Training in progress, step 1050, checkpoint
Browse files
last-checkpoint/adapter_model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 527048968
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ffadfcecebb0222aa110b4c04dc7dcf449445cb1257ad373d2c573864d4946f4
|
| 3 |
size 527048968
|
last-checkpoint/optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1054135994
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:48049032ce30e690a0e20e1c74a40a3d0d491ae37296b800adf28f948ac25c62
|
| 3 |
size 1054135994
|
last-checkpoint/rng_state.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 14244
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5033c3d16bccb36ffaebb5ce29ad27ffe0f183d458fcf132e776197413917bfa
|
| 3 |
size 14244
|
last-checkpoint/scheduler.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1064
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:15c1d156a1458cb6f00d5682720ac01d76d87fde2ae92aa38769a7cbc234762e
|
| 3 |
size 1064
|
last-checkpoint/trainer_state.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
"best_metric": 0.7167766094207764,
|
| 3 |
"best_model_checkpoint": "./output/checkpoint-450",
|
| 4 |
-
"epoch":
|
| 5 |
"eval_steps": 150,
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
@@ -346,6 +346,482 @@
|
|
| 346 |
"eval_samples_per_second": 24.61,
|
| 347 |
"eval_steps_per_second": 24.61,
|
| 348 |
"step": 450
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 349 |
}
|
| 350 |
],
|
| 351 |
"logging_steps": 10,
|
|
@@ -365,7 +841,7 @@
|
|
| 365 |
"attributes": {}
|
| 366 |
}
|
| 367 |
},
|
| 368 |
-
"total_flos":
|
| 369 |
"train_batch_size": 4,
|
| 370 |
"trial_name": null,
|
| 371 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": 0.7167766094207764,
|
| 3 |
"best_model_checkpoint": "./output/checkpoint-450",
|
| 4 |
+
"epoch": 45.65217391304348,
|
| 5 |
"eval_steps": 150,
|
| 6 |
+
"global_step": 1050,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
|
|
| 346 |
"eval_samples_per_second": 24.61,
|
| 347 |
"eval_steps_per_second": 24.61,
|
| 348 |
"step": 450
|
| 349 |
+
},
|
| 350 |
+
{
|
| 351 |
+
"epoch": 20.0,
|
| 352 |
+
"grad_norm": 2.1287033557891846,
|
| 353 |
+
"learning_rate": 2.9996147467351856e-05,
|
| 354 |
+
"loss": 0.5144,
|
| 355 |
+
"step": 460
|
| 356 |
+
},
|
| 357 |
+
{
|
| 358 |
+
"epoch": 20.434782608695652,
|
| 359 |
+
"grad_norm": 1.2749507427215576,
|
| 360 |
+
"learning_rate": 2.9995930474939773e-05,
|
| 361 |
+
"loss": 0.4784,
|
| 362 |
+
"step": 470
|
| 363 |
+
},
|
| 364 |
+
{
|
| 365 |
+
"epoch": 20.869565217391305,
|
| 366 |
+
"grad_norm": 1.4720127582550049,
|
| 367 |
+
"learning_rate": 2.9995707538619975e-05,
|
| 368 |
+
"loss": 0.5705,
|
| 369 |
+
"step": 480
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 21.304347826086957,
|
| 373 |
+
"grad_norm": 1.3214296102523804,
|
| 374 |
+
"learning_rate": 2.9995478658480822e-05,
|
| 375 |
+
"loss": 0.516,
|
| 376 |
+
"step": 490
|
| 377 |
+
},
|
| 378 |
+
{
|
| 379 |
+
"epoch": 21.73913043478261,
|
| 380 |
+
"grad_norm": 1.2429661750793457,
|
| 381 |
+
"learning_rate": 2.9995243834613043e-05,
|
| 382 |
+
"loss": 0.521,
|
| 383 |
+
"step": 500
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 22.17391304347826,
|
| 387 |
+
"grad_norm": 1.7754019498825073,
|
| 388 |
+
"learning_rate": 2.9995003067109707e-05,
|
| 389 |
+
"loss": 0.4834,
|
| 390 |
+
"step": 510
|
| 391 |
+
},
|
| 392 |
+
{
|
| 393 |
+
"epoch": 22.608695652173914,
|
| 394 |
+
"grad_norm": 1.5309511423110962,
|
| 395 |
+
"learning_rate": 2.9994756356066246e-05,
|
| 396 |
+
"loss": 0.5615,
|
| 397 |
+
"step": 520
|
| 398 |
+
},
|
| 399 |
+
{
|
| 400 |
+
"epoch": 23.043478260869566,
|
| 401 |
+
"grad_norm": 1.7414244413375854,
|
| 402 |
+
"learning_rate": 2.999450370158046e-05,
|
| 403 |
+
"loss": 0.4927,
|
| 404 |
+
"step": 530
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 23.47826086956522,
|
| 408 |
+
"grad_norm": 1.3099156618118286,
|
| 409 |
+
"learning_rate": 2.9994245103752478e-05,
|
| 410 |
+
"loss": 0.4382,
|
| 411 |
+
"step": 540
|
| 412 |
+
},
|
| 413 |
+
{
|
| 414 |
+
"epoch": 23.91304347826087,
|
| 415 |
+
"grad_norm": 1.2318459749221802,
|
| 416 |
+
"learning_rate": 2.999398056268481e-05,
|
| 417 |
+
"loss": 0.5265,
|
| 418 |
+
"step": 550
|
| 419 |
+
},
|
| 420 |
+
{
|
| 421 |
+
"epoch": 24.347826086956523,
|
| 422 |
+
"grad_norm": 1.406052589416504,
|
| 423 |
+
"learning_rate": 2.9993710078482306e-05,
|
| 424 |
+
"loss": 0.5204,
|
| 425 |
+
"step": 560
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 24.782608695652176,
|
| 429 |
+
"grad_norm": 0.9533604383468628,
|
| 430 |
+
"learning_rate": 2.9993433651252185e-05,
|
| 431 |
+
"loss": 0.4433,
|
| 432 |
+
"step": 570
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"epoch": 25.217391304347824,
|
| 436 |
+
"grad_norm": 1.7167927026748657,
|
| 437 |
+
"learning_rate": 2.9993151281104006e-05,
|
| 438 |
+
"loss": 0.5325,
|
| 439 |
+
"step": 580
|
| 440 |
+
},
|
| 441 |
+
{
|
| 442 |
+
"epoch": 25.652173913043477,
|
| 443 |
+
"grad_norm": 1.1355968713760376,
|
| 444 |
+
"learning_rate": 2.9992862968149695e-05,
|
| 445 |
+
"loss": 0.4736,
|
| 446 |
+
"step": 590
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 26.08695652173913,
|
| 450 |
+
"grad_norm": 1.1737916469573975,
|
| 451 |
+
"learning_rate": 2.9992568712503533e-05,
|
| 452 |
+
"loss": 0.4608,
|
| 453 |
+
"step": 600
|
| 454 |
+
},
|
| 455 |
+
{
|
| 456 |
+
"epoch": 26.08695652173913,
|
| 457 |
+
"eval_loss": 0.7202739715576172,
|
| 458 |
+
"eval_runtime": 0.5354,
|
| 459 |
+
"eval_samples_per_second": 18.678,
|
| 460 |
+
"eval_steps_per_second": 18.678,
|
| 461 |
+
"step": 600
|
| 462 |
+
},
|
| 463 |
+
{
|
| 464 |
+
"epoch": 26.52173913043478,
|
| 465 |
+
"grad_norm": 1.070996880531311,
|
| 466 |
+
"learning_rate": 2.9992268514282142e-05,
|
| 467 |
+
"loss": 0.5118,
|
| 468 |
+
"step": 610
|
| 469 |
+
},
|
| 470 |
+
{
|
| 471 |
+
"epoch": 26.956521739130434,
|
| 472 |
+
"grad_norm": 1.3263134956359863,
|
| 473 |
+
"learning_rate": 2.999196237360452e-05,
|
| 474 |
+
"loss": 0.4314,
|
| 475 |
+
"step": 620
|
| 476 |
+
},
|
| 477 |
+
{
|
| 478 |
+
"epoch": 27.391304347826086,
|
| 479 |
+
"grad_norm": 1.1889632940292358,
|
| 480 |
+
"learning_rate": 2.9991650290592016e-05,
|
| 481 |
+
"loss": 0.4761,
|
| 482 |
+
"step": 630
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 27.82608695652174,
|
| 486 |
+
"grad_norm": 1.3556241989135742,
|
| 487 |
+
"learning_rate": 2.999133226536832e-05,
|
| 488 |
+
"loss": 0.5014,
|
| 489 |
+
"step": 640
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 28.26086956521739,
|
| 493 |
+
"grad_norm": 1.3724006414413452,
|
| 494 |
+
"learning_rate": 2.9991008298059493e-05,
|
| 495 |
+
"loss": 0.4115,
|
| 496 |
+
"step": 650
|
| 497 |
+
},
|
| 498 |
+
{
|
| 499 |
+
"epoch": 28.695652173913043,
|
| 500 |
+
"grad_norm": 1.5488152503967285,
|
| 501 |
+
"learning_rate": 2.9990678388793944e-05,
|
| 502 |
+
"loss": 0.5074,
|
| 503 |
+
"step": 660
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"epoch": 29.130434782608695,
|
| 507 |
+
"grad_norm": 1.2838879823684692,
|
| 508 |
+
"learning_rate": 2.999034253770244e-05,
|
| 509 |
+
"loss": 0.4356,
|
| 510 |
+
"step": 670
|
| 511 |
+
},
|
| 512 |
+
{
|
| 513 |
+
"epoch": 29.565217391304348,
|
| 514 |
+
"grad_norm": 1.0596660375595093,
|
| 515 |
+
"learning_rate": 2.9990000744918097e-05,
|
| 516 |
+
"loss": 0.4714,
|
| 517 |
+
"step": 680
|
| 518 |
+
},
|
| 519 |
+
{
|
| 520 |
+
"epoch": 30.0,
|
| 521 |
+
"grad_norm": 2.439441680908203,
|
| 522 |
+
"learning_rate": 2.9989653010576392e-05,
|
| 523 |
+
"loss": 0.4153,
|
| 524 |
+
"step": 690
|
| 525 |
+
},
|
| 526 |
+
{
|
| 527 |
+
"epoch": 30.434782608695652,
|
| 528 |
+
"grad_norm": 1.2911003828048706,
|
| 529 |
+
"learning_rate": 2.9989299334815158e-05,
|
| 530 |
+
"loss": 0.4779,
|
| 531 |
+
"step": 700
|
| 532 |
+
},
|
| 533 |
+
{
|
| 534 |
+
"epoch": 30.869565217391305,
|
| 535 |
+
"grad_norm": 1.5854625701904297,
|
| 536 |
+
"learning_rate": 2.9988939717774578e-05,
|
| 537 |
+
"loss": 0.4132,
|
| 538 |
+
"step": 710
|
| 539 |
+
},
|
| 540 |
+
{
|
| 541 |
+
"epoch": 31.304347826086957,
|
| 542 |
+
"grad_norm": 0.8958747386932373,
|
| 543 |
+
"learning_rate": 2.9988574159597194e-05,
|
| 544 |
+
"loss": 0.4262,
|
| 545 |
+
"step": 720
|
| 546 |
+
},
|
| 547 |
+
{
|
| 548 |
+
"epoch": 31.73913043478261,
|
| 549 |
+
"grad_norm": 1.5975956916809082,
|
| 550 |
+
"learning_rate": 2.9988202660427907e-05,
|
| 551 |
+
"loss": 0.4839,
|
| 552 |
+
"step": 730
|
| 553 |
+
},
|
| 554 |
+
{
|
| 555 |
+
"epoch": 32.17391304347826,
|
| 556 |
+
"grad_norm": 1.1761164665222168,
|
| 557 |
+
"learning_rate": 2.9987825220413958e-05,
|
| 558 |
+
"loss": 0.4404,
|
| 559 |
+
"step": 740
|
| 560 |
+
},
|
| 561 |
+
{
|
| 562 |
+
"epoch": 32.608695652173914,
|
| 563 |
+
"grad_norm": 1.728130578994751,
|
| 564 |
+
"learning_rate": 2.998744183970496e-05,
|
| 565 |
+
"loss": 0.4754,
|
| 566 |
+
"step": 750
|
| 567 |
+
},
|
| 568 |
+
{
|
| 569 |
+
"epoch": 32.608695652173914,
|
| 570 |
+
"eval_loss": 0.7323614358901978,
|
| 571 |
+
"eval_runtime": 0.3964,
|
| 572 |
+
"eval_samples_per_second": 25.226,
|
| 573 |
+
"eval_steps_per_second": 25.226,
|
| 574 |
+
"step": 750
|
| 575 |
+
},
|
| 576 |
+
{
|
| 577 |
+
"epoch": 33.04347826086956,
|
| 578 |
+
"grad_norm": 1.4063504934310913,
|
| 579 |
+
"learning_rate": 2.998705251845287e-05,
|
| 580 |
+
"loss": 0.4318,
|
| 581 |
+
"step": 760
|
| 582 |
+
},
|
| 583 |
+
{
|
| 584 |
+
"epoch": 33.47826086956522,
|
| 585 |
+
"grad_norm": 1.542359709739685,
|
| 586 |
+
"learning_rate": 2.9986657256812e-05,
|
| 587 |
+
"loss": 0.4331,
|
| 588 |
+
"step": 770
|
| 589 |
+
},
|
| 590 |
+
{
|
| 591 |
+
"epoch": 33.91304347826087,
|
| 592 |
+
"grad_norm": 1.222367763519287,
|
| 593 |
+
"learning_rate": 2.9986256054939022e-05,
|
| 594 |
+
"loss": 0.411,
|
| 595 |
+
"step": 780
|
| 596 |
+
},
|
| 597 |
+
{
|
| 598 |
+
"epoch": 34.34782608695652,
|
| 599 |
+
"grad_norm": 1.40287446975708,
|
| 600 |
+
"learning_rate": 2.9985848912992956e-05,
|
| 601 |
+
"loss": 0.4068,
|
| 602 |
+
"step": 790
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 34.78260869565217,
|
| 606 |
+
"grad_norm": 1.3716740608215332,
|
| 607 |
+
"learning_rate": 2.9985435831135184e-05,
|
| 608 |
+
"loss": 0.3873,
|
| 609 |
+
"step": 800
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 35.21739130434783,
|
| 613 |
+
"grad_norm": 1.2356096506118774,
|
| 614 |
+
"learning_rate": 2.9985016809529437e-05,
|
| 615 |
+
"loss": 0.4798,
|
| 616 |
+
"step": 810
|
| 617 |
+
},
|
| 618 |
+
{
|
| 619 |
+
"epoch": 35.65217391304348,
|
| 620 |
+
"grad_norm": 1.2525038719177246,
|
| 621 |
+
"learning_rate": 2.9984591848341806e-05,
|
| 622 |
+
"loss": 0.4089,
|
| 623 |
+
"step": 820
|
| 624 |
+
},
|
| 625 |
+
{
|
| 626 |
+
"epoch": 36.08695652173913,
|
| 627 |
+
"grad_norm": 1.0272836685180664,
|
| 628 |
+
"learning_rate": 2.9984160947740723e-05,
|
| 629 |
+
"loss": 0.4241,
|
| 630 |
+
"step": 830
|
| 631 |
+
},
|
| 632 |
+
{
|
| 633 |
+
"epoch": 36.52173913043478,
|
| 634 |
+
"grad_norm": 1.0659395456314087,
|
| 635 |
+
"learning_rate": 2.9983724107896993e-05,
|
| 636 |
+
"loss": 0.3879,
|
| 637 |
+
"step": 840
|
| 638 |
+
},
|
| 639 |
+
{
|
| 640 |
+
"epoch": 36.95652173913044,
|
| 641 |
+
"grad_norm": 1.3044679164886475,
|
| 642 |
+
"learning_rate": 2.9983281328983757e-05,
|
| 643 |
+
"loss": 0.4575,
|
| 644 |
+
"step": 850
|
| 645 |
+
},
|
| 646 |
+
{
|
| 647 |
+
"epoch": 37.391304347826086,
|
| 648 |
+
"grad_norm": 1.8042947053909302,
|
| 649 |
+
"learning_rate": 2.9982832611176523e-05,
|
| 650 |
+
"loss": 0.4274,
|
| 651 |
+
"step": 860
|
| 652 |
+
},
|
| 653 |
+
{
|
| 654 |
+
"epoch": 37.82608695652174,
|
| 655 |
+
"grad_norm": 1.1544910669326782,
|
| 656 |
+
"learning_rate": 2.998237795465315e-05,
|
| 657 |
+
"loss": 0.3799,
|
| 658 |
+
"step": 870
|
| 659 |
+
},
|
| 660 |
+
{
|
| 661 |
+
"epoch": 38.26086956521739,
|
| 662 |
+
"grad_norm": 1.1834816932678223,
|
| 663 |
+
"learning_rate": 2.9981917359593843e-05,
|
| 664 |
+
"loss": 0.4118,
|
| 665 |
+
"step": 880
|
| 666 |
+
},
|
| 667 |
+
{
|
| 668 |
+
"epoch": 38.69565217391305,
|
| 669 |
+
"grad_norm": 1.3457635641098022,
|
| 670 |
+
"learning_rate": 2.9981450826181172e-05,
|
| 671 |
+
"loss": 0.3652,
|
| 672 |
+
"step": 890
|
| 673 |
+
},
|
| 674 |
+
{
|
| 675 |
+
"epoch": 39.130434782608695,
|
| 676 |
+
"grad_norm": 1.8477773666381836,
|
| 677 |
+
"learning_rate": 2.9980978354600057e-05,
|
| 678 |
+
"loss": 0.4751,
|
| 679 |
+
"step": 900
|
| 680 |
+
},
|
| 681 |
+
{
|
| 682 |
+
"epoch": 39.130434782608695,
|
| 683 |
+
"eval_loss": 0.7445621490478516,
|
| 684 |
+
"eval_runtime": 0.5125,
|
| 685 |
+
"eval_samples_per_second": 19.513,
|
| 686 |
+
"eval_steps_per_second": 19.513,
|
| 687 |
+
"step": 900
|
| 688 |
+
},
|
| 689 |
+
{
|
| 690 |
+
"epoch": 39.56521739130435,
|
| 691 |
+
"grad_norm": 1.402637004852295,
|
| 692 |
+
"learning_rate": 2.9980499945037765e-05,
|
| 693 |
+
"loss": 0.396,
|
| 694 |
+
"step": 910
|
| 695 |
+
},
|
| 696 |
+
{
|
| 697 |
+
"epoch": 40.0,
|
| 698 |
+
"grad_norm": 2.739563226699829,
|
| 699 |
+
"learning_rate": 2.998001559768393e-05,
|
| 700 |
+
"loss": 0.3981,
|
| 701 |
+
"step": 920
|
| 702 |
+
},
|
| 703 |
+
{
|
| 704 |
+
"epoch": 40.43478260869565,
|
| 705 |
+
"grad_norm": 1.420412302017212,
|
| 706 |
+
"learning_rate": 2.9979525312730525e-05,
|
| 707 |
+
"loss": 0.4639,
|
| 708 |
+
"step": 930
|
| 709 |
+
},
|
| 710 |
+
{
|
| 711 |
+
"epoch": 40.869565217391305,
|
| 712 |
+
"grad_norm": 1.4820829629898071,
|
| 713 |
+
"learning_rate": 2.9979029090371885e-05,
|
| 714 |
+
"loss": 0.3535,
|
| 715 |
+
"step": 940
|
| 716 |
+
},
|
| 717 |
+
{
|
| 718 |
+
"epoch": 41.30434782608695,
|
| 719 |
+
"grad_norm": 1.9887058734893799,
|
| 720 |
+
"learning_rate": 2.99785269308047e-05,
|
| 721 |
+
"loss": 0.357,
|
| 722 |
+
"step": 950
|
| 723 |
+
},
|
| 724 |
+
{
|
| 725 |
+
"epoch": 41.73913043478261,
|
| 726 |
+
"grad_norm": 1.3715981245040894,
|
| 727 |
+
"learning_rate": 2.9978018834228007e-05,
|
| 728 |
+
"loss": 0.3794,
|
| 729 |
+
"step": 960
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 42.17391304347826,
|
| 733 |
+
"grad_norm": 1.3865244388580322,
|
| 734 |
+
"learning_rate": 2.9977504800843197e-05,
|
| 735 |
+
"loss": 0.4517,
|
| 736 |
+
"step": 970
|
| 737 |
+
},
|
| 738 |
+
{
|
| 739 |
+
"epoch": 42.608695652173914,
|
| 740 |
+
"grad_norm": 1.3773282766342163,
|
| 741 |
+
"learning_rate": 2.9976984830854022e-05,
|
| 742 |
+
"loss": 0.393,
|
| 743 |
+
"step": 980
|
| 744 |
+
},
|
| 745 |
+
{
|
| 746 |
+
"epoch": 43.04347826086956,
|
| 747 |
+
"grad_norm": 1.4781227111816406,
|
| 748 |
+
"learning_rate": 2.997645892446658e-05,
|
| 749 |
+
"loss": 0.3833,
|
| 750 |
+
"step": 990
|
| 751 |
+
},
|
| 752 |
+
{
|
| 753 |
+
"epoch": 43.47826086956522,
|
| 754 |
+
"grad_norm": 1.2712657451629639,
|
| 755 |
+
"learning_rate": 2.9975927081889322e-05,
|
| 756 |
+
"loss": 0.4105,
|
| 757 |
+
"step": 1000
|
| 758 |
+
},
|
| 759 |
+
{
|
| 760 |
+
"epoch": 43.91304347826087,
|
| 761 |
+
"grad_norm": 1.098405122756958,
|
| 762 |
+
"learning_rate": 2.9975389303333047e-05,
|
| 763 |
+
"loss": 0.3641,
|
| 764 |
+
"step": 1010
|
| 765 |
+
},
|
| 766 |
+
{
|
| 767 |
+
"epoch": 44.34782608695652,
|
| 768 |
+
"grad_norm": 1.8786333799362183,
|
| 769 |
+
"learning_rate": 2.997484558901093e-05,
|
| 770 |
+
"loss": 0.4123,
|
| 771 |
+
"step": 1020
|
| 772 |
+
},
|
| 773 |
+
{
|
| 774 |
+
"epoch": 44.78260869565217,
|
| 775 |
+
"grad_norm": 1.525408148765564,
|
| 776 |
+
"learning_rate": 2.9974295939138465e-05,
|
| 777 |
+
"loss": 0.4029,
|
| 778 |
+
"step": 1030
|
| 779 |
+
},
|
| 780 |
+
{
|
| 781 |
+
"epoch": 45.21739130434783,
|
| 782 |
+
"grad_norm": 1.158229112625122,
|
| 783 |
+
"learning_rate": 2.9973740353933523e-05,
|
| 784 |
+
"loss": 0.2844,
|
| 785 |
+
"step": 1040
|
| 786 |
+
},
|
| 787 |
+
{
|
| 788 |
+
"epoch": 45.65217391304348,
|
| 789 |
+
"grad_norm": 1.5672129392623901,
|
| 790 |
+
"learning_rate": 2.997317883361632e-05,
|
| 791 |
+
"loss": 0.384,
|
| 792 |
+
"step": 1050
|
| 793 |
+
},
|
| 794 |
+
{
|
| 795 |
+
"epoch": 45.65217391304348,
|
| 796 |
+
"eval_loss": 0.7594717741012573,
|
| 797 |
+
"eval_runtime": 0.4425,
|
| 798 |
+
"eval_samples_per_second": 22.599,
|
| 799 |
+
"eval_steps_per_second": 22.599,
|
| 800 |
+
"step": 1050
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 45.65217391304348,
|
| 804 |
+
"eval_loss": 0.7558861970901489,
|
| 805 |
+
"eval_runtime": 0.3979,
|
| 806 |
+
"eval_samples_per_second": 25.133,
|
| 807 |
+
"eval_steps_per_second": 25.133,
|
| 808 |
+
"step": 1050
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 45.65217391304348,
|
| 812 |
+
"eval_loss": 0.7558861970901489,
|
| 813 |
+
"eval_runtime": 0.3911,
|
| 814 |
+
"eval_samples_per_second": 25.569,
|
| 815 |
+
"eval_steps_per_second": 25.569,
|
| 816 |
+
"step": 1050
|
| 817 |
+
},
|
| 818 |
+
{
|
| 819 |
+
"epoch": 45.65217391304348,
|
| 820 |
+
"eval_loss": 0.7558087110519409,
|
| 821 |
+
"eval_runtime": 0.3948,
|
| 822 |
+
"eval_samples_per_second": 25.331,
|
| 823 |
+
"eval_steps_per_second": 25.331,
|
| 824 |
+
"step": 1050
|
| 825 |
}
|
| 826 |
],
|
| 827 |
"logging_steps": 10,
|
|
|
|
| 841 |
"attributes": {}
|
| 842 |
}
|
| 843 |
},
|
| 844 |
+
"total_flos": 2.696323352857805e+16,
|
| 845 |
"train_batch_size": 4,
|
| 846 |
"trial_name": null,
|
| 847 |
"trial_params": null
|