add fp32 chk
Browse files- all_results.json +14 -14
- config.json +2 -2
- eval_results.json +10 -10
- pytorch_model.bin +2 -2
- tokenizer_config.json +1 -1
- train_results.json +5 -5
- trainer_state.json +152 -341
- training_args.bin +2 -2
all_results.json
CHANGED
|
@@ -1,18 +1,18 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_gen_len": 122.
|
| 4 |
-
"eval_loss": 2.
|
| 5 |
-
"eval_rouge1": 58.
|
| 6 |
-
"eval_rouge2":
|
| 7 |
-
"eval_rougeL": 44.
|
| 8 |
-
"eval_rougeLsum": 54.
|
| 9 |
-
"eval_runtime":
|
| 10 |
"eval_samples": 1000,
|
| 11 |
-
"eval_samples_per_second": 0.
|
| 12 |
-
"eval_steps_per_second": 0.
|
| 13 |
-
"train_loss": 0.
|
| 14 |
-
"train_runtime":
|
| 15 |
"train_samples": 29441,
|
| 16 |
-
"train_samples_per_second":
|
| 17 |
-
"train_steps_per_second": 0.
|
| 18 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"eval_gen_len": 122.85,
|
| 4 |
+
"eval_loss": 2.1855268478393555,
|
| 5 |
+
"eval_rouge1": 58.2278,
|
| 6 |
+
"eval_rouge2": 28.9939,
|
| 7 |
+
"eval_rougeL": 44.6946,
|
| 8 |
+
"eval_rougeLsum": 54.4068,
|
| 9 |
+
"eval_runtime": 3779.6108,
|
| 10 |
"eval_samples": 1000,
|
| 11 |
+
"eval_samples_per_second": 0.265,
|
| 12 |
+
"eval_steps_per_second": 0.066,
|
| 13 |
+
"train_loss": 0.10453087175669877,
|
| 14 |
+
"train_runtime": 27358.0249,
|
| 15 |
"train_samples": 29441,
|
| 16 |
+
"train_samples_per_second": 1.076,
|
| 17 |
+
"train_steps_per_second": 0.004
|
| 18 |
}
|
config.json
CHANGED
|
@@ -1,5 +1,5 @@
|
|
| 1 |
{
|
| 2 |
-
"_name_or_path": "pszemraj/flan-t5-large-stacked-
|
| 3 |
"architectures": [
|
| 4 |
"T5ForConditionalGeneration"
|
| 5 |
],
|
|
@@ -25,7 +25,7 @@
|
|
| 25 |
"relative_attention_max_distance": 128,
|
| 26 |
"relative_attention_num_buckets": 32,
|
| 27 |
"tie_word_embeddings": false,
|
| 28 |
-
"torch_dtype": "
|
| 29 |
"transformers_version": "4.26.0.dev0",
|
| 30 |
"use_cache": true,
|
| 31 |
"vocab_size": 32128
|
|
|
|
| 1 |
{
|
| 2 |
+
"_name_or_path": "pszemraj/flan-t5-large-stacked-samsum-1024-FP32",
|
| 3 |
"architectures": [
|
| 4 |
"T5ForConditionalGeneration"
|
| 5 |
],
|
|
|
|
| 25 |
"relative_attention_max_distance": 128,
|
| 26 |
"relative_attention_num_buckets": 32,
|
| 27 |
"tie_word_embeddings": false,
|
| 28 |
+
"torch_dtype": "float32",
|
| 29 |
"transformers_version": "4.26.0.dev0",
|
| 30 |
"use_cache": true,
|
| 31 |
"vocab_size": 32128
|
eval_results.json
CHANGED
|
@@ -1,13 +1,13 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_gen_len": 122.
|
| 4 |
-
"eval_loss": 2.
|
| 5 |
-
"eval_rouge1": 58.
|
| 6 |
-
"eval_rouge2":
|
| 7 |
-
"eval_rougeL": 44.
|
| 8 |
-
"eval_rougeLsum": 54.
|
| 9 |
-
"eval_runtime":
|
| 10 |
"eval_samples": 1000,
|
| 11 |
-
"eval_samples_per_second": 0.
|
| 12 |
-
"eval_steps_per_second": 0.
|
| 13 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"eval_gen_len": 122.85,
|
| 4 |
+
"eval_loss": 2.1855268478393555,
|
| 5 |
+
"eval_rouge1": 58.2278,
|
| 6 |
+
"eval_rouge2": 28.9939,
|
| 7 |
+
"eval_rougeL": 44.6946,
|
| 8 |
+
"eval_rougeLsum": 54.4068,
|
| 9 |
+
"eval_runtime": 3779.6108,
|
| 10 |
"eval_samples": 1000,
|
| 11 |
+
"eval_samples_per_second": 0.265,
|
| 12 |
+
"eval_steps_per_second": 0.066
|
| 13 |
}
|
pytorch_model.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:48e1444ded77a29b2a7173d934981b6e9107abf86815db08298cd227e8c8b5f9
|
| 3 |
+
size 3132793669
|
tokenizer_config.json
CHANGED
|
@@ -104,7 +104,7 @@
|
|
| 104 |
"eos_token": "</s>",
|
| 105 |
"extra_ids": 100,
|
| 106 |
"model_max_length": 512,
|
| 107 |
-
"name_or_path": "pszemraj/flan-t5-large-stacked-
|
| 108 |
"pad_token": "<pad>",
|
| 109 |
"sp_model_kwargs": {},
|
| 110 |
"special_tokens_map_file": "/home/younes_huggingface_co/.cache/huggingface/hub/models--google--t5-v1_1-large/snapshots/314bc112b191ec17b625ba81438dc73d6c23659d/special_tokens_map.json",
|
|
|
|
| 104 |
"eos_token": "</s>",
|
| 105 |
"extra_ids": 100,
|
| 106 |
"model_max_length": 512,
|
| 107 |
+
"name_or_path": "pszemraj/flan-t5-large-stacked-samsum-1024-FP32",
|
| 108 |
"pad_token": "<pad>",
|
| 109 |
"sp_model_kwargs": {},
|
| 110 |
"special_tokens_map_file": "/home/younes_huggingface_co/.cache/huggingface/hub/models--google--t5-v1_1-large/snapshots/314bc112b191ec17b625ba81438dc73d6c23659d/special_tokens_map.json",
|
train_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"train_loss": 0.
|
| 4 |
-
"train_runtime":
|
| 5 |
"train_samples": 29441,
|
| 6 |
-
"train_samples_per_second":
|
| 7 |
-
"train_steps_per_second": 0.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"train_loss": 0.10453087175669877,
|
| 4 |
+
"train_runtime": 27358.0249,
|
| 5 |
"train_samples": 29441,
|
| 6 |
+
"train_samples_per_second": 1.076,
|
| 7 |
+
"train_steps_per_second": 0.004
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,507 +1,318 @@
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
-
"epoch":
|
| 5 |
-
"global_step":
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
| 9 |
"log_history": [
|
| 10 |
{
|
| 11 |
"epoch": 0.03,
|
| 12 |
-
"learning_rate": 0.
|
| 13 |
-
"loss": 0.
|
| 14 |
"step": 3
|
| 15 |
},
|
| 16 |
{
|
| 17 |
"epoch": 0.05,
|
| 18 |
-
"learning_rate":
|
| 19 |
-
"loss": 0.
|
| 20 |
"step": 6
|
| 21 |
},
|
| 22 |
{
|
| 23 |
"epoch": 0.08,
|
| 24 |
-
"learning_rate":
|
| 25 |
-
"loss": 0.
|
| 26 |
"step": 9
|
| 27 |
},
|
| 28 |
{
|
| 29 |
"epoch": 0.1,
|
| 30 |
-
"learning_rate":
|
| 31 |
-
"loss": 0.
|
| 32 |
"step": 12
|
| 33 |
},
|
| 34 |
{
|
| 35 |
"epoch": 0.13,
|
| 36 |
-
"learning_rate":
|
| 37 |
-
"loss": 0.
|
| 38 |
"step": 15
|
| 39 |
},
|
| 40 |
{
|
| 41 |
"epoch": 0.16,
|
| 42 |
-
"learning_rate":
|
| 43 |
-
"loss": 0.
|
| 44 |
"step": 18
|
| 45 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 46 |
{
|
| 47 |
"epoch": 0.18,
|
| 48 |
-
"learning_rate":
|
| 49 |
-
"loss": 0.
|
| 50 |
"step": 21
|
| 51 |
},
|
| 52 |
{
|
| 53 |
"epoch": 0.21,
|
| 54 |
-
"learning_rate":
|
| 55 |
-
"loss": 0.
|
| 56 |
"step": 24
|
| 57 |
},
|
| 58 |
{
|
| 59 |
"epoch": 0.23,
|
| 60 |
-
"learning_rate":
|
| 61 |
-
"loss": 0.
|
| 62 |
"step": 27
|
| 63 |
},
|
| 64 |
{
|
| 65 |
"epoch": 0.26,
|
| 66 |
-
"learning_rate":
|
| 67 |
-
"loss": 0.
|
| 68 |
"step": 30
|
| 69 |
},
|
| 70 |
{
|
| 71 |
"epoch": 0.29,
|
| 72 |
-
"learning_rate":
|
| 73 |
-
"loss": 0.
|
| 74 |
"step": 33
|
| 75 |
},
|
| 76 |
{
|
| 77 |
"epoch": 0.31,
|
| 78 |
-
"learning_rate":
|
| 79 |
-
"loss": 0.
|
| 80 |
"step": 36
|
| 81 |
},
|
| 82 |
{
|
| 83 |
"epoch": 0.34,
|
| 84 |
-
"learning_rate":
|
| 85 |
-
"loss": 0.
|
| 86 |
"step": 39
|
| 87 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 88 |
{
|
| 89 |
"epoch": 0.37,
|
| 90 |
-
"learning_rate":
|
| 91 |
-
"loss": 0.
|
| 92 |
"step": 42
|
| 93 |
},
|
| 94 |
{
|
| 95 |
"epoch": 0.39,
|
| 96 |
-
"learning_rate":
|
| 97 |
-
"loss": 0.
|
| 98 |
"step": 45
|
| 99 |
},
|
| 100 |
{
|
| 101 |
"epoch": 0.42,
|
| 102 |
-
"learning_rate":
|
| 103 |
-
"loss": 0.
|
| 104 |
"step": 48
|
| 105 |
},
|
| 106 |
{
|
| 107 |
"epoch": 0.44,
|
| 108 |
-
"learning_rate":
|
| 109 |
-
"loss": 0.
|
| 110 |
"step": 51
|
| 111 |
},
|
| 112 |
{
|
| 113 |
"epoch": 0.47,
|
| 114 |
-
"learning_rate":
|
| 115 |
-
"loss": 0.
|
| 116 |
"step": 54
|
| 117 |
},
|
| 118 |
{
|
| 119 |
"epoch": 0.5,
|
| 120 |
-
"learning_rate":
|
| 121 |
-
"loss": 0.
|
| 122 |
"step": 57
|
| 123 |
},
|
| 124 |
{
|
| 125 |
"epoch": 0.52,
|
| 126 |
-
"learning_rate":
|
| 127 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 128 |
"step": 60
|
| 129 |
},
|
| 130 |
{
|
| 131 |
"epoch": 0.55,
|
| 132 |
-
"learning_rate":
|
| 133 |
-
"loss": 0.
|
| 134 |
"step": 63
|
| 135 |
},
|
| 136 |
{
|
| 137 |
"epoch": 0.57,
|
| 138 |
-
"learning_rate":
|
| 139 |
-
"loss": 0.
|
| 140 |
"step": 66
|
| 141 |
},
|
| 142 |
{
|
| 143 |
"epoch": 0.6,
|
| 144 |
-
"learning_rate":
|
| 145 |
-
"loss": 0.
|
| 146 |
"step": 69
|
| 147 |
},
|
| 148 |
{
|
| 149 |
"epoch": 0.63,
|
| 150 |
-
"learning_rate":
|
| 151 |
-
"loss": 0.
|
| 152 |
"step": 72
|
| 153 |
},
|
| 154 |
{
|
| 155 |
"epoch": 0.65,
|
| 156 |
-
"learning_rate":
|
| 157 |
-
"loss": 0.
|
| 158 |
"step": 75
|
| 159 |
},
|
| 160 |
{
|
| 161 |
"epoch": 0.68,
|
| 162 |
-
"learning_rate":
|
| 163 |
-
"loss": 0.
|
| 164 |
"step": 78
|
| 165 |
},
|
| 166 |
{
|
| 167 |
"epoch": 0.7,
|
| 168 |
-
"
|
| 169 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 170 |
"step": 81
|
| 171 |
},
|
| 172 |
{
|
| 173 |
"epoch": 0.73,
|
| 174 |
-
"learning_rate":
|
| 175 |
-
"loss": 0.
|
| 176 |
"step": 84
|
| 177 |
},
|
| 178 |
{
|
| 179 |
"epoch": 0.76,
|
| 180 |
-
"learning_rate":
|
| 181 |
-
"loss": 0.
|
| 182 |
"step": 87
|
| 183 |
},
|
| 184 |
{
|
| 185 |
"epoch": 0.78,
|
| 186 |
-
"learning_rate":
|
| 187 |
-
"loss": 0.
|
| 188 |
"step": 90
|
| 189 |
},
|
| 190 |
{
|
| 191 |
"epoch": 0.81,
|
| 192 |
-
"learning_rate":
|
| 193 |
-
"loss": 0.
|
| 194 |
"step": 93
|
| 195 |
},
|
| 196 |
{
|
| 197 |
"epoch": 0.83,
|
| 198 |
-
"learning_rate":
|
| 199 |
-
"loss": 0.
|
| 200 |
"step": 96
|
| 201 |
},
|
| 202 |
{
|
| 203 |
"epoch": 0.86,
|
| 204 |
-
"learning_rate":
|
| 205 |
-
"loss": 0.
|
| 206 |
"step": 99
|
| 207 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 208 |
{
|
| 209 |
"epoch": 0.89,
|
| 210 |
-
"learning_rate":
|
| 211 |
-
"loss": 0.
|
| 212 |
"step": 102
|
| 213 |
},
|
| 214 |
{
|
| 215 |
"epoch": 0.91,
|
| 216 |
-
"learning_rate":
|
| 217 |
-
"loss": 0.
|
| 218 |
"step": 105
|
| 219 |
},
|
| 220 |
{
|
| 221 |
"epoch": 0.94,
|
| 222 |
-
"learning_rate":
|
| 223 |
-
"loss": 0.
|
| 224 |
"step": 108
|
| 225 |
},
|
| 226 |
{
|
| 227 |
"epoch": 0.96,
|
| 228 |
-
"learning_rate":
|
| 229 |
-
"loss": 0.
|
| 230 |
"step": 111
|
| 231 |
},
|
| 232 |
{
|
| 233 |
"epoch": 0.99,
|
| 234 |
-
"learning_rate":
|
| 235 |
-
"loss": 0.
|
| 236 |
"step": 114
|
| 237 |
},
|
| 238 |
{
|
| 239 |
"epoch": 1.0,
|
| 240 |
-
"
|
| 241 |
-
"
|
| 242 |
-
"
|
| 243 |
-
"
|
| 244 |
-
"
|
| 245 |
-
"
|
| 246 |
-
"eval_runtime": 1692.8237,
|
| 247 |
-
"eval_samples_per_second": 0.591,
|
| 248 |
-
"eval_steps_per_second": 0.148,
|
| 249 |
-
"step": 115
|
| 250 |
-
},
|
| 251 |
-
{
|
| 252 |
-
"epoch": 1.02,
|
| 253 |
-
"learning_rate": 0.00030209437808938845,
|
| 254 |
-
"loss": 0.1607,
|
| 255 |
-
"step": 117
|
| 256 |
-
},
|
| 257 |
-
{
|
| 258 |
-
"epoch": 1.04,
|
| 259 |
-
"learning_rate": 0.0002895301509892498,
|
| 260 |
-
"loss": 0.1423,
|
| 261 |
-
"step": 120
|
| 262 |
-
},
|
| 263 |
-
{
|
| 264 |
-
"epoch": 1.07,
|
| 265 |
-
"learning_rate": 0.0002769842915619544,
|
| 266 |
-
"loss": 0.1361,
|
| 267 |
-
"step": 123
|
| 268 |
-
},
|
| 269 |
-
{
|
| 270 |
-
"epoch": 1.1,
|
| 271 |
-
"learning_rate": 0.0002644788095080497,
|
| 272 |
-
"loss": 0.1438,
|
| 273 |
-
"step": 126
|
| 274 |
-
},
|
| 275 |
-
{
|
| 276 |
-
"epoch": 1.12,
|
| 277 |
-
"learning_rate": 0.00025203564369244956,
|
| 278 |
-
"loss": 0.1367,
|
| 279 |
-
"step": 129
|
| 280 |
-
},
|
| 281 |
-
{
|
| 282 |
-
"epoch": 1.15,
|
| 283 |
-
"learning_rate": 0.00023967662365621063,
|
| 284 |
-
"loss": 0.1333,
|
| 285 |
-
"step": 132
|
| 286 |
-
},
|
| 287 |
-
{
|
| 288 |
-
"epoch": 1.17,
|
| 289 |
-
"learning_rate": 0.0002274234313200997,
|
| 290 |
-
"loss": 0.1325,
|
| 291 |
-
"step": 135
|
| 292 |
-
},
|
| 293 |
-
{
|
| 294 |
-
"epoch": 1.2,
|
| 295 |
-
"learning_rate": 0.000215297562947137,
|
| 296 |
-
"loss": 0.1367,
|
| 297 |
-
"step": 138
|
| 298 |
-
},
|
| 299 |
-
{
|
| 300 |
-
"epoch": 1.23,
|
| 301 |
-
"learning_rate": 0.00020332029143084668,
|
| 302 |
-
"loss": 0.1309,
|
| 303 |
-
"step": 141
|
| 304 |
-
},
|
| 305 |
-
{
|
| 306 |
-
"epoch": 1.25,
|
| 307 |
-
"learning_rate": 0.00019151262897537235,
|
| 308 |
-
"loss": 0.1356,
|
| 309 |
-
"step": 144
|
| 310 |
-
},
|
| 311 |
-
{
|
| 312 |
-
"epoch": 1.28,
|
| 313 |
-
"learning_rate": 0.00017989529023293153,
|
| 314 |
-
"loss": 0.128,
|
| 315 |
-
"step": 147
|
| 316 |
-
},
|
| 317 |
-
{
|
| 318 |
-
"epoch": 1.3,
|
| 319 |
-
"learning_rate": 0.00016848865596327673,
|
| 320 |
-
"loss": 0.1215,
|
| 321 |
-
"step": 150
|
| 322 |
-
},
|
| 323 |
-
{
|
| 324 |
-
"epoch": 1.33,
|
| 325 |
-
"learning_rate": 0.0001573127372789174,
|
| 326 |
-
"loss": 0.1244,
|
| 327 |
-
"step": 153
|
| 328 |
-
},
|
| 329 |
-
{
|
| 330 |
-
"epoch": 1.36,
|
| 331 |
-
"learning_rate": 0.00014638714053882856,
|
| 332 |
-
"loss": 0.1265,
|
| 333 |
-
"step": 156
|
| 334 |
-
},
|
| 335 |
-
{
|
| 336 |
-
"epoch": 1.38,
|
| 337 |
-
"learning_rate": 0.00013573103295223495,
|
| 338 |
-
"loss": 0.117,
|
| 339 |
-
"step": 159
|
| 340 |
-
},
|
| 341 |
-
{
|
| 342 |
-
"epoch": 1.41,
|
| 343 |
-
"learning_rate": 0.0001253631089528132,
|
| 344 |
-
"loss": 0.1216,
|
| 345 |
-
"step": 162
|
| 346 |
-
},
|
| 347 |
-
{
|
| 348 |
-
"epoch": 1.43,
|
| 349 |
-
"learning_rate": 0.0001153015574023025,
|
| 350 |
-
"loss": 0.1157,
|
| 351 |
-
"step": 165
|
| 352 |
-
},
|
| 353 |
-
{
|
| 354 |
-
"epoch": 1.46,
|
| 355 |
-
"learning_rate": 0.00010556402968106073,
|
| 356 |
-
"loss": 0.1197,
|
| 357 |
-
"step": 168
|
| 358 |
-
},
|
| 359 |
-
{
|
| 360 |
-
"epoch": 1.49,
|
| 361 |
-
"learning_rate": 9.616760872154511e-05,
|
| 362 |
-
"loss": 0.1227,
|
| 363 |
-
"step": 171
|
| 364 |
-
},
|
| 365 |
-
{
|
| 366 |
-
"epoch": 1.51,
|
| 367 |
-
"learning_rate": 8.712877903904379e-05,
|
| 368 |
-
"loss": 0.1156,
|
| 369 |
-
"step": 174
|
| 370 |
-
},
|
| 371 |
-
{
|
| 372 |
-
"epoch": 1.54,
|
| 373 |
-
"learning_rate": 7.846339781223482e-05,
|
| 374 |
-
"loss": 0.1198,
|
| 375 |
-
"step": 177
|
| 376 |
-
},
|
| 377 |
-
{
|
| 378 |
-
"epoch": 1.57,
|
| 379 |
-
"learning_rate": 7.018666706430662e-05,
|
| 380 |
-
"loss": 0.1192,
|
| 381 |
-
"step": 180
|
| 382 |
-
},
|
| 383 |
-
{
|
| 384 |
-
"epoch": 1.59,
|
| 385 |
-
"learning_rate": 6.231310699344282e-05,
|
| 386 |
-
"loss": 0.116,
|
| 387 |
-
"step": 183
|
| 388 |
-
},
|
| 389 |
-
{
|
| 390 |
-
"epoch": 1.62,
|
| 391 |
-
"learning_rate": 5.485653049946145e-05,
|
| 392 |
-
"loss": 0.119,
|
| 393 |
-
"step": 186
|
| 394 |
-
},
|
| 395 |
-
{
|
| 396 |
-
"epoch": 1.64,
|
| 397 |
-
"learning_rate": 4.7830018951294724e-05,
|
| 398 |
-
"loss": 0.1135,
|
| 399 |
-
"step": 189
|
| 400 |
-
},
|
| 401 |
-
{
|
| 402 |
-
"epoch": 1.67,
|
| 403 |
-
"learning_rate": 4.124589923782276e-05,
|
| 404 |
-
"loss": 0.1169,
|
| 405 |
-
"step": 192
|
| 406 |
-
},
|
| 407 |
-
{
|
| 408 |
-
"epoch": 1.7,
|
| 409 |
-
"learning_rate": 3.51157221423219e-05,
|
| 410 |
-
"loss": 0.115,
|
| 411 |
-
"step": 195
|
| 412 |
-
},
|
| 413 |
-
{
|
| 414 |
-
"epoch": 1.72,
|
| 415 |
-
"learning_rate": 2.945024207846589e-05,
|
| 416 |
-
"loss": 0.1156,
|
| 417 |
-
"step": 198
|
| 418 |
-
},
|
| 419 |
-
{
|
| 420 |
-
"epoch": 1.75,
|
| 421 |
-
"learning_rate": 2.425939822342968e-05,
|
| 422 |
-
"loss": 0.1157,
|
| 423 |
-
"step": 201
|
| 424 |
-
},
|
| 425 |
-
{
|
| 426 |
-
"epoch": 1.77,
|
| 427 |
-
"learning_rate": 1.9552297081195668e-05,
|
| 428 |
-
"loss": 0.1136,
|
| 429 |
-
"step": 204
|
| 430 |
-
},
|
| 431 |
-
{
|
| 432 |
-
"epoch": 1.8,
|
| 433 |
-
"learning_rate": 1.5337196506651038e-05,
|
| 434 |
-
"loss": 0.111,
|
| 435 |
-
"step": 207
|
| 436 |
-
},
|
| 437 |
-
{
|
| 438 |
-
"epoch": 1.83,
|
| 439 |
-
"learning_rate": 1.162149121850433e-05,
|
| 440 |
-
"loss": 0.113,
|
| 441 |
-
"step": 210
|
| 442 |
-
},
|
| 443 |
-
{
|
| 444 |
-
"epoch": 1.85,
|
| 445 |
-
"learning_rate": 8.411699826436147e-06,
|
| 446 |
-
"loss": 0.1098,
|
| 447 |
-
"step": 213
|
| 448 |
-
},
|
| 449 |
-
{
|
| 450 |
-
"epoch": 1.88,
|
| 451 |
-
"learning_rate": 5.7134533952425395e-06,
|
| 452 |
-
"loss": 0.1111,
|
| 453 |
-
"step": 216
|
| 454 |
-
},
|
| 455 |
-
{
|
| 456 |
-
"epoch": 1.9,
|
| 457 |
-
"learning_rate": 3.5314855660341646e-06,
|
| 458 |
-
"loss": 0.1098,
|
| 459 |
-
"step": 219
|
| 460 |
-
},
|
| 461 |
-
{
|
| 462 |
-
"epoch": 1.93,
|
| 463 |
-
"learning_rate": 1.869624251821089e-06,
|
| 464 |
-
"loss": 0.1111,
|
| 465 |
-
"step": 222
|
| 466 |
-
},
|
| 467 |
-
{
|
| 468 |
-
"epoch": 1.96,
|
| 469 |
-
"learning_rate": 7.307849220527406e-07,
|
| 470 |
-
"loss": 0.1121,
|
| 471 |
-
"step": 225
|
| 472 |
-
},
|
| 473 |
-
{
|
| 474 |
-
"epoch": 1.98,
|
| 475 |
-
"learning_rate": 1.1696548789369431e-07,
|
| 476 |
-
"loss": 0.1098,
|
| 477 |
-
"step": 228
|
| 478 |
-
},
|
| 479 |
-
{
|
| 480 |
-
"epoch": 2.0,
|
| 481 |
-
"eval_gen_len": 122.364,
|
| 482 |
-
"eval_loss": 2.1311135292053223,
|
| 483 |
-
"eval_rouge1": 58.1114,
|
| 484 |
-
"eval_rouge2": 29.339,
|
| 485 |
-
"eval_rougeL": 44.7611,
|
| 486 |
-
"eval_rougeLsum": 54.2823,
|
| 487 |
-
"eval_runtime": 1701.9255,
|
| 488 |
-
"eval_samples_per_second": 0.588,
|
| 489 |
-
"eval_steps_per_second": 0.147,
|
| 490 |
-
"step": 230
|
| 491 |
-
},
|
| 492 |
-
{
|
| 493 |
-
"epoch": 2.0,
|
| 494 |
-
"step": 230,
|
| 495 |
-
"total_flos": 2.7141869303916134e+17,
|
| 496 |
-
"train_loss": 0.16691997828690902,
|
| 497 |
-
"train_runtime": 19759.1358,
|
| 498 |
-
"train_samples_per_second": 2.98,
|
| 499 |
-
"train_steps_per_second": 0.012
|
| 500 |
}
|
| 501 |
],
|
| 502 |
-
"max_steps":
|
| 503 |
-
"num_train_epochs":
|
| 504 |
-
"total_flos":
|
| 505 |
"trial_name": null,
|
| 506 |
"trial_params": null
|
| 507 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.9997283346916599,
|
| 5 |
+
"global_step": 115,
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
| 9 |
"log_history": [
|
| 10 |
{
|
| 11 |
"epoch": 0.03,
|
| 12 |
+
"learning_rate": 0.0001,
|
| 13 |
+
"loss": 0.1341,
|
| 14 |
"step": 3
|
| 15 |
},
|
| 16 |
{
|
| 17 |
"epoch": 0.05,
|
| 18 |
+
"learning_rate": 9.982307470588098e-05,
|
| 19 |
+
"loss": 0.1267,
|
| 20 |
"step": 6
|
| 21 |
},
|
| 22 |
{
|
| 23 |
"epoch": 0.08,
|
| 24 |
+
"learning_rate": 9.92935509259118e-05,
|
| 25 |
+
"loss": 0.1213,
|
| 26 |
"step": 9
|
| 27 |
},
|
| 28 |
{
|
| 29 |
"epoch": 0.1,
|
| 30 |
+
"learning_rate": 9.841517610611309e-05,
|
| 31 |
+
"loss": 0.1211,
|
| 32 |
"step": 12
|
| 33 |
},
|
| 34 |
{
|
| 35 |
"epoch": 0.13,
|
| 36 |
+
"learning_rate": 9.719416651541839e-05,
|
| 37 |
+
"loss": 0.1172,
|
| 38 |
"step": 15
|
| 39 |
},
|
| 40 |
{
|
| 41 |
"epoch": 0.16,
|
| 42 |
+
"learning_rate": 9.563916325306594e-05,
|
| 43 |
+
"loss": 0.1195,
|
| 44 |
"step": 18
|
| 45 |
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.17,
|
| 48 |
+
"eval_gen_len": 121.8,
|
| 49 |
+
"eval_loss": 2.063453435897827,
|
| 50 |
+
"eval_rouge1": 57.8829,
|
| 51 |
+
"eval_rouge2": 28.7887,
|
| 52 |
+
"eval_rougeL": 44.4256,
|
| 53 |
+
"eval_rougeLsum": 54.1299,
|
| 54 |
+
"eval_runtime": 3417.7231,
|
| 55 |
+
"eval_samples_per_second": 0.293,
|
| 56 |
+
"eval_steps_per_second": 0.073,
|
| 57 |
+
"step": 20
|
| 58 |
+
},
|
| 59 |
{
|
| 60 |
"epoch": 0.18,
|
| 61 |
+
"learning_rate": 9.376117109543769e-05,
|
| 62 |
+
"loss": 0.1129,
|
| 63 |
"step": 21
|
| 64 |
},
|
| 65 |
{
|
| 66 |
"epoch": 0.21,
|
| 67 |
+
"learning_rate": 9.157348061512727e-05,
|
| 68 |
+
"loss": 0.1161,
|
| 69 |
"step": 24
|
| 70 |
},
|
| 71 |
{
|
| 72 |
"epoch": 0.23,
|
| 73 |
+
"learning_rate": 8.90915741234015e-05,
|
| 74 |
+
"loss": 0.1184,
|
| 75 |
"step": 27
|
| 76 |
},
|
| 77 |
{
|
| 78 |
"epoch": 0.26,
|
| 79 |
+
"learning_rate": 8.633301610170135e-05,
|
| 80 |
+
"loss": 0.1038,
|
| 81 |
"step": 30
|
| 82 |
},
|
| 83 |
{
|
| 84 |
"epoch": 0.29,
|
| 85 |
+
"learning_rate": 8.33173288976002e-05,
|
| 86 |
+
"loss": 0.1064,
|
| 87 |
"step": 33
|
| 88 |
},
|
| 89 |
{
|
| 90 |
"epoch": 0.31,
|
| 91 |
+
"learning_rate": 8.006585456492029e-05,
|
| 92 |
+
"loss": 0.1107,
|
| 93 |
"step": 36
|
| 94 |
},
|
| 95 |
{
|
| 96 |
"epoch": 0.34,
|
| 97 |
+
"learning_rate": 7.660160382576683e-05,
|
| 98 |
+
"loss": 0.1084,
|
| 99 |
"step": 39
|
| 100 |
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 0.35,
|
| 103 |
+
"eval_gen_len": 122.893,
|
| 104 |
+
"eval_loss": 2.117846965789795,
|
| 105 |
+
"eval_rouge1": 58.0416,
|
| 106 |
+
"eval_rouge2": 28.6487,
|
| 107 |
+
"eval_rougeL": 44.3905,
|
| 108 |
+
"eval_rougeLsum": 54.1557,
|
| 109 |
+
"eval_runtime": 3584.8893,
|
| 110 |
+
"eval_samples_per_second": 0.279,
|
| 111 |
+
"eval_steps_per_second": 0.07,
|
| 112 |
+
"step": 40
|
| 113 |
+
},
|
| 114 |
{
|
| 115 |
"epoch": 0.37,
|
| 116 |
+
"learning_rate": 7.294909322337689e-05,
|
| 117 |
+
"loss": 0.1026,
|
| 118 |
"step": 42
|
| 119 |
},
|
| 120 |
{
|
| 121 |
"epoch": 0.39,
|
| 122 |
+
"learning_rate": 6.91341716182545e-05,
|
| 123 |
+
"loss": 0.1059,
|
| 124 |
"step": 45
|
| 125 |
},
|
| 126 |
{
|
| 127 |
"epoch": 0.42,
|
| 128 |
+
"learning_rate": 6.518383725548074e-05,
|
| 129 |
+
"loss": 0.1078,
|
| 130 |
"step": 48
|
| 131 |
},
|
| 132 |
{
|
| 133 |
"epoch": 0.44,
|
| 134 |
+
"learning_rate": 6.112604669781572e-05,
|
| 135 |
+
"loss": 0.1053,
|
| 136 |
"step": 51
|
| 137 |
},
|
| 138 |
{
|
| 139 |
"epoch": 0.47,
|
| 140 |
+
"learning_rate": 5.698951697677498e-05,
|
| 141 |
+
"loss": 0.1027,
|
| 142 |
"step": 54
|
| 143 |
},
|
| 144 |
{
|
| 145 |
"epoch": 0.5,
|
| 146 |
+
"learning_rate": 5.2803522361859594e-05,
|
| 147 |
+
"loss": 0.0993,
|
| 148 |
"step": 57
|
| 149 |
},
|
| 150 |
{
|
| 151 |
"epoch": 0.52,
|
| 152 |
+
"learning_rate": 4.859768718620656e-05,
|
| 153 |
+
"loss": 0.1019,
|
| 154 |
+
"step": 60
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"epoch": 0.52,
|
| 158 |
+
"eval_gen_len": 120.524,
|
| 159 |
+
"eval_loss": 2.157585382461548,
|
| 160 |
+
"eval_rouge1": 57.816,
|
| 161 |
+
"eval_rouge2": 28.7069,
|
| 162 |
+
"eval_rougeL": 44.4242,
|
| 163 |
+
"eval_rougeLsum": 53.9598,
|
| 164 |
+
"eval_runtime": 3448.4317,
|
| 165 |
+
"eval_samples_per_second": 0.29,
|
| 166 |
+
"eval_steps_per_second": 0.072,
|
| 167 |
"step": 60
|
| 168 |
},
|
| 169 |
{
|
| 170 |
"epoch": 0.55,
|
| 171 |
+
"learning_rate": 4.4401776194834613e-05,
|
| 172 |
+
"loss": 0.0959,
|
| 173 |
"step": 63
|
| 174 |
},
|
| 175 |
{
|
| 176 |
"epoch": 0.57,
|
| 177 |
+
"learning_rate": 4.0245483899193595e-05,
|
| 178 |
+
"loss": 0.1051,
|
| 179 |
"step": 66
|
| 180 |
},
|
| 181 |
{
|
| 182 |
"epoch": 0.6,
|
| 183 |
+
"learning_rate": 3.6158224428757535e-05,
|
| 184 |
+
"loss": 0.0953,
|
| 185 |
"step": 69
|
| 186 |
},
|
| 187 |
{
|
| 188 |
"epoch": 0.63,
|
| 189 |
+
"learning_rate": 3.216892336688435e-05,
|
| 190 |
+
"loss": 0.0952,
|
| 191 |
"step": 72
|
| 192 |
},
|
| 193 |
{
|
| 194 |
"epoch": 0.65,
|
| 195 |
+
"learning_rate": 2.8305813044122097e-05,
|
| 196 |
+
"loss": 0.0957,
|
| 197 |
"step": 75
|
| 198 |
},
|
| 199 |
{
|
| 200 |
"epoch": 0.68,
|
| 201 |
+
"learning_rate": 2.459623273767354e-05,
|
| 202 |
+
"loss": 0.0975,
|
| 203 |
"step": 78
|
| 204 |
},
|
| 205 |
{
|
| 206 |
"epoch": 0.7,
|
| 207 |
+
"eval_gen_len": 121.793,
|
| 208 |
+
"eval_loss": 2.182112693786621,
|
| 209 |
+
"eval_rouge1": 57.9597,
|
| 210 |
+
"eval_rouge2": 28.8178,
|
| 211 |
+
"eval_rougeL": 44.4854,
|
| 212 |
+
"eval_rougeLsum": 54.068,
|
| 213 |
+
"eval_runtime": 3498.8291,
|
| 214 |
+
"eval_samples_per_second": 0.286,
|
| 215 |
+
"eval_steps_per_second": 0.071,
|
| 216 |
+
"step": 80
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 0.7,
|
| 220 |
+
"learning_rate": 2.1066435191009715e-05,
|
| 221 |
+
"loss": 0.0943,
|
| 222 |
"step": 81
|
| 223 |
},
|
| 224 |
{
|
| 225 |
"epoch": 0.73,
|
| 226 |
+
"learning_rate": 1.774140082289563e-05,
|
| 227 |
+
"loss": 0.0988,
|
| 228 |
"step": 84
|
| 229 |
},
|
| 230 |
{
|
| 231 |
"epoch": 0.76,
|
| 232 |
+
"learning_rate": 1.4644660940672627e-05,
|
| 233 |
+
"loss": 0.0954,
|
| 234 |
"step": 87
|
| 235 |
},
|
| 236 |
{
|
| 237 |
"epoch": 0.78,
|
| 238 |
+
"learning_rate": 1.1798131208919627e-05,
|
| 239 |
+
"loss": 0.0977,
|
| 240 |
"step": 90
|
| 241 |
},
|
| 242 |
{
|
| 243 |
"epoch": 0.81,
|
| 244 |
+
"learning_rate": 9.221956552036992e-06,
|
| 245 |
+
"loss": 0.0954,
|
| 246 |
"step": 93
|
| 247 |
},
|
| 248 |
{
|
| 249 |
"epoch": 0.83,
|
| 250 |
+
"learning_rate": 6.934368588379553e-06,
|
| 251 |
+
"loss": 0.0942,
|
| 252 |
"step": 96
|
| 253 |
},
|
| 254 |
{
|
| 255 |
"epoch": 0.86,
|
| 256 |
+
"learning_rate": 4.951556604879048e-06,
|
| 257 |
+
"loss": 0.0947,
|
| 258 |
"step": 99
|
| 259 |
},
|
| 260 |
+
{
|
| 261 |
+
"epoch": 0.87,
|
| 262 |
+
"eval_gen_len": 122.77,
|
| 263 |
+
"eval_loss": 2.1846070289611816,
|
| 264 |
+
"eval_rouge1": 57.9637,
|
| 265 |
+
"eval_rouge2": 28.7446,
|
| 266 |
+
"eval_rougeL": 44.3826,
|
| 267 |
+
"eval_rougeLsum": 54.0399,
|
| 268 |
+
"eval_runtime": 3495.1079,
|
| 269 |
+
"eval_samples_per_second": 0.286,
|
| 270 |
+
"eval_steps_per_second": 0.072,
|
| 271 |
+
"step": 100
|
| 272 |
+
},
|
| 273 |
{
|
| 274 |
"epoch": 0.89,
|
| 275 |
+
"learning_rate": 3.2875529852700147e-06,
|
| 276 |
+
"loss": 0.0952,
|
| 277 |
"step": 102
|
| 278 |
},
|
| 279 |
{
|
| 280 |
"epoch": 0.91,
|
| 281 |
+
"learning_rate": 1.9541339027450256e-06,
|
| 282 |
+
"loss": 0.1001,
|
| 283 |
"step": 105
|
| 284 |
},
|
| 285 |
{
|
| 286 |
"epoch": 0.94,
|
| 287 |
+
"learning_rate": 9.607359798384785e-07,
|
| 288 |
+
"loss": 0.0953,
|
| 289 |
"step": 108
|
| 290 |
},
|
| 291 |
{
|
| 292 |
"epoch": 0.96,
|
| 293 |
+
"learning_rate": 3.143895053378698e-07,
|
| 294 |
+
"loss": 0.0963,
|
| 295 |
"step": 111
|
| 296 |
},
|
| 297 |
{
|
| 298 |
"epoch": 0.99,
|
| 299 |
+
"learning_rate": 1.9668680847356735e-08,
|
| 300 |
+
"loss": 0.0924,
|
| 301 |
"step": 114
|
| 302 |
},
|
| 303 |
{
|
| 304 |
"epoch": 1.0,
|
| 305 |
+
"step": 115,
|
| 306 |
+
"total_flos": 1.3570473694593024e+17,
|
| 307 |
+
"train_loss": 0.10453087175669877,
|
| 308 |
+
"train_runtime": 27358.0249,
|
| 309 |
+
"train_samples_per_second": 1.076,
|
| 310 |
+
"train_steps_per_second": 0.004
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 311 |
}
|
| 312 |
],
|
| 313 |
+
"max_steps": 115,
|
| 314 |
+
"num_train_epochs": 1,
|
| 315 |
+
"total_flos": 1.3570473694593024e+17,
|
| 316 |
"trial_name": null,
|
| 317 |
"trial_params": null
|
| 318 |
}
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:874b9ef1a9fa7307bd62b5629cff23971f0e3f551dc50b0975c5d755424c33f1
|
| 3 |
+
size 3707
|