Commit
•
a71e823
1
Parent(s):
b853386
Upload 8 files
Browse files- checkpoint-15000/config.json +31 -0
- checkpoint-15000/generation_config.json +7 -0
- checkpoint-15000/optimizer.pt +3 -0
- checkpoint-15000/pytorch_model.bin +3 -0
- checkpoint-15000/rng_state.pth +3 -0
- checkpoint-15000/scheduler.pt +3 -0
- checkpoint-15000/trainer_state.json +490 -0
- checkpoint-15000/training_args.bin +3 -0
checkpoint-15000/config.json
ADDED
|
@@ -0,0 +1,31 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "google/mt5-small",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"MT5ForConditionalGeneration"
|
| 5 |
+
],
|
| 6 |
+
"d_ff": 1024,
|
| 7 |
+
"d_kv": 64,
|
| 8 |
+
"d_model": 512,
|
| 9 |
+
"decoder_start_token_id": 0,
|
| 10 |
+
"dense_act_fn": "gelu_new",
|
| 11 |
+
"dropout_rate": 0.1,
|
| 12 |
+
"eos_token_id": 1,
|
| 13 |
+
"feed_forward_proj": "gated-gelu",
|
| 14 |
+
"initializer_factor": 1.0,
|
| 15 |
+
"is_encoder_decoder": true,
|
| 16 |
+
"is_gated_act": true,
|
| 17 |
+
"layer_norm_epsilon": 1e-06,
|
| 18 |
+
"model_type": "mt5",
|
| 19 |
+
"num_decoder_layers": 8,
|
| 20 |
+
"num_heads": 6,
|
| 21 |
+
"num_layers": 8,
|
| 22 |
+
"pad_token_id": 0,
|
| 23 |
+
"relative_attention_max_distance": 128,
|
| 24 |
+
"relative_attention_num_buckets": 32,
|
| 25 |
+
"tie_word_embeddings": false,
|
| 26 |
+
"tokenizer_class": "T5Tokenizer",
|
| 27 |
+
"torch_dtype": "float32",
|
| 28 |
+
"transformers_version": "4.28.0",
|
| 29 |
+
"use_cache": true,
|
| 30 |
+
"vocab_size": 250112
|
| 31 |
+
}
|
checkpoint-15000/generation_config.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"decoder_start_token_id": 0,
|
| 4 |
+
"eos_token_id": 1,
|
| 5 |
+
"pad_token_id": 0,
|
| 6 |
+
"transformers_version": "4.28.0"
|
| 7 |
+
}
|
checkpoint-15000/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d0831c45eef4cb373d82a2158ac609ab45ac5d4d1db6209eb53cfbd94637349d
|
| 3 |
+
size 2401525449
|
checkpoint-15000/pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8b543b1ab713dc4649af695e62925584fd9c95a3a1e827cd2b0cf2dd263fb5c3
|
| 3 |
+
size 1200770757
|
checkpoint-15000/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:db24715fa1d181eb24364f66201adf2589f55a4fda0da6d4a2413db2c6f3a048
|
| 3 |
+
size 14503
|
checkpoint-15000/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:79ed0ebfeed37d4fae67cc150c5ffff9a16ccedbf987c79b5e3d183dd9c1434d
|
| 3 |
+
size 623
|
checkpoint-15000/trainer_state.json
ADDED
|
@@ -0,0 +1,490 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.7971753420380452,
|
| 5 |
+
"global_step": 15000,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.01,
|
| 12 |
+
"learning_rate": 1.9957482993197282e-05,
|
| 13 |
+
"loss": 31.2602,
|
| 14 |
+
"step": 200
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.02,
|
| 18 |
+
"learning_rate": 1.991496598639456e-05,
|
| 19 |
+
"loss": 18.3425,
|
| 20 |
+
"step": 400
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.03,
|
| 24 |
+
"learning_rate": 1.987244897959184e-05,
|
| 25 |
+
"loss": 11.5784,
|
| 26 |
+
"step": 600
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.04,
|
| 30 |
+
"learning_rate": 1.982993197278912e-05,
|
| 31 |
+
"loss": 7.843,
|
| 32 |
+
"step": 800
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.05,
|
| 36 |
+
"learning_rate": 1.9787414965986397e-05,
|
| 37 |
+
"loss": 4.9763,
|
| 38 |
+
"step": 1000
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.06,
|
| 42 |
+
"learning_rate": 1.9744897959183677e-05,
|
| 43 |
+
"loss": 3.365,
|
| 44 |
+
"step": 1200
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.07,
|
| 48 |
+
"learning_rate": 1.9702380952380954e-05,
|
| 49 |
+
"loss": 2.5671,
|
| 50 |
+
"step": 1400
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.09,
|
| 54 |
+
"learning_rate": 1.965986394557823e-05,
|
| 55 |
+
"loss": 1.9747,
|
| 56 |
+
"step": 1600
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 0.1,
|
| 60 |
+
"learning_rate": 1.961734693877551e-05,
|
| 61 |
+
"loss": 1.528,
|
| 62 |
+
"step": 1800
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 0.11,
|
| 66 |
+
"learning_rate": 1.9574829931972788e-05,
|
| 67 |
+
"loss": 1.1564,
|
| 68 |
+
"step": 2000
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 0.12,
|
| 72 |
+
"learning_rate": 1.953231292517007e-05,
|
| 73 |
+
"loss": 0.8765,
|
| 74 |
+
"step": 2200
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.13,
|
| 78 |
+
"learning_rate": 1.948979591836735e-05,
|
| 79 |
+
"loss": 0.6853,
|
| 80 |
+
"step": 2400
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.14,
|
| 84 |
+
"learning_rate": 1.9447278911564626e-05,
|
| 85 |
+
"loss": 0.5731,
|
| 86 |
+
"step": 2600
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.15,
|
| 90 |
+
"learning_rate": 1.9404761904761906e-05,
|
| 91 |
+
"loss": 0.4927,
|
| 92 |
+
"step": 2800
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 0.16,
|
| 96 |
+
"learning_rate": 1.9362244897959186e-05,
|
| 97 |
+
"loss": 0.4354,
|
| 98 |
+
"step": 3000
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 0.17,
|
| 102 |
+
"learning_rate": 1.9319727891156463e-05,
|
| 103 |
+
"loss": 0.3948,
|
| 104 |
+
"step": 3200
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 0.18,
|
| 108 |
+
"learning_rate": 1.9277210884353744e-05,
|
| 109 |
+
"loss": 0.3723,
|
| 110 |
+
"step": 3400
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 0.19,
|
| 114 |
+
"learning_rate": 1.9234693877551024e-05,
|
| 115 |
+
"loss": 0.346,
|
| 116 |
+
"step": 3600
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 0.2,
|
| 120 |
+
"learning_rate": 1.91921768707483e-05,
|
| 121 |
+
"loss": 0.3295,
|
| 122 |
+
"step": 3800
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.21,
|
| 126 |
+
"learning_rate": 1.914965986394558e-05,
|
| 127 |
+
"loss": 0.3101,
|
| 128 |
+
"step": 4000
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.21,
|
| 132 |
+
"eval_loss": 0.16350167989730835,
|
| 133 |
+
"eval_runtime": 1650.684,
|
| 134 |
+
"eval_samples_per_second": 81.06,
|
| 135 |
+
"eval_steps_per_second": 20.266,
|
| 136 |
+
"step": 4000
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.22,
|
| 140 |
+
"learning_rate": 1.910714285714286e-05,
|
| 141 |
+
"loss": 0.2984,
|
| 142 |
+
"step": 4200
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.23,
|
| 146 |
+
"learning_rate": 1.906462585034014e-05,
|
| 147 |
+
"loss": 0.2811,
|
| 148 |
+
"step": 4400
|
| 149 |
+
},
|
| 150 |
+
{
|
| 151 |
+
"epoch": 0.24,
|
| 152 |
+
"learning_rate": 1.9022108843537416e-05,
|
| 153 |
+
"loss": 0.2704,
|
| 154 |
+
"step": 4600
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"epoch": 0.26,
|
| 158 |
+
"learning_rate": 1.8979591836734696e-05,
|
| 159 |
+
"loss": 0.2612,
|
| 160 |
+
"step": 4800
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"epoch": 0.27,
|
| 164 |
+
"learning_rate": 1.8937074829931973e-05,
|
| 165 |
+
"loss": 0.2441,
|
| 166 |
+
"step": 5000
|
| 167 |
+
},
|
| 168 |
+
{
|
| 169 |
+
"epoch": 0.28,
|
| 170 |
+
"learning_rate": 1.8894557823129253e-05,
|
| 171 |
+
"loss": 0.2384,
|
| 172 |
+
"step": 5200
|
| 173 |
+
},
|
| 174 |
+
{
|
| 175 |
+
"epoch": 0.29,
|
| 176 |
+
"learning_rate": 1.885204081632653e-05,
|
| 177 |
+
"loss": 0.2265,
|
| 178 |
+
"step": 5400
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.3,
|
| 182 |
+
"learning_rate": 1.880952380952381e-05,
|
| 183 |
+
"loss": 0.2226,
|
| 184 |
+
"step": 5600
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.31,
|
| 188 |
+
"learning_rate": 1.876700680272109e-05,
|
| 189 |
+
"loss": 0.2197,
|
| 190 |
+
"step": 5800
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"epoch": 0.32,
|
| 194 |
+
"learning_rate": 1.8724489795918368e-05,
|
| 195 |
+
"loss": 0.2079,
|
| 196 |
+
"step": 6000
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 0.33,
|
| 200 |
+
"learning_rate": 1.8681972789115648e-05,
|
| 201 |
+
"loss": 0.204,
|
| 202 |
+
"step": 6200
|
| 203 |
+
},
|
| 204 |
+
{
|
| 205 |
+
"epoch": 0.34,
|
| 206 |
+
"learning_rate": 1.863945578231293e-05,
|
| 207 |
+
"loss": 0.1986,
|
| 208 |
+
"step": 6400
|
| 209 |
+
},
|
| 210 |
+
{
|
| 211 |
+
"epoch": 0.35,
|
| 212 |
+
"learning_rate": 1.8596938775510206e-05,
|
| 213 |
+
"loss": 0.1965,
|
| 214 |
+
"step": 6600
|
| 215 |
+
},
|
| 216 |
+
{
|
| 217 |
+
"epoch": 0.36,
|
| 218 |
+
"learning_rate": 1.8554421768707486e-05,
|
| 219 |
+
"loss": 0.1874,
|
| 220 |
+
"step": 6800
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.37,
|
| 224 |
+
"learning_rate": 1.8511904761904763e-05,
|
| 225 |
+
"loss": 0.1889,
|
| 226 |
+
"step": 7000
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.38,
|
| 230 |
+
"learning_rate": 1.8469387755102043e-05,
|
| 231 |
+
"loss": 0.1841,
|
| 232 |
+
"step": 7200
|
| 233 |
+
},
|
| 234 |
+
{
|
| 235 |
+
"epoch": 0.39,
|
| 236 |
+
"learning_rate": 1.842687074829932e-05,
|
| 237 |
+
"loss": 0.1808,
|
| 238 |
+
"step": 7400
|
| 239 |
+
},
|
| 240 |
+
{
|
| 241 |
+
"epoch": 0.4,
|
| 242 |
+
"learning_rate": 1.83843537414966e-05,
|
| 243 |
+
"loss": 0.1778,
|
| 244 |
+
"step": 7600
|
| 245 |
+
},
|
| 246 |
+
{
|
| 247 |
+
"epoch": 0.41,
|
| 248 |
+
"learning_rate": 1.8341836734693877e-05,
|
| 249 |
+
"loss": 0.1715,
|
| 250 |
+
"step": 7800
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"epoch": 0.43,
|
| 254 |
+
"learning_rate": 1.8299319727891158e-05,
|
| 255 |
+
"loss": 0.1717,
|
| 256 |
+
"step": 8000
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 0.43,
|
| 260 |
+
"eval_loss": 0.1131439059972763,
|
| 261 |
+
"eval_runtime": 1642.8688,
|
| 262 |
+
"eval_samples_per_second": 81.446,
|
| 263 |
+
"eval_steps_per_second": 20.362,
|
| 264 |
+
"step": 8000
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 0.44,
|
| 268 |
+
"learning_rate": 1.8256802721088435e-05,
|
| 269 |
+
"loss": 0.1673,
|
| 270 |
+
"step": 8200
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 0.45,
|
| 274 |
+
"learning_rate": 1.8214285714285715e-05,
|
| 275 |
+
"loss": 0.1633,
|
| 276 |
+
"step": 8400
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.46,
|
| 280 |
+
"learning_rate": 1.8171768707482995e-05,
|
| 281 |
+
"loss": 0.1608,
|
| 282 |
+
"step": 8600
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.47,
|
| 286 |
+
"learning_rate": 1.8129251700680272e-05,
|
| 287 |
+
"loss": 0.1623,
|
| 288 |
+
"step": 8800
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 0.48,
|
| 292 |
+
"learning_rate": 1.8086734693877553e-05,
|
| 293 |
+
"loss": 0.1583,
|
| 294 |
+
"step": 9000
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 0.49,
|
| 298 |
+
"learning_rate": 1.8044217687074833e-05,
|
| 299 |
+
"loss": 0.1548,
|
| 300 |
+
"step": 9200
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"epoch": 0.5,
|
| 304 |
+
"learning_rate": 1.800170068027211e-05,
|
| 305 |
+
"loss": 0.1553,
|
| 306 |
+
"step": 9400
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"epoch": 0.51,
|
| 310 |
+
"learning_rate": 1.795918367346939e-05,
|
| 311 |
+
"loss": 0.1491,
|
| 312 |
+
"step": 9600
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"epoch": 0.52,
|
| 316 |
+
"learning_rate": 1.7916666666666667e-05,
|
| 317 |
+
"loss": 0.1529,
|
| 318 |
+
"step": 9800
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.53,
|
| 322 |
+
"learning_rate": 1.7874149659863948e-05,
|
| 323 |
+
"loss": 0.1489,
|
| 324 |
+
"step": 10000
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.54,
|
| 328 |
+
"learning_rate": 1.7831632653061225e-05,
|
| 329 |
+
"loss": 0.1454,
|
| 330 |
+
"step": 10200
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"epoch": 0.55,
|
| 334 |
+
"learning_rate": 1.7789115646258505e-05,
|
| 335 |
+
"loss": 0.1458,
|
| 336 |
+
"step": 10400
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 0.56,
|
| 340 |
+
"learning_rate": 1.7746598639455782e-05,
|
| 341 |
+
"loss": 0.1401,
|
| 342 |
+
"step": 10600
|
| 343 |
+
},
|
| 344 |
+
{
|
| 345 |
+
"epoch": 0.57,
|
| 346 |
+
"learning_rate": 1.7704081632653062e-05,
|
| 347 |
+
"loss": 0.1407,
|
| 348 |
+
"step": 10800
|
| 349 |
+
},
|
| 350 |
+
{
|
| 351 |
+
"epoch": 0.58,
|
| 352 |
+
"learning_rate": 1.7661564625850343e-05,
|
| 353 |
+
"loss": 0.1377,
|
| 354 |
+
"step": 11000
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 0.6,
|
| 358 |
+
"learning_rate": 1.761904761904762e-05,
|
| 359 |
+
"loss": 0.1378,
|
| 360 |
+
"step": 11200
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.61,
|
| 364 |
+
"learning_rate": 1.75765306122449e-05,
|
| 365 |
+
"loss": 0.1344,
|
| 366 |
+
"step": 11400
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.62,
|
| 370 |
+
"learning_rate": 1.7534013605442177e-05,
|
| 371 |
+
"loss": 0.1331,
|
| 372 |
+
"step": 11600
|
| 373 |
+
},
|
| 374 |
+
{
|
| 375 |
+
"epoch": 0.63,
|
| 376 |
+
"learning_rate": 1.7491496598639457e-05,
|
| 377 |
+
"loss": 0.1333,
|
| 378 |
+
"step": 11800
|
| 379 |
+
},
|
| 380 |
+
{
|
| 381 |
+
"epoch": 0.64,
|
| 382 |
+
"learning_rate": 1.7448979591836738e-05,
|
| 383 |
+
"loss": 0.1317,
|
| 384 |
+
"step": 12000
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 0.64,
|
| 388 |
+
"eval_loss": 0.09435752034187317,
|
| 389 |
+
"eval_runtime": 1637.6898,
|
| 390 |
+
"eval_samples_per_second": 81.704,
|
| 391 |
+
"eval_steps_per_second": 20.426,
|
| 392 |
+
"step": 12000
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 0.65,
|
| 396 |
+
"learning_rate": 1.7406462585034015e-05,
|
| 397 |
+
"loss": 0.1309,
|
| 398 |
+
"step": 12200
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 0.66,
|
| 402 |
+
"learning_rate": 1.7363945578231295e-05,
|
| 403 |
+
"loss": 0.1308,
|
| 404 |
+
"step": 12400
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 0.67,
|
| 408 |
+
"learning_rate": 1.7321428571428572e-05,
|
| 409 |
+
"loss": 0.1303,
|
| 410 |
+
"step": 12600
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 0.68,
|
| 414 |
+
"learning_rate": 1.7278911564625852e-05,
|
| 415 |
+
"loss": 0.1258,
|
| 416 |
+
"step": 12800
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.69,
|
| 420 |
+
"learning_rate": 1.723639455782313e-05,
|
| 421 |
+
"loss": 0.1261,
|
| 422 |
+
"step": 13000
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.7,
|
| 426 |
+
"learning_rate": 1.719387755102041e-05,
|
| 427 |
+
"loss": 0.1261,
|
| 428 |
+
"step": 13200
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 0.71,
|
| 432 |
+
"learning_rate": 1.7151360544217686e-05,
|
| 433 |
+
"loss": 0.1236,
|
| 434 |
+
"step": 13400
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 0.72,
|
| 438 |
+
"learning_rate": 1.7108843537414967e-05,
|
| 439 |
+
"loss": 0.1219,
|
| 440 |
+
"step": 13600
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 0.73,
|
| 444 |
+
"learning_rate": 1.7066326530612247e-05,
|
| 445 |
+
"loss": 0.1223,
|
| 446 |
+
"step": 13800
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 0.74,
|
| 450 |
+
"learning_rate": 1.7023809523809524e-05,
|
| 451 |
+
"loss": 0.1219,
|
| 452 |
+
"step": 14000
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 0.75,
|
| 456 |
+
"learning_rate": 1.6981292517006804e-05,
|
| 457 |
+
"loss": 0.1188,
|
| 458 |
+
"step": 14200
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.77,
|
| 462 |
+
"learning_rate": 1.6938775510204085e-05,
|
| 463 |
+
"loss": 0.119,
|
| 464 |
+
"step": 14400
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.78,
|
| 468 |
+
"learning_rate": 1.6896258503401362e-05,
|
| 469 |
+
"loss": 0.118,
|
| 470 |
+
"step": 14600
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 0.79,
|
| 474 |
+
"learning_rate": 1.6853741496598642e-05,
|
| 475 |
+
"loss": 0.118,
|
| 476 |
+
"step": 14800
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 0.8,
|
| 480 |
+
"learning_rate": 1.681122448979592e-05,
|
| 481 |
+
"loss": 0.1156,
|
| 482 |
+
"step": 15000
|
| 483 |
+
}
|
| 484 |
+
],
|
| 485 |
+
"max_steps": 94080,
|
| 486 |
+
"num_train_epochs": 5,
|
| 487 |
+
"total_flos": 5.075997032448e+17,
|
| 488 |
+
"trial_name": null,
|
| 489 |
+
"trial_params": null
|
| 490 |
+
}
|
checkpoint-15000/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7c7a14b68a242d9adb4a5aed9daebd4e60b09b655362ead363872702c8708f4d
|
| 3 |
+
size 3695
|