VERSIL91 commited on
Commit
25d43ef
·
verified ·
1 Parent(s): f9e0033

Training in progress, step 100, checkpoint

Browse files
last-checkpoint/optimizer.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:9d6e3e3f8da5ca34e31e7e1da106d1a2de25558024735de31f395304b7ea671c
3
  size 43122580
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:06ee945e16ae22c9075277531a6f8bc0ec708f7e7d0f7f18b4a9e5b5ec85eb27
3
  size 43122580
last-checkpoint/rng_state.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:7f53083d6bf416fa674a4d7654c4dc44772d047f9fd45a1c2d70bd5cf789eb73
3
  size 14244
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f4c6893fc4a9ed236abb30b22cb769913941a10e57b959801e5127eb964077c8
3
  size 14244
last-checkpoint/scheduler.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:f23e2214bcafb439ebc7528dcc283ef6218d509a276c0baff0743503ecbe3d92
3
  size 1064
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:49d60a69e2379be2053e816cbaff31e6c931b5922dd86c71c9eaf473299cbf62
3
  size 1064
last-checkpoint/trainer_state.json CHANGED
@@ -1,9 +1,9 @@
1
  {
2
  "best_metric": null,
3
  "best_model_checkpoint": null,
4
- "epoch": 0.4355716878402904,
5
  "eval_steps": 25,
6
- "global_step": 75,
7
  "is_hyper_param_search": false,
8
  "is_local_process_zero": true,
9
  "is_world_process_zero": true,
@@ -564,6 +564,189 @@
564
  "eval_samples_per_second": 29.963,
565
  "eval_steps_per_second": 14.981,
566
  "step": 75
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
567
  }
568
  ],
569
  "logging_steps": 1,
@@ -578,12 +761,12 @@
578
  "should_evaluate": false,
579
  "should_log": false,
580
  "should_save": true,
581
- "should_training_stop": false
582
  },
583
  "attributes": {}
584
  }
585
  },
586
- "total_flos": 8.956142606195098e+16,
587
  "train_batch_size": 2,
588
  "trial_name": null,
589
  "trial_params": null
 
1
  {
2
  "best_metric": null,
3
  "best_model_checkpoint": null,
4
+ "epoch": 0.5807622504537205,
5
  "eval_steps": 25,
6
+ "global_step": 100,
7
  "is_hyper_param_search": false,
8
  "is_local_process_zero": true,
9
  "is_world_process_zero": true,
 
564
  "eval_samples_per_second": 29.963,
565
  "eval_steps_per_second": 14.981,
566
  "step": 75
567
+ },
568
+ {
569
+ "epoch": 0.4413793103448276,
570
+ "grad_norm": NaN,
571
+ "learning_rate": 1.6543469682057106e-05,
572
+ "loss": 65.3235,
573
+ "step": 76
574
+ },
575
+ {
576
+ "epoch": 0.4471869328493648,
577
+ "grad_norm": NaN,
578
+ "learning_rate": 1.526708147705013e-05,
579
+ "loss": 0.1186,
580
+ "step": 77
581
+ },
582
+ {
583
+ "epoch": 0.452994555353902,
584
+ "grad_norm": NaN,
585
+ "learning_rate": 1.4033009983067452e-05,
586
+ "loss": 58.8162,
587
+ "step": 78
588
+ },
589
+ {
590
+ "epoch": 0.4588021778584392,
591
+ "grad_norm": NaN,
592
+ "learning_rate": 1.2842758726130283e-05,
593
+ "loss": 21.2564,
594
+ "step": 79
595
+ },
596
+ {
597
+ "epoch": 0.4646098003629764,
598
+ "grad_norm": NaN,
599
+ "learning_rate": 1.1697777844051105e-05,
600
+ "loss": 0.0,
601
+ "step": 80
602
+ },
603
+ {
604
+ "epoch": 0.4704174228675136,
605
+ "grad_norm": NaN,
606
+ "learning_rate": 1.0599462319663905e-05,
607
+ "loss": 0.0,
608
+ "step": 81
609
+ },
610
+ {
611
+ "epoch": 0.4762250453720508,
612
+ "grad_norm": NaN,
613
+ "learning_rate": 9.549150281252633e-06,
614
+ "loss": 0.0,
615
+ "step": 82
616
+ },
617
+ {
618
+ "epoch": 0.482032667876588,
619
+ "grad_norm": NaN,
620
+ "learning_rate": 8.548121372247918e-06,
621
+ "loss": 0.0,
622
+ "step": 83
623
+ },
624
+ {
625
+ "epoch": 0.48784029038112525,
626
+ "grad_norm": NaN,
627
+ "learning_rate": 7.597595192178702e-06,
628
+ "loss": 2906.1953,
629
+ "step": 84
630
+ },
631
+ {
632
+ "epoch": 0.49364791288566245,
633
+ "grad_norm": NaN,
634
+ "learning_rate": 6.698729810778065e-06,
635
+ "loss": 0.4129,
636
+ "step": 85
637
+ },
638
+ {
639
+ "epoch": 0.49945553539019966,
640
+ "grad_norm": NaN,
641
+ "learning_rate": 5.852620357053651e-06,
642
+ "loss": 658.3425,
643
+ "step": 86
644
+ },
645
+ {
646
+ "epoch": 0.5052631578947369,
647
+ "grad_norm": NaN,
648
+ "learning_rate": 5.060297685041659e-06,
649
+ "loss": 0.0,
650
+ "step": 87
651
+ },
652
+ {
653
+ "epoch": 0.511070780399274,
654
+ "grad_norm": NaN,
655
+ "learning_rate": 4.322727117869951e-06,
656
+ "loss": 0.0,
657
+ "step": 88
658
+ },
659
+ {
660
+ "epoch": 0.5168784029038113,
661
+ "grad_norm": NaN,
662
+ "learning_rate": 3.6408072716606346e-06,
663
+ "loss": 2626.5364,
664
+ "step": 89
665
+ },
666
+ {
667
+ "epoch": 0.5226860254083484,
668
+ "grad_norm": NaN,
669
+ "learning_rate": 3.0153689607045845e-06,
670
+ "loss": 101.4575,
671
+ "step": 90
672
+ },
673
+ {
674
+ "epoch": 0.5284936479128857,
675
+ "grad_norm": NaN,
676
+ "learning_rate": 2.4471741852423237e-06,
677
+ "loss": 0.9226,
678
+ "step": 91
679
+ },
680
+ {
681
+ "epoch": 0.5343012704174228,
682
+ "grad_norm": NaN,
683
+ "learning_rate": 1.9369152030840556e-06,
684
+ "loss": 0.0,
685
+ "step": 92
686
+ },
687
+ {
688
+ "epoch": 0.5401088929219601,
689
+ "grad_norm": NaN,
690
+ "learning_rate": 1.4852136862001764e-06,
691
+ "loss": 0.0,
692
+ "step": 93
693
+ },
694
+ {
695
+ "epoch": 0.5459165154264973,
696
+ "grad_norm": NaN,
697
+ "learning_rate": 1.0926199633097157e-06,
698
+ "loss": 0.0,
699
+ "step": 94
700
+ },
701
+ {
702
+ "epoch": 0.5517241379310345,
703
+ "grad_norm": NaN,
704
+ "learning_rate": 7.596123493895991e-07,
705
+ "loss": 13.896,
706
+ "step": 95
707
+ },
708
+ {
709
+ "epoch": 0.5575317604355717,
710
+ "grad_norm": NaN,
711
+ "learning_rate": 4.865965629214819e-07,
712
+ "loss": 0.0,
713
+ "step": 96
714
+ },
715
+ {
716
+ "epoch": 0.5633393829401089,
717
+ "grad_norm": NaN,
718
+ "learning_rate": 2.7390523158633554e-07,
719
+ "loss": 0.0,
720
+ "step": 97
721
+ },
722
+ {
723
+ "epoch": 0.5691470054446461,
724
+ "grad_norm": NaN,
725
+ "learning_rate": 1.2179748700879012e-07,
726
+ "loss": 0.4251,
727
+ "step": 98
728
+ },
729
+ {
730
+ "epoch": 0.5749546279491833,
731
+ "grad_norm": NaN,
732
+ "learning_rate": 3.04586490452119e-08,
733
+ "loss": 2705.8181,
734
+ "step": 99
735
+ },
736
+ {
737
+ "epoch": 0.5807622504537205,
738
+ "grad_norm": NaN,
739
+ "learning_rate": 0.0,
740
+ "loss": 0.0,
741
+ "step": 100
742
+ },
743
+ {
744
+ "epoch": 0.5807622504537205,
745
+ "eval_loss": NaN,
746
+ "eval_runtime": 9.6878,
747
+ "eval_samples_per_second": 29.934,
748
+ "eval_steps_per_second": 14.967,
749
+ "step": 100
750
  }
751
  ],
752
  "logging_steps": 1,
 
761
  "should_evaluate": false,
762
  "should_log": false,
763
  "should_save": true,
764
+ "should_training_stop": true
765
  },
766
  "attributes": {}
767
  }
768
  },
769
+ "total_flos": 1.183490272961495e+17,
770
  "train_batch_size": 2,
771
  "trial_name": null,
772
  "trial_params": null