Upload folder using huggingface_hub
Browse files- rng_state.pth +1 -1
- scheduler.pt +1 -1
- trainer_state.json +2552 -377
rng_state.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 14645
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:cb8b281d3670780618442404eb45c98293a0d43e46f8e5ca3eb87cb4663d60e7
|
| 3 |
size 14645
|
scheduler.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1465
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5fa59c9185701213ec25411dab80244c017dba754eeac7bca5fb0c59c13e7e9c
|
| 3 |
size 1465
|
trainer_state.json
CHANGED
|
@@ -1,844 +1,3019 @@
|
|
| 1 |
{
|
| 2 |
-
"best_global_step":
|
| 3 |
-
"best_metric": 0.
|
| 4 |
-
"best_model_checkpoint": "/content/drive/MyDrive/lora_model/outputs/task15_microsoft/Phi-4-mini-instruct/checkpoint-
|
| 5 |
-
"epoch":
|
| 6 |
"eval_steps": 1,
|
| 7 |
-
"global_step":
|
| 8 |
"is_hyper_param_search": false,
|
| 9 |
"is_local_process_zero": true,
|
| 10 |
"is_world_process_zero": true,
|
| 11 |
"log_history": [
|
| 12 |
{
|
| 13 |
"epoch": 0.05263157894736842,
|
| 14 |
-
"grad_norm":
|
| 15 |
"learning_rate": 0.0,
|
| 16 |
-
"loss": 3.
|
| 17 |
"step": 1
|
| 18 |
},
|
| 19 |
{
|
| 20 |
"epoch": 0.05263157894736842,
|
| 21 |
-
"eval_loss": 3.
|
| 22 |
-
"eval_runtime":
|
| 23 |
-
"eval_samples_per_second":
|
| 24 |
-
"eval_steps_per_second":
|
| 25 |
"step": 1
|
| 26 |
},
|
| 27 |
{
|
| 28 |
"epoch": 0.10526315789473684,
|
| 29 |
-
"grad_norm":
|
| 30 |
"learning_rate": 3.3333333333333335e-05,
|
| 31 |
-
"loss": 3.
|
| 32 |
"step": 2
|
| 33 |
},
|
| 34 |
{
|
| 35 |
"epoch": 0.10526315789473684,
|
| 36 |
-
"eval_loss":
|
| 37 |
-
"eval_runtime":
|
| 38 |
-
"eval_samples_per_second":
|
| 39 |
-
"eval_steps_per_second":
|
| 40 |
"step": 2
|
| 41 |
},
|
| 42 |
{
|
| 43 |
"epoch": 0.15789473684210525,
|
| 44 |
-
"grad_norm":
|
| 45 |
"learning_rate": 6.666666666666667e-05,
|
| 46 |
-
"loss": 2.
|
| 47 |
"step": 3
|
| 48 |
},
|
| 49 |
{
|
| 50 |
"epoch": 0.15789473684210525,
|
| 51 |
-
"eval_loss": 2.
|
| 52 |
-
"eval_runtime":
|
| 53 |
-
"eval_samples_per_second":
|
| 54 |
-
"eval_steps_per_second":
|
| 55 |
"step": 3
|
| 56 |
},
|
| 57 |
{
|
| 58 |
"epoch": 0.21052631578947367,
|
| 59 |
-
"grad_norm":
|
| 60 |
"learning_rate": 0.0001,
|
| 61 |
-
"loss": 2.
|
| 62 |
"step": 4
|
| 63 |
},
|
| 64 |
{
|
| 65 |
"epoch": 0.21052631578947367,
|
| 66 |
-
"eval_loss":
|
| 67 |
-
"eval_runtime":
|
| 68 |
-
"eval_samples_per_second":
|
| 69 |
-
"eval_steps_per_second":
|
| 70 |
"step": 4
|
| 71 |
},
|
| 72 |
{
|
| 73 |
"epoch": 0.2631578947368421,
|
| 74 |
-
"grad_norm":
|
| 75 |
"learning_rate": 0.00013333333333333334,
|
| 76 |
-
"loss":
|
| 77 |
"step": 5
|
| 78 |
},
|
| 79 |
{
|
| 80 |
"epoch": 0.2631578947368421,
|
| 81 |
-
"eval_loss":
|
| 82 |
-
"eval_runtime":
|
| 83 |
-
"eval_samples_per_second":
|
| 84 |
-
"eval_steps_per_second":
|
| 85 |
"step": 5
|
| 86 |
},
|
| 87 |
{
|
| 88 |
"epoch": 0.3157894736842105,
|
| 89 |
-
"grad_norm":
|
| 90 |
"learning_rate": 0.00016666666666666666,
|
| 91 |
-
"loss":
|
| 92 |
"step": 6
|
| 93 |
},
|
| 94 |
{
|
| 95 |
"epoch": 0.3157894736842105,
|
| 96 |
-
"eval_loss":
|
| 97 |
-
"eval_runtime":
|
| 98 |
-
"eval_samples_per_second":
|
| 99 |
-
"eval_steps_per_second":
|
| 100 |
"step": 6
|
| 101 |
},
|
| 102 |
{
|
| 103 |
"epoch": 0.3684210526315789,
|
| 104 |
-
"grad_norm":
|
| 105 |
"learning_rate": 0.0002,
|
| 106 |
-
"loss":
|
| 107 |
"step": 7
|
| 108 |
},
|
| 109 |
{
|
| 110 |
"epoch": 0.3684210526315789,
|
| 111 |
-
"eval_loss":
|
| 112 |
-
"eval_runtime":
|
| 113 |
-
"eval_samples_per_second":
|
| 114 |
-
"eval_steps_per_second":
|
| 115 |
"step": 7
|
| 116 |
},
|
| 117 |
{
|
| 118 |
"epoch": 0.42105263157894735,
|
| 119 |
-
"grad_norm":
|
| 120 |
"learning_rate": 0.00023333333333333333,
|
| 121 |
-
"loss":
|
| 122 |
"step": 8
|
| 123 |
},
|
| 124 |
{
|
| 125 |
"epoch": 0.42105263157894735,
|
| 126 |
-
"eval_loss": 1.
|
| 127 |
-
"eval_runtime":
|
| 128 |
-
"eval_samples_per_second":
|
| 129 |
-
"eval_steps_per_second":
|
| 130 |
"step": 8
|
| 131 |
},
|
| 132 |
{
|
| 133 |
"epoch": 0.47368421052631576,
|
| 134 |
-
"grad_norm":
|
| 135 |
"learning_rate": 0.0002666666666666667,
|
| 136 |
-
"loss": 1.
|
| 137 |
"step": 9
|
| 138 |
},
|
| 139 |
{
|
| 140 |
"epoch": 0.47368421052631576,
|
| 141 |
-
"eval_loss": 1.
|
| 142 |
-
"eval_runtime":
|
| 143 |
-
"eval_samples_per_second":
|
| 144 |
-
"eval_steps_per_second":
|
| 145 |
"step": 9
|
| 146 |
},
|
| 147 |
{
|
| 148 |
"epoch": 0.5263157894736842,
|
| 149 |
-
"grad_norm":
|
| 150 |
"learning_rate": 0.0003,
|
| 151 |
-
"loss": 1.
|
| 152 |
"step": 10
|
| 153 |
},
|
| 154 |
{
|
| 155 |
"epoch": 0.5263157894736842,
|
| 156 |
-
"eval_loss": 1.
|
| 157 |
-
"eval_runtime":
|
| 158 |
-
"eval_samples_per_second":
|
| 159 |
-
"eval_steps_per_second":
|
| 160 |
"step": 10
|
| 161 |
},
|
| 162 |
{
|
| 163 |
"epoch": 0.5789473684210527,
|
| 164 |
-
"grad_norm":
|
| 165 |
"learning_rate": 0.0003333333333333333,
|
| 166 |
-
"loss": 1.
|
| 167 |
"step": 11
|
| 168 |
},
|
| 169 |
{
|
| 170 |
"epoch": 0.5789473684210527,
|
| 171 |
-
"eval_loss": 1.
|
| 172 |
-
"eval_runtime":
|
| 173 |
-
"eval_samples_per_second":
|
| 174 |
-
"eval_steps_per_second":
|
| 175 |
"step": 11
|
| 176 |
},
|
| 177 |
{
|
| 178 |
"epoch": 0.631578947368421,
|
| 179 |
-
"grad_norm":
|
| 180 |
"learning_rate": 0.00036666666666666667,
|
| 181 |
-
"loss": 1.
|
| 182 |
"step": 12
|
| 183 |
},
|
| 184 |
{
|
| 185 |
"epoch": 0.631578947368421,
|
| 186 |
-
"eval_loss": 1.
|
| 187 |
-
"eval_runtime":
|
| 188 |
-
"eval_samples_per_second":
|
| 189 |
-
"eval_steps_per_second":
|
| 190 |
"step": 12
|
| 191 |
},
|
| 192 |
{
|
| 193 |
"epoch": 0.6842105263157895,
|
| 194 |
-
"grad_norm":
|
| 195 |
"learning_rate": 0.0004,
|
| 196 |
-
"loss": 1.
|
| 197 |
"step": 13
|
| 198 |
},
|
| 199 |
{
|
| 200 |
"epoch": 0.6842105263157895,
|
| 201 |
-
"eval_loss": 1.
|
| 202 |
-
"eval_runtime":
|
| 203 |
-
"eval_samples_per_second":
|
| 204 |
-
"eval_steps_per_second":
|
| 205 |
"step": 13
|
| 206 |
},
|
| 207 |
{
|
| 208 |
"epoch": 0.7368421052631579,
|
| 209 |
-
"grad_norm":
|
| 210 |
"learning_rate": 0.00043333333333333337,
|
| 211 |
-
"loss": 1.
|
| 212 |
"step": 14
|
| 213 |
},
|
| 214 |
{
|
| 215 |
"epoch": 0.7368421052631579,
|
| 216 |
-
"eval_loss": 1.
|
| 217 |
-
"eval_runtime":
|
| 218 |
-
"eval_samples_per_second":
|
| 219 |
-
"eval_steps_per_second":
|
| 220 |
"step": 14
|
| 221 |
},
|
| 222 |
{
|
| 223 |
"epoch": 0.7894736842105263,
|
| 224 |
-
"grad_norm":
|
| 225 |
"learning_rate": 0.00046666666666666666,
|
| 226 |
-
"loss": 1.
|
| 227 |
"step": 15
|
| 228 |
},
|
| 229 |
{
|
| 230 |
"epoch": 0.7894736842105263,
|
| 231 |
-
"eval_loss": 1.
|
| 232 |
-
"eval_runtime":
|
| 233 |
-
"eval_samples_per_second":
|
| 234 |
-
"eval_steps_per_second":
|
| 235 |
"step": 15
|
| 236 |
},
|
| 237 |
{
|
| 238 |
"epoch": 0.8421052631578947,
|
| 239 |
-
"grad_norm":
|
| 240 |
"learning_rate": 0.0005,
|
| 241 |
-
"loss": 1.
|
| 242 |
"step": 16
|
| 243 |
},
|
| 244 |
{
|
| 245 |
"epoch": 0.8421052631578947,
|
| 246 |
-
"eval_loss":
|
| 247 |
-
"eval_runtime":
|
| 248 |
-
"eval_samples_per_second":
|
| 249 |
-
"eval_steps_per_second":
|
| 250 |
"step": 16
|
| 251 |
},
|
| 252 |
{
|
| 253 |
"epoch": 0.8947368421052632,
|
| 254 |
-
"grad_norm":
|
| 255 |
-
"learning_rate": 0.
|
| 256 |
-
"loss": 1.
|
| 257 |
"step": 17
|
| 258 |
},
|
| 259 |
{
|
| 260 |
"epoch": 0.8947368421052632,
|
| 261 |
-
"eval_loss":
|
| 262 |
-
"eval_runtime":
|
| 263 |
-
"eval_samples_per_second":
|
| 264 |
-
"eval_steps_per_second":
|
| 265 |
"step": 17
|
| 266 |
},
|
| 267 |
{
|
| 268 |
"epoch": 0.9473684210526315,
|
| 269 |
-
"grad_norm":
|
| 270 |
-
"learning_rate": 0.
|
| 271 |
-
"loss": 1.
|
| 272 |
"step": 18
|
| 273 |
},
|
| 274 |
{
|
| 275 |
"epoch": 0.9473684210526315,
|
| 276 |
-
"eval_loss": 1.
|
| 277 |
-
"eval_runtime":
|
| 278 |
-
"eval_samples_per_second":
|
| 279 |
-
"eval_steps_per_second":
|
| 280 |
"step": 18
|
| 281 |
},
|
| 282 |
{
|
| 283 |
"epoch": 1.0,
|
| 284 |
-
"grad_norm":
|
| 285 |
-
"learning_rate": 0.
|
| 286 |
-
"loss": 1.
|
| 287 |
"step": 19
|
| 288 |
},
|
| 289 |
{
|
| 290 |
"epoch": 1.0,
|
| 291 |
-
"eval_loss":
|
| 292 |
-
"eval_runtime":
|
| 293 |
-
"eval_samples_per_second":
|
| 294 |
-
"eval_steps_per_second":
|
| 295 |
"step": 19
|
| 296 |
},
|
| 297 |
{
|
| 298 |
"epoch": 1.0526315789473684,
|
| 299 |
-
"grad_norm":
|
| 300 |
-
"learning_rate": 0.
|
| 301 |
-
"loss":
|
| 302 |
"step": 20
|
| 303 |
},
|
| 304 |
{
|
| 305 |
"epoch": 1.0526315789473684,
|
| 306 |
-
"eval_loss":
|
| 307 |
-
"eval_runtime":
|
| 308 |
-
"eval_samples_per_second":
|
| 309 |
-
"eval_steps_per_second":
|
| 310 |
"step": 20
|
| 311 |
},
|
| 312 |
{
|
| 313 |
"epoch": 1.1052631578947367,
|
| 314 |
-
"grad_norm":
|
| 315 |
-
"learning_rate": 0.
|
| 316 |
-
"loss":
|
| 317 |
"step": 21
|
| 318 |
},
|
| 319 |
{
|
| 320 |
"epoch": 1.1052631578947367,
|
| 321 |
-
"eval_loss": 0.
|
| 322 |
-
"eval_runtime":
|
| 323 |
-
"eval_samples_per_second":
|
| 324 |
-
"eval_steps_per_second":
|
| 325 |
"step": 21
|
| 326 |
},
|
| 327 |
{
|
| 328 |
"epoch": 1.1578947368421053,
|
| 329 |
-
"grad_norm":
|
| 330 |
-
"learning_rate": 0.
|
| 331 |
-
"loss": 1.
|
| 332 |
"step": 22
|
| 333 |
},
|
| 334 |
{
|
| 335 |
"epoch": 1.1578947368421053,
|
| 336 |
-
"eval_loss": 0.
|
| 337 |
-
"eval_runtime":
|
| 338 |
-
"eval_samples_per_second":
|
| 339 |
-
"eval_steps_per_second":
|
| 340 |
"step": 22
|
| 341 |
},
|
| 342 |
{
|
| 343 |
"epoch": 1.2105263157894737,
|
| 344 |
-
"grad_norm":
|
| 345 |
-
"learning_rate": 0.
|
| 346 |
-
"loss": 0.
|
| 347 |
"step": 23
|
| 348 |
},
|
| 349 |
{
|
| 350 |
"epoch": 1.2105263157894737,
|
| 351 |
-
"eval_loss": 0.
|
| 352 |
-
"eval_runtime":
|
| 353 |
-
"eval_samples_per_second":
|
| 354 |
-
"eval_steps_per_second":
|
| 355 |
"step": 23
|
| 356 |
},
|
| 357 |
{
|
| 358 |
"epoch": 1.263157894736842,
|
| 359 |
-
"grad_norm":
|
| 360 |
-
"learning_rate": 0.
|
| 361 |
-
"loss":
|
| 362 |
"step": 24
|
| 363 |
},
|
| 364 |
{
|
| 365 |
"epoch": 1.263157894736842,
|
| 366 |
-
"eval_loss":
|
| 367 |
-
"eval_runtime":
|
| 368 |
-
"eval_samples_per_second":
|
| 369 |
-
"eval_steps_per_second":
|
| 370 |
"step": 24
|
| 371 |
},
|
| 372 |
{
|
| 373 |
"epoch": 1.3157894736842106,
|
| 374 |
-
"grad_norm":
|
| 375 |
-
"learning_rate": 0.
|
| 376 |
-
"loss":
|
| 377 |
"step": 25
|
| 378 |
},
|
| 379 |
{
|
| 380 |
"epoch": 1.3157894736842106,
|
| 381 |
-
"eval_loss": 0.
|
| 382 |
-
"eval_runtime":
|
| 383 |
-
"eval_samples_per_second":
|
| 384 |
-
"eval_steps_per_second":
|
| 385 |
"step": 25
|
| 386 |
},
|
| 387 |
{
|
| 388 |
"epoch": 1.368421052631579,
|
| 389 |
-
"grad_norm":
|
| 390 |
-
"learning_rate": 0.
|
| 391 |
-
"loss":
|
| 392 |
"step": 26
|
| 393 |
},
|
| 394 |
{
|
| 395 |
"epoch": 1.368421052631579,
|
| 396 |
-
"eval_loss": 0.
|
| 397 |
-
"eval_runtime":
|
| 398 |
-
"eval_samples_per_second":
|
| 399 |
-
"eval_steps_per_second":
|
| 400 |
"step": 26
|
| 401 |
},
|
| 402 |
{
|
| 403 |
"epoch": 1.4210526315789473,
|
| 404 |
-
"grad_norm":
|
| 405 |
-
"learning_rate": 0.
|
| 406 |
-
"loss":
|
| 407 |
"step": 27
|
| 408 |
},
|
| 409 |
{
|
| 410 |
"epoch": 1.4210526315789473,
|
| 411 |
-
"eval_loss": 0.
|
| 412 |
-
"eval_runtime":
|
| 413 |
-
"eval_samples_per_second":
|
| 414 |
-
"eval_steps_per_second":
|
| 415 |
"step": 27
|
| 416 |
},
|
| 417 |
{
|
| 418 |
"epoch": 1.4736842105263157,
|
| 419 |
-
"grad_norm":
|
| 420 |
-
"learning_rate": 0.
|
| 421 |
-
"loss": 0.
|
| 422 |
"step": 28
|
| 423 |
},
|
| 424 |
{
|
| 425 |
"epoch": 1.4736842105263157,
|
| 426 |
-
"eval_loss": 0.
|
| 427 |
-
"eval_runtime":
|
| 428 |
-
"eval_samples_per_second":
|
| 429 |
-
"eval_steps_per_second":
|
| 430 |
"step": 28
|
| 431 |
},
|
| 432 |
{
|
| 433 |
"epoch": 1.526315789473684,
|
| 434 |
-
"grad_norm":
|
| 435 |
-
"learning_rate": 0.
|
| 436 |
-
"loss": 1.
|
| 437 |
"step": 29
|
| 438 |
},
|
| 439 |
{
|
| 440 |
"epoch": 1.526315789473684,
|
| 441 |
-
"eval_loss": 0.
|
| 442 |
-
"eval_runtime":
|
| 443 |
-
"eval_samples_per_second":
|
| 444 |
-
"eval_steps_per_second":
|
| 445 |
"step": 29
|
| 446 |
},
|
| 447 |
{
|
| 448 |
"epoch": 1.5789473684210527,
|
| 449 |
-
"grad_norm":
|
| 450 |
-
"learning_rate": 0.
|
| 451 |
-
"loss": 0.
|
| 452 |
"step": 30
|
| 453 |
},
|
| 454 |
{
|
| 455 |
"epoch": 1.5789473684210527,
|
| 456 |
-
"eval_loss": 0.
|
| 457 |
-
"eval_runtime":
|
| 458 |
-
"eval_samples_per_second":
|
| 459 |
-
"eval_steps_per_second":
|
| 460 |
"step": 30
|
| 461 |
},
|
| 462 |
{
|
| 463 |
"epoch": 1.631578947368421,
|
| 464 |
-
"grad_norm":
|
| 465 |
-
"learning_rate": 0.
|
| 466 |
-
"loss":
|
| 467 |
"step": 31
|
| 468 |
},
|
| 469 |
{
|
| 470 |
"epoch": 1.631578947368421,
|
| 471 |
-
"eval_loss": 0.
|
| 472 |
-
"eval_runtime":
|
| 473 |
-
"eval_samples_per_second":
|
| 474 |
-
"eval_steps_per_second":
|
| 475 |
"step": 31
|
| 476 |
},
|
| 477 |
{
|
| 478 |
"epoch": 1.6842105263157894,
|
| 479 |
-
"grad_norm":
|
| 480 |
-
"learning_rate": 0.
|
| 481 |
-
"loss": 0.
|
| 482 |
"step": 32
|
| 483 |
},
|
| 484 |
{
|
| 485 |
"epoch": 1.6842105263157894,
|
| 486 |
-
"eval_loss": 0.
|
| 487 |
-
"eval_runtime":
|
| 488 |
-
"eval_samples_per_second":
|
| 489 |
-
"eval_steps_per_second":
|
| 490 |
"step": 32
|
| 491 |
},
|
| 492 |
{
|
| 493 |
"epoch": 1.736842105263158,
|
| 494 |
-
"grad_norm":
|
| 495 |
-
"learning_rate": 0.
|
| 496 |
-
"loss": 1.
|
| 497 |
"step": 33
|
| 498 |
},
|
| 499 |
{
|
| 500 |
"epoch": 1.736842105263158,
|
| 501 |
-
"eval_loss": 0.
|
| 502 |
-
"eval_runtime":
|
| 503 |
-
"eval_samples_per_second":
|
| 504 |
-
"eval_steps_per_second":
|
| 505 |
"step": 33
|
| 506 |
},
|
| 507 |
{
|
| 508 |
"epoch": 1.7894736842105263,
|
| 509 |
-
"grad_norm":
|
| 510 |
-
"learning_rate": 0.
|
| 511 |
-
"loss":
|
| 512 |
"step": 34
|
| 513 |
},
|
| 514 |
{
|
| 515 |
"epoch": 1.7894736842105263,
|
| 516 |
-
"eval_loss": 0.
|
| 517 |
-
"eval_runtime":
|
| 518 |
-
"eval_samples_per_second":
|
| 519 |
-
"eval_steps_per_second":
|
| 520 |
"step": 34
|
| 521 |
},
|
| 522 |
{
|
| 523 |
"epoch": 1.8421052631578947,
|
| 524 |
-
"grad_norm":
|
| 525 |
-
"learning_rate": 0.
|
| 526 |
-
"loss":
|
| 527 |
"step": 35
|
| 528 |
},
|
| 529 |
{
|
| 530 |
"epoch": 1.8421052631578947,
|
| 531 |
-
"eval_loss": 0.
|
| 532 |
-
"eval_runtime":
|
| 533 |
-
"eval_samples_per_second":
|
| 534 |
-
"eval_steps_per_second":
|
| 535 |
"step": 35
|
| 536 |
},
|
| 537 |
{
|
| 538 |
"epoch": 1.8947368421052633,
|
| 539 |
-
"grad_norm":
|
| 540 |
-
"learning_rate": 0.
|
| 541 |
-
"loss": 0.
|
| 542 |
"step": 36
|
| 543 |
},
|
| 544 |
{
|
| 545 |
"epoch": 1.8947368421052633,
|
| 546 |
-
"eval_loss": 0.
|
| 547 |
-
"eval_runtime":
|
| 548 |
-
"eval_samples_per_second":
|
| 549 |
-
"eval_steps_per_second":
|
| 550 |
"step": 36
|
| 551 |
},
|
| 552 |
{
|
| 553 |
"epoch": 1.9473684210526314,
|
| 554 |
-
"grad_norm":
|
| 555 |
-
"learning_rate": 0.
|
| 556 |
-
"loss": 1.
|
| 557 |
"step": 37
|
| 558 |
},
|
| 559 |
{
|
| 560 |
"epoch": 1.9473684210526314,
|
| 561 |
-
"eval_loss": 0.
|
| 562 |
-
"eval_runtime":
|
| 563 |
-
"eval_samples_per_second":
|
| 564 |
-
"eval_steps_per_second":
|
| 565 |
"step": 37
|
| 566 |
},
|
| 567 |
{
|
| 568 |
"epoch": 2.0,
|
| 569 |
-
"grad_norm":
|
| 570 |
-
"learning_rate": 0.
|
| 571 |
-
"loss": 1.
|
| 572 |
"step": 38
|
| 573 |
},
|
| 574 |
{
|
| 575 |
"epoch": 2.0,
|
| 576 |
-
"eval_loss": 0.
|
| 577 |
-
"eval_runtime":
|
| 578 |
-
"eval_samples_per_second":
|
| 579 |
-
"eval_steps_per_second":
|
| 580 |
"step": 38
|
| 581 |
},
|
| 582 |
{
|
| 583 |
"epoch": 2.0526315789473686,
|
| 584 |
-
"grad_norm":
|
| 585 |
-
"learning_rate": 0.
|
| 586 |
-
"loss": 0.
|
| 587 |
"step": 39
|
| 588 |
},
|
| 589 |
{
|
| 590 |
"epoch": 2.0526315789473686,
|
| 591 |
-
"eval_loss": 0.
|
| 592 |
-
"eval_runtime":
|
| 593 |
-
"eval_samples_per_second":
|
| 594 |
-
"eval_steps_per_second":
|
| 595 |
"step": 39
|
| 596 |
},
|
| 597 |
{
|
| 598 |
"epoch": 2.1052631578947367,
|
| 599 |
-
"grad_norm":
|
| 600 |
-
"learning_rate": 0.
|
| 601 |
-
"loss": 0.
|
| 602 |
"step": 40
|
| 603 |
},
|
| 604 |
{
|
| 605 |
"epoch": 2.1052631578947367,
|
| 606 |
-
"eval_loss": 0.
|
| 607 |
-
"eval_runtime":
|
| 608 |
-
"eval_samples_per_second":
|
| 609 |
-
"eval_steps_per_second":
|
| 610 |
"step": 40
|
| 611 |
},
|
| 612 |
{
|
| 613 |
"epoch": 2.1578947368421053,
|
| 614 |
-
"grad_norm":
|
| 615 |
-
"learning_rate": 0.
|
| 616 |
-
"loss": 0.
|
| 617 |
"step": 41
|
| 618 |
},
|
| 619 |
{
|
| 620 |
"epoch": 2.1578947368421053,
|
| 621 |
-
"eval_loss": 0.
|
| 622 |
-
"eval_runtime":
|
| 623 |
-
"eval_samples_per_second":
|
| 624 |
-
"eval_steps_per_second":
|
| 625 |
"step": 41
|
| 626 |
},
|
| 627 |
{
|
| 628 |
"epoch": 2.2105263157894735,
|
| 629 |
-
"grad_norm":
|
| 630 |
-
"learning_rate": 0.
|
| 631 |
-
"loss": 0.
|
| 632 |
"step": 42
|
| 633 |
},
|
| 634 |
{
|
| 635 |
"epoch": 2.2105263157894735,
|
| 636 |
-
"eval_loss": 0.
|
| 637 |
-
"eval_runtime":
|
| 638 |
-
"eval_samples_per_second":
|
| 639 |
-
"eval_steps_per_second":
|
| 640 |
"step": 42
|
| 641 |
},
|
| 642 |
{
|
| 643 |
"epoch": 2.263157894736842,
|
| 644 |
-
"grad_norm":
|
| 645 |
-
"learning_rate": 0.
|
| 646 |
-
"loss": 0.
|
| 647 |
"step": 43
|
| 648 |
},
|
| 649 |
{
|
| 650 |
"epoch": 2.263157894736842,
|
| 651 |
-
"eval_loss": 0.
|
| 652 |
-
"eval_runtime":
|
| 653 |
-
"eval_samples_per_second":
|
| 654 |
-
"eval_steps_per_second":
|
| 655 |
"step": 43
|
| 656 |
},
|
| 657 |
{
|
| 658 |
"epoch": 2.3157894736842106,
|
| 659 |
-
"grad_norm":
|
| 660 |
-
"learning_rate": 0.
|
| 661 |
-
"loss": 0.
|
| 662 |
"step": 44
|
| 663 |
},
|
| 664 |
{
|
| 665 |
"epoch": 2.3157894736842106,
|
| 666 |
-
"eval_loss": 0.
|
| 667 |
-
"eval_runtime":
|
| 668 |
-
"eval_samples_per_second":
|
| 669 |
-
"eval_steps_per_second":
|
| 670 |
"step": 44
|
| 671 |
},
|
| 672 |
{
|
| 673 |
"epoch": 2.3684210526315788,
|
| 674 |
-
"grad_norm":
|
| 675 |
-
"learning_rate": 0.
|
| 676 |
-
"loss": 0.
|
| 677 |
"step": 45
|
| 678 |
},
|
| 679 |
{
|
| 680 |
"epoch": 2.3684210526315788,
|
| 681 |
-
"eval_loss": 0.
|
| 682 |
-
"eval_runtime":
|
| 683 |
-
"eval_samples_per_second":
|
| 684 |
-
"eval_steps_per_second":
|
| 685 |
"step": 45
|
| 686 |
},
|
| 687 |
{
|
| 688 |
"epoch": 2.4210526315789473,
|
| 689 |
-
"grad_norm":
|
| 690 |
-
"learning_rate": 0.
|
| 691 |
-
"loss": 0.
|
| 692 |
"step": 46
|
| 693 |
},
|
| 694 |
{
|
| 695 |
"epoch": 2.4210526315789473,
|
| 696 |
-
"eval_loss": 0.
|
| 697 |
-
"eval_runtime":
|
| 698 |
-
"eval_samples_per_second":
|
| 699 |
-
"eval_steps_per_second":
|
| 700 |
"step": 46
|
| 701 |
},
|
| 702 |
{
|
| 703 |
"epoch": 2.473684210526316,
|
| 704 |
-
"grad_norm":
|
| 705 |
-
"learning_rate": 0.
|
| 706 |
-
"loss": 0.
|
| 707 |
"step": 47
|
| 708 |
},
|
| 709 |
{
|
| 710 |
"epoch": 2.473684210526316,
|
| 711 |
-
"eval_loss": 0.
|
| 712 |
-
"eval_runtime":
|
| 713 |
-
"eval_samples_per_second":
|
| 714 |
-
"eval_steps_per_second":
|
| 715 |
"step": 47
|
| 716 |
},
|
| 717 |
{
|
| 718 |
"epoch": 2.526315789473684,
|
| 719 |
-
"grad_norm":
|
| 720 |
-
"learning_rate":
|
| 721 |
-
"loss": 0.
|
| 722 |
"step": 48
|
| 723 |
},
|
| 724 |
{
|
| 725 |
"epoch": 2.526315789473684,
|
| 726 |
-
"eval_loss": 0.
|
| 727 |
-
"eval_runtime":
|
| 728 |
-
"eval_samples_per_second":
|
| 729 |
-
"eval_steps_per_second":
|
| 730 |
"step": 48
|
| 731 |
},
|
| 732 |
{
|
| 733 |
"epoch": 2.5789473684210527,
|
| 734 |
-
"grad_norm":
|
| 735 |
-
"learning_rate":
|
| 736 |
-
"loss": 0.
|
| 737 |
"step": 49
|
| 738 |
},
|
| 739 |
{
|
| 740 |
"epoch": 2.5789473684210527,
|
| 741 |
-
"eval_loss": 0.
|
| 742 |
-
"eval_runtime":
|
| 743 |
-
"eval_samples_per_second":
|
| 744 |
-
"eval_steps_per_second":
|
| 745 |
"step": 49
|
| 746 |
},
|
| 747 |
{
|
| 748 |
"epoch": 2.6315789473684212,
|
| 749 |
-
"grad_norm":
|
| 750 |
-
"learning_rate":
|
| 751 |
-
"loss": 0.
|
| 752 |
"step": 50
|
| 753 |
},
|
| 754 |
{
|
| 755 |
"epoch": 2.6315789473684212,
|
| 756 |
-
"eval_loss": 0.
|
| 757 |
-
"eval_runtime":
|
| 758 |
-
"eval_samples_per_second":
|
| 759 |
-
"eval_steps_per_second":
|
| 760 |
"step": 50
|
| 761 |
},
|
| 762 |
{
|
| 763 |
"epoch": 2.6842105263157894,
|
| 764 |
-
"grad_norm":
|
| 765 |
-
"learning_rate":
|
| 766 |
-
"loss": 0.
|
| 767 |
"step": 51
|
| 768 |
},
|
| 769 |
{
|
| 770 |
"epoch": 2.6842105263157894,
|
| 771 |
-
"eval_loss": 0.
|
| 772 |
-
"eval_runtime":
|
| 773 |
-
"eval_samples_per_second":
|
| 774 |
-
"eval_steps_per_second":
|
| 775 |
"step": 51
|
| 776 |
},
|
| 777 |
{
|
| 778 |
"epoch": 2.736842105263158,
|
| 779 |
-
"grad_norm":
|
| 780 |
-
"learning_rate":
|
| 781 |
-
"loss": 0.
|
| 782 |
"step": 52
|
| 783 |
},
|
| 784 |
{
|
| 785 |
"epoch": 2.736842105263158,
|
| 786 |
-
"eval_loss": 0.
|
| 787 |
-
"eval_runtime":
|
| 788 |
-
"eval_samples_per_second":
|
| 789 |
-
"eval_steps_per_second":
|
| 790 |
"step": 52
|
| 791 |
},
|
| 792 |
{
|
| 793 |
"epoch": 2.7894736842105265,
|
| 794 |
-
"grad_norm":
|
| 795 |
-
"learning_rate":
|
| 796 |
-
"loss": 0.
|
| 797 |
"step": 53
|
| 798 |
},
|
| 799 |
{
|
| 800 |
"epoch": 2.7894736842105265,
|
| 801 |
-
"eval_loss": 0.
|
| 802 |
-
"eval_runtime":
|
| 803 |
-
"eval_samples_per_second":
|
| 804 |
-
"eval_steps_per_second":
|
| 805 |
"step": 53
|
| 806 |
},
|
| 807 |
{
|
| 808 |
"epoch": 2.8421052631578947,
|
| 809 |
-
"grad_norm":
|
| 810 |
-
"learning_rate":
|
| 811 |
-
"loss": 0.
|
| 812 |
"step": 54
|
| 813 |
},
|
| 814 |
{
|
| 815 |
"epoch": 2.8421052631578947,
|
| 816 |
-
"eval_loss": 0.
|
| 817 |
-
"eval_runtime":
|
| 818 |
-
"eval_samples_per_second":
|
| 819 |
-
"eval_steps_per_second":
|
| 820 |
"step": 54
|
| 821 |
},
|
| 822 |
{
|
| 823 |
"epoch": 2.8947368421052633,
|
| 824 |
-
"grad_norm":
|
| 825 |
-
"learning_rate":
|
| 826 |
-
"loss": 0.
|
| 827 |
"step": 55
|
| 828 |
},
|
| 829 |
{
|
| 830 |
"epoch": 2.8947368421052633,
|
| 831 |
-
"eval_loss": 0.
|
| 832 |
-
"eval_runtime":
|
| 833 |
-
"eval_samples_per_second":
|
| 834 |
-
"eval_steps_per_second":
|
| 835 |
"step": 55
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 836 |
}
|
| 837 |
],
|
| 838 |
"logging_steps": 1,
|
| 839 |
-
"max_steps":
|
| 840 |
"num_input_tokens_seen": 0,
|
| 841 |
-
"num_train_epochs":
|
| 842 |
"save_steps": 5,
|
| 843 |
"stateful_callbacks": {
|
| 844 |
"TrainerControl": {
|
|
@@ -852,7 +3027,7 @@
|
|
| 852 |
"attributes": {}
|
| 853 |
}
|
| 854 |
},
|
| 855 |
-
"total_flos":
|
| 856 |
"train_batch_size": 1,
|
| 857 |
"trial_name": null,
|
| 858 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_global_step": 200,
|
| 3 |
+
"best_metric": 0.0016098986379802227,
|
| 4 |
+
"best_model_checkpoint": "/content/drive/MyDrive/lora_model/outputs/task15_microsoft/Phi-4-mini-instruct/checkpoint-200",
|
| 5 |
+
"epoch": 10.526315789473685,
|
| 6 |
"eval_steps": 1,
|
| 7 |
+
"global_step": 200,
|
| 8 |
"is_hyper_param_search": false,
|
| 9 |
"is_local_process_zero": true,
|
| 10 |
"is_world_process_zero": true,
|
| 11 |
"log_history": [
|
| 12 |
{
|
| 13 |
"epoch": 0.05263157894736842,
|
| 14 |
+
"grad_norm": 9.795289039611816,
|
| 15 |
"learning_rate": 0.0,
|
| 16 |
+
"loss": 3.2204,
|
| 17 |
"step": 1
|
| 18 |
},
|
| 19 |
{
|
| 20 |
"epoch": 0.05263157894736842,
|
| 21 |
+
"eval_loss": 3.1565215587615967,
|
| 22 |
+
"eval_runtime": 0.9831,
|
| 23 |
+
"eval_samples_per_second": 30.517,
|
| 24 |
+
"eval_steps_per_second": 4.069,
|
| 25 |
"step": 1
|
| 26 |
},
|
| 27 |
{
|
| 28 |
"epoch": 0.10526315789473684,
|
| 29 |
+
"grad_norm": 10.048436164855957,
|
| 30 |
"learning_rate": 3.3333333333333335e-05,
|
| 31 |
+
"loss": 3.1604,
|
| 32 |
"step": 2
|
| 33 |
},
|
| 34 |
{
|
| 35 |
"epoch": 0.10526315789473684,
|
| 36 |
+
"eval_loss": 2.4775681495666504,
|
| 37 |
+
"eval_runtime": 0.8971,
|
| 38 |
+
"eval_samples_per_second": 33.44,
|
| 39 |
+
"eval_steps_per_second": 4.459,
|
| 40 |
"step": 2
|
| 41 |
},
|
| 42 |
{
|
| 43 |
"epoch": 0.15789473684210525,
|
| 44 |
+
"grad_norm": 5.148971080780029,
|
| 45 |
"learning_rate": 6.666666666666667e-05,
|
| 46 |
+
"loss": 2.3511,
|
| 47 |
"step": 3
|
| 48 |
},
|
| 49 |
{
|
| 50 |
"epoch": 0.15789473684210525,
|
| 51 |
+
"eval_loss": 2.0030856132507324,
|
| 52 |
+
"eval_runtime": 0.8926,
|
| 53 |
+
"eval_samples_per_second": 33.611,
|
| 54 |
+
"eval_steps_per_second": 4.481,
|
| 55 |
"step": 3
|
| 56 |
},
|
| 57 |
{
|
| 58 |
"epoch": 0.21052631578947367,
|
| 59 |
+
"grad_norm": 4.8437819480896,
|
| 60 |
"learning_rate": 0.0001,
|
| 61 |
+
"loss": 2.0198,
|
| 62 |
"step": 4
|
| 63 |
},
|
| 64 |
{
|
| 65 |
"epoch": 0.21052631578947367,
|
| 66 |
+
"eval_loss": 1.6053706407546997,
|
| 67 |
+
"eval_runtime": 0.8924,
|
| 68 |
+
"eval_samples_per_second": 33.618,
|
| 69 |
+
"eval_steps_per_second": 4.482,
|
| 70 |
"step": 4
|
| 71 |
},
|
| 72 |
{
|
| 73 |
"epoch": 0.2631578947368421,
|
| 74 |
+
"grad_norm": 4.386927604675293,
|
| 75 |
"learning_rate": 0.00013333333333333334,
|
| 76 |
+
"loss": 1.6969,
|
| 77 |
"step": 5
|
| 78 |
},
|
| 79 |
{
|
| 80 |
"epoch": 0.2631578947368421,
|
| 81 |
+
"eval_loss": 1.4053733348846436,
|
| 82 |
+
"eval_runtime": 0.8951,
|
| 83 |
+
"eval_samples_per_second": 33.517,
|
| 84 |
+
"eval_steps_per_second": 4.469,
|
| 85 |
"step": 5
|
| 86 |
},
|
| 87 |
{
|
| 88 |
"epoch": 0.3157894736842105,
|
| 89 |
+
"grad_norm": 3.955519676208496,
|
| 90 |
"learning_rate": 0.00016666666666666666,
|
| 91 |
+
"loss": 1.4825,
|
| 92 |
"step": 6
|
| 93 |
},
|
| 94 |
{
|
| 95 |
"epoch": 0.3157894736842105,
|
| 96 |
+
"eval_loss": 1.3105080127716064,
|
| 97 |
+
"eval_runtime": 0.893,
|
| 98 |
+
"eval_samples_per_second": 33.593,
|
| 99 |
+
"eval_steps_per_second": 4.479,
|
| 100 |
"step": 6
|
| 101 |
},
|
| 102 |
{
|
| 103 |
"epoch": 0.3684210526315789,
|
| 104 |
+
"grad_norm": 3.6086604595184326,
|
| 105 |
"learning_rate": 0.0002,
|
| 106 |
+
"loss": 1.3404,
|
| 107 |
"step": 7
|
| 108 |
},
|
| 109 |
{
|
| 110 |
"epoch": 0.3684210526315789,
|
| 111 |
+
"eval_loss": 1.2445138692855835,
|
| 112 |
+
"eval_runtime": 0.8942,
|
| 113 |
+
"eval_samples_per_second": 33.549,
|
| 114 |
+
"eval_steps_per_second": 4.473,
|
| 115 |
"step": 7
|
| 116 |
},
|
| 117 |
{
|
| 118 |
"epoch": 0.42105263157894735,
|
| 119 |
+
"grad_norm": 2.691216230392456,
|
| 120 |
"learning_rate": 0.00023333333333333333,
|
| 121 |
+
"loss": 1.2627,
|
| 122 |
"step": 8
|
| 123 |
},
|
| 124 |
{
|
| 125 |
"epoch": 0.42105263157894735,
|
| 126 |
+
"eval_loss": 1.1471664905548096,
|
| 127 |
+
"eval_runtime": 0.8927,
|
| 128 |
+
"eval_samples_per_second": 33.606,
|
| 129 |
+
"eval_steps_per_second": 4.481,
|
| 130 |
"step": 8
|
| 131 |
},
|
| 132 |
{
|
| 133 |
"epoch": 0.47368421052631576,
|
| 134 |
+
"grad_norm": 2.5174126625061035,
|
| 135 |
"learning_rate": 0.0002666666666666667,
|
| 136 |
+
"loss": 1.2037,
|
| 137 |
"step": 9
|
| 138 |
},
|
| 139 |
{
|
| 140 |
"epoch": 0.47368421052631576,
|
| 141 |
+
"eval_loss": 1.1372406482696533,
|
| 142 |
+
"eval_runtime": 0.8947,
|
| 143 |
+
"eval_samples_per_second": 33.529,
|
| 144 |
+
"eval_steps_per_second": 4.471,
|
| 145 |
"step": 9
|
| 146 |
},
|
| 147 |
{
|
| 148 |
"epoch": 0.5263157894736842,
|
| 149 |
+
"grad_norm": 2.893831253051758,
|
| 150 |
"learning_rate": 0.0003,
|
| 151 |
+
"loss": 1.1793,
|
| 152 |
"step": 10
|
| 153 |
},
|
| 154 |
{
|
| 155 |
"epoch": 0.5263157894736842,
|
| 156 |
+
"eval_loss": 1.0686627626419067,
|
| 157 |
+
"eval_runtime": 0.8921,
|
| 158 |
+
"eval_samples_per_second": 33.628,
|
| 159 |
+
"eval_steps_per_second": 4.484,
|
| 160 |
"step": 10
|
| 161 |
},
|
| 162 |
{
|
| 163 |
"epoch": 0.5789473684210527,
|
| 164 |
+
"grad_norm": 2.5055713653564453,
|
| 165 |
"learning_rate": 0.0003333333333333333,
|
| 166 |
+
"loss": 1.201,
|
| 167 |
"step": 11
|
| 168 |
},
|
| 169 |
{
|
| 170 |
"epoch": 0.5789473684210527,
|
| 171 |
+
"eval_loss": 1.0994912385940552,
|
| 172 |
+
"eval_runtime": 0.8951,
|
| 173 |
+
"eval_samples_per_second": 33.517,
|
| 174 |
+
"eval_steps_per_second": 4.469,
|
| 175 |
"step": 11
|
| 176 |
},
|
| 177 |
{
|
| 178 |
"epoch": 0.631578947368421,
|
| 179 |
+
"grad_norm": 2.297982931137085,
|
| 180 |
"learning_rate": 0.00036666666666666667,
|
| 181 |
+
"loss": 1.177,
|
| 182 |
"step": 12
|
| 183 |
},
|
| 184 |
{
|
| 185 |
"epoch": 0.631578947368421,
|
| 186 |
+
"eval_loss": 1.0981471538543701,
|
| 187 |
+
"eval_runtime": 0.8926,
|
| 188 |
+
"eval_samples_per_second": 33.608,
|
| 189 |
+
"eval_steps_per_second": 4.481,
|
| 190 |
"step": 12
|
| 191 |
},
|
| 192 |
{
|
| 193 |
"epoch": 0.6842105263157895,
|
| 194 |
+
"grad_norm": 2.8536081314086914,
|
| 195 |
"learning_rate": 0.0004,
|
| 196 |
+
"loss": 1.2106,
|
| 197 |
"step": 13
|
| 198 |
},
|
| 199 |
{
|
| 200 |
"epoch": 0.6842105263157895,
|
| 201 |
+
"eval_loss": 1.0119823217391968,
|
| 202 |
+
"eval_runtime": 0.8936,
|
| 203 |
+
"eval_samples_per_second": 33.574,
|
| 204 |
+
"eval_steps_per_second": 4.477,
|
| 205 |
"step": 13
|
| 206 |
},
|
| 207 |
{
|
| 208 |
"epoch": 0.7368421052631579,
|
| 209 |
+
"grad_norm": 1.8637670278549194,
|
| 210 |
"learning_rate": 0.00043333333333333337,
|
| 211 |
+
"loss": 1.0688,
|
| 212 |
"step": 14
|
| 213 |
},
|
| 214 |
{
|
| 215 |
"epoch": 0.7368421052631579,
|
| 216 |
+
"eval_loss": 1.0545283555984497,
|
| 217 |
+
"eval_runtime": 0.893,
|
| 218 |
+
"eval_samples_per_second": 33.596,
|
| 219 |
+
"eval_steps_per_second": 4.479,
|
| 220 |
"step": 14
|
| 221 |
},
|
| 222 |
{
|
| 223 |
"epoch": 0.7894736842105263,
|
| 224 |
+
"grad_norm": 2.6284332275390625,
|
| 225 |
"learning_rate": 0.00046666666666666666,
|
| 226 |
+
"loss": 1.1661,
|
| 227 |
"step": 15
|
| 228 |
},
|
| 229 |
{
|
| 230 |
"epoch": 0.7894736842105263,
|
| 231 |
+
"eval_loss": 1.031855821609497,
|
| 232 |
+
"eval_runtime": 0.8928,
|
| 233 |
+
"eval_samples_per_second": 33.603,
|
| 234 |
+
"eval_steps_per_second": 4.48,
|
| 235 |
"step": 15
|
| 236 |
},
|
| 237 |
{
|
| 238 |
"epoch": 0.8421052631578947,
|
| 239 |
+
"grad_norm": 1.9439812898635864,
|
| 240 |
"learning_rate": 0.0005,
|
| 241 |
+
"loss": 1.1859,
|
| 242 |
"step": 16
|
| 243 |
},
|
| 244 |
{
|
| 245 |
"epoch": 0.8421052631578947,
|
| 246 |
+
"eval_loss": 0.9958587884902954,
|
| 247 |
+
"eval_runtime": 0.8982,
|
| 248 |
+
"eval_samples_per_second": 33.401,
|
| 249 |
+
"eval_steps_per_second": 4.453,
|
| 250 |
"step": 16
|
| 251 |
},
|
| 252 |
{
|
| 253 |
"epoch": 0.8947368421052632,
|
| 254 |
+
"grad_norm": 1.7199311256408691,
|
| 255 |
+
"learning_rate": 0.0004999776608025946,
|
| 256 |
+
"loss": 1.1636,
|
| 257 |
"step": 17
|
| 258 |
},
|
| 259 |
{
|
| 260 |
"epoch": 0.8947368421052632,
|
| 261 |
+
"eval_loss": 0.9928242564201355,
|
| 262 |
+
"eval_runtime": 0.8982,
|
| 263 |
+
"eval_samples_per_second": 33.398,
|
| 264 |
+
"eval_steps_per_second": 4.453,
|
| 265 |
"step": 17
|
| 266 |
},
|
| 267 |
{
|
| 268 |
"epoch": 0.9473684210526315,
|
| 269 |
+
"grad_norm": 5.788880825042725,
|
| 270 |
+
"learning_rate": 0.000499910647202696,
|
| 271 |
+
"loss": 1.0348,
|
| 272 |
"step": 18
|
| 273 |
},
|
| 274 |
{
|
| 275 |
"epoch": 0.9473684210526315,
|
| 276 |
+
"eval_loss": 1.0362129211425781,
|
| 277 |
+
"eval_runtime": 0.8954,
|
| 278 |
+
"eval_samples_per_second": 33.504,
|
| 279 |
+
"eval_steps_per_second": 4.467,
|
| 280 |
"step": 18
|
| 281 |
},
|
| 282 |
{
|
| 283 |
"epoch": 1.0,
|
| 284 |
+
"grad_norm": 2.156782388687134,
|
| 285 |
+
"learning_rate": 0.0004997989711765446,
|
| 286 |
+
"loss": 1.201,
|
| 287 |
"step": 19
|
| 288 |
},
|
| 289 |
{
|
| 290 |
"epoch": 1.0,
|
| 291 |
+
"eval_loss": 0.9807829260826111,
|
| 292 |
+
"eval_runtime": 0.8926,
|
| 293 |
+
"eval_samples_per_second": 33.611,
|
| 294 |
+
"eval_steps_per_second": 4.481,
|
| 295 |
"step": 19
|
| 296 |
},
|
| 297 |
{
|
| 298 |
"epoch": 1.0526315789473684,
|
| 299 |
+
"grad_norm": 1.9558554887771606,
|
| 300 |
+
"learning_rate": 0.0004996426526821629,
|
| 301 |
+
"loss": 0.8535,
|
| 302 |
"step": 20
|
| 303 |
},
|
| 304 |
{
|
| 305 |
"epoch": 1.0526315789473684,
|
| 306 |
+
"eval_loss": 0.9379722476005554,
|
| 307 |
+
"eval_runtime": 0.8935,
|
| 308 |
+
"eval_samples_per_second": 33.576,
|
| 309 |
+
"eval_steps_per_second": 4.477,
|
| 310 |
"step": 20
|
| 311 |
},
|
| 312 |
{
|
| 313 |
"epoch": 1.1052631578947367,
|
| 314 |
+
"grad_norm": 1.772550106048584,
|
| 315 |
+
"learning_rate": 0.0004994417196557883,
|
| 316 |
+
"loss": 0.968,
|
| 317 |
"step": 21
|
| 318 |
},
|
| 319 |
{
|
| 320 |
"epoch": 1.1052631578947367,
|
| 321 |
+
"eval_loss": 0.9845291376113892,
|
| 322 |
+
"eval_runtime": 0.8928,
|
| 323 |
+
"eval_samples_per_second": 33.603,
|
| 324 |
+
"eval_steps_per_second": 4.48,
|
| 325 |
"step": 21
|
| 326 |
},
|
| 327 |
{
|
| 328 |
"epoch": 1.1578947368421053,
|
| 329 |
+
"grad_norm": 2.108396291732788,
|
| 330 |
+
"learning_rate": 0.0004991962080068813,
|
| 331 |
+
"loss": 1.0552,
|
| 332 |
"step": 22
|
| 333 |
},
|
| 334 |
{
|
| 335 |
"epoch": 1.1578947368421053,
|
| 336 |
+
"eval_loss": 0.9239175319671631,
|
| 337 |
+
"eval_runtime": 0.893,
|
| 338 |
+
"eval_samples_per_second": 33.594,
|
| 339 |
+
"eval_steps_per_second": 4.479,
|
| 340 |
"step": 22
|
| 341 |
},
|
| 342 |
{
|
| 343 |
"epoch": 1.2105263157894737,
|
| 344 |
+
"grad_norm": 1.8215439319610596,
|
| 345 |
+
"learning_rate": 0.0004989061616117073,
|
| 346 |
+
"loss": 0.9825,
|
| 347 |
"step": 23
|
| 348 |
},
|
| 349 |
{
|
| 350 |
"epoch": 1.2105263157894737,
|
| 351 |
+
"eval_loss": 0.980516791343689,
|
| 352 |
+
"eval_runtime": 0.8952,
|
| 353 |
+
"eval_samples_per_second": 33.513,
|
| 354 |
+
"eval_steps_per_second": 4.468,
|
| 355 |
"step": 23
|
| 356 |
},
|
| 357 |
{
|
| 358 |
"epoch": 1.263157894736842,
|
| 359 |
+
"grad_norm": 20.904949188232422,
|
| 360 |
+
"learning_rate": 0.0004985716323054959,
|
| 361 |
+
"loss": 0.9563,
|
| 362 |
"step": 24
|
| 363 |
},
|
| 364 |
{
|
| 365 |
"epoch": 1.263157894736842,
|
| 366 |
+
"eval_loss": 1.112138271331787,
|
| 367 |
+
"eval_runtime": 0.8954,
|
| 368 |
+
"eval_samples_per_second": 33.505,
|
| 369 |
+
"eval_steps_per_second": 4.467,
|
| 370 |
"step": 24
|
| 371 |
},
|
| 372 |
{
|
| 373 |
"epoch": 1.3157894736842106,
|
| 374 |
+
"grad_norm": 2.785473585128784,
|
| 375 |
+
"learning_rate": 0.0004981926798731766,
|
| 376 |
+
"loss": 1.048,
|
| 377 |
"step": 25
|
| 378 |
},
|
| 379 |
{
|
| 380 |
"epoch": 1.3157894736842106,
|
| 381 |
+
"eval_loss": 0.9919915795326233,
|
| 382 |
+
"eval_runtime": 0.8932,
|
| 383 |
+
"eval_samples_per_second": 33.586,
|
| 384 |
+
"eval_steps_per_second": 4.478,
|
| 385 |
"step": 25
|
| 386 |
},
|
| 387 |
{
|
| 388 |
"epoch": 1.368421052631579,
|
| 389 |
+
"grad_norm": 1.8656138181686401,
|
| 390 |
+
"learning_rate": 0.000497769372038695,
|
| 391 |
+
"loss": 1.0315,
|
| 392 |
"step": 26
|
| 393 |
},
|
| 394 |
{
|
| 395 |
"epoch": 1.368421052631579,
|
| 396 |
+
"eval_loss": 0.9384483098983765,
|
| 397 |
+
"eval_runtime": 0.8931,
|
| 398 |
+
"eval_samples_per_second": 33.589,
|
| 399 |
+
"eval_steps_per_second": 4.479,
|
| 400 |
"step": 26
|
| 401 |
},
|
| 402 |
{
|
| 403 |
"epoch": 1.4210526315789473,
|
| 404 |
+
"grad_norm": 1.697496771812439,
|
| 405 |
+
"learning_rate": 0.0004973017844529094,
|
| 406 |
+
"loss": 1.0063,
|
| 407 |
"step": 27
|
| 408 |
},
|
| 409 |
{
|
| 410 |
"epoch": 1.4210526315789473,
|
| 411 |
+
"eval_loss": 0.904453694820404,
|
| 412 |
+
"eval_runtime": 0.8918,
|
| 413 |
+
"eval_samples_per_second": 33.64,
|
| 414 |
+
"eval_steps_per_second": 4.485,
|
| 415 |
"step": 27
|
| 416 |
},
|
| 417 |
{
|
| 418 |
"epoch": 1.4736842105263157,
|
| 419 |
+
"grad_norm": 1.7305934429168701,
|
| 420 |
+
"learning_rate": 0.0004967900006800708,
|
| 421 |
+
"loss": 0.8483,
|
| 422 |
"step": 28
|
| 423 |
},
|
| 424 |
{
|
| 425 |
"epoch": 1.4736842105263157,
|
| 426 |
+
"eval_loss": 0.876754879951477,
|
| 427 |
+
"eval_runtime": 0.8933,
|
| 428 |
+
"eval_samples_per_second": 33.584,
|
| 429 |
+
"eval_steps_per_second": 4.478,
|
| 430 |
"step": 28
|
| 431 |
},
|
| 432 |
{
|
| 433 |
"epoch": 1.526315789473684,
|
| 434 |
+
"grad_norm": 1.7766728401184082,
|
| 435 |
+
"learning_rate": 0.000496234112182889,
|
| 436 |
+
"loss": 1.0118,
|
| 437 |
"step": 29
|
| 438 |
},
|
| 439 |
{
|
| 440 |
"epoch": 1.526315789473684,
|
| 441 |
+
"eval_loss": 0.9041274785995483,
|
| 442 |
+
"eval_runtime": 0.8949,
|
| 443 |
+
"eval_samples_per_second": 33.524,
|
| 444 |
+
"eval_steps_per_second": 4.47,
|
| 445 |
"step": 29
|
| 446 |
},
|
| 447 |
{
|
| 448 |
"epoch": 1.5789473684210527,
|
| 449 |
+
"grad_norm": 1.9015165567398071,
|
| 450 |
+
"learning_rate": 0.000495634218306187,
|
| 451 |
+
"loss": 0.8917,
|
| 452 |
"step": 30
|
| 453 |
},
|
| 454 |
{
|
| 455 |
"epoch": 1.5789473684210527,
|
| 456 |
+
"eval_loss": 0.8897702693939209,
|
| 457 |
+
"eval_runtime": 0.8926,
|
| 458 |
+
"eval_samples_per_second": 33.611,
|
| 459 |
+
"eval_steps_per_second": 4.481,
|
| 460 |
"step": 30
|
| 461 |
},
|
| 462 |
{
|
| 463 |
"epoch": 1.631578947368421,
|
| 464 |
+
"grad_norm": 1.4804080724716187,
|
| 465 |
+
"learning_rate": 0.0004949904262591467,
|
| 466 |
+
"loss": 1.0084,
|
| 467 |
"step": 31
|
| 468 |
},
|
| 469 |
{
|
| 470 |
"epoch": 1.631578947368421,
|
| 471 |
+
"eval_loss": 0.885962188243866,
|
| 472 |
+
"eval_runtime": 0.8988,
|
| 473 |
+
"eval_samples_per_second": 33.378,
|
| 474 |
+
"eval_steps_per_second": 4.45,
|
| 475 |
"step": 31
|
| 476 |
},
|
| 477 |
{
|
| 478 |
"epoch": 1.6842105263157894,
|
| 479 |
+
"grad_norm": 1.819899320602417,
|
| 480 |
+
"learning_rate": 0.0004943028510961491,
|
| 481 |
+
"loss": 0.969,
|
| 482 |
"step": 32
|
| 483 |
},
|
| 484 |
{
|
| 485 |
"epoch": 1.6842105263157894,
|
| 486 |
+
"eval_loss": 0.8608292937278748,
|
| 487 |
+
"eval_runtime": 0.8958,
|
| 488 |
+
"eval_samples_per_second": 33.489,
|
| 489 |
+
"eval_steps_per_second": 4.465,
|
| 490 |
"step": 32
|
| 491 |
},
|
| 492 |
{
|
| 493 |
"epoch": 1.736842105263158,
|
| 494 |
+
"grad_norm": 2.8180196285247803,
|
| 495 |
+
"learning_rate": 0.0004935716156962127,
|
| 496 |
+
"loss": 1.1318,
|
| 497 |
"step": 33
|
| 498 |
},
|
| 499 |
{
|
| 500 |
"epoch": 1.736842105263158,
|
| 501 |
+
"eval_loss": 0.875141978263855,
|
| 502 |
+
"eval_runtime": 0.8971,
|
| 503 |
+
"eval_samples_per_second": 33.441,
|
| 504 |
+
"eval_steps_per_second": 4.459,
|
| 505 |
"step": 33
|
| 506 |
},
|
| 507 |
{
|
| 508 |
"epoch": 1.7894736842105263,
|
| 509 |
+
"grad_norm": 1.8047230243682861,
|
| 510 |
+
"learning_rate": 0.000492796850741033,
|
| 511 |
+
"loss": 1.0002,
|
| 512 |
"step": 34
|
| 513 |
},
|
| 514 |
{
|
| 515 |
"epoch": 1.7894736842105263,
|
| 516 |
+
"eval_loss": 0.89467453956604,
|
| 517 |
+
"eval_runtime": 0.8966,
|
| 518 |
+
"eval_samples_per_second": 33.46,
|
| 519 |
+
"eval_steps_per_second": 4.461,
|
| 520 |
"step": 34
|
| 521 |
},
|
| 522 |
{
|
| 523 |
"epoch": 1.8421052631578947,
|
| 524 |
+
"grad_norm": 2.6305246353149414,
|
| 525 |
+
"learning_rate": 0.0004919786946916281,
|
| 526 |
+
"loss": 1.1024,
|
| 527 |
"step": 35
|
| 528 |
},
|
| 529 |
{
|
| 530 |
"epoch": 1.8421052631578947,
|
| 531 |
+
"eval_loss": 0.8359136581420898,
|
| 532 |
+
"eval_runtime": 0.8971,
|
| 533 |
+
"eval_samples_per_second": 33.44,
|
| 534 |
+
"eval_steps_per_second": 4.459,
|
| 535 |
"step": 35
|
| 536 |
},
|
| 537 |
{
|
| 538 |
"epoch": 1.8947368421052633,
|
| 539 |
+
"grad_norm": 2.4130873680114746,
|
| 540 |
+
"learning_rate": 0.0004911172937635942,
|
| 541 |
+
"loss": 0.9314,
|
| 542 |
"step": 36
|
| 543 |
},
|
| 544 |
{
|
| 545 |
"epoch": 1.8947368421052633,
|
| 546 |
+
"eval_loss": 0.8058050274848938,
|
| 547 |
+
"eval_runtime": 0.8959,
|
| 548 |
+
"eval_samples_per_second": 33.487,
|
| 549 |
+
"eval_steps_per_second": 4.465,
|
| 550 |
"step": 36
|
| 551 |
},
|
| 552 |
{
|
| 553 |
"epoch": 1.9473684210526314,
|
| 554 |
+
"grad_norm": 1.580320119857788,
|
| 555 |
+
"learning_rate": 0.0004902128019009741,
|
| 556 |
+
"loss": 1.0036,
|
| 557 |
"step": 37
|
| 558 |
},
|
| 559 |
{
|
| 560 |
"epoch": 1.9473684210526314,
|
| 561 |
+
"eval_loss": 0.7546663880348206,
|
| 562 |
+
"eval_runtime": 0.8967,
|
| 563 |
+
"eval_samples_per_second": 33.457,
|
| 564 |
+
"eval_steps_per_second": 4.461,
|
| 565 |
"step": 37
|
| 566 |
},
|
| 567 |
{
|
| 568 |
"epoch": 2.0,
|
| 569 |
+
"grad_norm": 1.6066155433654785,
|
| 570 |
+
"learning_rate": 0.000489265380748746,
|
| 571 |
+
"loss": 1.094,
|
| 572 |
"step": 38
|
| 573 |
},
|
| 574 |
{
|
| 575 |
"epoch": 2.0,
|
| 576 |
+
"eval_loss": 0.8417730331420898,
|
| 577 |
+
"eval_runtime": 0.895,
|
| 578 |
+
"eval_samples_per_second": 33.519,
|
| 579 |
+
"eval_steps_per_second": 4.469,
|
| 580 |
"step": 38
|
| 581 |
},
|
| 582 |
{
|
| 583 |
"epoch": 2.0526315789473686,
|
| 584 |
+
"grad_norm": 2.4847571849823,
|
| 585 |
+
"learning_rate": 0.0004882751996239352,
|
| 586 |
+
"loss": 0.9106,
|
| 587 |
"step": 39
|
| 588 |
},
|
| 589 |
{
|
| 590 |
"epoch": 2.0526315789473686,
|
| 591 |
+
"eval_loss": 0.805930495262146,
|
| 592 |
+
"eval_runtime": 0.8985,
|
| 593 |
+
"eval_samples_per_second": 33.388,
|
| 594 |
+
"eval_steps_per_second": 4.452,
|
| 595 |
"step": 39
|
| 596 |
},
|
| 597 |
{
|
| 598 |
"epoch": 2.1052631578947367,
|
| 599 |
+
"grad_norm": 2.144543409347534,
|
| 600 |
+
"learning_rate": 0.0004872424354853545,
|
| 601 |
+
"loss": 0.8542,
|
| 602 |
"step": 40
|
| 603 |
},
|
| 604 |
{
|
| 605 |
"epoch": 2.1052631578947367,
|
| 606 |
+
"eval_loss": 0.7550076842308044,
|
| 607 |
+
"eval_runtime": 0.8977,
|
| 608 |
+
"eval_samples_per_second": 33.42,
|
| 609 |
+
"eval_steps_per_second": 4.456,
|
| 610 |
"step": 40
|
| 611 |
},
|
| 612 |
{
|
| 613 |
"epoch": 2.1578947368421053,
|
| 614 |
+
"grad_norm": 1.2767819166183472,
|
| 615 |
+
"learning_rate": 0.0004861672729019797,
|
| 616 |
+
"loss": 0.7569,
|
| 617 |
"step": 41
|
| 618 |
},
|
| 619 |
{
|
| 620 |
"epoch": 2.1578947368421053,
|
| 621 |
+
"eval_loss": 0.720465362071991,
|
| 622 |
+
"eval_runtime": 0.9013,
|
| 623 |
+
"eval_samples_per_second": 33.285,
|
| 624 |
+
"eval_steps_per_second": 4.438,
|
| 625 |
"step": 41
|
| 626 |
},
|
| 627 |
{
|
| 628 |
"epoch": 2.2105263157894735,
|
| 629 |
+
"grad_norm": 1.4606373310089111,
|
| 630 |
+
"learning_rate": 0.0004850499040199643,
|
| 631 |
+
"loss": 0.6198,
|
| 632 |
"step": 42
|
| 633 |
},
|
| 634 |
{
|
| 635 |
"epoch": 2.2105263157894735,
|
| 636 |
+
"eval_loss": 0.7800072431564331,
|
| 637 |
+
"eval_runtime": 0.8938,
|
| 638 |
+
"eval_samples_per_second": 33.564,
|
| 639 |
+
"eval_steps_per_second": 4.475,
|
| 640 |
"step": 42
|
| 641 |
},
|
| 642 |
{
|
| 643 |
"epoch": 2.263157894736842,
|
| 644 |
+
"grad_norm": 4.208314895629883,
|
| 645 |
+
"learning_rate": 0.0004838905285283005,
|
| 646 |
+
"loss": 0.8454,
|
| 647 |
"step": 43
|
| 648 |
},
|
| 649 |
{
|
| 650 |
"epoch": 2.263157894736842,
|
| 651 |
+
"eval_loss": 0.7882384657859802,
|
| 652 |
+
"eval_runtime": 0.8955,
|
| 653 |
+
"eval_samples_per_second": 33.502,
|
| 654 |
+
"eval_steps_per_second": 4.467,
|
| 655 |
"step": 43
|
| 656 |
},
|
| 657 |
{
|
| 658 |
"epoch": 2.3157894736842106,
|
| 659 |
+
"grad_norm": 2.8906519412994385,
|
| 660 |
+
"learning_rate": 0.00048268935362313215,
|
| 661 |
+
"loss": 0.8786,
|
| 662 |
"step": 44
|
| 663 |
},
|
| 664 |
{
|
| 665 |
"epoch": 2.3157894736842106,
|
| 666 |
+
"eval_loss": 0.7504675388336182,
|
| 667 |
+
"eval_runtime": 0.8973,
|
| 668 |
+
"eval_samples_per_second": 33.435,
|
| 669 |
+
"eval_steps_per_second": 4.458,
|
| 670 |
"step": 44
|
| 671 |
},
|
| 672 |
{
|
| 673 |
"epoch": 2.3684210526315788,
|
| 674 |
+
"grad_norm": 2.5608749389648438,
|
| 675 |
+
"learning_rate": 0.00048144659397072586,
|
| 676 |
+
"loss": 0.7165,
|
| 677 |
"step": 45
|
| 678 |
},
|
| 679 |
{
|
| 680 |
"epoch": 2.3684210526315788,
|
| 681 |
+
"eval_loss": 0.7160356640815735,
|
| 682 |
+
"eval_runtime": 0.8985,
|
| 683 |
+
"eval_samples_per_second": 33.389,
|
| 684 |
+
"eval_steps_per_second": 4.452,
|
| 685 |
"step": 45
|
| 686 |
},
|
| 687 |
{
|
| 688 |
"epoch": 2.4210526315789473,
|
| 689 |
+
"grad_norm": 2.237501621246338,
|
| 690 |
+
"learning_rate": 0.0004801624716691072,
|
| 691 |
+
"loss": 0.9232,
|
| 692 |
"step": 46
|
| 693 |
},
|
| 694 |
{
|
| 695 |
"epoch": 2.4210526315789473,
|
| 696 |
+
"eval_loss": 0.7007637619972229,
|
| 697 |
+
"eval_runtime": 0.8986,
|
| 698 |
+
"eval_samples_per_second": 33.387,
|
| 699 |
+
"eval_steps_per_second": 4.452,
|
| 700 |
"step": 46
|
| 701 |
},
|
| 702 |
{
|
| 703 |
"epoch": 2.473684210526316,
|
| 704 |
+
"grad_norm": 2.166039228439331,
|
| 705 |
+
"learning_rate": 0.00047883721620836894,
|
| 706 |
+
"loss": 0.782,
|
| 707 |
"step": 47
|
| 708 |
},
|
| 709 |
{
|
| 710 |
"epoch": 2.473684210526316,
|
| 711 |
+
"eval_loss": 0.6951841711997986,
|
| 712 |
+
"eval_runtime": 0.9007,
|
| 713 |
+
"eval_samples_per_second": 33.308,
|
| 714 |
+
"eval_steps_per_second": 4.441,
|
| 715 |
"step": 47
|
| 716 |
},
|
| 717 |
{
|
| 718 |
"epoch": 2.526315789473684,
|
| 719 |
+
"grad_norm": 1.6499485969543457,
|
| 720 |
+
"learning_rate": 0.0004774710644296578,
|
| 721 |
+
"loss": 0.7387,
|
| 722 |
"step": 48
|
| 723 |
},
|
| 724 |
{
|
| 725 |
"epoch": 2.526315789473684,
|
| 726 |
+
"eval_loss": 0.7041357755661011,
|
| 727 |
+
"eval_runtime": 0.8999,
|
| 728 |
+
"eval_samples_per_second": 33.337,
|
| 729 |
+
"eval_steps_per_second": 4.445,
|
| 730 |
"step": 48
|
| 731 |
},
|
| 732 |
{
|
| 733 |
"epoch": 2.5789473684210527,
|
| 734 |
+
"grad_norm": 2.833061456680298,
|
| 735 |
+
"learning_rate": 0.00047606426048284813,
|
| 736 |
+
"loss": 0.8343,
|
| 737 |
"step": 49
|
| 738 |
},
|
| 739 |
{
|
| 740 |
"epoch": 2.5789473684210527,
|
| 741 |
+
"eval_loss": 0.6822550296783447,
|
| 742 |
+
"eval_runtime": 0.9005,
|
| 743 |
+
"eval_samples_per_second": 33.316,
|
| 744 |
+
"eval_steps_per_second": 4.442,
|
| 745 |
"step": 49
|
| 746 |
},
|
| 747 |
{
|
| 748 |
"epoch": 2.6315789473684212,
|
| 749 |
+
"grad_norm": 2.0135650634765625,
|
| 750 |
+
"learning_rate": 0.00047461705578290833,
|
| 751 |
+
"loss": 0.7768,
|
| 752 |
"step": 50
|
| 753 |
},
|
| 754 |
{
|
| 755 |
"epoch": 2.6315789473684212,
|
| 756 |
+
"eval_loss": 0.6283606886863708,
|
| 757 |
+
"eval_runtime": 0.8974,
|
| 758 |
+
"eval_samples_per_second": 33.428,
|
| 759 |
+
"eval_steps_per_second": 4.457,
|
| 760 |
"step": 50
|
| 761 |
},
|
| 762 |
{
|
| 763 |
"epoch": 2.6842105263157894,
|
| 764 |
+
"grad_norm": 1.5658601522445679,
|
| 765 |
+
"learning_rate": 0.0004731297089649703,
|
| 766 |
+
"loss": 0.7418,
|
| 767 |
"step": 51
|
| 768 |
},
|
| 769 |
{
|
| 770 |
"epoch": 2.6842105263157894,
|
| 771 |
+
"eval_loss": 0.6374291181564331,
|
| 772 |
+
"eval_runtime": 0.8918,
|
| 773 |
+
"eval_samples_per_second": 33.641,
|
| 774 |
+
"eval_steps_per_second": 4.485,
|
| 775 |
"step": 51
|
| 776 |
},
|
| 777 |
{
|
| 778 |
"epoch": 2.736842105263158,
|
| 779 |
+
"grad_norm": 1.7403415441513062,
|
| 780 |
+
"learning_rate": 0.0004716024858381075,
|
| 781 |
+
"loss": 0.7866,
|
| 782 |
"step": 52
|
| 783 |
},
|
| 784 |
{
|
| 785 |
"epoch": 2.736842105263158,
|
| 786 |
+
"eval_loss": 0.6586597561836243,
|
| 787 |
+
"eval_runtime": 0.8957,
|
| 788 |
+
"eval_samples_per_second": 33.495,
|
| 789 |
+
"eval_steps_per_second": 4.466,
|
| 790 |
"step": 52
|
| 791 |
},
|
| 792 |
{
|
| 793 |
"epoch": 2.7894736842105265,
|
| 794 |
+
"grad_norm": 1.519404411315918,
|
| 795 |
+
"learning_rate": 0.00047003565933783123,
|
| 796 |
+
"loss": 0.8354,
|
| 797 |
"step": 53
|
| 798 |
},
|
| 799 |
{
|
| 800 |
"epoch": 2.7894736842105265,
|
| 801 |
+
"eval_loss": 0.691727340221405,
|
| 802 |
+
"eval_runtime": 0.8923,
|
| 803 |
+
"eval_samples_per_second": 33.62,
|
| 804 |
+
"eval_steps_per_second": 4.483,
|
| 805 |
"step": 53
|
| 806 |
},
|
| 807 |
{
|
| 808 |
"epoch": 2.8421052631578947,
|
| 809 |
+
"grad_norm": 1.5139788389205933,
|
| 810 |
+
"learning_rate": 0.0004684295094773134,
|
| 811 |
+
"loss": 0.7804,
|
| 812 |
"step": 54
|
| 813 |
},
|
| 814 |
{
|
| 815 |
"epoch": 2.8421052631578947,
|
| 816 |
+
"eval_loss": 0.6508743762969971,
|
| 817 |
+
"eval_runtime": 0.8929,
|
| 818 |
+
"eval_samples_per_second": 33.598,
|
| 819 |
+
"eval_steps_per_second": 4.48,
|
| 820 |
"step": 54
|
| 821 |
},
|
| 822 |
{
|
| 823 |
"epoch": 2.8947368421052633,
|
| 824 |
+
"grad_norm": 1.5480479001998901,
|
| 825 |
+
"learning_rate": 0.00046678432329734434,
|
| 826 |
+
"loss": 0.7253,
|
| 827 |
"step": 55
|
| 828 |
},
|
| 829 |
{
|
| 830 |
"epoch": 2.8947368421052633,
|
| 831 |
+
"eval_loss": 0.6439611911773682,
|
| 832 |
+
"eval_runtime": 0.894,
|
| 833 |
+
"eval_samples_per_second": 33.557,
|
| 834 |
+
"eval_steps_per_second": 4.474,
|
| 835 |
"step": 55
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 2.9473684210526314,
|
| 839 |
+
"grad_norm": 1.5994068384170532,
|
| 840 |
+
"learning_rate": 0.00046510039481503486,
|
| 841 |
+
"loss": 0.842,
|
| 842 |
+
"step": 56
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 2.9473684210526314,
|
| 846 |
+
"eval_loss": 0.6327024698257446,
|
| 847 |
+
"eval_runtime": 0.9041,
|
| 848 |
+
"eval_samples_per_second": 33.184,
|
| 849 |
+
"eval_steps_per_second": 4.424,
|
| 850 |
+
"step": 56
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 3.0,
|
| 854 |
+
"grad_norm": 1.6054733991622925,
|
| 855 |
+
"learning_rate": 0.00046337802497127117,
|
| 856 |
+
"loss": 0.8073,
|
| 857 |
+
"step": 57
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 3.0,
|
| 861 |
+
"eval_loss": 0.6213096976280212,
|
| 862 |
+
"eval_runtime": 0.8992,
|
| 863 |
+
"eval_samples_per_second": 33.362,
|
| 864 |
+
"eval_steps_per_second": 4.448,
|
| 865 |
+
"step": 57
|
| 866 |
+
},
|
| 867 |
+
{
|
| 868 |
+
"epoch": 3.0526315789473686,
|
| 869 |
+
"grad_norm": 2.5787405967712402,
|
| 870 |
+
"learning_rate": 0.00046161752157693284,
|
| 871 |
+
"loss": 0.6017,
|
| 872 |
+
"step": 58
|
| 873 |
+
},
|
| 874 |
+
{
|
| 875 |
+
"epoch": 3.0526315789473686,
|
| 876 |
+
"eval_loss": 0.5892248749732971,
|
| 877 |
+
"eval_runtime": 0.8922,
|
| 878 |
+
"eval_samples_per_second": 33.624,
|
| 879 |
+
"eval_steps_per_second": 4.483,
|
| 880 |
+
"step": 58
|
| 881 |
+
},
|
| 882 |
+
{
|
| 883 |
+
"epoch": 3.1052631578947367,
|
| 884 |
+
"grad_norm": 1.7601501941680908,
|
| 885 |
+
"learning_rate": 0.0004598191992578828,
|
| 886 |
+
"loss": 0.6071,
|
| 887 |
+
"step": 59
|
| 888 |
+
},
|
| 889 |
+
{
|
| 890 |
+
"epoch": 3.1052631578947367,
|
| 891 |
+
"eval_loss": 0.5735067129135132,
|
| 892 |
+
"eval_runtime": 0.8924,
|
| 893 |
+
"eval_samples_per_second": 33.618,
|
| 894 |
+
"eval_steps_per_second": 4.482,
|
| 895 |
+
"step": 59
|
| 896 |
+
},
|
| 897 |
+
{
|
| 898 |
+
"epoch": 3.1578947368421053,
|
| 899 |
+
"grad_norm": 1.7480543851852417,
|
| 900 |
+
"learning_rate": 0.00045798337939873923,
|
| 901 |
+
"loss": 0.6597,
|
| 902 |
+
"step": 60
|
| 903 |
+
},
|
| 904 |
+
{
|
| 905 |
+
"epoch": 3.1578947368421053,
|
| 906 |
+
"eval_loss": 0.5306870341300964,
|
| 907 |
+
"eval_runtime": 0.8938,
|
| 908 |
+
"eval_samples_per_second": 33.566,
|
| 909 |
+
"eval_steps_per_second": 4.475,
|
| 910 |
+
"step": 60
|
| 911 |
+
},
|
| 912 |
+
{
|
| 913 |
+
"epoch": 3.2105263157894735,
|
| 914 |
+
"grad_norm": 2.3808937072753906,
|
| 915 |
+
"learning_rate": 0.0004561103900854401,
|
| 916 |
+
"loss": 0.5372,
|
| 917 |
+
"step": 61
|
| 918 |
+
},
|
| 919 |
+
{
|
| 920 |
+
"epoch": 3.2105263157894735,
|
| 921 |
+
"eval_loss": 0.535223662853241,
|
| 922 |
+
"eval_runtime": 0.8966,
|
| 923 |
+
"eval_samples_per_second": 33.459,
|
| 924 |
+
"eval_steps_per_second": 4.461,
|
| 925 |
+
"step": 61
|
| 926 |
+
},
|
| 927 |
+
{
|
| 928 |
+
"epoch": 3.263157894736842,
|
| 929 |
+
"grad_norm": 1.8272178173065186,
|
| 930 |
+
"learning_rate": 0.0004542005660466094,
|
| 931 |
+
"loss": 0.5399,
|
| 932 |
+
"step": 62
|
| 933 |
+
},
|
| 934 |
+
{
|
| 935 |
+
"epoch": 3.263157894736842,
|
| 936 |
+
"eval_loss": 0.5316082239151001,
|
| 937 |
+
"eval_runtime": 0.8994,
|
| 938 |
+
"eval_samples_per_second": 33.354,
|
| 939 |
+
"eval_steps_per_second": 4.447,
|
| 940 |
+
"step": 62
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 3.3157894736842106,
|
| 944 |
+
"grad_norm": 2.0635435581207275,
|
| 945 |
+
"learning_rate": 0.0004522542485937369,
|
| 946 |
+
"loss": 0.5531,
|
| 947 |
+
"step": 63
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 3.3157894736842106,
|
| 951 |
+
"eval_loss": 0.5134085416793823,
|
| 952 |
+
"eval_runtime": 0.8937,
|
| 953 |
+
"eval_samples_per_second": 33.567,
|
| 954 |
+
"eval_steps_per_second": 4.476,
|
| 955 |
+
"step": 63
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 3.3684210526315788,
|
| 959 |
+
"grad_norm": 2.268183708190918,
|
| 960 |
+
"learning_rate": 0.0004502717855601809,
|
| 961 |
+
"loss": 0.5291,
|
| 962 |
+
"step": 64
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 3.3684210526315788,
|
| 966 |
+
"eval_loss": 0.5419598817825317,
|
| 967 |
+
"eval_runtime": 0.8959,
|
| 968 |
+
"eval_samples_per_second": 33.486,
|
| 969 |
+
"eval_steps_per_second": 4.465,
|
| 970 |
+
"step": 64
|
| 971 |
+
},
|
| 972 |
+
{
|
| 973 |
+
"epoch": 3.4210526315789473,
|
| 974 |
+
"grad_norm": 1.8800358772277832,
|
| 975 |
+
"learning_rate": 0.0004482535312390058,
|
| 976 |
+
"loss": 0.5501,
|
| 977 |
+
"step": 65
|
| 978 |
+
},
|
| 979 |
+
{
|
| 980 |
+
"epoch": 3.4210526315789473,
|
| 981 |
+
"eval_loss": 0.5209227800369263,
|
| 982 |
+
"eval_runtime": 0.8927,
|
| 983 |
+
"eval_samples_per_second": 33.606,
|
| 984 |
+
"eval_steps_per_second": 4.481,
|
| 985 |
+
"step": 65
|
| 986 |
+
},
|
| 987 |
+
{
|
| 988 |
+
"epoch": 3.473684210526316,
|
| 989 |
+
"grad_norm": 3.1507558822631836,
|
| 990 |
+
"learning_rate": 0.00044619984631966527,
|
| 991 |
+
"loss": 0.5309,
|
| 992 |
+
"step": 66
|
| 993 |
+
},
|
| 994 |
+
{
|
| 995 |
+
"epoch": 3.473684210526316,
|
| 996 |
+
"eval_loss": 0.536996603012085,
|
| 997 |
+
"eval_runtime": 0.8951,
|
| 998 |
+
"eval_samples_per_second": 33.517,
|
| 999 |
+
"eval_steps_per_second": 4.469,
|
| 1000 |
+
"step": 66
|
| 1001 |
+
},
|
| 1002 |
+
{
|
| 1003 |
+
"epoch": 3.526315789473684,
|
| 1004 |
+
"grad_norm": 3.5700478553771973,
|
| 1005 |
+
"learning_rate": 0.0004441110978235418,
|
| 1006 |
+
"loss": 0.7223,
|
| 1007 |
+
"step": 67
|
| 1008 |
+
},
|
| 1009 |
+
{
|
| 1010 |
+
"epoch": 3.526315789473684,
|
| 1011 |
+
"eval_loss": 0.5140640139579773,
|
| 1012 |
+
"eval_runtime": 0.8962,
|
| 1013 |
+
"eval_samples_per_second": 33.474,
|
| 1014 |
+
"eval_steps_per_second": 4.463,
|
| 1015 |
+
"step": 67
|
| 1016 |
+
},
|
| 1017 |
+
{
|
| 1018 |
+
"epoch": 3.5789473684210527,
|
| 1019 |
+
"grad_norm": 1.758971929550171,
|
| 1020 |
+
"learning_rate": 0.0004419876590383554,
|
| 1021 |
+
"loss": 0.6927,
|
| 1022 |
+
"step": 68
|
| 1023 |
+
},
|
| 1024 |
+
{
|
| 1025 |
+
"epoch": 3.5789473684210527,
|
| 1026 |
+
"eval_loss": 0.47072505950927734,
|
| 1027 |
+
"eval_runtime": 0.9127,
|
| 1028 |
+
"eval_samples_per_second": 32.87,
|
| 1029 |
+
"eval_steps_per_second": 4.383,
|
| 1030 |
+
"step": 68
|
| 1031 |
+
},
|
| 1032 |
+
{
|
| 1033 |
+
"epoch": 3.6315789473684212,
|
| 1034 |
+
"grad_norm": 1.5274709463119507,
|
| 1035 |
+
"learning_rate": 0.00043982990945145146,
|
| 1036 |
+
"loss": 0.4762,
|
| 1037 |
+
"step": 69
|
| 1038 |
+
},
|
| 1039 |
+
{
|
| 1040 |
+
"epoch": 3.6315789473684212,
|
| 1041 |
+
"eval_loss": 0.4518219828605652,
|
| 1042 |
+
"eval_runtime": 0.8967,
|
| 1043 |
+
"eval_samples_per_second": 33.456,
|
| 1044 |
+
"eval_steps_per_second": 4.461,
|
| 1045 |
+
"step": 69
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 3.6842105263157894,
|
| 1049 |
+
"grad_norm": 1.7685797214508057,
|
| 1050 |
+
"learning_rate": 0.0004376382346819819,
|
| 1051 |
+
"loss": 0.5629,
|
| 1052 |
+
"step": 70
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 3.6842105263157894,
|
| 1056 |
+
"eval_loss": 0.40707579255104065,
|
| 1057 |
+
"eval_runtime": 0.8934,
|
| 1058 |
+
"eval_samples_per_second": 33.581,
|
| 1059 |
+
"eval_steps_per_second": 4.478,
|
| 1060 |
+
"step": 70
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 3.736842105263158,
|
| 1064 |
+
"grad_norm": 1.6618574857711792,
|
| 1065 |
+
"learning_rate": 0.00043541302641198946,
|
| 1066 |
+
"loss": 0.5877,
|
| 1067 |
+
"step": 71
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 3.736842105263158,
|
| 1071 |
+
"eval_loss": 0.3780651390552521,
|
| 1072 |
+
"eval_runtime": 0.9024,
|
| 1073 |
+
"eval_samples_per_second": 33.246,
|
| 1074 |
+
"eval_steps_per_second": 4.433,
|
| 1075 |
+
"step": 71
|
| 1076 |
+
},
|
| 1077 |
+
{
|
| 1078 |
+
"epoch": 3.7894736842105265,
|
| 1079 |
+
"grad_norm": 1.542702317237854,
|
| 1080 |
+
"learning_rate": 0.00043315468231640834,
|
| 1081 |
+
"loss": 0.5222,
|
| 1082 |
+
"step": 72
|
| 1083 |
+
},
|
| 1084 |
+
{
|
| 1085 |
+
"epoch": 3.7894736842105265,
|
| 1086 |
+
"eval_loss": 0.3732970356941223,
|
| 1087 |
+
"eval_runtime": 0.9166,
|
| 1088 |
+
"eval_samples_per_second": 32.73,
|
| 1089 |
+
"eval_steps_per_second": 4.364,
|
| 1090 |
+
"step": 72
|
| 1091 |
+
},
|
| 1092 |
+
{
|
| 1093 |
+
"epoch": 3.8421052631578947,
|
| 1094 |
+
"grad_norm": 1.8039391040802002,
|
| 1095 |
+
"learning_rate": 0.00043086360599199516,
|
| 1096 |
+
"loss": 0.5238,
|
| 1097 |
+
"step": 73
|
| 1098 |
+
},
|
| 1099 |
+
{
|
| 1100 |
+
"epoch": 3.8421052631578947,
|
| 1101 |
+
"eval_loss": 0.3568810820579529,
|
| 1102 |
+
"eval_runtime": 0.9031,
|
| 1103 |
+
"eval_samples_per_second": 33.218,
|
| 1104 |
+
"eval_steps_per_second": 4.429,
|
| 1105 |
+
"step": 73
|
| 1106 |
+
},
|
| 1107 |
+
{
|
| 1108 |
+
"epoch": 3.8947368421052633,
|
| 1109 |
+
"grad_norm": 1.6215863227844238,
|
| 1110 |
+
"learning_rate": 0.0004285402068852002,
|
| 1111 |
+
"loss": 0.6504,
|
| 1112 |
+
"step": 74
|
| 1113 |
+
},
|
| 1114 |
+
{
|
| 1115 |
+
"epoch": 3.8947368421052633,
|
| 1116 |
+
"eval_loss": 0.3885921835899353,
|
| 1117 |
+
"eval_runtime": 0.896,
|
| 1118 |
+
"eval_samples_per_second": 33.483,
|
| 1119 |
+
"eval_steps_per_second": 4.464,
|
| 1120 |
+
"step": 74
|
| 1121 |
+
},
|
| 1122 |
+
{
|
| 1123 |
+
"epoch": 3.9473684210526314,
|
| 1124 |
+
"grad_norm": 1.5152952671051025,
|
| 1125 |
+
"learning_rate": 0.00042618490021899383,
|
| 1126 |
+
"loss": 0.5694,
|
| 1127 |
+
"step": 75
|
| 1128 |
+
},
|
| 1129 |
+
{
|
| 1130 |
+
"epoch": 3.9473684210526314,
|
| 1131 |
+
"eval_loss": 0.38745489716529846,
|
| 1132 |
+
"eval_runtime": 0.8939,
|
| 1133 |
+
"eval_samples_per_second": 33.562,
|
| 1134 |
+
"eval_steps_per_second": 4.475,
|
| 1135 |
+
"step": 75
|
| 1136 |
+
},
|
| 1137 |
+
{
|
| 1138 |
+
"epoch": 4.0,
|
| 1139 |
+
"grad_norm": 2.6989200115203857,
|
| 1140 |
+
"learning_rate": 0.00042379810691866064,
|
| 1141 |
+
"loss": 0.5849,
|
| 1142 |
+
"step": 76
|
| 1143 |
+
},
|
| 1144 |
+
{
|
| 1145 |
+
"epoch": 4.0,
|
| 1146 |
+
"eval_loss": 0.42535698413848877,
|
| 1147 |
+
"eval_runtime": 0.9073,
|
| 1148 |
+
"eval_samples_per_second": 33.066,
|
| 1149 |
+
"eval_steps_per_second": 4.409,
|
| 1150 |
+
"step": 76
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 4.052631578947368,
|
| 1154 |
+
"grad_norm": 1.7381691932678223,
|
| 1155 |
+
"learning_rate": 0.00042138025353657407,
|
| 1156 |
+
"loss": 0.3779,
|
| 1157 |
+
"step": 77
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 4.052631578947368,
|
| 1161 |
+
"eval_loss": 0.37115439772605896,
|
| 1162 |
+
"eval_runtime": 0.9112,
|
| 1163 |
+
"eval_samples_per_second": 32.922,
|
| 1164 |
+
"eval_steps_per_second": 4.39,
|
| 1165 |
+
"step": 77
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 4.105263157894737,
|
| 1169 |
+
"grad_norm": 2.188385248184204,
|
| 1170 |
+
"learning_rate": 0.00041893177217596633,
|
| 1171 |
+
"loss": 0.44,
|
| 1172 |
+
"step": 78
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 4.105263157894737,
|
| 1176 |
+
"eval_loss": 0.2926563322544098,
|
| 1177 |
+
"eval_runtime": 0.8982,
|
| 1178 |
+
"eval_samples_per_second": 33.401,
|
| 1179 |
+
"eval_steps_per_second": 4.453,
|
| 1180 |
+
"step": 78
|
| 1181 |
+
},
|
| 1182 |
+
{
|
| 1183 |
+
"epoch": 4.157894736842105,
|
| 1184 |
+
"grad_norm": 2.3652961254119873,
|
| 1185 |
+
"learning_rate": 0.0004164531004137049,
|
| 1186 |
+
"loss": 0.3639,
|
| 1187 |
+
"step": 79
|
| 1188 |
+
},
|
| 1189 |
+
{
|
| 1190 |
+
"epoch": 4.157894736842105,
|
| 1191 |
+
"eval_loss": 0.2751067876815796,
|
| 1192 |
+
"eval_runtime": 0.9146,
|
| 1193 |
+
"eval_samples_per_second": 32.8,
|
| 1194 |
+
"eval_steps_per_second": 4.373,
|
| 1195 |
+
"step": 79
|
| 1196 |
+
},
|
| 1197 |
+
{
|
| 1198 |
+
"epoch": 4.2105263157894735,
|
| 1199 |
+
"grad_norm": 2.165874719619751,
|
| 1200 |
+
"learning_rate": 0.0004139446812220924,
|
| 1201 |
+
"loss": 0.2683,
|
| 1202 |
+
"step": 80
|
| 1203 |
+
},
|
| 1204 |
+
{
|
| 1205 |
+
"epoch": 4.2105263157894735,
|
| 1206 |
+
"eval_loss": 0.2685202360153198,
|
| 1207 |
+
"eval_runtime": 0.9124,
|
| 1208 |
+
"eval_samples_per_second": 32.881,
|
| 1209 |
+
"eval_steps_per_second": 4.384,
|
| 1210 |
+
"step": 80
|
| 1211 |
+
},
|
| 1212 |
+
{
|
| 1213 |
+
"epoch": 4.2631578947368425,
|
| 1214 |
+
"grad_norm": 1.7391912937164307,
|
| 1215 |
+
"learning_rate": 0.0004114069628897006,
|
| 1216 |
+
"loss": 0.2993,
|
| 1217 |
+
"step": 81
|
| 1218 |
+
},
|
| 1219 |
+
{
|
| 1220 |
+
"epoch": 4.2631578947368425,
|
| 1221 |
+
"eval_loss": 0.33646491169929504,
|
| 1222 |
+
"eval_runtime": 0.8952,
|
| 1223 |
+
"eval_samples_per_second": 33.51,
|
| 1224 |
+
"eval_steps_per_second": 4.468,
|
| 1225 |
+
"step": 81
|
| 1226 |
+
},
|
| 1227 |
+
{
|
| 1228 |
+
"epoch": 4.315789473684211,
|
| 1229 |
+
"grad_norm": 3.65714693069458,
|
| 1230 |
+
"learning_rate": 0.0004088403989412559,
|
| 1231 |
+
"loss": 0.4252,
|
| 1232 |
+
"step": 82
|
| 1233 |
+
},
|
| 1234 |
+
{
|
| 1235 |
+
"epoch": 4.315789473684211,
|
| 1236 |
+
"eval_loss": 0.2839888632297516,
|
| 1237 |
+
"eval_runtime": 0.9057,
|
| 1238 |
+
"eval_samples_per_second": 33.123,
|
| 1239 |
+
"eval_steps_per_second": 4.416,
|
| 1240 |
+
"step": 82
|
| 1241 |
+
},
|
| 1242 |
+
{
|
| 1243 |
+
"epoch": 4.368421052631579,
|
| 1244 |
+
"grad_norm": 2.1762771606445312,
|
| 1245 |
+
"learning_rate": 0.00040624544805658794,
|
| 1246 |
+
"loss": 0.3304,
|
| 1247 |
+
"step": 83
|
| 1248 |
+
},
|
| 1249 |
+
{
|
| 1250 |
+
"epoch": 4.368421052631579,
|
| 1251 |
+
"eval_loss": 0.27002134919166565,
|
| 1252 |
+
"eval_runtime": 0.8939,
|
| 1253 |
+
"eval_samples_per_second": 33.562,
|
| 1254 |
+
"eval_steps_per_second": 4.475,
|
| 1255 |
+
"step": 83
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 4.421052631578947,
|
| 1259 |
+
"grad_norm": 2.1018354892730713,
|
| 1260 |
+
"learning_rate": 0.00040362257398865713,
|
| 1261 |
+
"loss": 0.4506,
|
| 1262 |
+
"step": 84
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 4.421052631578947,
|
| 1266 |
+
"eval_loss": 0.2557659149169922,
|
| 1267 |
+
"eval_runtime": 0.8969,
|
| 1268 |
+
"eval_samples_per_second": 33.45,
|
| 1269 |
+
"eval_steps_per_second": 4.46,
|
| 1270 |
+
"step": 84
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 4.473684210526316,
|
| 1274 |
+
"grad_norm": 1.7509180307388306,
|
| 1275 |
+
"learning_rate": 0.00040097224548067613,
|
| 1276 |
+
"loss": 0.3731,
|
| 1277 |
+
"step": 85
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 4.473684210526316,
|
| 1281 |
+
"eval_loss": 0.26859304308891296,
|
| 1282 |
+
"eval_runtime": 0.9009,
|
| 1283 |
+
"eval_samples_per_second": 33.299,
|
| 1284 |
+
"eval_steps_per_second": 4.44,
|
| 1285 |
+
"step": 85
|
| 1286 |
+
},
|
| 1287 |
+
{
|
| 1288 |
+
"epoch": 4.526315789473684,
|
| 1289 |
+
"grad_norm": 1.971816897392273,
|
| 1290 |
+
"learning_rate": 0.0003982949361823388,
|
| 1291 |
+
"loss": 0.38,
|
| 1292 |
+
"step": 86
|
| 1293 |
+
},
|
| 1294 |
+
{
|
| 1295 |
+
"epoch": 4.526315789473684,
|
| 1296 |
+
"eval_loss": 0.2624681293964386,
|
| 1297 |
+
"eval_runtime": 0.8949,
|
| 1298 |
+
"eval_samples_per_second": 33.524,
|
| 1299 |
+
"eval_steps_per_second": 4.47,
|
| 1300 |
+
"step": 86
|
| 1301 |
+
},
|
| 1302 |
+
{
|
| 1303 |
+
"epoch": 4.578947368421053,
|
| 1304 |
+
"grad_norm": 1.4714068174362183,
|
| 1305 |
+
"learning_rate": 0.0003955911245651726,
|
| 1306 |
+
"loss": 0.3944,
|
| 1307 |
+
"step": 87
|
| 1308 |
+
},
|
| 1309 |
+
{
|
| 1310 |
+
"epoch": 4.578947368421053,
|
| 1311 |
+
"eval_loss": 0.23652420938014984,
|
| 1312 |
+
"eval_runtime": 0.8952,
|
| 1313 |
+
"eval_samples_per_second": 33.511,
|
| 1314 |
+
"eval_steps_per_second": 4.468,
|
| 1315 |
+
"step": 87
|
| 1316 |
+
},
|
| 1317 |
+
{
|
| 1318 |
+
"epoch": 4.631578947368421,
|
| 1319 |
+
"grad_norm": 2.6970834732055664,
|
| 1320 |
+
"learning_rate": 0.0003928612938370292,
|
| 1321 |
+
"loss": 0.3374,
|
| 1322 |
+
"step": 88
|
| 1323 |
+
},
|
| 1324 |
+
{
|
| 1325 |
+
"epoch": 4.631578947368421,
|
| 1326 |
+
"eval_loss": 0.2716277241706848,
|
| 1327 |
+
"eval_runtime": 0.8932,
|
| 1328 |
+
"eval_samples_per_second": 33.588,
|
| 1329 |
+
"eval_steps_per_second": 4.478,
|
| 1330 |
+
"step": 88
|
| 1331 |
+
},
|
| 1332 |
+
{
|
| 1333 |
+
"epoch": 4.684210526315789,
|
| 1334 |
+
"grad_norm": 1.9066615104675293,
|
| 1335 |
+
"learning_rate": 0.00039010593185572867,
|
| 1336 |
+
"loss": 0.2442,
|
| 1337 |
+
"step": 89
|
| 1338 |
+
},
|
| 1339 |
+
{
|
| 1340 |
+
"epoch": 4.684210526315789,
|
| 1341 |
+
"eval_loss": 0.2999991476535797,
|
| 1342 |
+
"eval_runtime": 0.8939,
|
| 1343 |
+
"eval_samples_per_second": 33.559,
|
| 1344 |
+
"eval_steps_per_second": 4.475,
|
| 1345 |
+
"step": 89
|
| 1346 |
+
},
|
| 1347 |
+
{
|
| 1348 |
+
"epoch": 4.7368421052631575,
|
| 1349 |
+
"grad_norm": 2.6232354640960693,
|
| 1350 |
+
"learning_rate": 0.00038732553104187296,
|
| 1351 |
+
"loss": 0.2857,
|
| 1352 |
+
"step": 90
|
| 1353 |
+
},
|
| 1354 |
+
{
|
| 1355 |
+
"epoch": 4.7368421052631575,
|
| 1356 |
+
"eval_loss": 0.2302989959716797,
|
| 1357 |
+
"eval_runtime": 0.8938,
|
| 1358 |
+
"eval_samples_per_second": 33.564,
|
| 1359 |
+
"eval_steps_per_second": 4.475,
|
| 1360 |
+
"step": 90
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 4.7894736842105265,
|
| 1364 |
+
"grad_norm": 2.0710129737854004,
|
| 1365 |
+
"learning_rate": 0.0003845205882908432,
|
| 1366 |
+
"loss": 0.4195,
|
| 1367 |
+
"step": 91
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 4.7894736842105265,
|
| 1371 |
+
"eval_loss": 0.21816590428352356,
|
| 1372 |
+
"eval_runtime": 0.9251,
|
| 1373 |
+
"eval_samples_per_second": 32.429,
|
| 1374 |
+
"eval_steps_per_second": 4.324,
|
| 1375 |
+
"step": 91
|
| 1376 |
+
},
|
| 1377 |
+
{
|
| 1378 |
+
"epoch": 4.842105263157895,
|
| 1379 |
+
"grad_norm": 1.8006062507629395,
|
| 1380 |
+
"learning_rate": 0.0003816916048839979,
|
| 1381 |
+
"loss": 0.2859,
|
| 1382 |
+
"step": 92
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 4.842105263157895,
|
| 1386 |
+
"eval_loss": 0.21071405708789825,
|
| 1387 |
+
"eval_runtime": 0.8965,
|
| 1388 |
+
"eval_samples_per_second": 33.462,
|
| 1389 |
+
"eval_steps_per_second": 4.462,
|
| 1390 |
+
"step": 92
|
| 1391 |
+
},
|
| 1392 |
+
{
|
| 1393 |
+
"epoch": 4.894736842105263,
|
| 1394 |
+
"grad_norm": 1.6352888345718384,
|
| 1395 |
+
"learning_rate": 0.0003788390863990875,
|
| 1396 |
+
"loss": 0.4275,
|
| 1397 |
+
"step": 93
|
| 1398 |
+
},
|
| 1399 |
+
{
|
| 1400 |
+
"epoch": 4.894736842105263,
|
| 1401 |
+
"eval_loss": 0.20206846296787262,
|
| 1402 |
+
"eval_runtime": 0.9052,
|
| 1403 |
+
"eval_samples_per_second": 33.144,
|
| 1404 |
+
"eval_steps_per_second": 4.419,
|
| 1405 |
+
"step": 93
|
| 1406 |
+
},
|
| 1407 |
+
{
|
| 1408 |
+
"epoch": 4.947368421052632,
|
| 1409 |
+
"grad_norm": 1.6399378776550293,
|
| 1410 |
+
"learning_rate": 0.00037596354261990007,
|
| 1411 |
+
"loss": 0.389,
|
| 1412 |
+
"step": 94
|
| 1413 |
+
},
|
| 1414 |
+
{
|
| 1415 |
+
"epoch": 4.947368421052632,
|
| 1416 |
+
"eval_loss": 0.19467315077781677,
|
| 1417 |
+
"eval_runtime": 0.8973,
|
| 1418 |
+
"eval_samples_per_second": 33.435,
|
| 1419 |
+
"eval_steps_per_second": 4.458,
|
| 1420 |
+
"step": 94
|
| 1421 |
+
},
|
| 1422 |
+
{
|
| 1423 |
+
"epoch": 5.0,
|
| 1424 |
+
"grad_norm": 1.5680173635482788,
|
| 1425 |
+
"learning_rate": 0.0003730654874451569,
|
| 1426 |
+
"loss": 0.395,
|
| 1427 |
+
"step": 95
|
| 1428 |
+
},
|
| 1429 |
+
{
|
| 1430 |
+
"epoch": 5.0,
|
| 1431 |
+
"eval_loss": 0.19546455144882202,
|
| 1432 |
+
"eval_runtime": 0.91,
|
| 1433 |
+
"eval_samples_per_second": 32.968,
|
| 1434 |
+
"eval_steps_per_second": 4.396,
|
| 1435 |
+
"step": 95
|
| 1436 |
+
},
|
| 1437 |
+
{
|
| 1438 |
+
"epoch": 5.052631578947368,
|
| 1439 |
+
"grad_norm": 1.0308386087417603,
|
| 1440 |
+
"learning_rate": 0.00037014543879667093,
|
| 1441 |
+
"loss": 0.1384,
|
| 1442 |
+
"step": 96
|
| 1443 |
+
},
|
| 1444 |
+
{
|
| 1445 |
+
"epoch": 5.052631578947368,
|
| 1446 |
+
"eval_loss": 0.18969732522964478,
|
| 1447 |
+
"eval_runtime": 0.9021,
|
| 1448 |
+
"eval_samples_per_second": 33.258,
|
| 1449 |
+
"eval_steps_per_second": 4.434,
|
| 1450 |
+
"step": 96
|
| 1451 |
+
},
|
| 1452 |
+
{
|
| 1453 |
+
"epoch": 5.105263157894737,
|
| 1454 |
+
"grad_norm": 1.4042502641677856,
|
| 1455 |
+
"learning_rate": 0.0003672039185267878,
|
| 1456 |
+
"loss": 0.2291,
|
| 1457 |
+
"step": 97
|
| 1458 |
+
},
|
| 1459 |
+
{
|
| 1460 |
+
"epoch": 5.105263157894737,
|
| 1461 |
+
"eval_loss": 0.16800740361213684,
|
| 1462 |
+
"eval_runtime": 0.8938,
|
| 1463 |
+
"eval_samples_per_second": 33.563,
|
| 1464 |
+
"eval_steps_per_second": 4.475,
|
| 1465 |
+
"step": 97
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 5.157894736842105,
|
| 1469 |
+
"grad_norm": 1.6313552856445312,
|
| 1470 |
+
"learning_rate": 0.00036424145232512333,
|
| 1471 |
+
"loss": 0.1736,
|
| 1472 |
+
"step": 98
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 5.157894736842105,
|
| 1476 |
+
"eval_loss": 0.16714099049568176,
|
| 1477 |
+
"eval_runtime": 0.9009,
|
| 1478 |
+
"eval_samples_per_second": 33.301,
|
| 1479 |
+
"eval_steps_per_second": 4.44,
|
| 1480 |
+
"step": 98
|
| 1481 |
+
},
|
| 1482 |
+
{
|
| 1483 |
+
"epoch": 5.2105263157894735,
|
| 1484 |
+
"grad_norm": 1.8922698497772217,
|
| 1485 |
+
"learning_rate": 0.0003612585696246158,
|
| 1486 |
+
"loss": 0.1677,
|
| 1487 |
+
"step": 99
|
| 1488 |
+
},
|
| 1489 |
+
{
|
| 1490 |
+
"epoch": 5.2105263157894735,
|
| 1491 |
+
"eval_loss": 0.179762065410614,
|
| 1492 |
+
"eval_runtime": 0.9039,
|
| 1493 |
+
"eval_samples_per_second": 33.188,
|
| 1494 |
+
"eval_steps_per_second": 4.425,
|
| 1495 |
+
"step": 99
|
| 1496 |
+
},
|
| 1497 |
+
{
|
| 1498 |
+
"epoch": 5.2631578947368425,
|
| 1499 |
+
"grad_norm": 2.409526824951172,
|
| 1500 |
+
"learning_rate": 0.0003582558035069091,
|
| 1501 |
+
"loss": 0.2379,
|
| 1502 |
+
"step": 100
|
| 1503 |
+
},
|
| 1504 |
+
{
|
| 1505 |
+
"epoch": 5.2631578947368425,
|
| 1506 |
+
"eval_loss": 0.1902371197938919,
|
| 1507 |
+
"eval_runtime": 0.9097,
|
| 1508 |
+
"eval_samples_per_second": 32.98,
|
| 1509 |
+
"eval_steps_per_second": 4.397,
|
| 1510 |
+
"step": 100
|
| 1511 |
+
},
|
| 1512 |
+
{
|
| 1513 |
+
"epoch": 5.315789473684211,
|
| 1514 |
+
"grad_norm": 2.084869146347046,
|
| 1515 |
+
"learning_rate": 0.0003552336906070838,
|
| 1516 |
+
"loss": 0.2165,
|
| 1517 |
+
"step": 101
|
| 1518 |
+
},
|
| 1519 |
+
{
|
| 1520 |
+
"epoch": 5.315789473684211,
|
| 1521 |
+
"eval_loss": 0.17252177000045776,
|
| 1522 |
+
"eval_runtime": 0.8948,
|
| 1523 |
+
"eval_samples_per_second": 33.528,
|
| 1524 |
+
"eval_steps_per_second": 4.47,
|
| 1525 |
+
"step": 101
|
| 1526 |
+
},
|
| 1527 |
+
{
|
| 1528 |
+
"epoch": 5.368421052631579,
|
| 1529 |
+
"grad_norm": 1.655718207359314,
|
| 1530 |
+
"learning_rate": 0.000352192771017753,
|
| 1531 |
+
"loss": 0.223,
|
| 1532 |
+
"step": 102
|
| 1533 |
+
},
|
| 1534 |
+
{
|
| 1535 |
+
"epoch": 5.368421052631579,
|
| 1536 |
+
"eval_loss": 0.18867380917072296,
|
| 1537 |
+
"eval_runtime": 0.8956,
|
| 1538 |
+
"eval_samples_per_second": 33.495,
|
| 1539 |
+
"eval_steps_per_second": 4.466,
|
| 1540 |
+
"step": 102
|
| 1541 |
+
},
|
| 1542 |
+
{
|
| 1543 |
+
"epoch": 5.421052631578947,
|
| 1544 |
+
"grad_norm": 2.672633409500122,
|
| 1545 |
+
"learning_rate": 0.0003491335881925407,
|
| 1546 |
+
"loss": 0.161,
|
| 1547 |
+
"step": 103
|
| 1548 |
+
},
|
| 1549 |
+
{
|
| 1550 |
+
"epoch": 5.421052631578947,
|
| 1551 |
+
"eval_loss": 0.1944020837545395,
|
| 1552 |
+
"eval_runtime": 0.8924,
|
| 1553 |
+
"eval_samples_per_second": 33.616,
|
| 1554 |
+
"eval_steps_per_second": 4.482,
|
| 1555 |
+
"step": 103
|
| 1556 |
+
},
|
| 1557 |
+
{
|
| 1558 |
+
"epoch": 5.473684210526316,
|
| 1559 |
+
"grad_norm": 1.9712008237838745,
|
| 1560 |
+
"learning_rate": 0.0003460566888489593,
|
| 1561 |
+
"loss": 0.2525,
|
| 1562 |
+
"step": 104
|
| 1563 |
+
},
|
| 1564 |
+
{
|
| 1565 |
+
"epoch": 5.473684210526316,
|
| 1566 |
+
"eval_loss": 0.17671068012714386,
|
| 1567 |
+
"eval_runtime": 0.897,
|
| 1568 |
+
"eval_samples_per_second": 33.446,
|
| 1569 |
+
"eval_steps_per_second": 4.459,
|
| 1570 |
+
"step": 104
|
| 1571 |
+
},
|
| 1572 |
+
{
|
| 1573 |
+
"epoch": 5.526315789473684,
|
| 1574 |
+
"grad_norm": 2.2153072357177734,
|
| 1575 |
+
"learning_rate": 0.00034296262287070335,
|
| 1576 |
+
"loss": 0.2105,
|
| 1577 |
+
"step": 105
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 5.526315789473684,
|
| 1581 |
+
"eval_loss": 0.1715732216835022,
|
| 1582 |
+
"eval_runtime": 0.8951,
|
| 1583 |
+
"eval_samples_per_second": 33.514,
|
| 1584 |
+
"eval_steps_per_second": 4.469,
|
| 1585 |
+
"step": 105
|
| 1586 |
+
},
|
| 1587 |
+
{
|
| 1588 |
+
"epoch": 5.578947368421053,
|
| 1589 |
+
"grad_norm": 1.8106168508529663,
|
| 1590 |
+
"learning_rate": 0.0003398519432093782,
|
| 1591 |
+
"loss": 0.259,
|
| 1592 |
+
"step": 106
|
| 1593 |
+
},
|
| 1594 |
+
{
|
| 1595 |
+
"epoch": 5.578947368421053,
|
| 1596 |
+
"eval_loss": 0.1465868353843689,
|
| 1597 |
+
"eval_runtime": 0.9077,
|
| 1598 |
+
"eval_samples_per_second": 33.051,
|
| 1599 |
+
"eval_steps_per_second": 4.407,
|
| 1600 |
+
"step": 106
|
| 1601 |
+
},
|
| 1602 |
+
{
|
| 1603 |
+
"epoch": 5.631578947368421,
|
| 1604 |
+
"grad_norm": 2.1159439086914062,
|
| 1605 |
+
"learning_rate": 0.0003367252057856802,
|
| 1606 |
+
"loss": 0.2065,
|
| 1607 |
+
"step": 107
|
| 1608 |
+
},
|
| 1609 |
+
{
|
| 1610 |
+
"epoch": 5.631578947368421,
|
| 1611 |
+
"eval_loss": 0.14219093322753906,
|
| 1612 |
+
"eval_runtime": 0.9049,
|
| 1613 |
+
"eval_samples_per_second": 33.154,
|
| 1614 |
+
"eval_steps_per_second": 4.42,
|
| 1615 |
+
"step": 107
|
| 1616 |
+
},
|
| 1617 |
+
{
|
| 1618 |
+
"epoch": 5.684210526315789,
|
| 1619 |
+
"grad_norm": 1.4467761516571045,
|
| 1620 |
+
"learning_rate": 0.00033358296939004547,
|
| 1621 |
+
"loss": 0.2083,
|
| 1622 |
+
"step": 108
|
| 1623 |
+
},
|
| 1624 |
+
{
|
| 1625 |
+
"epoch": 5.684210526315789,
|
| 1626 |
+
"eval_loss": 0.1406753957271576,
|
| 1627 |
+
"eval_runtime": 0.8954,
|
| 1628 |
+
"eval_samples_per_second": 33.505,
|
| 1629 |
+
"eval_steps_per_second": 4.467,
|
| 1630 |
+
"step": 108
|
| 1631 |
+
},
|
| 1632 |
+
{
|
| 1633 |
+
"epoch": 5.7368421052631575,
|
| 1634 |
+
"grad_norm": 1.3671239614486694,
|
| 1635 |
+
"learning_rate": 0.00033042579558278717,
|
| 1636 |
+
"loss": 0.1825,
|
| 1637 |
+
"step": 109
|
| 1638 |
+
},
|
| 1639 |
+
{
|
| 1640 |
+
"epoch": 5.7368421052631575,
|
| 1641 |
+
"eval_loss": 0.13007155060768127,
|
| 1642 |
+
"eval_runtime": 0.8998,
|
| 1643 |
+
"eval_samples_per_second": 33.342,
|
| 1644 |
+
"eval_steps_per_second": 4.446,
|
| 1645 |
+
"step": 109
|
| 1646 |
+
},
|
| 1647 |
+
{
|
| 1648 |
+
"epoch": 5.7894736842105265,
|
| 1649 |
+
"grad_norm": 1.479944109916687,
|
| 1650 |
+
"learning_rate": 0.00032725424859373687,
|
| 1651 |
+
"loss": 0.2244,
|
| 1652 |
+
"step": 110
|
| 1653 |
+
},
|
| 1654 |
+
{
|
| 1655 |
+
"epoch": 5.7894736842105265,
|
| 1656 |
+
"eval_loss": 0.12692232429981232,
|
| 1657 |
+
"eval_runtime": 0.901,
|
| 1658 |
+
"eval_samples_per_second": 33.298,
|
| 1659 |
+
"eval_steps_per_second": 4.44,
|
| 1660 |
+
"step": 110
|
| 1661 |
+
},
|
| 1662 |
+
{
|
| 1663 |
+
"epoch": 5.842105263157895,
|
| 1664 |
+
"grad_norm": 1.5173969268798828,
|
| 1665 |
+
"learning_rate": 0.0003240688952214085,
|
| 1666 |
+
"loss": 0.2273,
|
| 1667 |
+
"step": 111
|
| 1668 |
+
},
|
| 1669 |
+
{
|
| 1670 |
+
"epoch": 5.842105263157895,
|
| 1671 |
+
"eval_loss": 0.12454597651958466,
|
| 1672 |
+
"eval_runtime": 0.8987,
|
| 1673 |
+
"eval_samples_per_second": 33.382,
|
| 1674 |
+
"eval_steps_per_second": 4.451,
|
| 1675 |
+
"step": 111
|
| 1676 |
+
},
|
| 1677 |
+
{
|
| 1678 |
+
"epoch": 5.894736842105263,
|
| 1679 |
+
"grad_norm": 2.7870988845825195,
|
| 1680 |
+
"learning_rate": 0.00032087030473170445,
|
| 1681 |
+
"loss": 0.2101,
|
| 1682 |
+
"step": 112
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 5.894736842105263,
|
| 1686 |
+
"eval_loss": 0.12002909928560257,
|
| 1687 |
+
"eval_runtime": 0.893,
|
| 1688 |
+
"eval_samples_per_second": 33.593,
|
| 1689 |
+
"eval_steps_per_second": 4.479,
|
| 1690 |
+
"step": 112
|
| 1691 |
+
},
|
| 1692 |
+
{
|
| 1693 |
+
"epoch": 5.947368421052632,
|
| 1694 |
+
"grad_norm": 1.3659342527389526,
|
| 1695 |
+
"learning_rate": 0.00031765904875617973,
|
| 1696 |
+
"loss": 0.1882,
|
| 1697 |
+
"step": 113
|
| 1698 |
+
},
|
| 1699 |
+
{
|
| 1700 |
+
"epoch": 5.947368421052632,
|
| 1701 |
+
"eval_loss": 0.10573837906122208,
|
| 1702 |
+
"eval_runtime": 0.8956,
|
| 1703 |
+
"eval_samples_per_second": 33.496,
|
| 1704 |
+
"eval_steps_per_second": 4.466,
|
| 1705 |
+
"step": 113
|
| 1706 |
+
},
|
| 1707 |
+
{
|
| 1708 |
+
"epoch": 6.0,
|
| 1709 |
+
"grad_norm": 1.8464044332504272,
|
| 1710 |
+
"learning_rate": 0.00031443570118988356,
|
| 1711 |
+
"loss": 0.2285,
|
| 1712 |
+
"step": 114
|
| 1713 |
+
},
|
| 1714 |
+
{
|
| 1715 |
+
"epoch": 6.0,
|
| 1716 |
+
"eval_loss": 0.10221625119447708,
|
| 1717 |
+
"eval_runtime": 0.8955,
|
| 1718 |
+
"eval_samples_per_second": 33.501,
|
| 1719 |
+
"eval_steps_per_second": 4.467,
|
| 1720 |
+
"step": 114
|
| 1721 |
+
},
|
| 1722 |
+
{
|
| 1723 |
+
"epoch": 6.052631578947368,
|
| 1724 |
+
"grad_norm": 1.3894392251968384,
|
| 1725 |
+
"learning_rate": 0.00031120083808879663,
|
| 1726 |
+
"loss": 0.1115,
|
| 1727 |
+
"step": 115
|
| 1728 |
+
},
|
| 1729 |
+
{
|
| 1730 |
+
"epoch": 6.052631578947368,
|
| 1731 |
+
"eval_loss": 0.09458151459693909,
|
| 1732 |
+
"eval_runtime": 0.8981,
|
| 1733 |
+
"eval_samples_per_second": 33.405,
|
| 1734 |
+
"eval_steps_per_second": 4.454,
|
| 1735 |
+
"step": 115
|
| 1736 |
+
},
|
| 1737 |
+
{
|
| 1738 |
+
"epoch": 6.105263157894737,
|
| 1739 |
+
"grad_norm": 0.933142364025116,
|
| 1740 |
+
"learning_rate": 0.0003079550375668821,
|
| 1741 |
+
"loss": 0.0888,
|
| 1742 |
+
"step": 116
|
| 1743 |
+
},
|
| 1744 |
+
{
|
| 1745 |
+
"epoch": 6.105263157894737,
|
| 1746 |
+
"eval_loss": 0.09364737570285797,
|
| 1747 |
+
"eval_runtime": 0.9403,
|
| 1748 |
+
"eval_samples_per_second": 31.905,
|
| 1749 |
+
"eval_steps_per_second": 4.254,
|
| 1750 |
+
"step": 116
|
| 1751 |
+
},
|
| 1752 |
+
{
|
| 1753 |
+
"epoch": 6.157894736842105,
|
| 1754 |
+
"grad_norm": 0.9676756262779236,
|
| 1755 |
+
"learning_rate": 0.00030469887969276877,
|
| 1756 |
+
"loss": 0.0785,
|
| 1757 |
+
"step": 117
|
| 1758 |
+
},
|
| 1759 |
+
{
|
| 1760 |
+
"epoch": 6.157894736842105,
|
| 1761 |
+
"eval_loss": 0.10635325312614441,
|
| 1762 |
+
"eval_runtime": 0.903,
|
| 1763 |
+
"eval_samples_per_second": 33.224,
|
| 1764 |
+
"eval_steps_per_second": 4.43,
|
| 1765 |
+
"step": 117
|
| 1766 |
+
},
|
| 1767 |
+
{
|
| 1768 |
+
"epoch": 6.2105263157894735,
|
| 1769 |
+
"grad_norm": 1.249068260192871,
|
| 1770 |
+
"learning_rate": 0.00030143294638608487,
|
| 1771 |
+
"loss": 0.0938,
|
| 1772 |
+
"step": 118
|
| 1773 |
+
},
|
| 1774 |
+
{
|
| 1775 |
+
"epoch": 6.2105263157894735,
|
| 1776 |
+
"eval_loss": 0.10837359726428986,
|
| 1777 |
+
"eval_runtime": 0.9003,
|
| 1778 |
+
"eval_samples_per_second": 33.323,
|
| 1779 |
+
"eval_steps_per_second": 4.443,
|
| 1780 |
+
"step": 118
|
| 1781 |
+
},
|
| 1782 |
+
{
|
| 1783 |
+
"epoch": 6.2631578947368425,
|
| 1784 |
+
"grad_norm": 2.1446304321289062,
|
| 1785 |
+
"learning_rate": 0.00029815782131346137,
|
| 1786 |
+
"loss": 0.1436,
|
| 1787 |
+
"step": 119
|
| 1788 |
+
},
|
| 1789 |
+
{
|
| 1790 |
+
"epoch": 6.2631578947368425,
|
| 1791 |
+
"eval_loss": 0.1047668606042862,
|
| 1792 |
+
"eval_runtime": 0.9066,
|
| 1793 |
+
"eval_samples_per_second": 33.09,
|
| 1794 |
+
"eval_steps_per_second": 4.412,
|
| 1795 |
+
"step": 119
|
| 1796 |
+
},
|
| 1797 |
+
{
|
| 1798 |
+
"epoch": 6.315789473684211,
|
| 1799 |
+
"grad_norm": 1.329365611076355,
|
| 1800 |
+
"learning_rate": 0.0002948740897842223,
|
| 1801 |
+
"loss": 0.1319,
|
| 1802 |
+
"step": 120
|
| 1803 |
+
},
|
| 1804 |
+
{
|
| 1805 |
+
"epoch": 6.315789473684211,
|
| 1806 |
+
"eval_loss": 0.10011889785528183,
|
| 1807 |
+
"eval_runtime": 0.9034,
|
| 1808 |
+
"eval_samples_per_second": 33.206,
|
| 1809 |
+
"eval_steps_per_second": 4.428,
|
| 1810 |
+
"step": 120
|
| 1811 |
+
},
|
| 1812 |
+
{
|
| 1813 |
+
"epoch": 6.368421052631579,
|
| 1814 |
+
"grad_norm": 1.4938923120498657,
|
| 1815 |
+
"learning_rate": 0.00029158233864578256,
|
| 1816 |
+
"loss": 0.1027,
|
| 1817 |
+
"step": 121
|
| 1818 |
+
},
|
| 1819 |
+
{
|
| 1820 |
+
"epoch": 6.368421052631579,
|
| 1821 |
+
"eval_loss": 0.09962069243192673,
|
| 1822 |
+
"eval_runtime": 0.8983,
|
| 1823 |
+
"eval_samples_per_second": 33.396,
|
| 1824 |
+
"eval_steps_per_second": 4.453,
|
| 1825 |
+
"step": 121
|
| 1826 |
+
},
|
| 1827 |
+
{
|
| 1828 |
+
"epoch": 6.421052631578947,
|
| 1829 |
+
"grad_norm": 1.295058250427246,
|
| 1830 |
+
"learning_rate": 0.00028828315617877,
|
| 1831 |
+
"loss": 0.0763,
|
| 1832 |
+
"step": 122
|
| 1833 |
+
},
|
| 1834 |
+
{
|
| 1835 |
+
"epoch": 6.421052631578947,
|
| 1836 |
+
"eval_loss": 0.10031073540449142,
|
| 1837 |
+
"eval_runtime": 0.9155,
|
| 1838 |
+
"eval_samples_per_second": 32.768,
|
| 1839 |
+
"eval_steps_per_second": 4.369,
|
| 1840 |
+
"step": 122
|
| 1841 |
+
},
|
| 1842 |
+
{
|
| 1843 |
+
"epoch": 6.473684210526316,
|
| 1844 |
+
"grad_norm": 1.8959721326828003,
|
| 1845 |
+
"learning_rate": 0.0002849771319918922,
|
| 1846 |
+
"loss": 0.1292,
|
| 1847 |
+
"step": 123
|
| 1848 |
+
},
|
| 1849 |
+
{
|
| 1850 |
+
"epoch": 6.473684210526316,
|
| 1851 |
+
"eval_loss": 0.11003147065639496,
|
| 1852 |
+
"eval_runtime": 0.919,
|
| 1853 |
+
"eval_samples_per_second": 32.644,
|
| 1854 |
+
"eval_steps_per_second": 4.353,
|
| 1855 |
+
"step": 123
|
| 1856 |
+
},
|
| 1857 |
+
{
|
| 1858 |
+
"epoch": 6.526315789473684,
|
| 1859 |
+
"grad_norm": 1.3598809242248535,
|
| 1860 |
+
"learning_rate": 0.00028166485691656423,
|
| 1861 |
+
"loss": 0.1272,
|
| 1862 |
+
"step": 124
|
| 1863 |
+
},
|
| 1864 |
+
{
|
| 1865 |
+
"epoch": 6.526315789473684,
|
| 1866 |
+
"eval_loss": 0.10435277968645096,
|
| 1867 |
+
"eval_runtime": 0.8989,
|
| 1868 |
+
"eval_samples_per_second": 33.374,
|
| 1869 |
+
"eval_steps_per_second": 4.45,
|
| 1870 |
+
"step": 124
|
| 1871 |
+
},
|
| 1872 |
+
{
|
| 1873 |
+
"epoch": 6.578947368421053,
|
| 1874 |
+
"grad_norm": 1.4015425443649292,
|
| 1875 |
+
"learning_rate": 0.00027834692290132053,
|
| 1876 |
+
"loss": 0.1348,
|
| 1877 |
+
"step": 125
|
| 1878 |
+
},
|
| 1879 |
+
{
|
| 1880 |
+
"epoch": 6.578947368421053,
|
| 1881 |
+
"eval_loss": 0.10004603117704391,
|
| 1882 |
+
"eval_runtime": 0.9016,
|
| 1883 |
+
"eval_samples_per_second": 33.273,
|
| 1884 |
+
"eval_steps_per_second": 4.436,
|
| 1885 |
+
"step": 125
|
| 1886 |
+
},
|
| 1887 |
+
{
|
| 1888 |
+
"epoch": 6.631578947368421,
|
| 1889 |
+
"grad_norm": 1.1642837524414062,
|
| 1890 |
+
"learning_rate": 0.0002750239229060246,
|
| 1891 |
+
"loss": 0.1121,
|
| 1892 |
+
"step": 126
|
| 1893 |
+
},
|
| 1894 |
+
{
|
| 1895 |
+
"epoch": 6.631578947368421,
|
| 1896 |
+
"eval_loss": 0.09635353088378906,
|
| 1897 |
+
"eval_runtime": 0.9239,
|
| 1898 |
+
"eval_samples_per_second": 32.47,
|
| 1899 |
+
"eval_steps_per_second": 4.329,
|
| 1900 |
+
"step": 126
|
| 1901 |
+
},
|
| 1902 |
+
{
|
| 1903 |
+
"epoch": 6.684210526315789,
|
| 1904 |
+
"grad_norm": 1.363749384880066,
|
| 1905 |
+
"learning_rate": 0.0002716964507958994,
|
| 1906 |
+
"loss": 0.1415,
|
| 1907 |
+
"step": 127
|
| 1908 |
+
},
|
| 1909 |
+
{
|
| 1910 |
+
"epoch": 6.684210526315789,
|
| 1911 |
+
"eval_loss": 0.07641066610813141,
|
| 1912 |
+
"eval_runtime": 0.912,
|
| 1913 |
+
"eval_samples_per_second": 32.894,
|
| 1914 |
+
"eval_steps_per_second": 4.386,
|
| 1915 |
+
"step": 127
|
| 1916 |
+
},
|
| 1917 |
+
{
|
| 1918 |
+
"epoch": 6.7368421052631575,
|
| 1919 |
+
"grad_norm": 1.292934775352478,
|
| 1920 |
+
"learning_rate": 0.0002683651012353955,
|
| 1921 |
+
"loss": 0.1513,
|
| 1922 |
+
"step": 128
|
| 1923 |
+
},
|
| 1924 |
+
{
|
| 1925 |
+
"epoch": 6.7368421052631575,
|
| 1926 |
+
"eval_loss": 0.07172319293022156,
|
| 1927 |
+
"eval_runtime": 0.8976,
|
| 1928 |
+
"eval_samples_per_second": 33.423,
|
| 1929 |
+
"eval_steps_per_second": 4.456,
|
| 1930 |
+
"step": 128
|
| 1931 |
+
},
|
| 1932 |
+
{
|
| 1933 |
+
"epoch": 6.7894736842105265,
|
| 1934 |
+
"grad_norm": 1.0078333616256714,
|
| 1935 |
+
"learning_rate": 0.0002650304695819168,
|
| 1936 |
+
"loss": 0.1185,
|
| 1937 |
+
"step": 129
|
| 1938 |
+
},
|
| 1939 |
+
{
|
| 1940 |
+
"epoch": 6.7894736842105265,
|
| 1941 |
+
"eval_loss": 0.06412829458713531,
|
| 1942 |
+
"eval_runtime": 0.8976,
|
| 1943 |
+
"eval_samples_per_second": 33.424,
|
| 1944 |
+
"eval_steps_per_second": 4.456,
|
| 1945 |
+
"step": 129
|
| 1946 |
+
},
|
| 1947 |
+
{
|
| 1948 |
+
"epoch": 6.842105263157895,
|
| 1949 |
+
"grad_norm": 1.173531413078308,
|
| 1950 |
+
"learning_rate": 0.00026169315177942135,
|
| 1951 |
+
"loss": 0.158,
|
| 1952 |
+
"step": 130
|
| 1953 |
+
},
|
| 1954 |
+
{
|
| 1955 |
+
"epoch": 6.842105263157895,
|
| 1956 |
+
"eval_loss": 0.05317940190434456,
|
| 1957 |
+
"eval_runtime": 0.895,
|
| 1958 |
+
"eval_samples_per_second": 33.52,
|
| 1959 |
+
"eval_steps_per_second": 4.469,
|
| 1960 |
+
"step": 130
|
| 1961 |
+
},
|
| 1962 |
+
{
|
| 1963 |
+
"epoch": 6.894736842105263,
|
| 1964 |
+
"grad_norm": 1.5312238931655884,
|
| 1965 |
+
"learning_rate": 0.0002583537442519187,
|
| 1966 |
+
"loss": 0.1335,
|
| 1967 |
+
"step": 131
|
| 1968 |
+
},
|
| 1969 |
+
{
|
| 1970 |
+
"epoch": 6.894736842105263,
|
| 1971 |
+
"eval_loss": 0.04491396248340607,
|
| 1972 |
+
"eval_runtime": 0.9089,
|
| 1973 |
+
"eval_samples_per_second": 33.006,
|
| 1974 |
+
"eval_steps_per_second": 4.401,
|
| 1975 |
+
"step": 131
|
| 1976 |
+
},
|
| 1977 |
+
{
|
| 1978 |
+
"epoch": 6.947368421052632,
|
| 1979 |
+
"grad_norm": 1.399732232093811,
|
| 1980 |
+
"learning_rate": 0.00025501284379688067,
|
| 1981 |
+
"loss": 0.1462,
|
| 1982 |
+
"step": 132
|
| 1983 |
+
},
|
| 1984 |
+
{
|
| 1985 |
+
"epoch": 6.947368421052632,
|
| 1986 |
+
"eval_loss": 0.050081584602594376,
|
| 1987 |
+
"eval_runtime": 0.9127,
|
| 1988 |
+
"eval_samples_per_second": 32.871,
|
| 1989 |
+
"eval_steps_per_second": 4.383,
|
| 1990 |
+
"step": 132
|
| 1991 |
+
},
|
| 1992 |
+
{
|
| 1993 |
+
"epoch": 7.0,
|
| 1994 |
+
"grad_norm": 1.5357416868209839,
|
| 1995 |
+
"learning_rate": 0.0002516710474785856,
|
| 1996 |
+
"loss": 0.1136,
|
| 1997 |
+
"step": 133
|
| 1998 |
+
},
|
| 1999 |
+
{
|
| 2000 |
+
"epoch": 7.0,
|
| 2001 |
+
"eval_loss": 0.05439286679029465,
|
| 2002 |
+
"eval_runtime": 0.8974,
|
| 2003 |
+
"eval_samples_per_second": 33.429,
|
| 2004 |
+
"eval_steps_per_second": 4.457,
|
| 2005 |
+
"step": 133
|
| 2006 |
+
},
|
| 2007 |
+
{
|
| 2008 |
+
"epoch": 7.052631578947368,
|
| 2009 |
+
"grad_norm": 0.9891072511672974,
|
| 2010 |
+
"learning_rate": 0.0002483289525214145,
|
| 2011 |
+
"loss": 0.0741,
|
| 2012 |
+
"step": 134
|
| 2013 |
+
},
|
| 2014 |
+
{
|
| 2015 |
+
"epoch": 7.052631578947368,
|
| 2016 |
+
"eval_loss": 0.05073266103863716,
|
| 2017 |
+
"eval_runtime": 0.8953,
|
| 2018 |
+
"eval_samples_per_second": 33.508,
|
| 2019 |
+
"eval_steps_per_second": 4.468,
|
| 2020 |
+
"step": 134
|
| 2021 |
+
},
|
| 2022 |
+
{
|
| 2023 |
+
"epoch": 7.105263157894737,
|
| 2024 |
+
"grad_norm": 0.9686666131019592,
|
| 2025 |
+
"learning_rate": 0.00024498715620311935,
|
| 2026 |
+
"loss": 0.0518,
|
| 2027 |
+
"step": 135
|
| 2028 |
+
},
|
| 2029 |
+
{
|
| 2030 |
+
"epoch": 7.105263157894737,
|
| 2031 |
+
"eval_loss": 0.05563385412096977,
|
| 2032 |
+
"eval_runtime": 0.9008,
|
| 2033 |
+
"eval_samples_per_second": 33.305,
|
| 2034 |
+
"eval_steps_per_second": 4.441,
|
| 2035 |
+
"step": 135
|
| 2036 |
+
},
|
| 2037 |
+
{
|
| 2038 |
+
"epoch": 7.157894736842105,
|
| 2039 |
+
"grad_norm": 1.2277772426605225,
|
| 2040 |
+
"learning_rate": 0.00024164625574808144,
|
| 2041 |
+
"loss": 0.0436,
|
| 2042 |
+
"step": 136
|
| 2043 |
+
},
|
| 2044 |
+
{
|
| 2045 |
+
"epoch": 7.157894736842105,
|
| 2046 |
+
"eval_loss": 0.058188486844301224,
|
| 2047 |
+
"eval_runtime": 0.9014,
|
| 2048 |
+
"eval_samples_per_second": 33.283,
|
| 2049 |
+
"eval_steps_per_second": 4.438,
|
| 2050 |
+
"step": 136
|
| 2051 |
+
},
|
| 2052 |
+
{
|
| 2053 |
+
"epoch": 7.2105263157894735,
|
| 2054 |
+
"grad_norm": 1.3704907894134521,
|
| 2055 |
+
"learning_rate": 0.00023830684822057877,
|
| 2056 |
+
"loss": 0.1041,
|
| 2057 |
+
"step": 137
|
| 2058 |
+
},
|
| 2059 |
+
{
|
| 2060 |
+
"epoch": 7.2105263157894735,
|
| 2061 |
+
"eval_loss": 0.06476210802793503,
|
| 2062 |
+
"eval_runtime": 0.9007,
|
| 2063 |
+
"eval_samples_per_second": 33.306,
|
| 2064 |
+
"eval_steps_per_second": 4.441,
|
| 2065 |
+
"step": 137
|
| 2066 |
+
},
|
| 2067 |
+
{
|
| 2068 |
+
"epoch": 7.2631578947368425,
|
| 2069 |
+
"grad_norm": 16.16583251953125,
|
| 2070 |
+
"learning_rate": 0.00023496953041808325,
|
| 2071 |
+
"loss": 0.0492,
|
| 2072 |
+
"step": 138
|
| 2073 |
+
},
|
| 2074 |
+
{
|
| 2075 |
+
"epoch": 7.2631578947368425,
|
| 2076 |
+
"eval_loss": 0.07608657330274582,
|
| 2077 |
+
"eval_runtime": 0.8935,
|
| 2078 |
+
"eval_samples_per_second": 33.576,
|
| 2079 |
+
"eval_steps_per_second": 4.477,
|
| 2080 |
+
"step": 138
|
| 2081 |
+
},
|
| 2082 |
+
{
|
| 2083 |
+
"epoch": 7.315789473684211,
|
| 2084 |
+
"grad_norm": 1.200278639793396,
|
| 2085 |
+
"learning_rate": 0.0002316348987646045,
|
| 2086 |
+
"loss": 0.0716,
|
| 2087 |
+
"step": 139
|
| 2088 |
+
},
|
| 2089 |
+
{
|
| 2090 |
+
"epoch": 7.315789473684211,
|
| 2091 |
+
"eval_loss": 0.07832919806241989,
|
| 2092 |
+
"eval_runtime": 0.8933,
|
| 2093 |
+
"eval_samples_per_second": 33.584,
|
| 2094 |
+
"eval_steps_per_second": 4.478,
|
| 2095 |
+
"step": 139
|
| 2096 |
+
},
|
| 2097 |
+
{
|
| 2098 |
+
"epoch": 7.368421052631579,
|
| 2099 |
+
"grad_norm": 1.10837984085083,
|
| 2100 |
+
"learning_rate": 0.00022830354920410064,
|
| 2101 |
+
"loss": 0.0867,
|
| 2102 |
+
"step": 140
|
| 2103 |
+
},
|
| 2104 |
+
{
|
| 2105 |
+
"epoch": 7.368421052631579,
|
| 2106 |
+
"eval_loss": 0.06757114827632904,
|
| 2107 |
+
"eval_runtime": 0.8916,
|
| 2108 |
+
"eval_samples_per_second": 33.649,
|
| 2109 |
+
"eval_steps_per_second": 4.487,
|
| 2110 |
+
"step": 140
|
| 2111 |
+
},
|
| 2112 |
+
{
|
| 2113 |
+
"epoch": 7.421052631578947,
|
| 2114 |
+
"grad_norm": 0.8154372572898865,
|
| 2115 |
+
"learning_rate": 0.0002249760770939754,
|
| 2116 |
+
"loss": 0.0596,
|
| 2117 |
+
"step": 141
|
| 2118 |
+
},
|
| 2119 |
+
{
|
| 2120 |
+
"epoch": 7.421052631578947,
|
| 2121 |
+
"eval_loss": 0.06439080089330673,
|
| 2122 |
+
"eval_runtime": 0.8926,
|
| 2123 |
+
"eval_samples_per_second": 33.61,
|
| 2124 |
+
"eval_steps_per_second": 4.481,
|
| 2125 |
+
"step": 141
|
| 2126 |
+
},
|
| 2127 |
+
{
|
| 2128 |
+
"epoch": 7.473684210526316,
|
| 2129 |
+
"grad_norm": 1.3668967485427856,
|
| 2130 |
+
"learning_rate": 0.0002216530770986795,
|
| 2131 |
+
"loss": 0.0742,
|
| 2132 |
+
"step": 142
|
| 2133 |
+
},
|
| 2134 |
+
{
|
| 2135 |
+
"epoch": 7.473684210526316,
|
| 2136 |
+
"eval_loss": 0.05956079065799713,
|
| 2137 |
+
"eval_runtime": 0.9129,
|
| 2138 |
+
"eval_samples_per_second": 32.861,
|
| 2139 |
+
"eval_steps_per_second": 4.381,
|
| 2140 |
+
"step": 142
|
| 2141 |
+
},
|
| 2142 |
+
{
|
| 2143 |
+
"epoch": 7.526315789473684,
|
| 2144 |
+
"grad_norm": 1.1893479824066162,
|
| 2145 |
+
"learning_rate": 0.0002183351430834358,
|
| 2146 |
+
"loss": 0.0885,
|
| 2147 |
+
"step": 143
|
| 2148 |
+
},
|
| 2149 |
+
{
|
| 2150 |
+
"epoch": 7.526315789473684,
|
| 2151 |
+
"eval_loss": 0.05919176712632179,
|
| 2152 |
+
"eval_runtime": 0.9013,
|
| 2153 |
+
"eval_samples_per_second": 33.285,
|
| 2154 |
+
"eval_steps_per_second": 4.438,
|
| 2155 |
+
"step": 143
|
| 2156 |
+
},
|
| 2157 |
+
{
|
| 2158 |
+
"epoch": 7.578947368421053,
|
| 2159 |
+
"grad_norm": 0.9393155574798584,
|
| 2160 |
+
"learning_rate": 0.0002150228680081079,
|
| 2161 |
+
"loss": 0.069,
|
| 2162 |
+
"step": 144
|
| 2163 |
+
},
|
| 2164 |
+
{
|
| 2165 |
+
"epoch": 7.578947368421053,
|
| 2166 |
+
"eval_loss": 0.055469710379838943,
|
| 2167 |
+
"eval_runtime": 0.8928,
|
| 2168 |
+
"eval_samples_per_second": 33.603,
|
| 2169 |
+
"eval_steps_per_second": 4.48,
|
| 2170 |
+
"step": 144
|
| 2171 |
+
},
|
| 2172 |
+
{
|
| 2173 |
+
"epoch": 7.631578947368421,
|
| 2174 |
+
"grad_norm": 1.1879485845565796,
|
| 2175 |
+
"learning_rate": 0.00021171684382123,
|
| 2176 |
+
"loss": 0.0636,
|
| 2177 |
+
"step": 145
|
| 2178 |
+
},
|
| 2179 |
+
{
|
| 2180 |
+
"epoch": 7.631578947368421,
|
| 2181 |
+
"eval_loss": 0.048830099403858185,
|
| 2182 |
+
"eval_runtime": 0.8962,
|
| 2183 |
+
"eval_samples_per_second": 33.476,
|
| 2184 |
+
"eval_steps_per_second": 4.463,
|
| 2185 |
+
"step": 145
|
| 2186 |
+
},
|
| 2187 |
+
{
|
| 2188 |
+
"epoch": 7.684210526315789,
|
| 2189 |
+
"grad_norm": 1.3696624040603638,
|
| 2190 |
+
"learning_rate": 0.0002084176613542175,
|
| 2191 |
+
"loss": 0.0769,
|
| 2192 |
+
"step": 146
|
| 2193 |
+
},
|
| 2194 |
+
{
|
| 2195 |
+
"epoch": 7.684210526315789,
|
| 2196 |
+
"eval_loss": 0.04780884087085724,
|
| 2197 |
+
"eval_runtime": 0.8944,
|
| 2198 |
+
"eval_samples_per_second": 33.543,
|
| 2199 |
+
"eval_steps_per_second": 4.472,
|
| 2200 |
+
"step": 146
|
| 2201 |
+
},
|
| 2202 |
+
{
|
| 2203 |
+
"epoch": 7.7368421052631575,
|
| 2204 |
+
"grad_norm": 0.8504798412322998,
|
| 2205 |
+
"learning_rate": 0.00020512591021577773,
|
| 2206 |
+
"loss": 0.0452,
|
| 2207 |
+
"step": 147
|
| 2208 |
+
},
|
| 2209 |
+
{
|
| 2210 |
+
"epoch": 7.7368421052631575,
|
| 2211 |
+
"eval_loss": 0.05237739533185959,
|
| 2212 |
+
"eval_runtime": 0.8936,
|
| 2213 |
+
"eval_samples_per_second": 33.572,
|
| 2214 |
+
"eval_steps_per_second": 4.476,
|
| 2215 |
+
"step": 147
|
| 2216 |
+
},
|
| 2217 |
+
{
|
| 2218 |
+
"epoch": 7.7894736842105265,
|
| 2219 |
+
"grad_norm": 1.4475505352020264,
|
| 2220 |
+
"learning_rate": 0.00020184217868653867,
|
| 2221 |
+
"loss": 0.0855,
|
| 2222 |
+
"step": 148
|
| 2223 |
+
},
|
| 2224 |
+
{
|
| 2225 |
+
"epoch": 7.7894736842105265,
|
| 2226 |
+
"eval_loss": 0.04543802887201309,
|
| 2227 |
+
"eval_runtime": 0.896,
|
| 2228 |
+
"eval_samples_per_second": 33.484,
|
| 2229 |
+
"eval_steps_per_second": 4.465,
|
| 2230 |
+
"step": 148
|
| 2231 |
+
},
|
| 2232 |
+
{
|
| 2233 |
+
"epoch": 7.842105263157895,
|
| 2234 |
+
"grad_norm": 1.5789515972137451,
|
| 2235 |
+
"learning_rate": 0.0001985670536139151,
|
| 2236 |
+
"loss": 0.0874,
|
| 2237 |
+
"step": 149
|
| 2238 |
+
},
|
| 2239 |
+
{
|
| 2240 |
+
"epoch": 7.842105263157895,
|
| 2241 |
+
"eval_loss": 0.0420089028775692,
|
| 2242 |
+
"eval_runtime": 0.9085,
|
| 2243 |
+
"eval_samples_per_second": 33.022,
|
| 2244 |
+
"eval_steps_per_second": 4.403,
|
| 2245 |
+
"step": 149
|
| 2246 |
+
},
|
| 2247 |
+
{
|
| 2248 |
+
"epoch": 7.894736842105263,
|
| 2249 |
+
"grad_norm": 0.9716910719871521,
|
| 2250 |
+
"learning_rate": 0.0001953011203072312,
|
| 2251 |
+
"loss": 0.0741,
|
| 2252 |
+
"step": 150
|
| 2253 |
+
},
|
| 2254 |
+
{
|
| 2255 |
+
"epoch": 7.894736842105263,
|
| 2256 |
+
"eval_loss": 0.053930822759866714,
|
| 2257 |
+
"eval_runtime": 0.8925,
|
| 2258 |
+
"eval_samples_per_second": 33.612,
|
| 2259 |
+
"eval_steps_per_second": 4.482,
|
| 2260 |
+
"step": 150
|
| 2261 |
+
},
|
| 2262 |
+
{
|
| 2263 |
+
"epoch": 7.947368421052632,
|
| 2264 |
+
"grad_norm": 1.258216142654419,
|
| 2265 |
+
"learning_rate": 0.00019204496243311792,
|
| 2266 |
+
"loss": 0.0988,
|
| 2267 |
+
"step": 151
|
| 2268 |
+
},
|
| 2269 |
+
{
|
| 2270 |
+
"epoch": 7.947368421052632,
|
| 2271 |
+
"eval_loss": 0.050727710127830505,
|
| 2272 |
+
"eval_runtime": 0.9129,
|
| 2273 |
+
"eval_samples_per_second": 32.861,
|
| 2274 |
+
"eval_steps_per_second": 4.381,
|
| 2275 |
+
"step": 151
|
| 2276 |
+
},
|
| 2277 |
+
{
|
| 2278 |
+
"epoch": 8.0,
|
| 2279 |
+
"grad_norm": 1.6167078018188477,
|
| 2280 |
+
"learning_rate": 0.00018879916191120349,
|
| 2281 |
+
"loss": 0.1526,
|
| 2282 |
+
"step": 152
|
| 2283 |
+
},
|
| 2284 |
+
{
|
| 2285 |
+
"epoch": 8.0,
|
| 2286 |
+
"eval_loss": 0.044940169900655746,
|
| 2287 |
+
"eval_runtime": 0.9149,
|
| 2288 |
+
"eval_samples_per_second": 32.792,
|
| 2289 |
+
"eval_steps_per_second": 4.372,
|
| 2290 |
+
"step": 152
|
| 2291 |
+
},
|
| 2292 |
+
{
|
| 2293 |
+
"epoch": 8.052631578947368,
|
| 2294 |
+
"grad_norm": 0.7703630328178406,
|
| 2295 |
+
"learning_rate": 0.00018556429881011656,
|
| 2296 |
+
"loss": 0.029,
|
| 2297 |
+
"step": 153
|
| 2298 |
+
},
|
| 2299 |
+
{
|
| 2300 |
+
"epoch": 8.052631578947368,
|
| 2301 |
+
"eval_loss": 0.04256557673215866,
|
| 2302 |
+
"eval_runtime": 0.8993,
|
| 2303 |
+
"eval_samples_per_second": 33.359,
|
| 2304 |
+
"eval_steps_per_second": 4.448,
|
| 2305 |
+
"step": 153
|
| 2306 |
+
},
|
| 2307 |
+
{
|
| 2308 |
+
"epoch": 8.105263157894736,
|
| 2309 |
+
"grad_norm": 0.7948728799819946,
|
| 2310 |
+
"learning_rate": 0.0001823409512438203,
|
| 2311 |
+
"loss": 0.0294,
|
| 2312 |
+
"step": 154
|
| 2313 |
+
},
|
| 2314 |
+
{
|
| 2315 |
+
"epoch": 8.105263157894736,
|
| 2316 |
+
"eval_loss": 0.0398668609559536,
|
| 2317 |
+
"eval_runtime": 0.9221,
|
| 2318 |
+
"eval_samples_per_second": 32.536,
|
| 2319 |
+
"eval_steps_per_second": 4.338,
|
| 2320 |
+
"step": 154
|
| 2321 |
+
},
|
| 2322 |
+
{
|
| 2323 |
+
"epoch": 8.157894736842104,
|
| 2324 |
+
"grad_norm": 0.5918542146682739,
|
| 2325 |
+
"learning_rate": 0.00017912969526829559,
|
| 2326 |
+
"loss": 0.0219,
|
| 2327 |
+
"step": 155
|
| 2328 |
+
},
|
| 2329 |
+
{
|
| 2330 |
+
"epoch": 8.157894736842104,
|
| 2331 |
+
"eval_loss": 0.03863578289747238,
|
| 2332 |
+
"eval_runtime": 0.8936,
|
| 2333 |
+
"eval_samples_per_second": 33.573,
|
| 2334 |
+
"eval_steps_per_second": 4.476,
|
| 2335 |
+
"step": 155
|
| 2336 |
+
},
|
| 2337 |
+
{
|
| 2338 |
+
"epoch": 8.210526315789474,
|
| 2339 |
+
"grad_norm": 0.5533296465873718,
|
| 2340 |
+
"learning_rate": 0.00017593110477859153,
|
| 2341 |
+
"loss": 0.0238,
|
| 2342 |
+
"step": 156
|
| 2343 |
+
},
|
| 2344 |
+
{
|
| 2345 |
+
"epoch": 8.210526315789474,
|
| 2346 |
+
"eval_loss": 0.03713521733880043,
|
| 2347 |
+
"eval_runtime": 0.8943,
|
| 2348 |
+
"eval_samples_per_second": 33.547,
|
| 2349 |
+
"eval_steps_per_second": 4.473,
|
| 2350 |
+
"step": 156
|
| 2351 |
+
},
|
| 2352 |
+
{
|
| 2353 |
+
"epoch": 8.263157894736842,
|
| 2354 |
+
"grad_norm": 0.5387775897979736,
|
| 2355 |
+
"learning_rate": 0.00017274575140626317,
|
| 2356 |
+
"loss": 0.0332,
|
| 2357 |
+
"step": 157
|
| 2358 |
+
},
|
| 2359 |
+
{
|
| 2360 |
+
"epoch": 8.263157894736842,
|
| 2361 |
+
"eval_loss": 0.0393383763730526,
|
| 2362 |
+
"eval_runtime": 0.8986,
|
| 2363 |
+
"eval_samples_per_second": 33.384,
|
| 2364 |
+
"eval_steps_per_second": 4.451,
|
| 2365 |
+
"step": 157
|
| 2366 |
+
},
|
| 2367 |
+
{
|
| 2368 |
+
"epoch": 8.31578947368421,
|
| 2369 |
+
"grad_norm": 2.716648578643799,
|
| 2370 |
+
"learning_rate": 0.00016957420441721284,
|
| 2371 |
+
"loss": 0.0508,
|
| 2372 |
+
"step": 158
|
| 2373 |
+
},
|
| 2374 |
+
{
|
| 2375 |
+
"epoch": 8.31578947368421,
|
| 2376 |
+
"eval_loss": 0.039231013506650925,
|
| 2377 |
+
"eval_runtime": 0.911,
|
| 2378 |
+
"eval_samples_per_second": 32.932,
|
| 2379 |
+
"eval_steps_per_second": 4.391,
|
| 2380 |
+
"step": 158
|
| 2381 |
+
},
|
| 2382 |
+
{
|
| 2383 |
+
"epoch": 8.368421052631579,
|
| 2384 |
+
"grad_norm": 0.6262527704238892,
|
| 2385 |
+
"learning_rate": 0.00016641703060995457,
|
| 2386 |
+
"loss": 0.0376,
|
| 2387 |
+
"step": 159
|
| 2388 |
+
},
|
| 2389 |
+
{
|
| 2390 |
+
"epoch": 8.368421052631579,
|
| 2391 |
+
"eval_loss": 0.03573182597756386,
|
| 2392 |
+
"eval_runtime": 0.9178,
|
| 2393 |
+
"eval_samples_per_second": 32.685,
|
| 2394 |
+
"eval_steps_per_second": 4.358,
|
| 2395 |
+
"step": 159
|
| 2396 |
+
},
|
| 2397 |
+
{
|
| 2398 |
+
"epoch": 8.421052631578947,
|
| 2399 |
+
"grad_norm": 1.1656262874603271,
|
| 2400 |
+
"learning_rate": 0.00016327479421431983,
|
| 2401 |
+
"loss": 0.0613,
|
| 2402 |
+
"step": 160
|
| 2403 |
+
},
|
| 2404 |
+
{
|
| 2405 |
+
"epoch": 8.421052631578947,
|
| 2406 |
+
"eval_loss": 0.029768355190753937,
|
| 2407 |
+
"eval_runtime": 0.8926,
|
| 2408 |
+
"eval_samples_per_second": 33.61,
|
| 2409 |
+
"eval_steps_per_second": 4.481,
|
| 2410 |
+
"step": 160
|
| 2411 |
+
},
|
| 2412 |
+
{
|
| 2413 |
+
"epoch": 8.473684210526315,
|
| 2414 |
+
"grad_norm": 0.9731020927429199,
|
| 2415 |
+
"learning_rate": 0.00016014805679062183,
|
| 2416 |
+
"loss": 0.0755,
|
| 2417 |
+
"step": 161
|
| 2418 |
+
},
|
| 2419 |
+
{
|
| 2420 |
+
"epoch": 8.473684210526315,
|
| 2421 |
+
"eval_loss": 0.022336162626743317,
|
| 2422 |
+
"eval_runtime": 0.9064,
|
| 2423 |
+
"eval_samples_per_second": 33.097,
|
| 2424 |
+
"eval_steps_per_second": 4.413,
|
| 2425 |
+
"step": 161
|
| 2426 |
+
},
|
| 2427 |
+
{
|
| 2428 |
+
"epoch": 8.526315789473685,
|
| 2429 |
+
"grad_norm": 0.9505934119224548,
|
| 2430 |
+
"learning_rate": 0.0001570373771292967,
|
| 2431 |
+
"loss": 0.0592,
|
| 2432 |
+
"step": 162
|
| 2433 |
+
},
|
| 2434 |
+
{
|
| 2435 |
+
"epoch": 8.526315789473685,
|
| 2436 |
+
"eval_loss": 0.019842755049467087,
|
| 2437 |
+
"eval_runtime": 0.9099,
|
| 2438 |
+
"eval_samples_per_second": 32.97,
|
| 2439 |
+
"eval_steps_per_second": 4.396,
|
| 2440 |
+
"step": 162
|
| 2441 |
+
},
|
| 2442 |
+
{
|
| 2443 |
+
"epoch": 8.578947368421053,
|
| 2444 |
+
"grad_norm": 0.709037184715271,
|
| 2445 |
+
"learning_rate": 0.00015394331115104075,
|
| 2446 |
+
"loss": 0.0386,
|
| 2447 |
+
"step": 163
|
| 2448 |
+
},
|
| 2449 |
+
{
|
| 2450 |
+
"epoch": 8.578947368421053,
|
| 2451 |
+
"eval_loss": 0.019390322268009186,
|
| 2452 |
+
"eval_runtime": 0.8955,
|
| 2453 |
+
"eval_samples_per_second": 33.501,
|
| 2454 |
+
"eval_steps_per_second": 4.467,
|
| 2455 |
+
"step": 163
|
| 2456 |
+
},
|
| 2457 |
+
{
|
| 2458 |
+
"epoch": 8.631578947368421,
|
| 2459 |
+
"grad_norm": 1.0144383907318115,
|
| 2460 |
+
"learning_rate": 0.00015086641180745932,
|
| 2461 |
+
"loss": 0.0392,
|
| 2462 |
+
"step": 164
|
| 2463 |
+
},
|
| 2464 |
+
{
|
| 2465 |
+
"epoch": 8.631578947368421,
|
| 2466 |
+
"eval_loss": 0.018627820536494255,
|
| 2467 |
+
"eval_runtime": 0.8942,
|
| 2468 |
+
"eval_samples_per_second": 33.549,
|
| 2469 |
+
"eval_steps_per_second": 4.473,
|
| 2470 |
+
"step": 164
|
| 2471 |
+
},
|
| 2472 |
+
{
|
| 2473 |
+
"epoch": 8.68421052631579,
|
| 2474 |
+
"grad_norm": 1.3723385334014893,
|
| 2475 |
+
"learning_rate": 0.00014780722898224708,
|
| 2476 |
+
"loss": 0.0286,
|
| 2477 |
+
"step": 165
|
| 2478 |
+
},
|
| 2479 |
+
{
|
| 2480 |
+
"epoch": 8.68421052631579,
|
| 2481 |
+
"eval_loss": 0.015356449410319328,
|
| 2482 |
+
"eval_runtime": 0.8932,
|
| 2483 |
+
"eval_samples_per_second": 33.586,
|
| 2484 |
+
"eval_steps_per_second": 4.478,
|
| 2485 |
+
"step": 165
|
| 2486 |
+
},
|
| 2487 |
+
{
|
| 2488 |
+
"epoch": 8.736842105263158,
|
| 2489 |
+
"grad_norm": 0.6372384428977966,
|
| 2490 |
+
"learning_rate": 0.0001447663093929163,
|
| 2491 |
+
"loss": 0.0425,
|
| 2492 |
+
"step": 166
|
| 2493 |
+
},
|
| 2494 |
+
{
|
| 2495 |
+
"epoch": 8.736842105263158,
|
| 2496 |
+
"eval_loss": 0.015127343125641346,
|
| 2497 |
+
"eval_runtime": 0.9004,
|
| 2498 |
+
"eval_samples_per_second": 33.319,
|
| 2499 |
+
"eval_steps_per_second": 4.443,
|
| 2500 |
+
"step": 166
|
| 2501 |
+
},
|
| 2502 |
+
{
|
| 2503 |
+
"epoch": 8.789473684210526,
|
| 2504 |
+
"grad_norm": 0.7628927826881409,
|
| 2505 |
+
"learning_rate": 0.00014174419649309089,
|
| 2506 |
+
"loss": 0.0218,
|
| 2507 |
+
"step": 167
|
| 2508 |
+
},
|
| 2509 |
+
{
|
| 2510 |
+
"epoch": 8.789473684210526,
|
| 2511 |
+
"eval_loss": 0.015446596778929234,
|
| 2512 |
+
"eval_runtime": 0.8932,
|
| 2513 |
+
"eval_samples_per_second": 33.586,
|
| 2514 |
+
"eval_steps_per_second": 4.478,
|
| 2515 |
+
"step": 167
|
| 2516 |
+
},
|
| 2517 |
+
{
|
| 2518 |
+
"epoch": 8.842105263157894,
|
| 2519 |
+
"grad_norm": 0.7694376111030579,
|
| 2520 |
+
"learning_rate": 0.00013874143037538418,
|
| 2521 |
+
"loss": 0.0251,
|
| 2522 |
+
"step": 168
|
| 2523 |
+
},
|
| 2524 |
+
{
|
| 2525 |
+
"epoch": 8.842105263157894,
|
| 2526 |
+
"eval_loss": 0.01555707585066557,
|
| 2527 |
+
"eval_runtime": 0.9268,
|
| 2528 |
+
"eval_samples_per_second": 32.368,
|
| 2529 |
+
"eval_steps_per_second": 4.316,
|
| 2530 |
+
"step": 168
|
| 2531 |
+
},
|
| 2532 |
+
{
|
| 2533 |
+
"epoch": 8.894736842105264,
|
| 2534 |
+
"grad_norm": 0.7292389869689941,
|
| 2535 |
+
"learning_rate": 0.0001357585476748766,
|
| 2536 |
+
"loss": 0.0345,
|
| 2537 |
+
"step": 169
|
| 2538 |
+
},
|
| 2539 |
+
{
|
| 2540 |
+
"epoch": 8.894736842105264,
|
| 2541 |
+
"eval_loss": 0.014117183163762093,
|
| 2542 |
+
"eval_runtime": 0.8989,
|
| 2543 |
+
"eval_samples_per_second": 33.374,
|
| 2544 |
+
"eval_steps_per_second": 4.45,
|
| 2545 |
+
"step": 169
|
| 2546 |
+
},
|
| 2547 |
+
{
|
| 2548 |
+
"epoch": 8.947368421052632,
|
| 2549 |
+
"grad_norm": 0.7417434453964233,
|
| 2550 |
+
"learning_rate": 0.00013279608147321223,
|
| 2551 |
+
"loss": 0.0355,
|
| 2552 |
+
"step": 170
|
| 2553 |
+
},
|
| 2554 |
+
{
|
| 2555 |
+
"epoch": 8.947368421052632,
|
| 2556 |
+
"eval_loss": 0.01502351462841034,
|
| 2557 |
+
"eval_runtime": 0.8919,
|
| 2558 |
+
"eval_samples_per_second": 33.634,
|
| 2559 |
+
"eval_steps_per_second": 4.485,
|
| 2560 |
+
"step": 170
|
| 2561 |
+
},
|
| 2562 |
+
{
|
| 2563 |
+
"epoch": 9.0,
|
| 2564 |
+
"grad_norm": 1.0023473501205444,
|
| 2565 |
+
"learning_rate": 0.00012985456120332905,
|
| 2566 |
+
"loss": 0.0463,
|
| 2567 |
+
"step": 171
|
| 2568 |
+
},
|
| 2569 |
+
{
|
| 2570 |
+
"epoch": 9.0,
|
| 2571 |
+
"eval_loss": 0.015432776883244514,
|
| 2572 |
+
"eval_runtime": 0.8928,
|
| 2573 |
+
"eval_samples_per_second": 33.604,
|
| 2574 |
+
"eval_steps_per_second": 4.481,
|
| 2575 |
+
"step": 171
|
| 2576 |
+
},
|
| 2577 |
+
{
|
| 2578 |
+
"epoch": 9.052631578947368,
|
| 2579 |
+
"grad_norm": 0.9472024440765381,
|
| 2580 |
+
"learning_rate": 0.00012693451255484312,
|
| 2581 |
+
"loss": 0.0164,
|
| 2582 |
+
"step": 172
|
| 2583 |
+
},
|
| 2584 |
+
{
|
| 2585 |
+
"epoch": 9.052631578947368,
|
| 2586 |
+
"eval_loss": 0.015417199581861496,
|
| 2587 |
+
"eval_runtime": 0.9209,
|
| 2588 |
+
"eval_samples_per_second": 32.576,
|
| 2589 |
+
"eval_steps_per_second": 4.344,
|
| 2590 |
+
"step": 172
|
| 2591 |
+
},
|
| 2592 |
+
{
|
| 2593 |
+
"epoch": 9.105263157894736,
|
| 2594 |
+
"grad_norm": 0.48799633979797363,
|
| 2595 |
+
"learning_rate": 0.00012403645738009997,
|
| 2596 |
+
"loss": 0.0112,
|
| 2597 |
+
"step": 173
|
| 2598 |
+
},
|
| 2599 |
+
{
|
| 2600 |
+
"epoch": 9.105263157894736,
|
| 2601 |
+
"eval_loss": 0.015746938064694405,
|
| 2602 |
+
"eval_runtime": 0.903,
|
| 2603 |
+
"eval_samples_per_second": 33.221,
|
| 2604 |
+
"eval_steps_per_second": 4.43,
|
| 2605 |
+
"step": 173
|
| 2606 |
+
},
|
| 2607 |
+
{
|
| 2608 |
+
"epoch": 9.157894736842104,
|
| 2609 |
+
"grad_norm": 0.38101622462272644,
|
| 2610 |
+
"learning_rate": 0.00012116091360091261,
|
| 2611 |
+
"loss": 0.0107,
|
| 2612 |
+
"step": 174
|
| 2613 |
+
},
|
| 2614 |
+
{
|
| 2615 |
+
"epoch": 9.157894736842104,
|
| 2616 |
+
"eval_loss": 0.016201062127947807,
|
| 2617 |
+
"eval_runtime": 0.9098,
|
| 2618 |
+
"eval_samples_per_second": 32.973,
|
| 2619 |
+
"eval_steps_per_second": 4.396,
|
| 2620 |
+
"step": 174
|
| 2621 |
+
},
|
| 2622 |
+
{
|
| 2623 |
+
"epoch": 9.210526315789474,
|
| 2624 |
+
"grad_norm": 0.5602852702140808,
|
| 2625 |
+
"learning_rate": 0.00011830839511600211,
|
| 2626 |
+
"loss": 0.0214,
|
| 2627 |
+
"step": 175
|
| 2628 |
+
},
|
| 2629 |
+
{
|
| 2630 |
+
"epoch": 9.210526315789474,
|
| 2631 |
+
"eval_loss": 0.01637989468872547,
|
| 2632 |
+
"eval_runtime": 0.8978,
|
| 2633 |
+
"eval_samples_per_second": 33.416,
|
| 2634 |
+
"eval_steps_per_second": 4.455,
|
| 2635 |
+
"step": 175
|
| 2636 |
+
},
|
| 2637 |
+
{
|
| 2638 |
+
"epoch": 9.263157894736842,
|
| 2639 |
+
"grad_norm": 0.48026910424232483,
|
| 2640 |
+
"learning_rate": 0.00011547941170915685,
|
| 2641 |
+
"loss": 0.0159,
|
| 2642 |
+
"step": 176
|
| 2643 |
+
},
|
| 2644 |
+
{
|
| 2645 |
+
"epoch": 9.263157894736842,
|
| 2646 |
+
"eval_loss": 0.01590169034898281,
|
| 2647 |
+
"eval_runtime": 0.8929,
|
| 2648 |
+
"eval_samples_per_second": 33.599,
|
| 2649 |
+
"eval_steps_per_second": 4.48,
|
| 2650 |
+
"step": 176
|
| 2651 |
+
},
|
| 2652 |
+
{
|
| 2653 |
+
"epoch": 9.31578947368421,
|
| 2654 |
+
"grad_norm": 0.42628395557403564,
|
| 2655 |
+
"learning_rate": 0.00011267446895812702,
|
| 2656 |
+
"loss": 0.0103,
|
| 2657 |
+
"step": 177
|
| 2658 |
+
},
|
| 2659 |
+
{
|
| 2660 |
+
"epoch": 9.31578947368421,
|
| 2661 |
+
"eval_loss": 0.016489733010530472,
|
| 2662 |
+
"eval_runtime": 0.9067,
|
| 2663 |
+
"eval_samples_per_second": 33.087,
|
| 2664 |
+
"eval_steps_per_second": 4.412,
|
| 2665 |
+
"step": 177
|
| 2666 |
+
},
|
| 2667 |
+
{
|
| 2668 |
+
"epoch": 9.368421052631579,
|
| 2669 |
+
"grad_norm": 0.31815841794013977,
|
| 2670 |
+
"learning_rate": 0.0001098940681442713,
|
| 2671 |
+
"loss": 0.0127,
|
| 2672 |
+
"step": 178
|
| 2673 |
+
},
|
| 2674 |
+
{
|
| 2675 |
+
"epoch": 9.368421052631579,
|
| 2676 |
+
"eval_loss": 0.016672790050506592,
|
| 2677 |
+
"eval_runtime": 0.9121,
|
| 2678 |
+
"eval_samples_per_second": 32.892,
|
| 2679 |
+
"eval_steps_per_second": 4.386,
|
| 2680 |
+
"step": 178
|
| 2681 |
+
},
|
| 2682 |
+
{
|
| 2683 |
+
"epoch": 9.421052631578947,
|
| 2684 |
+
"grad_norm": 0.9146761894226074,
|
| 2685 |
+
"learning_rate": 0.00010713870616297092,
|
| 2686 |
+
"loss": 0.0272,
|
| 2687 |
+
"step": 179
|
| 2688 |
+
},
|
| 2689 |
+
{
|
| 2690 |
+
"epoch": 9.421052631578947,
|
| 2691 |
+
"eval_loss": 0.016623031347990036,
|
| 2692 |
+
"eval_runtime": 0.8922,
|
| 2693 |
+
"eval_samples_per_second": 33.624,
|
| 2694 |
+
"eval_steps_per_second": 4.483,
|
| 2695 |
+
"step": 179
|
| 2696 |
+
},
|
| 2697 |
+
{
|
| 2698 |
+
"epoch": 9.473684210526315,
|
| 2699 |
+
"grad_norm": 0.7005583643913269,
|
| 2700 |
+
"learning_rate": 0.00010440887543482746,
|
| 2701 |
+
"loss": 0.0316,
|
| 2702 |
+
"step": 180
|
| 2703 |
+
},
|
| 2704 |
+
{
|
| 2705 |
+
"epoch": 9.473684210526315,
|
| 2706 |
+
"eval_loss": 0.01275827456265688,
|
| 2707 |
+
"eval_runtime": 0.8927,
|
| 2708 |
+
"eval_samples_per_second": 33.606,
|
| 2709 |
+
"eval_steps_per_second": 4.481,
|
| 2710 |
+
"step": 180
|
| 2711 |
+
},
|
| 2712 |
+
{
|
| 2713 |
+
"epoch": 9.526315789473685,
|
| 2714 |
+
"grad_norm": 0.650211751461029,
|
| 2715 |
+
"learning_rate": 0.0001017050638176612,
|
| 2716 |
+
"loss": 0.026,
|
| 2717 |
+
"step": 181
|
| 2718 |
+
},
|
| 2719 |
+
{
|
| 2720 |
+
"epoch": 9.526315789473685,
|
| 2721 |
+
"eval_loss": 0.00972173921763897,
|
| 2722 |
+
"eval_runtime": 0.8929,
|
| 2723 |
+
"eval_samples_per_second": 33.597,
|
| 2724 |
+
"eval_steps_per_second": 4.48,
|
| 2725 |
+
"step": 181
|
| 2726 |
+
},
|
| 2727 |
+
{
|
| 2728 |
+
"epoch": 9.578947368421053,
|
| 2729 |
+
"grad_norm": 0.6491077542304993,
|
| 2730 |
+
"learning_rate": 9.902775451932386e-05,
|
| 2731 |
+
"loss": 0.0118,
|
| 2732 |
+
"step": 182
|
| 2733 |
+
},
|
| 2734 |
+
{
|
| 2735 |
+
"epoch": 9.578947368421053,
|
| 2736 |
+
"eval_loss": 0.008100698702037334,
|
| 2737 |
+
"eval_runtime": 0.8982,
|
| 2738 |
+
"eval_samples_per_second": 33.399,
|
| 2739 |
+
"eval_steps_per_second": 4.453,
|
| 2740 |
+
"step": 182
|
| 2741 |
+
},
|
| 2742 |
+
{
|
| 2743 |
+
"epoch": 9.631578947368421,
|
| 2744 |
+
"grad_norm": 0.5663555264472961,
|
| 2745 |
+
"learning_rate": 9.637742601134286e-05,
|
| 2746 |
+
"loss": 0.0179,
|
| 2747 |
+
"step": 183
|
| 2748 |
+
},
|
| 2749 |
+
{
|
| 2750 |
+
"epoch": 9.631578947368421,
|
| 2751 |
+
"eval_loss": 0.007195114623755217,
|
| 2752 |
+
"eval_runtime": 0.8943,
|
| 2753 |
+
"eval_samples_per_second": 33.547,
|
| 2754 |
+
"eval_steps_per_second": 4.473,
|
| 2755 |
+
"step": 183
|
| 2756 |
+
},
|
| 2757 |
+
{
|
| 2758 |
+
"epoch": 9.68421052631579,
|
| 2759 |
+
"grad_norm": 0.45350518822669983,
|
| 2760 |
+
"learning_rate": 9.375455194341214e-05,
|
| 2761 |
+
"loss": 0.0133,
|
| 2762 |
+
"step": 184
|
| 2763 |
+
},
|
| 2764 |
+
{
|
| 2765 |
+
"epoch": 9.68421052631579,
|
| 2766 |
+
"eval_loss": 0.005673492327332497,
|
| 2767 |
+
"eval_runtime": 0.8957,
|
| 2768 |
+
"eval_samples_per_second": 33.493,
|
| 2769 |
+
"eval_steps_per_second": 4.466,
|
| 2770 |
+
"step": 184
|
| 2771 |
+
},
|
| 2772 |
+
{
|
| 2773 |
+
"epoch": 9.736842105263158,
|
| 2774 |
+
"grad_norm": 0.4562082886695862,
|
| 2775 |
+
"learning_rate": 9.11596010587441e-05,
|
| 2776 |
+
"loss": 0.0116,
|
| 2777 |
+
"step": 185
|
| 2778 |
+
},
|
| 2779 |
+
{
|
| 2780 |
+
"epoch": 9.736842105263158,
|
| 2781 |
+
"eval_loss": 0.005512699484825134,
|
| 2782 |
+
"eval_runtime": 0.8942,
|
| 2783 |
+
"eval_samples_per_second": 33.551,
|
| 2784 |
+
"eval_steps_per_second": 4.473,
|
| 2785 |
+
"step": 185
|
| 2786 |
+
},
|
| 2787 |
+
{
|
| 2788 |
+
"epoch": 9.789473684210526,
|
| 2789 |
+
"grad_norm": 0.4965287446975708,
|
| 2790 |
+
"learning_rate": 8.85930371102994e-05,
|
| 2791 |
+
"loss": 0.0175,
|
| 2792 |
+
"step": 186
|
| 2793 |
+
},
|
| 2794 |
+
{
|
| 2795 |
+
"epoch": 9.789473684210526,
|
| 2796 |
+
"eval_loss": 0.005058939103037119,
|
| 2797 |
+
"eval_runtime": 0.8924,
|
| 2798 |
+
"eval_samples_per_second": 33.619,
|
| 2799 |
+
"eval_steps_per_second": 4.482,
|
| 2800 |
+
"step": 186
|
| 2801 |
+
},
|
| 2802 |
+
{
|
| 2803 |
+
"epoch": 9.842105263157894,
|
| 2804 |
+
"grad_norm": 0.4823167324066162,
|
| 2805 |
+
"learning_rate": 8.605531877790762e-05,
|
| 2806 |
+
"loss": 0.0156,
|
| 2807 |
+
"step": 187
|
| 2808 |
+
},
|
| 2809 |
+
{
|
| 2810 |
+
"epoch": 9.842105263157894,
|
| 2811 |
+
"eval_loss": 0.004006177186965942,
|
| 2812 |
+
"eval_runtime": 0.8937,
|
| 2813 |
+
"eval_samples_per_second": 33.568,
|
| 2814 |
+
"eval_steps_per_second": 4.476,
|
| 2815 |
+
"step": 187
|
| 2816 |
+
},
|
| 2817 |
+
{
|
| 2818 |
+
"epoch": 9.894736842105264,
|
| 2819 |
+
"grad_norm": 0.5879040360450745,
|
| 2820 |
+
"learning_rate": 8.354689958629513e-05,
|
| 2821 |
+
"loss": 0.0147,
|
| 2822 |
+
"step": 188
|
| 2823 |
+
},
|
| 2824 |
+
{
|
| 2825 |
+
"epoch": 9.894736842105264,
|
| 2826 |
+
"eval_loss": 0.003014415269717574,
|
| 2827 |
+
"eval_runtime": 0.8965,
|
| 2828 |
+
"eval_samples_per_second": 33.465,
|
| 2829 |
+
"eval_steps_per_second": 4.462,
|
| 2830 |
+
"step": 188
|
| 2831 |
+
},
|
| 2832 |
+
{
|
| 2833 |
+
"epoch": 9.947368421052632,
|
| 2834 |
+
"grad_norm": 0.4576377868652344,
|
| 2835 |
+
"learning_rate": 8.106822782403376e-05,
|
| 2836 |
+
"loss": 0.0095,
|
| 2837 |
+
"step": 189
|
| 2838 |
+
},
|
| 2839 |
+
{
|
| 2840 |
+
"epoch": 9.947368421052632,
|
| 2841 |
+
"eval_loss": 0.002746094949543476,
|
| 2842 |
+
"eval_runtime": 0.8942,
|
| 2843 |
+
"eval_samples_per_second": 33.55,
|
| 2844 |
+
"eval_steps_per_second": 4.473,
|
| 2845 |
+
"step": 189
|
| 2846 |
+
},
|
| 2847 |
+
{
|
| 2848 |
+
"epoch": 10.0,
|
| 2849 |
+
"grad_norm": 0.3874748647212982,
|
| 2850 |
+
"learning_rate": 7.861974646342596e-05,
|
| 2851 |
+
"loss": 0.0065,
|
| 2852 |
+
"step": 190
|
| 2853 |
+
},
|
| 2854 |
+
{
|
| 2855 |
+
"epoch": 10.0,
|
| 2856 |
+
"eval_loss": 0.0022730662021785975,
|
| 2857 |
+
"eval_runtime": 0.8918,
|
| 2858 |
+
"eval_samples_per_second": 33.639,
|
| 2859 |
+
"eval_steps_per_second": 4.485,
|
| 2860 |
+
"step": 190
|
| 2861 |
+
},
|
| 2862 |
+
{
|
| 2863 |
+
"epoch": 10.052631578947368,
|
| 2864 |
+
"grad_norm": 0.19529208540916443,
|
| 2865 |
+
"learning_rate": 7.620189308133943e-05,
|
| 2866 |
+
"loss": 0.0038,
|
| 2867 |
+
"step": 191
|
| 2868 |
+
},
|
| 2869 |
+
{
|
| 2870 |
+
"epoch": 10.052631578947368,
|
| 2871 |
+
"eval_loss": 0.0020791899878531694,
|
| 2872 |
+
"eval_runtime": 0.8957,
|
| 2873 |
+
"eval_samples_per_second": 33.495,
|
| 2874 |
+
"eval_steps_per_second": 4.466,
|
| 2875 |
+
"step": 191
|
| 2876 |
+
},
|
| 2877 |
+
{
|
| 2878 |
+
"epoch": 10.105263157894736,
|
| 2879 |
+
"grad_norm": 0.11527393758296967,
|
| 2880 |
+
"learning_rate": 7.381509978100626e-05,
|
| 2881 |
+
"loss": 0.0022,
|
| 2882 |
+
"step": 192
|
| 2883 |
+
},
|
| 2884 |
+
{
|
| 2885 |
+
"epoch": 10.105263157894736,
|
| 2886 |
+
"eval_loss": 0.002016394166275859,
|
| 2887 |
+
"eval_runtime": 0.8996,
|
| 2888 |
+
"eval_samples_per_second": 33.347,
|
| 2889 |
+
"eval_steps_per_second": 4.446,
|
| 2890 |
+
"step": 192
|
| 2891 |
+
},
|
| 2892 |
+
{
|
| 2893 |
+
"epoch": 10.157894736842104,
|
| 2894 |
+
"grad_norm": 0.15251131355762482,
|
| 2895 |
+
"learning_rate": 7.145979311479986e-05,
|
| 2896 |
+
"loss": 0.003,
|
| 2897 |
+
"step": 193
|
| 2898 |
+
},
|
| 2899 |
+
{
|
| 2900 |
+
"epoch": 10.157894736842104,
|
| 2901 |
+
"eval_loss": 0.0021317771170288324,
|
| 2902 |
+
"eval_runtime": 0.8932,
|
| 2903 |
+
"eval_samples_per_second": 33.585,
|
| 2904 |
+
"eval_steps_per_second": 4.478,
|
| 2905 |
+
"step": 193
|
| 2906 |
+
},
|
| 2907 |
+
{
|
| 2908 |
+
"epoch": 10.210526315789474,
|
| 2909 |
+
"grad_norm": 0.16482071578502655,
|
| 2910 |
+
"learning_rate": 6.913639400800489e-05,
|
| 2911 |
+
"loss": 0.0024,
|
| 2912 |
+
"step": 194
|
| 2913 |
+
},
|
| 2914 |
+
{
|
| 2915 |
+
"epoch": 10.210526315789474,
|
| 2916 |
+
"eval_loss": 0.0021966167259961367,
|
| 2917 |
+
"eval_runtime": 0.8945,
|
| 2918 |
+
"eval_samples_per_second": 33.537,
|
| 2919 |
+
"eval_steps_per_second": 4.472,
|
| 2920 |
+
"step": 194
|
| 2921 |
+
},
|
| 2922 |
+
{
|
| 2923 |
+
"epoch": 10.263157894736842,
|
| 2924 |
+
"grad_norm": 0.14208117127418518,
|
| 2925 |
+
"learning_rate": 6.684531768359173e-05,
|
| 2926 |
+
"loss": 0.002,
|
| 2927 |
+
"step": 195
|
| 2928 |
+
},
|
| 2929 |
+
{
|
| 2930 |
+
"epoch": 10.263157894736842,
|
| 2931 |
+
"eval_loss": 0.0022034423891454935,
|
| 2932 |
+
"eval_runtime": 0.8952,
|
| 2933 |
+
"eval_samples_per_second": 33.511,
|
| 2934 |
+
"eval_steps_per_second": 4.468,
|
| 2935 |
+
"step": 195
|
| 2936 |
+
},
|
| 2937 |
+
{
|
| 2938 |
+
"epoch": 10.31578947368421,
|
| 2939 |
+
"grad_norm": 0.11844911426305771,
|
| 2940 |
+
"learning_rate": 6.458697358801061e-05,
|
| 2941 |
+
"loss": 0.0018,
|
| 2942 |
+
"step": 196
|
| 2943 |
+
},
|
| 2944 |
+
{
|
| 2945 |
+
"epoch": 10.31578947368421,
|
| 2946 |
+
"eval_loss": 0.002191495383158326,
|
| 2947 |
+
"eval_runtime": 0.8926,
|
| 2948 |
+
"eval_samples_per_second": 33.611,
|
| 2949 |
+
"eval_steps_per_second": 4.481,
|
| 2950 |
+
"step": 196
|
| 2951 |
+
},
|
| 2952 |
+
{
|
| 2953 |
+
"epoch": 10.368421052631579,
|
| 2954 |
+
"grad_norm": 0.25322437286376953,
|
| 2955 |
+
"learning_rate": 6.236176531801813e-05,
|
| 2956 |
+
"loss": 0.0049,
|
| 2957 |
+
"step": 197
|
| 2958 |
+
},
|
| 2959 |
+
{
|
| 2960 |
+
"epoch": 10.368421052631579,
|
| 2961 |
+
"eval_loss": 0.0022686992306262255,
|
| 2962 |
+
"eval_runtime": 0.8949,
|
| 2963 |
+
"eval_samples_per_second": 33.525,
|
| 2964 |
+
"eval_steps_per_second": 4.47,
|
| 2965 |
+
"step": 197
|
| 2966 |
+
},
|
| 2967 |
+
{
|
| 2968 |
+
"epoch": 10.421052631578947,
|
| 2969 |
+
"grad_norm": 0.29156965017318726,
|
| 2970 |
+
"learning_rate": 6.017009054854858e-05,
|
| 2971 |
+
"loss": 0.0045,
|
| 2972 |
+
"step": 198
|
| 2973 |
+
},
|
| 2974 |
+
{
|
| 2975 |
+
"epoch": 10.421052631578947,
|
| 2976 |
+
"eval_loss": 0.002286201808601618,
|
| 2977 |
+
"eval_runtime": 0.8929,
|
| 2978 |
+
"eval_samples_per_second": 33.597,
|
| 2979 |
+
"eval_steps_per_second": 4.48,
|
| 2980 |
+
"step": 198
|
| 2981 |
+
},
|
| 2982 |
+
{
|
| 2983 |
+
"epoch": 10.473684210526315,
|
| 2984 |
+
"grad_norm": 0.3855668306350708,
|
| 2985 |
+
"learning_rate": 5.801234096164468e-05,
|
| 2986 |
+
"loss": 0.0034,
|
| 2987 |
+
"step": 199
|
| 2988 |
+
},
|
| 2989 |
+
{
|
| 2990 |
+
"epoch": 10.473684210526315,
|
| 2991 |
+
"eval_loss": 0.0018616730812937021,
|
| 2992 |
+
"eval_runtime": 0.894,
|
| 2993 |
+
"eval_samples_per_second": 33.558,
|
| 2994 |
+
"eval_steps_per_second": 4.474,
|
| 2995 |
+
"step": 199
|
| 2996 |
+
},
|
| 2997 |
+
{
|
| 2998 |
+
"epoch": 10.526315789473685,
|
| 2999 |
+
"grad_norm": 0.2883719205856323,
|
| 3000 |
+
"learning_rate": 5.58889021764582e-05,
|
| 3001 |
+
"loss": 0.0044,
|
| 3002 |
+
"step": 200
|
| 3003 |
+
},
|
| 3004 |
+
{
|
| 3005 |
+
"epoch": 10.526315789473685,
|
| 3006 |
+
"eval_loss": 0.0016098986379802227,
|
| 3007 |
+
"eval_runtime": 0.8994,
|
| 3008 |
+
"eval_samples_per_second": 33.357,
|
| 3009 |
+
"eval_steps_per_second": 4.448,
|
| 3010 |
+
"step": 200
|
| 3011 |
}
|
| 3012 |
],
|
| 3013 |
"logging_steps": 1,
|
| 3014 |
+
"max_steps": 250,
|
| 3015 |
"num_input_tokens_seen": 0,
|
| 3016 |
+
"num_train_epochs": 14,
|
| 3017 |
"save_steps": 5,
|
| 3018 |
"stateful_callbacks": {
|
| 3019 |
"TrainerControl": {
|
|
|
|
| 3027 |
"attributes": {}
|
| 3028 |
}
|
| 3029 |
},
|
| 3030 |
+
"total_flos": 8525733259253760.0,
|
| 3031 |
"train_batch_size": 1,
|
| 3032 |
"trial_name": null,
|
| 3033 |
"trial_params": null
|