Upload 9 files
Browse files- all_results.json +5 -5
- train_results.json +5 -5
- trainer_state.json +306 -306
all_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch": 0.
|
| 3 |
"total_flos": 2305515375820800.0,
|
| 4 |
-
"train_loss": 1.
|
| 5 |
-
"train_runtime":
|
| 6 |
-
"train_samples_per_second": 0.
|
| 7 |
-
"train_steps_per_second": 0.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 0.8547008547008547,
|
| 3 |
"total_flos": 2305515375820800.0,
|
| 4 |
+
"train_loss": 1.7708107882738113,
|
| 5 |
+
"train_runtime": 5686.4644,
|
| 6 |
+
"train_samples_per_second": 0.018,
|
| 7 |
+
"train_steps_per_second": 0.018
|
| 8 |
}
|
train_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch": 0.
|
| 3 |
"total_flos": 2305515375820800.0,
|
| 4 |
-
"train_loss": 1.
|
| 5 |
-
"train_runtime":
|
| 6 |
-
"train_samples_per_second": 0.
|
| 7 |
-
"train_steps_per_second": 0.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 0.8547008547008547,
|
| 3 |
"total_flos": 2305515375820800.0,
|
| 4 |
+
"train_loss": 1.7708107882738113,
|
| 5 |
+
"train_runtime": 5686.4644,
|
| 6 |
+
"train_samples_per_second": 0.018,
|
| 7 |
+
"train_steps_per_second": 0.018
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,7 +1,7 @@
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
-
"epoch": 0.
|
| 5 |
"eval_steps": 500,
|
| 6 |
"global_step": 100,
|
| 7 |
"is_hyper_param_search": false,
|
|
@@ -9,713 +9,713 @@
|
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
-
"epoch": 0.
|
| 13 |
-
"grad_norm":
|
| 14 |
"learning_rate": 5e-06,
|
| 15 |
-
"loss":
|
| 16 |
"step": 1
|
| 17 |
},
|
| 18 |
{
|
| 19 |
-
"epoch": 0.
|
| 20 |
-
"grad_norm":
|
| 21 |
"learning_rate": 1e-05,
|
| 22 |
-
"loss":
|
| 23 |
"step": 2
|
| 24 |
},
|
| 25 |
{
|
| 26 |
-
"epoch": 0.
|
| 27 |
-
"grad_norm":
|
| 28 |
"learning_rate": 9.89795918367347e-06,
|
| 29 |
-
"loss":
|
| 30 |
"step": 3
|
| 31 |
},
|
| 32 |
{
|
| 33 |
-
"epoch": 0.
|
| 34 |
-
"grad_norm":
|
| 35 |
"learning_rate": 9.795918367346939e-06,
|
| 36 |
-
"loss": 1.
|
| 37 |
"step": 4
|
| 38 |
},
|
| 39 |
{
|
| 40 |
-
"epoch": 0.
|
| 41 |
-
"grad_norm":
|
| 42 |
"learning_rate": 9.693877551020408e-06,
|
| 43 |
-
"loss":
|
| 44 |
"step": 5
|
| 45 |
},
|
| 46 |
{
|
| 47 |
-
"epoch": 0.
|
| 48 |
-
"grad_norm":
|
| 49 |
"learning_rate": 9.591836734693878e-06,
|
| 50 |
-
"loss":
|
| 51 |
"step": 6
|
| 52 |
},
|
| 53 |
{
|
| 54 |
-
"epoch": 0.
|
| 55 |
-
"grad_norm":
|
| 56 |
"learning_rate": 9.489795918367348e-06,
|
| 57 |
-
"loss": 2.
|
| 58 |
"step": 7
|
| 59 |
},
|
| 60 |
{
|
| 61 |
-
"epoch": 0.
|
| 62 |
-
"grad_norm":
|
| 63 |
"learning_rate": 9.387755102040818e-06,
|
| 64 |
-
"loss":
|
| 65 |
"step": 8
|
| 66 |
},
|
| 67 |
{
|
| 68 |
-
"epoch": 0.
|
| 69 |
-
"grad_norm":
|
| 70 |
"learning_rate": 9.285714285714288e-06,
|
| 71 |
-
"loss": 1.
|
| 72 |
"step": 9
|
| 73 |
},
|
| 74 |
{
|
| 75 |
-
"epoch": 0.
|
| 76 |
-
"grad_norm":
|
| 77 |
"learning_rate": 9.183673469387756e-06,
|
| 78 |
-
"loss":
|
| 79 |
"step": 10
|
| 80 |
},
|
| 81 |
{
|
| 82 |
-
"epoch": 0.
|
| 83 |
-
"grad_norm":
|
| 84 |
"learning_rate": 9.081632653061225e-06,
|
| 85 |
-
"loss":
|
| 86 |
"step": 11
|
| 87 |
},
|
| 88 |
{
|
| 89 |
-
"epoch": 0.
|
| 90 |
-
"grad_norm":
|
| 91 |
"learning_rate": 8.979591836734695e-06,
|
| 92 |
-
"loss": 1.
|
| 93 |
"step": 12
|
| 94 |
},
|
| 95 |
{
|
| 96 |
-
"epoch": 0.
|
| 97 |
-
"grad_norm":
|
| 98 |
"learning_rate": 8.877551020408163e-06,
|
| 99 |
-
"loss":
|
| 100 |
"step": 13
|
| 101 |
},
|
| 102 |
{
|
| 103 |
-
"epoch": 0.
|
| 104 |
-
"grad_norm":
|
| 105 |
"learning_rate": 8.775510204081633e-06,
|
| 106 |
-
"loss": 1.
|
| 107 |
"step": 14
|
| 108 |
},
|
| 109 |
{
|
| 110 |
-
"epoch": 0.
|
| 111 |
-
"grad_norm":
|
| 112 |
"learning_rate": 8.673469387755103e-06,
|
| 113 |
-
"loss":
|
| 114 |
"step": 15
|
| 115 |
},
|
| 116 |
{
|
| 117 |
-
"epoch": 0.
|
| 118 |
-
"grad_norm":
|
| 119 |
"learning_rate": 8.571428571428571e-06,
|
| 120 |
-
"loss": 2.
|
| 121 |
"step": 16
|
| 122 |
},
|
| 123 |
{
|
| 124 |
-
"epoch": 0.
|
| 125 |
-
"grad_norm":
|
| 126 |
"learning_rate": 8.469387755102042e-06,
|
| 127 |
-
"loss":
|
| 128 |
"step": 17
|
| 129 |
},
|
| 130 |
{
|
| 131 |
-
"epoch": 0.
|
| 132 |
-
"grad_norm":
|
| 133 |
"learning_rate": 8.36734693877551e-06,
|
| 134 |
-
"loss": 2.
|
| 135 |
"step": 18
|
| 136 |
},
|
| 137 |
{
|
| 138 |
-
"epoch": 0.
|
| 139 |
-
"grad_norm":
|
| 140 |
"learning_rate": 8.26530612244898e-06,
|
| 141 |
-
"loss":
|
| 142 |
"step": 19
|
| 143 |
},
|
| 144 |
{
|
| 145 |
-
"epoch": 0.
|
| 146 |
-
"grad_norm":
|
| 147 |
"learning_rate": 8.16326530612245e-06,
|
| 148 |
-
"loss": 1.
|
| 149 |
"step": 20
|
| 150 |
},
|
| 151 |
{
|
| 152 |
-
"epoch": 0.
|
| 153 |
-
"grad_norm":
|
| 154 |
"learning_rate": 8.06122448979592e-06,
|
| 155 |
-
"loss":
|
| 156 |
"step": 21
|
| 157 |
},
|
| 158 |
{
|
| 159 |
-
"epoch": 0.
|
| 160 |
-
"grad_norm":
|
| 161 |
"learning_rate": 7.959183673469388e-06,
|
| 162 |
-
"loss": 2.
|
| 163 |
"step": 22
|
| 164 |
},
|
| 165 |
{
|
| 166 |
-
"epoch": 0.
|
| 167 |
-
"grad_norm":
|
| 168 |
"learning_rate": 7.857142857142858e-06,
|
| 169 |
-
"loss":
|
| 170 |
"step": 23
|
| 171 |
},
|
| 172 |
{
|
| 173 |
-
"epoch": 0.
|
| 174 |
-
"grad_norm":
|
| 175 |
"learning_rate": 7.755102040816327e-06,
|
| 176 |
-
"loss": 1.
|
| 177 |
"step": 24
|
| 178 |
},
|
| 179 |
{
|
| 180 |
-
"epoch": 0.
|
| 181 |
-
"grad_norm":
|
| 182 |
"learning_rate": 7.653061224489796e-06,
|
| 183 |
-
"loss":
|
| 184 |
"step": 25
|
| 185 |
},
|
| 186 |
{
|
| 187 |
-
"epoch": 0.
|
| 188 |
-
"grad_norm":
|
| 189 |
"learning_rate": 7.551020408163265e-06,
|
| 190 |
-
"loss":
|
| 191 |
"step": 26
|
| 192 |
},
|
| 193 |
{
|
| 194 |
-
"epoch": 0.
|
| 195 |
-
"grad_norm":
|
| 196 |
"learning_rate": 7.448979591836736e-06,
|
| 197 |
-
"loss": 1.
|
| 198 |
"step": 27
|
| 199 |
},
|
| 200 |
{
|
| 201 |
-
"epoch": 0.
|
| 202 |
-
"grad_norm":
|
| 203 |
"learning_rate": 7.346938775510205e-06,
|
| 204 |
-
"loss":
|
| 205 |
"step": 28
|
| 206 |
},
|
| 207 |
{
|
| 208 |
-
"epoch": 0.
|
| 209 |
-
"grad_norm":
|
| 210 |
"learning_rate": 7.244897959183675e-06,
|
| 211 |
-
"loss":
|
| 212 |
"step": 29
|
| 213 |
},
|
| 214 |
{
|
| 215 |
-
"epoch": 0.
|
| 216 |
-
"grad_norm":
|
| 217 |
"learning_rate": 7.1428571428571436e-06,
|
| 218 |
-
"loss":
|
| 219 |
"step": 30
|
| 220 |
},
|
| 221 |
{
|
| 222 |
-
"epoch": 0.
|
| 223 |
-
"grad_norm":
|
| 224 |
"learning_rate": 7.0408163265306125e-06,
|
| 225 |
-
"loss": 2.
|
| 226 |
"step": 31
|
| 227 |
},
|
| 228 |
{
|
| 229 |
-
"epoch": 0.
|
| 230 |
-
"grad_norm":
|
| 231 |
"learning_rate": 6.938775510204082e-06,
|
| 232 |
-
"loss": 1.
|
| 233 |
"step": 32
|
| 234 |
},
|
| 235 |
{
|
| 236 |
-
"epoch": 0.
|
| 237 |
-
"grad_norm": 31.
|
| 238 |
"learning_rate": 6.836734693877551e-06,
|
| 239 |
-
"loss": 1.
|
| 240 |
"step": 33
|
| 241 |
},
|
| 242 |
{
|
| 243 |
-
"epoch": 0.
|
| 244 |
-
"grad_norm":
|
| 245 |
"learning_rate": 6.734693877551021e-06,
|
| 246 |
-
"loss": 2.
|
| 247 |
"step": 34
|
| 248 |
},
|
| 249 |
{
|
| 250 |
-
"epoch": 0.
|
| 251 |
-
"grad_norm":
|
| 252 |
"learning_rate": 6.63265306122449e-06,
|
| 253 |
-
"loss":
|
| 254 |
"step": 35
|
| 255 |
},
|
| 256 |
{
|
| 257 |
-
"epoch": 0.
|
| 258 |
-
"grad_norm":
|
| 259 |
"learning_rate": 6.530612244897959e-06,
|
| 260 |
-
"loss":
|
| 261 |
"step": 36
|
| 262 |
},
|
| 263 |
{
|
| 264 |
-
"epoch": 0.
|
| 265 |
-
"grad_norm":
|
| 266 |
"learning_rate": 6.4285714285714295e-06,
|
| 267 |
-
"loss":
|
| 268 |
"step": 37
|
| 269 |
},
|
| 270 |
{
|
| 271 |
-
"epoch": 0.
|
| 272 |
-
"grad_norm":
|
| 273 |
"learning_rate": 6.326530612244899e-06,
|
| 274 |
-
"loss": 1.
|
| 275 |
"step": 38
|
| 276 |
},
|
| 277 |
{
|
| 278 |
-
"epoch": 0.
|
| 279 |
-
"grad_norm":
|
| 280 |
"learning_rate": 6.224489795918368e-06,
|
| 281 |
-
"loss":
|
| 282 |
"step": 39
|
| 283 |
},
|
| 284 |
{
|
| 285 |
-
"epoch": 0.
|
| 286 |
-
"grad_norm":
|
| 287 |
"learning_rate": 6.122448979591837e-06,
|
| 288 |
-
"loss":
|
| 289 |
"step": 40
|
| 290 |
},
|
| 291 |
{
|
| 292 |
-
"epoch": 0.
|
| 293 |
-
"grad_norm":
|
| 294 |
"learning_rate": 6.020408163265307e-06,
|
| 295 |
-
"loss":
|
| 296 |
"step": 41
|
| 297 |
},
|
| 298 |
{
|
| 299 |
-
"epoch": 0.
|
| 300 |
-
"grad_norm":
|
| 301 |
"learning_rate": 5.918367346938776e-06,
|
| 302 |
-
"loss": 1.
|
| 303 |
"step": 42
|
| 304 |
},
|
| 305 |
{
|
| 306 |
-
"epoch": 0.
|
| 307 |
-
"grad_norm":
|
| 308 |
"learning_rate": 5.816326530612246e-06,
|
| 309 |
-
"loss": 1.
|
| 310 |
"step": 43
|
| 311 |
},
|
| 312 |
{
|
| 313 |
-
"epoch": 0.
|
| 314 |
-
"grad_norm":
|
| 315 |
"learning_rate": 5.7142857142857145e-06,
|
| 316 |
-
"loss": 1.
|
| 317 |
"step": 44
|
| 318 |
},
|
| 319 |
{
|
| 320 |
-
"epoch": 0.
|
| 321 |
-
"grad_norm":
|
| 322 |
"learning_rate": 5.6122448979591834e-06,
|
| 323 |
-
"loss": 1.
|
| 324 |
"step": 45
|
| 325 |
},
|
| 326 |
{
|
| 327 |
-
"epoch": 0.
|
| 328 |
-
"grad_norm":
|
| 329 |
"learning_rate": 5.510204081632653e-06,
|
| 330 |
-
"loss": 1.
|
| 331 |
"step": 46
|
| 332 |
},
|
| 333 |
{
|
| 334 |
-
"epoch": 0.
|
| 335 |
-
"grad_norm":
|
| 336 |
"learning_rate": 5.408163265306123e-06,
|
| 337 |
-
"loss": 1.
|
| 338 |
"step": 47
|
| 339 |
},
|
| 340 |
{
|
| 341 |
-
"epoch": 0.
|
| 342 |
-
"grad_norm":
|
| 343 |
"learning_rate": 5.306122448979593e-06,
|
| 344 |
-
"loss": 1.
|
| 345 |
"step": 48
|
| 346 |
},
|
| 347 |
{
|
| 348 |
-
"epoch": 0.
|
| 349 |
-
"grad_norm": 31.
|
| 350 |
"learning_rate": 5.204081632653062e-06,
|
| 351 |
-
"loss":
|
| 352 |
"step": 49
|
| 353 |
},
|
| 354 |
{
|
| 355 |
-
"epoch": 0.
|
| 356 |
-
"grad_norm":
|
| 357 |
"learning_rate": 5.1020408163265315e-06,
|
| 358 |
-
"loss": 1.
|
| 359 |
"step": 50
|
| 360 |
},
|
| 361 |
{
|
| 362 |
-
"epoch": 0.
|
| 363 |
-
"grad_norm":
|
| 364 |
"learning_rate": 5e-06,
|
| 365 |
-
"loss": 1.
|
| 366 |
"step": 51
|
| 367 |
},
|
| 368 |
{
|
| 369 |
-
"epoch": 0.
|
| 370 |
-
"grad_norm":
|
| 371 |
"learning_rate": 4.897959183673469e-06,
|
| 372 |
-
"loss":
|
| 373 |
"step": 52
|
| 374 |
},
|
| 375 |
{
|
| 376 |
-
"epoch": 0.
|
| 377 |
-
"grad_norm":
|
| 378 |
"learning_rate": 4.795918367346939e-06,
|
| 379 |
-
"loss": 1.
|
| 380 |
"step": 53
|
| 381 |
},
|
| 382 |
{
|
| 383 |
-
"epoch": 0.
|
| 384 |
-
"grad_norm":
|
| 385 |
"learning_rate": 4.693877551020409e-06,
|
| 386 |
-
"loss": 1.
|
| 387 |
"step": 54
|
| 388 |
},
|
| 389 |
{
|
| 390 |
-
"epoch": 0.
|
| 391 |
-
"grad_norm":
|
| 392 |
"learning_rate": 4.591836734693878e-06,
|
| 393 |
-
"loss": 1.
|
| 394 |
"step": 55
|
| 395 |
},
|
| 396 |
{
|
| 397 |
-
"epoch": 0.
|
| 398 |
-
"grad_norm":
|
| 399 |
"learning_rate": 4.489795918367348e-06,
|
| 400 |
-
"loss": 1.
|
| 401 |
"step": 56
|
| 402 |
},
|
| 403 |
{
|
| 404 |
-
"epoch": 0.
|
| 405 |
-
"grad_norm":
|
| 406 |
"learning_rate": 4.3877551020408165e-06,
|
| 407 |
-
"loss":
|
| 408 |
"step": 57
|
| 409 |
},
|
| 410 |
{
|
| 411 |
-
"epoch": 0.
|
| 412 |
-
"grad_norm":
|
| 413 |
"learning_rate": 4.2857142857142855e-06,
|
| 414 |
-
"loss":
|
| 415 |
"step": 58
|
| 416 |
},
|
| 417 |
{
|
| 418 |
-
"epoch": 0.
|
| 419 |
-
"grad_norm":
|
| 420 |
"learning_rate": 4.183673469387755e-06,
|
| 421 |
-
"loss": 1.
|
| 422 |
"step": 59
|
| 423 |
},
|
| 424 |
{
|
| 425 |
-
"epoch": 0.
|
| 426 |
-
"grad_norm":
|
| 427 |
"learning_rate": 4.081632653061225e-06,
|
| 428 |
-
"loss": 1.
|
| 429 |
"step": 60
|
| 430 |
},
|
| 431 |
{
|
| 432 |
-
"epoch": 0.
|
| 433 |
-
"grad_norm":
|
| 434 |
"learning_rate": 3.979591836734694e-06,
|
| 435 |
-
"loss": 1.
|
| 436 |
"step": 61
|
| 437 |
},
|
| 438 |
{
|
| 439 |
-
"epoch": 0.
|
| 440 |
-
"grad_norm":
|
| 441 |
"learning_rate": 3.877551020408164e-06,
|
| 442 |
-
"loss":
|
| 443 |
"step": 62
|
| 444 |
},
|
| 445 |
{
|
| 446 |
-
"epoch": 0.
|
| 447 |
-
"grad_norm":
|
| 448 |
"learning_rate": 3.7755102040816327e-06,
|
| 449 |
-
"loss":
|
| 450 |
"step": 63
|
| 451 |
},
|
| 452 |
{
|
| 453 |
-
"epoch": 0.
|
| 454 |
-
"grad_norm":
|
| 455 |
"learning_rate": 3.6734693877551024e-06,
|
| 456 |
-
"loss": 1.
|
| 457 |
"step": 64
|
| 458 |
},
|
| 459 |
{
|
| 460 |
-
"epoch": 0.
|
| 461 |
-
"grad_norm":
|
| 462 |
"learning_rate": 3.5714285714285718e-06,
|
| 463 |
-
"loss": 1.
|
| 464 |
"step": 65
|
| 465 |
},
|
| 466 |
{
|
| 467 |
-
"epoch": 0.
|
| 468 |
-
"grad_norm":
|
| 469 |
"learning_rate": 3.469387755102041e-06,
|
| 470 |
-
"loss": 1.
|
| 471 |
"step": 66
|
| 472 |
},
|
| 473 |
{
|
| 474 |
-
"epoch": 0.
|
| 475 |
-
"grad_norm":
|
| 476 |
"learning_rate": 3.3673469387755105e-06,
|
| 477 |
-
"loss":
|
| 478 |
"step": 67
|
| 479 |
},
|
| 480 |
{
|
| 481 |
-
"epoch": 0.
|
| 482 |
-
"grad_norm":
|
| 483 |
"learning_rate": 3.2653061224489794e-06,
|
| 484 |
-
"loss": 1.
|
| 485 |
"step": 68
|
| 486 |
},
|
| 487 |
{
|
| 488 |
-
"epoch": 0.
|
| 489 |
-
"grad_norm":
|
| 490 |
"learning_rate": 3.1632653061224496e-06,
|
| 491 |
-
"loss": 1.
|
| 492 |
"step": 69
|
| 493 |
},
|
| 494 |
{
|
| 495 |
-
"epoch": 0.
|
| 496 |
-
"grad_norm":
|
| 497 |
"learning_rate": 3.0612244897959185e-06,
|
| 498 |
-
"loss": 1.
|
| 499 |
"step": 70
|
| 500 |
},
|
| 501 |
{
|
| 502 |
-
"epoch": 0.
|
| 503 |
-
"grad_norm":
|
| 504 |
"learning_rate": 2.959183673469388e-06,
|
| 505 |
-
"loss": 1.
|
| 506 |
"step": 71
|
| 507 |
},
|
| 508 |
{
|
| 509 |
-
"epoch": 0.
|
| 510 |
-
"grad_norm":
|
| 511 |
"learning_rate": 2.8571428571428573e-06,
|
| 512 |
-
"loss": 1.
|
| 513 |
"step": 72
|
| 514 |
},
|
| 515 |
{
|
| 516 |
-
"epoch": 0.
|
| 517 |
-
"grad_norm":
|
| 518 |
"learning_rate": 2.7551020408163266e-06,
|
| 519 |
-
"loss": 1.
|
| 520 |
"step": 73
|
| 521 |
},
|
| 522 |
{
|
| 523 |
-
"epoch": 0.
|
| 524 |
-
"grad_norm":
|
| 525 |
"learning_rate": 2.6530612244897964e-06,
|
| 526 |
-
"loss": 1.
|
| 527 |
"step": 74
|
| 528 |
},
|
| 529 |
{
|
| 530 |
-
"epoch": 0.
|
| 531 |
-
"grad_norm":
|
| 532 |
"learning_rate": 2.5510204081632657e-06,
|
| 533 |
-
"loss":
|
| 534 |
"step": 75
|
| 535 |
},
|
| 536 |
{
|
| 537 |
-
"epoch": 0.
|
| 538 |
-
"grad_norm":
|
| 539 |
"learning_rate": 2.4489795918367347e-06,
|
| 540 |
-
"loss": 1.
|
| 541 |
"step": 76
|
| 542 |
},
|
| 543 |
{
|
| 544 |
-
"epoch": 0.
|
| 545 |
-
"grad_norm":
|
| 546 |
"learning_rate": 2.3469387755102044e-06,
|
| 547 |
-
"loss": 1.
|
| 548 |
"step": 77
|
| 549 |
},
|
| 550 |
{
|
| 551 |
-
"epoch": 0.
|
| 552 |
-
"grad_norm":
|
| 553 |
"learning_rate": 2.244897959183674e-06,
|
| 554 |
-
"loss": 1.
|
| 555 |
"step": 78
|
| 556 |
},
|
| 557 |
{
|
| 558 |
-
"epoch": 0.
|
| 559 |
-
"grad_norm": 26.
|
| 560 |
"learning_rate": 2.1428571428571427e-06,
|
| 561 |
-
"loss":
|
| 562 |
"step": 79
|
| 563 |
},
|
| 564 |
{
|
| 565 |
-
"epoch": 0.
|
| 566 |
-
"grad_norm":
|
| 567 |
"learning_rate": 2.0408163265306125e-06,
|
| 568 |
-
"loss":
|
| 569 |
"step": 80
|
| 570 |
},
|
| 571 |
{
|
| 572 |
-
"epoch": 0.
|
| 573 |
-
"grad_norm":
|
| 574 |
"learning_rate": 1.938775510204082e-06,
|
| 575 |
-
"loss": 1.
|
| 576 |
"step": 81
|
| 577 |
},
|
| 578 |
{
|
| 579 |
-
"epoch": 0.
|
| 580 |
-
"grad_norm":
|
| 581 |
"learning_rate": 1.8367346938775512e-06,
|
| 582 |
-
"loss": 1.
|
| 583 |
"step": 82
|
| 584 |
},
|
| 585 |
{
|
| 586 |
-
"epoch": 0.
|
| 587 |
-
"grad_norm":
|
| 588 |
"learning_rate": 1.7346938775510206e-06,
|
| 589 |
-
"loss": 1.
|
| 590 |
"step": 83
|
| 591 |
},
|
| 592 |
{
|
| 593 |
-
"epoch": 0.
|
| 594 |
-
"grad_norm":
|
| 595 |
"learning_rate": 1.6326530612244897e-06,
|
| 596 |
-
"loss": 1.
|
| 597 |
"step": 84
|
| 598 |
},
|
| 599 |
{
|
| 600 |
-
"epoch": 0.
|
| 601 |
-
"grad_norm":
|
| 602 |
"learning_rate": 1.5306122448979593e-06,
|
| 603 |
-
"loss":
|
| 604 |
"step": 85
|
| 605 |
},
|
| 606 |
{
|
| 607 |
-
"epoch": 0.
|
| 608 |
-
"grad_norm":
|
| 609 |
"learning_rate": 1.4285714285714286e-06,
|
| 610 |
-
"loss":
|
| 611 |
"step": 86
|
| 612 |
},
|
| 613 |
{
|
| 614 |
-
"epoch": 0.
|
| 615 |
-
"grad_norm":
|
| 616 |
"learning_rate": 1.3265306122448982e-06,
|
| 617 |
-
"loss": 1.
|
| 618 |
"step": 87
|
| 619 |
},
|
| 620 |
{
|
| 621 |
-
"epoch": 0.
|
| 622 |
-
"grad_norm":
|
| 623 |
"learning_rate": 1.2244897959183673e-06,
|
| 624 |
-
"loss": 1.
|
| 625 |
"step": 88
|
| 626 |
},
|
| 627 |
{
|
| 628 |
-
"epoch": 0.
|
| 629 |
-
"grad_norm":
|
| 630 |
"learning_rate": 1.122448979591837e-06,
|
| 631 |
-
"loss": 1.
|
| 632 |
"step": 89
|
| 633 |
},
|
| 634 |
{
|
| 635 |
-
"epoch": 0.
|
| 636 |
-
"grad_norm":
|
| 637 |
"learning_rate": 1.0204081632653063e-06,
|
| 638 |
-
"loss":
|
| 639 |
"step": 90
|
| 640 |
},
|
| 641 |
{
|
| 642 |
-
"epoch": 0.
|
| 643 |
-
"grad_norm":
|
| 644 |
"learning_rate": 9.183673469387756e-07,
|
| 645 |
-
"loss": 1.
|
| 646 |
"step": 91
|
| 647 |
},
|
| 648 |
{
|
| 649 |
-
"epoch": 0.
|
| 650 |
-
"grad_norm":
|
| 651 |
"learning_rate": 8.163265306122449e-07,
|
| 652 |
-
"loss": 1.
|
| 653 |
"step": 92
|
| 654 |
},
|
| 655 |
{
|
| 656 |
-
"epoch": 0.
|
| 657 |
-
"grad_norm":
|
| 658 |
"learning_rate": 7.142857142857143e-07,
|
| 659 |
-
"loss": 1.
|
| 660 |
"step": 93
|
| 661 |
},
|
| 662 |
{
|
| 663 |
-
"epoch": 0.
|
| 664 |
-
"grad_norm":
|
| 665 |
"learning_rate": 6.122448979591837e-07,
|
| 666 |
-
"loss": 1.
|
| 667 |
"step": 94
|
| 668 |
},
|
| 669 |
{
|
| 670 |
-
"epoch": 0.
|
| 671 |
-
"grad_norm":
|
| 672 |
"learning_rate": 5.102040816326531e-07,
|
| 673 |
-
"loss": 1.
|
| 674 |
"step": 95
|
| 675 |
},
|
| 676 |
{
|
| 677 |
-
"epoch": 0.
|
| 678 |
-
"grad_norm":
|
| 679 |
"learning_rate": 4.0816326530612243e-07,
|
| 680 |
-
"loss": 1.
|
| 681 |
"step": 96
|
| 682 |
},
|
| 683 |
{
|
| 684 |
-
"epoch": 0.
|
| 685 |
-
"grad_norm":
|
| 686 |
"learning_rate": 3.0612244897959183e-07,
|
| 687 |
-
"loss": 1.
|
| 688 |
"step": 97
|
| 689 |
},
|
| 690 |
{
|
| 691 |
-
"epoch": 0.
|
| 692 |
-
"grad_norm":
|
| 693 |
"learning_rate": 2.0408163265306121e-07,
|
| 694 |
-
"loss": 1.
|
| 695 |
"step": 98
|
| 696 |
},
|
| 697 |
{
|
| 698 |
-
"epoch": 0.
|
| 699 |
-
"grad_norm":
|
| 700 |
"learning_rate": 1.0204081632653061e-07,
|
| 701 |
-
"loss": 1.
|
| 702 |
"step": 99
|
| 703 |
},
|
| 704 |
{
|
| 705 |
-
"epoch": 0.
|
| 706 |
-
"grad_norm":
|
| 707 |
"learning_rate": 0.0,
|
| 708 |
-
"loss": 1.
|
| 709 |
"step": 100
|
| 710 |
},
|
| 711 |
{
|
| 712 |
-
"epoch": 0.
|
| 713 |
"step": 100,
|
| 714 |
"total_flos": 2305515375820800.0,
|
| 715 |
-
"train_loss": 1.
|
| 716 |
-
"train_runtime":
|
| 717 |
-
"train_samples_per_second": 0.
|
| 718 |
-
"train_steps_per_second": 0.
|
| 719 |
}
|
| 720 |
],
|
| 721 |
"logging_steps": 1,
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.8547008547008547,
|
| 5 |
"eval_steps": 500,
|
| 6 |
"global_step": 100,
|
| 7 |
"is_hyper_param_search": false,
|
|
|
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
+
"epoch": 0.008547008547008548,
|
| 13 |
+
"grad_norm": 45.01607894897461,
|
| 14 |
"learning_rate": 5e-06,
|
| 15 |
+
"loss": 1.9033,
|
| 16 |
"step": 1
|
| 17 |
},
|
| 18 |
{
|
| 19 |
+
"epoch": 0.017094017094017096,
|
| 20 |
+
"grad_norm": 28.609533309936523,
|
| 21 |
"learning_rate": 1e-05,
|
| 22 |
+
"loss": 1.8163,
|
| 23 |
"step": 2
|
| 24 |
},
|
| 25 |
{
|
| 26 |
+
"epoch": 0.02564102564102564,
|
| 27 |
+
"grad_norm": 58.36526870727539,
|
| 28 |
"learning_rate": 9.89795918367347e-06,
|
| 29 |
+
"loss": 2.5933,
|
| 30 |
"step": 3
|
| 31 |
},
|
| 32 |
{
|
| 33 |
+
"epoch": 0.03418803418803419,
|
| 34 |
+
"grad_norm": 39.587982177734375,
|
| 35 |
"learning_rate": 9.795918367346939e-06,
|
| 36 |
+
"loss": 1.7991,
|
| 37 |
"step": 4
|
| 38 |
},
|
| 39 |
{
|
| 40 |
+
"epoch": 0.042735042735042736,
|
| 41 |
+
"grad_norm": 90.51757049560547,
|
| 42 |
"learning_rate": 9.693877551020408e-06,
|
| 43 |
+
"loss": 2.6511,
|
| 44 |
"step": 5
|
| 45 |
},
|
| 46 |
{
|
| 47 |
+
"epoch": 0.05128205128205128,
|
| 48 |
+
"grad_norm": 33.15379333496094,
|
| 49 |
"learning_rate": 9.591836734693878e-06,
|
| 50 |
+
"loss": 2.0931,
|
| 51 |
"step": 6
|
| 52 |
},
|
| 53 |
{
|
| 54 |
+
"epoch": 0.05982905982905983,
|
| 55 |
+
"grad_norm": 45.139949798583984,
|
| 56 |
"learning_rate": 9.489795918367348e-06,
|
| 57 |
+
"loss": 2.0398,
|
| 58 |
"step": 7
|
| 59 |
},
|
| 60 |
{
|
| 61 |
+
"epoch": 0.06837606837606838,
|
| 62 |
+
"grad_norm": 63.17831039428711,
|
| 63 |
"learning_rate": 9.387755102040818e-06,
|
| 64 |
+
"loss": 2.0816,
|
| 65 |
"step": 8
|
| 66 |
},
|
| 67 |
{
|
| 68 |
+
"epoch": 0.07692307692307693,
|
| 69 |
+
"grad_norm": 24.13720703125,
|
| 70 |
"learning_rate": 9.285714285714288e-06,
|
| 71 |
+
"loss": 1.3469,
|
| 72 |
"step": 9
|
| 73 |
},
|
| 74 |
{
|
| 75 |
+
"epoch": 0.08547008547008547,
|
| 76 |
+
"grad_norm": 130.69837951660156,
|
| 77 |
"learning_rate": 9.183673469387756e-06,
|
| 78 |
+
"loss": 1.8606,
|
| 79 |
"step": 10
|
| 80 |
},
|
| 81 |
{
|
| 82 |
+
"epoch": 0.09401709401709402,
|
| 83 |
+
"grad_norm": 41.71373748779297,
|
| 84 |
"learning_rate": 9.081632653061225e-06,
|
| 85 |
+
"loss": 2.0109,
|
| 86 |
"step": 11
|
| 87 |
},
|
| 88 |
{
|
| 89 |
+
"epoch": 0.10256410256410256,
|
| 90 |
+
"grad_norm": 41.24937438964844,
|
| 91 |
"learning_rate": 8.979591836734695e-06,
|
| 92 |
+
"loss": 1.929,
|
| 93 |
"step": 12
|
| 94 |
},
|
| 95 |
{
|
| 96 |
+
"epoch": 0.1111111111111111,
|
| 97 |
+
"grad_norm": 51.960426330566406,
|
| 98 |
"learning_rate": 8.877551020408163e-06,
|
| 99 |
+
"loss": 1.866,
|
| 100 |
"step": 13
|
| 101 |
},
|
| 102 |
{
|
| 103 |
+
"epoch": 0.11965811965811966,
|
| 104 |
+
"grad_norm": 42.9802360534668,
|
| 105 |
"learning_rate": 8.775510204081633e-06,
|
| 106 |
+
"loss": 1.8252,
|
| 107 |
"step": 14
|
| 108 |
},
|
| 109 |
{
|
| 110 |
+
"epoch": 0.1282051282051282,
|
| 111 |
+
"grad_norm": 50.58588409423828,
|
| 112 |
"learning_rate": 8.673469387755103e-06,
|
| 113 |
+
"loss": 2.1206,
|
| 114 |
"step": 15
|
| 115 |
},
|
| 116 |
{
|
| 117 |
+
"epoch": 0.13675213675213677,
|
| 118 |
+
"grad_norm": 98.52228546142578,
|
| 119 |
"learning_rate": 8.571428571428571e-06,
|
| 120 |
+
"loss": 2.1851,
|
| 121 |
"step": 16
|
| 122 |
},
|
| 123 |
{
|
| 124 |
+
"epoch": 0.1452991452991453,
|
| 125 |
+
"grad_norm": 187.91094970703125,
|
| 126 |
"learning_rate": 8.469387755102042e-06,
|
| 127 |
+
"loss": 2.2665,
|
| 128 |
"step": 17
|
| 129 |
},
|
| 130 |
{
|
| 131 |
+
"epoch": 0.15384615384615385,
|
| 132 |
+
"grad_norm": 44.042022705078125,
|
| 133 |
"learning_rate": 8.36734693877551e-06,
|
| 134 |
+
"loss": 2.1688,
|
| 135 |
"step": 18
|
| 136 |
},
|
| 137 |
{
|
| 138 |
+
"epoch": 0.1623931623931624,
|
| 139 |
+
"grad_norm": 36.91716766357422,
|
| 140 |
"learning_rate": 8.26530612244898e-06,
|
| 141 |
+
"loss": 2.0421,
|
| 142 |
"step": 19
|
| 143 |
},
|
| 144 |
{
|
| 145 |
+
"epoch": 0.17094017094017094,
|
| 146 |
+
"grad_norm": 27.246334075927734,
|
| 147 |
"learning_rate": 8.16326530612245e-06,
|
| 148 |
+
"loss": 1.9751,
|
| 149 |
"step": 20
|
| 150 |
},
|
| 151 |
{
|
| 152 |
+
"epoch": 0.1794871794871795,
|
| 153 |
+
"grad_norm": 32.8808479309082,
|
| 154 |
"learning_rate": 8.06122448979592e-06,
|
| 155 |
+
"loss": 2.1734,
|
| 156 |
"step": 21
|
| 157 |
},
|
| 158 |
{
|
| 159 |
+
"epoch": 0.18803418803418803,
|
| 160 |
+
"grad_norm": 33.08450698852539,
|
| 161 |
"learning_rate": 7.959183673469388e-06,
|
| 162 |
+
"loss": 2.3046,
|
| 163 |
"step": 22
|
| 164 |
},
|
| 165 |
{
|
| 166 |
+
"epoch": 0.19658119658119658,
|
| 167 |
+
"grad_norm": 34.17445373535156,
|
| 168 |
"learning_rate": 7.857142857142858e-06,
|
| 169 |
+
"loss": 2.122,
|
| 170 |
"step": 23
|
| 171 |
},
|
| 172 |
{
|
| 173 |
+
"epoch": 0.20512820512820512,
|
| 174 |
+
"grad_norm": 55.36966323852539,
|
| 175 |
"learning_rate": 7.755102040816327e-06,
|
| 176 |
+
"loss": 1.6824,
|
| 177 |
"step": 24
|
| 178 |
},
|
| 179 |
{
|
| 180 |
+
"epoch": 0.21367521367521367,
|
| 181 |
+
"grad_norm": 50.512596130371094,
|
| 182 |
"learning_rate": 7.653061224489796e-06,
|
| 183 |
+
"loss": 2.1274,
|
| 184 |
"step": 25
|
| 185 |
},
|
| 186 |
{
|
| 187 |
+
"epoch": 0.2222222222222222,
|
| 188 |
+
"grad_norm": 32.29758834838867,
|
| 189 |
"learning_rate": 7.551020408163265e-06,
|
| 190 |
+
"loss": 1.8956,
|
| 191 |
"step": 26
|
| 192 |
},
|
| 193 |
{
|
| 194 |
+
"epoch": 0.23076923076923078,
|
| 195 |
+
"grad_norm": 26.03915786743164,
|
| 196 |
"learning_rate": 7.448979591836736e-06,
|
| 197 |
+
"loss": 1.8013,
|
| 198 |
"step": 27
|
| 199 |
},
|
| 200 |
{
|
| 201 |
+
"epoch": 0.23931623931623933,
|
| 202 |
+
"grad_norm": 29.17389678955078,
|
| 203 |
"learning_rate": 7.346938775510205e-06,
|
| 204 |
+
"loss": 2.0998,
|
| 205 |
"step": 28
|
| 206 |
},
|
| 207 |
{
|
| 208 |
+
"epoch": 0.24786324786324787,
|
| 209 |
+
"grad_norm": 25.569332122802734,
|
| 210 |
"learning_rate": 7.244897959183675e-06,
|
| 211 |
+
"loss": 0.8649,
|
| 212 |
"step": 29
|
| 213 |
},
|
| 214 |
{
|
| 215 |
+
"epoch": 0.2564102564102564,
|
| 216 |
+
"grad_norm": 24.709575653076172,
|
| 217 |
"learning_rate": 7.1428571428571436e-06,
|
| 218 |
+
"loss": 1.4284,
|
| 219 |
"step": 30
|
| 220 |
},
|
| 221 |
{
|
| 222 |
+
"epoch": 0.26495726495726496,
|
| 223 |
+
"grad_norm": 40.75468063354492,
|
| 224 |
"learning_rate": 7.0408163265306125e-06,
|
| 225 |
+
"loss": 2.4333,
|
| 226 |
"step": 31
|
| 227 |
},
|
| 228 |
{
|
| 229 |
+
"epoch": 0.27350427350427353,
|
| 230 |
+
"grad_norm": 28.9516544342041,
|
| 231 |
"learning_rate": 6.938775510204082e-06,
|
| 232 |
+
"loss": 1.8231,
|
| 233 |
"step": 32
|
| 234 |
},
|
| 235 |
{
|
| 236 |
+
"epoch": 0.28205128205128205,
|
| 237 |
+
"grad_norm": 31.57037925720215,
|
| 238 |
"learning_rate": 6.836734693877551e-06,
|
| 239 |
+
"loss": 1.9139,
|
| 240 |
"step": 33
|
| 241 |
},
|
| 242 |
{
|
| 243 |
+
"epoch": 0.2905982905982906,
|
| 244 |
+
"grad_norm": 28.325578689575195,
|
| 245 |
"learning_rate": 6.734693877551021e-06,
|
| 246 |
+
"loss": 2.2195,
|
| 247 |
"step": 34
|
| 248 |
},
|
| 249 |
{
|
| 250 |
+
"epoch": 0.29914529914529914,
|
| 251 |
+
"grad_norm": 39.064449310302734,
|
| 252 |
"learning_rate": 6.63265306122449e-06,
|
| 253 |
+
"loss": 2.1197,
|
| 254 |
"step": 35
|
| 255 |
},
|
| 256 |
{
|
| 257 |
+
"epoch": 0.3076923076923077,
|
| 258 |
+
"grad_norm": 26.608673095703125,
|
| 259 |
"learning_rate": 6.530612244897959e-06,
|
| 260 |
+
"loss": 1.6402,
|
| 261 |
"step": 36
|
| 262 |
},
|
| 263 |
{
|
| 264 |
+
"epoch": 0.3162393162393162,
|
| 265 |
+
"grad_norm": 24.13069725036621,
|
| 266 |
"learning_rate": 6.4285714285714295e-06,
|
| 267 |
+
"loss": 2.3671,
|
| 268 |
"step": 37
|
| 269 |
},
|
| 270 |
{
|
| 271 |
+
"epoch": 0.3247863247863248,
|
| 272 |
+
"grad_norm": 25.87238311767578,
|
| 273 |
"learning_rate": 6.326530612244899e-06,
|
| 274 |
+
"loss": 1.5874,
|
| 275 |
"step": 38
|
| 276 |
},
|
| 277 |
{
|
| 278 |
+
"epoch": 0.3333333333333333,
|
| 279 |
+
"grad_norm": 29.62607765197754,
|
| 280 |
"learning_rate": 6.224489795918368e-06,
|
| 281 |
+
"loss": 1.6643,
|
| 282 |
"step": 39
|
| 283 |
},
|
| 284 |
{
|
| 285 |
+
"epoch": 0.3418803418803419,
|
| 286 |
+
"grad_norm": 30.726987838745117,
|
| 287 |
"learning_rate": 6.122448979591837e-06,
|
| 288 |
+
"loss": 1.583,
|
| 289 |
"step": 40
|
| 290 |
},
|
| 291 |
{
|
| 292 |
+
"epoch": 0.3504273504273504,
|
| 293 |
+
"grad_norm": 26.581026077270508,
|
| 294 |
"learning_rate": 6.020408163265307e-06,
|
| 295 |
+
"loss": 2.0036,
|
| 296 |
"step": 41
|
| 297 |
},
|
| 298 |
{
|
| 299 |
+
"epoch": 0.358974358974359,
|
| 300 |
+
"grad_norm": 26.827810287475586,
|
| 301 |
"learning_rate": 5.918367346938776e-06,
|
| 302 |
+
"loss": 1.4689,
|
| 303 |
"step": 42
|
| 304 |
},
|
| 305 |
{
|
| 306 |
+
"epoch": 0.36752136752136755,
|
| 307 |
+
"grad_norm": 25.676605224609375,
|
| 308 |
"learning_rate": 5.816326530612246e-06,
|
| 309 |
+
"loss": 1.875,
|
| 310 |
"step": 43
|
| 311 |
},
|
| 312 |
{
|
| 313 |
+
"epoch": 0.37606837606837606,
|
| 314 |
+
"grad_norm": 29.468841552734375,
|
| 315 |
"learning_rate": 5.7142857142857145e-06,
|
| 316 |
+
"loss": 1.5171,
|
| 317 |
"step": 44
|
| 318 |
},
|
| 319 |
{
|
| 320 |
+
"epoch": 0.38461538461538464,
|
| 321 |
+
"grad_norm": 23.370439529418945,
|
| 322 |
"learning_rate": 5.6122448979591834e-06,
|
| 323 |
+
"loss": 1.1047,
|
| 324 |
"step": 45
|
| 325 |
},
|
| 326 |
{
|
| 327 |
+
"epoch": 0.39316239316239315,
|
| 328 |
+
"grad_norm": 24.33774757385254,
|
| 329 |
"learning_rate": 5.510204081632653e-06,
|
| 330 |
+
"loss": 1.6296,
|
| 331 |
"step": 46
|
| 332 |
},
|
| 333 |
{
|
| 334 |
+
"epoch": 0.4017094017094017,
|
| 335 |
+
"grad_norm": 30.425437927246094,
|
| 336 |
"learning_rate": 5.408163265306123e-06,
|
| 337 |
+
"loss": 1.478,
|
| 338 |
"step": 47
|
| 339 |
},
|
| 340 |
{
|
| 341 |
+
"epoch": 0.41025641025641024,
|
| 342 |
+
"grad_norm": 19.86203956604004,
|
| 343 |
"learning_rate": 5.306122448979593e-06,
|
| 344 |
+
"loss": 1.846,
|
| 345 |
"step": 48
|
| 346 |
},
|
| 347 |
{
|
| 348 |
+
"epoch": 0.4188034188034188,
|
| 349 |
+
"grad_norm": 31.836505889892578,
|
| 350 |
"learning_rate": 5.204081632653062e-06,
|
| 351 |
+
"loss": 2.0027,
|
| 352 |
"step": 49
|
| 353 |
},
|
| 354 |
{
|
| 355 |
+
"epoch": 0.42735042735042733,
|
| 356 |
+
"grad_norm": 30.679075241088867,
|
| 357 |
"learning_rate": 5.1020408163265315e-06,
|
| 358 |
+
"loss": 1.6028,
|
| 359 |
"step": 50
|
| 360 |
},
|
| 361 |
{
|
| 362 |
+
"epoch": 0.4358974358974359,
|
| 363 |
+
"grad_norm": 25.78423309326172,
|
| 364 |
"learning_rate": 5e-06,
|
| 365 |
+
"loss": 1.2363,
|
| 366 |
"step": 51
|
| 367 |
},
|
| 368 |
{
|
| 369 |
+
"epoch": 0.4444444444444444,
|
| 370 |
+
"grad_norm": 27.172372817993164,
|
| 371 |
"learning_rate": 4.897959183673469e-06,
|
| 372 |
+
"loss": 1.6288,
|
| 373 |
"step": 52
|
| 374 |
},
|
| 375 |
{
|
| 376 |
+
"epoch": 0.452991452991453,
|
| 377 |
+
"grad_norm": 28.951725006103516,
|
| 378 |
"learning_rate": 4.795918367346939e-06,
|
| 379 |
+
"loss": 1.8334,
|
| 380 |
"step": 53
|
| 381 |
},
|
| 382 |
{
|
| 383 |
+
"epoch": 0.46153846153846156,
|
| 384 |
+
"grad_norm": 46.81833267211914,
|
| 385 |
"learning_rate": 4.693877551020409e-06,
|
| 386 |
+
"loss": 1.1823,
|
| 387 |
"step": 54
|
| 388 |
},
|
| 389 |
{
|
| 390 |
+
"epoch": 0.4700854700854701,
|
| 391 |
+
"grad_norm": 26.226045608520508,
|
| 392 |
"learning_rate": 4.591836734693878e-06,
|
| 393 |
+
"loss": 1.8236,
|
| 394 |
"step": 55
|
| 395 |
},
|
| 396 |
{
|
| 397 |
+
"epoch": 0.47863247863247865,
|
| 398 |
+
"grad_norm": 22.57502555847168,
|
| 399 |
"learning_rate": 4.489795918367348e-06,
|
| 400 |
+
"loss": 1.7485,
|
| 401 |
"step": 56
|
| 402 |
},
|
| 403 |
{
|
| 404 |
+
"epoch": 0.48717948717948717,
|
| 405 |
+
"grad_norm": 38.75823974609375,
|
| 406 |
"learning_rate": 4.3877551020408165e-06,
|
| 407 |
+
"loss": 2.315,
|
| 408 |
"step": 57
|
| 409 |
},
|
| 410 |
{
|
| 411 |
+
"epoch": 0.49572649572649574,
|
| 412 |
+
"grad_norm": 30.443164825439453,
|
| 413 |
"learning_rate": 4.2857142857142855e-06,
|
| 414 |
+
"loss": 1.4465,
|
| 415 |
"step": 58
|
| 416 |
},
|
| 417 |
{
|
| 418 |
+
"epoch": 0.5042735042735043,
|
| 419 |
+
"grad_norm": 30.137744903564453,
|
| 420 |
"learning_rate": 4.183673469387755e-06,
|
| 421 |
+
"loss": 1.8649,
|
| 422 |
"step": 59
|
| 423 |
},
|
| 424 |
{
|
| 425 |
+
"epoch": 0.5128205128205128,
|
| 426 |
+
"grad_norm": 30.10780143737793,
|
| 427 |
"learning_rate": 4.081632653061225e-06,
|
| 428 |
+
"loss": 1.2203,
|
| 429 |
"step": 60
|
| 430 |
},
|
| 431 |
{
|
| 432 |
+
"epoch": 0.5213675213675214,
|
| 433 |
+
"grad_norm": 30.356447219848633,
|
| 434 |
"learning_rate": 3.979591836734694e-06,
|
| 435 |
+
"loss": 1.5023,
|
| 436 |
"step": 61
|
| 437 |
},
|
| 438 |
{
|
| 439 |
+
"epoch": 0.5299145299145299,
|
| 440 |
+
"grad_norm": 23.183887481689453,
|
| 441 |
"learning_rate": 3.877551020408164e-06,
|
| 442 |
+
"loss": 2.1429,
|
| 443 |
"step": 62
|
| 444 |
},
|
| 445 |
{
|
| 446 |
+
"epoch": 0.5384615384615384,
|
| 447 |
+
"grad_norm": 23.465574264526367,
|
| 448 |
"learning_rate": 3.7755102040816327e-06,
|
| 449 |
+
"loss": 2.482,
|
| 450 |
"step": 63
|
| 451 |
},
|
| 452 |
{
|
| 453 |
+
"epoch": 0.5470085470085471,
|
| 454 |
+
"grad_norm": 27.807279586791992,
|
| 455 |
"learning_rate": 3.6734693877551024e-06,
|
| 456 |
+
"loss": 1.5094,
|
| 457 |
"step": 64
|
| 458 |
},
|
| 459 |
{
|
| 460 |
+
"epoch": 0.5555555555555556,
|
| 461 |
+
"grad_norm": 24.647205352783203,
|
| 462 |
"learning_rate": 3.5714285714285718e-06,
|
| 463 |
+
"loss": 1.3724,
|
| 464 |
"step": 65
|
| 465 |
},
|
| 466 |
{
|
| 467 |
+
"epoch": 0.5641025641025641,
|
| 468 |
+
"grad_norm": 23.685543060302734,
|
| 469 |
"learning_rate": 3.469387755102041e-06,
|
| 470 |
+
"loss": 1.5686,
|
| 471 |
"step": 66
|
| 472 |
},
|
| 473 |
{
|
| 474 |
+
"epoch": 0.5726495726495726,
|
| 475 |
+
"grad_norm": 44.072669982910156,
|
| 476 |
"learning_rate": 3.3673469387755105e-06,
|
| 477 |
+
"loss": 2.7541,
|
| 478 |
"step": 67
|
| 479 |
},
|
| 480 |
{
|
| 481 |
+
"epoch": 0.5811965811965812,
|
| 482 |
+
"grad_norm": 29.146209716796875,
|
| 483 |
"learning_rate": 3.2653061224489794e-06,
|
| 484 |
+
"loss": 1.6415,
|
| 485 |
"step": 68
|
| 486 |
},
|
| 487 |
{
|
| 488 |
+
"epoch": 0.5897435897435898,
|
| 489 |
+
"grad_norm": 31.696557998657227,
|
| 490 |
"learning_rate": 3.1632653061224496e-06,
|
| 491 |
+
"loss": 1.7669,
|
| 492 |
"step": 69
|
| 493 |
},
|
| 494 |
{
|
| 495 |
+
"epoch": 0.5982905982905983,
|
| 496 |
+
"grad_norm": 18.153284072875977,
|
| 497 |
"learning_rate": 3.0612244897959185e-06,
|
| 498 |
+
"loss": 1.763,
|
| 499 |
"step": 70
|
| 500 |
},
|
| 501 |
{
|
| 502 |
+
"epoch": 0.6068376068376068,
|
| 503 |
+
"grad_norm": 17.51633644104004,
|
| 504 |
"learning_rate": 2.959183673469388e-06,
|
| 505 |
+
"loss": 1.7882,
|
| 506 |
"step": 71
|
| 507 |
},
|
| 508 |
{
|
| 509 |
+
"epoch": 0.6153846153846154,
|
| 510 |
+
"grad_norm": 21.281356811523438,
|
| 511 |
"learning_rate": 2.8571428571428573e-06,
|
| 512 |
+
"loss": 1.2255,
|
| 513 |
"step": 72
|
| 514 |
},
|
| 515 |
{
|
| 516 |
+
"epoch": 0.6239316239316239,
|
| 517 |
+
"grad_norm": 30.624502182006836,
|
| 518 |
"learning_rate": 2.7551020408163266e-06,
|
| 519 |
+
"loss": 1.2902,
|
| 520 |
"step": 73
|
| 521 |
},
|
| 522 |
{
|
| 523 |
+
"epoch": 0.6324786324786325,
|
| 524 |
+
"grad_norm": 32.38172149658203,
|
| 525 |
"learning_rate": 2.6530612244897964e-06,
|
| 526 |
+
"loss": 1.8834,
|
| 527 |
"step": 74
|
| 528 |
},
|
| 529 |
{
|
| 530 |
+
"epoch": 0.6410256410256411,
|
| 531 |
+
"grad_norm": 29.069534301757812,
|
| 532 |
"learning_rate": 2.5510204081632657e-06,
|
| 533 |
+
"loss": 1.7467,
|
| 534 |
"step": 75
|
| 535 |
},
|
| 536 |
{
|
| 537 |
+
"epoch": 0.6495726495726496,
|
| 538 |
+
"grad_norm": 56.46368408203125,
|
| 539 |
"learning_rate": 2.4489795918367347e-06,
|
| 540 |
+
"loss": 1.6087,
|
| 541 |
"step": 76
|
| 542 |
},
|
| 543 |
{
|
| 544 |
+
"epoch": 0.6581196581196581,
|
| 545 |
+
"grad_norm": 29.552095413208008,
|
| 546 |
"learning_rate": 2.3469387755102044e-06,
|
| 547 |
+
"loss": 1.4008,
|
| 548 |
"step": 77
|
| 549 |
},
|
| 550 |
{
|
| 551 |
+
"epoch": 0.6666666666666666,
|
| 552 |
+
"grad_norm": 30.940513610839844,
|
| 553 |
"learning_rate": 2.244897959183674e-06,
|
| 554 |
+
"loss": 1.1384,
|
| 555 |
"step": 78
|
| 556 |
},
|
| 557 |
{
|
| 558 |
+
"epoch": 0.6752136752136753,
|
| 559 |
+
"grad_norm": 26.42818832397461,
|
| 560 |
"learning_rate": 2.1428571428571427e-06,
|
| 561 |
+
"loss": 1.6209,
|
| 562 |
"step": 79
|
| 563 |
},
|
| 564 |
{
|
| 565 |
+
"epoch": 0.6837606837606838,
|
| 566 |
+
"grad_norm": 23.604707717895508,
|
| 567 |
"learning_rate": 2.0408163265306125e-06,
|
| 568 |
+
"loss": 1.2335,
|
| 569 |
"step": 80
|
| 570 |
},
|
| 571 |
{
|
| 572 |
+
"epoch": 0.6923076923076923,
|
| 573 |
+
"grad_norm": 33.276771545410156,
|
| 574 |
"learning_rate": 1.938775510204082e-06,
|
| 575 |
+
"loss": 1.8955,
|
| 576 |
"step": 81
|
| 577 |
},
|
| 578 |
{
|
| 579 |
+
"epoch": 0.7008547008547008,
|
| 580 |
+
"grad_norm": 17.27667808532715,
|
| 581 |
"learning_rate": 1.8367346938775512e-06,
|
| 582 |
+
"loss": 1.8708,
|
| 583 |
"step": 82
|
| 584 |
},
|
| 585 |
{
|
| 586 |
+
"epoch": 0.7094017094017094,
|
| 587 |
+
"grad_norm": 34.4437255859375,
|
| 588 |
"learning_rate": 1.7346938775510206e-06,
|
| 589 |
+
"loss": 1.8406,
|
| 590 |
"step": 83
|
| 591 |
},
|
| 592 |
{
|
| 593 |
+
"epoch": 0.717948717948718,
|
| 594 |
+
"grad_norm": 27.164016723632812,
|
| 595 |
"learning_rate": 1.6326530612244897e-06,
|
| 596 |
+
"loss": 1.6318,
|
| 597 |
"step": 84
|
| 598 |
},
|
| 599 |
{
|
| 600 |
+
"epoch": 0.7264957264957265,
|
| 601 |
+
"grad_norm": 30.789165496826172,
|
| 602 |
"learning_rate": 1.5306122448979593e-06,
|
| 603 |
+
"loss": 2.1226,
|
| 604 |
"step": 85
|
| 605 |
},
|
| 606 |
{
|
| 607 |
+
"epoch": 0.7350427350427351,
|
| 608 |
+
"grad_norm": 24.16193962097168,
|
| 609 |
"learning_rate": 1.4285714285714286e-06,
|
| 610 |
+
"loss": 0.8153,
|
| 611 |
"step": 86
|
| 612 |
},
|
| 613 |
{
|
| 614 |
+
"epoch": 0.7435897435897436,
|
| 615 |
+
"grad_norm": 19.538379669189453,
|
| 616 |
"learning_rate": 1.3265306122448982e-06,
|
| 617 |
+
"loss": 1.9163,
|
| 618 |
"step": 87
|
| 619 |
},
|
| 620 |
{
|
| 621 |
+
"epoch": 0.7521367521367521,
|
| 622 |
+
"grad_norm": 26.798044204711914,
|
| 623 |
"learning_rate": 1.2244897959183673e-06,
|
| 624 |
+
"loss": 1.7253,
|
| 625 |
"step": 88
|
| 626 |
},
|
| 627 |
{
|
| 628 |
+
"epoch": 0.7606837606837606,
|
| 629 |
+
"grad_norm": 26.220504760742188,
|
| 630 |
"learning_rate": 1.122448979591837e-06,
|
| 631 |
+
"loss": 1.2386,
|
| 632 |
"step": 89
|
| 633 |
},
|
| 634 |
{
|
| 635 |
+
"epoch": 0.7692307692307693,
|
| 636 |
+
"grad_norm": 27.829275131225586,
|
| 637 |
"learning_rate": 1.0204081632653063e-06,
|
| 638 |
+
"loss": 2.0316,
|
| 639 |
"step": 90
|
| 640 |
},
|
| 641 |
{
|
| 642 |
+
"epoch": 0.7777777777777778,
|
| 643 |
+
"grad_norm": 19.041147232055664,
|
| 644 |
"learning_rate": 9.183673469387756e-07,
|
| 645 |
+
"loss": 1.8049,
|
| 646 |
"step": 91
|
| 647 |
},
|
| 648 |
{
|
| 649 |
+
"epoch": 0.7863247863247863,
|
| 650 |
+
"grad_norm": 18.105703353881836,
|
| 651 |
"learning_rate": 8.163265306122449e-07,
|
| 652 |
+
"loss": 1.5467,
|
| 653 |
"step": 92
|
| 654 |
},
|
| 655 |
{
|
| 656 |
+
"epoch": 0.7948717948717948,
|
| 657 |
+
"grad_norm": 25.5799560546875,
|
| 658 |
"learning_rate": 7.142857142857143e-07,
|
| 659 |
+
"loss": 1.5628,
|
| 660 |
"step": 93
|
| 661 |
},
|
| 662 |
{
|
| 663 |
+
"epoch": 0.8034188034188035,
|
| 664 |
+
"grad_norm": 24.031326293945312,
|
| 665 |
"learning_rate": 6.122448979591837e-07,
|
| 666 |
+
"loss": 1.8895,
|
| 667 |
"step": 94
|
| 668 |
},
|
| 669 |
{
|
| 670 |
+
"epoch": 0.811965811965812,
|
| 671 |
+
"grad_norm": 14.778085708618164,
|
| 672 |
"learning_rate": 5.102040816326531e-07,
|
| 673 |
+
"loss": 1.4871,
|
| 674 |
"step": 95
|
| 675 |
},
|
| 676 |
{
|
| 677 |
+
"epoch": 0.8205128205128205,
|
| 678 |
+
"grad_norm": 24.120147705078125,
|
| 679 |
"learning_rate": 4.0816326530612243e-07,
|
| 680 |
+
"loss": 1.507,
|
| 681 |
"step": 96
|
| 682 |
},
|
| 683 |
{
|
| 684 |
+
"epoch": 0.8290598290598291,
|
| 685 |
+
"grad_norm": 19.402923583984375,
|
| 686 |
"learning_rate": 3.0612244897959183e-07,
|
| 687 |
+
"loss": 1.2183,
|
| 688 |
"step": 97
|
| 689 |
},
|
| 690 |
{
|
| 691 |
+
"epoch": 0.8376068376068376,
|
| 692 |
+
"grad_norm": 26.27826499938965,
|
| 693 |
"learning_rate": 2.0408163265306121e-07,
|
| 694 |
+
"loss": 1.7308,
|
| 695 |
"step": 98
|
| 696 |
},
|
| 697 |
{
|
| 698 |
+
"epoch": 0.8461538461538461,
|
| 699 |
+
"grad_norm": 23.33223533630371,
|
| 700 |
"learning_rate": 1.0204081632653061e-07,
|
| 701 |
+
"loss": 1.3037,
|
| 702 |
"step": 99
|
| 703 |
},
|
| 704 |
{
|
| 705 |
+
"epoch": 0.8547008547008547,
|
| 706 |
+
"grad_norm": 24.819562911987305,
|
| 707 |
"learning_rate": 0.0,
|
| 708 |
+
"loss": 1.2484,
|
| 709 |
"step": 100
|
| 710 |
},
|
| 711 |
{
|
| 712 |
+
"epoch": 0.8547008547008547,
|
| 713 |
"step": 100,
|
| 714 |
"total_flos": 2305515375820800.0,
|
| 715 |
+
"train_loss": 1.7708107882738113,
|
| 716 |
+
"train_runtime": 5686.4644,
|
| 717 |
+
"train_samples_per_second": 0.018,
|
| 718 |
+
"train_steps_per_second": 0.018
|
| 719 |
}
|
| 720 |
],
|
| 721 |
"logging_steps": 1,
|