AlekseyKorshuk commited on
Commit
48763f7
1 Parent(s): d5a6f79

huggingartists

Browse files
README.md CHANGED
@@ -45,15 +45,15 @@ from datasets import load_dataset
45
  dataset = load_dataset("huggingartists/snoop-dogg")
46
  ```
47
 
48
- [Explore the data](https://wandb.ai/huggingartists/huggingartists/runs/255rv52h/artifacts), which is tracked with [W&B artifacts](https://docs.wandb.com/artifacts) at every step of the pipeline.
49
 
50
  ## Training procedure
51
 
52
  The model is based on a pre-trained [GPT-2](https://huggingface.co/gpt2) which is fine-tuned on Snoop Dogg's lyrics.
53
 
54
- Hyperparameters and metrics are recorded in the [W&B training run](https://wandb.ai/huggingartists/huggingartists/runs/xlo6sfmn) for full transparency and reproducibility.
55
 
56
- At the end of training, [the final model](https://wandb.ai/huggingartists/huggingartists/runs/xlo6sfmn/artifacts) is logged and versioned.
57
 
58
  ## How to use
59
 
 
45
  dataset = load_dataset("huggingartists/snoop-dogg")
46
  ```
47
 
48
+ [Explore the data](https://wandb.ai/huggingartists/huggingartists/runs/xru6xdjl/artifacts), which is tracked with [W&B artifacts](https://docs.wandb.com/artifacts) at every step of the pipeline.
49
 
50
  ## Training procedure
51
 
52
  The model is based on a pre-trained [GPT-2](https://huggingface.co/gpt2) which is fine-tuned on Snoop Dogg's lyrics.
53
 
54
+ Hyperparameters and metrics are recorded in the [W&B training run](https://wandb.ai/huggingartists/huggingartists/runs/1o72aoie) for full transparency and reproducibility.
55
 
56
+ At the end of training, [the final model](https://wandb.ai/huggingartists/huggingartists/runs/1o72aoie/artifacts) is logged and versioned.
57
 
58
  ## How to use
59
 
config.json CHANGED
@@ -1,5 +1,5 @@
1
  {
2
- "_name_or_path": "gpt2",
3
  "activation_function": "gelu_new",
4
  "architectures": [
5
  "GPT2LMHeadModel"
@@ -18,7 +18,9 @@
18
  "n_inner": null,
19
  "n_layer": 12,
20
  "n_positions": 1024,
 
21
  "resid_pdrop": 0.1,
 
22
  "scale_attn_weights": true,
23
  "summary_activation": null,
24
  "summary_first_dropout": 0.1,
@@ -35,7 +37,7 @@
35
  }
36
  },
37
  "torch_dtype": "float32",
38
- "transformers_version": "4.9.2",
39
  "use_cache": true,
40
  "vocab_size": 50257
41
  }
 
1
  {
2
+ "_name_or_path": "snoop-dogg",
3
  "activation_function": "gelu_new",
4
  "architectures": [
5
  "GPT2LMHeadModel"
 
18
  "n_inner": null,
19
  "n_layer": 12,
20
  "n_positions": 1024,
21
+ "reorder_and_upcast_attn": false,
22
  "resid_pdrop": 0.1,
23
+ "scale_attn_by_inverse_layer_idx": false,
24
  "scale_attn_weights": true,
25
  "summary_activation": null,
26
  "summary_first_dropout": 0.1,
 
37
  }
38
  },
39
  "torch_dtype": "float32",
40
+ "transformers_version": "4.18.0",
41
  "use_cache": true,
42
  "vocab_size": 50257
43
  }
evaluation.txt CHANGED
@@ -1 +1 @@
1
- {"eval_loss": 3.417607307434082, "eval_runtime": 35.077, "eval_samples_per_second": 21.467, "eval_steps_per_second": 2.708, "epoch": 1.0}
 
1
+ {"eval_loss": 3.1907694339752197, "eval_runtime": 18.6025, "eval_samples_per_second": 40.478, "eval_steps_per_second": 5.107, "epoch": 3.0}
flax_model.msgpack CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:82c220da81cdcb2810fb296930ad027595bc527561d769450b6c1004a1482d7f
3
  size 497764120
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3306f594fc320bfc82351986cb36289796a080b3749471511f81f224e16f7a3e
3
  size 497764120
optimizer.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:26c89df95a623615699b8cb15d454e0e51988ed659e4edea185b59d09df7320d
3
  size 995604017
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5cbc63612b54039ddb66de58d8e52f98d5b29c55b31218574bef41e8025e58ae
3
  size 995604017
pytorch_model.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:04b799dfe236e2ce13953e8b7372b70f8194c3a1fbb7eba80985f1a7710b60b3
3
- size 510403817
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e63ca0257324e3e10601effd0c9420a2d44bb90cb750193eab1ced9503347139
3
+ size 510396521
rng_state.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:ca4efa8760670b265a20db81d357178a028968297f1568342afc561ffd367cfb
3
- size 14503
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:032b548e33a403b4088f0e3fbd13e2e53669f3f9f0dfa3cebb95ac16d40d4c4f
3
+ size 14567
scheduler.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:b772c417587c5d01fc5e54458d542c1222437cf21ae50383dc362c9bbceaf117
3
  size 623
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7a287eb795244c2ae6a56f293ee55a0613fb0462f5a571b1ccc1a364ce3de351
3
  size 623
tokenizer.json CHANGED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json CHANGED
@@ -1 +1 @@
1
- {"unk_token": "<|endoftext|>", "bos_token": "<|endoftext|>", "eos_token": "<|endoftext|>", "add_prefix_space": false, "model_max_length": 1024, "special_tokens_map_file": null, "name_or_path": "gpt2", "tokenizer_class": "GPT2Tokenizer"}
 
1
+ {"unk_token": "<|endoftext|>", "bos_token": "<|endoftext|>", "eos_token": "<|endoftext|>", "add_prefix_space": false, "model_max_length": 1024, "special_tokens_map_file": null, "name_or_path": "huggingartists/snoop-dogg", "tokenizer_class": "GPT2Tokenizer"}
trainer_state.json CHANGED
@@ -1,8 +1,8 @@
1
  {
2
- "best_metric": 3.417607307434082,
3
- "best_model_checkpoint": "output/snoop-dogg/checkpoint-538",
4
- "epoch": 1.0,
5
- "global_step": 538,
6
  "is_hyper_param_search": false,
7
  "is_local_process_zero": true,
8
  "is_world_process_zero": true,
@@ -656,11 +656,1317 @@
656
  "eval_samples_per_second": 21.535,
657
  "eval_steps_per_second": 2.717,
658
  "step": 538
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
659
  }
660
  ],
661
- "max_steps": 538,
662
- "num_train_epochs": 1,
663
- "total_flos": 561908514816000.0,
664
  "trial_name": null,
665
  "trial_params": null
666
  }
 
1
  {
2
+ "best_metric": 3.1907694339752197,
3
+ "best_model_checkpoint": "output/snoop-dogg/checkpoint-1614",
4
+ "epoch": 3.0,
5
+ "global_step": 1614,
6
  "is_hyper_param_search": false,
7
  "is_local_process_zero": true,
8
  "is_world_process_zero": true,
 
656
  "eval_samples_per_second": 21.535,
657
  "eval_steps_per_second": 2.717,
658
  "step": 538
659
+ },
660
+ {
661
+ "epoch": 1.0,
662
+ "learning_rate": 4.678260156942859e-09,
663
+ "loss": 3.4076,
664
+ "step": 540
665
+ },
666
+ {
667
+ "epoch": 1.01,
668
+ "learning_rate": 5.730135925604576e-08,
669
+ "loss": 3.3309,
670
+ "step": 545
671
+ },
672
+ {
673
+ "epoch": 1.02,
674
+ "learning_rate": 1.683503770757803e-07,
675
+ "loss": 3.3307,
676
+ "step": 550
677
+ },
678
+ {
679
+ "epoch": 1.03,
680
+ "learning_rate": 3.377306552350485e-07,
681
+ "loss": 3.1417,
682
+ "step": 555
683
+ },
684
+ {
685
+ "epoch": 1.04,
686
+ "learning_rate": 5.6529781368228e-07,
687
+ "loss": 3.2028,
688
+ "step": 560
689
+ },
690
+ {
691
+ "epoch": 1.05,
692
+ "learning_rate": 8.508578737652368e-07,
693
+ "loss": 3.4381,
694
+ "step": 565
695
+ },
696
+ {
697
+ "epoch": 1.06,
698
+ "learning_rate": 1.1941674235787645e-06,
699
+ "loss": 3.2711,
700
+ "step": 570
701
+ },
702
+ {
703
+ "epoch": 1.07,
704
+ "learning_rate": 1.5949338254495494e-06,
705
+ "loss": 3.2736,
706
+ "step": 575
707
+ },
708
+ {
709
+ "epoch": 1.08,
710
+ "learning_rate": 2.0528154653809706e-06,
711
+ "loss": 3.2426,
712
+ "step": 580
713
+ },
714
+ {
715
+ "epoch": 1.09,
716
+ "learning_rate": 2.5674220442454935e-06,
717
+ "loss": 3.3709,
718
+ "step": 585
719
+ },
720
+ {
721
+ "epoch": 1.1,
722
+ "learning_rate": 3.13831491047635e-06,
723
+ "loss": 3.5776,
724
+ "step": 590
725
+ },
726
+ {
727
+ "epoch": 1.11,
728
+ "learning_rate": 3.7650074339748796e-06,
729
+ "loss": 3.3789,
730
+ "step": 595
731
+ },
732
+ {
733
+ "epoch": 1.12,
734
+ "learning_rate": 4.446965420914962e-06,
735
+ "loss": 3.4289,
736
+ "step": 600
737
+ },
738
+ {
739
+ "epoch": 1.12,
740
+ "learning_rate": 5.1836075690907296e-06,
741
+ "loss": 3.3452,
742
+ "step": 605
743
+ },
744
+ {
745
+ "epoch": 1.13,
746
+ "learning_rate": 5.974305963419616e-06,
747
+ "loss": 3.3496,
748
+ "step": 610
749
+ },
750
+ {
751
+ "epoch": 1.14,
752
+ "learning_rate": 6.818386611178326e-06,
753
+ "loss": 3.3166,
754
+ "step": 615
755
+ },
756
+ {
757
+ "epoch": 1.15,
758
+ "learning_rate": 7.715130016515405e-06,
759
+ "loss": 3.3951,
760
+ "step": 620
761
+ },
762
+ {
763
+ "epoch": 1.16,
764
+ "learning_rate": 8.66377179375079e-06,
765
+ "loss": 3.3529,
766
+ "step": 625
767
+ },
768
+ {
769
+ "epoch": 1.17,
770
+ "learning_rate": 9.66350331893958e-06,
771
+ "loss": 3.3422,
772
+ "step": 630
773
+ },
774
+ {
775
+ "epoch": 1.18,
776
+ "learning_rate": 1.0713472419144369e-05,
777
+ "loss": 3.3403,
778
+ "step": 635
779
+ },
780
+ {
781
+ "epoch": 1.19,
782
+ "learning_rate": 1.1812784098829224e-05,
783
+ "loss": 3.3656,
784
+ "step": 640
785
+ },
786
+ {
787
+ "epoch": 1.2,
788
+ "learning_rate": 1.2960501302755281e-05,
789
+ "loss": 3.2795,
790
+ "step": 645
791
+ },
792
+ {
793
+ "epoch": 1.21,
794
+ "learning_rate": 1.41556457147284e-05,
795
+ "loss": 3.3825,
796
+ "step": 650
797
+ },
798
+ {
799
+ "epoch": 1.22,
800
+ "learning_rate": 1.5397198591517605e-05,
801
+ "loss": 3.1902,
802
+ "step": 655
803
+ },
804
+ {
805
+ "epoch": 1.23,
806
+ "learning_rate": 1.668410163123363e-05,
807
+ "loss": 3.3563,
808
+ "step": 660
809
+ },
810
+ {
811
+ "epoch": 1.24,
812
+ "learning_rate": 1.8015257875427347e-05,
813
+ "loss": 3.2479,
814
+ "step": 665
815
+ },
816
+ {
817
+ "epoch": 1.25,
818
+ "learning_rate": 1.9389532644139112e-05,
819
+ "loss": 3.3764,
820
+ "step": 670
821
+ },
822
+ {
823
+ "epoch": 1.25,
824
+ "learning_rate": 2.080575450310203e-05,
825
+ "loss": 3.2164,
826
+ "step": 675
827
+ },
828
+ {
829
+ "epoch": 1.26,
830
+ "learning_rate": 2.226271626227443e-05,
831
+ "loss": 3.1518,
832
+ "step": 680
833
+ },
834
+ {
835
+ "epoch": 1.27,
836
+ "learning_rate": 2.375917600485109e-05,
837
+ "loss": 3.2922,
838
+ "step": 685
839
+ },
840
+ {
841
+ "epoch": 1.28,
842
+ "learning_rate": 2.5293858145875134e-05,
843
+ "loss": 3.4571,
844
+ "step": 690
845
+ },
846
+ {
847
+ "epoch": 1.29,
848
+ "learning_rate": 2.6865454519549044e-05,
849
+ "loss": 3.3273,
850
+ "step": 695
851
+ },
852
+ {
853
+ "epoch": 1.3,
854
+ "learning_rate": 2.8472625494317417e-05,
855
+ "loss": 3.2389,
856
+ "step": 700
857
+ },
858
+ {
859
+ "epoch": 1.31,
860
+ "learning_rate": 3.011400111477139e-05,
861
+ "loss": 3.2702,
862
+ "step": 705
863
+ },
864
+ {
865
+ "epoch": 1.32,
866
+ "learning_rate": 3.178818226940102e-05,
867
+ "loss": 3.6087,
868
+ "step": 710
869
+ },
870
+ {
871
+ "epoch": 1.33,
872
+ "learning_rate": 3.349374188320044e-05,
873
+ "loss": 3.3782,
874
+ "step": 715
875
+ },
876
+ {
877
+ "epoch": 1.34,
878
+ "learning_rate": 3.522922613410924e-05,
879
+ "loss": 3.428,
880
+ "step": 720
881
+ },
882
+ {
883
+ "epoch": 1.35,
884
+ "learning_rate": 3.69931556922527e-05,
885
+ "loss": 3.2301,
886
+ "step": 725
887
+ },
888
+ {
889
+ "epoch": 1.36,
890
+ "learning_rate": 3.878402698092561e-05,
891
+ "loss": 3.3942,
892
+ "step": 730
893
+ },
894
+ {
895
+ "epoch": 1.37,
896
+ "learning_rate": 4.0600313458243476e-05,
897
+ "loss": 3.3824,
898
+ "step": 735
899
+ },
900
+ {
901
+ "epoch": 1.38,
902
+ "learning_rate": 4.244046691836963e-05,
903
+ "loss": 3.4915,
904
+ "step": 740
905
+ },
906
+ {
907
+ "epoch": 1.38,
908
+ "learning_rate": 4.430291881120861e-05,
909
+ "loss": 3.2738,
910
+ "step": 745
911
+ },
912
+ {
913
+ "epoch": 1.39,
914
+ "learning_rate": 4.618608157944113e-05,
915
+ "loss": 3.3024,
916
+ "step": 750
917
+ },
918
+ {
919
+ "epoch": 1.4,
920
+ "learning_rate": 4.808835001176069e-05,
921
+ "loss": 3.3214,
922
+ "step": 755
923
+ },
924
+ {
925
+ "epoch": 1.41,
926
+ "learning_rate": 5.000810261115872e-05,
927
+ "loss": 3.5582,
928
+ "step": 760
929
+ },
930
+ {
931
+ "epoch": 1.42,
932
+ "learning_rate": 5.194370297709118e-05,
933
+ "loss": 3.2745,
934
+ "step": 765
935
+ },
936
+ {
937
+ "epoch": 1.43,
938
+ "learning_rate": 5.3893501200349724e-05,
939
+ "loss": 3.2672,
940
+ "step": 770
941
+ },
942
+ {
943
+ "epoch": 1.44,
944
+ "learning_rate": 5.585583526944705e-05,
945
+ "loss": 3.3916,
946
+ "step": 775
947
+ },
948
+ {
949
+ "epoch": 1.45,
950
+ "learning_rate": 5.7829032487318437e-05,
951
+ "loss": 3.5783,
952
+ "step": 780
953
+ },
954
+ {
955
+ "epoch": 1.46,
956
+ "learning_rate": 5.9811410897131736e-05,
957
+ "loss": 3.1478,
958
+ "step": 785
959
+ },
960
+ {
961
+ "epoch": 1.47,
962
+ "learning_rate": 6.180128071599044e-05,
963
+ "loss": 3.2909,
964
+ "step": 790
965
+ },
966
+ {
967
+ "epoch": 1.48,
968
+ "learning_rate": 6.379694577530755e-05,
969
+ "loss": 3.4915,
970
+ "step": 795
971
+ },
972
+ {
973
+ "epoch": 1.49,
974
+ "learning_rate": 6.57967049666228e-05,
975
+ "loss": 3.1955,
976
+ "step": 800
977
+ },
978
+ {
979
+ "epoch": 1.5,
980
+ "learning_rate": 6.77988536916305e-05,
981
+ "loss": 3.4232,
982
+ "step": 805
983
+ },
984
+ {
985
+ "epoch": 1.51,
986
+ "learning_rate": 6.980168531518204e-05,
987
+ "loss": 3.3181,
988
+ "step": 810
989
+ },
990
+ {
991
+ "epoch": 1.51,
992
+ "learning_rate": 7.1803492620025e-05,
993
+ "loss": 3.4153,
994
+ "step": 815
995
+ },
996
+ {
997
+ "epoch": 1.52,
998
+ "learning_rate": 7.380256926203807e-05,
999
+ "loss": 3.4825,
1000
+ "step": 820
1001
+ },
1002
+ {
1003
+ "epoch": 1.53,
1004
+ "learning_rate": 7.579721122472176e-05,
1005
+ "loss": 3.2802,
1006
+ "step": 825
1007
+ },
1008
+ {
1009
+ "epoch": 1.54,
1010
+ "learning_rate": 7.778571827170516e-05,
1011
+ "loss": 3.4058,
1012
+ "step": 830
1013
+ },
1014
+ {
1015
+ "epoch": 1.55,
1016
+ "learning_rate": 7.976639539603049e-05,
1017
+ "loss": 3.385,
1018
+ "step": 835
1019
+ },
1020
+ {
1021
+ "epoch": 1.56,
1022
+ "learning_rate": 8.173755426497999e-05,
1023
+ "loss": 3.361,
1024
+ "step": 840
1025
+ },
1026
+ {
1027
+ "epoch": 1.57,
1028
+ "learning_rate": 8.36975146592138e-05,
1029
+ "loss": 3.3453,
1030
+ "step": 845
1031
+ },
1032
+ {
1033
+ "epoch": 1.58,
1034
+ "learning_rate": 8.564460590499189e-05,
1035
+ "loss": 3.2614,
1036
+ "step": 850
1037
+ },
1038
+ {
1039
+ "epoch": 1.59,
1040
+ "learning_rate": 8.757716829825938e-05,
1041
+ "loss": 3.2934,
1042
+ "step": 855
1043
+ },
1044
+ {
1045
+ "epoch": 1.6,
1046
+ "learning_rate": 8.949355451938121e-05,
1047
+ "loss": 3.2375,
1048
+ "step": 860
1049
+ },
1050
+ {
1051
+ "epoch": 1.61,
1052
+ "learning_rate": 9.139213103732031e-05,
1053
+ "loss": 3.5485,
1054
+ "step": 865
1055
+ },
1056
+ {
1057
+ "epoch": 1.62,
1058
+ "learning_rate": 9.327127950206244e-05,
1059
+ "loss": 3.6656,
1060
+ "step": 870
1061
+ },
1062
+ {
1063
+ "epoch": 1.63,
1064
+ "learning_rate": 9.512939812410057e-05,
1065
+ "loss": 3.2755,
1066
+ "step": 875
1067
+ },
1068
+ {
1069
+ "epoch": 1.64,
1070
+ "learning_rate": 9.69649030398032e-05,
1071
+ "loss": 3.3555,
1072
+ "step": 880
1073
+ },
1074
+ {
1075
+ "epoch": 1.64,
1076
+ "learning_rate": 9.877622966150243e-05,
1077
+ "loss": 3.3756,
1078
+ "step": 885
1079
+ },
1080
+ {
1081
+ "epoch": 1.65,
1082
+ "learning_rate": 0.00010056183401115142,
1083
+ "loss": 3.0988,
1084
+ "step": 890
1085
+ },
1086
+ {
1087
+ "epoch": 1.66,
1088
+ "learning_rate": 0.00010232019403641382,
1089
+ "loss": 3.2778,
1090
+ "step": 895
1091
+ },
1092
+ {
1093
+ "epoch": 1.67,
1094
+ "learning_rate": 0.00010404981090806416,
1095
+ "loss": 3.4258,
1096
+ "step": 900
1097
+ },
1098
+ {
1099
+ "epoch": 1.68,
1100
+ "learning_rate": 0.00010574921029759243,
1101
+ "loss": 3.3742,
1102
+ "step": 905
1103
+ },
1104
+ {
1105
+ "epoch": 1.69,
1106
+ "learning_rate": 0.00010741694363392428,
1107
+ "loss": 3.3941,
1108
+ "step": 910
1109
+ },
1110
+ {
1111
+ "epoch": 1.7,
1112
+ "learning_rate": 0.00010905158933818573,
1113
+ "loss": 3.3793,
1114
+ "step": 915
1115
+ },
1116
+ {
1117
+ "epoch": 1.71,
1118
+ "learning_rate": 0.00011065175403545939,
1119
+ "loss": 3.3372,
1120
+ "step": 920
1121
+ },
1122
+ {
1123
+ "epoch": 1.72,
1124
+ "learning_rate": 0.00011221607374249983,
1125
+ "loss": 3.3405,
1126
+ "step": 925
1127
+ },
1128
+ {
1129
+ "epoch": 1.73,
1130
+ "learning_rate": 0.00011374321503039532,
1131
+ "loss": 3.2013,
1132
+ "step": 930
1133
+ },
1134
+ {
1135
+ "epoch": 1.74,
1136
+ "learning_rate": 0.000115231876161185,
1137
+ "loss": 3.3862,
1138
+ "step": 935
1139
+ },
1140
+ {
1141
+ "epoch": 1.75,
1142
+ "learning_rate": 0.00011668078819746258,
1143
+ "loss": 3.3706,
1144
+ "step": 940
1145
+ },
1146
+ {
1147
+ "epoch": 1.76,
1148
+ "learning_rate": 0.00011808871608402098,
1149
+ "loss": 3.5269,
1150
+ "step": 945
1151
+ },
1152
+ {
1153
+ "epoch": 1.77,
1154
+ "learning_rate": 0.00011945445970061549,
1155
+ "loss": 3.4064,
1156
+ "step": 950
1157
+ },
1158
+ {
1159
+ "epoch": 1.78,
1160
+ "learning_rate": 0.00012077685488494853,
1161
+ "loss": 3.4603,
1162
+ "step": 955
1163
+ },
1164
+ {
1165
+ "epoch": 1.78,
1166
+ "learning_rate": 0.00012205477442500345,
1167
+ "loss": 3.3909,
1168
+ "step": 960
1169
+ },
1170
+ {
1171
+ "epoch": 1.79,
1172
+ "learning_rate": 0.00012328712901988232,
1173
+ "loss": 3.3253,
1174
+ "step": 965
1175
+ },
1176
+ {
1177
+ "epoch": 1.8,
1178
+ "learning_rate": 0.0001244728682083278,
1179
+ "loss": 3.384,
1180
+ "step": 970
1181
+ },
1182
+ {
1183
+ "epoch": 1.81,
1184
+ "learning_rate": 0.00012561098126413823,
1185
+ "loss": 3.3282,
1186
+ "step": 975
1187
+ },
1188
+ {
1189
+ "epoch": 1.82,
1190
+ "learning_rate": 0.0001267004980577125,
1191
+ "loss": 3.3693,
1192
+ "step": 980
1193
+ },
1194
+ {
1195
+ "epoch": 1.83,
1196
+ "learning_rate": 0.0001277404898829901,
1197
+ "loss": 3.3716,
1198
+ "step": 985
1199
+ },
1200
+ {
1201
+ "epoch": 1.84,
1202
+ "learning_rate": 0.00012873007024908207,
1203
+ "loss": 3.4564,
1204
+ "step": 990
1205
+ },
1206
+ {
1207
+ "epoch": 1.85,
1208
+ "learning_rate": 0.0001296683956359172,
1209
+ "loss": 3.4711,
1210
+ "step": 995
1211
+ },
1212
+ {
1213
+ "epoch": 1.86,
1214
+ "learning_rate": 0.00013055466621326016,
1215
+ "loss": 3.2512,
1216
+ "step": 1000
1217
+ },
1218
+ {
1219
+ "epoch": 1.87,
1220
+ "learning_rate": 0.0001313881265224882,
1221
+ "loss": 3.3624,
1222
+ "step": 1005
1223
+ },
1224
+ {
1225
+ "epoch": 1.88,
1226
+ "learning_rate": 0.00013216806612054534,
1227
+ "loss": 3.3469,
1228
+ "step": 1010
1229
+ },
1230
+ {
1231
+ "epoch": 1.89,
1232
+ "learning_rate": 0.00013289382018552548,
1233
+ "loss": 3.3339,
1234
+ "step": 1015
1235
+ },
1236
+ {
1237
+ "epoch": 1.9,
1238
+ "learning_rate": 0.00013356477008336762,
1239
+ "loss": 3.5051,
1240
+ "step": 1020
1241
+ },
1242
+ {
1243
+ "epoch": 1.91,
1244
+ "learning_rate": 0.00013418034389518066,
1245
+ "loss": 3.6009,
1246
+ "step": 1025
1247
+ },
1248
+ {
1249
+ "epoch": 1.91,
1250
+ "learning_rate": 0.00013474001690474816,
1251
+ "loss": 3.4188,
1252
+ "step": 1030
1253
+ },
1254
+ {
1255
+ "epoch": 1.92,
1256
+ "learning_rate": 0.00013524331204579738,
1257
+ "loss": 3.3583,
1258
+ "step": 1035
1259
+ },
1260
+ {
1261
+ "epoch": 1.93,
1262
+ "learning_rate": 0.00013568980030865128,
1263
+ "loss": 3.4374,
1264
+ "step": 1040
1265
+ },
1266
+ {
1267
+ "epoch": 1.94,
1268
+ "learning_rate": 0.00013607910110591736,
1269
+ "loss": 3.5138,
1270
+ "step": 1045
1271
+ },
1272
+ {
1273
+ "epoch": 1.95,
1274
+ "learning_rate": 0.0001364108825969008,
1275
+ "loss": 3.2506,
1276
+ "step": 1050
1277
+ },
1278
+ {
1279
+ "epoch": 1.96,
1280
+ "learning_rate": 0.00013668486197046618,
1281
+ "loss": 3.313,
1282
+ "step": 1055
1283
+ },
1284
+ {
1285
+ "epoch": 1.97,
1286
+ "learning_rate": 0.00013690080568610629,
1287
+ "loss": 3.4122,
1288
+ "step": 1060
1289
+ },
1290
+ {
1291
+ "epoch": 1.98,
1292
+ "learning_rate": 0.00013705852967301225,
1293
+ "loss": 3.2444,
1294
+ "step": 1065
1295
+ },
1296
+ {
1297
+ "epoch": 1.99,
1298
+ "learning_rate": 0.00013715789948697598,
1299
+ "loss": 3.3139,
1300
+ "step": 1070
1301
+ },
1302
+ {
1303
+ "epoch": 2.0,
1304
+ "learning_rate": 0.0001371988304249906,
1305
+ "loss": 3.4089,
1306
+ "step": 1075
1307
+ },
1308
+ {
1309
+ "epoch": 2.0,
1310
+ "eval_loss": 3.2667410373687744,
1311
+ "eval_runtime": 18.6928,
1312
+ "eval_samples_per_second": 40.283,
1313
+ "eval_steps_per_second": 5.082,
1314
+ "step": 1076
1315
+ },
1316
+ {
1317
+ "epoch": 2.01,
1318
+ "learning_rate": 0.00013718128759745147,
1319
+ "loss": 3.1857,
1320
+ "step": 1080
1321
+ },
1322
+ {
1323
+ "epoch": 2.02,
1324
+ "learning_rate": 0.00013710528595789614,
1325
+ "loss": 3.1502,
1326
+ "step": 1085
1327
+ },
1328
+ {
1329
+ "epoch": 2.03,
1330
+ "learning_rate": 0.00013697089029025781,
1331
+ "loss": 3.244,
1332
+ "step": 1090
1333
+ },
1334
+ {
1335
+ "epoch": 2.04,
1336
+ "learning_rate": 0.00013677821515364362,
1337
+ "loss": 2.9782,
1338
+ "step": 1095
1339
+ },
1340
+ {
1341
+ "epoch": 2.04,
1342
+ "learning_rate": 0.00013652742478468407,
1343
+ "loss": 3.4309,
1344
+ "step": 1100
1345
+ },
1346
+ {
1347
+ "epoch": 2.05,
1348
+ "learning_rate": 0.00013621873295753753,
1349
+ "loss": 3.3141,
1350
+ "step": 1105
1351
+ },
1352
+ {
1353
+ "epoch": 2.06,
1354
+ "learning_rate": 0.0001358524028016687,
1355
+ "loss": 3.2167,
1356
+ "step": 1110
1357
+ },
1358
+ {
1359
+ "epoch": 2.07,
1360
+ "learning_rate": 0.0001354287465775566,
1361
+ "loss": 3.2457,
1362
+ "step": 1115
1363
+ },
1364
+ {
1365
+ "epoch": 2.08,
1366
+ "learning_rate": 0.00013494812541052316,
1367
+ "loss": 3.2054,
1368
+ "step": 1120
1369
+ },
1370
+ {
1371
+ "epoch": 2.09,
1372
+ "learning_rate": 0.00013441094898290952,
1373
+ "loss": 3.1213,
1374
+ "step": 1125
1375
+ },
1376
+ {
1377
+ "epoch": 2.1,
1378
+ "learning_rate": 0.00013381767518486174,
1379
+ "loss": 3.1849,
1380
+ "step": 1130
1381
+ },
1382
+ {
1383
+ "epoch": 2.11,
1384
+ "learning_rate": 0.00013316880972402475,
1385
+ "loss": 3.2974,
1386
+ "step": 1135
1387
+ },
1388
+ {
1389
+ "epoch": 2.12,
1390
+ "learning_rate": 0.0001324649056944761,
1391
+ "loss": 3.0743,
1392
+ "step": 1140
1393
+ },
1394
+ {
1395
+ "epoch": 2.13,
1396
+ "learning_rate": 0.0001317065631052675,
1397
+ "loss": 3.3105,
1398
+ "step": 1145
1399
+ },
1400
+ {
1401
+ "epoch": 2.14,
1402
+ "learning_rate": 0.00013089442836897625,
1403
+ "loss": 3.0161,
1404
+ "step": 1150
1405
+ },
1406
+ {
1407
+ "epoch": 2.15,
1408
+ "learning_rate": 0.00013002919375070177,
1409
+ "loss": 3.1369,
1410
+ "step": 1155
1411
+ },
1412
+ {
1413
+ "epoch": 2.16,
1414
+ "learning_rate": 0.0001291115967779778,
1415
+ "loss": 3.2544,
1416
+ "step": 1160
1417
+ },
1418
+ {
1419
+ "epoch": 2.17,
1420
+ "learning_rate": 0.00012814241961210242,
1421
+ "loss": 3.2921,
1422
+ "step": 1165
1423
+ },
1424
+ {
1425
+ "epoch": 2.17,
1426
+ "learning_rate": 0.00012712248838142262,
1427
+ "loss": 3.2056,
1428
+ "step": 1170
1429
+ },
1430
+ {
1431
+ "epoch": 2.18,
1432
+ "learning_rate": 0.00012605267247714066,
1433
+ "loss": 3.0081,
1434
+ "step": 1175
1435
+ },
1436
+ {
1437
+ "epoch": 2.19,
1438
+ "learning_rate": 0.00012493388381224383,
1439
+ "loss": 3.3426,
1440
+ "step": 1180
1441
+ },
1442
+ {
1443
+ "epoch": 2.2,
1444
+ "learning_rate": 0.00012376707604418792,
1445
+ "loss": 3.4155,
1446
+ "step": 1185
1447
+ },
1448
+ {
1449
+ "epoch": 2.21,
1450
+ "learning_rate": 0.00012255324376199802,
1451
+ "loss": 3.2321,
1452
+ "step": 1190
1453
+ },
1454
+ {
1455
+ "epoch": 2.22,
1456
+ "learning_rate": 0.0001212934216384791,
1457
+ "loss": 3.3063,
1458
+ "step": 1195
1459
+ },
1460
+ {
1461
+ "epoch": 2.23,
1462
+ "learning_rate": 0.00011998868354825889,
1463
+ "loss": 3.3722,
1464
+ "step": 1200
1465
+ },
1466
+ {
1467
+ "epoch": 2.24,
1468
+ "learning_rate": 0.0001186401416524155,
1469
+ "loss": 3.0143,
1470
+ "step": 1205
1471
+ },
1472
+ {
1473
+ "epoch": 2.25,
1474
+ "learning_rate": 0.000117248945450469,
1475
+ "loss": 3.0218,
1476
+ "step": 1210
1477
+ },
1478
+ {
1479
+ "epoch": 2.26,
1480
+ "learning_rate": 0.00011581628080054626,
1481
+ "loss": 3.0963,
1482
+ "step": 1215
1483
+ },
1484
+ {
1485
+ "epoch": 2.27,
1486
+ "learning_rate": 0.00011434336890855335,
1487
+ "loss": 3.2636,
1488
+ "step": 1220
1489
+ },
1490
+ {
1491
+ "epoch": 2.28,
1492
+ "learning_rate": 0.00011283146528721704,
1493
+ "loss": 3.147,
1494
+ "step": 1225
1495
+ },
1496
+ {
1497
+ "epoch": 2.29,
1498
+ "learning_rate": 0.00011128185868588423,
1499
+ "loss": 3.1549,
1500
+ "step": 1230
1501
+ },
1502
+ {
1503
+ "epoch": 2.3,
1504
+ "learning_rate": 0.0001096958699919889,
1505
+ "loss": 3.2418,
1506
+ "step": 1235
1507
+ },
1508
+ {
1509
+ "epoch": 2.3,
1510
+ "learning_rate": 0.00010807485110512604,
1511
+ "loss": 3.3284,
1512
+ "step": 1240
1513
+ },
1514
+ {
1515
+ "epoch": 2.31,
1516
+ "learning_rate": 0.00010642018378468939,
1517
+ "loss": 2.9988,
1518
+ "step": 1245
1519
+ },
1520
+ {
1521
+ "epoch": 2.32,
1522
+ "learning_rate": 0.00010473327847205782,
1523
+ "loss": 3.2944,
1524
+ "step": 1250
1525
+ },
1526
+ {
1527
+ "epoch": 2.33,
1528
+ "learning_rate": 0.00010301557308833204,
1529
+ "loss": 3.3507,
1530
+ "step": 1255
1531
+ },
1532
+ {
1533
+ "epoch": 2.34,
1534
+ "learning_rate": 0.00010126853180864858,
1535
+ "loss": 3.1687,
1536
+ "step": 1260
1537
+ },
1538
+ {
1539
+ "epoch": 2.35,
1540
+ "learning_rate": 9.949364381411411e-05,
1541
+ "loss": 3.0591,
1542
+ "step": 1265
1543
+ },
1544
+ {
1545
+ "epoch": 2.36,
1546
+ "learning_rate": 9.769242202242448e-05,
1547
+ "loss": 3.1366,
1548
+ "step": 1270
1549
+ },
1550
+ {
1551
+ "epoch": 2.37,
1552
+ "learning_rate": 9.586640179825159e-05,
1553
+ "loss": 3.1332,
1554
+ "step": 1275
1555
+ },
1556
+ {
1557
+ "epoch": 2.38,
1558
+ "learning_rate": 9.401713964449514e-05,
1559
+ "loss": 3.17,
1560
+ "step": 1280
1561
+ },
1562
+ {
1563
+ "epoch": 2.39,
1564
+ "learning_rate": 9.21462118755175e-05,
1565
+ "loss": 3.2294,
1566
+ "step": 1285
1567
+ },
1568
+ {
1569
+ "epoch": 2.4,
1570
+ "learning_rate": 9.025521327348971e-05,
1571
+ "loss": 2.9817,
1572
+ "step": 1290
1573
+ },
1574
+ {
1575
+ "epoch": 2.41,
1576
+ "learning_rate": 8.834575572899717e-05,
1577
+ "loss": 3.1762,
1578
+ "step": 1295
1579
+ },
1580
+ {
1581
+ "epoch": 2.42,
1582
+ "learning_rate": 8.641946686706039e-05,
1583
+ "loss": 3.3385,
1584
+ "step": 1300
1585
+ },
1586
+ {
1587
+ "epoch": 2.43,
1588
+ "learning_rate": 8.447798865974526e-05,
1589
+ "loss": 3.0338,
1590
+ "step": 1305
1591
+ },
1592
+ {
1593
+ "epoch": 2.43,
1594
+ "learning_rate": 8.252297602654276e-05,
1595
+ "loss": 3.1138,
1596
+ "step": 1310
1597
+ },
1598
+ {
1599
+ "epoch": 2.44,
1600
+ "learning_rate": 8.055609542371231e-05,
1601
+ "loss": 3.3565,
1602
+ "step": 1315
1603
+ },
1604
+ {
1605
+ "epoch": 2.45,
1606
+ "learning_rate": 7.857902342379239e-05,
1607
+ "loss": 3.3677,
1608
+ "step": 1320
1609
+ },
1610
+ {
1611
+ "epoch": 2.46,
1612
+ "learning_rate": 7.659344528648596e-05,
1613
+ "loss": 3.0411,
1614
+ "step": 1325
1615
+ },
1616
+ {
1617
+ "epoch": 2.47,
1618
+ "learning_rate": 7.46010535221433e-05,
1619
+ "loss": 3.1615,
1620
+ "step": 1330
1621
+ },
1622
+ {
1623
+ "epoch": 2.48,
1624
+ "learning_rate": 7.26035464490619e-05,
1625
+ "loss": 3.3576,
1626
+ "step": 1335
1627
+ },
1628
+ {
1629
+ "epoch": 2.49,
1630
+ "learning_rate": 7.060262674583818e-05,
1631
+ "loss": 3.1091,
1632
+ "step": 1340
1633
+ },
1634
+ {
1635
+ "epoch": 2.5,
1636
+ "learning_rate": 6.860000000000001e-05,
1637
+ "loss": 3.4431,
1638
+ "step": 1345
1639
+ },
1640
+ {
1641
+ "epoch": 2.51,
1642
+ "learning_rate": 6.659737325416186e-05,
1643
+ "loss": 3.2144,
1644
+ "step": 1350
1645
+ },
1646
+ {
1647
+ "epoch": 2.52,
1648
+ "learning_rate": 6.459645355093813e-05,
1649
+ "loss": 3.1546,
1650
+ "step": 1355
1651
+ },
1652
+ {
1653
+ "epoch": 2.53,
1654
+ "learning_rate": 6.259894647785675e-05,
1655
+ "loss": 2.963,
1656
+ "step": 1360
1657
+ },
1658
+ {
1659
+ "epoch": 2.54,
1660
+ "learning_rate": 6.060655471351407e-05,
1661
+ "loss": 3.1648,
1662
+ "step": 1365
1663
+ },
1664
+ {
1665
+ "epoch": 2.55,
1666
+ "learning_rate": 5.862097657620766e-05,
1667
+ "loss": 3.1867,
1668
+ "step": 1370
1669
+ },
1670
+ {
1671
+ "epoch": 2.56,
1672
+ "learning_rate": 5.664390457628773e-05,
1673
+ "loss": 3.11,
1674
+ "step": 1375
1675
+ },
1676
+ {
1677
+ "epoch": 2.57,
1678
+ "learning_rate": 5.467702397345729e-05,
1679
+ "loss": 3.0585,
1680
+ "step": 1380
1681
+ },
1682
+ {
1683
+ "epoch": 2.57,
1684
+ "learning_rate": 5.272201134025478e-05,
1685
+ "loss": 3.0417,
1686
+ "step": 1385
1687
+ },
1688
+ {
1689
+ "epoch": 2.58,
1690
+ "learning_rate": 5.078053313293964e-05,
1691
+ "loss": 3.078,
1692
+ "step": 1390
1693
+ },
1694
+ {
1695
+ "epoch": 2.59,
1696
+ "learning_rate": 4.885424427100287e-05,
1697
+ "loss": 3.2723,
1698
+ "step": 1395
1699
+ },
1700
+ {
1701
+ "epoch": 2.6,
1702
+ "learning_rate": 4.694478672651034e-05,
1703
+ "loss": 3.0448,
1704
+ "step": 1400
1705
+ },
1706
+ {
1707
+ "epoch": 2.61,
1708
+ "learning_rate": 4.505378812448254e-05,
1709
+ "loss": 3.1823,
1710
+ "step": 1405
1711
+ },
1712
+ {
1713
+ "epoch": 2.62,
1714
+ "learning_rate": 4.31828603555049e-05,
1715
+ "loss": 3.1987,
1716
+ "step": 1410
1717
+ },
1718
+ {
1719
+ "epoch": 2.63,
1720
+ "learning_rate": 4.133359820174844e-05,
1721
+ "loss": 3.1579,
1722
+ "step": 1415
1723
+ },
1724
+ {
1725
+ "epoch": 2.64,
1726
+ "learning_rate": 3.950757797757557e-05,
1727
+ "loss": 3.0757,
1728
+ "step": 1420
1729
+ },
1730
+ {
1731
+ "epoch": 2.65,
1732
+ "learning_rate": 3.7706356185885915e-05,
1733
+ "loss": 3.1036,
1734
+ "step": 1425
1735
+ },
1736
+ {
1737
+ "epoch": 2.66,
1738
+ "learning_rate": 3.593146819135146e-05,
1739
+ "loss": 3.0992,
1740
+ "step": 1430
1741
+ },
1742
+ {
1743
+ "epoch": 2.67,
1744
+ "learning_rate": 3.418442691166801e-05,
1745
+ "loss": 3.1336,
1746
+ "step": 1435
1747
+ },
1748
+ {
1749
+ "epoch": 2.68,
1750
+ "learning_rate": 3.246672152794222e-05,
1751
+ "loss": 3.1688,
1752
+ "step": 1440
1753
+ },
1754
+ {
1755
+ "epoch": 2.69,
1756
+ "learning_rate": 3.077981621531067e-05,
1757
+ "loss": 2.9663,
1758
+ "step": 1445
1759
+ },
1760
+ {
1761
+ "epoch": 2.7,
1762
+ "learning_rate": 2.9125148894873993e-05,
1763
+ "loss": 3.057,
1764
+ "step": 1450
1765
+ },
1766
+ {
1767
+ "epoch": 2.7,
1768
+ "learning_rate": 2.750413000801114e-05,
1769
+ "loss": 3.1822,
1770
+ "step": 1455
1771
+ },
1772
+ {
1773
+ "epoch": 2.71,
1774
+ "learning_rate": 2.5918141314115796e-05,
1775
+ "loss": 2.9519,
1776
+ "step": 1460
1777
+ },
1778
+ {
1779
+ "epoch": 2.72,
1780
+ "learning_rate": 2.436853471278299e-05,
1781
+ "loss": 3.0403,
1782
+ "step": 1465
1783
+ },
1784
+ {
1785
+ "epoch": 2.73,
1786
+ "learning_rate": 2.2856631091446682e-05,
1787
+ "loss": 3.0835,
1788
+ "step": 1470
1789
+ },
1790
+ {
1791
+ "epoch": 2.74,
1792
+ "learning_rate": 2.138371919945377e-05,
1793
+ "loss": 3.0139,
1794
+ "step": 1475
1795
+ },
1796
+ {
1797
+ "epoch": 2.75,
1798
+ "learning_rate": 1.9951054549531022e-05,
1799
+ "loss": 3.0177,
1800
+ "step": 1480
1801
+ },
1802
+ {
1803
+ "epoch": 2.76,
1804
+ "learning_rate": 1.85598583475845e-05,
1805
+ "loss": 3.3141,
1806
+ "step": 1485
1807
+ },
1808
+ {
1809
+ "epoch": 2.77,
1810
+ "learning_rate": 1.7211316451741122e-05,
1811
+ "loss": 3.1076,
1812
+ "step": 1490
1813
+ },
1814
+ {
1815
+ "epoch": 2.78,
1816
+ "learning_rate": 1.59065783615209e-05,
1817
+ "loss": 3.1387,
1818
+ "step": 1495
1819
+ },
1820
+ {
1821
+ "epoch": 2.79,
1822
+ "learning_rate": 1.4646756238001999e-05,
1823
+ "loss": 3.1493,
1824
+ "step": 1500
1825
+ },
1826
+ {
1827
+ "epoch": 2.8,
1828
+ "learning_rate": 1.3432923955812079e-05,
1829
+ "loss": 3.2037,
1830
+ "step": 1505
1831
+ },
1832
+ {
1833
+ "epoch": 2.81,
1834
+ "learning_rate": 1.22661161877562e-05,
1835
+ "loss": 3.033,
1836
+ "step": 1510
1837
+ },
1838
+ {
1839
+ "epoch": 2.82,
1840
+ "learning_rate": 1.114732752285933e-05,
1841
+ "loss": 3.1567,
1842
+ "step": 1515
1843
+ },
1844
+ {
1845
+ "epoch": 2.83,
1846
+ "learning_rate": 1.0077511618577358e-05,
1847
+ "loss": 2.9584,
1848
+ "step": 1520
1849
+ },
1850
+ {
1851
+ "epoch": 2.83,
1852
+ "learning_rate": 9.057580387897574e-06,
1853
+ "loss": 3.1304,
1854
+ "step": 1525
1855
+ },
1856
+ {
1857
+ "epoch": 2.84,
1858
+ "learning_rate": 8.088403222022203e-06,
1859
+ "loss": 3.0946,
1860
+ "step": 1530
1861
+ },
1862
+ {
1863
+ "epoch": 2.85,
1864
+ "learning_rate": 7.1708062492982514e-06,
1865
+ "loss": 3.1109,
1866
+ "step": 1535
1867
+ },
1868
+ {
1869
+ "epoch": 2.86,
1870
+ "learning_rate": 6.3055716310237545e-06,
1871
+ "loss": 3.1375,
1872
+ "step": 1540
1873
+ },
1874
+ {
1875
+ "epoch": 2.87,
1876
+ "learning_rate": 5.493436894732511e-06,
1877
+ "loss": 3.1381,
1878
+ "step": 1545
1879
+ },
1880
+ {
1881
+ "epoch": 2.88,
1882
+ "learning_rate": 4.735094305523896e-06,
1883
+ "loss": 3.1063,
1884
+ "step": 1550
1885
+ },
1886
+ {
1887
+ "epoch": 2.89,
1888
+ "learning_rate": 4.031190275975246e-06,
1889
+ "loss": 3.1137,
1890
+ "step": 1555
1891
+ },
1892
+ {
1893
+ "epoch": 2.9,
1894
+ "learning_rate": 3.3823248151382478e-06,
1895
+ "loss": 3.1414,
1896
+ "step": 1560
1897
+ },
1898
+ {
1899
+ "epoch": 2.91,
1900
+ "learning_rate": 2.789051017090463e-06,
1901
+ "loss": 3.2499,
1902
+ "step": 1565
1903
+ },
1904
+ {
1905
+ "epoch": 2.92,
1906
+ "learning_rate": 2.251874589476814e-06,
1907
+ "loss": 3.0676,
1908
+ "step": 1570
1909
+ },
1910
+ {
1911
+ "epoch": 2.93,
1912
+ "learning_rate": 1.7712534224433918e-06,
1913
+ "loss": 3.0984,
1914
+ "step": 1575
1915
+ },
1916
+ {
1917
+ "epoch": 2.94,
1918
+ "learning_rate": 1.3475971983312994e-06,
1919
+ "loss": 3.2583,
1920
+ "step": 1580
1921
+ },
1922
+ {
1923
+ "epoch": 2.95,
1924
+ "learning_rate": 9.812670424624475e-07,
1925
+ "loss": 3.1486,
1926
+ "step": 1585
1927
+ },
1928
+ {
1929
+ "epoch": 2.96,
1930
+ "learning_rate": 6.725752153159165e-07,
1931
+ "loss": 3.1602,
1932
+ "step": 1590
1933
+ },
1934
+ {
1935
+ "epoch": 2.96,
1936
+ "learning_rate": 4.217848463563714e-07,
1937
+ "loss": 3.0449,
1938
+ "step": 1595
1939
+ },
1940
+ {
1941
+ "epoch": 2.97,
1942
+ "learning_rate": 2.291097097421942e-07,
1943
+ "loss": 3.1203,
1944
+ "step": 1600
1945
+ },
1946
+ {
1947
+ "epoch": 2.98,
1948
+ "learning_rate": 9.471404210387107e-08,
1949
+ "loss": 3.1678,
1950
+ "step": 1605
1951
+ },
1952
+ {
1953
+ "epoch": 2.99,
1954
+ "learning_rate": 1.871240254851856e-08,
1955
+ "loss": 3.3214,
1956
+ "step": 1610
1957
+ },
1958
+ {
1959
+ "epoch": 3.0,
1960
+ "eval_loss": 3.1907694339752197,
1961
+ "eval_runtime": 18.9242,
1962
+ "eval_samples_per_second": 39.79,
1963
+ "eval_steps_per_second": 5.02,
1964
+ "step": 1614
1965
  }
1966
  ],
1967
+ "max_steps": 1614,
1968
+ "num_train_epochs": 3,
1969
+ "total_flos": 1685464252416000.0,
1970
  "trial_name": null,
1971
  "trial_params": null
1972
  }
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:921fbeeaca9d2d7fa8d5c81ad9c59179822d44721ad18de326d17b3f14cc0e37
3
- size 2671
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ae886514c23914b629ae7350b2b20693c40738f5357f3e721515ec87c4a311f3
3
+ size 3055