Hugging Face's logo

Group11AIProject
/
Ambiguity_Detector

TensorBoard
Safetensors
Model card Files
xet
 Metrics Community
mcclellanc commited on
Commit
16702cf
·
verified ·
1 Parent(s): 2652329

Upload 42 files

Browse files
Files changed (36) hide show
  1. fine_tuned_model/checkpoint-4054/config.json +40 -0
  2. fine_tuned_model/checkpoint-4054/generation_config.json +6 -0
  3. fine_tuned_model/checkpoint-4054/merges.txt +0 -0
  4. fine_tuned_model/checkpoint-4054/model.safetensors +3 -0
  5. fine_tuned_model/checkpoint-4054/optimizer.pt +3 -0
  6. fine_tuned_model/checkpoint-4054/rng_state.pth +3 -0
  7. fine_tuned_model/checkpoint-4054/scaler.pt +3 -0
  8. fine_tuned_model/checkpoint-4054/scheduler.pt +3 -0
  9. fine_tuned_model/checkpoint-4054/special_tokens_map.json +6 -0
  10. fine_tuned_model/checkpoint-4054/tokenizer.json +0 -0
  11. fine_tuned_model/checkpoint-4054/tokenizer_config.json +21 -0
  12. fine_tuned_model/checkpoint-4054/trainer_state.json +2869 -0
  13. fine_tuned_model/checkpoint-4054/training_args.bin +3 -0
  14. fine_tuned_model/checkpoint-4054/vocab.json +0 -0
  15. fine_tuned_model/checkpoint-6081/config.json +40 -0
  16. fine_tuned_model/checkpoint-6081/generation_config.json +6 -0
  17. fine_tuned_model/checkpoint-6081/merges.txt +0 -0
  18. fine_tuned_model/checkpoint-6081/model.safetensors +3 -0
  19. fine_tuned_model/checkpoint-6081/optimizer.pt +3 -0
  20. fine_tuned_model/checkpoint-6081/rng_state.pth +3 -0
  21. fine_tuned_model/checkpoint-6081/scaler.pt +3 -0
  22. fine_tuned_model/checkpoint-6081/scheduler.pt +3 -0
  23. fine_tuned_model/checkpoint-6081/special_tokens_map.json +6 -0
  24. fine_tuned_model/checkpoint-6081/tokenizer.json +0 -0
  25. fine_tuned_model/checkpoint-6081/tokenizer_config.json +21 -0
  26. fine_tuned_model/checkpoint-6081/trainer_state.json +0 -0
  27. fine_tuned_model/checkpoint-6081/training_args.bin +3 -0
  28. fine_tuned_model/checkpoint-6081/vocab.json +0 -0
  29. fine_tuned_model/config.json +5 -3
  30. fine_tuned_model/logs/events.out.tfevents.1746766596.BallTwister.15212.0 +3 -0
  31. fine_tuned_model/logs/events.out.tfevents.1746769125.BallTwister.19464.0 +3 -0
  32. fine_tuned_model/logs/events.out.tfevents.1746769341.BallTwister.19392.0 +3 -0
  33. fine_tuned_model/logs/events.out.tfevents.1746769713.BallTwister.15164.0 +3 -0
  34. fine_tuned_model/logs/events.out.tfevents.1746769988.BallTwister.17104.0 +3 -0
  35. fine_tuned_model/model.safetensors +2 -2
  36. fine_tuned_model/training_args.bin +1 -1
fine_tuned_model/checkpoint-4054/config.json ADDED
@@ -0,0 +1,40 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "activation_function": "gelu_new",
3
+ "architectures": [
4
+ "GPT2LMHeadModel"
5
+ ],
6
+ "attn_pdrop": 0.1,
7
+ "bos_token_id": 50256,
8
+ "embd_pdrop": 0.1,
9
+ "eos_token_id": 50256,
10
+ "initializer_range": 0.02,
11
+ "layer_norm_epsilon": 1e-05,
12
+ "model_type": "gpt2",
13
+ "n_ctx": 1024,
14
+ "n_embd": 1024,
15
+ "n_head": 16,
16
+ "n_inner": null,
17
+ "n_layer": 24,
18
+ "n_positions": 1024,
19
+ "n_special": 0,
20
+ "predict_special_tokens": true,
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,
27
+ "summary_proj_to_labels": true,
28
+ "summary_type": "cls_index",
29
+ "summary_use_proj": true,
30
+ "task_specific_params": {
31
+ "text-generation": {
32
+ "do_sample": true,
33
+ "max_length": 50
34
+ }
35
+ },
36
+ "torch_dtype": "float32",
37
+ "transformers_version": "4.51.3",
38
+ "use_cache": true,
39
+ "vocab_size": 50257
40
+ }
fine_tuned_model/checkpoint-4054/generation_config.json ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 50256,
4
+ "eos_token_id": 50256,
5
+ "transformers_version": "4.51.3"
6
+ }
fine_tuned_model/checkpoint-4054/merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
fine_tuned_model/checkpoint-4054/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0881d33f81ec2c86d9d3839891e7a8513f3b41167f1a2c4d0f1408e065cfe62e
3
+ size 1419322880
fine_tuned_model/checkpoint-4054/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:558ee6d185645a2b3016c52b7c904cc98979c5dc51910c4ce33bb81928ffce72
3
+ size 2838829707
fine_tuned_model/checkpoint-4054/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6afd3fc0aaa45e5cdf823f9c4552b1b5a876eb1ddef54daea0684a7ff54ad946
3
+ size 14645
fine_tuned_model/checkpoint-4054/scaler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2ae053a0683285f4687930f4055756a14a4ab293e05251bb93015ab25e8075df
3
+ size 1383
fine_tuned_model/checkpoint-4054/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cedb1f457bb169fb83bb1e6a6a58d3bd862c65c445ba3de1707e76328735d56f
3
+ size 1465
fine_tuned_model/checkpoint-4054/special_tokens_map.json ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "<|endoftext|>",
3
+ "eos_token": "<|endoftext|>",
4
+ "pad_token": "<|endoftext|>",
5
+ "unk_token": "<|endoftext|>"
6
+ }
fine_tuned_model/checkpoint-4054/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
fine_tuned_model/checkpoint-4054/tokenizer_config.json ADDED
@@ -0,0 +1,21 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": false,
3
+ "added_tokens_decoder": {
4
+ "50256": {
5
+ "content": "<|endoftext|>",
6
+ "lstrip": false,
7
+ "normalized": true,
8
+ "rstrip": false,
9
+ "single_word": false,
10
+ "special": true
11
+ }
12
+ },
13
+ "bos_token": "<|endoftext|>",
14
+ "clean_up_tokenization_spaces": false,
15
+ "eos_token": "<|endoftext|>",
16
+ "extra_special_tokens": {},
17
+ "model_max_length": 1024,
18
+ "pad_token": "<|endoftext|>",
19
+ "tokenizer_class": "GPT2Tokenizer",
20
+ "unk_token": "<|endoftext|>"
21
+ }
fine_tuned_model/checkpoint-4054/trainer_state.json ADDED
@@ -0,0 +1,2869 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 2.0,
6
+ "eval_steps": 500,
7
+ "global_step": 4054,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.00493339911198816,
14
+ "grad_norm": 32.17447280883789,
15
+ "learning_rate": 2.5e-05,
16
+ "loss": 4.0607,
17
+ "step": 10
18
+ },
19
+ {
20
+ "epoch": 0.00986679822397632,
21
+ "grad_norm": 30.161712646484375,
22
+ "learning_rate": 4.995882062263219e-05,
23
+ "loss": 3.7481,
24
+ "step": 20
25
+ },
26
+ {
27
+ "epoch": 0.01480019733596448,
28
+ "grad_norm": 38.9721565246582,
29
+ "learning_rate": 4.9884697743370127e-05,
30
+ "loss": 3.9839,
31
+ "step": 30
32
+ },
33
+ {
34
+ "epoch": 0.01973359644795264,
35
+ "grad_norm": 29.38846778869629,
36
+ "learning_rate": 4.9802338988634494e-05,
37
+ "loss": 3.7565,
38
+ "step": 40
39
+ },
40
+ {
41
+ "epoch": 0.0246669955599408,
42
+ "grad_norm": 32.933860778808594,
43
+ "learning_rate": 4.9728216109372424e-05,
44
+ "loss": 3.7562,
45
+ "step": 50
46
+ },
47
+ {
48
+ "epoch": 0.02960039467192896,
49
+ "grad_norm": 24.409399032592773,
50
+ "learning_rate": 4.96458573546368e-05,
51
+ "loss": 3.3008,
52
+ "step": 60
53
+ },
54
+ {
55
+ "epoch": 0.03453379378391712,
56
+ "grad_norm": 25.322418212890625,
57
+ "learning_rate": 4.956349859990117e-05,
58
+ "loss": 3.0558,
59
+ "step": 70
60
+ },
61
+ {
62
+ "epoch": 0.03946719289590528,
63
+ "grad_norm": 28.60622215270996,
64
+ "learning_rate": 4.948113984516554e-05,
65
+ "loss": 3.4003,
66
+ "step": 80
67
+ },
68
+ {
69
+ "epoch": 0.04440059200789344,
70
+ "grad_norm": 15.609489440917969,
71
+ "learning_rate": 4.9398781090429915e-05,
72
+ "loss": 3.0121,
73
+ "step": 90
74
+ },
75
+ {
76
+ "epoch": 0.0493339911198816,
77
+ "grad_norm": 22.34126091003418,
78
+ "learning_rate": 4.931642233569428e-05,
79
+ "loss": 3.1712,
80
+ "step": 100
81
+ },
82
+ {
83
+ "epoch": 0.05426739023186976,
84
+ "grad_norm": 19.47034454345703,
85
+ "learning_rate": 4.923406358095866e-05,
86
+ "loss": 3.2003,
87
+ "step": 110
88
+ },
89
+ {
90
+ "epoch": 0.05920078934385792,
91
+ "grad_norm": 20.92117691040039,
92
+ "learning_rate": 4.9151704826223025e-05,
93
+ "loss": 3.2641,
94
+ "step": 120
95
+ },
96
+ {
97
+ "epoch": 0.06413418845584608,
98
+ "grad_norm": 10.978850364685059,
99
+ "learning_rate": 4.90693460714874e-05,
100
+ "loss": 3.0391,
101
+ "step": 130
102
+ },
103
+ {
104
+ "epoch": 0.06906758756783424,
105
+ "grad_norm": 12.89031982421875,
106
+ "learning_rate": 4.8986987316751774e-05,
107
+ "loss": 2.7204,
108
+ "step": 140
109
+ },
110
+ {
111
+ "epoch": 0.0740009866798224,
112
+ "grad_norm": 9.097904205322266,
113
+ "learning_rate": 4.890462856201615e-05,
114
+ "loss": 2.9097,
115
+ "step": 150
116
+ },
117
+ {
118
+ "epoch": 0.07893438579181056,
119
+ "grad_norm": 10.250645637512207,
120
+ "learning_rate": 4.8822269807280516e-05,
121
+ "loss": 2.8115,
122
+ "step": 160
123
+ },
124
+ {
125
+ "epoch": 0.08386778490379872,
126
+ "grad_norm": 9.343478202819824,
127
+ "learning_rate": 4.873991105254489e-05,
128
+ "loss": 3.2778,
129
+ "step": 170
130
+ },
131
+ {
132
+ "epoch": 0.08880118401578688,
133
+ "grad_norm": 10.233019828796387,
134
+ "learning_rate": 4.8657552297809264e-05,
135
+ "loss": 2.8652,
136
+ "step": 180
137
+ },
138
+ {
139
+ "epoch": 0.09373458312777504,
140
+ "grad_norm": 9.129461288452148,
141
+ "learning_rate": 4.857519354307363e-05,
142
+ "loss": 2.7035,
143
+ "step": 190
144
+ },
145
+ {
146
+ "epoch": 0.0986679822397632,
147
+ "grad_norm": 6.655858039855957,
148
+ "learning_rate": 4.8492834788338006e-05,
149
+ "loss": 2.8944,
150
+ "step": 200
151
+ },
152
+ {
153
+ "epoch": 0.10360138135175136,
154
+ "grad_norm": 8.336820602416992,
155
+ "learning_rate": 4.8410476033602374e-05,
156
+ "loss": 2.4224,
157
+ "step": 210
158
+ },
159
+ {
160
+ "epoch": 0.10853478046373952,
161
+ "grad_norm": 16.6269474029541,
162
+ "learning_rate": 4.832811727886675e-05,
163
+ "loss": 2.9059,
164
+ "step": 220
165
+ },
166
+ {
167
+ "epoch": 0.11346817957572768,
168
+ "grad_norm": 14.81032657623291,
169
+ "learning_rate": 4.8245758524131116e-05,
170
+ "loss": 3.026,
171
+ "step": 230
172
+ },
173
+ {
174
+ "epoch": 0.11840157868771584,
175
+ "grad_norm": 5.8194193840026855,
176
+ "learning_rate": 4.816339976939549e-05,
177
+ "loss": 2.6034,
178
+ "step": 240
179
+ },
180
+ {
181
+ "epoch": 0.123334977799704,
182
+ "grad_norm": 12.556567192077637,
183
+ "learning_rate": 4.808104101465986e-05,
184
+ "loss": 2.9635,
185
+ "step": 250
186
+ },
187
+ {
188
+ "epoch": 0.12826837691169216,
189
+ "grad_norm": 11.739566802978516,
190
+ "learning_rate": 4.799868225992423e-05,
191
+ "loss": 3.1042,
192
+ "step": 260
193
+ },
194
+ {
195
+ "epoch": 0.13320177602368033,
196
+ "grad_norm": 9.672442436218262,
197
+ "learning_rate": 4.791632350518861e-05,
198
+ "loss": 3.0241,
199
+ "step": 270
200
+ },
201
+ {
202
+ "epoch": 0.13813517513566848,
203
+ "grad_norm": 5.8154497146606445,
204
+ "learning_rate": 4.7833964750452975e-05,
205
+ "loss": 2.7709,
206
+ "step": 280
207
+ },
208
+ {
209
+ "epoch": 0.14306857424765665,
210
+ "grad_norm": 20.954362869262695,
211
+ "learning_rate": 4.775160599571735e-05,
212
+ "loss": 2.8549,
213
+ "step": 290
214
+ },
215
+ {
216
+ "epoch": 0.1480019733596448,
217
+ "grad_norm": 9.915385246276855,
218
+ "learning_rate": 4.7669247240981717e-05,
219
+ "loss": 2.9146,
220
+ "step": 300
221
+ },
222
+ {
223
+ "epoch": 0.15293537247163297,
224
+ "grad_norm": 7.093989372253418,
225
+ "learning_rate": 4.758688848624609e-05,
226
+ "loss": 2.6771,
227
+ "step": 310
228
+ },
229
+ {
230
+ "epoch": 0.15786877158362111,
231
+ "grad_norm": 8.882306098937988,
232
+ "learning_rate": 4.750452973151046e-05,
233
+ "loss": 2.6295,
234
+ "step": 320
235
+ },
236
+ {
237
+ "epoch": 0.1628021706956093,
238
+ "grad_norm": 17.28571128845215,
239
+ "learning_rate": 4.742217097677483e-05,
240
+ "loss": 2.492,
241
+ "step": 330
242
+ },
243
+ {
244
+ "epoch": 0.16773556980759743,
245
+ "grad_norm": 18.01361846923828,
246
+ "learning_rate": 4.73398122220392e-05,
247
+ "loss": 3.288,
248
+ "step": 340
249
+ },
250
+ {
251
+ "epoch": 0.1726689689195856,
252
+ "grad_norm": 10.41086196899414,
253
+ "learning_rate": 4.7257453467303575e-05,
254
+ "loss": 2.6676,
255
+ "step": 350
256
+ },
257
+ {
258
+ "epoch": 0.17760236803157375,
259
+ "grad_norm": 12.40413761138916,
260
+ "learning_rate": 4.717509471256795e-05,
261
+ "loss": 2.8085,
262
+ "step": 360
263
+ },
264
+ {
265
+ "epoch": 0.18253576714356193,
266
+ "grad_norm": 13.229728698730469,
267
+ "learning_rate": 4.709273595783232e-05,
268
+ "loss": 2.2978,
269
+ "step": 370
270
+ },
271
+ {
272
+ "epoch": 0.18746916625555007,
273
+ "grad_norm": 7.240847587585449,
274
+ "learning_rate": 4.701037720309669e-05,
275
+ "loss": 2.7408,
276
+ "step": 380
277
+ },
278
+ {
279
+ "epoch": 0.19240256536753825,
280
+ "grad_norm": 10.95034122467041,
281
+ "learning_rate": 4.692801844836106e-05,
282
+ "loss": 2.7323,
283
+ "step": 390
284
+ },
285
+ {
286
+ "epoch": 0.1973359644795264,
287
+ "grad_norm": 11.429134368896484,
288
+ "learning_rate": 4.6845659693625434e-05,
289
+ "loss": 2.6121,
290
+ "step": 400
291
+ },
292
+ {
293
+ "epoch": 0.20226936359151457,
294
+ "grad_norm": 14.395548820495605,
295
+ "learning_rate": 4.67633009388898e-05,
296
+ "loss": 3.0402,
297
+ "step": 410
298
+ },
299
+ {
300
+ "epoch": 0.2072027627035027,
301
+ "grad_norm": 5.770746231079102,
302
+ "learning_rate": 4.668094218415418e-05,
303
+ "loss": 2.4771,
304
+ "step": 420
305
+ },
306
+ {
307
+ "epoch": 0.2121361618154909,
308
+ "grad_norm": 8.25928783416748,
309
+ "learning_rate": 4.659858342941855e-05,
310
+ "loss": 2.1011,
311
+ "step": 430
312
+ },
313
+ {
314
+ "epoch": 0.21706956092747903,
315
+ "grad_norm": 14.545477867126465,
316
+ "learning_rate": 4.6516224674682924e-05,
317
+ "loss": 3.0616,
318
+ "step": 440
319
+ },
320
+ {
321
+ "epoch": 0.2220029600394672,
322
+ "grad_norm": 11.545022964477539,
323
+ "learning_rate": 4.643386591994729e-05,
324
+ "loss": 2.2621,
325
+ "step": 450
326
+ },
327
+ {
328
+ "epoch": 0.22693635915145535,
329
+ "grad_norm": 25.751686096191406,
330
+ "learning_rate": 4.6351507165211666e-05,
331
+ "loss": 2.7998,
332
+ "step": 460
333
+ },
334
+ {
335
+ "epoch": 0.23186975826344353,
336
+ "grad_norm": 8.96506118774414,
337
+ "learning_rate": 4.626914841047604e-05,
338
+ "loss": 2.2543,
339
+ "step": 470
340
+ },
341
+ {
342
+ "epoch": 0.23680315737543167,
343
+ "grad_norm": 8.918927192687988,
344
+ "learning_rate": 4.618678965574041e-05,
345
+ "loss": 2.2573,
346
+ "step": 480
347
+ },
348
+ {
349
+ "epoch": 0.24173655648741985,
350
+ "grad_norm": 8.464729309082031,
351
+ "learning_rate": 4.610443090100478e-05,
352
+ "loss": 2.7079,
353
+ "step": 490
354
+ },
355
+ {
356
+ "epoch": 0.246669955599408,
357
+ "grad_norm": 9.733293533325195,
358
+ "learning_rate": 4.602207214626915e-05,
359
+ "loss": 3.423,
360
+ "step": 500
361
+ },
362
+ {
363
+ "epoch": 0.25160335471139617,
364
+ "grad_norm": 11.454011917114258,
365
+ "learning_rate": 4.5939713391533525e-05,
366
+ "loss": 3.1205,
367
+ "step": 510
368
+ },
369
+ {
370
+ "epoch": 0.2565367538233843,
371
+ "grad_norm": 8.79787826538086,
372
+ "learning_rate": 4.585735463679789e-05,
373
+ "loss": 2.4529,
374
+ "step": 520
375
+ },
376
+ {
377
+ "epoch": 0.26147015293537246,
378
+ "grad_norm": 6.300372123718262,
379
+ "learning_rate": 4.577499588206227e-05,
380
+ "loss": 2.3711,
381
+ "step": 530
382
+ },
383
+ {
384
+ "epoch": 0.26640355204736066,
385
+ "grad_norm": 5.140571117401123,
386
+ "learning_rate": 4.5692637127326634e-05,
387
+ "loss": 2.3833,
388
+ "step": 540
389
+ },
390
+ {
391
+ "epoch": 0.2713369511593488,
392
+ "grad_norm": 11.252778053283691,
393
+ "learning_rate": 4.561027837259101e-05,
394
+ "loss": 2.3281,
395
+ "step": 550
396
+ },
397
+ {
398
+ "epoch": 0.27627035027133695,
399
+ "grad_norm": 10.254585266113281,
400
+ "learning_rate": 4.552791961785538e-05,
401
+ "loss": 3.0383,
402
+ "step": 560
403
+ },
404
+ {
405
+ "epoch": 0.2812037493833251,
406
+ "grad_norm": 8.8512544631958,
407
+ "learning_rate": 4.544556086311975e-05,
408
+ "loss": 2.3813,
409
+ "step": 570
410
+ },
411
+ {
412
+ "epoch": 0.2861371484953133,
413
+ "grad_norm": 4.089067459106445,
414
+ "learning_rate": 4.5363202108384125e-05,
415
+ "loss": 2.5559,
416
+ "step": 580
417
+ },
418
+ {
419
+ "epoch": 0.29107054760730144,
420
+ "grad_norm": 5.320517539978027,
421
+ "learning_rate": 4.528084335364849e-05,
422
+ "loss": 2.7508,
423
+ "step": 590
424
+ },
425
+ {
426
+ "epoch": 0.2960039467192896,
427
+ "grad_norm": 6.5414862632751465,
428
+ "learning_rate": 4.519848459891287e-05,
429
+ "loss": 2.3911,
430
+ "step": 600
431
+ },
432
+ {
433
+ "epoch": 0.30093734583127774,
434
+ "grad_norm": 7.56878137588501,
435
+ "learning_rate": 4.5116125844177235e-05,
436
+ "loss": 2.9435,
437
+ "step": 610
438
+ },
439
+ {
440
+ "epoch": 0.30587074494326594,
441
+ "grad_norm": 10.586882591247559,
442
+ "learning_rate": 4.503376708944161e-05,
443
+ "loss": 2.6231,
444
+ "step": 620
445
+ },
446
+ {
447
+ "epoch": 0.3108041440552541,
448
+ "grad_norm": 9.455619812011719,
449
+ "learning_rate": 4.495140833470598e-05,
450
+ "loss": 2.8628,
451
+ "step": 630
452
+ },
453
+ {
454
+ "epoch": 0.31573754316724223,
455
+ "grad_norm": 5.5599565505981445,
456
+ "learning_rate": 4.486904957997035e-05,
457
+ "loss": 2.2629,
458
+ "step": 640
459
+ },
460
+ {
461
+ "epoch": 0.3206709422792304,
462
+ "grad_norm": 5.07106351852417,
463
+ "learning_rate": 4.4786690825234726e-05,
464
+ "loss": 2.1062,
465
+ "step": 650
466
+ },
467
+ {
468
+ "epoch": 0.3256043413912186,
469
+ "grad_norm": 8.023340225219727,
470
+ "learning_rate": 4.470433207049909e-05,
471
+ "loss": 2.4445,
472
+ "step": 660
473
+ },
474
+ {
475
+ "epoch": 0.3305377405032067,
476
+ "grad_norm": 12.941463470458984,
477
+ "learning_rate": 4.462197331576347e-05,
478
+ "loss": 2.3829,
479
+ "step": 670
480
+ },
481
+ {
482
+ "epoch": 0.33547113961519487,
483
+ "grad_norm": 9.589372634887695,
484
+ "learning_rate": 4.4539614561027835e-05,
485
+ "loss": 2.8525,
486
+ "step": 680
487
+ },
488
+ {
489
+ "epoch": 0.340404538727183,
490
+ "grad_norm": 29.511812210083008,
491
+ "learning_rate": 4.445725580629221e-05,
492
+ "loss": 2.4082,
493
+ "step": 690
494
+ },
495
+ {
496
+ "epoch": 0.3453379378391712,
497
+ "grad_norm": 17.187841415405273,
498
+ "learning_rate": 4.437489705155658e-05,
499
+ "loss": 2.3018,
500
+ "step": 700
501
+ },
502
+ {
503
+ "epoch": 0.35027133695115936,
504
+ "grad_norm": 6.561307430267334,
505
+ "learning_rate": 4.429253829682096e-05,
506
+ "loss": 2.4964,
507
+ "step": 710
508
+ },
509
+ {
510
+ "epoch": 0.3552047360631475,
511
+ "grad_norm": 8.075435638427734,
512
+ "learning_rate": 4.4210179542085326e-05,
513
+ "loss": 2.4871,
514
+ "step": 720
515
+ },
516
+ {
517
+ "epoch": 0.36013813517513565,
518
+ "grad_norm": 8.039571762084961,
519
+ "learning_rate": 4.41278207873497e-05,
520
+ "loss": 2.4563,
521
+ "step": 730
522
+ },
523
+ {
524
+ "epoch": 0.36507153428712386,
525
+ "grad_norm": 24.152629852294922,
526
+ "learning_rate": 4.404546203261407e-05,
527
+ "loss": 2.549,
528
+ "step": 740
529
+ },
530
+ {
531
+ "epoch": 0.370004933399112,
532
+ "grad_norm": 2.6132395267486572,
533
+ "learning_rate": 4.396310327787844e-05,
534
+ "loss": 2.6625,
535
+ "step": 750
536
+ },
537
+ {
538
+ "epoch": 0.37493833251110015,
539
+ "grad_norm": 11.909361839294434,
540
+ "learning_rate": 4.388074452314282e-05,
541
+ "loss": 3.0034,
542
+ "step": 760
543
+ },
544
+ {
545
+ "epoch": 0.3798717316230883,
546
+ "grad_norm": 6.0602240562438965,
547
+ "learning_rate": 4.3798385768407185e-05,
548
+ "loss": 2.8432,
549
+ "step": 770
550
+ },
551
+ {
552
+ "epoch": 0.3848051307350765,
553
+ "grad_norm": 6.063085079193115,
554
+ "learning_rate": 4.371602701367156e-05,
555
+ "loss": 2.4882,
556
+ "step": 780
557
+ },
558
+ {
559
+ "epoch": 0.38973852984706464,
560
+ "grad_norm": 12.514052391052246,
561
+ "learning_rate": 4.363366825893593e-05,
562
+ "loss": 2.5734,
563
+ "step": 790
564
+ },
565
+ {
566
+ "epoch": 0.3946719289590528,
567
+ "grad_norm": 5.626593589782715,
568
+ "learning_rate": 4.35513095042003e-05,
569
+ "loss": 1.8927,
570
+ "step": 800
571
+ },
572
+ {
573
+ "epoch": 0.39960532807104093,
574
+ "grad_norm": 6.073607444763184,
575
+ "learning_rate": 4.346895074946467e-05,
576
+ "loss": 2.4321,
577
+ "step": 810
578
+ },
579
+ {
580
+ "epoch": 0.40453872718302913,
581
+ "grad_norm": 7.1377763748168945,
582
+ "learning_rate": 4.338659199472904e-05,
583
+ "loss": 3.0152,
584
+ "step": 820
585
+ },
586
+ {
587
+ "epoch": 0.4094721262950173,
588
+ "grad_norm": 5.029105186462402,
589
+ "learning_rate": 4.330423323999341e-05,
590
+ "loss": 2.3915,
591
+ "step": 830
592
+ },
593
+ {
594
+ "epoch": 0.4144055254070054,
595
+ "grad_norm": 5.954276084899902,
596
+ "learning_rate": 4.3221874485257785e-05,
597
+ "loss": 2.5008,
598
+ "step": 840
599
+ },
600
+ {
601
+ "epoch": 0.41933892451899357,
602
+ "grad_norm": 7.565403938293457,
603
+ "learning_rate": 4.313951573052216e-05,
604
+ "loss": 2.5035,
605
+ "step": 850
606
+ },
607
+ {
608
+ "epoch": 0.4242723236309818,
609
+ "grad_norm": 7.7029314041137695,
610
+ "learning_rate": 4.305715697578653e-05,
611
+ "loss": 2.4186,
612
+ "step": 860
613
+ },
614
+ {
615
+ "epoch": 0.4292057227429699,
616
+ "grad_norm": 8.457364082336426,
617
+ "learning_rate": 4.29747982210509e-05,
618
+ "loss": 2.5165,
619
+ "step": 870
620
+ },
621
+ {
622
+ "epoch": 0.43413912185495807,
623
+ "grad_norm": 10.918882369995117,
624
+ "learning_rate": 4.289243946631527e-05,
625
+ "loss": 2.4591,
626
+ "step": 880
627
+ },
628
+ {
629
+ "epoch": 0.4390725209669462,
630
+ "grad_norm": 9.553812026977539,
631
+ "learning_rate": 4.2810080711579644e-05,
632
+ "loss": 2.6194,
633
+ "step": 890
634
+ },
635
+ {
636
+ "epoch": 0.4440059200789344,
637
+ "grad_norm": 9.679872512817383,
638
+ "learning_rate": 4.272772195684401e-05,
639
+ "loss": 2.9103,
640
+ "step": 900
641
+ },
642
+ {
643
+ "epoch": 0.44893931919092256,
644
+ "grad_norm": 5.805199146270752,
645
+ "learning_rate": 4.2645363202108386e-05,
646
+ "loss": 2.202,
647
+ "step": 910
648
+ },
649
+ {
650
+ "epoch": 0.4538727183029107,
651
+ "grad_norm": 7.699798107147217,
652
+ "learning_rate": 4.256300444737275e-05,
653
+ "loss": 3.0242,
654
+ "step": 920
655
+ },
656
+ {
657
+ "epoch": 0.45880611741489885,
658
+ "grad_norm": 5.121403217315674,
659
+ "learning_rate": 4.248064569263713e-05,
660
+ "loss": 2.2204,
661
+ "step": 930
662
+ },
663
+ {
664
+ "epoch": 0.46373951652688705,
665
+ "grad_norm": 5.217260837554932,
666
+ "learning_rate": 4.23982869379015e-05,
667
+ "loss": 2.6708,
668
+ "step": 940
669
+ },
670
+ {
671
+ "epoch": 0.4686729156388752,
672
+ "grad_norm": 9.238679885864258,
673
+ "learning_rate": 4.231592818316587e-05,
674
+ "loss": 2.7795,
675
+ "step": 950
676
+ },
677
+ {
678
+ "epoch": 0.47360631475086334,
679
+ "grad_norm": 4.4800496101379395,
680
+ "learning_rate": 4.2233569428430244e-05,
681
+ "loss": 2.5765,
682
+ "step": 960
683
+ },
684
+ {
685
+ "epoch": 0.4785397138628515,
686
+ "grad_norm": 11.48628044128418,
687
+ "learning_rate": 4.215121067369461e-05,
688
+ "loss": 2.7761,
689
+ "step": 970
690
+ },
691
+ {
692
+ "epoch": 0.4834731129748397,
693
+ "grad_norm": 8.910355567932129,
694
+ "learning_rate": 4.2068851918958986e-05,
695
+ "loss": 2.8094,
696
+ "step": 980
697
+ },
698
+ {
699
+ "epoch": 0.48840651208682784,
700
+ "grad_norm": 7.16322660446167,
701
+ "learning_rate": 4.198649316422336e-05,
702
+ "loss": 2.3914,
703
+ "step": 990
704
+ },
705
+ {
706
+ "epoch": 0.493339911198816,
707
+ "grad_norm": 15.859028816223145,
708
+ "learning_rate": 4.1904134409487735e-05,
709
+ "loss": 2.5077,
710
+ "step": 1000
711
+ },
712
+ {
713
+ "epoch": 0.49827331031080413,
714
+ "grad_norm": 10.420004844665527,
715
+ "learning_rate": 4.18217756547521e-05,
716
+ "loss": 2.4869,
717
+ "step": 1010
718
+ },
719
+ {
720
+ "epoch": 0.5032067094227923,
721
+ "grad_norm": 5.477315902709961,
722
+ "learning_rate": 4.173941690001648e-05,
723
+ "loss": 2.8359,
724
+ "step": 1020
725
+ },
726
+ {
727
+ "epoch": 0.5081401085347804,
728
+ "grad_norm": 5.169394493103027,
729
+ "learning_rate": 4.1657058145280845e-05,
730
+ "loss": 2.4023,
731
+ "step": 1030
732
+ },
733
+ {
734
+ "epoch": 0.5130735076467686,
735
+ "grad_norm": 8.969653129577637,
736
+ "learning_rate": 4.157469939054522e-05,
737
+ "loss": 2.7727,
738
+ "step": 1040
739
+ },
740
+ {
741
+ "epoch": 0.5180069067587568,
742
+ "grad_norm": 7.325026512145996,
743
+ "learning_rate": 4.149234063580959e-05,
744
+ "loss": 2.3592,
745
+ "step": 1050
746
+ },
747
+ {
748
+ "epoch": 0.5229403058707449,
749
+ "grad_norm": 5.182565212249756,
750
+ "learning_rate": 4.140998188107396e-05,
751
+ "loss": 2.4861,
752
+ "step": 1060
753
+ },
754
+ {
755
+ "epoch": 0.5278737049827331,
756
+ "grad_norm": 6.98890495300293,
757
+ "learning_rate": 4.1327623126338335e-05,
758
+ "loss": 1.9875,
759
+ "step": 1070
760
+ },
761
+ {
762
+ "epoch": 0.5328071040947213,
763
+ "grad_norm": 7.717536449432373,
764
+ "learning_rate": 4.12452643716027e-05,
765
+ "loss": 1.7559,
766
+ "step": 1080
767
+ },
768
+ {
769
+ "epoch": 0.5377405032067094,
770
+ "grad_norm": 8.016460418701172,
771
+ "learning_rate": 4.116290561686708e-05,
772
+ "loss": 2.8117,
773
+ "step": 1090
774
+ },
775
+ {
776
+ "epoch": 0.5426739023186976,
777
+ "grad_norm": 6.484166622161865,
778
+ "learning_rate": 4.1080546862131445e-05,
779
+ "loss": 2.4036,
780
+ "step": 1100
781
+ },
782
+ {
783
+ "epoch": 0.5476073014306857,
784
+ "grad_norm": 4.496994495391846,
785
+ "learning_rate": 4.099818810739582e-05,
786
+ "loss": 2.4948,
787
+ "step": 1110
788
+ },
789
+ {
790
+ "epoch": 0.5525407005426739,
791
+ "grad_norm": 6.8036065101623535,
792
+ "learning_rate": 4.091582935266019e-05,
793
+ "loss": 2.4636,
794
+ "step": 1120
795
+ },
796
+ {
797
+ "epoch": 0.5574740996546621,
798
+ "grad_norm": 8.330781936645508,
799
+ "learning_rate": 4.083347059792456e-05,
800
+ "loss": 2.6983,
801
+ "step": 1130
802
+ },
803
+ {
804
+ "epoch": 0.5624074987666502,
805
+ "grad_norm": 6.868759632110596,
806
+ "learning_rate": 4.0751111843188936e-05,
807
+ "loss": 2.5327,
808
+ "step": 1140
809
+ },
810
+ {
811
+ "epoch": 0.5673408978786384,
812
+ "grad_norm": 7.583151340484619,
813
+ "learning_rate": 4.0668753088453303e-05,
814
+ "loss": 2.4815,
815
+ "step": 1150
816
+ },
817
+ {
818
+ "epoch": 0.5722742969906266,
819
+ "grad_norm": 7.823330402374268,
820
+ "learning_rate": 4.058639433371768e-05,
821
+ "loss": 2.2532,
822
+ "step": 1160
823
+ },
824
+ {
825
+ "epoch": 0.5772076961026147,
826
+ "grad_norm": 6.149056434631348,
827
+ "learning_rate": 4.0504035578982045e-05,
828
+ "loss": 2.762,
829
+ "step": 1170
830
+ },
831
+ {
832
+ "epoch": 0.5821410952146029,
833
+ "grad_norm": 9.300018310546875,
834
+ "learning_rate": 4.042167682424642e-05,
835
+ "loss": 2.505,
836
+ "step": 1180
837
+ },
838
+ {
839
+ "epoch": 0.587074494326591,
840
+ "grad_norm": 10.552204132080078,
841
+ "learning_rate": 4.033931806951079e-05,
842
+ "loss": 2.6128,
843
+ "step": 1190
844
+ },
845
+ {
846
+ "epoch": 0.5920078934385792,
847
+ "grad_norm": 10.5072603225708,
848
+ "learning_rate": 4.025695931477516e-05,
849
+ "loss": 2.6206,
850
+ "step": 1200
851
+ },
852
+ {
853
+ "epoch": 0.5969412925505674,
854
+ "grad_norm": 5.355686664581299,
855
+ "learning_rate": 4.017460056003953e-05,
856
+ "loss": 2.8745,
857
+ "step": 1210
858
+ },
859
+ {
860
+ "epoch": 0.6018746916625555,
861
+ "grad_norm": 4.0471649169921875,
862
+ "learning_rate": 4.0092241805303904e-05,
863
+ "loss": 2.7503,
864
+ "step": 1220
865
+ },
866
+ {
867
+ "epoch": 0.6068080907745437,
868
+ "grad_norm": 6.991344928741455,
869
+ "learning_rate": 4.000988305056828e-05,
870
+ "loss": 2.686,
871
+ "step": 1230
872
+ },
873
+ {
874
+ "epoch": 0.6117414898865319,
875
+ "grad_norm": 4.674221515655518,
876
+ "learning_rate": 3.9927524295832646e-05,
877
+ "loss": 2.5496,
878
+ "step": 1240
879
+ },
880
+ {
881
+ "epoch": 0.61667488899852,
882
+ "grad_norm": 23.224143981933594,
883
+ "learning_rate": 3.984516554109702e-05,
884
+ "loss": 3.0394,
885
+ "step": 1250
886
+ },
887
+ {
888
+ "epoch": 0.6216082881105082,
889
+ "grad_norm": 8.688756942749023,
890
+ "learning_rate": 3.976280678636139e-05,
891
+ "loss": 2.7851,
892
+ "step": 1260
893
+ },
894
+ {
895
+ "epoch": 0.6265416872224963,
896
+ "grad_norm": 6.090677261352539,
897
+ "learning_rate": 3.968044803162576e-05,
898
+ "loss": 2.2354,
899
+ "step": 1270
900
+ },
901
+ {
902
+ "epoch": 0.6314750863344845,
903
+ "grad_norm": 9.011785507202148,
904
+ "learning_rate": 3.959808927689014e-05,
905
+ "loss": 2.1909,
906
+ "step": 1280
907
+ },
908
+ {
909
+ "epoch": 0.6364084854464727,
910
+ "grad_norm": 9.320941925048828,
911
+ "learning_rate": 3.951573052215451e-05,
912
+ "loss": 2.1964,
913
+ "step": 1290
914
+ },
915
+ {
916
+ "epoch": 0.6413418845584608,
917
+ "grad_norm": 6.079468250274658,
918
+ "learning_rate": 3.943337176741888e-05,
919
+ "loss": 2.1266,
920
+ "step": 1300
921
+ },
922
+ {
923
+ "epoch": 0.646275283670449,
924
+ "grad_norm": 6.427783489227295,
925
+ "learning_rate": 3.935101301268325e-05,
926
+ "loss": 2.1273,
927
+ "step": 1310
928
+ },
929
+ {
930
+ "epoch": 0.6512086827824372,
931
+ "grad_norm": 5.442935943603516,
932
+ "learning_rate": 3.926865425794762e-05,
933
+ "loss": 2.3792,
934
+ "step": 1320
935
+ },
936
+ {
937
+ "epoch": 0.6561420818944252,
938
+ "grad_norm": 5.503900527954102,
939
+ "learning_rate": 3.9186295503211995e-05,
940
+ "loss": 2.3938,
941
+ "step": 1330
942
+ },
943
+ {
944
+ "epoch": 0.6610754810064134,
945
+ "grad_norm": 9.838761329650879,
946
+ "learning_rate": 3.910393674847637e-05,
947
+ "loss": 2.2793,
948
+ "step": 1340
949
+ },
950
+ {
951
+ "epoch": 0.6660088801184015,
952
+ "grad_norm": 7.890000343322754,
953
+ "learning_rate": 3.902157799374074e-05,
954
+ "loss": 2.3842,
955
+ "step": 1350
956
+ },
957
+ {
958
+ "epoch": 0.6709422792303897,
959
+ "grad_norm": 7.321061134338379,
960
+ "learning_rate": 3.893921923900511e-05,
961
+ "loss": 2.3586,
962
+ "step": 1360
963
+ },
964
+ {
965
+ "epoch": 0.6758756783423779,
966
+ "grad_norm": 5.176478385925293,
967
+ "learning_rate": 3.885686048426948e-05,
968
+ "loss": 2.6807,
969
+ "step": 1370
970
+ },
971
+ {
972
+ "epoch": 0.680809077454366,
973
+ "grad_norm": 4.9928975105285645,
974
+ "learning_rate": 3.8774501729533854e-05,
975
+ "loss": 1.8495,
976
+ "step": 1380
977
+ },
978
+ {
979
+ "epoch": 0.6857424765663542,
980
+ "grad_norm": 6.170323371887207,
981
+ "learning_rate": 3.869214297479822e-05,
982
+ "loss": 2.415,
983
+ "step": 1390
984
+ },
985
+ {
986
+ "epoch": 0.6906758756783424,
987
+ "grad_norm": 10.49080753326416,
988
+ "learning_rate": 3.8609784220062596e-05,
989
+ "loss": 1.8006,
990
+ "step": 1400
991
+ },
992
+ {
993
+ "epoch": 0.6956092747903305,
994
+ "grad_norm": 16.418981552124023,
995
+ "learning_rate": 3.852742546532696e-05,
996
+ "loss": 2.5657,
997
+ "step": 1410
998
+ },
999
+ {
1000
+ "epoch": 0.7005426739023187,
1001
+ "grad_norm": 4.002269268035889,
1002
+ "learning_rate": 3.844506671059134e-05,
1003
+ "loss": 2.5927,
1004
+ "step": 1420
1005
+ },
1006
+ {
1007
+ "epoch": 0.7054760730143068,
1008
+ "grad_norm": 7.075521469116211,
1009
+ "learning_rate": 3.836270795585571e-05,
1010
+ "loss": 1.8786,
1011
+ "step": 1430
1012
+ },
1013
+ {
1014
+ "epoch": 0.710409472126295,
1015
+ "grad_norm": 6.2812981605529785,
1016
+ "learning_rate": 3.828034920112008e-05,
1017
+ "loss": 2.1746,
1018
+ "step": 1440
1019
+ },
1020
+ {
1021
+ "epoch": 0.7153428712382832,
1022
+ "grad_norm": 7.315672874450684,
1023
+ "learning_rate": 3.8197990446384454e-05,
1024
+ "loss": 2.4418,
1025
+ "step": 1450
1026
+ },
1027
+ {
1028
+ "epoch": 0.7202762703502713,
1029
+ "grad_norm": 7.978590488433838,
1030
+ "learning_rate": 3.811563169164882e-05,
1031
+ "loss": 2.3702,
1032
+ "step": 1460
1033
+ },
1034
+ {
1035
+ "epoch": 0.7252096694622595,
1036
+ "grad_norm": 8.856497764587402,
1037
+ "learning_rate": 3.8033272936913196e-05,
1038
+ "loss": 2.7627,
1039
+ "step": 1470
1040
+ },
1041
+ {
1042
+ "epoch": 0.7301430685742477,
1043
+ "grad_norm": 5.009411811828613,
1044
+ "learning_rate": 3.7950914182177564e-05,
1045
+ "loss": 2.1163,
1046
+ "step": 1480
1047
+ },
1048
+ {
1049
+ "epoch": 0.7350764676862358,
1050
+ "grad_norm": 6.4281134605407715,
1051
+ "learning_rate": 3.786855542744194e-05,
1052
+ "loss": 2.554,
1053
+ "step": 1490
1054
+ },
1055
+ {
1056
+ "epoch": 0.740009866798224,
1057
+ "grad_norm": 6.119235515594482,
1058
+ "learning_rate": 3.7786196672706306e-05,
1059
+ "loss": 2.1492,
1060
+ "step": 1500
1061
+ },
1062
+ {
1063
+ "epoch": 0.7449432659102121,
1064
+ "grad_norm": 4.099040508270264,
1065
+ "learning_rate": 3.770383791797068e-05,
1066
+ "loss": 2.9967,
1067
+ "step": 1510
1068
+ },
1069
+ {
1070
+ "epoch": 0.7498766650222003,
1071
+ "grad_norm": 6.364965438842773,
1072
+ "learning_rate": 3.7621479163235055e-05,
1073
+ "loss": 2.7822,
1074
+ "step": 1520
1075
+ },
1076
+ {
1077
+ "epoch": 0.7548100641341885,
1078
+ "grad_norm": 13.07193374633789,
1079
+ "learning_rate": 3.753912040849942e-05,
1080
+ "loss": 2.1634,
1081
+ "step": 1530
1082
+ },
1083
+ {
1084
+ "epoch": 0.7597434632461766,
1085
+ "grad_norm": 4.285301685333252,
1086
+ "learning_rate": 3.74567616537638e-05,
1087
+ "loss": 2.49,
1088
+ "step": 1540
1089
+ },
1090
+ {
1091
+ "epoch": 0.7646768623581648,
1092
+ "grad_norm": 8.684725761413574,
1093
+ "learning_rate": 3.7374402899028164e-05,
1094
+ "loss": 2.4202,
1095
+ "step": 1550
1096
+ },
1097
+ {
1098
+ "epoch": 0.769610261470153,
1099
+ "grad_norm": 7.857943058013916,
1100
+ "learning_rate": 3.7292044144292545e-05,
1101
+ "loss": 2.5489,
1102
+ "step": 1560
1103
+ },
1104
+ {
1105
+ "epoch": 0.7745436605821411,
1106
+ "grad_norm": 10.40462589263916,
1107
+ "learning_rate": 3.720968538955691e-05,
1108
+ "loss": 2.5975,
1109
+ "step": 1570
1110
+ },
1111
+ {
1112
+ "epoch": 0.7794770596941293,
1113
+ "grad_norm": 3.614388942718506,
1114
+ "learning_rate": 3.712732663482129e-05,
1115
+ "loss": 2.5092,
1116
+ "step": 1580
1117
+ },
1118
+ {
1119
+ "epoch": 0.7844104588061174,
1120
+ "grad_norm": 4.367038726806641,
1121
+ "learning_rate": 3.7044967880085655e-05,
1122
+ "loss": 2.4972,
1123
+ "step": 1590
1124
+ },
1125
+ {
1126
+ "epoch": 0.7893438579181056,
1127
+ "grad_norm": 8.629862785339355,
1128
+ "learning_rate": 3.696260912535003e-05,
1129
+ "loss": 2.8085,
1130
+ "step": 1600
1131
+ },
1132
+ {
1133
+ "epoch": 0.7942772570300938,
1134
+ "grad_norm": 4.455677032470703,
1135
+ "learning_rate": 3.68802503706144e-05,
1136
+ "loss": 2.2631,
1137
+ "step": 1610
1138
+ },
1139
+ {
1140
+ "epoch": 0.7992106561420819,
1141
+ "grad_norm": 3.8020565509796143,
1142
+ "learning_rate": 3.679789161587877e-05,
1143
+ "loss": 2.1875,
1144
+ "step": 1620
1145
+ },
1146
+ {
1147
+ "epoch": 0.8041440552540701,
1148
+ "grad_norm": 8.94434928894043,
1149
+ "learning_rate": 3.6715532861143146e-05,
1150
+ "loss": 2.4996,
1151
+ "step": 1630
1152
+ },
1153
+ {
1154
+ "epoch": 0.8090774543660583,
1155
+ "grad_norm": 6.989756107330322,
1156
+ "learning_rate": 3.6633174106407514e-05,
1157
+ "loss": 2.0738,
1158
+ "step": 1640
1159
+ },
1160
+ {
1161
+ "epoch": 0.8140108534780464,
1162
+ "grad_norm": 6.769233226776123,
1163
+ "learning_rate": 3.655081535167189e-05,
1164
+ "loss": 2.4975,
1165
+ "step": 1650
1166
+ },
1167
+ {
1168
+ "epoch": 0.8189442525900346,
1169
+ "grad_norm": 5.789883136749268,
1170
+ "learning_rate": 3.6468456596936256e-05,
1171
+ "loss": 2.1423,
1172
+ "step": 1660
1173
+ },
1174
+ {
1175
+ "epoch": 0.8238776517020227,
1176
+ "grad_norm": 8.308130264282227,
1177
+ "learning_rate": 3.638609784220063e-05,
1178
+ "loss": 2.364,
1179
+ "step": 1670
1180
+ },
1181
+ {
1182
+ "epoch": 0.8288110508140109,
1183
+ "grad_norm": 7.824023246765137,
1184
+ "learning_rate": 3.6303739087465e-05,
1185
+ "loss": 2.001,
1186
+ "step": 1680
1187
+ },
1188
+ {
1189
+ "epoch": 0.833744449925999,
1190
+ "grad_norm": 6.625056266784668,
1191
+ "learning_rate": 3.622138033272937e-05,
1192
+ "loss": 2.4029,
1193
+ "step": 1690
1194
+ },
1195
+ {
1196
+ "epoch": 0.8386778490379871,
1197
+ "grad_norm": 3.973498582839966,
1198
+ "learning_rate": 3.613902157799374e-05,
1199
+ "loss": 2.766,
1200
+ "step": 1700
1201
+ },
1202
+ {
1203
+ "epoch": 0.8436112481499753,
1204
+ "grad_norm": 6.130688667297363,
1205
+ "learning_rate": 3.6056662823258114e-05,
1206
+ "loss": 2.4471,
1207
+ "step": 1710
1208
+ },
1209
+ {
1210
+ "epoch": 0.8485446472619635,
1211
+ "grad_norm": 4.856011867523193,
1212
+ "learning_rate": 3.597430406852249e-05,
1213
+ "loss": 2.035,
1214
+ "step": 1720
1215
+ },
1216
+ {
1217
+ "epoch": 0.8534780463739516,
1218
+ "grad_norm": 5.570549488067627,
1219
+ "learning_rate": 3.5891945313786856e-05,
1220
+ "loss": 1.8751,
1221
+ "step": 1730
1222
+ },
1223
+ {
1224
+ "epoch": 0.8584114454859398,
1225
+ "grad_norm": 8.646450996398926,
1226
+ "learning_rate": 3.580958655905123e-05,
1227
+ "loss": 2.0868,
1228
+ "step": 1740
1229
+ },
1230
+ {
1231
+ "epoch": 0.8633448445979279,
1232
+ "grad_norm": 5.265438556671143,
1233
+ "learning_rate": 3.57272278043156e-05,
1234
+ "loss": 2.5141,
1235
+ "step": 1750
1236
+ },
1237
+ {
1238
+ "epoch": 0.8682782437099161,
1239
+ "grad_norm": 11.451949119567871,
1240
+ "learning_rate": 3.564486904957997e-05,
1241
+ "loss": 2.3301,
1242
+ "step": 1760
1243
+ },
1244
+ {
1245
+ "epoch": 0.8732116428219043,
1246
+ "grad_norm": 8.040514945983887,
1247
+ "learning_rate": 3.556251029484434e-05,
1248
+ "loss": 2.4714,
1249
+ "step": 1770
1250
+ },
1251
+ {
1252
+ "epoch": 0.8781450419338924,
1253
+ "grad_norm": 6.2226481437683105,
1254
+ "learning_rate": 3.5480151540108715e-05,
1255
+ "loss": 2.667,
1256
+ "step": 1780
1257
+ },
1258
+ {
1259
+ "epoch": 0.8830784410458806,
1260
+ "grad_norm": 9.629585266113281,
1261
+ "learning_rate": 3.539779278537308e-05,
1262
+ "loss": 2.6848,
1263
+ "step": 1790
1264
+ },
1265
+ {
1266
+ "epoch": 0.8880118401578688,
1267
+ "grad_norm": 8.996950149536133,
1268
+ "learning_rate": 3.5315434030637457e-05,
1269
+ "loss": 2.4991,
1270
+ "step": 1800
1271
+ },
1272
+ {
1273
+ "epoch": 0.8929452392698569,
1274
+ "grad_norm": 7.078036785125732,
1275
+ "learning_rate": 3.523307527590183e-05,
1276
+ "loss": 2.5703,
1277
+ "step": 1810
1278
+ },
1279
+ {
1280
+ "epoch": 0.8978786383818451,
1281
+ "grad_norm": 6.913575649261475,
1282
+ "learning_rate": 3.51507165211662e-05,
1283
+ "loss": 2.5782,
1284
+ "step": 1820
1285
+ },
1286
+ {
1287
+ "epoch": 0.9028120374938332,
1288
+ "grad_norm": 8.875394821166992,
1289
+ "learning_rate": 3.506835776643057e-05,
1290
+ "loss": 2.1843,
1291
+ "step": 1830
1292
+ },
1293
+ {
1294
+ "epoch": 0.9077454366058214,
1295
+ "grad_norm": 12.219450950622559,
1296
+ "learning_rate": 3.498599901169494e-05,
1297
+ "loss": 2.7252,
1298
+ "step": 1840
1299
+ },
1300
+ {
1301
+ "epoch": 0.9126788357178096,
1302
+ "grad_norm": 8.12746524810791,
1303
+ "learning_rate": 3.490364025695932e-05,
1304
+ "loss": 2.2132,
1305
+ "step": 1850
1306
+ },
1307
+ {
1308
+ "epoch": 0.9176122348297977,
1309
+ "grad_norm": 17.30431365966797,
1310
+ "learning_rate": 3.482128150222369e-05,
1311
+ "loss": 2.807,
1312
+ "step": 1860
1313
+ },
1314
+ {
1315
+ "epoch": 0.9225456339417859,
1316
+ "grad_norm": 4.950164794921875,
1317
+ "learning_rate": 3.4738922747488064e-05,
1318
+ "loss": 2.2689,
1319
+ "step": 1870
1320
+ },
1321
+ {
1322
+ "epoch": 0.9274790330537741,
1323
+ "grad_norm": 6.925759315490723,
1324
+ "learning_rate": 3.465656399275243e-05,
1325
+ "loss": 2.3312,
1326
+ "step": 1880
1327
+ },
1328
+ {
1329
+ "epoch": 0.9324124321657622,
1330
+ "grad_norm": 5.608890533447266,
1331
+ "learning_rate": 3.4574205238016806e-05,
1332
+ "loss": 2.305,
1333
+ "step": 1890
1334
+ },
1335
+ {
1336
+ "epoch": 0.9373458312777504,
1337
+ "grad_norm": 8.436676025390625,
1338
+ "learning_rate": 3.4491846483281173e-05,
1339
+ "loss": 2.4468,
1340
+ "step": 1900
1341
+ },
1342
+ {
1343
+ "epoch": 0.9422792303897385,
1344
+ "grad_norm": 14.684228897094727,
1345
+ "learning_rate": 3.440948772854555e-05,
1346
+ "loss": 2.4759,
1347
+ "step": 1910
1348
+ },
1349
+ {
1350
+ "epoch": 0.9472126295017267,
1351
+ "grad_norm": 9.368124961853027,
1352
+ "learning_rate": 3.432712897380992e-05,
1353
+ "loss": 2.5555,
1354
+ "step": 1920
1355
+ },
1356
+ {
1357
+ "epoch": 0.9521460286137149,
1358
+ "grad_norm": 8.614909172058105,
1359
+ "learning_rate": 3.424477021907429e-05,
1360
+ "loss": 2.351,
1361
+ "step": 1930
1362
+ },
1363
+ {
1364
+ "epoch": 0.957079427725703,
1365
+ "grad_norm": 10.137232780456543,
1366
+ "learning_rate": 3.4162411464338664e-05,
1367
+ "loss": 2.2095,
1368
+ "step": 1940
1369
+ },
1370
+ {
1371
+ "epoch": 0.9620128268376912,
1372
+ "grad_norm": 5.5300493240356445,
1373
+ "learning_rate": 3.408005270960303e-05,
1374
+ "loss": 2.2628,
1375
+ "step": 1950
1376
+ },
1377
+ {
1378
+ "epoch": 0.9669462259496794,
1379
+ "grad_norm": 5.127216815948486,
1380
+ "learning_rate": 3.3997693954867406e-05,
1381
+ "loss": 1.9502,
1382
+ "step": 1960
1383
+ },
1384
+ {
1385
+ "epoch": 0.9718796250616675,
1386
+ "grad_norm": 4.680507659912109,
1387
+ "learning_rate": 3.3915335200131774e-05,
1388
+ "loss": 1.8751,
1389
+ "step": 1970
1390
+ },
1391
+ {
1392
+ "epoch": 0.9768130241736557,
1393
+ "grad_norm": 6.784450531005859,
1394
+ "learning_rate": 3.383297644539615e-05,
1395
+ "loss": 2.477,
1396
+ "step": 1980
1397
+ },
1398
+ {
1399
+ "epoch": 0.9817464232856438,
1400
+ "grad_norm": 6.171282768249512,
1401
+ "learning_rate": 3.3750617690660516e-05,
1402
+ "loss": 2.4942,
1403
+ "step": 1990
1404
+ },
1405
+ {
1406
+ "epoch": 0.986679822397632,
1407
+ "grad_norm": 6.970947265625,
1408
+ "learning_rate": 3.366825893592489e-05,
1409
+ "loss": 2.2718,
1410
+ "step": 2000
1411
+ },
1412
+ {
1413
+ "epoch": 0.9916132215096202,
1414
+ "grad_norm": 4.48728084564209,
1415
+ "learning_rate": 3.3585900181189265e-05,
1416
+ "loss": 2.6097,
1417
+ "step": 2010
1418
+ },
1419
+ {
1420
+ "epoch": 0.9965466206216083,
1421
+ "grad_norm": 13.015890121459961,
1422
+ "learning_rate": 3.350354142645363e-05,
1423
+ "loss": 2.7752,
1424
+ "step": 2020
1425
+ },
1426
+ {
1427
+ "epoch": 1.0014800197335965,
1428
+ "grad_norm": 7.251813888549805,
1429
+ "learning_rate": 3.342118267171801e-05,
1430
+ "loss": 2.0151,
1431
+ "step": 2030
1432
+ },
1433
+ {
1434
+ "epoch": 1.0064134188455847,
1435
+ "grad_norm": 5.12528133392334,
1436
+ "learning_rate": 3.3338823916982374e-05,
1437
+ "loss": 2.0101,
1438
+ "step": 2040
1439
+ },
1440
+ {
1441
+ "epoch": 1.0113468179575729,
1442
+ "grad_norm": 5.179903507232666,
1443
+ "learning_rate": 3.325646516224675e-05,
1444
+ "loss": 2.0539,
1445
+ "step": 2050
1446
+ },
1447
+ {
1448
+ "epoch": 1.0162802170695608,
1449
+ "grad_norm": 13.52995777130127,
1450
+ "learning_rate": 3.3174106407511116e-05,
1451
+ "loss": 1.9679,
1452
+ "step": 2060
1453
+ },
1454
+ {
1455
+ "epoch": 1.021213616181549,
1456
+ "grad_norm": 10.211468696594238,
1457
+ "learning_rate": 3.309174765277549e-05,
1458
+ "loss": 2.4251,
1459
+ "step": 2070
1460
+ },
1461
+ {
1462
+ "epoch": 1.0261470152935372,
1463
+ "grad_norm": 11.468674659729004,
1464
+ "learning_rate": 3.300938889803986e-05,
1465
+ "loss": 2.026,
1466
+ "step": 2080
1467
+ },
1468
+ {
1469
+ "epoch": 1.0310804144055254,
1470
+ "grad_norm": 9.60104751586914,
1471
+ "learning_rate": 3.292703014330423e-05,
1472
+ "loss": 2.0974,
1473
+ "step": 2090
1474
+ },
1475
+ {
1476
+ "epoch": 1.0360138135175136,
1477
+ "grad_norm": 16.93134307861328,
1478
+ "learning_rate": 3.284467138856861e-05,
1479
+ "loss": 1.8676,
1480
+ "step": 2100
1481
+ },
1482
+ {
1483
+ "epoch": 1.0409472126295016,
1484
+ "grad_norm": 7.654803276062012,
1485
+ "learning_rate": 3.2762312633832975e-05,
1486
+ "loss": 1.6796,
1487
+ "step": 2110
1488
+ },
1489
+ {
1490
+ "epoch": 1.0458806117414898,
1491
+ "grad_norm": 7.115079879760742,
1492
+ "learning_rate": 3.267995387909735e-05,
1493
+ "loss": 1.9246,
1494
+ "step": 2120
1495
+ },
1496
+ {
1497
+ "epoch": 1.050814010853478,
1498
+ "grad_norm": 5.4859185218811035,
1499
+ "learning_rate": 3.2597595124361724e-05,
1500
+ "loss": 2.0991,
1501
+ "step": 2130
1502
+ },
1503
+ {
1504
+ "epoch": 1.0557474099654662,
1505
+ "grad_norm": 16.93896484375,
1506
+ "learning_rate": 3.25152363696261e-05,
1507
+ "loss": 1.87,
1508
+ "step": 2140
1509
+ },
1510
+ {
1511
+ "epoch": 1.0606808090774544,
1512
+ "grad_norm": 8.788496017456055,
1513
+ "learning_rate": 3.2432877614890466e-05,
1514
+ "loss": 1.8518,
1515
+ "step": 2150
1516
+ },
1517
+ {
1518
+ "epoch": 1.0656142081894426,
1519
+ "grad_norm": 5.605971336364746,
1520
+ "learning_rate": 3.235051886015484e-05,
1521
+ "loss": 1.4863,
1522
+ "step": 2160
1523
+ },
1524
+ {
1525
+ "epoch": 1.0705476073014306,
1526
+ "grad_norm": 8.495088577270508,
1527
+ "learning_rate": 3.226816010541921e-05,
1528
+ "loss": 1.7369,
1529
+ "step": 2170
1530
+ },
1531
+ {
1532
+ "epoch": 1.0754810064134188,
1533
+ "grad_norm": 9.221963882446289,
1534
+ "learning_rate": 3.218580135068358e-05,
1535
+ "loss": 2.0833,
1536
+ "step": 2180
1537
+ },
1538
+ {
1539
+ "epoch": 1.080414405525407,
1540
+ "grad_norm": 5.741699695587158,
1541
+ "learning_rate": 3.210344259594795e-05,
1542
+ "loss": 2.4029,
1543
+ "step": 2190
1544
+ },
1545
+ {
1546
+ "epoch": 1.0853478046373952,
1547
+ "grad_norm": 9.234859466552734,
1548
+ "learning_rate": 3.2021083841212324e-05,
1549
+ "loss": 1.5287,
1550
+ "step": 2200
1551
+ },
1552
+ {
1553
+ "epoch": 1.0902812037493834,
1554
+ "grad_norm": 7.735116004943848,
1555
+ "learning_rate": 3.19387250864767e-05,
1556
+ "loss": 1.9516,
1557
+ "step": 2210
1558
+ },
1559
+ {
1560
+ "epoch": 1.0952146028613714,
1561
+ "grad_norm": 6.103670120239258,
1562
+ "learning_rate": 3.1856366331741066e-05,
1563
+ "loss": 1.9241,
1564
+ "step": 2220
1565
+ },
1566
+ {
1567
+ "epoch": 1.1001480019733596,
1568
+ "grad_norm": 5.348116397857666,
1569
+ "learning_rate": 3.177400757700544e-05,
1570
+ "loss": 1.6599,
1571
+ "step": 2230
1572
+ },
1573
+ {
1574
+ "epoch": 1.1050814010853478,
1575
+ "grad_norm": 11.653718948364258,
1576
+ "learning_rate": 3.169164882226981e-05,
1577
+ "loss": 1.9352,
1578
+ "step": 2240
1579
+ },
1580
+ {
1581
+ "epoch": 1.110014800197336,
1582
+ "grad_norm": 8.342635154724121,
1583
+ "learning_rate": 3.160929006753418e-05,
1584
+ "loss": 1.963,
1585
+ "step": 2250
1586
+ },
1587
+ {
1588
+ "epoch": 1.1149481993093242,
1589
+ "grad_norm": 6.9188103675842285,
1590
+ "learning_rate": 3.152693131279855e-05,
1591
+ "loss": 1.9066,
1592
+ "step": 2260
1593
+ },
1594
+ {
1595
+ "epoch": 1.1198815984213122,
1596
+ "grad_norm": 7.183344841003418,
1597
+ "learning_rate": 3.1444572558062925e-05,
1598
+ "loss": 1.9223,
1599
+ "step": 2270
1600
+ },
1601
+ {
1602
+ "epoch": 1.1248149975333004,
1603
+ "grad_norm": 6.872612476348877,
1604
+ "learning_rate": 3.136221380332729e-05,
1605
+ "loss": 1.5956,
1606
+ "step": 2280
1607
+ },
1608
+ {
1609
+ "epoch": 1.1297483966452886,
1610
+ "grad_norm": 7.547828674316406,
1611
+ "learning_rate": 3.127985504859167e-05,
1612
+ "loss": 1.7239,
1613
+ "step": 2290
1614
+ },
1615
+ {
1616
+ "epoch": 1.1346817957572768,
1617
+ "grad_norm": 9.748641014099121,
1618
+ "learning_rate": 3.119749629385604e-05,
1619
+ "loss": 1.455,
1620
+ "step": 2300
1621
+ },
1622
+ {
1623
+ "epoch": 1.139615194869265,
1624
+ "grad_norm": 13.310020446777344,
1625
+ "learning_rate": 3.111513753912041e-05,
1626
+ "loss": 1.9041,
1627
+ "step": 2310
1628
+ },
1629
+ {
1630
+ "epoch": 1.144548593981253,
1631
+ "grad_norm": 8.848477363586426,
1632
+ "learning_rate": 3.103277878438478e-05,
1633
+ "loss": 2.0644,
1634
+ "step": 2320
1635
+ },
1636
+ {
1637
+ "epoch": 1.1494819930932412,
1638
+ "grad_norm": 7.570590019226074,
1639
+ "learning_rate": 3.095042002964915e-05,
1640
+ "loss": 2.1607,
1641
+ "step": 2330
1642
+ },
1643
+ {
1644
+ "epoch": 1.1544153922052294,
1645
+ "grad_norm": 7.833923816680908,
1646
+ "learning_rate": 3.0868061274913525e-05,
1647
+ "loss": 1.5976,
1648
+ "step": 2340
1649
+ },
1650
+ {
1651
+ "epoch": 1.1593487913172176,
1652
+ "grad_norm": 6.807453632354736,
1653
+ "learning_rate": 3.078570252017789e-05,
1654
+ "loss": 1.8354,
1655
+ "step": 2350
1656
+ },
1657
+ {
1658
+ "epoch": 1.1642821904292058,
1659
+ "grad_norm": 5.707053184509277,
1660
+ "learning_rate": 3.070334376544227e-05,
1661
+ "loss": 2.0815,
1662
+ "step": 2360
1663
+ },
1664
+ {
1665
+ "epoch": 1.169215589541194,
1666
+ "grad_norm": 3.920215606689453,
1667
+ "learning_rate": 3.0620985010706635e-05,
1668
+ "loss": 2.1072,
1669
+ "step": 2370
1670
+ },
1671
+ {
1672
+ "epoch": 1.174148988653182,
1673
+ "grad_norm": 8.955766677856445,
1674
+ "learning_rate": 3.053862625597101e-05,
1675
+ "loss": 1.9202,
1676
+ "step": 2380
1677
+ },
1678
+ {
1679
+ "epoch": 1.1790823877651702,
1680
+ "grad_norm": 6.6745219230651855,
1681
+ "learning_rate": 3.045626750123538e-05,
1682
+ "loss": 1.9565,
1683
+ "step": 2390
1684
+ },
1685
+ {
1686
+ "epoch": 1.1840157868771584,
1687
+ "grad_norm": 5.365530967712402,
1688
+ "learning_rate": 3.037390874649975e-05,
1689
+ "loss": 1.8332,
1690
+ "step": 2400
1691
+ },
1692
+ {
1693
+ "epoch": 1.1889491859891466,
1694
+ "grad_norm": 4.238736629486084,
1695
+ "learning_rate": 3.0291549991764122e-05,
1696
+ "loss": 1.6349,
1697
+ "step": 2410
1698
+ },
1699
+ {
1700
+ "epoch": 1.1938825851011348,
1701
+ "grad_norm": 4.686330795288086,
1702
+ "learning_rate": 3.02091912370285e-05,
1703
+ "loss": 2.0218,
1704
+ "step": 2420
1705
+ },
1706
+ {
1707
+ "epoch": 1.1988159842131227,
1708
+ "grad_norm": 10.073396682739258,
1709
+ "learning_rate": 3.012683248229287e-05,
1710
+ "loss": 2.1104,
1711
+ "step": 2430
1712
+ },
1713
+ {
1714
+ "epoch": 1.203749383325111,
1715
+ "grad_norm": 7.226747035980225,
1716
+ "learning_rate": 3.0044473727557242e-05,
1717
+ "loss": 1.7695,
1718
+ "step": 2440
1719
+ },
1720
+ {
1721
+ "epoch": 1.2086827824370991,
1722
+ "grad_norm": 6.546811103820801,
1723
+ "learning_rate": 2.9962114972821613e-05,
1724
+ "loss": 1.5873,
1725
+ "step": 2450
1726
+ },
1727
+ {
1728
+ "epoch": 1.2136161815490873,
1729
+ "grad_norm": 7.798117637634277,
1730
+ "learning_rate": 2.9879756218085987e-05,
1731
+ "loss": 1.7036,
1732
+ "step": 2460
1733
+ },
1734
+ {
1735
+ "epoch": 1.2185495806610755,
1736
+ "grad_norm": 5.489384174346924,
1737
+ "learning_rate": 2.979739746335036e-05,
1738
+ "loss": 1.9574,
1739
+ "step": 2470
1740
+ },
1741
+ {
1742
+ "epoch": 1.2234829797730638,
1743
+ "grad_norm": 11.180368423461914,
1744
+ "learning_rate": 2.971503870861473e-05,
1745
+ "loss": 1.7365,
1746
+ "step": 2480
1747
+ },
1748
+ {
1749
+ "epoch": 1.2284163788850517,
1750
+ "grad_norm": 6.837683200836182,
1751
+ "learning_rate": 2.96326799538791e-05,
1752
+ "loss": 1.8239,
1753
+ "step": 2490
1754
+ },
1755
+ {
1756
+ "epoch": 1.23334977799704,
1757
+ "grad_norm": 6.739687919616699,
1758
+ "learning_rate": 2.955032119914347e-05,
1759
+ "loss": 1.6268,
1760
+ "step": 2500
1761
+ },
1762
+ {
1763
+ "epoch": 1.2382831771090281,
1764
+ "grad_norm": 6.837540149688721,
1765
+ "learning_rate": 2.9467962444407842e-05,
1766
+ "loss": 1.9372,
1767
+ "step": 2510
1768
+ },
1769
+ {
1770
+ "epoch": 1.2432165762210163,
1771
+ "grad_norm": 12.321346282958984,
1772
+ "learning_rate": 2.9385603689672213e-05,
1773
+ "loss": 1.7772,
1774
+ "step": 2520
1775
+ },
1776
+ {
1777
+ "epoch": 1.2481499753330045,
1778
+ "grad_norm": 5.676684379577637,
1779
+ "learning_rate": 2.9303244934936584e-05,
1780
+ "loss": 2.2801,
1781
+ "step": 2530
1782
+ },
1783
+ {
1784
+ "epoch": 1.2530833744449925,
1785
+ "grad_norm": 8.29293155670166,
1786
+ "learning_rate": 2.9220886180200955e-05,
1787
+ "loss": 1.9417,
1788
+ "step": 2540
1789
+ },
1790
+ {
1791
+ "epoch": 1.2580167735569807,
1792
+ "grad_norm": 10.451891899108887,
1793
+ "learning_rate": 2.913852742546533e-05,
1794
+ "loss": 1.9312,
1795
+ "step": 2550
1796
+ },
1797
+ {
1798
+ "epoch": 1.262950172668969,
1799
+ "grad_norm": 6.803496837615967,
1800
+ "learning_rate": 2.90561686707297e-05,
1801
+ "loss": 2.1992,
1802
+ "step": 2560
1803
+ },
1804
+ {
1805
+ "epoch": 1.2678835717809571,
1806
+ "grad_norm": 7.187436580657959,
1807
+ "learning_rate": 2.8973809915994072e-05,
1808
+ "loss": 1.9083,
1809
+ "step": 2570
1810
+ },
1811
+ {
1812
+ "epoch": 1.2728169708929453,
1813
+ "grad_norm": 8.493294715881348,
1814
+ "learning_rate": 2.8891451161258443e-05,
1815
+ "loss": 1.796,
1816
+ "step": 2580
1817
+ },
1818
+ {
1819
+ "epoch": 1.2777503700049335,
1820
+ "grad_norm": 10.225448608398438,
1821
+ "learning_rate": 2.8809092406522814e-05,
1822
+ "loss": 2.2556,
1823
+ "step": 2590
1824
+ },
1825
+ {
1826
+ "epoch": 1.2826837691169215,
1827
+ "grad_norm": 5.342253684997559,
1828
+ "learning_rate": 2.8726733651787185e-05,
1829
+ "loss": 1.8728,
1830
+ "step": 2600
1831
+ },
1832
+ {
1833
+ "epoch": 1.2876171682289097,
1834
+ "grad_norm": 6.041278839111328,
1835
+ "learning_rate": 2.8644374897051556e-05,
1836
+ "loss": 1.9532,
1837
+ "step": 2610
1838
+ },
1839
+ {
1840
+ "epoch": 1.292550567340898,
1841
+ "grad_norm": 6.2567853927612305,
1842
+ "learning_rate": 2.8562016142315927e-05,
1843
+ "loss": 2.0493,
1844
+ "step": 2620
1845
+ },
1846
+ {
1847
+ "epoch": 1.297483966452886,
1848
+ "grad_norm": 9.334773063659668,
1849
+ "learning_rate": 2.8479657387580298e-05,
1850
+ "loss": 1.573,
1851
+ "step": 2630
1852
+ },
1853
+ {
1854
+ "epoch": 1.302417365564874,
1855
+ "grad_norm": 13.7177734375,
1856
+ "learning_rate": 2.8397298632844672e-05,
1857
+ "loss": 2.2434,
1858
+ "step": 2640
1859
+ },
1860
+ {
1861
+ "epoch": 1.3073507646768623,
1862
+ "grad_norm": 11.203829765319824,
1863
+ "learning_rate": 2.8314939878109043e-05,
1864
+ "loss": 2.157,
1865
+ "step": 2650
1866
+ },
1867
+ {
1868
+ "epoch": 1.3122841637888505,
1869
+ "grad_norm": 7.4039483070373535,
1870
+ "learning_rate": 2.8232581123373414e-05,
1871
+ "loss": 2.0443,
1872
+ "step": 2660
1873
+ },
1874
+ {
1875
+ "epoch": 1.3172175629008387,
1876
+ "grad_norm": 4.927941799163818,
1877
+ "learning_rate": 2.8150222368637785e-05,
1878
+ "loss": 2.3465,
1879
+ "step": 2670
1880
+ },
1881
+ {
1882
+ "epoch": 1.322150962012827,
1883
+ "grad_norm": 11.956111907958984,
1884
+ "learning_rate": 2.8067863613902156e-05,
1885
+ "loss": 1.9842,
1886
+ "step": 2680
1887
+ },
1888
+ {
1889
+ "epoch": 1.327084361124815,
1890
+ "grad_norm": 5.877150535583496,
1891
+ "learning_rate": 2.7985504859166527e-05,
1892
+ "loss": 2.1889,
1893
+ "step": 2690
1894
+ },
1895
+ {
1896
+ "epoch": 1.3320177602368033,
1897
+ "grad_norm": 5.822342395782471,
1898
+ "learning_rate": 2.7903146104430905e-05,
1899
+ "loss": 1.6998,
1900
+ "step": 2700
1901
+ },
1902
+ {
1903
+ "epoch": 1.3369511593487913,
1904
+ "grad_norm": 5.760201930999756,
1905
+ "learning_rate": 2.7820787349695276e-05,
1906
+ "loss": 1.7595,
1907
+ "step": 2710
1908
+ },
1909
+ {
1910
+ "epoch": 1.3418845584607795,
1911
+ "grad_norm": 4.266313552856445,
1912
+ "learning_rate": 2.7738428594959647e-05,
1913
+ "loss": 1.4794,
1914
+ "step": 2720
1915
+ },
1916
+ {
1917
+ "epoch": 1.3468179575727677,
1918
+ "grad_norm": 9.27362060546875,
1919
+ "learning_rate": 2.7656069840224018e-05,
1920
+ "loss": 1.9813,
1921
+ "step": 2730
1922
+ },
1923
+ {
1924
+ "epoch": 1.3517513566847559,
1925
+ "grad_norm": 6.628077507019043,
1926
+ "learning_rate": 2.757371108548839e-05,
1927
+ "loss": 1.7057,
1928
+ "step": 2740
1929
+ },
1930
+ {
1931
+ "epoch": 1.3566847557967439,
1932
+ "grad_norm": 4.694432735443115,
1933
+ "learning_rate": 2.7491352330752764e-05,
1934
+ "loss": 1.8465,
1935
+ "step": 2750
1936
+ },
1937
+ {
1938
+ "epoch": 1.361618154908732,
1939
+ "grad_norm": 6.031017780303955,
1940
+ "learning_rate": 2.7408993576017135e-05,
1941
+ "loss": 1.2289,
1942
+ "step": 2760
1943
+ },
1944
+ {
1945
+ "epoch": 1.3665515540207203,
1946
+ "grad_norm": 5.344706058502197,
1947
+ "learning_rate": 2.7326634821281506e-05,
1948
+ "loss": 1.9155,
1949
+ "step": 2770
1950
+ },
1951
+ {
1952
+ "epoch": 1.3714849531327085,
1953
+ "grad_norm": 5.284564018249512,
1954
+ "learning_rate": 2.7244276066545877e-05,
1955
+ "loss": 1.7855,
1956
+ "step": 2780
1957
+ },
1958
+ {
1959
+ "epoch": 1.3764183522446967,
1960
+ "grad_norm": 4.38336181640625,
1961
+ "learning_rate": 2.7161917311810248e-05,
1962
+ "loss": 2.0217,
1963
+ "step": 2790
1964
+ },
1965
+ {
1966
+ "epoch": 1.3813517513566849,
1967
+ "grad_norm": 10.120343208312988,
1968
+ "learning_rate": 2.707955855707462e-05,
1969
+ "loss": 2.19,
1970
+ "step": 2800
1971
+ },
1972
+ {
1973
+ "epoch": 1.3862851504686728,
1974
+ "grad_norm": 6.627511024475098,
1975
+ "learning_rate": 2.699719980233899e-05,
1976
+ "loss": 1.9309,
1977
+ "step": 2810
1978
+ },
1979
+ {
1980
+ "epoch": 1.391218549580661,
1981
+ "grad_norm": 2.8276286125183105,
1982
+ "learning_rate": 2.691484104760336e-05,
1983
+ "loss": 1.8075,
1984
+ "step": 2820
1985
+ },
1986
+ {
1987
+ "epoch": 1.3961519486926492,
1988
+ "grad_norm": 10.60366153717041,
1989
+ "learning_rate": 2.6832482292867732e-05,
1990
+ "loss": 1.4447,
1991
+ "step": 2830
1992
+ },
1993
+ {
1994
+ "epoch": 1.4010853478046374,
1995
+ "grad_norm": 5.9414286613464355,
1996
+ "learning_rate": 2.6750123538132106e-05,
1997
+ "loss": 1.8544,
1998
+ "step": 2840
1999
+ },
2000
+ {
2001
+ "epoch": 1.4060187469166254,
2002
+ "grad_norm": 5.864882946014404,
2003
+ "learning_rate": 2.6667764783396477e-05,
2004
+ "loss": 2.1942,
2005
+ "step": 2850
2006
+ },
2007
+ {
2008
+ "epoch": 1.4109521460286136,
2009
+ "grad_norm": 5.399135112762451,
2010
+ "learning_rate": 2.6585406028660848e-05,
2011
+ "loss": 1.9905,
2012
+ "step": 2860
2013
+ },
2014
+ {
2015
+ "epoch": 1.4158855451406018,
2016
+ "grad_norm": 4.467281341552734,
2017
+ "learning_rate": 2.650304727392522e-05,
2018
+ "loss": 1.7024,
2019
+ "step": 2870
2020
+ },
2021
+ {
2022
+ "epoch": 1.42081894425259,
2023
+ "grad_norm": 7.034926414489746,
2024
+ "learning_rate": 2.642068851918959e-05,
2025
+ "loss": 1.8154,
2026
+ "step": 2880
2027
+ },
2028
+ {
2029
+ "epoch": 1.4257523433645782,
2030
+ "grad_norm": 4.346402168273926,
2031
+ "learning_rate": 2.633832976445396e-05,
2032
+ "loss": 1.7937,
2033
+ "step": 2890
2034
+ },
2035
+ {
2036
+ "epoch": 1.4306857424765664,
2037
+ "grad_norm": 9.743393898010254,
2038
+ "learning_rate": 2.6255971009718332e-05,
2039
+ "loss": 1.7868,
2040
+ "step": 2900
2041
+ },
2042
+ {
2043
+ "epoch": 1.4356191415885546,
2044
+ "grad_norm": 7.675935745239258,
2045
+ "learning_rate": 2.6173612254982703e-05,
2046
+ "loss": 1.8108,
2047
+ "step": 2910
2048
+ },
2049
+ {
2050
+ "epoch": 1.4405525407005426,
2051
+ "grad_norm": 5.96721076965332,
2052
+ "learning_rate": 2.6091253500247074e-05,
2053
+ "loss": 1.8259,
2054
+ "step": 2920
2055
+ },
2056
+ {
2057
+ "epoch": 1.4454859398125308,
2058
+ "grad_norm": 6.192174911499023,
2059
+ "learning_rate": 2.600889474551145e-05,
2060
+ "loss": 1.7254,
2061
+ "step": 2930
2062
+ },
2063
+ {
2064
+ "epoch": 1.450419338924519,
2065
+ "grad_norm": 7.611389636993408,
2066
+ "learning_rate": 2.592653599077582e-05,
2067
+ "loss": 2.1246,
2068
+ "step": 2940
2069
+ },
2070
+ {
2071
+ "epoch": 1.4553527380365072,
2072
+ "grad_norm": 8.755757331848145,
2073
+ "learning_rate": 2.584417723604019e-05,
2074
+ "loss": 1.2897,
2075
+ "step": 2950
2076
+ },
2077
+ {
2078
+ "epoch": 1.4602861371484952,
2079
+ "grad_norm": 17.625520706176758,
2080
+ "learning_rate": 2.5761818481304562e-05,
2081
+ "loss": 1.8039,
2082
+ "step": 2960
2083
+ },
2084
+ {
2085
+ "epoch": 1.4652195362604834,
2086
+ "grad_norm": 6.4828057289123535,
2087
+ "learning_rate": 2.5679459726568933e-05,
2088
+ "loss": 1.7734,
2089
+ "step": 2970
2090
+ },
2091
+ {
2092
+ "epoch": 1.4701529353724716,
2093
+ "grad_norm": 4.072720527648926,
2094
+ "learning_rate": 2.5597100971833304e-05,
2095
+ "loss": 2.1754,
2096
+ "step": 2980
2097
+ },
2098
+ {
2099
+ "epoch": 1.4750863344844598,
2100
+ "grad_norm": 5.953226089477539,
2101
+ "learning_rate": 2.551474221709768e-05,
2102
+ "loss": 1.7654,
2103
+ "step": 2990
2104
+ },
2105
+ {
2106
+ "epoch": 1.480019733596448,
2107
+ "grad_norm": 5.490500450134277,
2108
+ "learning_rate": 2.5432383462362053e-05,
2109
+ "loss": 1.9371,
2110
+ "step": 3000
2111
+ },
2112
+ {
2113
+ "epoch": 1.4849531327084362,
2114
+ "grad_norm": 5.547089576721191,
2115
+ "learning_rate": 2.5350024707626424e-05,
2116
+ "loss": 1.8543,
2117
+ "step": 3010
2118
+ },
2119
+ {
2120
+ "epoch": 1.4898865318204244,
2121
+ "grad_norm": 7.270637512207031,
2122
+ "learning_rate": 2.5267665952890795e-05,
2123
+ "loss": 1.6728,
2124
+ "step": 3020
2125
+ },
2126
+ {
2127
+ "epoch": 1.4948199309324124,
2128
+ "grad_norm": 10.142009735107422,
2129
+ "learning_rate": 2.5185307198155166e-05,
2130
+ "loss": 1.7656,
2131
+ "step": 3030
2132
+ },
2133
+ {
2134
+ "epoch": 1.4997533300444006,
2135
+ "grad_norm": 11.43400764465332,
2136
+ "learning_rate": 2.510294844341954e-05,
2137
+ "loss": 1.7731,
2138
+ "step": 3040
2139
+ },
2140
+ {
2141
+ "epoch": 1.5046867291563888,
2142
+ "grad_norm": 8.225043296813965,
2143
+ "learning_rate": 2.502058968868391e-05,
2144
+ "loss": 2.0266,
2145
+ "step": 3050
2146
+ },
2147
+ {
2148
+ "epoch": 1.5096201282683768,
2149
+ "grad_norm": 8.20263385772705,
2150
+ "learning_rate": 2.4938230933948282e-05,
2151
+ "loss": 1.9749,
2152
+ "step": 3060
2153
+ },
2154
+ {
2155
+ "epoch": 1.514553527380365,
2156
+ "grad_norm": 8.855592727661133,
2157
+ "learning_rate": 2.4855872179212653e-05,
2158
+ "loss": 1.9333,
2159
+ "step": 3070
2160
+ },
2161
+ {
2162
+ "epoch": 1.5194869264923532,
2163
+ "grad_norm": 5.7763752937316895,
2164
+ "learning_rate": 2.4773513424477024e-05,
2165
+ "loss": 1.5184,
2166
+ "step": 3080
2167
+ },
2168
+ {
2169
+ "epoch": 1.5244203256043414,
2170
+ "grad_norm": 8.338578224182129,
2171
+ "learning_rate": 2.4691154669741395e-05,
2172
+ "loss": 1.9215,
2173
+ "step": 3090
2174
+ },
2175
+ {
2176
+ "epoch": 1.5293537247163296,
2177
+ "grad_norm": 5.367077350616455,
2178
+ "learning_rate": 2.4608795915005766e-05,
2179
+ "loss": 1.973,
2180
+ "step": 3100
2181
+ },
2182
+ {
2183
+ "epoch": 1.5342871238283178,
2184
+ "grad_norm": 9.6682710647583,
2185
+ "learning_rate": 2.4526437160270137e-05,
2186
+ "loss": 1.6508,
2187
+ "step": 3110
2188
+ },
2189
+ {
2190
+ "epoch": 1.539220522940306,
2191
+ "grad_norm": 5.364202499389648,
2192
+ "learning_rate": 2.4444078405534508e-05,
2193
+ "loss": 1.9087,
2194
+ "step": 3120
2195
+ },
2196
+ {
2197
+ "epoch": 1.5441539220522942,
2198
+ "grad_norm": 9.031702995300293,
2199
+ "learning_rate": 2.4361719650798882e-05,
2200
+ "loss": 1.759,
2201
+ "step": 3130
2202
+ },
2203
+ {
2204
+ "epoch": 1.5490873211642822,
2205
+ "grad_norm": 6.895035743713379,
2206
+ "learning_rate": 2.4279360896063253e-05,
2207
+ "loss": 1.7942,
2208
+ "step": 3140
2209
+ },
2210
+ {
2211
+ "epoch": 1.5540207202762704,
2212
+ "grad_norm": 7.573614120483398,
2213
+ "learning_rate": 2.4197002141327624e-05,
2214
+ "loss": 2.0408,
2215
+ "step": 3150
2216
+ },
2217
+ {
2218
+ "epoch": 1.5589541193882586,
2219
+ "grad_norm": 14.733696937561035,
2220
+ "learning_rate": 2.4114643386591996e-05,
2221
+ "loss": 1.748,
2222
+ "step": 3160
2223
+ },
2224
+ {
2225
+ "epoch": 1.5638875185002465,
2226
+ "grad_norm": 7.862844944000244,
2227
+ "learning_rate": 2.4032284631856367e-05,
2228
+ "loss": 1.8507,
2229
+ "step": 3170
2230
+ },
2231
+ {
2232
+ "epoch": 1.5688209176122347,
2233
+ "grad_norm": 4.0979390144348145,
2234
+ "learning_rate": 2.3949925877120738e-05,
2235
+ "loss": 2.0245,
2236
+ "step": 3180
2237
+ },
2238
+ {
2239
+ "epoch": 1.573754316724223,
2240
+ "grad_norm": 6.260437965393066,
2241
+ "learning_rate": 2.386756712238511e-05,
2242
+ "loss": 1.7502,
2243
+ "step": 3190
2244
+ },
2245
+ {
2246
+ "epoch": 1.5786877158362111,
2247
+ "grad_norm": 8.188227653503418,
2248
+ "learning_rate": 2.3785208367649483e-05,
2249
+ "loss": 1.959,
2250
+ "step": 3200
2251
+ },
2252
+ {
2253
+ "epoch": 1.5836211149481993,
2254
+ "grad_norm": 5.43758487701416,
2255
+ "learning_rate": 2.3702849612913854e-05,
2256
+ "loss": 1.7763,
2257
+ "step": 3210
2258
+ },
2259
+ {
2260
+ "epoch": 1.5885545140601876,
2261
+ "grad_norm": 5.603811740875244,
2262
+ "learning_rate": 2.3620490858178225e-05,
2263
+ "loss": 1.7247,
2264
+ "step": 3220
2265
+ },
2266
+ {
2267
+ "epoch": 1.5934879131721758,
2268
+ "grad_norm": 4.974462509155273,
2269
+ "learning_rate": 2.35381321034426e-05,
2270
+ "loss": 1.8819,
2271
+ "step": 3230
2272
+ },
2273
+ {
2274
+ "epoch": 1.598421312284164,
2275
+ "grad_norm": 5.835883617401123,
2276
+ "learning_rate": 2.345577334870697e-05,
2277
+ "loss": 1.9156,
2278
+ "step": 3240
2279
+ },
2280
+ {
2281
+ "epoch": 1.603354711396152,
2282
+ "grad_norm": 10.42984676361084,
2283
+ "learning_rate": 2.337341459397134e-05,
2284
+ "loss": 1.6488,
2285
+ "step": 3250
2286
+ },
2287
+ {
2288
+ "epoch": 1.6082881105081401,
2289
+ "grad_norm": 10.849640846252441,
2290
+ "learning_rate": 2.3291055839235712e-05,
2291
+ "loss": 1.9802,
2292
+ "step": 3260
2293
+ },
2294
+ {
2295
+ "epoch": 1.6132215096201281,
2296
+ "grad_norm": 6.4310784339904785,
2297
+ "learning_rate": 2.3208697084500083e-05,
2298
+ "loss": 1.9685,
2299
+ "step": 3270
2300
+ },
2301
+ {
2302
+ "epoch": 1.6181549087321163,
2303
+ "grad_norm": 4.533667087554932,
2304
+ "learning_rate": 2.3126338329764454e-05,
2305
+ "loss": 1.5295,
2306
+ "step": 3280
2307
+ },
2308
+ {
2309
+ "epoch": 1.6230883078441045,
2310
+ "grad_norm": 7.374139785766602,
2311
+ "learning_rate": 2.3043979575028825e-05,
2312
+ "loss": 1.8251,
2313
+ "step": 3290
2314
+ },
2315
+ {
2316
+ "epoch": 1.6280217069560927,
2317
+ "grad_norm": 7.097417831420898,
2318
+ "learning_rate": 2.2961620820293196e-05,
2319
+ "loss": 1.7943,
2320
+ "step": 3300
2321
+ },
2322
+ {
2323
+ "epoch": 1.632955106068081,
2324
+ "grad_norm": 9.811117172241211,
2325
+ "learning_rate": 2.2879262065557567e-05,
2326
+ "loss": 1.5591,
2327
+ "step": 3310
2328
+ },
2329
+ {
2330
+ "epoch": 1.6378885051800691,
2331
+ "grad_norm": 2.922159194946289,
2332
+ "learning_rate": 2.2796903310821942e-05,
2333
+ "loss": 1.5074,
2334
+ "step": 3320
2335
+ },
2336
+ {
2337
+ "epoch": 1.6428219042920573,
2338
+ "grad_norm": 5.961956977844238,
2339
+ "learning_rate": 2.2714544556086313e-05,
2340
+ "loss": 2.1771,
2341
+ "step": 3330
2342
+ },
2343
+ {
2344
+ "epoch": 1.6477553034040455,
2345
+ "grad_norm": 6.768845558166504,
2346
+ "learning_rate": 2.2632185801350687e-05,
2347
+ "loss": 1.7893,
2348
+ "step": 3340
2349
+ },
2350
+ {
2351
+ "epoch": 1.6526887025160335,
2352
+ "grad_norm": 5.665631294250488,
2353
+ "learning_rate": 2.2549827046615058e-05,
2354
+ "loss": 1.8398,
2355
+ "step": 3350
2356
+ },
2357
+ {
2358
+ "epoch": 1.6576221016280217,
2359
+ "grad_norm": 8.640176773071289,
2360
+ "learning_rate": 2.246746829187943e-05,
2361
+ "loss": 1.9559,
2362
+ "step": 3360
2363
+ },
2364
+ {
2365
+ "epoch": 1.66255550074001,
2366
+ "grad_norm": 6.773179531097412,
2367
+ "learning_rate": 2.23851095371438e-05,
2368
+ "loss": 1.7461,
2369
+ "step": 3370
2370
+ },
2371
+ {
2372
+ "epoch": 1.6674888998519979,
2373
+ "grad_norm": 5.100921630859375,
2374
+ "learning_rate": 2.230275078240817e-05,
2375
+ "loss": 1.9243,
2376
+ "step": 3380
2377
+ },
2378
+ {
2379
+ "epoch": 1.672422298963986,
2380
+ "grad_norm": 8.315923690795898,
2381
+ "learning_rate": 2.2220392027672542e-05,
2382
+ "loss": 1.8641,
2383
+ "step": 3390
2384
+ },
2385
+ {
2386
+ "epoch": 1.6773556980759743,
2387
+ "grad_norm": 6.231566429138184,
2388
+ "learning_rate": 2.2138033272936913e-05,
2389
+ "loss": 1.9068,
2390
+ "step": 3400
2391
+ },
2392
+ {
2393
+ "epoch": 1.6822890971879625,
2394
+ "grad_norm": 4.701402187347412,
2395
+ "learning_rate": 2.2055674518201284e-05,
2396
+ "loss": 1.9061,
2397
+ "step": 3410
2398
+ },
2399
+ {
2400
+ "epoch": 1.6872224962999507,
2401
+ "grad_norm": 5.9488325119018555,
2402
+ "learning_rate": 2.197331576346566e-05,
2403
+ "loss": 2.0304,
2404
+ "step": 3420
2405
+ },
2406
+ {
2407
+ "epoch": 1.692155895411939,
2408
+ "grad_norm": 9.257286071777344,
2409
+ "learning_rate": 2.189095700873003e-05,
2410
+ "loss": 1.8544,
2411
+ "step": 3430
2412
+ },
2413
+ {
2414
+ "epoch": 1.697089294523927,
2415
+ "grad_norm": 6.57626485824585,
2416
+ "learning_rate": 2.18085982539944e-05,
2417
+ "loss": 1.9199,
2418
+ "step": 3440
2419
+ },
2420
+ {
2421
+ "epoch": 1.7020226936359153,
2422
+ "grad_norm": 6.381181240081787,
2423
+ "learning_rate": 2.1726239499258772e-05,
2424
+ "loss": 2.0876,
2425
+ "step": 3450
2426
+ },
2427
+ {
2428
+ "epoch": 1.7069560927479033,
2429
+ "grad_norm": 3.832770586013794,
2430
+ "learning_rate": 2.1643880744523143e-05,
2431
+ "loss": 1.7948,
2432
+ "step": 3460
2433
+ },
2434
+ {
2435
+ "epoch": 1.7118894918598915,
2436
+ "grad_norm": 20.409229278564453,
2437
+ "learning_rate": 2.1561521989787514e-05,
2438
+ "loss": 1.9628,
2439
+ "step": 3470
2440
+ },
2441
+ {
2442
+ "epoch": 1.7168228909718797,
2443
+ "grad_norm": 12.321199417114258,
2444
+ "learning_rate": 2.1479163235051885e-05,
2445
+ "loss": 2.149,
2446
+ "step": 3480
2447
+ },
2448
+ {
2449
+ "epoch": 1.7217562900838677,
2450
+ "grad_norm": 7.48840856552124,
2451
+ "learning_rate": 2.139680448031626e-05,
2452
+ "loss": 1.6974,
2453
+ "step": 3490
2454
+ },
2455
+ {
2456
+ "epoch": 1.7266896891958559,
2457
+ "grad_norm": 10.339478492736816,
2458
+ "learning_rate": 2.131444572558063e-05,
2459
+ "loss": 2.0896,
2460
+ "step": 3500
2461
+ },
2462
+ {
2463
+ "epoch": 1.731623088307844,
2464
+ "grad_norm": 8.04315185546875,
2465
+ "learning_rate": 2.1232086970845e-05,
2466
+ "loss": 1.6795,
2467
+ "step": 3510
2468
+ },
2469
+ {
2470
+ "epoch": 1.7365564874198323,
2471
+ "grad_norm": 9.33163070678711,
2472
+ "learning_rate": 2.1149728216109376e-05,
2473
+ "loss": 1.7942,
2474
+ "step": 3520
2475
+ },
2476
+ {
2477
+ "epoch": 1.7414898865318205,
2478
+ "grad_norm": 11.18399715423584,
2479
+ "learning_rate": 2.1067369461373747e-05,
2480
+ "loss": 1.4243,
2481
+ "step": 3530
2482
+ },
2483
+ {
2484
+ "epoch": 1.7464232856438087,
2485
+ "grad_norm": 9.912873268127441,
2486
+ "learning_rate": 2.0985010706638118e-05,
2487
+ "loss": 1.5473,
2488
+ "step": 3540
2489
+ },
2490
+ {
2491
+ "epoch": 1.7513566847557969,
2492
+ "grad_norm": 8.763083457946777,
2493
+ "learning_rate": 2.090265195190249e-05,
2494
+ "loss": 1.8,
2495
+ "step": 3550
2496
+ },
2497
+ {
2498
+ "epoch": 1.756290083867785,
2499
+ "grad_norm": 8.042789459228516,
2500
+ "learning_rate": 2.082029319716686e-05,
2501
+ "loss": 1.9486,
2502
+ "step": 3560
2503
+ },
2504
+ {
2505
+ "epoch": 1.761223482979773,
2506
+ "grad_norm": 5.028835296630859,
2507
+ "learning_rate": 2.073793444243123e-05,
2508
+ "loss": 1.8425,
2509
+ "step": 3570
2510
+ },
2511
+ {
2512
+ "epoch": 1.7661568820917612,
2513
+ "grad_norm": 5.263244152069092,
2514
+ "learning_rate": 2.0655575687695602e-05,
2515
+ "loss": 2.3425,
2516
+ "step": 3580
2517
+ },
2518
+ {
2519
+ "epoch": 1.7710902812037492,
2520
+ "grad_norm": 6.808823585510254,
2521
+ "learning_rate": 2.0573216932959973e-05,
2522
+ "loss": 1.7344,
2523
+ "step": 3590
2524
+ },
2525
+ {
2526
+ "epoch": 1.7760236803157374,
2527
+ "grad_norm": 7.849328994750977,
2528
+ "learning_rate": 2.0490858178224344e-05,
2529
+ "loss": 1.9272,
2530
+ "step": 3600
2531
+ },
2532
+ {
2533
+ "epoch": 1.7809570794277256,
2534
+ "grad_norm": 5.156803131103516,
2535
+ "learning_rate": 2.0408499423488718e-05,
2536
+ "loss": 1.9647,
2537
+ "step": 3610
2538
+ },
2539
+ {
2540
+ "epoch": 1.7858904785397138,
2541
+ "grad_norm": 8.30761432647705,
2542
+ "learning_rate": 2.032614066875309e-05,
2543
+ "loss": 1.5823,
2544
+ "step": 3620
2545
+ },
2546
+ {
2547
+ "epoch": 1.790823877651702,
2548
+ "grad_norm": 9.06478214263916,
2549
+ "learning_rate": 2.0243781914017464e-05,
2550
+ "loss": 2.03,
2551
+ "step": 3630
2552
+ },
2553
+ {
2554
+ "epoch": 1.7957572767636902,
2555
+ "grad_norm": 13.680024147033691,
2556
+ "learning_rate": 2.0161423159281835e-05,
2557
+ "loss": 1.9991,
2558
+ "step": 3640
2559
+ },
2560
+ {
2561
+ "epoch": 1.8006906758756784,
2562
+ "grad_norm": 6.494818210601807,
2563
+ "learning_rate": 2.0079064404546206e-05,
2564
+ "loss": 1.8554,
2565
+ "step": 3650
2566
+ },
2567
+ {
2568
+ "epoch": 1.8056240749876666,
2569
+ "grad_norm": 7.450560092926025,
2570
+ "learning_rate": 1.9996705649810577e-05,
2571
+ "loss": 1.9436,
2572
+ "step": 3660
2573
+ },
2574
+ {
2575
+ "epoch": 1.8105574740996546,
2576
+ "grad_norm": 6.066117763519287,
2577
+ "learning_rate": 1.9914346895074948e-05,
2578
+ "loss": 1.7343,
2579
+ "step": 3670
2580
+ },
2581
+ {
2582
+ "epoch": 1.8154908732116428,
2583
+ "grad_norm": 4.272149562835693,
2584
+ "learning_rate": 1.983198814033932e-05,
2585
+ "loss": 1.8145,
2586
+ "step": 3680
2587
+ },
2588
+ {
2589
+ "epoch": 1.820424272323631,
2590
+ "grad_norm": 7.599696636199951,
2591
+ "learning_rate": 1.974962938560369e-05,
2592
+ "loss": 1.9128,
2593
+ "step": 3690
2594
+ },
2595
+ {
2596
+ "epoch": 1.825357671435619,
2597
+ "grad_norm": 8.375687599182129,
2598
+ "learning_rate": 1.966727063086806e-05,
2599
+ "loss": 1.568,
2600
+ "step": 3700
2601
+ },
2602
+ {
2603
+ "epoch": 1.8302910705476072,
2604
+ "grad_norm": 12.533012390136719,
2605
+ "learning_rate": 1.9584911876132435e-05,
2606
+ "loss": 1.7363,
2607
+ "step": 3710
2608
+ },
2609
+ {
2610
+ "epoch": 1.8352244696595954,
2611
+ "grad_norm": 7.711116313934326,
2612
+ "learning_rate": 1.9502553121396806e-05,
2613
+ "loss": 1.9017,
2614
+ "step": 3720
2615
+ },
2616
+ {
2617
+ "epoch": 1.8401578687715836,
2618
+ "grad_norm": 4.572812080383301,
2619
+ "learning_rate": 1.9420194366661177e-05,
2620
+ "loss": 1.5797,
2621
+ "step": 3730
2622
+ },
2623
+ {
2624
+ "epoch": 1.8450912678835718,
2625
+ "grad_norm": 4.474880695343018,
2626
+ "learning_rate": 1.9337835611925548e-05,
2627
+ "loss": 2.0278,
2628
+ "step": 3740
2629
+ },
2630
+ {
2631
+ "epoch": 1.85002466699556,
2632
+ "grad_norm": 5.497013092041016,
2633
+ "learning_rate": 1.925547685718992e-05,
2634
+ "loss": 1.757,
2635
+ "step": 3750
2636
+ },
2637
+ {
2638
+ "epoch": 1.8549580661075482,
2639
+ "grad_norm": 8.392544746398926,
2640
+ "learning_rate": 1.917311810245429e-05,
2641
+ "loss": 1.6158,
2642
+ "step": 3760
2643
+ },
2644
+ {
2645
+ "epoch": 1.8598914652195364,
2646
+ "grad_norm": 8.011329650878906,
2647
+ "learning_rate": 1.9090759347718665e-05,
2648
+ "loss": 1.8829,
2649
+ "step": 3770
2650
+ },
2651
+ {
2652
+ "epoch": 1.8648248643315244,
2653
+ "grad_norm": 6.119569778442383,
2654
+ "learning_rate": 1.9008400592983036e-05,
2655
+ "loss": 2.5915,
2656
+ "step": 3780
2657
+ },
2658
+ {
2659
+ "epoch": 1.8697582634435126,
2660
+ "grad_norm": 4.356236934661865,
2661
+ "learning_rate": 1.8926041838247407e-05,
2662
+ "loss": 1.7601,
2663
+ "step": 3790
2664
+ },
2665
+ {
2666
+ "epoch": 1.8746916625555008,
2667
+ "grad_norm": 3.654446601867676,
2668
+ "learning_rate": 1.8843683083511778e-05,
2669
+ "loss": 1.5618,
2670
+ "step": 3800
2671
+ },
2672
+ {
2673
+ "epoch": 1.8796250616674888,
2674
+ "grad_norm": 10.40364933013916,
2675
+ "learning_rate": 1.8761324328776152e-05,
2676
+ "loss": 2.1173,
2677
+ "step": 3810
2678
+ },
2679
+ {
2680
+ "epoch": 1.884558460779477,
2681
+ "grad_norm": 11.642656326293945,
2682
+ "learning_rate": 1.8678965574040523e-05,
2683
+ "loss": 1.9532,
2684
+ "step": 3820
2685
+ },
2686
+ {
2687
+ "epoch": 1.8894918598914652,
2688
+ "grad_norm": 7.247965335845947,
2689
+ "learning_rate": 1.8596606819304894e-05,
2690
+ "loss": 1.8573,
2691
+ "step": 3830
2692
+ },
2693
+ {
2694
+ "epoch": 1.8944252590034534,
2695
+ "grad_norm": 8.925628662109375,
2696
+ "learning_rate": 1.8514248064569265e-05,
2697
+ "loss": 1.6098,
2698
+ "step": 3840
2699
+ },
2700
+ {
2701
+ "epoch": 1.8993586581154416,
2702
+ "grad_norm": 6.515012264251709,
2703
+ "learning_rate": 1.8431889309833636e-05,
2704
+ "loss": 1.7684,
2705
+ "step": 3850
2706
+ },
2707
+ {
2708
+ "epoch": 1.9042920572274298,
2709
+ "grad_norm": 9.089420318603516,
2710
+ "learning_rate": 1.8349530555098007e-05,
2711
+ "loss": 2.1597,
2712
+ "step": 3860
2713
+ },
2714
+ {
2715
+ "epoch": 1.909225456339418,
2716
+ "grad_norm": 6.92453670501709,
2717
+ "learning_rate": 1.8267171800362378e-05,
2718
+ "loss": 1.8394,
2719
+ "step": 3870
2720
+ },
2721
+ {
2722
+ "epoch": 1.9141588554514062,
2723
+ "grad_norm": 7.409922122955322,
2724
+ "learning_rate": 1.818481304562675e-05,
2725
+ "loss": 1.8576,
2726
+ "step": 3880
2727
+ },
2728
+ {
2729
+ "epoch": 1.9190922545633942,
2730
+ "grad_norm": 5.864316463470459,
2731
+ "learning_rate": 1.810245429089112e-05,
2732
+ "loss": 1.7566,
2733
+ "step": 3890
2734
+ },
2735
+ {
2736
+ "epoch": 1.9240256536753824,
2737
+ "grad_norm": 7.16778039932251,
2738
+ "learning_rate": 1.8020095536155494e-05,
2739
+ "loss": 1.7765,
2740
+ "step": 3900
2741
+ },
2742
+ {
2743
+ "epoch": 1.9289590527873703,
2744
+ "grad_norm": 7.447871208190918,
2745
+ "learning_rate": 1.7937736781419865e-05,
2746
+ "loss": 1.5466,
2747
+ "step": 3910
2748
+ },
2749
+ {
2750
+ "epoch": 1.9338924518993585,
2751
+ "grad_norm": 8.456513404846191,
2752
+ "learning_rate": 1.785537802668424e-05,
2753
+ "loss": 1.6053,
2754
+ "step": 3920
2755
+ },
2756
+ {
2757
+ "epoch": 1.9388258510113467,
2758
+ "grad_norm": 5.673243045806885,
2759
+ "learning_rate": 1.777301927194861e-05,
2760
+ "loss": 1.6813,
2761
+ "step": 3930
2762
+ },
2763
+ {
2764
+ "epoch": 1.943759250123335,
2765
+ "grad_norm": 13.05078125,
2766
+ "learning_rate": 1.7690660517212982e-05,
2767
+ "loss": 1.8932,
2768
+ "step": 3940
2769
+ },
2770
+ {
2771
+ "epoch": 1.9486926492353231,
2772
+ "grad_norm": 5.095945358276367,
2773
+ "learning_rate": 1.7608301762477353e-05,
2774
+ "loss": 1.9467,
2775
+ "step": 3950
2776
+ },
2777
+ {
2778
+ "epoch": 1.9536260483473114,
2779
+ "grad_norm": 7.766141414642334,
2780
+ "learning_rate": 1.7525943007741724e-05,
2781
+ "loss": 1.9318,
2782
+ "step": 3960
2783
+ },
2784
+ {
2785
+ "epoch": 1.9585594474592996,
2786
+ "grad_norm": 6.2664690017700195,
2787
+ "learning_rate": 1.7443584253006095e-05,
2788
+ "loss": 1.5792,
2789
+ "step": 3970
2790
+ },
2791
+ {
2792
+ "epoch": 1.9634928465712878,
2793
+ "grad_norm": 9.3766508102417,
2794
+ "learning_rate": 1.7361225498270466e-05,
2795
+ "loss": 1.7359,
2796
+ "step": 3980
2797
+ },
2798
+ {
2799
+ "epoch": 1.9684262456832757,
2800
+ "grad_norm": 8.860836029052734,
2801
+ "learning_rate": 1.7278866743534837e-05,
2802
+ "loss": 1.7544,
2803
+ "step": 3990
2804
+ },
2805
+ {
2806
+ "epoch": 1.973359644795264,
2807
+ "grad_norm": 6.054210662841797,
2808
+ "learning_rate": 1.719650798879921e-05,
2809
+ "loss": 1.5618,
2810
+ "step": 4000
2811
+ },
2812
+ {
2813
+ "epoch": 1.9782930439072521,
2814
+ "grad_norm": 11.120077133178711,
2815
+ "learning_rate": 1.7114149234063582e-05,
2816
+ "loss": 1.8174,
2817
+ "step": 4010
2818
+ },
2819
+ {
2820
+ "epoch": 1.9832264430192401,
2821
+ "grad_norm": 6.720516681671143,
2822
+ "learning_rate": 1.7031790479327953e-05,
2823
+ "loss": 1.613,
2824
+ "step": 4020
2825
+ },
2826
+ {
2827
+ "epoch": 1.9881598421312283,
2828
+ "grad_norm": 6.063734531402588,
2829
+ "learning_rate": 1.6949431724592324e-05,
2830
+ "loss": 1.9363,
2831
+ "step": 4030
2832
+ },
2833
+ {
2834
+ "epoch": 1.9930932412432165,
2835
+ "grad_norm": 5.5154523849487305,
2836
+ "learning_rate": 1.6867072969856695e-05,
2837
+ "loss": 1.6523,
2838
+ "step": 4040
2839
+ },
2840
+ {
2841
+ "epoch": 1.9980266403552047,
2842
+ "grad_norm": 10.039024353027344,
2843
+ "learning_rate": 1.6784714215121066e-05,
2844
+ "loss": 1.8938,
2845
+ "step": 4050
2846
+ }
2847
+ ],
2848
+ "logging_steps": 10,
2849
+ "max_steps": 6081,
2850
+ "num_input_tokens_seen": 0,
2851
+ "num_train_epochs": 3,
2852
+ "save_steps": 500,
2853
+ "stateful_callbacks": {
2854
+ "TrainerControl": {
2855
+ "args": {
2856
+ "should_epoch_stop": false,
2857
+ "should_evaluate": false,
2858
+ "should_log": false,
2859
+ "should_save": true,
2860
+ "should_training_stop": false
2861
+ },
2862
+ "attributes": {}
2863
+ }
2864
+ },
2865
+ "total_flos": 1.5059810462466048e+16,
2866
+ "train_batch_size": 1,
2867
+ "trial_name": null,
2868
+ "trial_params": null
2869
+ }
fine_tuned_model/checkpoint-4054/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:37bd4209646599a5612b713c1e2ec16f0b3ad2d6bbec8a9566a22a9bc23ae4c3
3
+ size 5713
fine_tuned_model/checkpoint-4054/vocab.json ADDED
The diff for this file is too large to render. See raw diff
 
fine_tuned_model/checkpoint-6081/config.json ADDED
@@ -0,0 +1,40 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "activation_function": "gelu_new",
3
+ "architectures": [
4
+ "GPT2LMHeadModel"
5
+ ],
6
+ "attn_pdrop": 0.1,
7
+ "bos_token_id": 50256,
8
+ "embd_pdrop": 0.1,
9
+ "eos_token_id": 50256,
10
+ "initializer_range": 0.02,
11
+ "layer_norm_epsilon": 1e-05,
12
+ "model_type": "gpt2",
13
+ "n_ctx": 1024,
14
+ "n_embd": 1024,
15
+ "n_head": 16,
16
+ "n_inner": null,
17
+ "n_layer": 24,
18
+ "n_positions": 1024,
19
+ "n_special": 0,
20
+ "predict_special_tokens": true,
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,
27
+ "summary_proj_to_labels": true,
28
+ "summary_type": "cls_index",
29
+ "summary_use_proj": true,
30
+ "task_specific_params": {
31
+ "text-generation": {
32
+ "do_sample": true,
33
+ "max_length": 50
34
+ }
35
+ },
36
+ "torch_dtype": "float32",
37
+ "transformers_version": "4.51.3",
38
+ "use_cache": true,
39
+ "vocab_size": 50257
40
+ }
fine_tuned_model/checkpoint-6081/generation_config.json ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 50256,
4
+ "eos_token_id": 50256,
5
+ "transformers_version": "4.51.3"
6
+ }
fine_tuned_model/checkpoint-6081/merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
fine_tuned_model/checkpoint-6081/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:162664b502c33c6c04c5776498b5ad474fe88e760bd08e54a8522cc34cb75783
3
+ size 1419322880
fine_tuned_model/checkpoint-6081/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:75045fe6bf0ca885fbd9725126b28b7daffa52fbd333e967efbab668a47faba8
3
+ size 2838829707
fine_tuned_model/checkpoint-6081/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:aaa63fd256a5f01aeb345d3ad52c158a7976dd34cea5db2e9a7731fb389a0dae
3
+ size 14645
fine_tuned_model/checkpoint-6081/scaler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:651382b0a958510091a8d0a09db80a458d157f97a432c7e82225934dfc51b2ef
3
+ size 1383
fine_tuned_model/checkpoint-6081/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:58537a0cdf22ec7a35e0ce5c51dbd4f927a5fa170abeb3c0ada3851788eb069f
3
+ size 1465
fine_tuned_model/checkpoint-6081/special_tokens_map.json ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "<|endoftext|>",
3
+ "eos_token": "<|endoftext|>",
4
+ "pad_token": "<|endoftext|>",
5
+ "unk_token": "<|endoftext|>"
6
+ }
fine_tuned_model/checkpoint-6081/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
fine_tuned_model/checkpoint-6081/tokenizer_config.json ADDED
@@ -0,0 +1,21 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": false,
3
+ "added_tokens_decoder": {
4
+ "50256": {
5
+ "content": "<|endoftext|>",
6
+ "lstrip": false,
7
+ "normalized": true,
8
+ "rstrip": false,
9
+ "single_word": false,
10
+ "special": true
11
+ }
12
+ },
13
+ "bos_token": "<|endoftext|>",
14
+ "clean_up_tokenization_spaces": false,
15
+ "eos_token": "<|endoftext|>",
16
+ "extra_special_tokens": {},
17
+ "model_max_length": 1024,
18
+ "pad_token": "<|endoftext|>",
19
+ "tokenizer_class": "GPT2Tokenizer",
20
+ "unk_token": "<|endoftext|>"
21
+ }
fine_tuned_model/checkpoint-6081/trainer_state.json ADDED
The diff for this file is too large to render. See raw diff
 
fine_tuned_model/checkpoint-6081/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:37bd4209646599a5612b713c1e2ec16f0b3ad2d6bbec8a9566a22a9bc23ae4c3
3
+ size 5713
fine_tuned_model/checkpoint-6081/vocab.json ADDED
The diff for this file is too large to render. See raw diff
 
fine_tuned_model/config.json CHANGED
@@ -11,11 +11,13 @@
11
  "layer_norm_epsilon": 1e-05,
12
  "model_type": "gpt2",
13
  "n_ctx": 1024,
14
- "n_embd": 768,
15
- "n_head": 12,
16
  "n_inner": null,
17
- "n_layer": 12,
18
  "n_positions": 1024,
 
 
19
  "reorder_and_upcast_attn": false,
20
  "resid_pdrop": 0.1,
21
  "scale_attn_by_inverse_layer_idx": false,
 
11
  "layer_norm_epsilon": 1e-05,
12
  "model_type": "gpt2",
13
  "n_ctx": 1024,
14
+ "n_embd": 1024,
15
+ "n_head": 16,
16
  "n_inner": null,
17
+ "n_layer": 24,
18
  "n_positions": 1024,
19
+ "n_special": 0,
20
+ "predict_special_tokens": true,
21
  "reorder_and_upcast_attn": false,
22
  "resid_pdrop": 0.1,
23
  "scale_attn_by_inverse_layer_idx": false,
fine_tuned_model/logs/events.out.tfevents.1746766596.BallTwister.15212.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:874d62355b3cd90f43732ecdfe955af64eefed3aaa2bddb3d4cf3d1eb85fee9c
3
+ size 5232
fine_tuned_model/logs/events.out.tfevents.1746769125.BallTwister.19464.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:252440ce133561e2c4420614f2141fccfda17b267fdeb7304f621781fd80be87
3
+ size 5232
fine_tuned_model/logs/events.out.tfevents.1746769341.BallTwister.19392.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:89ded71ebc065190398cff675688e567f2408dd6cd58c5677f8d71a2c387e087
3
+ size 5232
fine_tuned_model/logs/events.out.tfevents.1746769713.BallTwister.15164.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:119882eb2420aac920fe258684c484a8459bc755c4e5cf4ca1b5191a4ba51afb
3
+ size 5232
fine_tuned_model/logs/events.out.tfevents.1746769988.BallTwister.17104.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1b97a6b3fdc209ff5c3a299fbdf0a10ff456232c7eed7a9141c41812a735b336
3
+ size 133826
fine_tuned_model/model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:ae99eb8101f81c938c96b73fc2d85ead70e74fabf76021a6b35899fed60c5585
3
- size 497774208
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:162664b502c33c6c04c5776498b5ad474fe88e760bd08e54a8522cc34cb75783
3
+ size 1419322880
fine_tuned_model/training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:01255d5daed1e0195768aa4479272763e7c5e49e048202ed7ff4039df3f097d5
3
  size 5713
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:37bd4209646599a5612b713c1e2ec16f0b3ad2d6bbec8a9566a22a9bc23ae4c3
3
  size 5713