Commit
•
a9a6c32
1
Parent(s):
c7cca67
Model save
Browse files- README.md +4 -4
- all_results.json +6 -6
- generation_config.json +3 -2
- train_results.json +6 -6
- trainer_state.json +329 -434
README.md
CHANGED
|
@@ -20,7 +20,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 20 |
|
| 21 |
This model is a fine-tuned version of [mistralai/Mistral-7B-v0.3](https://huggingface.co/mistralai/Mistral-7B-v0.3) on the generator dataset.
|
| 22 |
It achieves the following results on the evaluation set:
|
| 23 |
-
- Loss:
|
| 24 |
|
| 25 |
## Model description
|
| 26 |
|
|
@@ -55,9 +55,9 @@ The following hyperparameters were used during training:
|
|
| 55 |
|
| 56 |
### Training results
|
| 57 |
|
| 58 |
-
| Training Loss | Epoch
|
| 59 |
-
|
| 60 |
-
|
|
| 61 |
|
| 62 |
|
| 63 |
### Framework versions
|
|
|
|
| 20 |
|
| 21 |
This model is a fine-tuned version of [mistralai/Mistral-7B-v0.3](https://huggingface.co/mistralai/Mistral-7B-v0.3) on the generator dataset.
|
| 22 |
It achieves the following results on the evaluation set:
|
| 23 |
+
- Loss: 7.8271
|
| 24 |
|
| 25 |
## Model description
|
| 26 |
|
|
|
|
| 55 |
|
| 56 |
### Training results
|
| 57 |
|
| 58 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 59 |
+
|:-------------:|:------:|:----:|:---------------:|
|
| 60 |
+
| 7.9839 | 0.9987 | 386 | 7.8271 |
|
| 61 |
|
| 62 |
|
| 63 |
### Framework versions
|
all_results.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"total_flos":
|
| 4 |
-
"train_loss":
|
| 5 |
-
"train_runtime":
|
| 6 |
"train_samples": 77486,
|
| 7 |
-
"train_samples_per_second":
|
| 8 |
-
"train_steps_per_second": 0.
|
| 9 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 0.9987063389391979,
|
| 3 |
+
"total_flos": 161536404357120.0,
|
| 4 |
+
"train_loss": 11.514628536342958,
|
| 5 |
+
"train_runtime": 5043.2067,
|
| 6 |
"train_samples": 77486,
|
| 7 |
+
"train_samples_per_second": 19.608,
|
| 8 |
+
"train_steps_per_second": 0.077
|
| 9 |
}
|
generation_config.json
CHANGED
|
@@ -1,6 +1,7 @@
|
|
| 1 |
{
|
| 2 |
"_from_model_config": true,
|
| 3 |
-
"bos_token_id":
|
| 4 |
-
"eos_token_id":
|
|
|
|
| 5 |
"transformers_version": "4.44.1"
|
| 6 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 100263,
|
| 4 |
+
"eos_token_id": 100264,
|
| 5 |
+
"pad_token_id": 100264,
|
| 6 |
"transformers_version": "4.44.1"
|
| 7 |
}
|
train_results.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"total_flos":
|
| 4 |
-
"train_loss":
|
| 5 |
-
"train_runtime":
|
| 6 |
"train_samples": 77486,
|
| 7 |
-
"train_samples_per_second":
|
| 8 |
-
"train_steps_per_second": 0.
|
| 9 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 0.9987063389391979,
|
| 3 |
+
"total_flos": 161536404357120.0,
|
| 4 |
+
"train_loss": 11.514628536342958,
|
| 5 |
+
"train_runtime": 5043.2067,
|
| 6 |
"train_samples": 77486,
|
| 7 |
+
"train_samples_per_second": 19.608,
|
| 8 |
+
"train_steps_per_second": 0.077
|
| 9 |
}
|
trainer_state.json
CHANGED
|
@@ -1,684 +1,579 @@
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
-
"epoch":
|
| 5 |
"eval_steps": 500,
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
-
"epoch": 0.
|
| 13 |
-
"grad_norm":
|
| 14 |
-
"learning_rate": 1.
|
| 15 |
-
"loss":
|
| 16 |
"step": 1
|
| 17 |
},
|
| 18 |
{
|
| 19 |
-
"epoch": 0.
|
| 20 |
-
"grad_norm":
|
| 21 |
-
"learning_rate":
|
| 22 |
-
"loss":
|
| 23 |
"step": 5
|
| 24 |
},
|
| 25 |
{
|
| 26 |
-
"epoch": 0.
|
| 27 |
-
"grad_norm":
|
| 28 |
-
"learning_rate": 1.
|
| 29 |
-
"loss":
|
| 30 |
"step": 10
|
| 31 |
},
|
| 32 |
{
|
| 33 |
-
"epoch": 0.
|
| 34 |
-
"grad_norm":
|
| 35 |
-
"learning_rate": 1.
|
| 36 |
-
"loss":
|
| 37 |
"step": 15
|
| 38 |
},
|
| 39 |
{
|
| 40 |
-
"epoch": 0.
|
| 41 |
-
"grad_norm":
|
| 42 |
-
"learning_rate": 2.
|
| 43 |
-
"loss":
|
| 44 |
"step": 20
|
| 45 |
},
|
| 46 |
{
|
| 47 |
-
"epoch": 0.
|
| 48 |
-
"grad_norm":
|
| 49 |
-
"learning_rate":
|
| 50 |
-
"loss":
|
| 51 |
"step": 25
|
| 52 |
},
|
| 53 |
{
|
| 54 |
-
"epoch": 0.
|
| 55 |
-
"grad_norm":
|
| 56 |
-
"learning_rate": 3.
|
| 57 |
-
"loss":
|
| 58 |
"step": 30
|
| 59 |
},
|
| 60 |
{
|
| 61 |
-
"epoch": 0.
|
| 62 |
-
"grad_norm":
|
| 63 |
-
"learning_rate":
|
| 64 |
-
"loss":
|
| 65 |
"step": 35
|
| 66 |
},
|
| 67 |
{
|
| 68 |
-
"epoch": 0.
|
| 69 |
-
"grad_norm":
|
| 70 |
-
"learning_rate": 4.
|
| 71 |
-
"loss":
|
| 72 |
"step": 40
|
| 73 |
},
|
| 74 |
{
|
| 75 |
-
"epoch": 0.
|
| 76 |
-
"grad_norm":
|
| 77 |
-
"learning_rate": 4.
|
| 78 |
-
"loss":
|
| 79 |
"step": 45
|
| 80 |
},
|
| 81 |
{
|
| 82 |
-
"epoch": 0.
|
| 83 |
-
"grad_norm":
|
| 84 |
-
"learning_rate": 4.
|
| 85 |
-
"loss":
|
| 86 |
"step": 50
|
| 87 |
},
|
| 88 |
{
|
| 89 |
-
"epoch": 0.
|
| 90 |
-
"grad_norm":
|
| 91 |
-
"learning_rate": 4.
|
| 92 |
-
"loss":
|
| 93 |
"step": 55
|
| 94 |
},
|
| 95 |
{
|
| 96 |
-
"epoch": 0.
|
| 97 |
-
"grad_norm":
|
| 98 |
-
"learning_rate": 4.
|
| 99 |
-
"loss":
|
| 100 |
"step": 60
|
| 101 |
},
|
| 102 |
{
|
| 103 |
-
"epoch": 0.
|
| 104 |
-
"grad_norm":
|
| 105 |
-
"learning_rate": 4.
|
| 106 |
-
"loss":
|
| 107 |
"step": 65
|
| 108 |
},
|
| 109 |
{
|
| 110 |
-
"epoch": 0.
|
| 111 |
-
"grad_norm":
|
| 112 |
-
"learning_rate": 4.
|
| 113 |
-
"loss":
|
| 114 |
"step": 70
|
| 115 |
},
|
| 116 |
{
|
| 117 |
-
"epoch": 0.
|
| 118 |
-
"grad_norm":
|
| 119 |
-
"learning_rate": 4.
|
| 120 |
-
"loss":
|
| 121 |
"step": 75
|
| 122 |
},
|
| 123 |
{
|
| 124 |
-
"epoch": 0.
|
| 125 |
-
"grad_norm":
|
| 126 |
-
"learning_rate": 4.
|
| 127 |
-
"loss":
|
| 128 |
"step": 80
|
| 129 |
},
|
| 130 |
{
|
| 131 |
-
"epoch": 0.
|
| 132 |
-
"grad_norm":
|
| 133 |
-
"learning_rate": 4.
|
| 134 |
-
"loss":
|
| 135 |
"step": 85
|
| 136 |
},
|
| 137 |
{
|
| 138 |
-
"epoch": 0.
|
| 139 |
-
"grad_norm":
|
| 140 |
-
"learning_rate": 4.
|
| 141 |
-
"loss":
|
| 142 |
"step": 90
|
| 143 |
},
|
| 144 |
{
|
| 145 |
-
"epoch": 0.
|
| 146 |
-
"grad_norm":
|
| 147 |
-
"learning_rate": 4.
|
| 148 |
-
"loss":
|
| 149 |
"step": 95
|
| 150 |
},
|
| 151 |
{
|
| 152 |
-
"epoch": 0.
|
| 153 |
-
"grad_norm":
|
| 154 |
-
"learning_rate": 4.
|
| 155 |
-
"loss":
|
| 156 |
"step": 100
|
| 157 |
},
|
| 158 |
{
|
| 159 |
-
"epoch": 0.
|
| 160 |
-
"grad_norm":
|
| 161 |
-
"learning_rate": 4.
|
| 162 |
-
"loss":
|
| 163 |
"step": 105
|
| 164 |
},
|
| 165 |
{
|
| 166 |
-
"epoch": 0.
|
| 167 |
-
"grad_norm":
|
| 168 |
-
"learning_rate": 4.
|
| 169 |
-
"loss":
|
| 170 |
"step": 110
|
| 171 |
},
|
| 172 |
{
|
| 173 |
-
"epoch": 0.
|
| 174 |
-
"grad_norm":
|
| 175 |
-
"learning_rate": 4.
|
| 176 |
-
"loss":
|
| 177 |
"step": 115
|
| 178 |
},
|
| 179 |
{
|
| 180 |
-
"epoch": 0.
|
| 181 |
-
"grad_norm":
|
| 182 |
-
"learning_rate": 4.
|
| 183 |
-
"loss":
|
| 184 |
"step": 120
|
| 185 |
},
|
| 186 |
{
|
| 187 |
-
"epoch": 0.
|
| 188 |
-
"grad_norm":
|
| 189 |
-
"learning_rate": 4.
|
| 190 |
-
"loss":
|
| 191 |
"step": 125
|
| 192 |
},
|
| 193 |
{
|
| 194 |
-
"epoch": 0.
|
| 195 |
-
"grad_norm":
|
| 196 |
-
"learning_rate": 4.
|
| 197 |
-
"loss":
|
| 198 |
"step": 130
|
| 199 |
},
|
| 200 |
{
|
| 201 |
-
"epoch": 0.
|
| 202 |
-
"grad_norm":
|
| 203 |
-
"learning_rate": 4.
|
| 204 |
-
"loss":
|
| 205 |
"step": 135
|
| 206 |
},
|
| 207 |
{
|
| 208 |
-
"epoch": 0.
|
| 209 |
-
"grad_norm":
|
| 210 |
-
"learning_rate": 4.
|
| 211 |
-
"loss":
|
| 212 |
"step": 140
|
| 213 |
},
|
| 214 |
{
|
| 215 |
-
"epoch": 0.
|
| 216 |
-
"grad_norm":
|
| 217 |
-
"learning_rate":
|
| 218 |
-
"loss":
|
| 219 |
"step": 145
|
| 220 |
},
|
| 221 |
{
|
| 222 |
-
"epoch": 0.
|
| 223 |
-
"grad_norm":
|
| 224 |
-
"learning_rate":
|
| 225 |
-
"loss":
|
| 226 |
"step": 150
|
| 227 |
},
|
| 228 |
{
|
| 229 |
-
"epoch": 0.
|
| 230 |
-
"grad_norm":
|
| 231 |
-
"learning_rate":
|
| 232 |
-
"loss":
|
| 233 |
"step": 155
|
| 234 |
},
|
| 235 |
{
|
| 236 |
-
"epoch": 0.
|
| 237 |
-
"grad_norm":
|
| 238 |
-
"learning_rate":
|
| 239 |
-
"loss":
|
| 240 |
"step": 160
|
| 241 |
},
|
| 242 |
{
|
| 243 |
-
"epoch": 0.
|
| 244 |
-
"grad_norm":
|
| 245 |
-
"learning_rate":
|
| 246 |
-
"loss":
|
| 247 |
"step": 165
|
| 248 |
},
|
| 249 |
{
|
| 250 |
-
"epoch": 0.
|
| 251 |
-
"grad_norm":
|
| 252 |
-
"learning_rate": 3.
|
| 253 |
-
"loss":
|
| 254 |
"step": 170
|
| 255 |
},
|
| 256 |
{
|
| 257 |
-
"epoch": 0.
|
| 258 |
-
"grad_norm":
|
| 259 |
-
"learning_rate": 3.
|
| 260 |
-
"loss":
|
| 261 |
"step": 175
|
| 262 |
},
|
| 263 |
{
|
| 264 |
-
"epoch": 0.
|
| 265 |
-
"grad_norm":
|
| 266 |
-
"learning_rate": 3.
|
| 267 |
-
"loss":
|
| 268 |
"step": 180
|
| 269 |
},
|
| 270 |
{
|
| 271 |
-
"epoch": 0.
|
| 272 |
-
"grad_norm":
|
| 273 |
-
"learning_rate": 3.
|
| 274 |
-
"loss":
|
| 275 |
"step": 185
|
| 276 |
},
|
| 277 |
{
|
| 278 |
-
"epoch": 0.
|
| 279 |
-
"grad_norm":
|
| 280 |
-
"learning_rate": 3.
|
| 281 |
-
"loss":
|
| 282 |
"step": 190
|
| 283 |
},
|
| 284 |
{
|
| 285 |
-
"epoch": 0.
|
| 286 |
-
"grad_norm":
|
| 287 |
-
"learning_rate":
|
| 288 |
-
"loss":
|
| 289 |
"step": 195
|
| 290 |
},
|
| 291 |
{
|
| 292 |
-
"epoch": 0.
|
| 293 |
-
"grad_norm":
|
| 294 |
-
"learning_rate":
|
| 295 |
-
"loss":
|
| 296 |
"step": 200
|
| 297 |
},
|
| 298 |
{
|
| 299 |
-
"epoch": 0.
|
| 300 |
-
"grad_norm":
|
| 301 |
-
"learning_rate":
|
| 302 |
-
"loss":
|
| 303 |
"step": 205
|
| 304 |
},
|
| 305 |
{
|
| 306 |
-
"epoch": 0.
|
| 307 |
-
"grad_norm":
|
| 308 |
-
"learning_rate":
|
| 309 |
-
"loss":
|
| 310 |
"step": 210
|
| 311 |
},
|
| 312 |
{
|
| 313 |
-
"epoch": 0.
|
| 314 |
-
"grad_norm":
|
| 315 |
-
"learning_rate":
|
| 316 |
-
"loss":
|
| 317 |
"step": 215
|
| 318 |
},
|
| 319 |
{
|
| 320 |
-
"epoch": 0.
|
| 321 |
-
"grad_norm":
|
| 322 |
-
"learning_rate":
|
| 323 |
-
"loss":
|
| 324 |
"step": 220
|
| 325 |
},
|
| 326 |
{
|
| 327 |
-
"epoch": 0.
|
| 328 |
-
"grad_norm":
|
| 329 |
-
"learning_rate":
|
| 330 |
-
"loss":
|
| 331 |
"step": 225
|
| 332 |
},
|
| 333 |
{
|
| 334 |
-
"epoch": 0.
|
| 335 |
-
"grad_norm":
|
| 336 |
-
"learning_rate": 2.
|
| 337 |
-
"loss":
|
| 338 |
"step": 230
|
| 339 |
},
|
| 340 |
{
|
| 341 |
-
"epoch": 0.
|
| 342 |
-
"grad_norm":
|
| 343 |
-
"learning_rate":
|
| 344 |
-
"loss":
|
| 345 |
"step": 235
|
| 346 |
},
|
| 347 |
{
|
| 348 |
-
"epoch": 0.
|
| 349 |
-
"grad_norm":
|
| 350 |
-
"learning_rate":
|
| 351 |
-
"loss":
|
| 352 |
"step": 240
|
| 353 |
},
|
| 354 |
{
|
| 355 |
-
"epoch": 0.
|
| 356 |
-
"grad_norm":
|
| 357 |
-
"learning_rate":
|
| 358 |
-
"loss":
|
| 359 |
"step": 245
|
| 360 |
},
|
| 361 |
{
|
| 362 |
-
"epoch": 0.
|
| 363 |
-
"grad_norm":
|
| 364 |
-
"learning_rate":
|
| 365 |
-
"loss":
|
| 366 |
"step": 250
|
| 367 |
},
|
| 368 |
{
|
| 369 |
-
"epoch": 0.
|
| 370 |
-
"grad_norm":
|
| 371 |
-
"learning_rate":
|
| 372 |
-
"loss":
|
| 373 |
"step": 255
|
| 374 |
},
|
| 375 |
{
|
| 376 |
-
"epoch": 0.
|
| 377 |
-
"grad_norm":
|
| 378 |
-
"learning_rate":
|
| 379 |
-
"loss":
|
| 380 |
"step": 260
|
| 381 |
},
|
| 382 |
{
|
| 383 |
-
"epoch": 0.
|
| 384 |
-
"grad_norm":
|
| 385 |
-
"learning_rate":
|
| 386 |
-
"loss":
|
| 387 |
"step": 265
|
| 388 |
},
|
| 389 |
{
|
| 390 |
-
"epoch": 0.
|
| 391 |
-
"grad_norm":
|
| 392 |
-
"learning_rate":
|
| 393 |
-
"loss":
|
| 394 |
"step": 270
|
| 395 |
},
|
| 396 |
{
|
| 397 |
-
"epoch": 0.
|
| 398 |
-
"grad_norm":
|
| 399 |
-
"learning_rate":
|
| 400 |
-
"loss":
|
| 401 |
"step": 275
|
| 402 |
},
|
| 403 |
{
|
| 404 |
-
"epoch": 0.
|
| 405 |
-
"grad_norm":
|
| 406 |
-
"learning_rate":
|
| 407 |
-
"loss":
|
| 408 |
"step": 280
|
| 409 |
},
|
| 410 |
{
|
| 411 |
-
"epoch": 0.
|
| 412 |
-
"grad_norm":
|
| 413 |
-
"learning_rate":
|
| 414 |
-
"loss":
|
| 415 |
"step": 285
|
| 416 |
},
|
| 417 |
{
|
| 418 |
-
"epoch": 0.
|
| 419 |
-
"grad_norm":
|
| 420 |
-
"learning_rate":
|
| 421 |
-
"loss":
|
| 422 |
"step": 290
|
| 423 |
},
|
| 424 |
{
|
| 425 |
-
"epoch": 0.
|
| 426 |
-
"grad_norm":
|
| 427 |
-
"learning_rate":
|
| 428 |
-
"loss":
|
| 429 |
"step": 295
|
| 430 |
},
|
| 431 |
{
|
| 432 |
-
"epoch": 0.
|
| 433 |
-
"grad_norm":
|
| 434 |
-
"learning_rate":
|
| 435 |
-
"loss":
|
| 436 |
"step": 300
|
| 437 |
},
|
| 438 |
{
|
| 439 |
-
"epoch": 0.
|
| 440 |
-
"grad_norm":
|
| 441 |
-
"learning_rate":
|
| 442 |
-
"loss":
|
| 443 |
"step": 305
|
| 444 |
},
|
| 445 |
{
|
| 446 |
-
"epoch": 0.
|
| 447 |
-
"grad_norm":
|
| 448 |
-
"learning_rate":
|
| 449 |
-
"loss":
|
| 450 |
"step": 310
|
| 451 |
},
|
| 452 |
{
|
| 453 |
-
"epoch": 0.
|
| 454 |
-
"grad_norm":
|
| 455 |
-
"learning_rate":
|
| 456 |
-
"loss":
|
| 457 |
"step": 315
|
| 458 |
},
|
| 459 |
{
|
| 460 |
-
"epoch": 0.
|
| 461 |
-
"grad_norm":
|
| 462 |
-
"learning_rate":
|
| 463 |
-
"loss":
|
| 464 |
"step": 320
|
| 465 |
},
|
| 466 |
{
|
| 467 |
-
"epoch": 0.
|
| 468 |
-
"grad_norm":
|
| 469 |
-
"learning_rate":
|
| 470 |
-
"loss":
|
| 471 |
"step": 325
|
| 472 |
},
|
| 473 |
{
|
| 474 |
-
"epoch": 0.
|
| 475 |
-
"grad_norm":
|
| 476 |
-
"learning_rate":
|
| 477 |
-
"loss":
|
| 478 |
"step": 330
|
| 479 |
},
|
| 480 |
{
|
| 481 |
-
"epoch": 0.
|
| 482 |
-
"grad_norm":
|
| 483 |
-
"learning_rate":
|
| 484 |
-
"loss":
|
| 485 |
"step": 335
|
| 486 |
},
|
| 487 |
{
|
| 488 |
-
"epoch": 0.
|
| 489 |
-
"grad_norm":
|
| 490 |
-
"learning_rate":
|
| 491 |
-
"loss": 0
|
| 492 |
"step": 340
|
| 493 |
},
|
| 494 |
{
|
| 495 |
-
"epoch": 0.
|
| 496 |
-
"grad_norm":
|
| 497 |
-
"learning_rate":
|
| 498 |
-
"loss":
|
| 499 |
"step": 345
|
| 500 |
},
|
| 501 |
{
|
| 502 |
-
"epoch": 0.
|
| 503 |
-
"grad_norm":
|
| 504 |
-
"learning_rate":
|
| 505 |
-
"loss":
|
| 506 |
"step": 350
|
| 507 |
},
|
| 508 |
{
|
| 509 |
-
"epoch": 0.
|
| 510 |
-
"grad_norm":
|
| 511 |
-
"learning_rate":
|
| 512 |
-
"loss":
|
| 513 |
"step": 355
|
| 514 |
},
|
| 515 |
{
|
| 516 |
-
"epoch": 0.
|
| 517 |
-
"grad_norm":
|
| 518 |
-
"learning_rate": 6.
|
| 519 |
-
"loss":
|
| 520 |
"step": 360
|
| 521 |
},
|
| 522 |
{
|
| 523 |
-
"epoch": 0.
|
| 524 |
-
"grad_norm":
|
| 525 |
-
"learning_rate":
|
| 526 |
-
"loss":
|
| 527 |
"step": 365
|
| 528 |
},
|
| 529 |
{
|
| 530 |
-
"epoch": 0.
|
| 531 |
-
"grad_norm":
|
| 532 |
-
"learning_rate":
|
| 533 |
-
"loss":
|
| 534 |
"step": 370
|
| 535 |
},
|
| 536 |
{
|
| 537 |
-
"epoch": 0.
|
| 538 |
-
"grad_norm":
|
| 539 |
-
"learning_rate":
|
| 540 |
-
"loss":
|
| 541 |
"step": 375
|
| 542 |
},
|
| 543 |
{
|
| 544 |
-
"epoch": 0.
|
| 545 |
-
"grad_norm":
|
| 546 |
-
"learning_rate":
|
| 547 |
-
"loss":
|
| 548 |
"step": 380
|
| 549 |
},
|
| 550 |
{
|
| 551 |
-
"epoch": 0.
|
| 552 |
-
"grad_norm":
|
| 553 |
-
"learning_rate":
|
| 554 |
-
"loss":
|
| 555 |
"step": 385
|
| 556 |
},
|
| 557 |
{
|
| 558 |
-
"epoch": 0.
|
| 559 |
-
"
|
| 560 |
-
"
|
| 561 |
-
"
|
| 562 |
-
"
|
|
|
|
| 563 |
},
|
| 564 |
{
|
| 565 |
-
"epoch": 0.
|
| 566 |
-
"
|
| 567 |
-
"
|
| 568 |
-
"
|
| 569 |
-
"
|
| 570 |
-
|
| 571 |
-
|
| 572 |
-
"epoch": 0.8676789587852495,
|
| 573 |
-
"grad_norm": 2.1291510379356753,
|
| 574 |
-
"learning_rate": 2.6308749729643058e-08,
|
| 575 |
-
"loss": 0.9651,
|
| 576 |
-
"step": 400
|
| 577 |
-
},
|
| 578 |
-
{
|
| 579 |
-
"epoch": 0.8785249457700651,
|
| 580 |
-
"grad_norm": 2.1035620640567734,
|
| 581 |
-
"learning_rate": 2.2235135302712092e-08,
|
| 582 |
-
"loss": 0.9952,
|
| 583 |
-
"step": 405
|
| 584 |
-
},
|
| 585 |
-
{
|
| 586 |
-
"epoch": 0.8893709327548807,
|
| 587 |
-
"grad_norm": 2.0424482935650614,
|
| 588 |
-
"learning_rate": 1.8489369753048682e-08,
|
| 589 |
-
"loss": 0.9631,
|
| 590 |
-
"step": 410
|
| 591 |
-
},
|
| 592 |
-
{
|
| 593 |
-
"epoch": 0.9002169197396963,
|
| 594 |
-
"grad_norm": 2.0939056533766847,
|
| 595 |
-
"learning_rate": 1.507684480352292e-08,
|
| 596 |
-
"loss": 0.9747,
|
| 597 |
-
"step": 415
|
| 598 |
-
},
|
| 599 |
-
{
|
| 600 |
-
"epoch": 0.911062906724512,
|
| 601 |
-
"grad_norm": 4.429046210547233,
|
| 602 |
-
"learning_rate": 1.2002472504440807e-08,
|
| 603 |
-
"loss": 0.9843,
|
| 604 |
-
"step": 420
|
| 605 |
-
},
|
| 606 |
-
{
|
| 607 |
-
"epoch": 0.9219088937093276,
|
| 608 |
-
"grad_norm": 2.1047314550346297,
|
| 609 |
-
"learning_rate": 9.270678163050217e-09,
|
| 610 |
-
"loss": 0.9738,
|
| 611 |
-
"step": 425
|
| 612 |
-
},
|
| 613 |
-
{
|
| 614 |
-
"epoch": 0.9327548806941431,
|
| 615 |
-
"grad_norm": 2.0998363541587888,
|
| 616 |
-
"learning_rate": 6.885393973673298e-09,
|
| 617 |
-
"loss": 0.9509,
|
| 618 |
-
"step": 430
|
| 619 |
-
},
|
| 620 |
-
{
|
| 621 |
-
"epoch": 0.9436008676789588,
|
| 622 |
-
"grad_norm": 2.138035512894866,
|
| 623 |
-
"learning_rate": 4.850053357634693e-09,
|
| 624 |
-
"loss": 0.9799,
|
| 625 |
-
"step": 435
|
| 626 |
-
},
|
| 627 |
-
{
|
| 628 |
-
"epoch": 0.9544468546637744,
|
| 629 |
-
"grad_norm": 2.053994471901985,
|
| 630 |
-
"learning_rate": 3.1675860211325954e-09,
|
| 631 |
-
"loss": 0.9762,
|
| 632 |
-
"step": 440
|
| 633 |
-
},
|
| 634 |
-
{
|
| 635 |
-
"epoch": 0.96529284164859,
|
| 636 |
-
"grad_norm": 2.054984790757136,
|
| 637 |
-
"learning_rate": 1.840413738166402e-09,
|
| 638 |
-
"loss": 0.9499,
|
| 639 |
-
"step": 445
|
| 640 |
-
},
|
| 641 |
-
{
|
| 642 |
-
"epoch": 0.9761388286334056,
|
| 643 |
-
"grad_norm": 2.0710147404144097,
|
| 644 |
-
"learning_rate": 8.704468645914787e-10,
|
| 645 |
-
"loss": 0.9692,
|
| 646 |
-
"step": 450
|
| 647 |
-
},
|
| 648 |
-
{
|
| 649 |
-
"epoch": 0.9869848156182213,
|
| 650 |
-
"grad_norm": 2.067619347023652,
|
| 651 |
-
"learning_rate": 2.5908158831811077e-10,
|
| 652 |
-
"loss": 0.9994,
|
| 653 |
-
"step": 455
|
| 654 |
-
},
|
| 655 |
-
{
|
| 656 |
-
"epoch": 0.9978308026030369,
|
| 657 |
-
"grad_norm": 2.095822455094207,
|
| 658 |
-
"learning_rate": 7.197919613455284e-12,
|
| 659 |
-
"loss": 0.9524,
|
| 660 |
-
"step": 460
|
| 661 |
-
},
|
| 662 |
-
{
|
| 663 |
-
"epoch": 1.0,
|
| 664 |
-
"eval_loss": 1.0288481712341309,
|
| 665 |
-
"eval_runtime": 3.4782,
|
| 666 |
-
"eval_samples_per_second": 74.465,
|
| 667 |
-
"eval_steps_per_second": 1.438,
|
| 668 |
-
"step": 461
|
| 669 |
-
},
|
| 670 |
-
{
|
| 671 |
-
"epoch": 1.0,
|
| 672 |
-
"step": 461,
|
| 673 |
-
"total_flos": 192943352709120.0,
|
| 674 |
-
"train_loss": 1.0081440616326323,
|
| 675 |
-
"train_runtime": 5423.7345,
|
| 676 |
-
"train_samples_per_second": 21.752,
|
| 677 |
-
"train_steps_per_second": 0.085
|
| 678 |
}
|
| 679 |
],
|
| 680 |
"logging_steps": 5,
|
| 681 |
-
"max_steps":
|
| 682 |
"num_input_tokens_seen": 0,
|
| 683 |
"num_train_epochs": 1,
|
| 684 |
"save_steps": 100,
|
|
@@ -694,7 +589,7 @@
|
|
| 694 |
"attributes": {}
|
| 695 |
}
|
| 696 |
},
|
| 697 |
-
"total_flos":
|
| 698 |
"train_batch_size": 8,
|
| 699 |
"trial_name": null,
|
| 700 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.9987063389391979,
|
| 5 |
"eval_steps": 500,
|
| 6 |
+
"global_step": 386,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
+
"epoch": 0.00258732212160414,
|
| 13 |
+
"grad_norm": 1.5696258219863308e+34,
|
| 14 |
+
"learning_rate": 1.282051282051282e-08,
|
| 15 |
+
"loss": 34.5991,
|
| 16 |
"step": 1
|
| 17 |
},
|
| 18 |
{
|
| 19 |
+
"epoch": 0.0129366106080207,
|
| 20 |
+
"grad_norm": 2.4800051592297763e+32,
|
| 21 |
+
"learning_rate": 6.410256410256409e-08,
|
| 22 |
+
"loss": 35.1328,
|
| 23 |
"step": 5
|
| 24 |
},
|
| 25 |
{
|
| 26 |
+
"epoch": 0.0258732212160414,
|
| 27 |
+
"grad_norm": 2.54523168619682e+30,
|
| 28 |
+
"learning_rate": 1.2820512820512818e-07,
|
| 29 |
+
"loss": 35.3514,
|
| 30 |
"step": 10
|
| 31 |
},
|
| 32 |
{
|
| 33 |
+
"epoch": 0.03880983182406209,
|
| 34 |
+
"grad_norm": 2.9350637382111784e+28,
|
| 35 |
+
"learning_rate": 1.9230769230769231e-07,
|
| 36 |
+
"loss": 35.2418,
|
| 37 |
"step": 15
|
| 38 |
},
|
| 39 |
{
|
| 40 |
+
"epoch": 0.0517464424320828,
|
| 41 |
+
"grad_norm": 6.845379509396579e+24,
|
| 42 |
+
"learning_rate": 2.5641025641025636e-07,
|
| 43 |
+
"loss": 35.4029,
|
| 44 |
"step": 20
|
| 45 |
},
|
| 46 |
{
|
| 47 |
+
"epoch": 0.0646830530401035,
|
| 48 |
+
"grad_norm": 9.432421146054794e+16,
|
| 49 |
+
"learning_rate": 3.2051282051282055e-07,
|
| 50 |
+
"loss": 35.4307,
|
| 51 |
"step": 25
|
| 52 |
},
|
| 53 |
{
|
| 54 |
+
"epoch": 0.07761966364812418,
|
| 55 |
+
"grad_norm": 103996205868.7185,
|
| 56 |
+
"learning_rate": 3.8461538461538463e-07,
|
| 57 |
+
"loss": 35.3465,
|
| 58 |
"step": 30
|
| 59 |
},
|
| 60 |
{
|
| 61 |
+
"epoch": 0.09055627425614489,
|
| 62 |
+
"grad_norm": 7902.581384813984,
|
| 63 |
+
"learning_rate": 4.487179487179487e-07,
|
| 64 |
+
"loss": 35.2609,
|
| 65 |
"step": 35
|
| 66 |
},
|
| 67 |
{
|
| 68 |
+
"epoch": 0.1034928848641656,
|
| 69 |
+
"grad_norm": 3891.3695056510064,
|
| 70 |
+
"learning_rate": 4.999897541535663e-07,
|
| 71 |
+
"loss": 30.0438,
|
| 72 |
"step": 40
|
| 73 |
},
|
| 74 |
{
|
| 75 |
+
"epoch": 0.11642949547218628,
|
| 76 |
+
"grad_norm": 2358.218700340259,
|
| 77 |
+
"learning_rate": 4.996312377016688e-07,
|
| 78 |
+
"loss": 20.8297,
|
| 79 |
"step": 45
|
| 80 |
},
|
| 81 |
{
|
| 82 |
+
"epoch": 0.129366106080207,
|
| 83 |
+
"grad_norm": 330.05121170654246,
|
| 84 |
+
"learning_rate": 4.987612684376705e-07,
|
| 85 |
+
"loss": 12.961,
|
| 86 |
"step": 50
|
| 87 |
},
|
| 88 |
{
|
| 89 |
+
"epoch": 0.1423027166882277,
|
| 90 |
+
"grad_norm": 193.68319402032964,
|
| 91 |
+
"learning_rate": 4.973816287836379e-07,
|
| 92 |
+
"loss": 10.809,
|
| 93 |
"step": 55
|
| 94 |
},
|
| 95 |
{
|
| 96 |
+
"epoch": 0.15523932729624837,
|
| 97 |
+
"grad_norm": 63.544778628785686,
|
| 98 |
+
"learning_rate": 4.954951453913442e-07,
|
| 99 |
+
"loss": 10.385,
|
| 100 |
"step": 60
|
| 101 |
},
|
| 102 |
{
|
| 103 |
+
"epoch": 0.16817593790426907,
|
| 104 |
+
"grad_norm": 26.370287565962407,
|
| 105 |
+
"learning_rate": 4.931056833509313e-07,
|
| 106 |
+
"loss": 10.1709,
|
| 107 |
"step": 65
|
| 108 |
},
|
| 109 |
{
|
| 110 |
+
"epoch": 0.18111254851228978,
|
| 111 |
+
"grad_norm": 99.65626298246401,
|
| 112 |
+
"learning_rate": 4.902181382719843e-07,
|
| 113 |
+
"loss": 10.0785,
|
| 114 |
"step": 70
|
| 115 |
},
|
| 116 |
{
|
| 117 |
+
"epoch": 0.19404915912031048,
|
| 118 |
+
"grad_norm": 91.97251247083376,
|
| 119 |
+
"learning_rate": 4.868384262532425e-07,
|
| 120 |
+
"loss": 10.0305,
|
| 121 |
"step": 75
|
| 122 |
},
|
| 123 |
{
|
| 124 |
+
"epoch": 0.2069857697283312,
|
| 125 |
+
"grad_norm": 133.24722674771854,
|
| 126 |
+
"learning_rate": 4.829734717614995e-07,
|
| 127 |
+
"loss": 9.9683,
|
| 128 |
"step": 80
|
| 129 |
},
|
| 130 |
{
|
| 131 |
+
"epoch": 0.21992238033635186,
|
| 132 |
+
"grad_norm": 49.44784904214244,
|
| 133 |
+
"learning_rate": 4.78631193444524e-07,
|
| 134 |
+
"loss": 9.8086,
|
| 135 |
"step": 85
|
| 136 |
},
|
| 137 |
{
|
| 138 |
+
"epoch": 0.23285899094437257,
|
| 139 |
+
"grad_norm": 39.83457542693803,
|
| 140 |
+
"learning_rate": 4.738204879070702e-07,
|
| 141 |
+
"loss": 9.6444,
|
| 142 |
"step": 90
|
| 143 |
},
|
| 144 |
{
|
| 145 |
+
"epoch": 0.24579560155239327,
|
| 146 |
+
"grad_norm": 35.524332578393064,
|
| 147 |
+
"learning_rate": 4.6855121148321705e-07,
|
| 148 |
+
"loss": 9.5317,
|
| 149 |
"step": 95
|
| 150 |
},
|
| 151 |
{
|
| 152 |
+
"epoch": 0.258732212160414,
|
| 153 |
+
"grad_norm": 21.89346239249435,
|
| 154 |
+
"learning_rate": 4.6283416004238185e-07,
|
| 155 |
+
"loss": 9.4199,
|
| 156 |
"step": 100
|
| 157 |
},
|
| 158 |
{
|
| 159 |
+
"epoch": 0.2716688227684347,
|
| 160 |
+
"grad_norm": 18.371198207871824,
|
| 161 |
+
"learning_rate": 4.566810468703828e-07,
|
| 162 |
+
"loss": 9.3422,
|
| 163 |
"step": 105
|
| 164 |
},
|
| 165 |
{
|
| 166 |
+
"epoch": 0.2846054333764554,
|
| 167 |
+
"grad_norm": 14.363049133916364,
|
| 168 |
+
"learning_rate": 4.5010447867086775e-07,
|
| 169 |
+
"loss": 9.2499,
|
| 170 |
"step": 110
|
| 171 |
},
|
| 172 |
{
|
| 173 |
+
"epoch": 0.2975420439844761,
|
| 174 |
+
"grad_norm": 25.239020822491135,
|
| 175 |
+
"learning_rate": 4.431179297362797e-07,
|
| 176 |
+
"loss": 9.1466,
|
| 177 |
"step": 115
|
| 178 |
},
|
| 179 |
{
|
| 180 |
+
"epoch": 0.31047865459249674,
|
| 181 |
+
"grad_norm": 50.305464578443626,
|
| 182 |
+
"learning_rate": 4.3573571434127553e-07,
|
| 183 |
+
"loss": 9.1168,
|
| 184 |
"step": 120
|
| 185 |
},
|
| 186 |
{
|
| 187 |
+
"epoch": 0.32341526520051744,
|
| 188 |
+
"grad_norm": 29.946521630452818,
|
| 189 |
+
"learning_rate": 4.2797295741516337e-07,
|
| 190 |
+
"loss": 9.0978,
|
| 191 |
"step": 125
|
| 192 |
},
|
| 193 |
{
|
| 194 |
+
"epoch": 0.33635187580853815,
|
| 195 |
+
"grad_norm": 35.470152215729236,
|
| 196 |
+
"learning_rate": 4.1984556355344205e-07,
|
| 197 |
+
"loss": 9.0445,
|
| 198 |
"step": 130
|
| 199 |
},
|
| 200 |
{
|
| 201 |
+
"epoch": 0.34928848641655885,
|
| 202 |
+
"grad_norm": 26.3769064790735,
|
| 203 |
+
"learning_rate": 4.1137018443193496e-07,
|
| 204 |
+
"loss": 9.0052,
|
| 205 |
"step": 135
|
| 206 |
},
|
| 207 |
{
|
| 208 |
+
"epoch": 0.36222509702457956,
|
| 209 |
+
"grad_norm": 28.18463108090832,
|
| 210 |
+
"learning_rate": 4.025641846902812e-07,
|
| 211 |
+
"loss": 8.9554,
|
| 212 |
"step": 140
|
| 213 |
},
|
| 214 |
{
|
| 215 |
+
"epoch": 0.37516170763260026,
|
| 216 |
+
"grad_norm": 23.521159788503216,
|
| 217 |
+
"learning_rate": 3.9344560635468183e-07,
|
| 218 |
+
"loss": 8.9328,
|
| 219 |
"step": 145
|
| 220 |
},
|
| 221 |
{
|
| 222 |
+
"epoch": 0.38809831824062097,
|
| 223 |
+
"grad_norm": 20.007227770493905,
|
| 224 |
+
"learning_rate": 3.8403313187279446e-07,
|
| 225 |
+
"loss": 8.8458,
|
| 226 |
"step": 150
|
| 227 |
},
|
| 228 |
{
|
| 229 |
+
"epoch": 0.40103492884864167,
|
| 230 |
+
"grad_norm": 20.506423559359675,
|
| 231 |
+
"learning_rate": 3.743460458365114e-07,
|
| 232 |
+
"loss": 8.8147,
|
| 233 |
"step": 155
|
| 234 |
},
|
| 235 |
{
|
| 236 |
+
"epoch": 0.4139715394566624,
|
| 237 |
+
"grad_norm": 20.107542907366796,
|
| 238 |
+
"learning_rate": 3.644041954710432e-07,
|
| 239 |
+
"loss": 8.7916,
|
| 240 |
"step": 160
|
| 241 |
},
|
| 242 |
{
|
| 243 |
+
"epoch": 0.4269081500646831,
|
| 244 |
+
"grad_norm": 18.743137987266405,
|
| 245 |
+
"learning_rate": 3.5422794997126223e-07,
|
| 246 |
+
"loss": 8.7202,
|
| 247 |
"step": 165
|
| 248 |
},
|
| 249 |
{
|
| 250 |
+
"epoch": 0.4398447606727037,
|
| 251 |
+
"grad_norm": 21.53271448781464,
|
| 252 |
+
"learning_rate": 3.438381587686152e-07,
|
| 253 |
+
"loss": 8.6901,
|
| 254 |
"step": 170
|
| 255 |
},
|
| 256 |
{
|
| 257 |
+
"epoch": 0.45278137128072443,
|
| 258 |
+
"grad_norm": 29.119808165824566,
|
| 259 |
+
"learning_rate": 3.3325610881411314e-07,
|
| 260 |
+
"loss": 8.6836,
|
| 261 |
"step": 175
|
| 262 |
},
|
| 263 |
{
|
| 264 |
+
"epoch": 0.46571798188874514,
|
| 265 |
+
"grad_norm": 19.101582760449293,
|
| 266 |
+
"learning_rate": 3.225034809649149e-07,
|
| 267 |
+
"loss": 8.6107,
|
| 268 |
"step": 180
|
| 269 |
},
|
| 270 |
{
|
| 271 |
+
"epoch": 0.47865459249676584,
|
| 272 |
+
"grad_norm": 26.344817395383185,
|
| 273 |
+
"learning_rate": 3.116023055638638e-07,
|
| 274 |
+
"loss": 8.5785,
|
| 275 |
"step": 185
|
| 276 |
},
|
| 277 |
{
|
| 278 |
+
"epoch": 0.49159120310478654,
|
| 279 |
+
"grad_norm": 22.757551882374766,
|
| 280 |
+
"learning_rate": 3.005749173029856e-07,
|
| 281 |
+
"loss": 8.5407,
|
| 282 |
"step": 190
|
| 283 |
},
|
| 284 |
{
|
| 285 |
+
"epoch": 0.5045278137128072,
|
| 286 |
+
"grad_norm": 17.287888797801536,
|
| 287 |
+
"learning_rate": 2.894439094634258e-07,
|
| 288 |
+
"loss": 8.4908,
|
| 289 |
"step": 195
|
| 290 |
},
|
| 291 |
{
|
| 292 |
+
"epoch": 0.517464424320828,
|
| 293 |
+
"grad_norm": 18.78518698061505,
|
| 294 |
+
"learning_rate": 2.782320876255818e-07,
|
| 295 |
+
"loss": 8.4408,
|
| 296 |
"step": 200
|
| 297 |
},
|
| 298 |
{
|
| 299 |
+
"epoch": 0.5304010349288486,
|
| 300 |
+
"grad_norm": 26.147386948788057,
|
| 301 |
+
"learning_rate": 2.6696242294426794e-07,
|
| 302 |
+
"loss": 8.4371,
|
| 303 |
"step": 205
|
| 304 |
},
|
| 305 |
{
|
| 306 |
+
"epoch": 0.5433376455368694,
|
| 307 |
+
"grad_norm": 20.864739238751334,
|
| 308 |
+
"learning_rate": 2.5565800508464693e-07,
|
| 309 |
+
"loss": 8.3946,
|
| 310 |
"step": 210
|
| 311 |
},
|
| 312 |
{
|
| 313 |
+
"epoch": 0.55627425614489,
|
| 314 |
+
"grad_norm": 25.03041019448578,
|
| 315 |
+
"learning_rate": 2.443419949153531e-07,
|
| 316 |
+
"loss": 8.3438,
|
| 317 |
"step": 215
|
| 318 |
},
|
| 319 |
{
|
| 320 |
+
"epoch": 0.5692108667529108,
|
| 321 |
+
"grad_norm": 20.710123234104667,
|
| 322 |
+
"learning_rate": 2.3303757705573201e-07,
|
| 323 |
+
"loss": 8.3685,
|
| 324 |
"step": 220
|
| 325 |
},
|
| 326 |
{
|
| 327 |
+
"epoch": 0.5821474773609314,
|
| 328 |
+
"grad_norm": 20.005194039944502,
|
| 329 |
+
"learning_rate": 2.217679123744182e-07,
|
| 330 |
+
"loss": 8.2943,
|
| 331 |
"step": 225
|
| 332 |
},
|
| 333 |
{
|
| 334 |
+
"epoch": 0.5950840879689522,
|
| 335 |
+
"grad_norm": 20.64021311062467,
|
| 336 |
+
"learning_rate": 2.1055609053657423e-07,
|
| 337 |
+
"loss": 8.2681,
|
| 338 |
"step": 230
|
| 339 |
},
|
| 340 |
{
|
| 341 |
+
"epoch": 0.6080206985769728,
|
| 342 |
+
"grad_norm": 18.041357467771427,
|
| 343 |
+
"learning_rate": 1.9942508269701447e-07,
|
| 344 |
+
"loss": 8.2574,
|
| 345 |
"step": 235
|
| 346 |
},
|
| 347 |
{
|
| 348 |
+
"epoch": 0.6209573091849935,
|
| 349 |
+
"grad_norm": 24.18533351853082,
|
| 350 |
+
"learning_rate": 1.883976944361362e-07,
|
| 351 |
+
"loss": 8.2044,
|
| 352 |
"step": 240
|
| 353 |
},
|
| 354 |
{
|
| 355 |
+
"epoch": 0.6338939197930142,
|
| 356 |
+
"grad_norm": 19.308910630600565,
|
| 357 |
+
"learning_rate": 1.7749651903508505e-07,
|
| 358 |
+
"loss": 8.1439,
|
| 359 |
"step": 245
|
| 360 |
},
|
| 361 |
{
|
| 362 |
+
"epoch": 0.6468305304010349,
|
| 363 |
+
"grad_norm": 18.940923202491184,
|
| 364 |
+
"learning_rate": 1.6674389118588684e-07,
|
| 365 |
+
"loss": 8.2016,
|
| 366 |
"step": 250
|
| 367 |
},
|
| 368 |
{
|
| 369 |
+
"epoch": 0.6597671410090556,
|
| 370 |
+
"grad_norm": 22.62903264366213,
|
| 371 |
+
"learning_rate": 1.5616184123138476e-07,
|
| 372 |
+
"loss": 8.1482,
|
| 373 |
"step": 255
|
| 374 |
},
|
| 375 |
{
|
| 376 |
+
"epoch": 0.6727037516170763,
|
| 377 |
+
"grad_norm": 18.71414418245413,
|
| 378 |
+
"learning_rate": 1.457720500287379e-07,
|
| 379 |
+
"loss": 8.1104,
|
| 380 |
"step": 260
|
| 381 |
},
|
| 382 |
{
|
| 383 |
+
"epoch": 0.685640362225097,
|
| 384 |
+
"grad_norm": 18.16814176627225,
|
| 385 |
+
"learning_rate": 1.3559580452895682e-07,
|
| 386 |
+
"loss": 8.1109,
|
| 387 |
"step": 265
|
| 388 |
},
|
| 389 |
{
|
| 390 |
+
"epoch": 0.6985769728331177,
|
| 391 |
+
"grad_norm": 19.84573072806573,
|
| 392 |
+
"learning_rate": 1.2565395416348867e-07,
|
| 393 |
+
"loss": 8.0936,
|
| 394 |
"step": 270
|
| 395 |
},
|
| 396 |
{
|
| 397 |
+
"epoch": 0.7115135834411385,
|
| 398 |
+
"grad_norm": 19.40761310343157,
|
| 399 |
+
"learning_rate": 1.1596686812720555e-07,
|
| 400 |
+
"loss": 8.081,
|
| 401 |
"step": 275
|
| 402 |
},
|
| 403 |
{
|
| 404 |
+
"epoch": 0.7244501940491591,
|
| 405 |
+
"grad_norm": 17.694041027215786,
|
| 406 |
+
"learning_rate": 1.065543936453182e-07,
|
| 407 |
+
"loss": 8.0583,
|
| 408 |
"step": 280
|
| 409 |
},
|
| 410 |
{
|
| 411 |
+
"epoch": 0.7373868046571799,
|
| 412 |
+
"grad_norm": 19.473414715193275,
|
| 413 |
+
"learning_rate": 9.743581530971878e-08,
|
| 414 |
+
"loss": 8.0536,
|
| 415 |
"step": 285
|
| 416 |
},
|
| 417 |
{
|
| 418 |
+
"epoch": 0.7503234152652005,
|
| 419 |
+
"grad_norm": 20.279917617864715,
|
| 420 |
+
"learning_rate": 8.862981556806499e-08,
|
| 421 |
+
"loss": 8.0876,
|
| 422 |
"step": 290
|
| 423 |
},
|
| 424 |
{
|
| 425 |
+
"epoch": 0.7632600258732212,
|
| 426 |
+
"grad_norm": 18.215356879442886,
|
| 427 |
+
"learning_rate": 8.0154436446558e-08,
|
| 428 |
+
"loss": 8.0188,
|
| 429 |
"step": 295
|
| 430 |
},
|
| 431 |
{
|
| 432 |
+
"epoch": 0.7761966364812419,
|
| 433 |
+
"grad_norm": 17.250109490543473,
|
| 434 |
+
"learning_rate": 7.202704258483663e-08,
|
| 435 |
+
"loss": 8.0066,
|
| 436 |
"step": 300
|
| 437 |
},
|
| 438 |
{
|
| 439 |
+
"epoch": 0.7891332470892626,
|
| 440 |
+
"grad_norm": 16.664504705841736,
|
| 441 |
+
"learning_rate": 6.426428565872443e-08,
|
| 442 |
+
"loss": 8.0198,
|
| 443 |
"step": 305
|
| 444 |
},
|
| 445 |
{
|
| 446 |
+
"epoch": 0.8020698576972833,
|
| 447 |
+
"grad_norm": 21.103826588222557,
|
| 448 |
+
"learning_rate": 5.688207026372027e-08,
|
| 449 |
+
"loss": 8.0141,
|
| 450 |
"step": 310
|
| 451 |
},
|
| 452 |
{
|
| 453 |
+
"epoch": 0.815006468305304,
|
| 454 |
+
"grad_norm": 17.388005585117728,
|
| 455 |
+
"learning_rate": 4.989552132913219e-08,
|
| 456 |
+
"loss": 7.9909,
|
| 457 |
"step": 315
|
| 458 |
},
|
| 459 |
{
|
| 460 |
+
"epoch": 0.8279430789133247,
|
| 461 |
+
"grad_norm": 17.869968121368707,
|
| 462 |
+
"learning_rate": 4.331895312961725e-08,
|
| 463 |
+
"loss": 7.9866,
|
| 464 |
"step": 320
|
| 465 |
},
|
| 466 |
{
|
| 467 |
+
"epoch": 0.8408796895213454,
|
| 468 |
+
"grad_norm": 19.683008826871088,
|
| 469 |
+
"learning_rate": 3.7165839957618156e-08,
|
| 470 |
+
"loss": 7.9899,
|
| 471 |
"step": 325
|
| 472 |
},
|
| 473 |
{
|
| 474 |
+
"epoch": 0.8538163001293662,
|
| 475 |
+
"grad_norm": 17.460030572155425,
|
| 476 |
+
"learning_rate": 3.144878851678298e-08,
|
| 477 |
+
"loss": 7.9855,
|
| 478 |
"step": 330
|
| 479 |
},
|
| 480 |
{
|
| 481 |
+
"epoch": 0.8667529107373868,
|
| 482 |
+
"grad_norm": 17.02791948891275,
|
| 483 |
+
"learning_rate": 2.617951209292979e-08,
|
| 484 |
+
"loss": 8.0126,
|
| 485 |
"step": 335
|
| 486 |
},
|
| 487 |
{
|
| 488 |
+
"epoch": 0.8796895213454075,
|
| 489 |
+
"grad_norm": 20.515442084242775,
|
| 490 |
+
"learning_rate": 2.136880655547596e-08,
|
| 491 |
+
"loss": 8.0,
|
| 492 |
"step": 340
|
| 493 |
},
|
| 494 |
{
|
| 495 |
+
"epoch": 0.8926261319534282,
|
| 496 |
+
"grad_norm": 16.829511963680876,
|
| 497 |
+
"learning_rate": 1.7026528238500426e-08,
|
| 498 |
+
"loss": 7.9638,
|
| 499 |
"step": 345
|
| 500 |
},
|
| 501 |
{
|
| 502 |
+
"epoch": 0.9055627425614489,
|
| 503 |
+
"grad_norm": 17.7377760938193,
|
| 504 |
+
"learning_rate": 1.3161573746757415e-08,
|
| 505 |
+
"loss": 7.923,
|
| 506 |
"step": 350
|
| 507 |
},
|
| 508 |
{
|
| 509 |
+
"epoch": 0.9184993531694696,
|
| 510 |
+
"grad_norm": 17.924298918004375,
|
| 511 |
+
"learning_rate": 9.78186172801565e-09,
|
| 512 |
+
"loss": 7.9911,
|
| 513 |
"step": 355
|
| 514 |
},
|
| 515 |
{
|
| 516 |
+
"epoch": 0.9314359637774903,
|
| 517 |
+
"grad_norm": 17.683222436895043,
|
| 518 |
+
"learning_rate": 6.894316649068643e-09,
|
| 519 |
+
"loss": 7.9886,
|
| 520 |
"step": 360
|
| 521 |
},
|
| 522 |
{
|
| 523 |
+
"epoch": 0.944372574385511,
|
| 524 |
+
"grad_norm": 17.323184239913775,
|
| 525 |
+
"learning_rate": 4.50485460865585e-09,
|
| 526 |
+
"loss": 7.9649,
|
| 527 |
"step": 365
|
| 528 |
},
|
| 529 |
{
|
| 530 |
+
"epoch": 0.9573091849935317,
|
| 531 |
+
"grad_norm": 16.759552285384956,
|
| 532 |
+
"learning_rate": 2.6183712163621308e-09,
|
| 533 |
+
"loss": 7.9909,
|
| 534 |
"step": 370
|
| 535 |
},
|
| 536 |
{
|
| 537 |
+
"epoch": 0.9702457956015524,
|
| 538 |
+
"grad_norm": 16.40682058120129,
|
| 539 |
+
"learning_rate": 1.2387315623294536e-09,
|
| 540 |
+
"loss": 8.0056,
|
| 541 |
"step": 375
|
| 542 |
},
|
| 543 |
{
|
| 544 |
+
"epoch": 0.9831824062095731,
|
| 545 |
+
"grad_norm": 16.270242929964763,
|
| 546 |
+
"learning_rate": 3.6876229833118776e-10,
|
| 547 |
+
"loss": 7.9802,
|
| 548 |
"step": 380
|
| 549 |
},
|
| 550 |
{
|
| 551 |
+
"epoch": 0.9961190168175937,
|
| 552 |
+
"grad_norm": 17.17813750433213,
|
| 553 |
+
"learning_rate": 1.0245846433665217e-11,
|
| 554 |
+
"loss": 7.9839,
|
| 555 |
"step": 385
|
| 556 |
},
|
| 557 |
{
|
| 558 |
+
"epoch": 0.9987063389391979,
|
| 559 |
+
"eval_loss": 7.827143669128418,
|
| 560 |
+
"eval_runtime": 3.6254,
|
| 561 |
+
"eval_samples_per_second": 62.613,
|
| 562 |
+
"eval_steps_per_second": 1.103,
|
| 563 |
+
"step": 386
|
| 564 |
},
|
| 565 |
{
|
| 566 |
+
"epoch": 0.9987063389391979,
|
| 567 |
+
"step": 386,
|
| 568 |
+
"total_flos": 161536404357120.0,
|
| 569 |
+
"train_loss": 11.514628536342958,
|
| 570 |
+
"train_runtime": 5043.2067,
|
| 571 |
+
"train_samples_per_second": 19.608,
|
| 572 |
+
"train_steps_per_second": 0.077
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 573 |
}
|
| 574 |
],
|
| 575 |
"logging_steps": 5,
|
| 576 |
+
"max_steps": 386,
|
| 577 |
"num_input_tokens_seen": 0,
|
| 578 |
"num_train_epochs": 1,
|
| 579 |
"save_steps": 100,
|
|
|
|
| 589 |
"attributes": {}
|
| 590 |
}
|
| 591 |
},
|
| 592 |
+
"total_flos": 161536404357120.0,
|
| 593 |
"train_batch_size": 8,
|
| 594 |
"trial_name": null,
|
| 595 |
"trial_params": null
|