████▌| 23365/24420 [9:45:31<26:09,  1.49s/it]
                                                       
{'loss': 0.0725, 'learning_rate': 5e-05, 'epoch': 4.78}

 96%|█████████▌| 23365/24420 [9:45:31<26:09,  1.49s/it]
 96%|█████████▌| 23366/24420 [9:45:32<26:12,  1.49s/it]
 96%|█████████▌| 23367/24420 [9:45:34<26:08,  1.49s/it]
 96%|█████████▌| 23368/24420 [9:45:35<26:08,  1.49s/it]
 96%|█████████▌| 23369/24420 [9:45:37<26:11,  1.50s/it]
 96%|█████████▌| 23370/24420 [9:45:38<26:09,  1.49s/it]
                                                       
{'loss': 0.0702, 'learning_rate': 5e-05, 'epoch': 4.79}

 96%|█████████▌| 23370/24420 [9:45:38<26:09,  1.49s/it]
 96%|█████████▌| 23371/24420 [9:45:40<26:06,  1.49s/it]
 96%|█████████▌| 23372/24420 [9:45:41<26:00,  1.49s/it]
 96%|█████████▌| 23373/24420 [9:45:43<25:58,  1.49s/it]
 96%|█████████▌| 23374/24420 [9:45:44<25:54,  1.49s/it]
 96%|█████████▌| 23375/24420 [9:45:46<25:55,  1.49s/it]
                                                       
{'loss': 0.0836, 'learning_rate': 5e-05, 'epoch': 4.79}

 96%|█████████▌| 23375/24420 [9:45:46<25:55,  1.49s/it]
 96%|█████████▌| 23376/24420 [9:45:47<25:51,  1.49s/it]
 96%|█████████▌| 23377/24420 [9:45:48<25:49,  1.49s/it]
 96%|█████████▌| 23378/24420 [9:45:50<25:45,  1.48s/it]
 96%|█████████▌| 23379/24420 [9:45:51<25:44,  1.48s/it]
 96%|█████████▌| 23380/24420 [9:45:53<25:42,  1.48s/it]
                                                       
{'loss': 0.0708, 'learning_rate': 5e-05, 'epoch': 4.79}

 96%|█████████▌| 23380/24420 [9:45:53<25:42,  1.48s/it]
 96%|█████████▌| 23381/24420 [9:45:54<25:47,  1.49s/it]
 96%|█████████▌| 23382/24420 [9:45:56<25:44,  1.49s/it]
 96%|█████████▌| 23383/24420 [9:45:57<25:40,  1.49s/it]
 96%|█████████▌| 23384/24420 [9:45:59<25:40,  1.49s/it]
 96%|█████████▌| 23385/24420 [9:46:00<25:39,  1.49s/it]
                                                       
{'loss': 0.082, 'learning_rate': 5e-05, 'epoch': 4.79}

 96%|█████████▌| 23385/24420 [9:46:00<25:39,  1.49s/it]
 96%|█████████▌| 23386/24420 [9:46:02<25:39,  1.49s/it]
 96%|█████████▌| 23387/24420 [9:46:03<25:37,  1.49s/it]
 96%|█████████▌| 23388/24420 [9:46:05<25:32,  1.49s/it]
 96%|█████████▌| 23389/24420 [9:46:06<25:32,  1.49s/it]
 96%|█████████▌| 23390/24420 [9:46:08<25:30,  1.49s/it]
                                                       
{'loss': 0.071, 'learning_rate': 5e-05, 'epoch': 4.79}

 96%|█████████▌| 23390/24420 [9:46:08<25:30,  1.49s/it]
 96%|█████████▌| 23391/24420 [9:46:09<25:32,  1.49s/it]
 96%|█████████▌| 23392/24420 [9:46:11<25:30,  1.49s/it]
 96%|█████████▌| 23393/24420 [9:46:12<25:27,  1.49s/it]
 96%|█████████▌| 23394/24420 [9:46:14<25:30,  1.49s/it]
 96%|█████████▌| 23395/24420 [9:46:15<25:28,  1.49s/it]
                                                       
{'loss': 0.072, 'learning_rate': 5e-05, 'epoch': 4.79}

 96%|█████████▌| 23395/24420 [9:46:15<25:28,  1.49s/it]
 96%|█████████▌| 23396/24420 [9:46:17<25:25,  1.49s/it]
 96%|█████████▌| 23397/24420 [9:46:18<25:22,  1.49s/it]
 96%|█████████▌| 23398/24420 [9:46:20<25:20,  1.49s/it]
 96%|█████████▌| 23399/24420 [9:46:21<25:19,  1.49s/it]
 96%|█████████▌| 23400/24420 [9:46:23<25:18,  1.49s/it]
                                                       
{'loss': 0.0699, 'learning_rate': 5e-05, 'epoch': 4.79}

 96%|█████████▌| 23400/24420 [9:46:23<25:18,  1.49s/it]
 96%|█████████▌| 23401/24420 [9:46:27<41:05,  2.42s/it]
 96%|█████████▌| 23402/24420 [9:46:29<36:13,  2.14s/it]
 96%|█████████▌| 23403/24420 [9:46:30<32:50,  1.94s/it]
 96%|█████████▌| 23404/24420 [9:46:32<30:27,  1.80s/it]
 96%|█████████▌| 23405/24420 [9:46:33<28:55,  1.71s/it]
                                                       
{'loss': 0.0805, 'learning_rate': 5e-05, 'epoch': 4.79}

 96%|█████████▌| 23405/24420 [9:46:33<28:55,  1.71s/it]
 96%|█████████▌| 23406/24420 [9:46:35<27:47,  1.64s/it]
 96%|█████████▌| 23407/24420 [9:46:36<26:55,  1.60s/it]
 96%|█████████▌| 23408/24420 [9:46:38<26:21,  1.56s/it]
 96%|█████████▌| 23409/24420 [9:46:39<25:54,  1.54s/it]
 96%|█████████▌| 23410/24420 [9:46:41<25:36,  1.52s/it]
                                                       
{'loss': 0.0639, 'learning_rate': 5e-05, 'epoch': 4.79}

 96%|█████████▌| 23410/24420 [9:46:41<25:36,  1.52s/it]
 96%|█████████▌| 23411/24420 [9:46:42<25:22,  1.51s/it]
 96%|█████████▌| 23412/24420 [9:46:44<25:12,  1.50s/it]
 96%|█████████▌| 23413/24420 [9:46:45<25:09,  1.50s/it]
 96%|█████████▌| 23414/24420 [9:46:47<25:09,  1.50s/it]
 96%|█████████▌| 23415/24420 [9:46:48<25:02,  1.50s/it]
                                                       
{'loss': 0.0861, 'learning_rate': 5e-05, 'epoch': 4.79}

 96%|█████████▌| 23415/24420 [9:46:48<25:02,  1.50s/it]
 96%|█████████▌| 23416/24420 [9:46:50<24:59,  1.49s/it]
 96%|█████████▌| 23417/24420 [9:46:51<24:54,  1.49s/it]
 96%|█████████▌| 23418/24420 [9:46:53<24:51,  1.49s/it]
 96%|█████████▌| 23419/24420 [9:46:54<24:48,  1.49s/it]
 96%|█████████▌| 23420/24420 [9:46:56<24:46,  1.49s/it]
                                                       
{'loss': 0.0811, 'learning_rate': 5e-05, 'epoch': 4.8}

 96%|█████████▌| 23420/24420 [9:46:56<24:46,  1.49s/it]
 96%|█████████▌| 23421/24420 [9:46:57<24:46,  1.49s/it]
 96%|█████████▌| 23422/24420 [9:46:58<24:44,  1.49s/it]
 96%|█████████▌| 23423/24420 [9:47:00<24:42,  1.49s/it]
 96%|█████████▌| 23424/24420 [9:47:01<24:42,  1.49s/it]
 96%|█████████▌| 23425/24420 [9:47:03<24:38,  1.49s/it]
                                                       
{'loss': 0.0729, 'learning_rate': 5e-05, 'epoch': 4.8}

 96%|█████████▌| 23425/24420 [9:47:03<24:38,  1.49s/it]
 96%|█████████▌| 23426/24420 [9:47:04<24:39,  1.49s/it]
 96%|█████████▌| 23427/24420 [9:47:06<24:36,  1.49s/it]
 96%|█████████▌| 23428/24420 [9:47:07<24:35,  1.49s/it]
 96%|█████████▌| 23429/24420 [9:47:09<24:34,  1.49s/it]
 96%|█████████▌| 23430/24420 [9:47:10<24:33,  1.49s/it]
                                                       
{'loss': 0.0737, 'learning_rate': 5e-05, 'epoch': 4.8}

 96%|█████████▌| 23430/24420 [9:47:10<24:33,  1.49s/it]
 96%|█████████▌| 23431/24420 [9:47:12<24:30,  1.49s/it]
 96%|█████████▌| 23432/24420 [9:47:13<24:29,  1.49s/it]
 96%|█████████▌| 23433/24420 [9:47:15<24:27,  1.49s/it]
 96%|█████████▌| 23434/24420 [9:47:16<24:26,  1.49s/it]
 96%|█████████▌| 23435/24420 [9:47:18<24:24,  1.49s/it]
                                                       
{'loss': 0.0855, 'learning_rate': 5e-05, 'epoch': 4.8}

 96%|█████████▌| 23435/24420 [9:47:18<24:24,  1.49s/it]
 96%|█████████▌| 23436/24420 [9:47:19<24:23,  1.49s/it]
 96%|█████████▌| 23437/24420 [9:47:21<24:24,  1.49s/it]
 96%|█████████▌| 23438/24420 [9:47:22<24:22,  1.49s/it]
 96%|█████████▌| 23439/24420 [9:47:24<24:20,  1.49s/it]
 96%|█████████▌| 23440/24420 [9:47:25<24:19,  1.49s/it]
                                                       
{'loss': 0.08, 'learning_rate': 5e-05, 'epoch': 4.8}

 96%|█████████▌| 23440/24420 [9:47:25<24:19,  1.49s/it]
 96%|█████████▌| 23441/24420 [9:47:27<24:19,  1.49s/it]
 96%|█████████▌| 23442/24420 [9:47:28<24:16,  1.49s/it]
 96%|█████████▌| 23443/24420 [9:47:30<24:11,  1.49s/it]
 96%|█████████▌| 23444/24420 [9:47:31<24:12,  1.49s/it]
 96%|█████████▌| 23445/24420 [9:47:33<24:11,  1.49s/it]
                                                       
{'loss': 0.0773, 'learning_rate': 5e-05, 'epoch': 4.8}

 96%|█████████▌| 23445/24420 [9:47:33<24:11,  1.49s/it]
 96%|█████████▌| 23446/24420 [9:47:34<24:08,  1.49s/it]
 96%|█████████▌| 23447/24420 [9:47:36<24:06,  1.49s/it]
 96%|█████████▌| 23448/24420 [9:47:37<24:03,  1.48s/it]
 96%|█████████▌| 23449/24420 [9:47:39<24:02,  1.49s/it]
 96%|█████████▌| 23450/24420 [9:47:40<24:00,  1.49s/it]
                                                       
{'loss': 0.0673, 'learning_rate': 5e-05, 'epoch': 4.8}

 96%|█████████▌| 23450/24420 [9:47:40<24:00,  1.49s/it]
 96%|█████████▌| 23451/24420 [9:47:42<23:58,  1.48s/it]
 96%|█████████▌| 23452/24420 [9:47:43<23:56,  1.48s/it]
 96%|█████████▌| 23453/24420 [9:47:45<23:55,  1.48s/it]
 96%|█████████▌| 23454/24420 [9:47:46<23:53,  1.48s/it]
 96%|█████████▌| 23455/24420 [9:47:48<23:53,  1.49s/it]
                                                       
{'loss': 0.0785, 'learning_rate': 5e-05, 'epoch': 4.8}

 96%|█████████▌| 23455/24420 [9:47:48<23:53,  1.49s/it]
 96%|█████████▌| 23456/24420 [9:47:49<23:54,  1.49s/it]
 96%|█████████▌| 23457/24420 [9:47:51<23:51,  1.49s/it]
 96%|█████████▌| 23458/24420 [9:47:52<23:51,  1.49s/it]
 96%|█████████▌| 23459/24420 [9:47:54<23:50,  1.49s/it]
 96%|█████████▌| 23460/24420 [9:47:55<23:46,  1.49s/it]
                                                       
{'loss': 0.0783, 'learning_rate': 5e-05, 'epoch': 4.8}

 96%|█████████▌| 23460/24420 [9:47:55<23:46,  1.49s/it]
 96%|█████████▌| 23461/24420 [9:47:56<23:46,  1.49s/it]
 96%|█████████▌| 23462/24420 [9:47:58<23:43,  1.49s/it]
 96%|█████████▌| 23463/24420 [9:47:59<23:39,  1.48s/it]
 96%|█████████▌| 23464/24420 [9:48:01<23:38,  1.48s/it]
 96%|█████████▌| 23465/24420 [9:48:02<23:40,  1.49s/it]
                                                       
{'loss': 0.0905, 'learning_rate': 5e-05, 'epoch': 4.8}

 96%|█████████▌| 23465/24420 [9:48:02<23:40,  1.49s/it]
 96%|█████████▌| 23466/24420 [9:48:04<23:37,  1.49s/it]
 96%|█████████▌| 23467/24420 [9:48:05<23:34,  1.48s/it]
 96%|█████████▌| 23468/24420 [9:48:07<23:38,  1.49s/it]
 96%|█████████▌| 23469/24420 [9:48:08<23:36,  1.49s/it]
 96%|█████████▌| 23470/24420 [9:48:10<23:35,  1.49s/it]
                                                       
{'loss': 0.0784, 'learning_rate': 5e-05, 'epoch': 4.81}

 96%|█████████▌| 23470/24420 [9:48:10<23:35,  1.49s/it]
 96%|█████████▌| 23471/24420 [9:48:11<23:38,  1.50s/it]
 96%|█████████▌| 23472/24420 [9:48:13<23:35,  1.49s/it]
 96%|█████████▌| 23473/24420 [9:48:14<23:31,  1.49s/it]
 96%|█████████▌| 23474/24420 [9:48:16<23:32,  1.49s/it]
 96%|█████████▌| 23475/24420 [9:48:17<23:29,  1.49s/it]
                                                       
{'loss': 0.0828, 'learning_rate': 5e-05, 'epoch': 4.81}

 96%|█████████▌| 23475/24420 [9:48:17<23:29,  1.49s/it]
 96%|█████████▌| 23476/24420 [9:48:19<23:26,  1.49s/it]
 96%|█████████▌| 23477/24420 [9:48:20<23:24,  1.49s/it]
 96%|█████████▌| 23478/24420 [9:48:22<23:28,  1.50s/it]
 96%|█████████▌| 23479/24420 [9:48:23<23:25,  1.49s/it]
 96%|█████████▌| 23480/24420 [9:48:25<23:22,  1.49s/it]
                                                       
{'loss': 0.0789, 'learning_rate': 5e-05, 'epoch': 4.81}

 96%|█████████▌| 23480/24420 [9:48:25<23:22,  1.49s/it]
 96%|█████████▌| 23481/24420 [9:48:26<23:20,  1.49s/it]
 96%|█████████▌| 23482/24420 [9:48:28<23:21,  1.49s/it]
 96%|█████████▌| 23483/24420 [9:48:29<23:17,  1.49s/it]
 96%|█████████▌| 23484/24420 [9:48:31<23:16,  1.49s/it]
 96%|█████████▌| 23485/24420 [9:48:32<23:14,  1.49s/it]
                                                       
{'loss': 0.0701, 'learning_rate': 5e-05, 'epoch': 4.81}

 96%|█████████▌| 23485/24420 [9:48:32<23:14,  1.49s/it]
 96%|█████████▌| 23486/24420 [9:48:34<23:13,  1.49s/it]
 96%|█████████▌| 23487/24420 [9:48:35<23:09,  1.49s/it]
 96%|█████████▌| 23488/24420 [9:48:37<23:08,  1.49s/it]
 96%|█████████▌| 23489/24420 [9:48:38<23:06,  1.49s/it]
 96%|█████████▌| 23490/24420 [9:48:40<23:04,  1.49s/it]
                                                       
{'loss': 0.0683, 'learning_rate': 5e-05, 'epoch': 4.81}

 96%|█████████▌| 23490/24420 [9:48:40<23:04,  1.49s/it]
 96%|█████████▌| 23491/24420 [9:48:41<23:04,  1.49s/it]
 96%|█████████▌| 23492/24420 [9:48:43<23:05,  1.49s/it]
 96%|█████████▌| 23493/24420 [9:48:44<23:02,  1.49s/it]
 96%|█████████▌| 23494/24420 [9:48:46<22:58,  1.49s/it]
 96%|█████████▌| 23495/24420 [9:48:47<22:57,  1.49s/it]
                                                       
{'loss': 0.0855, 'learning_rate': 5e-05, 'epoch': 4.81}

 96%|█████████▌| 23495/24420 [9:48:47<22:57,  1.49s/it]
 96%|█████████▌| 23496/24420 [9:48:49<22:58,  1.49s/it]
 96%|█████████▌| 23497/24420 [9:48:50<22:55,  1.49s/it]
 96%|█████████▌| 23498/24420 [9:48:52<22:54,  1.49s/it]
 96%|█████████▌| 23499/24420 [9:48:53<22:51,  1.49s/it]
 96%|█████████▌| 23500/24420 [9:48:55<22:49,  1.49s/it]
                                                       
{'loss': 0.083, 'learning_rate': 5e-05, 'epoch': 4.81}

 96%|█████████▌| 23500/24420 [9:48:55<22:49,  1.49s/it]
 96%|█████████▌| 23501/24420 [9:48:56<22:48,  1.49s/it]
 96%|█████████▌| 23502/24420 [9:48:58<22:45,  1.49s/it]
 96%|█████████▌| 23503/24420 [9:48:59<22:42,  1.49s/it]
 96%|█████████▌| 23504/24420 [9:49:01<22:44,  1.49s/it]
 96%|█████████▋| 23505/24420 [9:49:02<22:40,  1.49s/it]
                                                       
{'loss': 0.0785, 'learning_rate': 5e-05, 'epoch': 4.81}

 96%|█████████▋| 23505/24420 [9:49:02<22:40,  1.49s/it]
 96%|█████████▋| 23506/24420 [9:49:04<22:37,  1.49s/it]
 96%|█████████▋| 23507/24420 [9:49:05<22:34,  1.48s/it]
 96%|█████████▋| 23508/24420 [9:49:06<22:32,  1.48s/it]
 96%|█████████▋| 23509/24420 [9:49:08<22:31,  1.48s/it]
 96%|█████████▋| 23510/24420 [9:49:09<22:34,  1.49s/it]
                                                       
{'loss': 0.072, 'learning_rate': 5e-05, 'epoch': 4.81}

 96%|█████████▋| 23510/24420 [9:49:09<22:34,  1.49s/it]
 96%|█████████▋| 23511/24420 [9:49:11<22:32,  1.49s/it]
 96%|█████████▋| 23512/24420 [9:49:12<22:29,  1.49s/it]
 96%|█████████▋| 23513/24420 [9:49:14<22:26,  1.48s/it]
 96%|█████████▋| 23514/24420 [9:49:15<22:26,  1.49s/it]
 96%|█████████▋| 23515/24420 [9:49:17<22:24,  1.49s/it]
                                                       
{'loss': 0.073, 'learning_rate': 5e-05, 'epoch': 4.81}

 96%|█████████▋| 23515/24420 [9:49:17<22:24,  1.49s/it]
 96%|█████████▋| 23516/24420 [9:49:18<22:24,  1.49s/it]
 96%|█████████▋| 23517/24420 [9:49:20<22:24,  1.49s/it]
 96%|█████████▋| 23518/24420 [9:49:21<22:25,  1.49s/it]
 96%|█████████▋| 23519/24420 [9:49:23<22:21,  1.49s/it]
 96%|█████████▋| 23520/24420 [9:49:24<22:18,  1.49s/it]
                                                       
{'loss': 0.0835, 'learning_rate': 5e-05, 'epoch': 4.82}

 96%|█████████▋| 23520/24420 [9:49:24<22:18,  1.49s/it]
 96%|█████████▋| 23521/24420 [9:49:26<22:16,  1.49s/it]
 96%|█████████▋| 23522/24420 [9:49:27<22:16,  1.49s/it]
 96%|█████████▋| 23523/24420 [9:49:29<22:19,  1.49s/it]
 96%|█████████▋| 23524/24420 [9:49:30<22:15,  1.49s/it]
 96%|█████████▋| 23525/24420 [9:49:32<22:11,  1.49s/it]
                                                       
{'loss': 0.0931, 'learning_rate': 5e-05, 'epoch': 4.82}

 96%|█████████▋| 23525/24420 [9:49:32<22:11,  1.49s/it]
 96%|█████████▋| 23526/24420 [9:49:33<22:09,  1.49s/it]
 96%|█████████▋| 23527/24420 [9:49:35<22:08,  1.49s/it]
 96%|█████████▋| 23528/24420 [9:49:36<22:06,  1.49s/it]
 96%|█████████▋| 23529/24420 [9:49:38<22:03,  1.49s/it]
 96%|█████████▋| 23530/24420 [9:49:39<22:04,  1.49s/it]
                                                       
{'loss': 0.0709, 'learning_rate': 5e-05, 'epoch': 4.82}

 96%|█████████▋| 23530/24420 [9:49:39<22:04,  1.49s/it]
 96%|█████████▋| 23531/24420 [9:49:41<22:00,  1.49s/it]
 96%|█████████▋| 23532/24420 [9:49:42<21:58,  1.48s/it]
 96%|█████████▋| 23533/24420 [9:49:44<21:58,  1.49s/it]
 96%|█████████▋| 23534/24420 [9:49:45<21:59,  1.49s/it]
 96%|█████████▋| 23535/24420 [9:49:47<21:58,  1.49s/it]
                                                       
{'loss': 0.0883, 'learning_rate': 5e-05, 'epoch': 4.82}

 96%|█████████▋| 23535/24420 [9:49:47<21:58,  1.49s/it]
 96%|█████████▋| 23536/24420 [9:49:48<21:58,  1.49s/it]
 96%|█████████▋| 23537/24420 [9:49:50<21:57,  1.49s/it]
 96%|█████████▋| 23538/24420 [9:49:51<21:54,  1.49s/it]
 96%|█████████▋| 23539/24420 [9:49:53<21:51,  1.49s/it]
 96%|█████████▋| 23540/24420 [9:49:54<21:49,  1.49s/it]
                                                       
{'loss': 0.0585, 'learning_rate': 5e-05, 'epoch': 4.82}

 96%|█████████▋| 23540/24420 [9:49:54<21:49,  1.49s/it]
 96%|█████████▋| 23541/24420 [9:49:56<21:47,  1.49s/it]
 96%|█████████▋| 23542/24420 [9:49:57<21:45,  1.49s/it]
 96%|█████████▋| 23543/24420 [9:49:59<21:43,  1.49s/it]
 96%|█████████▋| 23544/24420 [9:50:00<21:42,  1.49s/it]
 96%|█████████▋| 23545/24420 [9:50:02<21:39,  1.48s/it]
                                                       
{'loss': 0.0742, 'learning_rate': 5e-05, 'epoch': 4.82}

 96%|█████████▋| 23545/24420 [9:50:02<21:39,  1.48s/it]
 96%|█████████▋| 23546/24420 [9:50:03<21:39,  1.49s/it]
 96%|█████████▋| 23547/24420 [9:50:04<21:35,  1.48s/it]
 96%|█████████▋| 23548/24420 [9:50:06<21:35,  1.49s/it]
 96%|█████████▋| 23549/24420 [9:50:07<21:36,  1.49s/it]
 96%|█████████▋| 23550/24420 [9:50:09<21:33,  1.49s/it]
                                                       
{'loss': 0.0757, 'learning_rate': 5e-05, 'epoch': 4.82}

 96%|█████████▋| 23550/24420 [9:50:09<21:33,  1.49s/it]
 96%|█████████▋| 23551/24420 [9:50:10<21:33,  1.49s/it]
 96%|█████████▋| 23552/24420 [9:50:12<21:31,  1.49s/it]
 96%|█████████▋| 23553/24420 [9:50:13<21:31,  1.49s/it]
 96%|█████████▋| 23554/24420 [9:50:15<21:29,  1.49s/it]
 96%|█████████▋| 23555/24420 [9:50:16<21:24,  1.49s/it]
                                                       
{'loss': 0.0701, 'learning_rate': 5e-05, 'epoch': 4.82}

 96%|█████████▋| 23555/24420 [9:50:16<21:24,  1.49s/it]
 96%|█████████▋| 23556/24420 [9:50:18<21:26,  1.49s/it]
 96%|█████████▋| 23557/24420 [9:50:19<21:25,  1.49s/it]
 96%|█████████▋| 23558/24420 [9:50:21<21:23,  1.49s/it]
 96%|█████████▋| 23559/24420 [9:50:22<21:21,  1.49s/it]
 96%|█████████▋| 23560/24420 [9:50:24<21:22,  1.49s/it]
                                                       
{'loss': 0.0738, 'learning_rate': 5e-05, 'epoch': 4.82}

 96%|█████████▋| 23560/24420 [9:50:24<21:22,  1.49s/it]
 96%|█████████▋| 23561/24420 [9:50:25<21:19,  1.49s/it]
 96%|█████████▋| 23562/24420 [9:50:27<21:19,  1.49s/it]
 96%|█████████▋| 23563/24420 [9:50:28<21:16,  1.49s/it]
 96%|█████████▋| 23564/24420 [9:50:30<21:15,  1.49s/it]
 96%|█████████▋| 23565/24420 [9:50:31<21:13,  1.49s/it]
                                                       
{'loss': 0.0638, 'learning_rate': 5e-05, 'epoch': 4.82}

 96%|█████████▋| 23565/24420 [9:50:31<21:13,  1.49s/it]
 97%|█████████▋| 23566/24420 [9:50:33<21:10,  1.49s/it]
 97%|█████████▋| 23567/24420 [9:50:34<21:09,  1.49s/it]
 97%|█████████▋| 23568/24420 [9:50:36<21:18,  1.50s/it]
 97%|█████████▋| 23569/24420 [9:50:37<21:13,  1.50s/it]
 97%|█████████▋| 23570/24420 [9:50:39<21:10,  1.50s/it]
                                                       
{'loss': 0.0781, 'learning_rate': 5e-05, 'epoch': 4.83}

 97%|█████████▋| 23570/24420 [9:50:39<21:10,  1.50s/it]
 97%|█████████▋| 23571/24420 [9:50:40<21:08,  1.49s/it]
 97%|█████████▋| 23572/24420 [9:50:42<21:04,  1.49s/it]
 97%|█████████▋| 23573/24420 [9:50:43<21:00,  1.49s/it]
 97%|█████████▋| 23574/24420 [9:50:45<20:58,  1.49s/it]
 97%|█████████▋| 23575/24420 [9:50:46<20:56,  1.49s/it]
                                                       
{'loss': 0.0611, 'learning_rate': 5e-05, 'epoch': 4.83}

 97%|█████████▋| 23575/24420 [9:50:46<20:56,  1.49s/it]
 97%|█████████▋| 23576/24420 [9:50:48<20:53,  1.49s/it]
 97%|█████████▋| 23577/24420 [9:50:49<20:51,  1.48s/it]
 97%|█████████▋| 23578/24420 [9:50:51<20:53,  1.49s/it]
 97%|█████████▋| 23579/24420 [9:50:52<20:49,  1.49s/it]
 97%|█████████▋| 23580/24420 [9:50:54<20:48,  1.49s/it]
                                                       
{'loss': 0.0674, 'learning_rate': 5e-05, 'epoch': 4.83}

 97%|█████████▋| 23580/24420 [9:50:54<20:48,  1.49s/it]
 97%|█████████▋| 23581/24420 [9:50:55<20:47,  1.49s/it]
 97%|█████████▋| 23582/24420 [9:50:57<20:48,  1.49s/it]
 97%|█████████▋| 23583/24420 [9:50:58<20:46,  1.49s/it]
 97%|█████████▋| 23584/24420 [9:51:00<20:42,  1.49s/it]
 97%|█████████▋| 23585/24420 [9:51:01<20:38,  1.48s/it]
                                                       
{'loss': 0.0788, 'learning_rate': 5e-05, 'epoch': 4.83}

 97%|█████████▋| 23585/24420 [9:51:01<20:38,  1.48s/it]
 97%|█████████▋| 23586/24420 [9:51:03<20:36,  1.48s/it]
 97%|█████████▋| 23587/24420 [9:51:04<20:34,  1.48s/it]
 97%|█████████▋| 23588/24420 [9:51:06<20:33,  1.48s/it]
 97%|█████████▋| 23589/24420 [9:51:07<20:33,  1.48s/it]
 97%|█████████▋| 23590/24420 [9:51:09<20:33,  1.49s/it]
                                                       
{'loss': 0.0634, 'learning_rate': 5e-05, 'epoch': 4.83}

 97%|█████████▋| 23590/24420 [9:51:09<20:33,  1.49s/it]
 97%|█████████▋| 23591/24420 [9:51:10<20:32,  1.49s/it]
 97%|█████████▋| 23592/24420 [9:51:11<20:31,  1.49s/it]
 97%|█████████▋| 23593/24420 [9:51:13<20:29,  1.49s/it]
 97%|█████████▋| 23594/24420 [9:51:14<20:28,  1.49s/it]
 97%|█████████▋| 23595/24420 [9:51:16<20:28,  1.49s/it]
                                                       
{'loss': 0.0709, 'learning_rate': 5e-05, 'epoch': 4.83}

 97%|█████████▋| 23595/24420 [9:51:16<20:28,  1.49s/it]
 97%|█████████▋| 23596/24420 [9:51:17<20:25,  1.49s/it]
 97%|█████████▋| 23597/24420 [9:51:19<20:26,  1.49s/it]
 97%|█████████▋| 23598/24420 [9:51:20<20:23,  1.49s/it]
 97%|█████████▋| 23599/24420 [9:51:22<20:24,  1.49s/it]
 97%|█████████▋| 23600/24420 [9:51:23<20:22,  1.49s/it]
                                                       
{'loss': 0.0619, 'learning_rate': 5e-05, 'epoch': 4.83}

 97%|█████████▋| 23600/24420 [9:51:23<20:22,  1.49s/it]
 97%|█████████▋| 23601/24420 [9:51:28<33:20,  2.44s/it]
 97%|█████████▋| 23602/24420 [9:51:30<29:21,  2.15s/it]
 97%|█████████▋| 23603/24420 [9:51:31<26:33,  1.95s/it]
 97%|█████████▋| 23604/24420 [9:51:32<24:36,  1.81s/it]
 97%|█████████▋| 23605/24420 [9:51:34<23:15,  1.71s/it]
                                                       
{'loss': 0.0689, 'learning_rate': 5e-05, 'epoch': 4.83}

 97%|█████████▋| 23605/24420 [9:51:34<23:15,  1.71s/it]
 97%|█████████▋| 23606/24420 [9:51:35<22:18,  1.64s/it]
 97%|█████████▋| 23607/24420 [9:51:37<21:40,  1.60s/it]
 97%|█████████▋| 23608/24420 [9:51:38<21:13,  1.57s/it]
 97%|█████████▋| 23609/24420 [9:51:40<20:49,  1.54s/it]
 97%|█████████▋| 23610/24420 [9:51:41<20:33,  1.52s/it]
                                                       
{'loss': 0.0641, 'learning_rate': 5e-05, 'epoch': 4.83}

 97%|█████████▋| 23610/24420 [9:51:41<20:33,  1.52s/it]
 97%|█████████▋| 23611/24420 [9:51:43<20:24,  1.51s/it]
 97%|█████████▋| 23612/24420 [9:51:44<20:14,  1.50s/it]
 97%|█████████▋| 23613/24420 [9:51:46<20:09,  1.50s/it]
 97%|█████████▋| 23614/24420 [9:51:47<20:05,  1.50s/it]
 97%|█████████▋| 23615/24420 [9:51:49<20:01,  1.49s/it]
                                                       
{'loss': 0.0693, 'learning_rate': 5e-05, 'epoch': 4.84}

 97%|█████████▋| 23615/24420 [9:51:49<20:01,  1.49s/it]
 97%|█████████▋| 23616/24420 [9:51:50<19:57,  1.49s/it]
 97%|█████████▋| 23617/24420 [9:51:52<19:54,  1.49s/it]
 97%|█████████▋| 23618/24420 [9:51:53<19:53,  1.49s/it]
 97%|█████████▋| 23619/24420 [9:51:55<19:51,  1.49s/it]
 97%|█████████▋| 23620/24420 [9:51:56<19:50,  1.49s/it]
                                                       
{'loss': 0.0662, 'learning_rate': 5e-05, 'epoch': 4.84}

 97%|█████████▋| 23620/24420 [9:51:56<19:50,  1.49s/it]
 97%|█████████▋| 23621/24420 [9:51:58<19:48,  1.49s/it]
 97%|█████████▋| 23622/24420 [9:51:59<19:48,  1.49s/it]
 97%|█████████▋| 23623/24420 [9:52:01<19:45,  1.49s/it]
 97%|█████████▋| 23624/24420 [9:52:02<19:44,  1.49s/it]
 97%|█████████▋| 23625/24420 [9:52:04<19:42,  1.49s/it]
                                                       
{'loss': 0.0627, 'learning_rate': 5e-05, 'epoch': 4.84}

 97%|█████████▋| 23625/24420 [9:52:04<19:42,  1.49s/it]
 97%|█████████▋| 23626/24420 [9:52:05<19:42,  1.49s/it]
 97%|█████████▋| 23627/24420 [9:52:07<19:38,  1.49s/it]
 97%|█████████▋| 23628/24420 [9:52:08<19:34,  1.48s/it]
 97%|█████████▋| 23629/24420 [9:52:10<19:33,  1.48s/it]
 97%|█████████▋| 23630/24420 [9:52:11<19:32,  1.48s/it]
                                                       
{'loss': 0.0693, 'learning_rate': 5e-05, 'epoch': 4.84}

 97%|█████████▋| 23630/24420 [9:52:11<19:32,  1.48s/it]
 97%|█████████▋| 23631/24420 [9:52:13<19:31,  1.48s/it]
 97%|█████████▋| 23632/24420 [9:52:14<19:30,  1.49s/it]
 97%|█████████▋| 23633/24420 [9:52:16<19:30,  1.49s/it]
 97%|█████████▋| 23634/24420 [9:52:17<19:30,  1.49s/it]
 97%|█████████▋| 23635/24420 [9:52:19<19:28,  1.49s/it]
                                                       
{'loss': 0.0624, 'learning_rate': 5e-05, 'epoch': 4.84}

 97%|█████████▋| 23635/24420 [9:52:19<19:28,  1.49s/it]
 97%|█████████▋| 23636/24420 [9:52:20<19:28,  1.49s/it]
 97%|█████████▋| 23637/24420 [9:52:22<19:27,  1.49s/it]
 97%|█████████▋| 23638/24420 [9:52:23<19:24,  1.49s/it]
 97%|█████████▋| 23639/24420 [9:52:25<19:22,  1.49s/it]
 97%|█████████▋| 23640/24420 [9:52:26<19:20,  1.49s/it]
                                                       
{'loss': 0.0746, 'learning_rate': 5e-05, 'epoch': 4.84}

 97%|█████████▋| 23640/24420 [9:52:26<19:20,  1.49s/it]
 97%|█████████▋| 23641/24420 [9:52:28<19:20,  1.49s/it]
 97%|█████████▋| 23642/24420 [9:52:29<19:17,  1.49s/it]
 97%|█████████▋| 23643/24420 [9:52:30<19:15,  1.49s/it]
 97%|█████████▋| 23644/24420 [9:52:32<19:14,  1.49s/it]
 97%|█████████▋| 23645/24420 [9:52:33<19:10,  1.48s/it]
                                                       
{'loss': 0.0715, 'learning_rate': 5e-05, 'epoch': 4.84}

 97%|█████████▋| 23645/24420 [9:52:33<19:10,  1.48s/it]
 97%|█████████▋| 23646/24420 [9:52:35<19:10,  1.49s/it]
 97%|█████████▋| 23647/24420 [9:52:36<19:08,  1.49s/it]
 97%|█████████▋| 23648/24420 [9:52:38<19:08,  1.49s/it]
 97%|█████████▋| 23649/24420 [9:52:39<19:05,  1.49s/it]
 97%|█████████▋| 23650/24420 [9:52:41<19:04,  1.49s/it]
                                                       
{'loss': 0.0655, 'learning_rate': 5e-05, 'epoch': 4.84}

 97%|█████████▋| 23650/24420 [9:52:41<19:04,  1.49s/it]
 97%|█████████▋| 23651/24420 [9:52:42<19:05,  1.49s/it]
 97%|█████████▋| 23652/24420 [9:52:44<19:01,  1.49s/it]
 97%|█████████▋| 23653/24420 [9:52:45<19:01,  1.49s/it]
 97%|█████████▋| 23654/24420 [9:52:47<18:59,  1.49s/it]
 97%|█████████▋| 23655/24420 [9:52:48<18:57,  1.49s/it]
                                                       
{'loss': 0.0812, 'learning_rate': 5e-05, 'epoch': 4.84}

 97%|█████████▋| 23655/24420 [9:52:48<18:57,  1.49s/it]
 97%|█████████▋| 23656/24420 [9:52:50<18:55,  1.49s/it]
 97%|█████████▋| 23657/24420 [9:52:51<18:55,  1.49s/it]
 97%|█████████▋| 23658/24420 [9:52:53<18:53,  1.49s/it]
 97%|█████████▋| 23659/24420 [9:52:54<18:52,  1.49s/it]
 97%|█████████▋| 23660/24420 [9:52:56<18:49,  1.49s/it]
                                                       
{'loss': 0.0845, 'learning_rate': 5e-05, 'epoch': 4.84}

 97%|█████████▋| 23660/24420 [9:52:56<18:49,  1.49s/it]
 97%|█████████▋| 23661/24420 [9:52:57<18:49,  1.49s/it]
 97%|█████████▋| 23662/24420 [9:52:59<18:47,  1.49s/it]
 97%|█████████▋| 23663/24420 [9:53:00<18:47,  1.49s/it]
 97%|█████████▋| 23664/24420 [9:53:02<18:45,  1.49s/it]
 97%|█████████▋| 23665/24420 [9:53:03<18:43,  1.49s/it]
                                                       
{'loss': 0.0714, 'learning_rate': 5e-05, 'epoch': 4.85}

 97%|█████████▋| 23665/24420 [9:53:03<18:43,  1.49s/it]
 97%|█████████▋| 23666/24420 [9:53:05<18:44,  1.49s/it]
 97%|█████████▋| 23667/24420 [9:53:06<18:43,  1.49s/it]
 97%|█████████▋| 23668/24420 [9:53:08<18:41,  1.49s/it]
 97%|█████████▋| 23669/24420 [9:53:09<18:39,  1.49s/it]
 97%|█████████▋| 23670/24420 [9:53:11<18:37,  1.49s/it]
                                                       
{'loss': 0.0638, 'learning_rate': 5e-05, 'epoch': 4.85}

 97%|█████████▋| 23670/24420 [9:53:11<18:37,  1.49s/it]
 97%|█████████▋| 23671/24420 [9:53:12<18:36,  1.49s/it]
 97%|█████████▋| 23672/24420 [9:53:14<18:36,  1.49s/it]
 97%|█████████▋| 23673/24420 [9:53:15<18:34,  1.49s/it]
 97%|█████████▋| 23674/24420 [9:53:17<18:32,  1.49s/it]
 97%|█████████▋| 23675/24420 [9:53:18<18:30,  1.49s/it]
                                                       
{'loss': 0.0758, 'learning_rate': 5e-05, 'epoch': 4.85}

 97%|█████████▋| 23675/24420 [9:53:18<18:30,  1.49s/it]
 97%|█████████▋| 23676/24420 [9:53:20<18:29,  1.49s/it]
 97%|█████████▋| 23677/24420 [9:53:21<18:27,  1.49s/it]
 97%|█████████▋| 23678/24420 [9:53:23<18:25,  1.49s/it]
 97%|█████████▋| 23679/24420 [9:53:24<18:25,  1.49s/it]
 97%|█████████▋| 23680/24420 [9:53:26<18:23,  1.49s/it]
                                                       
{'loss': 0.0689, 'learning_rate': 5e-05, 'epoch': 4.85}

 97%|█████████▋| 23680/24420 [9:53:26<18:23,  1.49s/it]
 97%|█████████▋| 23681/24420 [9:53:27<18:22,  1.49s/it]
 97%|█████████▋| 23682/24420 [9:53:29<18:18,  1.49s/it]
 97%|█████████▋| 23683/24420 [9:53:30<18:16,  1.49s/it]
 97%|█████████▋| 23684/24420 [9:53:32<18:16,  1.49s/it]
 97%|█████████▋| 23685/24420 [9:53:33<18:15,  1.49s/it]
                                                       
{'loss': 0.0719, 'learning_rate': 5e-05, 'epoch': 4.85}

 97%|█████████▋| 23685/24420 [9:53:33<18:15,  1.49s/it]
 97%|█████████▋| 23686/24420 [9:53:35<18:14,  1.49s/it]
 97%|█████████▋| 23687/24420 [9:53:36<18:14,  1.49s/it]
 97%|█████████▋| 23688/24420 [9:53:38<18:12,  1.49s/it]
 97%|█████████▋| 23689/24420 [9:53:39<18:09,  1.49s/it]
 97%|█████████▋| 23690/24420 [9:53:41<18:10,  1.49s/it]
                                                       
{'loss': 0.0715, 'learning_rate': 5e-05, 'epoch': 4.85}

 97%|█████████▋| 23690/24420 [9:53:41<18:10,  1.49s/it]
 97%|█████████▋| 23691/24420 [9:53:42<18:07,  1.49s/it]
 97%|█████████▋| 23692/24420 [9:53:43<18:05,  1.49s/it]
 97%|█████████▋| 23693/24420 [9:53:45<18:02,  1.49s/it]
 97%|█████████▋| 23694/24420 [9:53:46<18:00,  1.49s/it]
 97%|█████████▋| 23695/24420 [9:53:48<17:58,  1.49s/it]
                                                       
{'loss': 0.0786, 'learning_rate': 5e-05, 'epoch': 4.85}

 97%|█████████▋| 23695/24420 [9:53:48<17:58,  1.49s/it]
 97%|█████████▋| 23696/24420 [9:53:49<17:57,  1.49s/it]
 97%|█████████▋| 23697/24420 [9:53:51<17:53,  1.49s/it]
 97%|█████████▋| 23698/24420 [9:53:52<17:54,  1.49s/it]
 97%|█████████▋| 23699/24420 [9:53:54<17:55,  1.49s/it]
 97%|█████████▋| 23700/24420 [9:53:55<17:51,  1.49s/it]
                                                       
{'loss': 0.0789, 'learning_rate': 5e-05, 'epoch': 4.85}

 97%|█████████▋| 23700/24420 [9:53:55<17:51,  1.49s/it]
 97%|█████████▋| 23701/24420 [9:53:57<17:50,  1.49s/it]
 97%|█████████▋| 23702/24420 [9:53:58<17:49,  1.49s/it]
 97%|█████████▋| 23703/24420 [9:54:00<17:48,  1.49s/it]
 97%|█████████▋| 23704/24420 [9:54:01<17:45,  1.49s/it]
 97%|█████████▋| 23705/24420 [9:54:03<17:43,  1.49s/it]
                                                       
{'loss': 0.0751, 'learning_rate': 5e-05, 'epoch': 4.85}

 97%|█████████▋| 23705/24420 [9:54:03<17:43,  1.49s/it]
 97%|█████████▋| 23706/24420 [9:54:04<17:43,  1.49s/it]
 97%|█████████▋| 23707/24420 [9:54:06<17:39,  1.49s/it]
 97%|█████████▋| 23708/24420 [9:54:07<17:36,  1.48s/it]
 97%|█████████▋| 23709/24420 [9:54:09<17:36,  1.49s/it]
 97%|█████████▋| 23710/24420 [9:54:10<17:35,  1.49s/it]
                                                       
{'loss': 0.0631, 'learning_rate': 5e-05, 'epoch': 4.85}

 97%|█████████▋| 23710/24420 [9:54:10<17:35,  1.49s/it]
 97%|█████████▋| 23711/24420 [9:54:12<17:34,  1.49s/it]
 97%|█████████▋| 23712/24420 [9:54:13<17:30,  1.48s/it]
 97%|█████████▋| 23713/24420 [9:54:15<17:29,  1.48s/it]
 97%|█████████▋| 23714/24420 [9:54:16<17:27,  1.48s/it]
 97%|█████████▋| 23715/24420 [9:54:18<17:26,  1.48s/it]
                                                       
{'loss': 0.0548, 'learning_rate': 5e-05, 'epoch': 4.86}

 97%|█████████▋| 23715/24420 [9:54:18<17:26,  1.48s/it]
 97%|█████████▋| 23716/24420 [9:54:19<17:26,  1.49s/it]
 97%|█████████▋| 23717/24420 [9:54:21<17:26,  1.49s/it]
 97%|█████████▋| 23718/24420 [9:54:22<17:23,  1.49s/it]
 97%|█████████▋| 23719/24420 [9:54:24<17:21,  1.49s/it]
 97%|█████████▋| 23720/24420 [9:54:25<17:21,  1.49s/it]
                                                       
{'loss': 0.0896, 'learning_rate': 5e-05, 'epoch': 4.86}

 97%|█████████▋| 23720/24420 [9:54:25<17:21,  1.49s/it]
 97%|█████████▋| 23721/24420 [9:54:27<17:20,  1.49s/it]
 97%|█████████▋| 23722/24420 [9:54:28<17:18,  1.49s/it]
 97%|█████████▋| 23723/24420 [9:54:30<17:15,  1.49s/it]
 97%|█████████▋| 23724/24420 [9:54:31<17:13,  1.49s/it]
 97%|█████████▋| 23725/24420 [9:54:33<17:12,  1.49s/it]
                                                       
{'loss': 0.092, 'learning_rate': 5e-05, 'epoch': 4.86}

 97%|█████████▋| 23725/24420 [9:54:33<17:12,  1.49s/it]
 97%|█████████▋| 23726/24420 [9:54:34<17:11,  1.49s/it]
 97%|█████████▋| 23727/24420 [9:54:36<17:12,  1.49s/it]
 97%|█████████▋| 23728/24420 [9:54:37<17:10,  1.49s/it]
 97%|█████████▋| 23729/24420 [9:54:39<17:08,  1.49s/it]
 97%|█████████▋| 23730/24420 [9:54:40<17:04,  1.49s/it]
                                                       
{'loss': 0.0772, 'learning_rate': 5e-05, 'epoch': 4.86}

 97%|█████████▋| 23730/24420 [9:54:40<17:04,  1.49s/it]
 97%|█████████▋| 23731/24420 [9:54:41<17:03,  1.49s/it]
 97%|█████████▋| 23732/24420 [9:54:43<17:01,  1.48s/it]
 97%|█████████▋| 23733/24420 [9:54:44<17:01,  1.49s/it]
 97%|█████████▋| 23734/24420 [9:54:46<17:01,  1.49s/it]
 97%|█████████▋| 23735/24420 [9:54:47<17:00,  1.49s/it]
                                                       
{'loss': 0.0769, 'learning_rate': 5e-05, 'epoch': 4.86}

 97%|█████████▋| 23735/24420 [9:54:47<17:00,  1.49s/it]
 97%|█████████▋| 23736/24420 [9:54:49<16:58,  1.49s/it]
 97%|█████████▋| 23737/24420 [9:54:50<16:55,  1.49s/it]
 97%|█████████▋| 23738/24420 [9:54:52<16:53,  1.49s/it]
 97%|█████████▋| 23739/24420 [9:54:53<16:53,  1.49s/it]
 97%|█████████▋| 23740/24420 [9:54:55<16:52,  1.49s/it]
                                                       
{'loss': 0.0731, 'learning_rate': 5e-05, 'epoch': 4.86}

 97%|█████████▋| 23740/24420 [9:54:55<16:52,  1.49s/it]
 97%|█████████▋| 23741/24420 [9:54:56<16:52,  1.49s/it]
 97%|█████████▋| 23742/24420 [9:54:58<16:51,  1.49s/it]
 97%|█████████▋| 23743/24420 [9:54:59<16:48,  1.49s/it]
 97%|█████████▋| 23744/24420 [9:55:01<16:48,  1.49s/it]
 97%|█████████▋| 23745/24420 [9:55:02<16:46,  1.49s/it]
                                                       
{'loss': 0.0746, 'learning_rate': 5e-05, 'epoch': 4.86}

 97%|█████████▋| 23745/24420 [9:55:02<16:46,  1.49s/it]
 97%|█████████▋| 23746/24420 [9:55:04<16:43,  1.49s/it]
 97%|█████████▋| 23747/24420 [9:55:05<16:42,  1.49s/it]
 97%|█████████▋| 23748/24420 [9:55:07<16:41,  1.49s/it]
 97%|█████████▋| 23749/24420 [9:55:08<16:38,  1.49s/it]
 97%|█████████▋| 23750/24420 [9:55:10<16:37,  1.49s/it]
                                                       
{'loss': 0.0825, 'learning_rate': 5e-05, 'epoch': 4.86}

 97%|█████████▋| 23750/24420 [9:55:10<16:37,  1.49s/it]
 97%|█████████▋| 23751/24420 [9:55:11<16:40,  1.50s/it]
 97%|█████████▋| 23752/24420 [9:55:13<16:39,  1.50s/it]
 97%|█████████▋| 23753/24420 [9:55:14<16:37,  1.50s/it]
 97%|█████████▋| 23754/24420 [9:55:16<16:35,  1.49s/it]
 97%|█████████▋| 23755/24420 [9:55:17<16:33,  1.49s/it]
                                                       
{'loss': 0.0823, 'learning_rate': 5e-05, 'epoch': 4.86}

 97%|█████████▋| 23755/24420 [9:55:17<16:33,  1.49s/it]
 97%|█████████▋| 23756/24420 [9:55:19<16:31,  1.49s/it]
 97%|█████████▋| 23757/24420 [9:55:20<16:29,  1.49s/it]
 97%|█████████▋| 23758/24420 [9:55:22<16:26,  1.49s/it]
 97%|█████████▋| 23759/24420 [9:55:23<16:23,  1.49s/it]
 97%|█████████▋| 23760/24420 [9:55:25<16:22,  1.49s/it]
                                                       
{'loss': 0.0788, 'learning_rate': 5e-05, 'epoch': 4.86}

 97%|█████████▋| 23760/24420 [9:55:25<16:22,  1.49s/it]
 97%|█████████▋| 23761/24420 [9:55:26<16:22,  1.49s/it]
 97%|█████████▋| 23762/24420 [9:55:28<16:20,  1.49s/it]
 97%|█████████▋| 23763/24420 [9:55:29<16:19,  1.49s/it]
 97%|█████████▋| 23764/24420 [9:55:31<16:15,  1.49s/it]
 97%|█████████▋| 23765/24420 [9:55:32<16:14,  1.49s/it]
                                                       
{'loss': 0.0563, 'learning_rate': 5e-05, 'epoch': 4.87}

 97%|█████████▋| 23765/24420 [9:55:32<16:14,  1.49s/it]
 97%|█████████▋| 23766/24420 [9:55:34<16:12,  1.49s/it]
 97%|█████████▋| 23767/24420 [9:55:35<16:12,  1.49s/it]
 97%|█████████▋| 23768/24420 [9:55:37<16:11,  1.49s/it]
 97%|█████████▋| 23769/24420 [9:55:38<16:09,  1.49s/it]
 97%|█████████▋| 23770/24420 [9:55:40<16:08,  1.49s/it]
                                                       
{'loss': 0.0615, 'learning_rate': 5e-05, 'epoch': 4.87}

 97%|█████████▋| 23770/24420 [9:55:40<16:08,  1.49s/it]
 97%|█████████▋| 23771/24420 [9:55:41<16:07,  1.49s/it]
 97%|█████████▋| 23772/24420 [9:55:43<16:06,  1.49s/it]
 97%|█████████▋| 23773/24420 [9:55:44<16:04,  1.49s/it]
 97%|█████████▋| 23774/24420 [9:55:46<16:03,  1.49s/it]
 97%|█████████▋| 23775/24420 [9:55:47<16:04,  1.49s/it]
                                                       
{'loss': 0.0855, 'learning_rate': 5e-05, 'epoch': 4.87}

 97%|█████████▋| 23775/24420 [9:55:47<16:04,  1.49s/it]
 97%|█████████▋| 23776/24420 [9:55:49<16:03,  1.50s/it]
 97%|█████████▋| 23777/24420 [9:55:50<16:03,  1.50s/it]
 97%|█████████▋| 23778/24420 [9:55:52<16:01,  1.50s/it]
 97%|█████████▋| 23779/24420 [9:55:53<15:58,  1.49s/it]
 97%|█████████▋| 23780/24420 [9:55:55<15:55,  1.49s/it]
                                                       
{'loss': 0.0645, 'learning_rate': 5e-05, 'epoch': 4.87}

 97%|█████████▋| 23780/24420 [9:55:55<15:55,  1.49s/it]
 97%|█████████▋| 23781/24420 [9:55:56<15:53,  1.49s/it]
 97%|█████████▋| 23782/24420 [9:55:58<15:50,  1.49s/it]
 97%|█████████▋| 23783/24420 [9:55:59<15:50,  1.49s/it]
 97%|█████████▋| 23784/24420 [9:56:00<15:47,  1.49s/it]
 97%|█████████▋| 23785/24420 [9:56:02<15:46,  1.49s/it]
                                                       
{'loss': 0.0742, 'learning_rate': 5e-05, 'epoch': 4.87}

 97%|█████████▋| 23785/24420 [9:56:02<15:46,  1.49s/it]
 97%|█████████▋| 23786/24420 [9:56:03<15:45,  1.49s/it]
 97%|█████████▋| 23787/24420 [9:56:05<15:45,  1.49s/it]
 97%|█████████▋| 23788/24420 [9:56:06<15:42,  1.49s/it]
 97%|█████████▋| 23789/24420 [9:56:08<15:38,  1.49s/it]
 97%|█████████▋| 23790/24420 [9:56:09<15:38,  1.49s/it]
                                                       
{'loss': 0.0691, 'learning_rate': 5e-05, 'epoch': 4.87}

 97%|█████████▋| 23790/24420 [9:56:09<15:38,  1.49s/it]
 97%|█████████▋| 23791/24420 [9:56:11<15:38,  1.49s/it]
 97%|█████████▋| 23792/24420 [9:56:12<15:35,  1.49s/it]
 97%|█████████▋| 23793/24420 [9:56:14<15:33,  1.49s/it]
 97%|█████████▋| 23794/24420 [9:56:15<15:31,  1.49s/it]
 97%|█████████▋| 23795/24420 [9:56:17<15:30,  1.49s/it]
                                                       
{'loss': 0.0764, 'learning_rate': 5e-05, 'epoch': 4.87}

 97%|█████████▋| 23795/24420 [9:56:17<15:30,  1.49s/it]
 97%|█████████▋| 23796/24420 [9:56:19<17:20,  1.67s/it]
 97%|█████████▋| 23797/24420 [9:56:20<16:45,  1.61s/it]
 97%|█████████▋| 23798/24420 [9:56:22<16:21,  1.58s/it]
 97%|█████████▋| 23799/24420 [9:56:23<16:03,  1.55s/it]
 97%|█████████▋| 23800/24420 [9:56:25<15:50,  1.53s/it]
                                                       
{'loss': 0.0796, 'learning_rate': 5e-05, 'epoch': 4.87}

 97%|█████████▋| 23800/24420 [9:56:25<15:50,  1.53s/it]
 97%|█████████▋| 23801/24420 [9:56:30<25:24,  2.46s/it]
 97%|█████████▋| 23802/24420 [9:56:31<22:20,  2.17s/it]
 97%|█████████▋| 23803/24420 [9:56:33<20:11,  1.96s/it]
 97%|█████████▋| 23804/24420 [9:56:34<18:40,  1.82s/it]
 97%|█████████▋| 23805/24420 [9:56:35<17:37,  1.72s/it]
                                                       
{'loss': 0.0624, 'learning_rate': 5e-05, 'epoch': 4.87}

 97%|█████████▋| 23805/24420 [9:56:35<17:37,  1.72s/it]
 97%|█████████▋| 23806/24420 [9:56:37<16:53,  1.65s/it]
 97%|█████████▋| 23807/24420 [9:56:38<16:22,  1.60s/it]
 97%|█████████▋| 23808/24420 [9:56:40<16:00,  1.57s/it]
 97%|█████████▋| 23809/24420 [9:56:41<15:43,  1.54s/it]
 98%|█████████▊| 23810/24420 [9:56:43<15:30,  1.53s/it]
                                                       
{'loss': 0.0729, 'learning_rate': 5e-05, 'epoch': 4.88}

 98%|█████████▊| 23810/24420 [9:56:43<15:30,  1.53s/it]
 98%|█████████▊| 23811/24420 [9:56:44<15:22,  1.52s/it]
 98%|█████████▊| 23812/24420 [9:56:46<15:17,  1.51s/it]
 98%|█████████▊| 23813/24420 [9:56:47<15:11,  1.50s/it]
 98%|█████████▊| 23814/24420 [9:56:49<15:07,  1.50s/it]
 98%|█████████▊| 23815/24420 [9:56:50<15:04,  1.50s/it]
                                                       
{'loss': 0.0774, 'learning_rate': 5e-05, 'epoch': 4.88}

 98%|█████████▊| 23815/24420 [9:56:50<15:04,  1.50s/it]
 98%|█████████▊| 23816/24420 [9:56:52<15:01,  1.49s/it]
 98%|█████████▊| 23817/24420 [9:56:53<14:57,  1.49s/it]
 98%|█████████▊| 23818/24420 [9:56:55<14:58,  1.49s/it]
 98%|█████████▊| 23819/24420 [9:56:56<14:58,  1.49s/it]
 98%|█████████▊| 23820/24420 [9:56:58<14:55,  1.49s/it]
                                                       
{'loss': 0.0778, 'learning_rate': 5e-05, 'epoch': 4.88}

 98%|█████████▊| 23820/24420 [9:56:58<14:55,  1.49s/it]
 98%|█████████▊| 23821/24420 [9:56:59<14:53,  1.49s/it]
 98%|█████████▊| 23822/24420 [9:57:01<14:50,  1.49s/it]
 98%|█████████▊| 23823/24420 [9:57:02<14:49,  1.49s/it]
 98%|█████████▊| 23824/24420 [9:57:04<14:46,  1.49s/it]
 98%|█████████▊| 23825/24420 [9:57:05<14:45,  1.49s/it]
                                                       
{'loss': 0.0805, 'learning_rate': 5e-05, 'epoch': 4.88}

 98%|█████████▊| 23825/24420 [9:57:05<14:45,  1.49s/it]
 98%|█████████▊| 23826/24420 [9:57:07<14:43,  1.49s/it]
 98%|█████████▊| 23827/24420 [9:57:08<14:41,  1.49s/it]
 98%|█████████▊| 23828/24420 [9:57:10<14:38,  1.48s/it]
 98%|█████████▊| 23829/24420 [9:57:11<14:38,  1.49s/it]
 98%|█████████▊| 23830/24420 [9:57:13<14:38,  1.49s/it]
                                                       
{'loss': 0.0739, 'learning_rate': 5e-05, 'epoch': 4.88}

 98%|█████████▊| 23830/24420 [9:57:13<14:38,  1.49s/it]
 98%|█████████▊| 23831/24420 [9:57:14<14:36,  1.49s/it]
 98%|█████████▊| 23832/24420 [9:57:16<14:35,  1.49s/it]
 98%|█████████▊| 23833/24420 [9:57:17<14:32,  1.49s/it]
 98%|█████████▊| 23834/24420 [9:57:19<14:30,  1.49s/it]
 98%|█████████▊| 23835/24420 [9:57:20<14:30,  1.49s/it]
                                                       
{'loss': 0.083, 'learning_rate': 5e-05, 'epoch': 4.88}

 98%|█████████▊| 23835/24420 [9:57:20<14:30,  1.49s/it]
 98%|█████████▊| 23836/24420 [9:57:22<14:28,  1.49s/it]
 98%|█████████▊| 23837/24420 [9:57:23<14:28,  1.49s/it]
 98%|█████████▊| 23838/24420 [9:57:25<14:27,  1.49s/it]
 98%|█████████▊| 23839/24420 [9:57:26<14:25,  1.49s/it]
 98%|█████████▊| 23840/24420 [9:57:28<14:22,  1.49s/it]
                                                       
{'loss': 0.0644, 'learning_rate': 5e-05, 'epoch': 4.88}

 98%|█████████▊| 23840/24420 [9:57:28<14:22,  1.49s/it]
 98%|█████████▊| 23841/24420 [9:57:29<14:21,  1.49s/it]
 98%|█████████▊| 23842/24420 [9:57:31<14:18,  1.48s/it]
 98%|█████████▊| 23843/24420 [9:57:32<14:17,  1.49s/it]
 98%|█████████▊| 23844/24420 [9:57:34<14:16,  1.49s/it]
 98%|█████████▊| 23845/24420 [9:57:35<14:15,  1.49s/it]
                                                       
{'loss': 0.0667, 'learning_rate': 5e-05, 'epoch': 4.88}

 98%|█████████▊| 23845/24420 [9:57:35<14:15,  1.49s/it]
 98%|█████████▊| 23846/24420 [9:57:37<14:13,  1.49s/it]
 98%|█████████▊| 23847/24420 [9:57:38<14:14,  1.49s/it]
 98%|█████████▊| 23848/24420 [9:57:40<14:14,  1.49s/it]
 98%|█████████▊| 23849/24420 [9:57:41<14:11,  1.49s/it]
 98%|█████████▊| 23850/24420 [9:57:42<14:09,  1.49s/it]
                                                       
{'loss': 0.0793, 'learning_rate': 5e-05, 'epoch': 4.88}

 98%|█████████▊| 23850/24420 [9:57:42<14:09,  1.49s/it]
 98%|█████████▊| 23851/24420 [9:57:44<14:09,  1.49s/it]
 98%|█████████▊| 23852/24420 [9:57:45<14:07,  1.49s/it]
 98%|█████████▊| 23853/24420 [9:57:47<14:05,  1.49s/it]
 98%|█████████▊| 23854/24420 [9:57:48<14:04,  1.49s/it]
 98%|█████████▊| 23855/24420 [9:57:50<14:02,  1.49s/it]
                                                       
{'loss': 0.0809, 'learning_rate': 5e-05, 'epoch': 4.88}

 98%|█████████▊| 23855/24420 [9:57:50<14:02,  1.49s/it]
 98%|█████████▊| 23856/24420 [9:57:51<14:01,  1.49s/it]
 98%|█████████▊| 23857/24420 [9:57:53<13:59,  1.49s/it]
 98%|█████████▊| 23858/24420 [9:57:54<13:56,  1.49s/it]
 98%|█████████▊| 23859/24420 [9:57:56<13:54,  1.49s/it]
 98%|█████████▊| 23860/24420 [9:57:57<13:50,  1.48s/it]
                                                       
{'loss': 0.064, 'learning_rate': 5e-05, 'epoch': 4.89}

 98%|█████████▊| 23860/24420 [9:57:57<13:50,  1.48s/it]
 98%|█████████▊| 23861/24420 [9:57:59<13:50,  1.49s/it]
 98%|█████████▊| 23862/24420 [9:58:00<13:48,  1.48s/it]
 98%|█████████▊| 23863/24420 [9:58:02<13:48,  1.49s/it]
 98%|█████████▊| 23864/24420 [9:58:03<13:46,  1.49s/it]
 98%|█████████▊| 23865/24420 [9:58:05<13:45,  1.49s/it]
                                                       
{'loss': 0.0714, 'learning_rate': 5e-05, 'epoch': 4.89}

 98%|█████████▊| 23865/24420 [9:58:05<13:45,  1.49s/it]
 98%|█████████▊| 23866/24420 [9:58:06<13:45,  1.49s/it]
 98%|█████████▊| 23867/24420 [9:58:08<13:43,  1.49s/it]
 98%|█████████▊| 23868/24420 [9:58:09<13:42,  1.49s/it]
 98%|█████████▊| 23869/24420 [9:58:11<13:42,  1.49s/it]
 98%|█████████▊| 23870/24420 [9:58:12<13:39,  1.49s/it]
                                                       
{'loss': 0.0671, 'learning_rate': 5e-05, 'epoch': 4.89}

 98%|█████████▊| 23870/24420 [9:58:12<13:39,  1.49s/it]
 98%|█████████▊| 23871/24420 [9:58:14<13:39,  1.49s/it]
 98%|█████████▊| 23872/24420 [9:58:15<13:35,  1.49s/it]
 98%|█████████▊| 23873/24420 [9:58:17<13:33,  1.49s/it]
 98%|█████████▊| 23874/24420 [9:58:18<13:33,  1.49s/it]
 98%|█████████▊| 23875/24420 [9:58:20<13:32,  1.49s/it]
                                                       
{'loss': 0.0711, 'learning_rate': 5e-05, 'epoch': 4.89}

 98%|█████████▊| 23875/24420 [9:58:20<13:32,  1.49s/it]
 98%|█████████▊| 23876/24420 [9:58:21<13:32,  1.49s/it]
 98%|█████████▊| 23877/24420 [9:58:23<13:35,  1.50s/it]
 98%|█████████▊| 23878/24420 [9:58:24<13:32,  1.50s/it]
 98%|█████████▊| 23879/24420 [9:58:26<13:29,  1.50s/it]
 98%|█████████▊| 23880/24420 [9:58:27<13:28,  1.50s/it]
                                                       
{'loss': 0.0752, 'learning_rate': 5e-05, 'epoch': 4.89}

 98%|█████████▊| 23880/24420 [9:58:27<13:28,  1.50s/it]
 98%|█████████▊| 23881/24420 [9:58:29<13:25,  1.49s/it]
 98%|█████████▊| 23882/24420 [9:58:30<13:24,  1.49s/it]
 98%|█████████▊| 23883/24420 [9:58:32<13:20,  1.49s/it]
 98%|█████████▊| 23884/24420 [9:58:33<13:18,  1.49s/it]
 98%|█████████▊| 23885/24420 [9:58:35<13:16,  1.49s/it]
                                                       
{'loss': 0.0734, 'learning_rate': 5e-05, 'epoch': 4.89}

 98%|█████████▊| 23885/24420 [9:58:35<13:16,  1.49s/it]
 98%|█████████▊| 23886/24420 [9:58:36<13:15,  1.49s/it]
 98%|█████████▊| 23887/24420 [9:58:38<13:13,  1.49s/it]
 98%|█████████▊| 23888/24420 [9:58:39<13:11,  1.49s/it]
 98%|█████████▊| 23889/24420 [9:58:41<13:10,  1.49s/it]
 98%|█████████▊| 23890/24420 [9:58:42<13:09,  1.49s/it]
                                                       
{'loss': 0.0697, 'learning_rate': 5e-05, 'epoch': 4.89}

 98%|█████████▊| 23890/24420 [9:58:42<13:09,  1.49s/it]
 98%|█████████▊| 23891/24420 [9:58:44<13:08,  1.49s/it]
 98%|█████████▊| 23892/24420 [9:58:45<13:06,  1.49s/it]
 98%|█████████▊| 23893/24420 [9:58:47<13:05,  1.49s/it]
 98%|█████████▊| 23894/24420 [9:58:48<13:04,  1.49s/it]
 98%|█████████▊| 23895/24420 [9:58:50<13:01,  1.49s/it]
                                                       
{'loss': 0.0876, 'learning_rate': 5e-05, 'epoch': 4.89}

 98%|█████████▊| 23895/24420 [9:58:50<13:01,  1.49s/it]
 98%|█████████▊| 23896/24420 [9:58:51<13:01,  1.49s/it]
 98%|█████████▊| 23897/24420 [9:58:53<12:58,  1.49s/it]
 98%|█████████▊| 23898/24420 [9:58:54<12:56,  1.49s/it]
 98%|█████████▊| 23899/24420 [9:58:55<12:54,  1.49s/it]
 98%|█████████▊| 23900/24420 [9:58:57<12:53,  1.49s/it]
                                                       
{'loss': 0.0823, 'learning_rate': 5e-05, 'epoch': 4.89}

 98%|█████████▊| 23900/24420 [9:58:57<12:53,  1.49s/it]
 98%|█████████▊| 23901/24420 [9:58:58<12:52,  1.49s/it]
 98%|█████████▊| 23902/24420 [9:59:00<12:50,  1.49s/it]
 98%|█████████▊| 23903/24420 [9:59:01<12:49,  1.49s/it]
 98%|█████████▊| 23904/24420 [9:59:03<12:47,  1.49s/it]
 98%|█████████▊| 23905/24420 [9:59:04<12:44,  1.48s/it]
                                                       
{'loss': 0.0664, 'learning_rate': 5e-05, 'epoch': 4.89}

 98%|█████████▊| 23905/24420 [9:59:04<12:44,  1.48s/it]
 98%|█████████▊| 23906/24420 [9:59:06<12:43,  1.48s/it]
 98%|█████████▊| 23907/24420 [9:59:07<12:42,  1.49s/it]
 98%|█████████▊| 23908/24420 [9:59:09<12:41,  1.49s/it]
 98%|█████████▊| 23909/24420 [9:59:10<12:39,  1.49s/it]
 98%|█████████▊| 23910/24420 [9:59:12<12:37,  1.49s/it]
                                                       
{'loss': 0.0784, 'learning_rate': 5e-05, 'epoch': 4.9}

 98%|█████████▊| 23910/24420 [9:59:12<12:37,  1.49s/it]
 98%|█████████▊| 23911/24420 [9:59:13<12:36,  1.49s/it]
 98%|█████████▊| 23912/24420 [9:59:15<12:35,  1.49s/it]
 98%|█████████▊| 23913/24420 [9:59:16<12:34,  1.49s/it]
 98%|█████████▊| 23914/24420 [9:59:18<12:32,  1.49s/it]
 98%|█████████▊| 23915/24420 [9:59:19<12:31,  1.49s/it]
                                                       
{'loss': 0.0829, 'learning_rate': 5e-05, 'epoch': 4.9}

 98%|█████████▊| 23915/24420 [9:59:19<12:31,  1.49s/it]
 98%|█████████▊| 23916/24420 [9:59:21<12:30,  1.49s/it]
 98%|█████████▊| 23917/24420 [9:59:22<12:28,  1.49s/it]
 98%|█████████▊| 23918/24420 [9:59:24<12:30,  1.49s/it]
 98%|█████████▊| 23919/24420 [9:59:25<12:30,  1.50s/it]
 98%|█████████▊| 23920/24420 [9:59:27<12:27,  1.49s/it]
                                                       
{'loss': 0.0806, 'learning_rate': 5e-05, 'epoch': 4.9}

 98%|█████████▊| 23920/24420 [9:59:27<12:27,  1.49s/it]
 98%|█████████▊| 23921/24420 [9:59:28<12:24,  1.49s/it]
 98%|█████████▊| 23922/24420 [9:59:30<12:21,  1.49s/it]
 98%|█████████▊| 23923/24420 [9:59:31<12:20,  1.49s/it]
 98%|█████████▊| 23924/24420 [9:59:33<12:18,  1.49s/it]
 98%|█████████▊| 23925/24420 [9:59:35<14:16,  1.73s/it]
                                                       
{'loss': 0.0639, 'learning_rate': 5e-05, 'epoch': 4.9}

 98%|█████████▊| 23925/24420 [9:59:35<14:16,  1.73s/it]
 98%|█████████▊| 23926/24420 [9:59:36<13:38,  1.66s/it]
 98%|█████████▊| 23927/24420 [9:59:38<13:10,  1.60s/it]
 98%|█████████▊| 23928/24420 [9:59:39<12:51,  1.57s/it]
 98%|█████████▊| 23929/24420 [9:59:41<12:37,  1.54s/it]
 98%|█████████▊| 23930/24420 [9:59:42<12:27,  1.52s/it]
                                                       
{'loss': 0.067, 'learning_rate': 5e-05, 'epoch': 4.9}

 98%|█████████▊| 23930/24420 [9:59:42<12:27,  1.52s/it]
 98%|█████████▊| 23931/24420 [9:59:44<12:19,  1.51s/it]
 98%|█████████▊| 23932/24420 [9:59:45<12:13,  1.50s/it]
 98%|█████████▊| 23933/24420 [9:59:47<12:10,  1.50s/it]
 98%|█████████▊| 23934/24420 [9:59:48<12:08,  1.50s/it]
 98%|█████████▊| 23935/24420 [9:59:50<12:05,  1.50s/it]
                                                       
{'loss': 0.0636, 'learning_rate': 5e-05, 'epoch': 4.9}

 98%|█████████▊| 23935/24420 [9:59:50<12:05,  1.50s/it]
 98%|█████████▊| 23936/24420 [9:59:51<12:04,  1.50s/it]
 98%|█████████▊| 23937/24420 [9:59:53<12:00,  1.49s/it]
 98%|█████████▊| 23938/24420 [9:59:54<11:58,  1.49s/it]
 98%|█████████▊| 23939/24420 [9:59:56<11:56,  1.49s/it]
 98%|█████████▊| 23940/24420 [9:59:57<11:55,  1.49s/it]
                                                       
{'loss': 0.0669, 'learning_rate': 5e-05, 'epoch': 4.9}

 98%|█████████▊| 23940/24420 [9:59:57<11:55,  1.49s/it]
 98%|█████████▊| 23941/24420 [9:59:59<11:53,  1.49s/it]
 98%|█████████▊| 23942/24420 [10:00:00<11:52,  1.49s/it]
 98%|█████████▊| 23943/24420 [10:00:02<11:49,  1.49s/it]
 98%|█████████▊| 23944/24420 [10:00:03<11:47,  1.49s/it]
 98%|█████████▊| 23945/24420 [10:00:05<11:46,  1.49s/it]
                                                        
{'loss': 0.0704, 'learning_rate': 5e-05, 'epoch': 4.9}

 98%|█████████▊| 23945/24420 [10:00:05<11:46,  1.49s/it]
 98%|█████████▊| 23946/24420 [10:00:06<11:44,  1.49s/it]
 98%|█████████▊| 23947/24420 [10:00:08<11:42,  1.48s/it]
 98%|█████████▊| 23948/24420 [10:00:09<11:41,  1.49s/it]
 98%|█████████▊| 23949/24420 [10:00:11<11:42,  1.49s/it]
 98%|█████████▊| 23950/24420 [10:00:12<11:39,  1.49s/it]
                                                        
{'loss': 0.0734, 'learning_rate': 5e-05, 'epoch': 4.9}

 98%|█████████▊| 23950/24420 [10:00:12<11:39,  1.49s/it]
 98%|█████████▊| 23951/24420 [10:00:14<11:40,  1.49s/it]
 98%|█████████▊| 23952/24420 [10:00:15<11:38,  1.49s/it]
 98%|█████████▊| 23953/24420 [10:00:17<11:36,  1.49s/it]
 98%|█████████▊| 23954/24420 [10:00:18<11:33,  1.49s/it]
 98%|█████████▊| 23955/24420 [10:00:20<11:33,  1.49s/it]
                                                        
{'loss': 0.0757, 'learning_rate': 5e-05, 'epoch': 4.9}

 98%|█████████▊| 23955/24420 [10:00:20<11:33,  1.49s/it]
 98%|█████████▊| 23956/24420 [10:00:21<11:31,  1.49s/it]
 98%|█████████▊| 23957/24420 [10:00:23<11:28,  1.49s/it]
 98%|█████████▊| 23958/24420 [10:00:24<11:28,  1.49s/it]
 98%|█████████▊| 23959/24420 [10:00:26<11:25,  1.49s/it]
 98%|█████████▊| 23960/24420 [10:00:27<11:24,  1.49s/it]
                                                        
{'loss': 0.0691, 'learning_rate': 5e-05, 'epoch': 4.91}

 98%|█████████▊| 23960/24420 [10:00:27<11:24,  1.49s/it]
 98%|█████████▊| 23961/24420 [10:00:29<11:22,  1.49s/it]
 98%|█████████▊| 23962/24420 [10:00:30<11:20,  1.49s/it]
 98%|█████████▊| 23963/24420 [10:00:32<11:18,  1.48s/it]
 98%|█████████▊| 23964/24420 [10:00:33<11:17,  1.49s/it]
 98%|█████████▊| 23965/24420 [10:00:35<11:15,  1.49s/it]
                                                        
{'loss': 0.0668, 'learning_rate': 5e-05, 'epoch': 4.91}

 98%|█████████▊| 23965/24420 [10:00:35<11:15,  1.49s/it]
 98%|█████████▊| 23966/24420 [10:00:36<11:14,  1.49s/it]
 98%|█████████▊| 23967/24420 [10:00:37<11:12,  1.49s/it]
 98%|█████████▊| 23968/24420 [10:00:39<11:12,  1.49s/it]
 98%|█████████▊| 23969/24420 [10:00:40<11:10,  1.49s/it]
 98%|█████████▊| 23970/24420 [10:00:42<11:11,  1.49s/it]
                                                        
{'loss': 0.065, 'learning_rate': 5e-05, 'epoch': 4.91}

 98%|█████████▊| 23970/24420 [10:00:42<11:11,  1.49s/it]
 98%|█████████▊| 23971/24420 [10:00:43<11:09,  1.49s/it]
 98%|█████████▊| 23972/24420 [10:00:45<11:06,  1.49s/it]
 98%|█████████▊| 23973/24420 [10:00:46<11:05,  1.49s/it]
 98%|█████████▊| 23974/24420 [10:00:48<11:02,  1.49s/it]
 98%|█████████▊| 23975/24420 [10:00:49<11:01,  1.49s/it]
                                                        
{'loss': 0.0655, 'learning_rate': 5e-05, 'epoch': 4.91}

 98%|█████████▊| 23975/24420 [10:00:49<11:01,  1.49s/it]
 98%|█████████▊| 23976/24420 [10:00:51<10:59,  1.49s/it]
 98%|█████████▊| 23977/24420 [10:00:52<10:58,  1.49s/it]
 98%|█████████▊| 23978/24420 [10:00:54<10:58,  1.49s/it]
 98%|█████████▊| 23979/24420 [10:00:55<10:57,  1.49s/it]
 98%|█████████▊| 23980/24420 [10:00:57<10:56,  1.49s/it]
                                                        
{'loss': 0.0803, 'learning_rate': 5e-05, 'epoch': 4.91}

 98%|█████████▊| 23980/24420 [10:00:57<10:56,  1.49s/it]
 98%|█████████▊| 23981/24420 [10:00:58<10:54,  1.49s/it]
 98%|█████████▊| 23982/24420 [10:01:00<10:53,  1.49s/it]
 98%|█████████▊| 23983/24420 [10:01:01<10:51,  1.49s/it]
 98%|█████████▊| 23984/24420 [10:01:03<10:49,  1.49s/it]
 98%|█████████▊| 23985/24420 [10:01:04<10:47,  1.49s/it]
                                                        
{'loss': 0.0705, 'learning_rate': 5e-05, 'epoch': 4.91}

 98%|█████████▊| 23985/24420 [10:01:04<10:47,  1.49s/it]
 98%|█████████▊| 23986/24420 [10:01:06<10:46,  1.49s/it]
 98%|█████████▊| 23987/24420 [10:01:07<10:45,  1.49s/it]
 98%|█████████▊| 23988/24420 [10:01:09<10:42,  1.49s/it]
 98%|█████████▊| 23989/24420 [10:01:10<10:40,  1.49s/it]
 98%|█████████▊| 23990/24420 [10:01:12<10:39,  1.49s/it]
                                                        
{'loss': 0.0815, 'learning_rate': 5e-05, 'epoch': 4.91}

 98%|█████████▊| 23990/24420 [10:01:12<10:39,  1.49s/it]
 98%|█████████▊| 23991/24420 [10:01:13<10:37,  1.49s/it]
 98%|█████████▊| 23992/24420 [10:01:15<10:36,  1.49s/it]
 98%|█████████▊| 23993/24420 [10:01:16<10:35,  1.49s/it]
 98%|█████████▊| 23994/24420 [10:01:18<10:32,  1.49s/it]
 98%|█████████▊| 23995/24420 [10:01:19<10:31,  1.49s/it]
                                                        
{'loss': 0.0657, 'learning_rate': 5e-05, 'epoch': 4.91}

 98%|█████████▊| 23995/24420 [10:01:19<10:31,  1.49s/it]
 98%|█████████▊| 23996/24420 [10:01:21<10:29,  1.49s/it]
 98%|█████████▊| 23997/24420 [10:01:22<10:27,  1.48s/it]
 98%|█████████▊| 23998/24420 [10:01:24<10:26,  1.49s/it]
 98%|█████████▊| 23999/24420 [10:01:25<10:26,  1.49s/it]
 98%|█████████▊| 24000/24420 [10:01:27<10:25,  1.49s/it]
                                                        
{'loss': 0.073, 'learning_rate': 5e-05, 'epoch': 4.91}

 98%|█████████▊| 24000/24420 [10:01:27<10:25,  1.49s/it]
 98%|█████████▊| 24001/24420 [10:01:31<17:06,  2.45s/it]
 98%|█████████▊| 24002/24420 [10:01:33<15:02,  2.16s/it]
 98%|█████████▊| 24003/24420 [10:01:34<13:36,  1.96s/it]
 98%|█████████▊| 24004/24420 [10:01:36<12:35,  1.82s/it]
 98%|█████████▊| 24005/24420 [10:01:37<11:54,  1.72s/it]
                                                        
{'loss': 0.0706, 'learning_rate': 5e-05, 'epoch': 4.92}

 98%|█████████▊| 24005/24420 [10:01:37<11:54,  1.72s/it]
 98%|█████████▊| 24006/24420 [10:01:39<11:25,  1.65s/it]
 98%|█████████▊| 24007/24420 [10:01:40<11:02,  1.60s/it]
 98%|█████████▊| 24008/24420 [10:01:42<10:44,  1.56s/it]
 98%|█████████▊| 24009/24420 [10:01:43<10:33,  1.54s/it]
 98%|█████████▊| 24010/24420 [10:01:45<10:25,  1.53s/it]
                                                        
{'loss': 0.0839, 'learning_rate': 5e-05, 'epoch': 4.92}

 98%|█████████▊| 24010/24420 [10:01:45<10:25,  1.53s/it]
 98%|█████████▊| 24011/24420 [10:01:46<10:19,  1.51s/it]
 98%|█████████▊| 24012/24420 [10:01:48<10:14,  1.51s/it]
 98%|█████████▊| 24013/24420 [10:01:49<10:09,  1.50s/it]
 98%|█████████▊| 24014/24420 [10:01:51<10:07,  1.50s/it]
 98%|█████████▊| 24015/24420 [10:01:52<10:04,  1.49s/it]
                                                        
{'loss': 0.0706, 'learning_rate': 5e-05, 'epoch': 4.92}

 98%|█████████▊| 24015/24420 [10:01:52<10:04,  1.49s/it]
 98%|█████████▊| 24016/24420 [10:01:54<10:02,  1.49s/it]
 98%|█████████▊| 24017/24420 [10:01:55<10:00,  1.49s/it]
 98%|█████████▊| 24018/24420 [10:01:57<09:59,  1.49s/it]
 98%|█████████▊| 24019/24420 [10:01:58<09:58,  1.49s/it]
 98%|█████████▊| 24020/24420 [10:02:00<09:56,  1.49s/it]
                                                        
{'loss': 0.077, 'learning_rate': 5e-05, 'epoch': 4.92}

 98%|█████████▊| 24020/24420 [10:02:00<09:56,  1.49s/it]
 98%|█████████▊| 24021/24420 [10:02:01<09:53,  1.49s/it]
 98%|█████████▊| 24022/24420 [10:02:03<09:51,  1.49s/it]
 98%|█████████▊| 24023/24420 [10:02:04<09:50,  1.49s/it]
 98%|█████████▊| 24024/24420 [10:02:05<09:49,  1.49s/it]
 98%|█████████▊| 24025/24420 [10:02:07<09:49,  1.49s/it]
                                                        
{'loss': 0.0586, 'learning_rate': 5e-05, 'epoch': 4.92}

 98%|█████████▊| 24025/24420 [10:02:07<09:49,  1.49s/it]
 98%|█████████▊| 24026/24420 [10:02:08<09:46,  1.49s/it]
 98%|█████████▊| 24027/24420 [10:02:10<09:45,  1.49s/it]
 98%|█████████▊| 24028/24420 [10:02:11<09:42,  1.49s/it]
 98%|█████████▊| 24029/24420 [10:02:13<09:41,  1.49s/it]
 98%|█████████▊| 24030/24420 [10:02:14<09:40,  1.49s/it]
                                                        
{'loss': 0.0606, 'learning_rate': 5e-05, 'epoch': 4.92}

 98%|█████████▊| 24030/24420 [10:02:14<09:40,  1.49s/it]
 98%|█████████▊| 24031/24420 [10:02:16<09:39,  1.49s/it]
 98%|█████████▊| 24032/24420 [10:02:17<09:39,  1.49s/it]
 98%|█████████▊| 24033/24420 [10:02:19<09:37,  1.49s/it]
 98%|█████████▊| 24034/24420 [10:02:20<09:34,  1.49s/it]
 98%|█████████▊| 24035/24420 [10:02:22<09:33,  1.49s/it]
                                                        
{'loss': 0.0761, 'learning_rate': 5e-05, 'epoch': 4.92}

 98%|█████████▊| 24035/24420 [10:02:22<09:33,  1.49s/it]
 98%|█████████▊| 24036/24420 [10:02:23<09:32,  1.49s/it]
 98%|█████████▊| 24037/24420 [10:02:25<09:30,  1.49s/it]
 98%|█████████▊| 24038/24420 [10:02:26<09:29,  1.49s/it]
 98%|█████████▊| 24039/24420 [10:02:28<09:27,  1.49s/it]
 98%|█████████▊| 24040/24420 [10:02:29<09:27,  1.49s/it]
                                                        
{'loss': 0.0722, 'learning_rate': 5e-05, 'epoch': 4.92}

 98%|█████████▊| 24040/24420 [10:02:29<09:27,  1.49s/it]
 98%|█████████▊| 24041/24420 [10:02:31<09:24,  1.49s/it]
 98%|█████████▊| 24042/24420 [10:02:32<09:23,  1.49s/it]
 98%|█████████▊| 24043/24420 [10:02:34<09:22,  1.49s/it]
 98%|█████████▊| 24044/24420 [10:02:35<09:20,  1.49s/it]
 98%|█████████▊| 24045/24420 [10:02:37<09:19,  1.49s/it]
                                                        
{'loss': 0.0947, 'learning_rate': 5e-05, 'epoch': 4.92}

 98%|█████████▊| 24045/24420 [10:02:37<09:19,  1.49s/it]
 98%|█████████▊| 24046/24420 [10:02:38<09:17,  1.49s/it]
 98%|█████████▊| 24047/24420 [10:02:40<09:15,  1.49s/it]
 98%|█████████▊| 24048/24420 [10:02:41<09:14,  1.49s/it]
 98%|█████████▊| 24049/24420 [10:02:43<09:12,  1.49s/it]
 98%|█████████▊| 24050/24420 [10:02:44<09:11,  1.49s/it]
                                                        
{'loss': 0.067, 'learning_rate': 5e-05, 'epoch': 4.92}

 98%|█████████▊| 24050/24420 [10:02:44<09:11,  1.49s/it]
 98%|█████████▊| 24051/24420 [10:02:46<09:09,  1.49s/it]
 98%|█████████▊| 24052/24420 [10:02:47<09:07,  1.49s/it]
 98%|█████████▊| 24053/24420 [10:02:49<09:07,  1.49s/it]
 99%|█████████▊| 24054/24420 [10:02:50<09:04,  1.49s/it]
 99%|█████████▊| 24055/24420 [10:02:52<09:02,  1.49s/it]
                                                        
{'loss': 0.0662, 'learning_rate': 5e-05, 'epoch': 4.93}

 99%|█████████▊| 24055/24420 [10:02:52<09:02,  1.49s/it]
 99%|█████████▊| 24056/24420 [10:02:53<09:15,  1.53s/it]
 99%|█████████▊| 24057/24420 [10:02:55<09:09,  1.51s/it]
 99%|█████████▊| 24058/24420 [10:02:56<09:04,  1.50s/it]
 99%|█████████▊| 24059/24420 [10:02:58<09:00,  1.50s/it]
 99%|█████████▊| 24060/24420 [10:02:59<08:57,  1.49s/it]
                                                        
{'loss': 0.082, 'learning_rate': 5e-05, 'epoch': 4.93}

 99%|█████████▊| 24060/24420 [10:02:59<08:57,  1.49s/it]
 99%|█████████▊| 24061/24420 [10:03:01<08:56,  1.49s/it]
 99%|█████████▊| 24062/24420 [10:03:02<08:54,  1.49s/it]
 99%|█████████▊| 24063/24420 [10:03:04<08:52,  1.49s/it]
 99%|█████████▊| 24064/24420 [10:03:05<08:51,  1.49s/it]
 99%|█████████▊| 24065/24420 [10:03:07<08:49,  1.49s/it]
                                                        
{'loss': 0.0811, 'learning_rate': 5e-05, 'epoch': 4.93}

 99%|█████████▊| 24065/24420 [10:03:07<08:49,  1.49s/it]
 99%|█████████▊| 24066/24420 [10:03:08<08:47,  1.49s/it]
 99%|█████████▊| 24067/24420 [10:03:10<08:44,  1.49s/it]
 99%|█████████▊| 24068/24420 [10:03:11<08:43,  1.49s/it]
 99%|█████████▊| 24069/24420 [10:03:13<08:42,  1.49s/it]
 99%|█████████▊| 24070/24420 [10:03:14<08:40,  1.49s/it]
                                                        
{'loss': 0.086, 'learning_rate': 5e-05, 'epoch': 4.93}

 99%|█████████▊| 24070/24420 [10:03:14<08:40,  1.49s/it]
 99%|█████████▊| 24071/24420 [10:03:16<08:39,  1.49s/it]
 99%|█████████▊| 24072/24420 [10:03:17<08:37,  1.49s/it]
 99%|█████████▊| 24073/24420 [10:03:19<08:35,  1.49s/it]
 99%|█████████▊| 24074/24420 [10:03:20<08:34,  1.49s/it]
 99%|█████████▊| 24075/24420 [10:03:22<08:32,  1.49s/it]
                                                        
{'loss': 0.0613, 'learning_rate': 5e-05, 'epoch': 4.93}

 99%|█████████▊| 24075/24420 [10:03:22<08:32,  1.49s/it]
 99%|█████████▊| 24076/24420 [10:03:23<08:31,  1.49s/it]
 99%|█████████▊| 24077/24420 [10:03:25<08:31,  1.49s/it]
 99%|█████████▊| 24078/24420 [10:03:26<08:30,  1.49s/it]
 99%|█████████▊| 24079/24420 [10:03:28<08:28,  1.49s/it]
 99%|█████████▊| 24080/24420 [10:03:29<08:25,  1.49s/it]
                                                        
{'loss': 0.0754, 'learning_rate': 5e-05, 'epoch': 4.93}

 99%|█████████▊| 24080/24420 [10:03:29<08:25,  1.49s/it]
 99%|█████████▊| 24081/24420 [10:03:30<08:23,  1.49s/it]
 99%|█████████▊| 24082/24420 [10:03:32<08:22,  1.49s/it]
 99%|█████████▊| 24083/24420 [10:03:33<08:20,  1.49s/it]
 99%|█████████▊| 24084/24420 [10:03:35<08:19,  1.49s/it]
 99%|█████████▊| 24085/24420 [10:03:36<08:17,  1.49s/it]
                                                        
{'loss': 0.0665, 'learning_rate': 5e-05, 'epoch': 4.93}

 99%|█████████▊| 24085/24420 [10:03:36<08:17,  1.49s/it]
 99%|█████████▊| 24086/24420 [10:03:38<08:16,  1.49s/it]
 99%|█████████▊| 24087/24420 [10:03:39<08:14,  1.49s/it]
 99%|█████████▊| 24088/24420 [10:03:41<08:13,  1.49s/it]
 99%|█████████▊| 24089/24420 [10:03:42<08:13,  1.49s/it]
 99%|█████████▊| 24090/24420 [10:03:44<08:11,  1.49s/it]
                                                        
{'loss': 0.0667, 'learning_rate': 5e-05, 'epoch': 4.93}

 99%|█████████▊| 24090/24420 [10:03:44<08:11,  1.49s/it]
 99%|█████████▊| 24091/24420 [10:03:45<08:09,  1.49s/it]
 99%|█████████▊| 24092/24420 [10:03:47<08:07,  1.49s/it]
 99%|█████████▊| 24093/24420 [10:03:48<08:07,  1.49s/it]
 99%|█████████▊| 24094/24420 [10:03:50<08:05,  1.49s/it]
 99%|█████████▊| 24095/24420 [10:03:51<08:03,  1.49s/it]
                                                        
{'loss': 0.0867, 'learning_rate': 5e-05, 'epoch': 4.93}

 99%|█████████▊| 24095/24420 [10:03:51<08:03,  1.49s/it]
 99%|█████████▊| 24096/24420 [10:03:53<08:02,  1.49s/it]
 99%|█████████▊| 24097/24420 [10:03:54<08:01,  1.49s/it]
 99%|█████████▊| 24098/24420 [10:03:56<07:58,  1.49s/it]
 99%|█████████▊| 24099/24420 [10:03:57<07:56,  1.49s/it]
 99%|█████████▊| 24100/24420 [10:03:59<07:55,  1.48s/it]
                                                        
{'loss': 0.0707, 'learning_rate': 5e-05, 'epoch': 4.93}

 99%|█████████▊| 24100/24420 [10:03:59<07:55,  1.48s/it]
 99%|█████████▊| 24101/24420 [10:04:00<07:52,  1.48s/it]
 99%|█████████▊| 24102/24420 [10:04:02<07:52,  1.49s/it]
 99%|█████████▊| 24103/24420 [10:04:03<07:50,  1.48s/it]
 99%|█████████▊| 24104/24420 [10:04:05<07:49,  1.49s/it]
 99%|█████████▊| 24105/24420 [10:04:06<07:47,  1.49s/it]
                                                        
{'loss': 0.0743, 'learning_rate': 5e-05, 'epoch': 4.94}

 99%|█████████▊| 24105/24420 [10:04:06<07:47,  1.49s/it]
 99%|█████████▊| 24106/24420 [10:04:08<07:46,  1.49s/it]
 99%|█████████▊| 24107/24420 [10:04:09<07:46,  1.49s/it]
 99%|█████████▊| 24108/24420 [10:04:11<07:44,  1.49s/it]
 99%|█████████▊| 24109/24420 [10:04:12<07:43,  1.49s/it]
 99%|█████████▊| 24110/24420 [10:04:14<07:41,  1.49s/it]
                                                        
{'loss': 0.0755, 'learning_rate': 5e-05, 'epoch': 4.94}

 99%|█████████▊| 24110/24420 [10:04:14<07:41,  1.49s/it]
 99%|█████████▊| 24111/24420 [10:04:15<07:40,  1.49s/it]
 99%|█████████▊| 24112/24420 [10:04:17<07:39,  1.49s/it]
 99%|█████████▊| 24113/24420 [10:04:18<07:37,  1.49s/it]
 99%|█████████▊| 24114/24420 [10:04:20<07:34,  1.49s/it]
 99%|█████████▉| 24115/24420 [10:04:21<07:33,  1.49s/it]
                                                        
{'loss': 0.0676, 'learning_rate': 5e-05, 'epoch': 4.94}

 99%|█████████▉| 24115/24420 [10:04:21<07:33,  1.49s/it]
 99%|█████████▉| 24116/24420 [10:04:23<07:31,  1.49s/it]
 99%|█████████▉| 24117/24420 [10:04:24<07:29,  1.49s/it]
 99%|█████████▉| 24118/24420 [10:04:25<07:28,  1.49s/it]
 99%|█████████▉| 24119/24420 [10:04:27<07:26,  1.48s/it]
 99%|█████████▉| 24120/24420 [10:04:28<07:27,  1.49s/it]
                                                        
{'loss': 0.0694, 'learning_rate': 5e-05, 'epoch': 4.94}

 99%|█████████▉| 24120/24420 [10:04:28<07:27,  1.49s/it]
 99%|█████████▉| 24121/24420 [10:04:30<07:26,  1.49s/it]
 99%|█████████▉| 24122/24420 [10:04:31<07:24,  1.49s/it]
 99%|█████████▉| 24123/24420 [10:04:33<07:22,  1.49s/it]
 99%|█████████▉| 24124/24420 [10:04:34<07:22,  1.49s/it]
 99%|█████████▉| 24125/24420 [10:04:36<07:20,  1.49s/it]
                                                        
{'loss': 0.0676, 'learning_rate': 5e-05, 'epoch': 4.94}

 99%|█████████▉| 24125/24420 [10:04:36<07:20,  1.49s/it]
 99%|█████████▉| 24126/24420 [10:04:37<07:18,  1.49s/it]
 99%|█████████▉| 24127/24420 [10:04:39<07:17,  1.49s/it]
 99%|█████████▉| 24128/24420 [10:04:40<07:16,  1.50s/it]
 99%|█████████▉| 24129/24420 [10:04:42<07:14,  1.49s/it]
 99%|█████████▉| 24130/24420 [10:04:43<07:12,  1.49s/it]
                                                        
{'loss': 0.0819, 'learning_rate': 5e-05, 'epoch': 4.94}

 99%|█████████▉| 24130/24420 [10:04:43<07:12,  1.49s/it]
 99%|█████████▉| 24131/24420 [10:04:45<07:12,  1.50s/it]
 99%|█████████▉| 24132/24420 [10:04:46<07:10,  1.49s/it]
 99%|█████████▉| 24133/24420 [10:04:48<07:08,  1.49s/it]
 99%|█████████▉| 24134/24420 [10:04:49<07:06,  1.49s/it]
 99%|█████████▉| 24135/24420 [10:04:51<07:05,  1.49s/it]
                                                        
{'loss': 0.0554, 'learning_rate': 5e-05, 'epoch': 4.94}

 99%|█████████▉| 24135/24420 [10:04:51<07:05,  1.49s/it]
 99%|█████████▉| 24136/24420 [10:04:52<07:04,  1.49s/it]
 99%|█████████▉| 24137/24420 [10:04:54<07:02,  1.49s/it]
 99%|█████████▉| 24138/24420 [10:04:55<07:00,  1.49s/it]
 99%|█████████▉| 24139/24420 [10:04:57<06:59,  1.49s/it]
 99%|█████████▉| 24140/24420 [10:04:58<06:57,  1.49s/it]
                                                        
{'loss': 0.0695, 'learning_rate': 5e-05, 'epoch': 4.94}

 99%|█████████▉| 24140/24420 [10:04:58<06:57,  1.49s/it]
 99%|█████████▉| 24141/24420 [10:05:00<06:56,  1.49s/it]
 99%|█████████▉| 24142/24420 [10:05:01<06:54,  1.49s/it]
 99%|█████████▉| 24143/24420 [10:05:03<06:52,  1.49s/it]
 99%|█████████▉| 24144/24420 [10:05:04<06:50,  1.49s/it]
 99%|█████████▉| 24145/24420 [10:05:06<06:48,  1.49s/it]
                                                        
{'loss': 0.0799, 'learning_rate': 5e-05, 'epoch': 4.94}

 99%|█████████▉| 24145/24420 [10:05:06<06:48,  1.49s/it]
 99%|█████████▉| 24146/24420 [10:05:07<06:47,  1.49s/it]
 99%|█████████▉| 24147/24420 [10:05:09<06:45,  1.49s/it]
 99%|█████████▉| 24148/24420 [10:05:10<06:44,  1.49s/it]
 99%|█████████▉| 24149/24420 [10:05:12<06:43,  1.49s/it]
 99%|█████████▉| 24150/24420 [10:05:13<06:41,  1.49s/it]
                                                        
{'loss': 0.0664, 'learning_rate': 5e-05, 'epoch': 4.94}

 99%|█████████▉| 24150/24420 [10:05:13<06:41,  1.49s/it]
 99%|█████████▉| 24151/24420 [10:05:15<06:40,  1.49s/it]
 99%|█████████▉| 24152/24420 [10:05:16<06:38,  1.49s/it]
 99%|█████████▉| 24153/24420 [10:05:18<06:37,  1.49s/it]
 99%|█████████▉| 24154/24420 [10:05:19<06:35,  1.49s/it]
 99%|█████████▉| 24155/24420 [10:05:21<06:34,  1.49s/it]
                                                        
{'loss': 0.0709, 'learning_rate': 5e-05, 'epoch': 4.95}

 99%|█████████▉| 24155/24420 [10:05:21<06:34,  1.49s/it]
 99%|█████████▉| 24156/24420 [10:05:22<06:34,  1.49s/it]
 99%|█████████▉| 24157/24420 [10:05:24<06:32,  1.49s/it]
 99%|█████████▉| 24158/24420 [10:05:25<06:30,  1.49s/it]
 99%|█████████▉| 24159/24420 [10:05:27<06:28,  1.49s/it]
 99%|█████████▉| 24160/24420 [10:05:28<06:27,  1.49s/it]
                                                        
{'loss': 0.0752, 'learning_rate': 5e-05, 'epoch': 4.95}

 99%|█████████▉| 24160/24420 [10:05:28<06:27,  1.49s/it]
 99%|█████████▉| 24161/24420 [10:05:30<06:25,  1.49s/it]
 99%|█████████▉| 24162/24420 [10:05:31<06:24,  1.49s/it]
 99%|█████████▉| 24163/24420 [10:05:33<06:22,  1.49s/it]
 99%|█████████▉| 24164/24420 [10:05:34<06:21,  1.49s/it]
 99%|█████████▉| 24165/24420 [10:05:36<06:19,  1.49s/it]
                                                        
{'loss': 0.0738, 'learning_rate': 5e-05, 'epoch': 4.95}

 99%|█████████▉| 24165/24420 [10:05:36<06:19,  1.49s/it]
 99%|█████████▉| 24166/24420 [10:05:37<06:19,  1.49s/it]
 99%|█████████▉| 24167/24420 [10:05:39<06:17,  1.49s/it]
 99%|█████████▉| 24168/24420 [10:05:40<06:15,  1.49s/it]
 99%|█████████▉| 24169/24420 [10:05:42<06:13,  1.49s/it]
 99%|█████████▉| 24170/24420 [10:05:43<06:11,  1.49s/it]
                                                        
{'loss': 0.0648, 'learning_rate': 5e-05, 'epoch': 4.95}

 99%|█████████▉| 24170/24420 [10:05:43<06:11,  1.49s/it]
 99%|█████████▉| 24171/24420 [10:05:44<06:10,  1.49s/it]
 99%|█████████▉| 24172/24420 [10:05:46<06:08,  1.49s/it]
 99%|█████████▉| 24173/24420 [10:05:47<06:07,  1.49s/it]
 99%|█████████▉| 24174/24420 [10:05:49<06:05,  1.49s/it]
 99%|█████████▉| 24175/24420 [10:05:50<06:04,  1.49s/it]
                                                        
{'loss': 0.0621, 'learning_rate': 5e-05, 'epoch': 4.95}

 99%|█████████▉| 24175/24420 [10:05:50<06:04,  1.49s/it]
 99%|█████████▉| 24176/24420 [10:05:52<06:03,  1.49s/it]
 99%|█████████▉| 24177/24420 [10:05:53<06:02,  1.49s/it]
 99%|█████████▉| 24178/24420 [10:05:55<06:00,  1.49s/it]
 99%|█████████▉| 24179/24420 [10:05:56<05:58,  1.49s/it]
 99%|█████████▉| 24180/24420 [10:05:58<05:56,  1.49s/it]
                                                        
{'loss': 0.0775, 'learning_rate': 5e-05, 'epoch': 4.95}

 99%|█████████▉| 24180/24420 [10:05:58<05:56,  1.49s/it]
 99%|█████████▉| 24181/24420 [10:05:59<05:54,  1.48s/it]
 99%|█████████▉| 24182/24420 [10:06:01<05:53,  1.49s/it]
 99%|█████████▉| 24183/24420 [10:06:02<05:53,  1.49s/it]
 99%|█████████▉| 24184/24420 [10:06:04<05:51,  1.49s/it]
 99%|█████████▉| 24185/24420 [10:06:05<05:50,  1.49s/it]
                                                        
{'loss': 0.0801, 'learning_rate': 5e-05, 'epoch': 4.95}

 99%|█████████▉| 24185/24420 [10:06:05<05:50,  1.49s/it]
 99%|█████████▉| 24186/24420 [10:06:07<05:48,  1.49s/it]
 99%|█████████▉| 24187/24420 [10:06:08<05:47,  1.49s/it]
 99%|█████████▉| 24188/24420 [10:06:10<05:45,  1.49s/it]
 99%|█████████▉| 24189/24420 [10:06:11<05:44,  1.49s/it]
 99%|█████████▉| 24190/24420 [10:06:13<05:42,  1.49s/it]
                                                        
{'loss': 0.0626, 'learning_rate': 5e-05, 'epoch': 4.95}

 99%|█████████▉| 24190/24420 [10:06:13<05:42,  1.49s/it]
 99%|█████████▉| 24191/24420 [10:06:14<05:41,  1.49s/it]
 99%|█████████▉| 24192/24420 [10:06:16<05:40,  1.49s/it]
 99%|█████████▉| 24193/24420 [10:06:17<05:38,  1.49s/it]
 99%|█████████▉| 24194/24420 [10:06:19<05:36,  1.49s/it]
 99%|█████████▉| 24195/24420 [10:06:20<05:35,  1.49s/it]
                                                        
{'loss': 0.0838, 'learning_rate': 5e-05, 'epoch': 4.95}

 99%|█████████▉| 24195/24420 [10:06:20<05:35,  1.49s/it]
 99%|█████████▉| 24196/24420 [10:06:22<05:34,  1.49s/it]
 99%|█████████▉| 24197/24420 [10:06:23<05:32,  1.49s/it]
 99%|█████████▉| 24198/24420 [10:06:25<05:30,  1.49s/it]
 99%|█████████▉| 24199/24420 [10:06:26<05:28,  1.49s/it]
 99%|█████████▉| 24200/24420 [10:06:28<05:26,  1.48s/it]
                                                        
{'loss': 0.0691, 'learning_rate': 5e-05, 'epoch': 4.95}

 99%|█████████▉| 24200/24420 [10:06:28<05:26,  1.48s/it]
 99%|█████████▉| 24201/24420 [10:06:32<08:54,  2.44s/it]
 99%|█████████▉| 24202/24420 [10:06:34<07:49,  2.15s/it]
 99%|█████████▉| 24203/24420 [10:06:35<07:03,  1.95s/it]
 99%|█████████▉| 24204/24420 [10:06:37<06:30,  1.81s/it]
 99%|█████████▉| 24205/24420 [10:06:38<06:07,  1.71s/it]
                                                        
{'loss': 0.0591, 'learning_rate': 5e-05, 'epoch': 4.96}

 99%|█████████▉| 24205/24420 [10:06:38<06:07,  1.71s/it]
 99%|█████████▉| 24206/24420 [10:06:40<05:51,  1.64s/it]
 99%|█████████▉| 24207/24420 [10:06:41<05:39,  1.59s/it]
 99%|█████████▉| 24208/24420 [10:06:43<05:31,  1.56s/it]
 99%|█████████▉| 24209/24420 [10:06:44<05:24,  1.54s/it]
 99%|█████████▉| 24210/24420 [10:06:46<05:19,  1.52s/it]
                                                        
{'loss': 0.0697, 'learning_rate': 5e-05, 'epoch': 4.96}

 99%|█████████▉| 24210/24420 [10:06:46<05:19,  1.52s/it]
 99%|█████████▉| 24211/24420 [10:06:47<05:16,  1.51s/it]
 99%|█████████▉| 24212/24420 [10:06:49<05:12,  1.50s/it]
 99%|█████████▉| 24213/24420 [10:06:50<05:10,  1.50s/it]
 99%|█████████▉| 24214/24420 [10:06:52<05:07,  1.49s/it]
 99%|█████████▉| 24215/24420 [10:06:53<05:05,  1.49s/it]
                                                        
{'loss': 0.0621, 'learning_rate': 5e-05, 'epoch': 4.96}

 99%|█████████▉| 24215/24420 [10:06:53<05:05,  1.49s/it]
 99%|█████████▉| 24216/24420 [10:06:55<05:03,  1.49s/it]
 99%|█████████▉| 24217/24420 [10:06:56<05:02,  1.49s/it]
 99%|█████████▉| 24218/24420 [10:06:58<05:01,  1.49s/it]
 99%|█████████▉| 24219/24420 [10:06:59<04:59,  1.49s/it]
 99%|█████████▉| 24220/24420 [10:07:01<04:58,  1.49s/it]
                                                        
{'loss': 0.0731, 'learning_rate': 5e-05, 'epoch': 4.96}

 99%|█████████▉| 24220/24420 [10:07:01<04:58,  1.49s/it]
 99%|█████████▉| 24221/24420 [10:07:02<04:56,  1.49s/it]
 99%|█████████▉| 24222/24420 [10:07:04<04:58,  1.51s/it]
 99%|█████████▉| 24223/24420 [10:07:05<04:55,  1.50s/it]
 99%|█████████▉| 24224/24420 [10:07:07<04:52,  1.49s/it]
 99%|█████████▉| 24225/24420 [10:07:08<04:51,  1.49s/it]
                                                        
{'loss': 0.0944, 'learning_rate': 5e-05, 'epoch': 4.96}

 99%|█████████▉| 24225/24420 [10:07:08<04:51,  1.49s/it]
 99%|█████████▉| 24226/24420 [10:07:10<04:50,  1.50s/it]
 99%|█████████▉| 24227/24420 [10:07:11<04:48,  1.50s/it]
 99%|█████████▉| 24228/24420 [10:07:13<04:46,  1.49s/it]
 99%|█████████▉| 24229/24420 [10:07:14<04:44,  1.49s/it]
 99%|█████████▉| 24230/24420 [10:07:16<04:54,  1.55s/it]
                                                        
{'loss': 0.0892, 'learning_rate': 5e-05, 'epoch': 4.96}

 99%|█████████▉| 24230/24420 [10:07:16<04:54,  1.55s/it]
 99%|█████████▉| 24231/24420 [10:07:17<04:49,  1.53s/it]
 99%|█████████▉| 24232/24420 [10:07:19<04:45,  1.52s/it]
 99%|█████████▉| 24233/24420 [10:07:20<04:42,  1.51s/it]
 99%|█████████▉| 24234/24420 [10:07:22<04:39,  1.50s/it]
 99%|█████████▉| 24235/24420 [10:07:23<04:37,  1.50s/it]
                                                        
{'loss': 0.0822, 'learning_rate': 5e-05, 'epoch': 4.96}

 99%|█████████▉| 24235/24420 [10:07:23<04:37,  1.50s/it]
 99%|█████████▉| 24236/24420 [10:07:25<04:35,  1.50s/it]
 99%|█████████▉| 24237/24420 [10:07:26<04:33,  1.49s/it]
 99%|█████████▉| 24238/24420 [10:07:28<04:31,  1.49s/it]
 99%|█████████▉| 24239/24420 [10:07:29<04:29,  1.49s/it]
 99%|█████████▉| 24240/24420 [10:07:31<04:28,  1.49s/it]
                                                        
{'loss': 0.0658, 'learning_rate': 5e-05, 'epoch': 4.96}

 99%|█████████▉| 24240/24420 [10:07:31<04:28,  1.49s/it]
 99%|█████████▉| 24241/24420 [10:07:32<04:27,  1.50s/it]
 99%|█████████▉| 24242/24420 [10:07:34<04:26,  1.49s/it]
 99%|█████████▉| 24243/24420 [10:07:35<04:24,  1.50s/it]
 99%|█████████▉| 24244/24420 [10:07:37<04:22,  1.49s/it]
 99%|█████████▉| 24245/24420 [10:07:38<04:20,  1.49s/it]
                                                        
{'loss': 0.0714, 'learning_rate': 5e-05, 'epoch': 4.96}

 99%|█████████▉| 24245/24420 [10:07:38<04:20,  1.49s/it]
 99%|█████████▉| 24246/24420 [10:07:40<04:18,  1.49s/it]
 99%|█████████▉| 24247/24420 [10:07:41<04:17,  1.49s/it]
 99%|█████████▉| 24248/24420 [10:07:43<04:15,  1.49s/it]
 99%|█████████▉| 24249/24420 [10:07:44<04:14,  1.49s/it]
 99%|█████████▉| 24250/24420 [10:07:45<04:12,  1.49s/it]
                                                        
{'loss': 0.0636, 'learning_rate': 5e-05, 'epoch': 4.97}

 99%|█████████▉| 24250/24420 [10:07:45<04:12,  1.49s/it]
 99%|█████████▉| 24251/24420 [10:07:47<04:11,  1.49s/it]
 99%|█████████▉| 24252/24420 [10:07:48<04:10,  1.49s/it]
 99%|█████████▉| 24253/24420 [10:07:50<04:08,  1.49s/it]
 99%|█████████▉| 24254/24420 [10:07:51<04:06,  1.48s/it]
 99%|█████████▉| 24255/24420 [10:07:53<04:05,  1.49s/it]
                                                        
{'loss': 0.091, 'learning_rate': 5e-05, 'epoch': 4.97}

 99%|█████████▉| 24255/24420 [10:07:53<04:05,  1.49s/it]
 99%|█████████▉| 24256/24420 [10:07:54<04:03,  1.49s/it]
 99%|█████████▉| 24257/24420 [10:07:56<04:02,  1.49s/it]
 99%|█████████▉| 24258/24420 [10:07:57<04:01,  1.49s/it]
 99%|█████████▉| 24259/24420 [10:07:59<03:58,  1.48s/it]
 99%|█████████▉| 24260/24420 [10:08:00<03:58,  1.49s/it]
                                                        
{'loss': 0.065, 'learning_rate': 5e-05, 'epoch': 4.97}

 99%|█████████▉| 24260/24420 [10:08:00<03:58,  1.49s/it]
 99%|█████████▉| 24261/24420 [10:08:02<03:56,  1.49s/it]
 99%|█████████▉| 24262/24420 [10:08:03<03:54,  1.49s/it]
 99%|█████████▉| 24263/24420 [10:08:05<03:53,  1.49s/it]
 99%|█████████▉| 24264/24420 [10:08:06<03:51,  1.49s/it]
 99%|█████████▉| 24265/24420 [10:08:08<03:50,  1.49s/it]
                                                        
{'loss': 0.0642, 'learning_rate': 5e-05, 'epoch': 4.97}

 99%|█████████▉| 24265/24420 [10:08:08<03:50,  1.49s/it]
 99%|█████████▉| 24266/24420 [10:08:09<03:48,  1.49s/it]
 99%|█████████▉| 24267/24420 [10:08:11<03:47,  1.49s/it]
 99%|█████████▉| 24268/24420 [10:08:12<03:45,  1.49s/it]
 99%|█████████▉| 24269/24420 [10:08:14<03:45,  1.49s/it]
 99%|█████████▉| 24270/24420 [10:08:15<03:43,  1.49s/it]
                                                        
{'loss': 0.0692, 'learning_rate': 5e-05, 'epoch': 4.97}

 99%|█████████▉| 24270/24420 [10:08:15<03:43,  1.49s/it]
 99%|█████████▉| 24271/24420 [10:08:17<03:41,  1.49s/it]
 99%|█████████▉| 24272/24420 [10:08:18<03:40,  1.49s/it]
 99%|█████████▉| 24273/24420 [10:08:20<03:38,  1.49s/it]
 99%|█████████▉| 24274/24420 [10:08:21<03:37,  1.49s/it]
 99%|█████████▉| 24275/24420 [10:08:23<03:36,  1.49s/it]
                                                        
{'loss': 0.071, 'learning_rate': 5e-05, 'epoch': 4.97}

 99%|█████████▉| 24275/24420 [10:08:23<03:36,  1.49s/it]
 99%|█████████▉| 24276/24420 [10:08:24<03:34,  1.49s/it]
 99%|█████████▉| 24277/24420 [10:08:26<03:32,  1.49s/it]
 99%|█████████▉| 24278/24420 [10:08:27<03:31,  1.49s/it]
 99%|█████████▉| 24279/24420 [10:08:29<03:29,  1.49s/it]
 99%|█████████▉| 24280/24420 [10:08:30<03:28,  1.49s/it]
                                                        
{'loss': 0.0654, 'learning_rate': 5e-05, 'epoch': 4.97}

 99%|█████████▉| 24280/24420 [10:08:30<03:28,  1.49s/it]
 99%|█████████▉| 24281/24420 [10:08:32<03:26,  1.49s/it]
 99%|█████████▉| 24282/24420 [10:08:33<03:25,  1.49s/it]
 99%|█████████▉| 24283/24420 [10:08:35<03:23,  1.49s/it]
 99%|█████████▉| 24284/24420 [10:08:36<03:22,  1.49s/it]
 99%|█████████▉| 24285/24420 [10:08:38<03:20,  1.49s/it]
                                                        
{'loss': 0.0779, 'learning_rate': 5e-05, 'epoch': 4.97}

 99%|█████████▉| 24285/24420 [10:08:38<03:20,  1.49s/it]
 99%|█████████▉| 24286/24420 [10:08:39<03:19,  1.49s/it]
 99%|█████████▉| 24287/24420 [10:08:41<03:17,  1.49s/it]
 99%|█████████▉| 24288/24420 [10:08:42<03:16,  1.49s/it]
 99%|█████████▉| 24289/24420 [10:08:44<03:14,  1.49s/it]
 99%|█████████▉| 24290/24420 [10:08:45<03:13,  1.49s/it]
                                                        
{'loss': 0.076, 'learning_rate': 5e-05, 'epoch': 4.97}

 99%|█████████▉| 24290/24420 [10:08:45<03:13,  1.49s/it]
 99%|█████████▉| 24291/24420 [10:08:46<03:11,  1.49s/it]
 99%|█████████▉| 24292/24420 [10:08:48<03:10,  1.49s/it]
 99%|█████████▉| 24293/24420 [10:08:49<03:09,  1.49s/it]
 99%|█████████▉| 24294/24420 [10:08:51<03:07,  1.49s/it]
 99%|█████████▉| 24295/24420 [10:08:52<03:06,  1.49s/it]
                                                        
{'loss': 0.0613, 'learning_rate': 5e-05, 'epoch': 4.97}

 99%|█████████▉| 24295/24420 [10:08:52<03:06,  1.49s/it]
 99%|█████████▉| 24296/24420 [10:08:54<03:04,  1.49s/it]
 99%|█████████▉| 24297/24420 [10:08:55<03:03,  1.49s/it]
100%|█████████▉| 24298/24420 [10:08:57<03:01,  1.49s/it]
100%|█████████▉| 24299/24420 [10:08:58<03:00,  1.49s/it]
100%|█████████▉| 24300/24420 [10:09:00<02:58,  1.49s/it]
                                                        
{'loss': 0.0674, 'learning_rate': 5e-05, 'epoch': 4.98}

100%|█████████▉| 24300/24420 [10:09:00<02:58,  1.49s/it]
100%|█████████▉| 24301/24420 [10:09:01<02:57,  1.49s/it]
100%|█████████▉| 24302/24420 [10:09:03<02:55,  1.49s/it]
100%|█████████▉| 24303/24420 [10:09:04<02:54,  1.49s/it]
100%|█████████▉| 24304/24420 [10:09:06<02:52,  1.49s/it]
100%|█████████▉| 24305/24420 [10:09:07<02:51,  1.49s/it]
                                                        
{'loss': 0.0783, 'learning_rate': 5e-05, 'epoch': 4.98}

100%|█████████▉| 24305/24420 [10:09:07<02:51,  1.49s/it]
100%|█████████▉| 24306/24420 [10:09:09<02:49,  1.49s/it]
100%|█████████▉| 24307/24420 [10:09:10<02:48,  1.49s/it]
100%|█████████▉| 24308/24420 [10:09:12<02:46,  1.49s/it]
100%|█████████▉| 24309/24420 [10:09:13<02:45,  1.49s/it]
100%|█████████▉| 24310/24420 [10:09:15<02:43,  1.49s/it]
                                                        
{'loss': 0.0751, 'learning_rate': 5e-05, 'epoch': 4.98}

100%|█████████▉| 24310/24420 [10:09:15<02:43,  1.49s/it]
100%|█████████▉| 24311/24420 [10:09:16<02:42,  1.49s/it]
100%|█████████▉| 24312/24420 [10:09:18<02:40,  1.49s/it]
100%|█████████▉| 24313/24420 [10:09:19<02:39,  1.49s/it]
100%|█████████▉| 24314/24420 [10:09:21<02:37,  1.49s/it]
100%|█████████▉| 24315/24420 [10:09:22<02:36,  1.49s/it]
                                                        
{'loss': 0.066, 'learning_rate': 5e-05, 'epoch': 4.98}

100%|█████████▉| 24315/24420 [10:09:22<02:36,  1.49s/it]
100%|█████████▉| 24316/24420 [10:09:24<02:34,  1.49s/it]
100%|█████████▉| 24317/24420 [10:09:25<02:33,  1.49s/it]
100%|█████████▉| 24318/24420 [10:09:27<02:31,  1.49s/it]
100%|█████████▉| 24319/24420 [10:09:28<02:30,  1.49s/it]
100%|█████████▉| 24320/24420 [10:09:30<02:29,  1.49s/it]
                                                        
{'loss': 0.0668, 'learning_rate': 5e-05, 'epoch': 4.98}

100%|█████████▉| 24320/24420 [10:09:30<02:29,  1.49s/it]
100%|█████████▉| 24321/24420 [10:09:31<02:27,  1.49s/it]
100%|█████████▉| 24322/24420 [10:09:33<02:26,  1.49s/it]
100%|█████████▉| 24323/24420 [10:09:34<02:24,  1.49s/it]
100%|█████████▉| 24324/24420 [10:09:36<02:23,  1.50s/it]
100%|█████████▉| 24325/24420 [10:09:37<02:22,  1.50s/it]
                                                        
{'loss': 0.0754, 'learning_rate': 5e-05, 'epoch': 4.98}

100%|█████████▉| 24325/24420 [10:09:37<02:22,  1.50s/it]
100%|█████████▉| 24326/24420 [10:09:39<02:20,  1.50s/it]
100%|█████████▉| 24327/24420 [10:09:40<02:19,  1.50s/it]
100%|█████████▉| 24328/24420 [10:09:42<02:17,  1.50s/it]
100%|█████████▉| 24329/24420 [10:09:43<02:16,  1.50s/it]
100%|█████████▉| 24330/24420 [10:09:45<02:14,  1.50s/it]
                                                        
{'loss': 0.0845, 'learning_rate': 5e-05, 'epoch': 4.98}

100%|█████████▉| 24330/24420 [10:09:45<02:14,  1.50s/it]
100%|█████████▉| 24331/24420 [10:09:46<02:12,  1.49s/it]
100%|█████████▉| 24332/24420 [10:09:48<02:11,  1.49s/it]
100%|█████████▉| 24333/24420 [10:09:49<02:09,  1.49s/it]
100%|█████████▉| 24334/24420 [10:09:51<02:08,  1.49s/it]
100%|█████████▉| 24335/24420 [10:09:52<02:06,  1.49s/it]
                                                        
{'loss': 0.0653, 'learning_rate': 5e-05, 'epoch': 4.98}

100%|█████████▉| 24335/24420 [10:09:52<02:06,  1.49s/it]
100%|█████████▉| 24336/24420 [10:09:54<02:05,  1.49s/it]
100%|█████████▉| 24337/24420 [10:09:55<02:04,  1.49s/it]
100%|█████████▉| 24338/24420 [10:09:57<02:02,  1.49s/it]
100%|█████████▉| 24339/24420 [10:09:58<02:01,  1.49s/it]
100%|█████████▉| 24340/24420 [10:10:00<01:59,  1.49s/it]
                                                        
{'loss': 0.0707, 'learning_rate': 5e-05, 'epoch': 4.98}

100%|█████████▉| 24340/24420 [10:10:00<01:59,  1.49s/it]
100%|█████████▉| 24341/24420 [10:10:01<01:57,  1.49s/it]
100%|█████████▉| 24342/24420 [10:10:03<01:56,  1.49s/it]
100%|█████████▉| 24343/24420 [10:10:04<01:55,  1.49s/it]
100%|█████████▉| 24344/24420 [10:10:06<01:53,  1.49s/it]
100%|█████████▉| 24345/24420 [10:10:07<01:52,  1.49s/it]
                                                        
{'loss': 0.0759, 'learning_rate': 5e-05, 'epoch': 4.98}

100%|█████████▉| 24345/24420 [10:10:07<01:52,  1.49s/it]
100%|█████████▉| 24346/24420 [10:10:09<01:50,  1.49s/it]
100%|█████████▉| 24347/24420 [10:10:10<01:49,  1.49s/it]
100%|█████████▉| 24348/24420 [10:10:12<01:47,  1.49s/it]
100%|█████████▉| 24349/24420 [10:10:13<01:45,  1.49s/it]
100%|█████████▉| 24350/24420 [10:10:14<01:44,  1.49s/it]
                                                        
{'loss': 0.0838, 'learning_rate': 5e-05, 'epoch': 4.99}

100%|█████████▉| 24350/24420 [10:10:14<01:44,  1.49s/it]
100%|█████████▉| 24351/24420 [10:10:16<01:43,  1.49s/it]
100%|█████████▉| 24352/24420 [10:10:17<01:41,  1.49s/it]
100%|█████████▉| 24353/24420 [10:10:19<01:40,  1.49s/it]
100%|█████████▉| 24354/24420 [10:10:20<01:38,  1.49s/it]
100%|█████████▉| 24355/24420 [10:10:22<01:36,  1.49s/it]
                                                        
{'loss': 0.0783, 'learning_rate': 5e-05, 'epoch': 4.99}

100%|█████████▉| 24355/24420 [10:10:22<01:36,  1.49s/it]
100%|█████████▉| 24356/24420 [10:10:23<01:35,  1.49s/it]
100%|█████████▉| 24357/24420 [10:10:25<01:33,  1.49s/it]
100%|█████████▉| 24358/24420 [10:10:26<01:32,  1.49s/it]
100%|█████████▉| 24359/24420 [10:10:28<01:30,  1.49s/it]
100%|█████████▉| 24360/24420 [10:10:29<01:29,  1.49s/it]
                                                        
{'loss': 0.0766, 'learning_rate': 5e-05, 'epoch': 4.99}

100%|█████████▉| 24360/24420 [10:10:29<01:29,  1.49s/it]
100%|█████████▉| 24361/24420 [10:10:31<01:27,  1.49s/it]
100%|█████████▉| 24362/24420 [10:10:32<01:26,  1.49s/it]
100%|█████████▉| 24363/24420 [10:10:34<01:24,  1.49s/it]
100%|█████████▉| 24364/24420 [10:10:35<01:24,  1.50s/it]
100%|█████████▉| 24365/24420 [10:10:37<01:22,  1.50s/it]
                                                        
{'loss': 0.0698, 'learning_rate': 5e-05, 'epoch': 4.99}

100%|█████████▉| 24365/24420 [10:10:37<01:22,  1.50s/it]
100%|█████████▉| 24366/24420 [10:10:38<01:20,  1.50s/it]
100%|█████████▉| 24367/24420 [10:10:40<01:19,  1.51s/it]
100%|█████████▉| 24368/24420 [10:10:41<01:18,  1.50s/it]
100%|█████████▉| 24369/24420 [10:10:43<01:16,  1.50s/it]
100%|█████████▉| 24370/24420 [10:10:44<01:14,  1.50s/it]
                                                        
{'loss': 0.0635, 'learning_rate': 5e-05, 'epoch': 4.99}

100%|█████████▉| 24370/24420 [10:10:44<01:14,  1.50s/it]
100%|█████████▉| 24371/24420 [10:10:46<01:13,  1.50s/it]
100%|█████████▉| 24372/24420 [10:10:47<01:12,  1.50s/it]
100%|█████████▉| 24373/24420 [10:10:49<01:10,  1.50s/it]
100%|█████████▉| 24374/24420 [10:10:50<01:08,  1.50s/it]
100%|█████████▉| 24375/24420 [10:10:52<01:07,  1.50s/it]
                                                        
{'loss': 0.0677, 'learning_rate': 5e-05, 'epoch': 4.99}

100%|█████████▉| 24375/24420 [10:10:52<01:07,  1.50s/it]
100%|█████████▉| 24376/24420 [10:10:53<01:05,  1.50s/it]
100%|█████████▉| 24377/24420 [10:10:55<01:04,  1.49s/it]
100%|█████████▉| 24378/24420 [10:10:56<01:02,  1.49s/it]
100%|█████████▉| 24379/24420 [10:10:58<01:01,  1.50s/it]
100%|█████████▉| 24380/24420 [10:10:59<00:59,  1.50s/it]
                                                        
{'loss': 0.0717, 'learning_rate': 5e-05, 'epoch': 4.99}

100%|█████████▉| 24380/24420 [10:10:59<00:59,  1.50s/it]
100%|█████████▉| 24381/24420 [10:11:01<00:58,  1.49s/it]
100%|█████████▉| 24382/24420 [10:11:02<00:56,  1.50s/it]
100%|█████████▉| 24383/24420 [10:11:04<00:55,  1.49s/it]
100%|█████████▉| 24384/24420 [10:11:05<00:53,  1.49s/it]
100%|█████████▉| 24385/24420 [10:11:07<00:52,  1.49s/it]
                                                        
{'loss': 0.0706, 'learning_rate': 5e-05, 'epoch': 4.99}

100%|█████████▉| 24385/24420 [10:11:07<00:52,  1.49s/it]
100%|█████████▉| 24386/24420 [10:11:08<00:50,  1.49s/it]
100%|█████████▉| 24387/24420 [10:11:10<00:49,  1.49s/it]
100%|█████████▉| 24388/24420 [10:11:11<00:47,  1.49s/it]
100%|█████████▉| 24389/24420 [10:11:13<00:46,  1.49s/it]
100%|█████████▉| 24390/24420 [10:11:14<00:44,  1.49s/it]
                                                        
{'loss': 0.0665, 'learning_rate': 5e-05, 'epoch': 4.99}

100%|█████████▉| 24390/24420 [10:11:14<00:44,  1.49s/it]
100%|█████████▉| 24391/24420 [10:11:16<00:43,  1.49s/it]
100%|█████████▉| 24392/24420 [10:11:17<00:41,  1.49s/it]
100%|█████████▉| 24393/24420 [10:11:19<00:40,  1.49s/it]
100%|█████████▉| 24394/24420 [10:11:20<00:38,  1.49s/it]
100%|█████████▉| 24395/24420 [10:11:22<00:37,  1.49s/it]
                                                        
{'loss': 0.0681, 'learning_rate': 5e-05, 'epoch': 4.99}

100%|█████████▉| 24395/24420 [10:11:22<00:37,  1.49s/it]
100%|█████████▉| 24396/24420 [10:11:23<00:35,  1.49s/it]
100%|█████████▉| 24397/24420 [10:11:25<00:34,  1.49s/it]
100%|█████████▉| 24398/24420 [10:11:26<00:32,  1.49s/it]
100%|█████████▉| 24399/24420 [10:11:28<00:31,  1.49s/it]
100%|█████████▉| 24400/24420 [10:11:29<00:29,  1.49s/it]
                                                        
{'loss': 0.0654, 'learning_rate': 5e-05, 'epoch': 5.0}

100%|█████████▉| 24400/24420 [10:11:29<00:29,  1.49s/it]
100%|█████████▉| 24401/24420 [10:11:35<00:53,  2.83s/it]
100%|█████████▉| 24402/24420 [10:11:37<00:43,  2.43s/it]
100%|█████████▉| 24403/24420 [10:11:38<00:36,  2.15s/it]
100%|█████████▉| 24404/24420 [10:11:40<00:31,  1.95s/it]
100%|█████████▉| 24405/24420 [10:11:41<00:27,  1.81s/it]
                                                        
{'loss': 0.0926, 'learning_rate': 5e-05, 'epoch': 5.0}

100%|█████████▉| 24405/24420 [10:11:41<00:27,  1.81s/it]
100%|█████████▉| 24406/24420 [10:11:43<00:24,  1.72s/it]
100%|█████████▉| 24407/24420 [10:11:44<00:21,  1.65s/it]
100%|█████████▉| 24408/24420 [10:11:46<00:19,  1.60s/it]
100%|█████████▉| 24409/24420 [10:11:47<00:17,  1.56s/it]
100%|█████████▉| 24410/24420 [10:11:49<00:15,  1.54s/it]
                                                        
{'loss': 0.0682, 'learning_rate': 5e-05, 'epoch': 5.0}

100%|█████████▉| 24410/24420 [10:11:49<00:15,  1.54s/it]
100%|█████████▉| 24411/24420 [10:11:50<00:13,  1.53s/it]
100%|█████████▉| 24412/24420 [10:11:52<00:12,  1.52s/it]
100%|█████████▉| 24413/24420 [10:11:53<00:10,  1.51s/it]
100%|█████████▉| 24414/24420 [10:11:55<00:09,  1.51s/it]
100%|█████████▉| 24415/24420 [10:11:56<00:07,  1.50s/it]
                                                        
{'loss': 0.0853, 'learning_rate': 5e-05, 'epoch': 5.0}

100%|█████████▉| 24415/24420 [10:11:56<00:07,  1.50s/it]
100%|█████████▉| 24416/24420 [10:11:57<00:05,  1.50s/it]
100%|█████████▉| 24417/24420 [10:11:59<00:04,  1.49s/it]
100%|█████████▉| 24418/24420 [10:12:00<00:02,  1.49s/it]
100%|█████████▉| 24419/24420 [10:12:02<00:01,  1.49s/it]
100%|██████████| 24420/24420 [10:12:03<00:00,  1.49s/it]
                                                        
{'loss': 0.0634, 'learning_rate': 5e-05, 'epoch': 5.0}

100%|██████████| 24420/24420 [10:12:03<00:00,  1.49s/it]Time: 36724.34
Samples/second: 15.96

                                                        
Time: 36724.34{'train_runtime': 36723.963, 'train_samples_per_second': 15.961, 'train_steps_per_second': 0.665, 'train_loss': 0.12059639945784911, 'epoch': 5.0}

Samples/second: 15.96

100%|██████████| 24420/24420 [10:12:03<00:00,  1.49s/it]GPU memory occupied: 47582 MB.
GPU memory occupied: 47582 MB.

100%|██████████| 24420/24420 [10:12:03<00:00,  1.50s/it]
Time: 36723.96
Samples/second: 15.96
GPU memory occupied: 47582 MB.
Time: 36724.47
Samples/second: 15.96
GPU memory occupied: 47582 MB.
[2022-12-21 03:32:45,648] [INFO] [launch.py:350:main] Process 2267 exits successfully.
[2022-12-21 03:32:46,650] [INFO] [launch.py:350:main] Process 2269 exits successfully.
[2022-12-21 03:32:46,650] [INFO] [launch.py:350:main] Process 2270 exits successfully.
[2022-12-21 03:32:46,652] [INFO] [launch.py:350:main] Process 2268 exits successfully.