Wilsonwin commited on
Commit
75f3748
·
verified ·
1 Parent(s): 7947b33

Training in progress, step 1000, checkpoint

Browse files
last-checkpoint/model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:c706455b22491373a3d05c14ba83b357c853d8374d88791c7676b0991261ada4
3
  size 328277848
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f6a6ed44cd2dfba870cb534237ba4896f3e7dc134dd8f5e7b12dba7ffa27c335
3
  size 328277848
last-checkpoint/optimizer.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:f4a4b6282ca780e8cd61a5761889ffdafaa2e5be8ebe2483e19dae424d92b8f8
3
  size 318646859
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a3d65452f71865cff75ce7bd9061bd7c195d9f7790eb08651bef46b28c8cf5db
3
  size 318646859
last-checkpoint/rng_state.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:577f0b7cee8114520712167e366c7fd1da45e6cc96d21bb243dccf2193c0d60f
3
  size 14645
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a141ddada80b12146ad2875b480471ca4604a84a507446df6ce95668765adaf4
3
  size 14645
last-checkpoint/scheduler.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:1d6f69db244377919464a655cd8dc38a3ecd36738e8ab7fd789261995d39f57a
3
  size 1465
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:641712abd35039f810da46b5ecace55e8c31f5b5a7d2cfa0aaa8182597f8aad6
3
  size 1465
last-checkpoint/trainer_state.json CHANGED
@@ -2,9 +2,9 @@
2
  "best_global_step": null,
3
  "best_metric": null,
4
  "best_model_checkpoint": null,
5
- "epoch": 0.08447372867038351,
6
  "eval_steps": 500,
7
- "global_step": 500,
8
  "is_hyper_param_search": false,
9
  "is_local_process_zero": true,
10
  "is_world_process_zero": true,
@@ -373,6 +373,364 @@
373
  "eval_samples_per_second": 248.538,
374
  "eval_steps_per_second": 5.219,
375
  "step": 500
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
376
  }
377
  ],
378
  "logging_steps": 10,
@@ -392,7 +750,7 @@
392
  "attributes": {}
393
  }
394
  },
395
- "total_flos": 1.6722841042944e+16,
396
  "train_batch_size": 48,
397
  "trial_name": null,
398
  "trial_params": null
 
2
  "best_global_step": null,
3
  "best_metric": null,
4
  "best_model_checkpoint": null,
5
+ "epoch": 0.16894745734076702,
6
  "eval_steps": 500,
7
+ "global_step": 1000,
8
  "is_hyper_param_search": false,
9
  "is_local_process_zero": true,
10
  "is_world_process_zero": true,
 
373
  "eval_samples_per_second": 248.538,
374
  "eval_steps_per_second": 5.219,
375
  "step": 500
376
+ },
377
+ {
378
+ "epoch": 0.08616320324379118,
379
+ "grad_norm": 0.6618097424507141,
380
+ "learning_rate": 7.635e-05,
381
+ "loss": 7.392906188964844,
382
+ "step": 510
383
+ },
384
+ {
385
+ "epoch": 0.08785267781719885,
386
+ "grad_norm": 0.6140709519386292,
387
+ "learning_rate": 7.785e-05,
388
+ "loss": 7.364067840576172,
389
+ "step": 520
390
+ },
391
+ {
392
+ "epoch": 0.08954215239060652,
393
+ "grad_norm": 0.6116703748703003,
394
+ "learning_rate": 7.934999999999999e-05,
395
+ "loss": 7.337810516357422,
396
+ "step": 530
397
+ },
398
+ {
399
+ "epoch": 0.0912316269640142,
400
+ "grad_norm": 0.8000091314315796,
401
+ "learning_rate": 8.085e-05,
402
+ "loss": 7.299466705322265,
403
+ "step": 540
404
+ },
405
+ {
406
+ "epoch": 0.09292110153742186,
407
+ "grad_norm": 0.5890388488769531,
408
+ "learning_rate": 8.235e-05,
409
+ "loss": 7.308570098876953,
410
+ "step": 550
411
+ },
412
+ {
413
+ "epoch": 0.09461057611082953,
414
+ "grad_norm": 1.0396614074707031,
415
+ "learning_rate": 8.385e-05,
416
+ "loss": 7.27392349243164,
417
+ "step": 560
418
+ },
419
+ {
420
+ "epoch": 0.0963000506842372,
421
+ "grad_norm": 0.5742290019989014,
422
+ "learning_rate": 8.534999999999999e-05,
423
+ "loss": 7.271208953857422,
424
+ "step": 570
425
+ },
426
+ {
427
+ "epoch": 0.09798952525764487,
428
+ "grad_norm": 0.684992790222168,
429
+ "learning_rate": 8.684999999999998e-05,
430
+ "loss": 7.2550514221191404,
431
+ "step": 580
432
+ },
433
+ {
434
+ "epoch": 0.09967899983105254,
435
+ "grad_norm": 1.2290043830871582,
436
+ "learning_rate": 8.834999999999999e-05,
437
+ "loss": 7.2304443359375,
438
+ "step": 590
439
+ },
440
+ {
441
+ "epoch": 0.10136847440446022,
442
+ "grad_norm": 0.7645843029022217,
443
+ "learning_rate": 8.984999999999999e-05,
444
+ "loss": 7.205104064941406,
445
+ "step": 600
446
+ },
447
+ {
448
+ "epoch": 0.10305794897786788,
449
+ "grad_norm": 0.730484664440155,
450
+ "learning_rate": 9.134999999999998e-05,
451
+ "loss": 7.210204315185547,
452
+ "step": 610
453
+ },
454
+ {
455
+ "epoch": 0.10474742355127556,
456
+ "grad_norm": 0.7423863410949707,
457
+ "learning_rate": 9.285e-05,
458
+ "loss": 7.166588592529297,
459
+ "step": 620
460
+ },
461
+ {
462
+ "epoch": 0.10643689812468322,
463
+ "grad_norm": 0.888006329536438,
464
+ "learning_rate": 9.434999999999999e-05,
465
+ "loss": 7.162047576904297,
466
+ "step": 630
467
+ },
468
+ {
469
+ "epoch": 0.10812637269809089,
470
+ "grad_norm": 0.9920506477355957,
471
+ "learning_rate": 9.585e-05,
472
+ "loss": 7.145941925048828,
473
+ "step": 640
474
+ },
475
+ {
476
+ "epoch": 0.10981584727149857,
477
+ "grad_norm": 0.8996961712837219,
478
+ "learning_rate": 9.735e-05,
479
+ "loss": 7.110871124267578,
480
+ "step": 650
481
+ },
482
+ {
483
+ "epoch": 0.11150532184490623,
484
+ "grad_norm": 0.7783015370368958,
485
+ "learning_rate": 9.884999999999999e-05,
486
+ "loss": 7.120133972167968,
487
+ "step": 660
488
+ },
489
+ {
490
+ "epoch": 0.1131947964183139,
491
+ "grad_norm": 0.8237811923027039,
492
+ "learning_rate": 0.00010035,
493
+ "loss": 7.082501220703125,
494
+ "step": 670
495
+ },
496
+ {
497
+ "epoch": 0.11488427099172158,
498
+ "grad_norm": 0.8586721420288086,
499
+ "learning_rate": 0.00010185,
500
+ "loss": 7.066880798339843,
501
+ "step": 680
502
+ },
503
+ {
504
+ "epoch": 0.11657374556512924,
505
+ "grad_norm": 0.9714040160179138,
506
+ "learning_rate": 0.00010334999999999998,
507
+ "loss": 7.058338928222656,
508
+ "step": 690
509
+ },
510
+ {
511
+ "epoch": 0.11826322013853692,
512
+ "grad_norm": 0.8379534482955933,
513
+ "learning_rate": 0.00010484999999999999,
514
+ "loss": 7.033222198486328,
515
+ "step": 700
516
+ },
517
+ {
518
+ "epoch": 0.11995269471194459,
519
+ "grad_norm": 1.147356629371643,
520
+ "learning_rate": 0.00010634999999999998,
521
+ "loss": 7.022679138183594,
522
+ "step": 710
523
+ },
524
+ {
525
+ "epoch": 0.12164216928535225,
526
+ "grad_norm": 0.946237325668335,
527
+ "learning_rate": 0.00010784999999999999,
528
+ "loss": 6.991328430175781,
529
+ "step": 720
530
+ },
531
+ {
532
+ "epoch": 0.12333164385875993,
533
+ "grad_norm": 1.0189383029937744,
534
+ "learning_rate": 0.00010934999999999999,
535
+ "loss": 6.9945930480957035,
536
+ "step": 730
537
+ },
538
+ {
539
+ "epoch": 0.1250211184321676,
540
+ "grad_norm": 1.0500218868255615,
541
+ "learning_rate": 0.00011084999999999998,
542
+ "loss": 6.963920593261719,
543
+ "step": 740
544
+ },
545
+ {
546
+ "epoch": 0.12671059300557527,
547
+ "grad_norm": 0.9184631109237671,
548
+ "learning_rate": 0.00011235,
549
+ "loss": 6.966143798828125,
550
+ "step": 750
551
+ },
552
+ {
553
+ "epoch": 0.12840006757898292,
554
+ "grad_norm": 0.7820301651954651,
555
+ "learning_rate": 0.00011384999999999999,
556
+ "loss": 6.938487243652344,
557
+ "step": 760
558
+ },
559
+ {
560
+ "epoch": 0.1300895421523906,
561
+ "grad_norm": 0.861544668674469,
562
+ "learning_rate": 0.00011535,
563
+ "loss": 6.920912170410157,
564
+ "step": 770
565
+ },
566
+ {
567
+ "epoch": 0.13177901672579828,
568
+ "grad_norm": 0.7475805878639221,
569
+ "learning_rate": 0.00011685,
570
+ "loss": 6.9139961242675785,
571
+ "step": 780
572
+ },
573
+ {
574
+ "epoch": 0.13346849129920596,
575
+ "grad_norm": 1.1264002323150635,
576
+ "learning_rate": 0.00011834999999999999,
577
+ "loss": 6.893434143066406,
578
+ "step": 790
579
+ },
580
+ {
581
+ "epoch": 0.1351579658726136,
582
+ "grad_norm": 0.869057834148407,
583
+ "learning_rate": 0.00011985,
584
+ "loss": 6.887288665771484,
585
+ "step": 800
586
+ },
587
+ {
588
+ "epoch": 0.1368474404460213,
589
+ "grad_norm": 0.831230878829956,
590
+ "learning_rate": 0.00012135,
591
+ "loss": 6.863740539550781,
592
+ "step": 810
593
+ },
594
+ {
595
+ "epoch": 0.13853691501942897,
596
+ "grad_norm": 1.068192720413208,
597
+ "learning_rate": 0.00012284999999999998,
598
+ "loss": 6.873618316650391,
599
+ "step": 820
600
+ },
601
+ {
602
+ "epoch": 0.14022638959283662,
603
+ "grad_norm": 0.9752544164657593,
604
+ "learning_rate": 0.00012435,
605
+ "loss": 6.84736328125,
606
+ "step": 830
607
+ },
608
+ {
609
+ "epoch": 0.1419158641662443,
610
+ "grad_norm": 0.9449293613433838,
611
+ "learning_rate": 0.00012585,
612
+ "loss": 6.829524230957031,
613
+ "step": 840
614
+ },
615
+ {
616
+ "epoch": 0.14360533873965198,
617
+ "grad_norm": 1.104444980621338,
618
+ "learning_rate": 0.00012734999999999998,
619
+ "loss": 6.8291679382324215,
620
+ "step": 850
621
+ },
622
+ {
623
+ "epoch": 0.14529481331305963,
624
+ "grad_norm": 0.9382540583610535,
625
+ "learning_rate": 0.00012885,
626
+ "loss": 6.8139289855957035,
627
+ "step": 860
628
+ },
629
+ {
630
+ "epoch": 0.1469842878864673,
631
+ "grad_norm": 0.7313889861106873,
632
+ "learning_rate": 0.00013035,
633
+ "loss": 6.798196411132812,
634
+ "step": 870
635
+ },
636
+ {
637
+ "epoch": 0.148673762459875,
638
+ "grad_norm": 0.7715932130813599,
639
+ "learning_rate": 0.00013184999999999998,
640
+ "loss": 6.805503845214844,
641
+ "step": 880
642
+ },
643
+ {
644
+ "epoch": 0.15036323703328264,
645
+ "grad_norm": 1.0334839820861816,
646
+ "learning_rate": 0.00013335,
647
+ "loss": 6.745892333984375,
648
+ "step": 890
649
+ },
650
+ {
651
+ "epoch": 0.15205271160669032,
652
+ "grad_norm": 1.1189385652542114,
653
+ "learning_rate": 0.00013485,
654
+ "loss": 6.769204711914062,
655
+ "step": 900
656
+ },
657
+ {
658
+ "epoch": 0.153742186180098,
659
+ "grad_norm": 1.289933681488037,
660
+ "learning_rate": 0.00013634999999999998,
661
+ "loss": 6.737556457519531,
662
+ "step": 910
663
+ },
664
+ {
665
+ "epoch": 0.15543166075350565,
666
+ "grad_norm": 1.0107234716415405,
667
+ "learning_rate": 0.00013785,
668
+ "loss": 6.7412353515625,
669
+ "step": 920
670
+ },
671
+ {
672
+ "epoch": 0.15712113532691332,
673
+ "grad_norm": 0.9233148097991943,
674
+ "learning_rate": 0.00013935,
675
+ "loss": 6.707360076904297,
676
+ "step": 930
677
+ },
678
+ {
679
+ "epoch": 0.158810609900321,
680
+ "grad_norm": 0.9500652551651001,
681
+ "learning_rate": 0.00014084999999999998,
682
+ "loss": 6.697336578369141,
683
+ "step": 940
684
+ },
685
+ {
686
+ "epoch": 0.16050008447372868,
687
+ "grad_norm": 1.0929033756256104,
688
+ "learning_rate": 0.00014235,
689
+ "loss": 6.673794555664062,
690
+ "step": 950
691
+ },
692
+ {
693
+ "epoch": 0.16218955904713633,
694
+ "grad_norm": 1.0387179851531982,
695
+ "learning_rate": 0.00014384999999999997,
696
+ "loss": 6.667636871337891,
697
+ "step": 960
698
+ },
699
+ {
700
+ "epoch": 0.163879033620544,
701
+ "grad_norm": 1.1298182010650635,
702
+ "learning_rate": 0.00014534999999999998,
703
+ "loss": 6.645402526855468,
704
+ "step": 970
705
+ },
706
+ {
707
+ "epoch": 0.1655685081939517,
708
+ "grad_norm": 0.9608763456344604,
709
+ "learning_rate": 0.00014685,
710
+ "loss": 6.658983612060547,
711
+ "step": 980
712
+ },
713
+ {
714
+ "epoch": 0.16725798276735934,
715
+ "grad_norm": 1.2303314208984375,
716
+ "learning_rate": 0.00014834999999999997,
717
+ "loss": 6.633333587646485,
718
+ "step": 990
719
+ },
720
+ {
721
+ "epoch": 0.16894745734076702,
722
+ "grad_norm": 0.9978023767471313,
723
+ "learning_rate": 0.00014984999999999998,
724
+ "loss": 6.6367958068847654,
725
+ "step": 1000
726
+ },
727
+ {
728
+ "epoch": 0.16894745734076702,
729
+ "eval_loss": 6.617567539215088,
730
+ "eval_runtime": 3.6651,
731
+ "eval_samples_per_second": 272.845,
732
+ "eval_steps_per_second": 5.73,
733
+ "step": 1000
734
  }
735
  ],
736
  "logging_steps": 10,
 
750
  "attributes": {}
751
  }
752
  },
753
+ "total_flos": 3.3445682085888e+16,
754
  "train_batch_size": 48,
755
  "trial_name": null,
756
  "trial_params": null