Training in progress, step 100, 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 389074464
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0866b0718ac5f568089e57be66e100c3130e11978f3b12a029a7b68c5a5ccede
|
| 3 |
size 389074464
|
last-checkpoint/optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 778374186
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f3b82b43ef0af7db935769ed3301b8cf121529a7cd3663da2bcbc8082d0623d1
|
| 3 |
size 778374186
|
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 sha256:771c75945d7b61f9bd5b7f6677420c61a6e339b42a93da85745b20c97ed87696
|
| 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 sha256:5c9fd70aa4cf68daad2242bc04a8a03f22adc681e42e4ebf5294902cea9d0a87
|
| 3 |
size 1064
|
last-checkpoint/trainer_state.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric": 1.
|
| 3 |
-
"best_model_checkpoint": "miner_id_24/checkpoint-
|
| 4 |
-
"epoch": 0.
|
| 5 |
"eval_steps": 50,
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
@@ -373,6 +373,364 @@
|
|
| 373 |
"eval_samples_per_second": 14.23,
|
| 374 |
"eval_steps_per_second": 7.115,
|
| 375 |
"step": 50
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 376 |
}
|
| 377 |
],
|
| 378 |
"logging_steps": 1,
|
|
@@ -401,7 +759,7 @@
|
|
| 401 |
"attributes": {}
|
| 402 |
}
|
| 403 |
},
|
| 404 |
-
"total_flos":
|
| 405 |
"train_batch_size": 8,
|
| 406 |
"trial_name": null,
|
| 407 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 1.1527422666549683,
|
| 3 |
+
"best_model_checkpoint": "miner_id_24/checkpoint-100",
|
| 4 |
+
"epoch": 0.005614901949774702,
|
| 5 |
"eval_steps": 50,
|
| 6 |
+
"global_step": 100,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
|
|
| 373 |
"eval_samples_per_second": 14.23,
|
| 374 |
"eval_steps_per_second": 7.115,
|
| 375 |
"step": 50
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 0.0028635999943850982,
|
| 379 |
+
"grad_norm": 1.7174652814865112,
|
| 380 |
+
"learning_rate": 9.628188298907782e-05,
|
| 381 |
+
"loss": 1.4528,
|
| 382 |
+
"step": 51
|
| 383 |
+
},
|
| 384 |
+
{
|
| 385 |
+
"epoch": 0.002919749013882845,
|
| 386 |
+
"grad_norm": 1.5406450033187866,
|
| 387 |
+
"learning_rate": 9.592434928729616e-05,
|
| 388 |
+
"loss": 1.4099,
|
| 389 |
+
"step": 52
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 0.002975898033380592,
|
| 393 |
+
"grad_norm": 1.2001056671142578,
|
| 394 |
+
"learning_rate": 9.555113246230442e-05,
|
| 395 |
+
"loss": 1.3917,
|
| 396 |
+
"step": 53
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"epoch": 0.0030320470528783392,
|
| 400 |
+
"grad_norm": 0.9413912296295166,
|
| 401 |
+
"learning_rate": 9.516235996730645e-05,
|
| 402 |
+
"loss": 1.2862,
|
| 403 |
+
"step": 54
|
| 404 |
+
},
|
| 405 |
+
{
|
| 406 |
+
"epoch": 0.003088196072376086,
|
| 407 |
+
"grad_norm": 0.6574397087097168,
|
| 408 |
+
"learning_rate": 9.475816456775313e-05,
|
| 409 |
+
"loss": 1.2614,
|
| 410 |
+
"step": 55
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 0.003144345091873833,
|
| 414 |
+
"grad_norm": 0.6218005418777466,
|
| 415 |
+
"learning_rate": 9.43386842960031e-05,
|
| 416 |
+
"loss": 1.2038,
|
| 417 |
+
"step": 56
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 0.0032004941113715803,
|
| 421 |
+
"grad_norm": 0.7471184730529785,
|
| 422 |
+
"learning_rate": 9.39040624041849e-05,
|
| 423 |
+
"loss": 1.1666,
|
| 424 |
+
"step": 57
|
| 425 |
+
},
|
| 426 |
+
{
|
| 427 |
+
"epoch": 0.003256643130869327,
|
| 428 |
+
"grad_norm": 0.7934604287147522,
|
| 429 |
+
"learning_rate": 9.345444731527642e-05,
|
| 430 |
+
"loss": 1.2472,
|
| 431 |
+
"step": 58
|
| 432 |
+
},
|
| 433 |
+
{
|
| 434 |
+
"epoch": 0.0033127921503670744,
|
| 435 |
+
"grad_norm": 0.8308062553405762,
|
| 436 |
+
"learning_rate": 9.298999257241863e-05,
|
| 437 |
+
"loss": 1.2442,
|
| 438 |
+
"step": 59
|
| 439 |
+
},
|
| 440 |
+
{
|
| 441 |
+
"epoch": 0.0033689411698648213,
|
| 442 |
+
"grad_norm": 0.6573333740234375,
|
| 443 |
+
"learning_rate": 9.251085678648072e-05,
|
| 444 |
+
"loss": 1.1986,
|
| 445 |
+
"step": 60
|
| 446 |
+
},
|
| 447 |
+
{
|
| 448 |
+
"epoch": 0.003425090189362568,
|
| 449 |
+
"grad_norm": 0.6570683717727661,
|
| 450 |
+
"learning_rate": 9.201720358189464e-05,
|
| 451 |
+
"loss": 1.2302,
|
| 452 |
+
"step": 61
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 0.0034812392088603154,
|
| 456 |
+
"grad_norm": 0.6069157719612122,
|
| 457 |
+
"learning_rate": 9.150920154077754e-05,
|
| 458 |
+
"loss": 1.1475,
|
| 459 |
+
"step": 62
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 0.0035373882283580623,
|
| 463 |
+
"grad_norm": 0.600216269493103,
|
| 464 |
+
"learning_rate": 9.098702414536107e-05,
|
| 465 |
+
"loss": 1.102,
|
| 466 |
+
"step": 63
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"epoch": 0.003593537247855809,
|
| 470 |
+
"grad_norm": 0.6119481921195984,
|
| 471 |
+
"learning_rate": 9.045084971874738e-05,
|
| 472 |
+
"loss": 1.1455,
|
| 473 |
+
"step": 64
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 0.0036496862673535565,
|
| 477 |
+
"grad_norm": 0.6701284646987915,
|
| 478 |
+
"learning_rate": 8.9900861364012e-05,
|
| 479 |
+
"loss": 1.192,
|
| 480 |
+
"step": 65
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 0.0037058352868513033,
|
| 484 |
+
"grad_norm": 0.6195027828216553,
|
| 485 |
+
"learning_rate": 8.933724690167417e-05,
|
| 486 |
+
"loss": 1.1548,
|
| 487 |
+
"step": 66
|
| 488 |
+
},
|
| 489 |
+
{
|
| 490 |
+
"epoch": 0.00376198430634905,
|
| 491 |
+
"grad_norm": 0.6576624512672424,
|
| 492 |
+
"learning_rate": 8.876019880555649e-05,
|
| 493 |
+
"loss": 1.23,
|
| 494 |
+
"step": 67
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 0.0038181333258467975,
|
| 498 |
+
"grad_norm": 0.6199018955230713,
|
| 499 |
+
"learning_rate": 8.816991413705516e-05,
|
| 500 |
+
"loss": 1.085,
|
| 501 |
+
"step": 68
|
| 502 |
+
},
|
| 503 |
+
{
|
| 504 |
+
"epoch": 0.0038742823453445444,
|
| 505 |
+
"grad_norm": 0.6421352624893188,
|
| 506 |
+
"learning_rate": 8.756659447784368e-05,
|
| 507 |
+
"loss": 1.1397,
|
| 508 |
+
"step": 69
|
| 509 |
+
},
|
| 510 |
+
{
|
| 511 |
+
"epoch": 0.003930431364842292,
|
| 512 |
+
"grad_norm": 0.7040902376174927,
|
| 513 |
+
"learning_rate": 8.695044586103296e-05,
|
| 514 |
+
"loss": 1.1982,
|
| 515 |
+
"step": 70
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 0.003986580384340038,
|
| 519 |
+
"grad_norm": 0.6545359492301941,
|
| 520 |
+
"learning_rate": 8.632167870081121e-05,
|
| 521 |
+
"loss": 1.1224,
|
| 522 |
+
"step": 71
|
| 523 |
+
},
|
| 524 |
+
{
|
| 525 |
+
"epoch": 0.004042729403837785,
|
| 526 |
+
"grad_norm": 0.7151196002960205,
|
| 527 |
+
"learning_rate": 8.568050772058762e-05,
|
| 528 |
+
"loss": 1.1129,
|
| 529 |
+
"step": 72
|
| 530 |
+
},
|
| 531 |
+
{
|
| 532 |
+
"epoch": 0.004098878423335533,
|
| 533 |
+
"grad_norm": 0.6587552428245544,
|
| 534 |
+
"learning_rate": 8.502715187966455e-05,
|
| 535 |
+
"loss": 1.1292,
|
| 536 |
+
"step": 73
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 0.004155027442833279,
|
| 540 |
+
"grad_norm": 0.6608380079269409,
|
| 541 |
+
"learning_rate": 8.436183429846313e-05,
|
| 542 |
+
"loss": 1.124,
|
| 543 |
+
"step": 74
|
| 544 |
+
},
|
| 545 |
+
{
|
| 546 |
+
"epoch": 0.004211176462331026,
|
| 547 |
+
"grad_norm": 0.6702746152877808,
|
| 548 |
+
"learning_rate": 8.368478218232787e-05,
|
| 549 |
+
"loss": 1.1656,
|
| 550 |
+
"step": 75
|
| 551 |
+
},
|
| 552 |
+
{
|
| 553 |
+
"epoch": 0.004267325481828774,
|
| 554 |
+
"grad_norm": 0.6599065661430359,
|
| 555 |
+
"learning_rate": 8.299622674393614e-05,
|
| 556 |
+
"loss": 1.1543,
|
| 557 |
+
"step": 76
|
| 558 |
+
},
|
| 559 |
+
{
|
| 560 |
+
"epoch": 0.004323474501326521,
|
| 561 |
+
"grad_norm": 0.6634683012962341,
|
| 562 |
+
"learning_rate": 8.229640312433937e-05,
|
| 563 |
+
"loss": 1.193,
|
| 564 |
+
"step": 77
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 0.004379623520824267,
|
| 568 |
+
"grad_norm": 0.683447539806366,
|
| 569 |
+
"learning_rate": 8.158555031266254e-05,
|
| 570 |
+
"loss": 1.1215,
|
| 571 |
+
"step": 78
|
| 572 |
+
},
|
| 573 |
+
{
|
| 574 |
+
"epoch": 0.004435772540322015,
|
| 575 |
+
"grad_norm": 0.6692266464233398,
|
| 576 |
+
"learning_rate": 8.086391106448965e-05,
|
| 577 |
+
"loss": 1.0958,
|
| 578 |
+
"step": 79
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 0.004491921559819762,
|
| 582 |
+
"grad_norm": 0.6918296813964844,
|
| 583 |
+
"learning_rate": 8.013173181896283e-05,
|
| 584 |
+
"loss": 1.2255,
|
| 585 |
+
"step": 80
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 0.0045480705793175084,
|
| 589 |
+
"grad_norm": 0.6787101626396179,
|
| 590 |
+
"learning_rate": 7.938926261462366e-05,
|
| 591 |
+
"loss": 1.1807,
|
| 592 |
+
"step": 81
|
| 593 |
+
},
|
| 594 |
+
{
|
| 595 |
+
"epoch": 0.004604219598815256,
|
| 596 |
+
"grad_norm": 0.7259939908981323,
|
| 597 |
+
"learning_rate": 7.863675700402526e-05,
|
| 598 |
+
"loss": 1.1612,
|
| 599 |
+
"step": 82
|
| 600 |
+
},
|
| 601 |
+
{
|
| 602 |
+
"epoch": 0.004660368618313003,
|
| 603 |
+
"grad_norm": 0.6921144723892212,
|
| 604 |
+
"learning_rate": 7.787447196714427e-05,
|
| 605 |
+
"loss": 1.1331,
|
| 606 |
+
"step": 83
|
| 607 |
+
},
|
| 608 |
+
{
|
| 609 |
+
"epoch": 0.0047165176378107495,
|
| 610 |
+
"grad_norm": 0.7044004797935486,
|
| 611 |
+
"learning_rate": 7.710266782362247e-05,
|
| 612 |
+
"loss": 1.1469,
|
| 613 |
+
"step": 84
|
| 614 |
+
},
|
| 615 |
+
{
|
| 616 |
+
"epoch": 0.004772666657308497,
|
| 617 |
+
"grad_norm": 0.7107269763946533,
|
| 618 |
+
"learning_rate": 7.63216081438678e-05,
|
| 619 |
+
"loss": 1.2156,
|
| 620 |
+
"step": 85
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"epoch": 0.004828815676806244,
|
| 624 |
+
"grad_norm": 0.7129175662994385,
|
| 625 |
+
"learning_rate": 7.553155965904535e-05,
|
| 626 |
+
"loss": 1.118,
|
| 627 |
+
"step": 86
|
| 628 |
+
},
|
| 629 |
+
{
|
| 630 |
+
"epoch": 0.0048849646963039905,
|
| 631 |
+
"grad_norm": 0.6944028735160828,
|
| 632 |
+
"learning_rate": 7.473279216998895e-05,
|
| 633 |
+
"loss": 1.1684,
|
| 634 |
+
"step": 87
|
| 635 |
+
},
|
| 636 |
+
{
|
| 637 |
+
"epoch": 0.004941113715801738,
|
| 638 |
+
"grad_norm": 0.6979909539222717,
|
| 639 |
+
"learning_rate": 7.392557845506432e-05,
|
| 640 |
+
"loss": 1.1503,
|
| 641 |
+
"step": 88
|
| 642 |
+
},
|
| 643 |
+
{
|
| 644 |
+
"epoch": 0.004997262735299485,
|
| 645 |
+
"grad_norm": 0.6969708800315857,
|
| 646 |
+
"learning_rate": 7.311019417701566e-05,
|
| 647 |
+
"loss": 1.0991,
|
| 648 |
+
"step": 89
|
| 649 |
+
},
|
| 650 |
+
{
|
| 651 |
+
"epoch": 0.0050534117547972315,
|
| 652 |
+
"grad_norm": 0.7195575833320618,
|
| 653 |
+
"learning_rate": 7.228691778882693e-05,
|
| 654 |
+
"loss": 1.1541,
|
| 655 |
+
"step": 90
|
| 656 |
+
},
|
| 657 |
+
{
|
| 658 |
+
"epoch": 0.005109560774294979,
|
| 659 |
+
"grad_norm": 0.7229387164115906,
|
| 660 |
+
"learning_rate": 7.145603043863045e-05,
|
| 661 |
+
"loss": 1.0439,
|
| 662 |
+
"step": 91
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 0.005165709793792726,
|
| 666 |
+
"grad_norm": 0.7308394908905029,
|
| 667 |
+
"learning_rate": 7.061781587369519e-05,
|
| 668 |
+
"loss": 1.1037,
|
| 669 |
+
"step": 92
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 0.0052218588132904725,
|
| 673 |
+
"grad_norm": 0.7165358662605286,
|
| 674 |
+
"learning_rate": 6.977256034352712e-05,
|
| 675 |
+
"loss": 1.0528,
|
| 676 |
+
"step": 93
|
| 677 |
+
},
|
| 678 |
+
{
|
| 679 |
+
"epoch": 0.00527800783278822,
|
| 680 |
+
"grad_norm": 0.7242848873138428,
|
| 681 |
+
"learning_rate": 6.892055250211552e-05,
|
| 682 |
+
"loss": 1.0637,
|
| 683 |
+
"step": 94
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"epoch": 0.005334156852285967,
|
| 687 |
+
"grad_norm": 0.7282512784004211,
|
| 688 |
+
"learning_rate": 6.806208330935766e-05,
|
| 689 |
+
"loss": 1.0407,
|
| 690 |
+
"step": 95
|
| 691 |
+
},
|
| 692 |
+
{
|
| 693 |
+
"epoch": 0.0053903058717837135,
|
| 694 |
+
"grad_norm": 0.7488335967063904,
|
| 695 |
+
"learning_rate": 6.719744593169641e-05,
|
| 696 |
+
"loss": 1.0557,
|
| 697 |
+
"step": 96
|
| 698 |
+
},
|
| 699 |
+
{
|
| 700 |
+
"epoch": 0.005446454891281461,
|
| 701 |
+
"grad_norm": 0.8261488676071167,
|
| 702 |
+
"learning_rate": 6.632693564200416e-05,
|
| 703 |
+
"loss": 1.1658,
|
| 704 |
+
"step": 97
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 0.005502603910779208,
|
| 708 |
+
"grad_norm": 0.7976979613304138,
|
| 709 |
+
"learning_rate": 6.545084971874738e-05,
|
| 710 |
+
"loss": 1.0287,
|
| 711 |
+
"step": 98
|
| 712 |
+
},
|
| 713 |
+
{
|
| 714 |
+
"epoch": 0.005558752930276955,
|
| 715 |
+
"grad_norm": 0.8135994076728821,
|
| 716 |
+
"learning_rate": 6.456948734446624e-05,
|
| 717 |
+
"loss": 0.9347,
|
| 718 |
+
"step": 99
|
| 719 |
+
},
|
| 720 |
+
{
|
| 721 |
+
"epoch": 0.005614901949774702,
|
| 722 |
+
"grad_norm": 0.9864936470985413,
|
| 723 |
+
"learning_rate": 6.368314950360415e-05,
|
| 724 |
+
"loss": 1.0265,
|
| 725 |
+
"step": 100
|
| 726 |
+
},
|
| 727 |
+
{
|
| 728 |
+
"epoch": 0.005614901949774702,
|
| 729 |
+
"eval_loss": 1.1527422666549683,
|
| 730 |
+
"eval_runtime": 2109.9145,
|
| 731 |
+
"eval_samples_per_second": 14.216,
|
| 732 |
+
"eval_steps_per_second": 7.108,
|
| 733 |
+
"step": 100
|
| 734 |
}
|
| 735 |
],
|
| 736 |
"logging_steps": 1,
|
|
|
|
| 759 |
"attributes": {}
|
| 760 |
}
|
| 761 |
},
|
| 762 |
+
"total_flos": 5.73309481648128e+16,
|
| 763 |
"train_batch_size": 8,
|
| 764 |
"trial_name": null,
|
| 765 |
"trial_params": null
|