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 871457448
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:63303b71e5d312d2644b2da9490486b4efc29510df35aca1172c5a8c1a23f988
|
| 3 |
size 871457448
|
last-checkpoint/optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 443280340
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:488609b0ab4180cbb26f0114c39b3ea751b0ac1f3ea8004801d22047c0e3c9dd
|
| 3 |
size 443280340
|
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:fd23bfead1e29533b5da69cff0304a9ddd119a6d662e90bbef7e9b33dcf44989
|
| 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 sha256:0ddb9588ea654e56e83effcf81a2bc03480954babcf6415cb44d41d3bfb8039f
|
| 3 |
size 1064
|
last-checkpoint/trainer_state.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric": 0.
|
| 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": 9.837,
|
| 374 |
"eval_steps_per_second": 2.464,
|
| 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": 0.15341179072856903,
|
| 3 |
+
"best_model_checkpoint": "miner_id_24/checkpoint-100",
|
| 4 |
+
"epoch": 0.17006802721088435,
|
| 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": 9.837,
|
| 374 |
"eval_steps_per_second": 2.464,
|
| 375 |
"step": 50
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 0.08673469387755102,
|
| 379 |
+
"grad_norm": 0.32864144444465637,
|
| 380 |
+
"learning_rate": 8.894386393810563e-05,
|
| 381 |
+
"loss": 0.1396,
|
| 382 |
+
"step": 51
|
| 383 |
+
},
|
| 384 |
+
{
|
| 385 |
+
"epoch": 0.08843537414965986,
|
| 386 |
+
"grad_norm": 0.3204747438430786,
|
| 387 |
+
"learning_rate": 8.842005554284296e-05,
|
| 388 |
+
"loss": 0.0863,
|
| 389 |
+
"step": 52
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 0.09013605442176871,
|
| 393 |
+
"grad_norm": 0.2585567831993103,
|
| 394 |
+
"learning_rate": 8.788574348801675e-05,
|
| 395 |
+
"loss": 0.1298,
|
| 396 |
+
"step": 53
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"epoch": 0.09183673469387756,
|
| 400 |
+
"grad_norm": 0.1529138833284378,
|
| 401 |
+
"learning_rate": 8.73410738492077e-05,
|
| 402 |
+
"loss": 0.067,
|
| 403 |
+
"step": 54
|
| 404 |
+
},
|
| 405 |
+
{
|
| 406 |
+
"epoch": 0.0935374149659864,
|
| 407 |
+
"grad_norm": 0.2032083421945572,
|
| 408 |
+
"learning_rate": 8.678619553365659e-05,
|
| 409 |
+
"loss": 0.123,
|
| 410 |
+
"step": 55
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 0.09523809523809523,
|
| 414 |
+
"grad_norm": 0.25678551197052,
|
| 415 |
+
"learning_rate": 8.622126023955446e-05,
|
| 416 |
+
"loss": 0.1174,
|
| 417 |
+
"step": 56
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 0.09693877551020408,
|
| 421 |
+
"grad_norm": 0.2589610815048218,
|
| 422 |
+
"learning_rate": 8.564642241456986e-05,
|
| 423 |
+
"loss": 0.1145,
|
| 424 |
+
"step": 57
|
| 425 |
+
},
|
| 426 |
+
{
|
| 427 |
+
"epoch": 0.09863945578231292,
|
| 428 |
+
"grad_norm": 0.3110625743865967,
|
| 429 |
+
"learning_rate": 8.506183921362443e-05,
|
| 430 |
+
"loss": 0.14,
|
| 431 |
+
"step": 58
|
| 432 |
+
},
|
| 433 |
+
{
|
| 434 |
+
"epoch": 0.10034013605442177,
|
| 435 |
+
"grad_norm": 0.20775605738162994,
|
| 436 |
+
"learning_rate": 8.44676704559283e-05,
|
| 437 |
+
"loss": 0.1232,
|
| 438 |
+
"step": 59
|
| 439 |
+
},
|
| 440 |
+
{
|
| 441 |
+
"epoch": 0.10204081632653061,
|
| 442 |
+
"grad_norm": 0.17721042037010193,
|
| 443 |
+
"learning_rate": 8.386407858128706e-05,
|
| 444 |
+
"loss": 0.1048,
|
| 445 |
+
"step": 60
|
| 446 |
+
},
|
| 447 |
+
{
|
| 448 |
+
"epoch": 0.10374149659863946,
|
| 449 |
+
"grad_norm": 0.20299284160137177,
|
| 450 |
+
"learning_rate": 8.32512286056924e-05,
|
| 451 |
+
"loss": 0.1269,
|
| 452 |
+
"step": 61
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 0.1054421768707483,
|
| 456 |
+
"grad_norm": 0.2592518627643585,
|
| 457 |
+
"learning_rate": 8.262928807620843e-05,
|
| 458 |
+
"loss": 0.1556,
|
| 459 |
+
"step": 62
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 0.10714285714285714,
|
| 463 |
+
"grad_norm": 0.28578001260757446,
|
| 464 |
+
"learning_rate": 8.199842702516583e-05,
|
| 465 |
+
"loss": 0.1339,
|
| 466 |
+
"step": 63
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"epoch": 0.10884353741496598,
|
| 470 |
+
"grad_norm": 0.21084251999855042,
|
| 471 |
+
"learning_rate": 8.135881792367686e-05,
|
| 472 |
+
"loss": 0.1197,
|
| 473 |
+
"step": 64
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 0.11054421768707483,
|
| 477 |
+
"grad_norm": 0.20630519092082977,
|
| 478 |
+
"learning_rate": 8.07106356344834e-05,
|
| 479 |
+
"loss": 0.1317,
|
| 480 |
+
"step": 65
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 0.11224489795918367,
|
| 484 |
+
"grad_norm": 0.22667892277240753,
|
| 485 |
+
"learning_rate": 8.005405736415126e-05,
|
| 486 |
+
"loss": 0.1253,
|
| 487 |
+
"step": 66
|
| 488 |
+
},
|
| 489 |
+
{
|
| 490 |
+
"epoch": 0.11394557823129252,
|
| 491 |
+
"grad_norm": 0.20838917791843414,
|
| 492 |
+
"learning_rate": 7.938926261462366e-05,
|
| 493 |
+
"loss": 0.1009,
|
| 494 |
+
"step": 67
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 0.11564625850340136,
|
| 498 |
+
"grad_norm": 0.2214667648077011,
|
| 499 |
+
"learning_rate": 7.871643313414718e-05,
|
| 500 |
+
"loss": 0.1302,
|
| 501 |
+
"step": 68
|
| 502 |
+
},
|
| 503 |
+
{
|
| 504 |
+
"epoch": 0.11734693877551021,
|
| 505 |
+
"grad_norm": 0.24575111269950867,
|
| 506 |
+
"learning_rate": 7.803575286758364e-05,
|
| 507 |
+
"loss": 0.1205,
|
| 508 |
+
"step": 69
|
| 509 |
+
},
|
| 510 |
+
{
|
| 511 |
+
"epoch": 0.11904761904761904,
|
| 512 |
+
"grad_norm": 0.26279327273368835,
|
| 513 |
+
"learning_rate": 7.734740790612136e-05,
|
| 514 |
+
"loss": 0.131,
|
| 515 |
+
"step": 70
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 0.12074829931972789,
|
| 519 |
+
"grad_norm": 0.25295835733413696,
|
| 520 |
+
"learning_rate": 7.66515864363997e-05,
|
| 521 |
+
"loss": 0.1435,
|
| 522 |
+
"step": 71
|
| 523 |
+
},
|
| 524 |
+
{
|
| 525 |
+
"epoch": 0.12244897959183673,
|
| 526 |
+
"grad_norm": 0.23764437437057495,
|
| 527 |
+
"learning_rate": 7.594847868906076e-05,
|
| 528 |
+
"loss": 0.1187,
|
| 529 |
+
"step": 72
|
| 530 |
+
},
|
| 531 |
+
{
|
| 532 |
+
"epoch": 0.12414965986394558,
|
| 533 |
+
"grad_norm": 0.20362688601016998,
|
| 534 |
+
"learning_rate": 7.52382768867422e-05,
|
| 535 |
+
"loss": 0.1213,
|
| 536 |
+
"step": 73
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 0.12585034013605442,
|
| 540 |
+
"grad_norm": 0.2588404715061188,
|
| 541 |
+
"learning_rate": 7.452117519152542e-05,
|
| 542 |
+
"loss": 0.1361,
|
| 543 |
+
"step": 74
|
| 544 |
+
},
|
| 545 |
+
{
|
| 546 |
+
"epoch": 0.12755102040816327,
|
| 547 |
+
"grad_norm": 0.48782673478126526,
|
| 548 |
+
"learning_rate": 7.379736965185368e-05,
|
| 549 |
+
"loss": 0.1394,
|
| 550 |
+
"step": 75
|
| 551 |
+
},
|
| 552 |
+
{
|
| 553 |
+
"epoch": 0.1292517006802721,
|
| 554 |
+
"grad_norm": 0.3100239038467407,
|
| 555 |
+
"learning_rate": 7.30670581489344e-05,
|
| 556 |
+
"loss": 0.1434,
|
| 557 |
+
"step": 76
|
| 558 |
+
},
|
| 559 |
+
{
|
| 560 |
+
"epoch": 0.13095238095238096,
|
| 561 |
+
"grad_norm": 0.28053736686706543,
|
| 562 |
+
"learning_rate": 7.233044034264034e-05,
|
| 563 |
+
"loss": 0.1234,
|
| 564 |
+
"step": 77
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 0.1326530612244898,
|
| 568 |
+
"grad_norm": 0.26575592160224915,
|
| 569 |
+
"learning_rate": 7.158771761692464e-05,
|
| 570 |
+
"loss": 0.1418,
|
| 571 |
+
"step": 78
|
| 572 |
+
},
|
| 573 |
+
{
|
| 574 |
+
"epoch": 0.13435374149659865,
|
| 575 |
+
"grad_norm": 0.26024481654167175,
|
| 576 |
+
"learning_rate": 7.083909302476453e-05,
|
| 577 |
+
"loss": 0.136,
|
| 578 |
+
"step": 79
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 0.1360544217687075,
|
| 582 |
+
"grad_norm": 0.44965115189552307,
|
| 583 |
+
"learning_rate": 7.008477123264848e-05,
|
| 584 |
+
"loss": 0.2536,
|
| 585 |
+
"step": 80
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 0.1377551020408163,
|
| 589 |
+
"grad_norm": 0.3972369134426117,
|
| 590 |
+
"learning_rate": 6.932495846462261e-05,
|
| 591 |
+
"loss": 0.2138,
|
| 592 |
+
"step": 81
|
| 593 |
+
},
|
| 594 |
+
{
|
| 595 |
+
"epoch": 0.13945578231292516,
|
| 596 |
+
"grad_norm": 0.2681017518043518,
|
| 597 |
+
"learning_rate": 6.855986244591104e-05,
|
| 598 |
+
"loss": 0.1231,
|
| 599 |
+
"step": 82
|
| 600 |
+
},
|
| 601 |
+
{
|
| 602 |
+
"epoch": 0.141156462585034,
|
| 603 |
+
"grad_norm": 0.42896729707717896,
|
| 604 |
+
"learning_rate": 6.778969234612584e-05,
|
| 605 |
+
"loss": 0.2301,
|
| 606 |
+
"step": 83
|
| 607 |
+
},
|
| 608 |
+
{
|
| 609 |
+
"epoch": 0.14285714285714285,
|
| 610 |
+
"grad_norm": 0.28568339347839355,
|
| 611 |
+
"learning_rate": 6.701465872208216e-05,
|
| 612 |
+
"loss": 0.1233,
|
| 613 |
+
"step": 84
|
| 614 |
+
},
|
| 615 |
+
{
|
| 616 |
+
"epoch": 0.1445578231292517,
|
| 617 |
+
"grad_norm": 0.38649600744247437,
|
| 618 |
+
"learning_rate": 6.623497346023418e-05,
|
| 619 |
+
"loss": 0.193,
|
| 620 |
+
"step": 85
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"epoch": 0.14625850340136054,
|
| 624 |
+
"grad_norm": 0.33405882120132446,
|
| 625 |
+
"learning_rate": 6.545084971874738e-05,
|
| 626 |
+
"loss": 0.1168,
|
| 627 |
+
"step": 86
|
| 628 |
+
},
|
| 629 |
+
{
|
| 630 |
+
"epoch": 0.14795918367346939,
|
| 631 |
+
"grad_norm": 0.5408915281295776,
|
| 632 |
+
"learning_rate": 6.466250186922325e-05,
|
| 633 |
+
"loss": 0.1856,
|
| 634 |
+
"step": 87
|
| 635 |
+
},
|
| 636 |
+
{
|
| 637 |
+
"epoch": 0.14965986394557823,
|
| 638 |
+
"grad_norm": 0.41401568055152893,
|
| 639 |
+
"learning_rate": 6.387014543809223e-05,
|
| 640 |
+
"loss": 0.1562,
|
| 641 |
+
"step": 88
|
| 642 |
+
},
|
| 643 |
+
{
|
| 644 |
+
"epoch": 0.15136054421768708,
|
| 645 |
+
"grad_norm": 0.36673107743263245,
|
| 646 |
+
"learning_rate": 6.307399704769099e-05,
|
| 647 |
+
"loss": 0.1655,
|
| 648 |
+
"step": 89
|
| 649 |
+
},
|
| 650 |
+
{
|
| 651 |
+
"epoch": 0.15306122448979592,
|
| 652 |
+
"grad_norm": 0.4040523171424866,
|
| 653 |
+
"learning_rate": 6.227427435703997e-05,
|
| 654 |
+
"loss": 0.1879,
|
| 655 |
+
"step": 90
|
| 656 |
+
},
|
| 657 |
+
{
|
| 658 |
+
"epoch": 0.15476190476190477,
|
| 659 |
+
"grad_norm": 0.4495351314544678,
|
| 660 |
+
"learning_rate": 6.147119600233758e-05,
|
| 661 |
+
"loss": 0.2504,
|
| 662 |
+
"step": 91
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 0.1564625850340136,
|
| 666 |
+
"grad_norm": 0.3567620515823364,
|
| 667 |
+
"learning_rate": 6.066498153718735e-05,
|
| 668 |
+
"loss": 0.1414,
|
| 669 |
+
"step": 92
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 0.15816326530612246,
|
| 673 |
+
"grad_norm": 0.3996597230434418,
|
| 674 |
+
"learning_rate": 5.985585137257401e-05,
|
| 675 |
+
"loss": 0.1539,
|
| 676 |
+
"step": 93
|
| 677 |
+
},
|
| 678 |
+
{
|
| 679 |
+
"epoch": 0.1598639455782313,
|
| 680 |
+
"grad_norm": 0.6240476965904236,
|
| 681 |
+
"learning_rate": 5.90440267166055e-05,
|
| 682 |
+
"loss": 0.3179,
|
| 683 |
+
"step": 94
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"epoch": 0.16156462585034015,
|
| 687 |
+
"grad_norm": 0.5150894522666931,
|
| 688 |
+
"learning_rate": 5.8229729514036705e-05,
|
| 689 |
+
"loss": 0.2054,
|
| 690 |
+
"step": 95
|
| 691 |
+
},
|
| 692 |
+
{
|
| 693 |
+
"epoch": 0.16326530612244897,
|
| 694 |
+
"grad_norm": 0.46117132902145386,
|
| 695 |
+
"learning_rate": 5.74131823855921e-05,
|
| 696 |
+
"loss": 0.1798,
|
| 697 |
+
"step": 96
|
| 698 |
+
},
|
| 699 |
+
{
|
| 700 |
+
"epoch": 0.1649659863945578,
|
| 701 |
+
"grad_norm": 0.6154991984367371,
|
| 702 |
+
"learning_rate": 5.6594608567103456e-05,
|
| 703 |
+
"loss": 0.196,
|
| 704 |
+
"step": 97
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 0.16666666666666666,
|
| 708 |
+
"grad_norm": 0.5747442245483398,
|
| 709 |
+
"learning_rate": 5.577423184847932e-05,
|
| 710 |
+
"loss": 0.245,
|
| 711 |
+
"step": 98
|
| 712 |
+
},
|
| 713 |
+
{
|
| 714 |
+
"epoch": 0.1683673469387755,
|
| 715 |
+
"grad_norm": 0.7409622669219971,
|
| 716 |
+
"learning_rate": 5.495227651252315e-05,
|
| 717 |
+
"loss": 0.2667,
|
| 718 |
+
"step": 99
|
| 719 |
+
},
|
| 720 |
+
{
|
| 721 |
+
"epoch": 0.17006802721088435,
|
| 722 |
+
"grad_norm": 1.218652367591858,
|
| 723 |
+
"learning_rate": 5.4128967273616625e-05,
|
| 724 |
+
"loss": 0.246,
|
| 725 |
+
"step": 100
|
| 726 |
+
},
|
| 727 |
+
{
|
| 728 |
+
"epoch": 0.17006802721088435,
|
| 729 |
+
"eval_loss": 0.15341179072856903,
|
| 730 |
+
"eval_runtime": 100.6674,
|
| 731 |
+
"eval_samples_per_second": 9.834,
|
| 732 |
+
"eval_steps_per_second": 2.464,
|
| 733 |
+
"step": 100
|
| 734 |
}
|
| 735 |
],
|
| 736 |
"logging_steps": 1,
|
|
|
|
| 759 |
"attributes": {}
|
| 760 |
}
|
| 761 |
},
|
| 762 |
+
"total_flos": 1.84380938187178e+17,
|
| 763 |
"train_batch_size": 8,
|
| 764 |
"trial_name": null,
|
| 765 |
"trial_params": null
|