Upload experiments/2022-10-19-515cf3b9155fd406d5067b25b7a969d2fc7be8e238d63667d772142982e8e3ff with huggingface_hub

experiments/2022-10-19-515cf3b9155fd406d5067b25b7a969d2fc7be8e238d63667d772142982e8e3ff/configs.json

@@ -0,0 +1,25 @@
+{
+    "output_dir": "./results",
+    "evaluation_strategy": "no",
+    "do_eval": false,
+    "eval_steps": 0,
+    "log_level": "error",
+    "logging_first_step": true,
+    "logging_steps": 5,
+    "logging_dir": "./logs",
+    "save_steps": 200,
+    "save_total_limit": 2,
+    "num_train_epochs": 5,
+    "per_device_train_batch_size": 12,
+    "optim": "adamw_torch",
+    "gradient_accumulation_steps": 4,
+    "dataloader_drop_last": true,
+    "warmup_steps": 100,
+    "weight_decay": 0.01,
+    "learning_rate": 5e-05,
+    "fp16": true,
+    "total_gpus": 1,
+    "v_cpus": 6,
+    "total_memory_in_gb": 23040.7890625,
+    "dataset_limit": 100
+}

experiments/2022-10-19-515cf3b9155fd406d5067b25b7a969d2fc7be8e238d63667d772142982e8e3ff/final_checkpoint/config.json

@@ -0,0 +1,54 @@
+{
+  "_name_or_path": "EleutherAI/gpt-neo-125M",
+  "activation_function": "gelu_new",
+  "architectures": [
+    "GPTNeoForCausalLM"
+  ],
+  "attention_dropout": 0,
+  "attention_layers": [
+    "global",
+    "local",
+    "global",
+    "local",
+    "global",
+    "local",
+    "global",
+    "local",
+    "global",
+    "local",
+    "global",
+    "local"
+  ],
+  "attention_types": [
+    [
+      [
+        "global",
+        "local"
+      ],
+      6
+    ]
+  ],
+  "bos_token_id": 50256,
+  "embed_dropout": 0,
+  "eos_token_id": 50256,
+  "gradient_checkpointing": false,
+  "hidden_size": 768,
+  "initializer_range": 0.02,
+  "intermediate_size": null,
+  "layer_norm_epsilon": 1e-05,
+  "max_position_embeddings": 2048,
+  "model_type": "gpt_neo",
+  "num_heads": 12,
+  "num_layers": 12,
+  "resid_dropout": 0,
+  "summary_activation": null,
+  "summary_first_dropout": 0.1,
+  "summary_proj_to_labels": true,
+  "summary_type": "cls_index",
+  "summary_use_proj": true,
+  "torch_dtype": "float32",
+  "transformers_version": "4.23.1",
+  "use_cache": true,
+  "vocab_size": 50259,
+  "window_size": 256
+}

experiments/2022-10-19-515cf3b9155fd406d5067b25b7a969d2fc7be8e238d63667d772142982e8e3ff/final_checkpoint/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ede1dadf6abcfb32d6b34e431179bc5dab08c31b2637c3c4c3e2798f42b7859c
+size 551191249

experiments/2022-10-19-515cf3b9155fd406d5067b25b7a969d2fc7be8e238d63667d772142982e8e3ff/logs/1666220848.4318457/events.out.tfevents.1666220848.recd68n2rbwktkbnt.4475.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ec3fef782ad78bdeb10bbbcb409e2d7b969965e2f316c5571f8c16e5b188b3bc
+size 5413

experiments/2022-10-19-515cf3b9155fd406d5067b25b7a969d2fc7be8e238d63667d772142982e8e3ff/logs/events.out.tfevents.1666220848.recd68n2rbwktkbnt.4475.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e3a062804908bee000b28c420e1909b82c34a084b0730684d34eec9686a0dead
+size 9341

experiments/2022-10-19-515cf3b9155fd406d5067b25b7a969d2fc7be8e238d63667d772142982e8e3ff/output.log

@@ -0,0 +1,46 @@
+nohup: ignoring input
+Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
+Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
+No config specified, defaulting to: apps/all
+Found cached dataset apps (/home/user/.cache/huggingface/datasets/codeparrot___apps/all/0.0.0/04ac807715d07d6e5cc580f59cdc8213cd7dc4529d0bb819cca72c9f8e8c1aa5)
+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 :-).
+GPU memory occupied: 2667 MB.
+
  0%|          | 0/155 [00:00<?, ?it/s]
  1%|          | 1/155 [00:03<09:53,  3.85s/it]
                                               
+
  1%|          | 1/155 [00:03<09:53,  3.85s/it]
  1%|▏         | 2/155 [00:07<08:58,  3.52s/it]
  2%|▏         | 3/155 [00:10<08:38,  3.41s/it]
  3%|▎         | 4/155 [00:13<08:29,  3.37s/it]
  3%|▎         | 5/155 [00:17<08:22,  3.35s/it]
                                               
+
  3%|▎         | 5/155 [00:17<08:22,  3.35s/it]
  4%|▍         | 6/155 [00:20<08:16,  3.33s/it]
  5%|▍         | 7/155 [00:23<08:12,  3.33s/it]
  5%|▌         | 8/155 [00:26<08:08,  3.32s/it]
  6%|▌         | 9/155 [00:30<08:04,  3.32s/it]
  6%|▋         | 10/155 [00:33<08:01,  3.32s/it]
                                                
+
  6%|▋         | 10/155 [00:33<08:01,  3.32s/it]
  7%|▋         | 11/155 [00:36<07:57,  3.32s/it]
  8%|▊         | 12/155 [00:40<07:54,  3.32s/it]
  8%|▊         | 13/155 [00:43<07:51,  3.32s/it]
  9%|▉         | 14/155 [00:46<07:48,  3.32s/it]
 10%|▉         | 15/155 [00:50<07:45,  3.32s/it]
                                                
+
 10%|▉         | 15/155 [00:50<07:45,  3.32s/it]
 10%|█         | 16/155 [00:53<07:41,  3.32s/it]
 11%|█         | 17/155 [00:56<07:38,  3.32s/it]
 12%|█▏        | 18/155 [01:00<07:35,  3.32s/it]
 12%|█▏        | 19/155 [01:03<07:32,  3.32s/it]
 13%|█▎        | 20/155 [01:06<07:28,  3.32s/it]
                                                
+
 13%|█▎        | 20/155 [01:06<07:28,  3.32s/it]
 14%|█▎        | 21/155 [01:10<07:25,  3.32s/it]
 14%|█▍        | 22/155 [01:13<07:22,  3.33s/it]
 15%|█▍        | 23/155 [01:16<07:19,  3.33s/it]
 15%|█▌        | 24/155 [01:20<07:15,  3.33s/it]
 16%|█▌        | 25/155 [01:23<07:12,  3.33s/it]
                                                
+
 16%|█▌        | 25/155 [01:23<07:12,  3.33s/it]
 17%|█▋        | 26/155 [01:26<07:09,  3.33s/it]
 17%|█▋        | 27/155 [01:30<07:06,  3.33s/it]
 18%|█▊        | 28/155 [01:33<07:02,  3.33s/it]
 19%|█▊        | 29/155 [01:36<06:59,  3.33s/it]
 19%|█▉        | 30/155 [01:40<06:56,  3.33s/it]
                                                
+
 19%|█▉        | 30/155 [01:40<06:56,  3.33s/it]
 20%|██        | 31/155 [01:43<06:53,  3.33s/it]
 21%|██        | 32/155 [01:49<08:21,  4.08s/it]
 21%|██▏       | 33/155 [01:52<07:50,  3.85s/it]
 22%|██▏       | 34/155 [01:55<07:27,  3.70s/it]
 23%|██▎       | 35/155 [01:59<07:10,  3.59s/it]
                                                
+
 23%|██▎       | 35/155 [01:59<07:10,  3.59s/it]
 23%|██▎       | 36/155 [02:02<06:57,  3.51s/it]
 24%|██▍       | 37/155 [02:05<06:48,  3.46s/it]
 25%|██▍       | 38/155 [02:09<06:40,  3.42s/it]
 25%|██▌       | 39/155 [02:12<06:33,  3.39s/it]
 26%|██▌       | 40/155 [02:15<06:27,  3.37s/it]
                                                
+
 26%|██▌       | 40/155 [02:15<06:27,  3.37s/it]
 26%|██▋       | 41/155 [02:19<06:23,  3.36s/it]
 27%|██▋       | 42/155 [02:22<06:18,  3.35s/it]
 28%|██▊       | 43/155 [02:25<06:14,  3.34s/it]
 28%|██▊       | 44/155 [02:29<06:10,  3.34s/it]
 29%|██▉       | 45/155 [02:32<06:06,  3.34s/it]
                                                
+
 29%|██▉       | 45/155 [02:32<06:06,  3.34s/it]
 30%|██▉       | 46/155 [02:35<06:03,  3.33s/it]
 30%|███       | 47/155 [02:39<05:59,  3.33s/it]
 31%|███       | 48/155 [02:42<05:56,  3.33s/it]
 32%|███���      | 49/155 [02:45<05:53,  3.33s/it]
 32%|███▏      | 50/155 [02:49<05:49,  3.33s/it]
                                                
+
 32%|███▏      | 50/155 [02:49<05:49,  3.33s/it]
 33%|███▎      | 51/155 [02:52<05:46,  3.33s/it]
 34%|███▎      | 52/155 [02:55<05:43,  3.33s/it]
 34%|███▍      | 53/155 [02:59<05:39,  3.33s/it]
 35%|███▍      | 54/155 [03:02<05:36,  3.33s/it]
 35%|███▌      | 55/155 [03:05<05:33,  3.33s/it]
                                                
+
 35%|███▌      | 55/155 [03:05<05:33,  3.33s/it]
 36%|███▌      | 56/155 [03:09<05:29,  3.33s/it]
 37%|███▋      | 57/155 [03:12<05:26,  3.33s/it]
 37%|███▋      | 58/155 [03:15<05:22,  3.33s/it]
 38%|███▊      | 59/155 [03:19<05:19,  3.33s/it]
 39%|███▊      | 60/155 [03:22<05:16,  3.33s/it]
                                                
+
 39%|███▊      | 60/155 [03:22<05:16,  3.33s/it]
 39%|███▉      | 61/155 [03:25<05:12,  3.33s/it]
 40%|████      | 62/155 [03:29<05:09,  3.33s/it]
 41%|████      | 63/155 [03:34<06:14,  4.07s/it]
 41%|████▏     | 64/155 [03:38<05:50,  3.85s/it]
 42%|████▏     | 65/155 [03:41<05:32,  3.69s/it]
                                                
+
 42%|████▏     | 65/155 [03:41<05:32,  3.69s/it]
 43%|████▎     | 66/155 [03:44<05:19,  3.59s/it]
 43%|████▎     | 67/155 [03:48<05:08,  3.51s/it]
 44%|████▍     | 68/155 [03:51<05:00,  3.45s/it]
 45%|████▍     | 69/155 [03:54<04:53,  3.42s/it]
 45%|████▌     | 70/155 [03:58<04:48,  3.39s/it]
                                                
+
 45%|████▌     | 70/155 [03:58<04:48,  3.39s/it]
 46%|████▌     | 71/155 [04:01<04:43,  3.37s/it]
 46%|████▋     | 72/155 [04:04<04:38,  3.36s/it]
 47%|████▋     | 73/155 [04:08<04:34,  3.35s/it]
 48%|████▊     | 74/155 [04:11<04:30,  3.34s/it]
 48%|████▊     | 75/155 [04:14<04:27,  3.34s/it]
                                                
+
 48%|████▊     | 75/155 [04:14<04:27,  3.34s/it]
 49%|████▉     | 76/155 [04:18<04:23,  3.34s/it]
 50%|████▉     | 77/155 [04:21<04:20,  3.33s/it]
 50%|█████     | 78/155 [04:24<04:16,  3.33s/it]
 51%|█████     | 79/155 [04:28<04:13,  3.33s/it]
 52%|█████▏    | 80/155 [04:31<04:09,  3.33s/it]
                                                
+
 52%|█████▏    | 80/155 [04:31<04:09,  3.33s/it]
 52%|█████▏    | 81/155 [04:34<04:06,  3.33s/it]
 53%|█████▎    | 82/155 [04:38<04:03,  3.33s/it]
 54%|█████▎    | 83/155 [04:41<03:59,  3.33s/it]
 54%|█████▍    | 84/155 [04:44<03:56,  3.33s/it]
 55%|█████▍    | 85/155 [04:48<03:52,  3.33s/it]
                                                
+
 55%|█████▍    | 85/155 [04:48<03:52,  3.33s/it]
 55%|█████▌    | 86/155 [04:51<03:49,  3.33s/it]
 56%|█████▌    | 87/155 [04:54<03:46,  3.33s/it]
 57%|█████▋    | 88/155 [04:58<03:43,  3.33s/it]
 57%|█████▋    | 89/155 [05:01<03:39,  3.33s/it]
 58%|█████▊    | 90/155 [05:04<03:36,  3.33s/it]
                                                
+
 58%|█████▊    | 90/155 [05:04<03:36,  3.33s/it]
 59%|█████▊    | 91/155 [05:08<03:33,  3.33s/it]
 59%|█████▉    | 92/155 [05:11<03:29,  3.33s/it]
 60%|██████    | 93/155 [05:14<03:26,  3.33s/it]
 61%|██████    | 94/155 [05:20<04:08,  4.07s/it]
 61%|██████▏   | 95/155 [05:23<03:51,  3.85s/it]
                                                
+
 61%|██████▏   | 95/155 [05:23<03:51,  3.85s/it]
 62%|██████▏   | 96/155 [05:27<03:37,  3.69s/it]
 63%|██████▎   | 97/155 [05:30<03:27,  3.58s/it]
 63%|██████▎   | 98/155 [05:33<03:19,  3.51s/it]
 64%|██████▍   | 99/155 [05:37<03:13,  3.45s/it]
 65%|██████▍   | 100/155 [05:40<03:07,  3.42s/it]
                                                 
+
 65%|██████▍   | 100/155 [05:40<03:07,  3.42s/it]
 65%|██████▌   | 101/155 [05:43<03:03,  3.39s/it]
 66%|██████▌   | 102/155 [05:47<02:58,  3.37s/it]
 66%|█���████▋   | 103/155 [05:50<02:54,  3.36s/it]
 67%|██████▋   | 104/155 [05:53<02:50,  3.35s/it]
 68%|██████▊   | 105/155 [05:57<02:46,  3.34s/it]
                                                 
+
 68%|██████▊   | 105/155 [05:57<02:46,  3.34s/it]
 68%|██████▊   | 106/155 [06:00<02:43,  3.34s/it]
 69%|██████▉   | 107/155 [06:03<02:39,  3.33s/it]
 70%|██████▉   | 108/155 [06:07<02:36,  3.33s/it]
 70%|███████   | 109/155 [06:10<02:33,  3.33s/it]
 71%|███████   | 110/155 [06:13<02:29,  3.33s/it]
                                                 
+
 71%|███████   | 110/155 [06:13<02:29,  3.33s/it]
 72%|███████▏  | 111/155 [06:17<02:26,  3.33s/it]
 72%|███████▏  | 112/155 [06:20<02:23,  3.33s/it]
 73%|███████▎  | 113/155 [06:23<02:19,  3.33s/it]
 74%|███████▎  | 114/155 [06:27<02:16,  3.33s/it]
 74%|███████▍  | 115/155 [06:30<02:13,  3.33s/it]
                                                 
+
 74%|███████▍  | 115/155 [06:30<02:13,  3.33s/it]
 75%|███████▍  | 116/155 [06:33<02:09,  3.33s/it]
 75%|███████▌  | 117/155 [06:37<02:06,  3.33s/it]
 76%|███████▌  | 118/155 [06:40<02:03,  3.33s/it]
 77%|███████▋  | 119/155 [06:43<01:59,  3.33s/it]
 77%|███████▋  | 120/155 [06:47<01:56,  3.33s/it]
                                                 
+
 77%|███████▋  | 120/155 [06:47<01:56,  3.33s/it]
 78%|███████▊  | 121/155 [06:50<01:53,  3.33s/it]
 79%|███████▊  | 122/155 [06:53<01:49,  3.33s/it]
 79%|███████▉  | 123/155 [06:57<01:46,  3.32s/it]
 80%|████████  | 124/155 [07:00<01:43,  3.32s/it]
 81%|████████  | 125/155 [07:06<02:02,  4.07s/it]
                                                 
+
 81%|████████  | 125/155 [07:06<02:02,  4.07s/it]
 81%|████████▏ | 126/155 [07:09<01:51,  3.85s/it]
 82%|████████▏ | 127/155 [07:12<01:43,  3.69s/it]
 83%|████████▎ | 128/155 [07:16<01:36,  3.58s/it]
 83%|████████▎ | 129/155 [07:19<01:31,  3.50s/it]
 84%|████████▍ | 130/155 [07:22<01:26,  3.45s/it]
                                                 
+
 84%|████████▍ | 130/155 [07:22<01:26,  3.45s/it]
 85%|████████▍ | 131/155 [07:26<01:21,  3.41s/it]
 85%|████████▌ | 132/155 [07:29<01:17,  3.39s/it]
 86%|████████▌ | 133/155 [07:32<01:14,  3.37s/it]
 86%|████████▋ | 134/155 [07:36<01:10,  3.36s/it]
 87%|████████▋ | 135/155 [07:39<01:06,  3.35s/it]
                                                 
+
 87%|████████▋ | 135/155 [07:39<01:06,  3.35s/it]
 88%|████████▊ | 136/155 [07:42<01:03,  3.34s/it]
 88%|████████▊ | 137/155 [07:46<01:00,  3.33s/it]
 89%|████████▉ | 138/155 [07:49<00:56,  3.33s/it]
 90%|████████▉ | 139/155 [07:52<00:53,  3.33s/it]
 90%|█████████ | 140/155 [07:56<00:49,  3.33s/it]
                                                 
+
 90%|█████████ | 140/155 [07:56<00:49,  3.33s/it]
 91%|█████████ | 141/155 [07:59<00:46,  3.33s/it]
 92%|█████████▏| 142/155 [08:02<00:43,  3.32s/it]
 92%|█████████▏| 143/155 [08:06<00:39,  3.32s/it]
 93%|█████████▎| 144/155 [08:09<00:36,  3.32s/it]
 94%|█████████▎| 145/155 [08:12<00:33,  3.32s/it]
                                                 
+
 94%|█████████▎| 145/155 [08:12<00:33,  3.32s/it]
 94%|█████████▍| 146/155 [08:16<00:29,  3.32s/it]
 95%|█████████▍| 147/155 [08:19<00:26,  3.32s/it]
 95%|█████████▌| 148/155 [08:22<00:23,  3.32s/it]
 96%|█████████▌| 149/155 [08:26<00:19,  3.33s/it]
 97%|█████████▋| 150/155 [08:29<00:16,  3.33s/it]
                                                 
+
 97%|█████████▋| 150/155 [08:29<00:16,  3.33s/it]
 97%|█████████▋| 151/155 [08:32<00:13,  3.33s/it]
 98%|█████████▊| 152/155 [08:36<00:09,  3.33s/it]
 99%|█████████▊| 153/155 [08:39<00:06,  3.33s/it]
 99%|█████████▉| 154/155 [08:42<00:03,  3.33s/it]
                                                 
+
                                                 
+
+Time: 526.03
+Samples/second: 14.54
+GPU memory occupied: 81547 MB.