Hugging Face's logo

derhan
/
IndoBert-QA

Question Answering
Transformers
Safetensors
Indonesian
bert
qa
indobert
Inference Endpoints
Model card Files Community
derhan commited on
Commit
82cf387
·
verified ·
1 Parent(s): b256a08

Upload folder using huggingface_hub

Browse files
Files changed (50) hide show
  1. checkpoint-1000/config.json +29 -0
  2. checkpoint-1000/model.safetensors +3 -0
  3. checkpoint-1000/optimizer.pt +3 -0
  4. checkpoint-1000/rng_state.pth +3 -0
  5. checkpoint-1000/scheduler.pt +3 -0
  6. checkpoint-1000/special_tokens_map.json +37 -0
  7. checkpoint-1000/tokenizer.json +0 -0
  8. checkpoint-1000/tokenizer_config.json +65 -0
  9. checkpoint-1000/trainer_state.json +0 -0
  10. checkpoint-1000/training_args.bin +3 -0
  11. checkpoint-1000/vocab.txt +0 -0
  12. checkpoint-1500/config.json +29 -0
  13. checkpoint-1500/model.safetensors +3 -0
  14. checkpoint-1500/optimizer.pt +3 -0
  15. checkpoint-1500/rng_state.pth +3 -0
  16. checkpoint-1500/scheduler.pt +3 -0
  17. checkpoint-1500/special_tokens_map.json +37 -0
  18. checkpoint-1500/tokenizer.json +0 -0
  19. checkpoint-1500/tokenizer_config.json +65 -0
  20. checkpoint-1500/trainer_state.json +0 -0
  21. checkpoint-1500/training_args.bin +3 -0
  22. checkpoint-1500/vocab.txt +0 -0
  23. checkpoint-1902/config.json +29 -0
  24. checkpoint-1902/model.safetensors +3 -0
  25. checkpoint-1902/optimizer.pt +3 -0
  26. checkpoint-1902/rng_state.pth +3 -0
  27. checkpoint-1902/scheduler.pt +3 -0
  28. checkpoint-1902/special_tokens_map.json +37 -0
  29. checkpoint-1902/tokenizer.json +0 -0
  30. checkpoint-1902/tokenizer_config.json +65 -0
  31. checkpoint-1902/trainer_state.json +0 -0
  32. checkpoint-1902/training_args.bin +3 -0
  33. checkpoint-1902/vocab.txt +0 -0
  34. checkpoint-500/config.json +29 -0
  35. checkpoint-500/model.safetensors +3 -0
  36. checkpoint-500/optimizer.pt +3 -0
  37. checkpoint-500/rng_state.pth +3 -0
  38. checkpoint-500/scheduler.pt +3 -0
  39. checkpoint-500/special_tokens_map.json +37 -0
  40. checkpoint-500/tokenizer.json +0 -0
  41. checkpoint-500/tokenizer_config.json +65 -0
  42. checkpoint-500/trainer_state.json +3533 -0
  43. checkpoint-500/training_args.bin +3 -0
  44. checkpoint-500/vocab.txt +0 -0
  45. config.json +29 -0
  46. model.safetensors +3 -0
  47. special_tokens_map.json +37 -0
  48. tokenizer.json +0 -0
  49. tokenizer_config.json +65 -0
  50. vocab.txt +0 -0
checkpoint-1000/config.json ADDED
@@ -0,0 +1,29 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "Rifky/Indobert-QA",
3
+ "architectures": [
4
+ "BertForQuestionAnswering"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.1,
7
+ "bos_token_id": 0,
8
+ "classifier_dropout": null,
9
+ "eos_token_ids": 0,
10
+ "gradient_checkpointing": false,
11
+ "hidden_act": "gelu",
12
+ "hidden_dropout_prob": 0.1,
13
+ "hidden_size": 768,
14
+ "initializer_range": 0.02,
15
+ "intermediate_size": 3072,
16
+ "layer_norm_eps": 1e-12,
17
+ "max_position_embeddings": 512,
18
+ "model_type": "bert",
19
+ "num_attention_heads": 12,
20
+ "num_hidden_layers": 12,
21
+ "output_past": true,
22
+ "pad_token_id": 0,
23
+ "position_embedding_type": "absolute",
24
+ "torch_dtype": "float32",
25
+ "transformers_version": "4.47.1",
26
+ "type_vocab_size": 2,
27
+ "use_cache": true,
28
+ "vocab_size": 31923
29
+ }
checkpoint-1000/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2dc700789a6bf0cc44421541512395d7ea98fc9d24bb58f963a7466dd08fa07c
3
+ size 439899960
checkpoint-1000/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:429666d3ed4346a3b8eb978688f3753aa6849dfab2bc2ee12130ca0981fcd3b6
3
+ size 879919674
checkpoint-1000/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:073fa0f9995165ecfb72f4e15c27b31be2214074bf7fe9126d15a99df0a3ed92
3
+ size 14244
checkpoint-1000/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4f526ddb850b01c827c72d370550a48a8bf2a3fdae4067f2ac9555c308e5af24
3
+ size 1064
checkpoint-1000/special_tokens_map.json ADDED
@@ -0,0 +1,37 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "cls_token": {
3
+ "content": "[CLS]",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "mask_token": {
10
+ "content": "[MASK]",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": {
17
+ "content": "[PAD]",
18
+ "lstrip": false,
19
+ "normalized": false,
20
+ "rstrip": false,
21
+ "single_word": false
22
+ },
23
+ "sep_token": {
24
+ "content": "[SEP]",
25
+ "lstrip": false,
26
+ "normalized": false,
27
+ "rstrip": false,
28
+ "single_word": false
29
+ },
30
+ "unk_token": {
31
+ "content": "[UNK]",
32
+ "lstrip": false,
33
+ "normalized": false,
34
+ "rstrip": false,
35
+ "single_word": false
36
+ }
37
+ }
checkpoint-1000/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-1000/tokenizer_config.json ADDED
@@ -0,0 +1,65 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "[PAD]",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "1": {
12
+ "content": "[UNK]",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "2": {
20
+ "content": "[MASK]",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "3": {
28
+ "content": "[CLS]",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "4": {
36
+ "content": "[SEP]",
37
+ "lstrip": false,
38
+ "normalized": false,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ }
43
+ },
44
+ "clean_up_tokenization_spaces": false,
45
+ "cls_token": "[CLS]",
46
+ "do_basic_tokenize": true,
47
+ "do_lower_case": true,
48
+ "extra_special_tokens": {},
49
+ "mask_token": "[MASK]",
50
+ "max_length": 384,
51
+ "model_max_length": 1000000000000000019884624838656,
52
+ "never_split": null,
53
+ "pad_to_multiple_of": null,
54
+ "pad_token": "[PAD]",
55
+ "pad_token_type_id": 0,
56
+ "padding_side": "right",
57
+ "sep_token": "[SEP]",
58
+ "stride": 128,
59
+ "strip_accents": null,
60
+ "tokenize_chinese_chars": true,
61
+ "tokenizer_class": "BertTokenizer",
62
+ "truncation_side": "right",
63
+ "truncation_strategy": "only_second",
64
+ "unk_token": "[UNK]"
65
+ }
checkpoint-1000/trainer_state.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-1000/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4c46477fc32a2baa50a9a31f5fc414e1b1586156a0273be927ce2a5de7f607f3
3
+ size 5304
checkpoint-1000/vocab.txt ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-1500/config.json ADDED
@@ -0,0 +1,29 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "Rifky/Indobert-QA",
3
+ "architectures": [
4
+ "BertForQuestionAnswering"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.1,
7
+ "bos_token_id": 0,
8
+ "classifier_dropout": null,
9
+ "eos_token_ids": 0,
10
+ "gradient_checkpointing": false,
11
+ "hidden_act": "gelu",
12
+ "hidden_dropout_prob": 0.1,
13
+ "hidden_size": 768,
14
+ "initializer_range": 0.02,
15
+ "intermediate_size": 3072,
16
+ "layer_norm_eps": 1e-12,
17
+ "max_position_embeddings": 512,
18
+ "model_type": "bert",
19
+ "num_attention_heads": 12,
20
+ "num_hidden_layers": 12,
21
+ "output_past": true,
22
+ "pad_token_id": 0,
23
+ "position_embedding_type": "absolute",
24
+ "torch_dtype": "float32",
25
+ "transformers_version": "4.47.1",
26
+ "type_vocab_size": 2,
27
+ "use_cache": true,
28
+ "vocab_size": 31923
29
+ }
checkpoint-1500/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a2d76a96dda4fdbbf6175b53c3edf1368ea63c0cb55b6247cbfd6259e1335c83
3
+ size 439899960
checkpoint-1500/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:322b6d71025c987640cc710f084482a733bcc705e162d04f2b91da2479ea4a62
3
+ size 879919674
checkpoint-1500/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b4a9dc4c953dc275bfa194b3c10c4f53ea8f3e6390307fd173eae7ee4eb9db84
3
+ size 14244
checkpoint-1500/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d5003dff628027838dfd9f17f17d526a92f4ebc27faf17b48f38862a58e19053
3
+ size 1064
checkpoint-1500/special_tokens_map.json ADDED
@@ -0,0 +1,37 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "cls_token": {
3
+ "content": "[CLS]",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "mask_token": {
10
+ "content": "[MASK]",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": {
17
+ "content": "[PAD]",
18
+ "lstrip": false,
19
+ "normalized": false,
20
+ "rstrip": false,
21
+ "single_word": false
22
+ },
23
+ "sep_token": {
24
+ "content": "[SEP]",
25
+ "lstrip": false,
26
+ "normalized": false,
27
+ "rstrip": false,
28
+ "single_word": false
29
+ },
30
+ "unk_token": {
31
+ "content": "[UNK]",
32
+ "lstrip": false,
33
+ "normalized": false,
34
+ "rstrip": false,
35
+ "single_word": false
36
+ }
37
+ }
checkpoint-1500/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-1500/tokenizer_config.json ADDED
@@ -0,0 +1,65 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "[PAD]",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "1": {
12
+ "content": "[UNK]",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "2": {
20
+ "content": "[MASK]",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "3": {
28
+ "content": "[CLS]",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "4": {
36
+ "content": "[SEP]",
37
+ "lstrip": false,
38
+ "normalized": false,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ }
43
+ },
44
+ "clean_up_tokenization_spaces": false,
45
+ "cls_token": "[CLS]",
46
+ "do_basic_tokenize": true,
47
+ "do_lower_case": true,
48
+ "extra_special_tokens": {},
49
+ "mask_token": "[MASK]",
50
+ "max_length": 384,
51
+ "model_max_length": 1000000000000000019884624838656,
52
+ "never_split": null,
53
+ "pad_to_multiple_of": null,
54
+ "pad_token": "[PAD]",
55
+ "pad_token_type_id": 0,
56
+ "padding_side": "right",
57
+ "sep_token": "[SEP]",
58
+ "stride": 128,
59
+ "strip_accents": null,
60
+ "tokenize_chinese_chars": true,
61
+ "tokenizer_class": "BertTokenizer",
62
+ "truncation_side": "right",
63
+ "truncation_strategy": "only_second",
64
+ "unk_token": "[UNK]"
65
+ }
checkpoint-1500/trainer_state.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-1500/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4c46477fc32a2baa50a9a31f5fc414e1b1586156a0273be927ce2a5de7f607f3
3
+ size 5304
checkpoint-1500/vocab.txt ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-1902/config.json ADDED
@@ -0,0 +1,29 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "Rifky/Indobert-QA",
3
+ "architectures": [
4
+ "BertForQuestionAnswering"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.1,
7
+ "bos_token_id": 0,
8
+ "classifier_dropout": null,
9
+ "eos_token_ids": 0,
10
+ "gradient_checkpointing": false,
11
+ "hidden_act": "gelu",
12
+ "hidden_dropout_prob": 0.1,
13
+ "hidden_size": 768,
14
+ "initializer_range": 0.02,
15
+ "intermediate_size": 3072,
16
+ "layer_norm_eps": 1e-12,
17
+ "max_position_embeddings": 512,
18
+ "model_type": "bert",
19
+ "num_attention_heads": 12,
20
+ "num_hidden_layers": 12,
21
+ "output_past": true,
22
+ "pad_token_id": 0,
23
+ "position_embedding_type": "absolute",
24
+ "torch_dtype": "float32",
25
+ "transformers_version": "4.47.1",
26
+ "type_vocab_size": 2,
27
+ "use_cache": true,
28
+ "vocab_size": 31923
29
+ }
checkpoint-1902/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d5b8499ed6aefc2d805cc18cd56df708dde1898d0c86166f20c35b590ffafac0
3
+ size 439899960
checkpoint-1902/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a109944bcf9e43f3b3f0064c766b3e7b9d632c23446ec27a52fed0f239ddb22d
3
+ size 879919674
checkpoint-1902/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a1cbd9df30705dac11a780ba82b404be260045778f6a028dd7faf02241142555
3
+ size 14244
checkpoint-1902/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9a8c54765710720a99456eb46c389e0fcec26e1e24fa93b35a7ff7f505d2cb80
3
+ size 1064
checkpoint-1902/special_tokens_map.json ADDED
@@ -0,0 +1,37 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "cls_token": {
3
+ "content": "[CLS]",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "mask_token": {
10
+ "content": "[MASK]",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": {
17
+ "content": "[PAD]",
18
+ "lstrip": false,
19
+ "normalized": false,
20
+ "rstrip": false,
21
+ "single_word": false
22
+ },
23
+ "sep_token": {
24
+ "content": "[SEP]",
25
+ "lstrip": false,
26
+ "normalized": false,
27
+ "rstrip": false,
28
+ "single_word": false
29
+ },
30
+ "unk_token": {
31
+ "content": "[UNK]",
32
+ "lstrip": false,
33
+ "normalized": false,
34
+ "rstrip": false,
35
+ "single_word": false
36
+ }
37
+ }
checkpoint-1902/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-1902/tokenizer_config.json ADDED
@@ -0,0 +1,65 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "[PAD]",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "1": {
12
+ "content": "[UNK]",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "2": {
20
+ "content": "[MASK]",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "3": {
28
+ "content": "[CLS]",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "4": {
36
+ "content": "[SEP]",
37
+ "lstrip": false,
38
+ "normalized": false,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ }
43
+ },
44
+ "clean_up_tokenization_spaces": false,
45
+ "cls_token": "[CLS]",
46
+ "do_basic_tokenize": true,
47
+ "do_lower_case": true,
48
+ "extra_special_tokens": {},
49
+ "mask_token": "[MASK]",
50
+ "max_length": 384,
51
+ "model_max_length": 1000000000000000019884624838656,
52
+ "never_split": null,
53
+ "pad_to_multiple_of": null,
54
+ "pad_token": "[PAD]",
55
+ "pad_token_type_id": 0,
56
+ "padding_side": "right",
57
+ "sep_token": "[SEP]",
58
+ "stride": 128,
59
+ "strip_accents": null,
60
+ "tokenize_chinese_chars": true,
61
+ "tokenizer_class": "BertTokenizer",
62
+ "truncation_side": "right",
63
+ "truncation_strategy": "only_second",
64
+ "unk_token": "[UNK]"
65
+ }
checkpoint-1902/trainer_state.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-1902/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4c46477fc32a2baa50a9a31f5fc414e1b1586156a0273be927ce2a5de7f607f3
3
+ size 5304
checkpoint-1902/vocab.txt ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-500/config.json ADDED
@@ -0,0 +1,29 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "Rifky/Indobert-QA",
3
+ "architectures": [
4
+ "BertForQuestionAnswering"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.1,
7
+ "bos_token_id": 0,
8
+ "classifier_dropout": null,
9
+ "eos_token_ids": 0,
10
+ "gradient_checkpointing": false,
11
+ "hidden_act": "gelu",
12
+ "hidden_dropout_prob": 0.1,
13
+ "hidden_size": 768,
14
+ "initializer_range": 0.02,
15
+ "intermediate_size": 3072,
16
+ "layer_norm_eps": 1e-12,
17
+ "max_position_embeddings": 512,
18
+ "model_type": "bert",
19
+ "num_attention_heads": 12,
20
+ "num_hidden_layers": 12,
21
+ "output_past": true,
22
+ "pad_token_id": 0,
23
+ "position_embedding_type": "absolute",
24
+ "torch_dtype": "float32",
25
+ "transformers_version": "4.47.1",
26
+ "type_vocab_size": 2,
27
+ "use_cache": true,
28
+ "vocab_size": 31923
29
+ }
checkpoint-500/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:815d2185cc07a36c812063d4080c05dd7e98b71194b1baeaf4f024c63b9383d8
3
+ size 439899960
checkpoint-500/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:97c215fd9fac8e16bf1569954d5215eb3b2300a6d39505cd3cd97cb5fc52ad0f
3
+ size 879919674
checkpoint-500/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9f750406fdaee45131a56c5eae5d4879e22e4db2d1d927adb51eb19c9a9d2409
3
+ size 14244
checkpoint-500/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:55dde4b8f70d371aaa290025b7825f1d9e4c0814f7d919138b63c00fb84da204
3
+ size 1064
checkpoint-500/special_tokens_map.json ADDED
@@ -0,0 +1,37 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "cls_token": {
3
+ "content": "[CLS]",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "mask_token": {
10
+ "content": "[MASK]",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": {
17
+ "content": "[PAD]",
18
+ "lstrip": false,
19
+ "normalized": false,
20
+ "rstrip": false,
21
+ "single_word": false
22
+ },
23
+ "sep_token": {
24
+ "content": "[SEP]",
25
+ "lstrip": false,
26
+ "normalized": false,
27
+ "rstrip": false,
28
+ "single_word": false
29
+ },
30
+ "unk_token": {
31
+ "content": "[UNK]",
32
+ "lstrip": false,
33
+ "normalized": false,
34
+ "rstrip": false,
35
+ "single_word": false
36
+ }
37
+ }
checkpoint-500/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-500/tokenizer_config.json ADDED
@@ -0,0 +1,65 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "[PAD]",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "1": {
12
+ "content": "[UNK]",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "2": {
20
+ "content": "[MASK]",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "3": {
28
+ "content": "[CLS]",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "4": {
36
+ "content": "[SEP]",
37
+ "lstrip": false,
38
+ "normalized": false,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ }
43
+ },
44
+ "clean_up_tokenization_spaces": false,
45
+ "cls_token": "[CLS]",
46
+ "do_basic_tokenize": true,
47
+ "do_lower_case": true,
48
+ "extra_special_tokens": {},
49
+ "mask_token": "[MASK]",
50
+ "max_length": 384,
51
+ "model_max_length": 1000000000000000019884624838656,
52
+ "never_split": null,
53
+ "pad_to_multiple_of": null,
54
+ "pad_token": "[PAD]",
55
+ "pad_token_type_id": 0,
56
+ "padding_side": "right",
57
+ "sep_token": "[SEP]",
58
+ "stride": 128,
59
+ "strip_accents": null,
60
+ "tokenize_chinese_chars": true,
61
+ "tokenizer_class": "BertTokenizer",
62
+ "truncation_side": "right",
63
+ "truncation_strategy": "only_second",
64
+ "unk_token": "[UNK]"
65
+ }
checkpoint-500/trainer_state.json ADDED
@@ -0,0 +1,3533 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 0.7886435331230284,
5
+ "eval_steps": 500,
6
+ "global_step": 500,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.0015772870662460567,
13
+ "grad_norm": 20.9893741607666,
14
+ "learning_rate": 2.998422712933754e-05,
15
+ "loss": 1.6281,
16
+ "step": 1
17
+ },
18
+ {
19
+ "epoch": 0.0031545741324921135,
20
+ "grad_norm": 19.62778663635254,
21
+ "learning_rate": 2.996845425867508e-05,
22
+ "loss": 0.3011,
23
+ "step": 2
24
+ },
25
+ {
26
+ "epoch": 0.00473186119873817,
27
+ "grad_norm": 12.428619384765625,
28
+ "learning_rate": 2.995268138801262e-05,
29
+ "loss": 0.3921,
30
+ "step": 3
31
+ },
32
+ {
33
+ "epoch": 0.006309148264984227,
34
+ "grad_norm": 21.599090576171875,
35
+ "learning_rate": 2.993690851735016e-05,
36
+ "loss": 1.1201,
37
+ "step": 4
38
+ },
39
+ {
40
+ "epoch": 0.007886435331230283,
41
+ "grad_norm": 37.74925231933594,
42
+ "learning_rate": 2.99211356466877e-05,
43
+ "loss": 0.9841,
44
+ "step": 5
45
+ },
46
+ {
47
+ "epoch": 0.00946372239747634,
48
+ "grad_norm": 21.6423397064209,
49
+ "learning_rate": 2.990536277602524e-05,
50
+ "loss": 0.4028,
51
+ "step": 6
52
+ },
53
+ {
54
+ "epoch": 0.011041009463722398,
55
+ "grad_norm": 16.425018310546875,
56
+ "learning_rate": 2.9889589905362776e-05,
57
+ "loss": 0.5775,
58
+ "step": 7
59
+ },
60
+ {
61
+ "epoch": 0.012618296529968454,
62
+ "grad_norm": 16.20684814453125,
63
+ "learning_rate": 2.9873817034700316e-05,
64
+ "loss": 0.4639,
65
+ "step": 8
66
+ },
67
+ {
68
+ "epoch": 0.014195583596214511,
69
+ "grad_norm": 26.340354919433594,
70
+ "learning_rate": 2.9858044164037856e-05,
71
+ "loss": 1.0636,
72
+ "step": 9
73
+ },
74
+ {
75
+ "epoch": 0.015772870662460567,
76
+ "grad_norm": 16.133371353149414,
77
+ "learning_rate": 2.9842271293375395e-05,
78
+ "loss": 0.3065,
79
+ "step": 10
80
+ },
81
+ {
82
+ "epoch": 0.017350157728706624,
83
+ "grad_norm": 49.714969635009766,
84
+ "learning_rate": 2.9826498422712935e-05,
85
+ "loss": 1.167,
86
+ "step": 11
87
+ },
88
+ {
89
+ "epoch": 0.01892744479495268,
90
+ "grad_norm": 23.211793899536133,
91
+ "learning_rate": 2.9810725552050475e-05,
92
+ "loss": 0.3201,
93
+ "step": 12
94
+ },
95
+ {
96
+ "epoch": 0.02050473186119874,
97
+ "grad_norm": 30.51961326599121,
98
+ "learning_rate": 2.9794952681388015e-05,
99
+ "loss": 1.3932,
100
+ "step": 13
101
+ },
102
+ {
103
+ "epoch": 0.022082018927444796,
104
+ "grad_norm": 27.843469619750977,
105
+ "learning_rate": 2.9779179810725555e-05,
106
+ "loss": 0.5147,
107
+ "step": 14
108
+ },
109
+ {
110
+ "epoch": 0.02365930599369085,
111
+ "grad_norm": 10.2225341796875,
112
+ "learning_rate": 2.9763406940063094e-05,
113
+ "loss": 0.162,
114
+ "step": 15
115
+ },
116
+ {
117
+ "epoch": 0.025236593059936908,
118
+ "grad_norm": 31.443710327148438,
119
+ "learning_rate": 2.9747634069400634e-05,
120
+ "loss": 1.0028,
121
+ "step": 16
122
+ },
123
+ {
124
+ "epoch": 0.026813880126182965,
125
+ "grad_norm": 50.728050231933594,
126
+ "learning_rate": 2.973186119873817e-05,
127
+ "loss": 1.119,
128
+ "step": 17
129
+ },
130
+ {
131
+ "epoch": 0.028391167192429023,
132
+ "grad_norm": 22.010009765625,
133
+ "learning_rate": 2.971608832807571e-05,
134
+ "loss": 0.6629,
135
+ "step": 18
136
+ },
137
+ {
138
+ "epoch": 0.02996845425867508,
139
+ "grad_norm": 65.1316146850586,
140
+ "learning_rate": 2.970031545741325e-05,
141
+ "loss": 1.3804,
142
+ "step": 19
143
+ },
144
+ {
145
+ "epoch": 0.031545741324921134,
146
+ "grad_norm": 10.089284896850586,
147
+ "learning_rate": 2.9684542586750787e-05,
148
+ "loss": 0.4699,
149
+ "step": 20
150
+ },
151
+ {
152
+ "epoch": 0.033123028391167195,
153
+ "grad_norm": 40.829681396484375,
154
+ "learning_rate": 2.9668769716088326e-05,
155
+ "loss": 1.5305,
156
+ "step": 21
157
+ },
158
+ {
159
+ "epoch": 0.03470031545741325,
160
+ "grad_norm": 37.95436096191406,
161
+ "learning_rate": 2.9652996845425866e-05,
162
+ "loss": 1.9942,
163
+ "step": 22
164
+ },
165
+ {
166
+ "epoch": 0.03627760252365931,
167
+ "grad_norm": 37.55937194824219,
168
+ "learning_rate": 2.9637223974763406e-05,
169
+ "loss": 1.3291,
170
+ "step": 23
171
+ },
172
+ {
173
+ "epoch": 0.03785488958990536,
174
+ "grad_norm": 23.157485961914062,
175
+ "learning_rate": 2.9621451104100946e-05,
176
+ "loss": 0.5593,
177
+ "step": 24
178
+ },
179
+ {
180
+ "epoch": 0.03943217665615142,
181
+ "grad_norm": 17.775590896606445,
182
+ "learning_rate": 2.9605678233438486e-05,
183
+ "loss": 0.5526,
184
+ "step": 25
185
+ },
186
+ {
187
+ "epoch": 0.04100946372239748,
188
+ "grad_norm": 30.24547004699707,
189
+ "learning_rate": 2.9589905362776025e-05,
190
+ "loss": 0.9385,
191
+ "step": 26
192
+ },
193
+ {
194
+ "epoch": 0.04258675078864353,
195
+ "grad_norm": 25.479963302612305,
196
+ "learning_rate": 2.9574132492113565e-05,
197
+ "loss": 0.7802,
198
+ "step": 27
199
+ },
200
+ {
201
+ "epoch": 0.04416403785488959,
202
+ "grad_norm": 40.21939468383789,
203
+ "learning_rate": 2.9558359621451105e-05,
204
+ "loss": 1.6281,
205
+ "step": 28
206
+ },
207
+ {
208
+ "epoch": 0.04574132492113565,
209
+ "grad_norm": 48.497589111328125,
210
+ "learning_rate": 2.9542586750788645e-05,
211
+ "loss": 1.1922,
212
+ "step": 29
213
+ },
214
+ {
215
+ "epoch": 0.0473186119873817,
216
+ "grad_norm": 29.533870697021484,
217
+ "learning_rate": 2.9526813880126185e-05,
218
+ "loss": 0.6093,
219
+ "step": 30
220
+ },
221
+ {
222
+ "epoch": 0.04889589905362776,
223
+ "grad_norm": 31.541582107543945,
224
+ "learning_rate": 2.9511041009463724e-05,
225
+ "loss": 1.7102,
226
+ "step": 31
227
+ },
228
+ {
229
+ "epoch": 0.050473186119873815,
230
+ "grad_norm": 17.051698684692383,
231
+ "learning_rate": 2.949526813880126e-05,
232
+ "loss": 0.4954,
233
+ "step": 32
234
+ },
235
+ {
236
+ "epoch": 0.052050473186119876,
237
+ "grad_norm": 20.815147399902344,
238
+ "learning_rate": 2.94794952681388e-05,
239
+ "loss": 0.8868,
240
+ "step": 33
241
+ },
242
+ {
243
+ "epoch": 0.05362776025236593,
244
+ "grad_norm": 31.399736404418945,
245
+ "learning_rate": 2.946372239747634e-05,
246
+ "loss": 2.1807,
247
+ "step": 34
248
+ },
249
+ {
250
+ "epoch": 0.055205047318611984,
251
+ "grad_norm": 24.065919876098633,
252
+ "learning_rate": 2.944794952681388e-05,
253
+ "loss": 0.6677,
254
+ "step": 35
255
+ },
256
+ {
257
+ "epoch": 0.056782334384858045,
258
+ "grad_norm": 21.20488166809082,
259
+ "learning_rate": 2.943217665615142e-05,
260
+ "loss": 1.0656,
261
+ "step": 36
262
+ },
263
+ {
264
+ "epoch": 0.0583596214511041,
265
+ "grad_norm": 19.37374496459961,
266
+ "learning_rate": 2.941640378548896e-05,
267
+ "loss": 0.337,
268
+ "step": 37
269
+ },
270
+ {
271
+ "epoch": 0.05993690851735016,
272
+ "grad_norm": 16.723129272460938,
273
+ "learning_rate": 2.94006309148265e-05,
274
+ "loss": 0.7522,
275
+ "step": 38
276
+ },
277
+ {
278
+ "epoch": 0.061514195583596214,
279
+ "grad_norm": 28.176803588867188,
280
+ "learning_rate": 2.938485804416404e-05,
281
+ "loss": 0.7646,
282
+ "step": 39
283
+ },
284
+ {
285
+ "epoch": 0.06309148264984227,
286
+ "grad_norm": 25.508466720581055,
287
+ "learning_rate": 2.936908517350158e-05,
288
+ "loss": 0.6824,
289
+ "step": 40
290
+ },
291
+ {
292
+ "epoch": 0.06466876971608833,
293
+ "grad_norm": 50.56950378417969,
294
+ "learning_rate": 2.935331230283912e-05,
295
+ "loss": 1.2007,
296
+ "step": 41
297
+ },
298
+ {
299
+ "epoch": 0.06624605678233439,
300
+ "grad_norm": 32.97687911987305,
301
+ "learning_rate": 2.933753943217666e-05,
302
+ "loss": 0.9715,
303
+ "step": 42
304
+ },
305
+ {
306
+ "epoch": 0.06782334384858044,
307
+ "grad_norm": 46.321529388427734,
308
+ "learning_rate": 2.93217665615142e-05,
309
+ "loss": 1.3589,
310
+ "step": 43
311
+ },
312
+ {
313
+ "epoch": 0.0694006309148265,
314
+ "grad_norm": 24.557514190673828,
315
+ "learning_rate": 2.9305993690851738e-05,
316
+ "loss": 0.6534,
317
+ "step": 44
318
+ },
319
+ {
320
+ "epoch": 0.07097791798107256,
321
+ "grad_norm": 50.76677703857422,
322
+ "learning_rate": 2.9290220820189275e-05,
323
+ "loss": 1.311,
324
+ "step": 45
325
+ },
326
+ {
327
+ "epoch": 0.07255520504731862,
328
+ "grad_norm": 27.278282165527344,
329
+ "learning_rate": 2.9274447949526814e-05,
330
+ "loss": 0.5609,
331
+ "step": 46
332
+ },
333
+ {
334
+ "epoch": 0.07413249211356467,
335
+ "grad_norm": 16.322616577148438,
336
+ "learning_rate": 2.9258675078864354e-05,
337
+ "loss": 0.9416,
338
+ "step": 47
339
+ },
340
+ {
341
+ "epoch": 0.07570977917981073,
342
+ "grad_norm": 18.981054306030273,
343
+ "learning_rate": 2.9242902208201894e-05,
344
+ "loss": 0.5823,
345
+ "step": 48
346
+ },
347
+ {
348
+ "epoch": 0.07728706624605679,
349
+ "grad_norm": 29.21390724182129,
350
+ "learning_rate": 2.9227129337539434e-05,
351
+ "loss": 0.8926,
352
+ "step": 49
353
+ },
354
+ {
355
+ "epoch": 0.07886435331230283,
356
+ "grad_norm": 38.420040130615234,
357
+ "learning_rate": 2.9211356466876974e-05,
358
+ "loss": 0.6849,
359
+ "step": 50
360
+ },
361
+ {
362
+ "epoch": 0.0804416403785489,
363
+ "grad_norm": 39.35421371459961,
364
+ "learning_rate": 2.919558359621451e-05,
365
+ "loss": 0.4574,
366
+ "step": 51
367
+ },
368
+ {
369
+ "epoch": 0.08201892744479496,
370
+ "grad_norm": 22.953149795532227,
371
+ "learning_rate": 2.917981072555205e-05,
372
+ "loss": 0.5486,
373
+ "step": 52
374
+ },
375
+ {
376
+ "epoch": 0.083596214511041,
377
+ "grad_norm": 20.978830337524414,
378
+ "learning_rate": 2.916403785488959e-05,
379
+ "loss": 1.1119,
380
+ "step": 53
381
+ },
382
+ {
383
+ "epoch": 0.08517350157728706,
384
+ "grad_norm": 36.50906753540039,
385
+ "learning_rate": 2.914826498422713e-05,
386
+ "loss": 0.7033,
387
+ "step": 54
388
+ },
389
+ {
390
+ "epoch": 0.08675078864353312,
391
+ "grad_norm": 38.198158264160156,
392
+ "learning_rate": 2.913249211356467e-05,
393
+ "loss": 0.9468,
394
+ "step": 55
395
+ },
396
+ {
397
+ "epoch": 0.08832807570977919,
398
+ "grad_norm": 24.576414108276367,
399
+ "learning_rate": 2.911671924290221e-05,
400
+ "loss": 1.1177,
401
+ "step": 56
402
+ },
403
+ {
404
+ "epoch": 0.08990536277602523,
405
+ "grad_norm": 27.50937271118164,
406
+ "learning_rate": 2.910094637223975e-05,
407
+ "loss": 0.8692,
408
+ "step": 57
409
+ },
410
+ {
411
+ "epoch": 0.0914826498422713,
412
+ "grad_norm": 24.555320739746094,
413
+ "learning_rate": 2.9085173501577285e-05,
414
+ "loss": 0.6637,
415
+ "step": 58
416
+ },
417
+ {
418
+ "epoch": 0.09305993690851735,
419
+ "grad_norm": 18.396305084228516,
420
+ "learning_rate": 2.9069400630914825e-05,
421
+ "loss": 0.9046,
422
+ "step": 59
423
+ },
424
+ {
425
+ "epoch": 0.0946372239747634,
426
+ "grad_norm": 24.06279754638672,
427
+ "learning_rate": 2.9053627760252365e-05,
428
+ "loss": 1.1363,
429
+ "step": 60
430
+ },
431
+ {
432
+ "epoch": 0.09621451104100946,
433
+ "grad_norm": 41.308319091796875,
434
+ "learning_rate": 2.9037854889589905e-05,
435
+ "loss": 1.0696,
436
+ "step": 61
437
+ },
438
+ {
439
+ "epoch": 0.09779179810725552,
440
+ "grad_norm": 24.500062942504883,
441
+ "learning_rate": 2.9022082018927444e-05,
442
+ "loss": 0.7277,
443
+ "step": 62
444
+ },
445
+ {
446
+ "epoch": 0.09936908517350158,
447
+ "grad_norm": 75.01914978027344,
448
+ "learning_rate": 2.9006309148264984e-05,
449
+ "loss": 1.8137,
450
+ "step": 63
451
+ },
452
+ {
453
+ "epoch": 0.10094637223974763,
454
+ "grad_norm": 36.68043899536133,
455
+ "learning_rate": 2.8990536277602524e-05,
456
+ "loss": 0.8721,
457
+ "step": 64
458
+ },
459
+ {
460
+ "epoch": 0.10252365930599369,
461
+ "grad_norm": 12.270452499389648,
462
+ "learning_rate": 2.8974763406940064e-05,
463
+ "loss": 0.4874,
464
+ "step": 65
465
+ },
466
+ {
467
+ "epoch": 0.10410094637223975,
468
+ "grad_norm": 59.047279357910156,
469
+ "learning_rate": 2.8958990536277604e-05,
470
+ "loss": 1.3081,
471
+ "step": 66
472
+ },
473
+ {
474
+ "epoch": 0.1056782334384858,
475
+ "grad_norm": 50.99152755737305,
476
+ "learning_rate": 2.8943217665615143e-05,
477
+ "loss": 1.4375,
478
+ "step": 67
479
+ },
480
+ {
481
+ "epoch": 0.10725552050473186,
482
+ "grad_norm": 22.043882369995117,
483
+ "learning_rate": 2.8927444794952683e-05,
484
+ "loss": 1.2062,
485
+ "step": 68
486
+ },
487
+ {
488
+ "epoch": 0.10883280757097792,
489
+ "grad_norm": 19.774150848388672,
490
+ "learning_rate": 2.8911671924290223e-05,
491
+ "loss": 0.4599,
492
+ "step": 69
493
+ },
494
+ {
495
+ "epoch": 0.11041009463722397,
496
+ "grad_norm": 40.691856384277344,
497
+ "learning_rate": 2.8895899053627763e-05,
498
+ "loss": 1.3628,
499
+ "step": 70
500
+ },
501
+ {
502
+ "epoch": 0.11198738170347003,
503
+ "grad_norm": 43.96192169189453,
504
+ "learning_rate": 2.88801261829653e-05,
505
+ "loss": 1.2163,
506
+ "step": 71
507
+ },
508
+ {
509
+ "epoch": 0.11356466876971609,
510
+ "grad_norm": 43.754364013671875,
511
+ "learning_rate": 2.886435331230284e-05,
512
+ "loss": 1.0337,
513
+ "step": 72
514
+ },
515
+ {
516
+ "epoch": 0.11514195583596215,
517
+ "grad_norm": 29.008777618408203,
518
+ "learning_rate": 2.884858044164038e-05,
519
+ "loss": 1.0086,
520
+ "step": 73
521
+ },
522
+ {
523
+ "epoch": 0.1167192429022082,
524
+ "grad_norm": 16.263519287109375,
525
+ "learning_rate": 2.883280757097792e-05,
526
+ "loss": 0.715,
527
+ "step": 74
528
+ },
529
+ {
530
+ "epoch": 0.11829652996845426,
531
+ "grad_norm": 36.508995056152344,
532
+ "learning_rate": 2.881703470031546e-05,
533
+ "loss": 1.3844,
534
+ "step": 75
535
+ },
536
+ {
537
+ "epoch": 0.11987381703470032,
538
+ "grad_norm": 36.9827880859375,
539
+ "learning_rate": 2.8801261829652998e-05,
540
+ "loss": 0.7432,
541
+ "step": 76
542
+ },
543
+ {
544
+ "epoch": 0.12145110410094637,
545
+ "grad_norm": 19.185775756835938,
546
+ "learning_rate": 2.8785488958990538e-05,
547
+ "loss": 0.6099,
548
+ "step": 77
549
+ },
550
+ {
551
+ "epoch": 0.12302839116719243,
552
+ "grad_norm": 48.45042037963867,
553
+ "learning_rate": 2.8769716088328078e-05,
554
+ "loss": 1.0526,
555
+ "step": 78
556
+ },
557
+ {
558
+ "epoch": 0.12460567823343849,
559
+ "grad_norm": 43.460269927978516,
560
+ "learning_rate": 2.8753943217665618e-05,
561
+ "loss": 1.2607,
562
+ "step": 79
563
+ },
564
+ {
565
+ "epoch": 0.12618296529968454,
566
+ "grad_norm": 25.808408737182617,
567
+ "learning_rate": 2.8738170347003157e-05,
568
+ "loss": 0.3468,
569
+ "step": 80
570
+ },
571
+ {
572
+ "epoch": 0.1277602523659306,
573
+ "grad_norm": 26.330102920532227,
574
+ "learning_rate": 2.8722397476340697e-05,
575
+ "loss": 0.5094,
576
+ "step": 81
577
+ },
578
+ {
579
+ "epoch": 0.12933753943217666,
580
+ "grad_norm": 36.02430725097656,
581
+ "learning_rate": 2.8706624605678237e-05,
582
+ "loss": 1.7968,
583
+ "step": 82
584
+ },
585
+ {
586
+ "epoch": 0.1309148264984227,
587
+ "grad_norm": 27.477529525756836,
588
+ "learning_rate": 2.8690851735015773e-05,
589
+ "loss": 0.5631,
590
+ "step": 83
591
+ },
592
+ {
593
+ "epoch": 0.13249211356466878,
594
+ "grad_norm": 27.334575653076172,
595
+ "learning_rate": 2.8675078864353313e-05,
596
+ "loss": 1.0695,
597
+ "step": 84
598
+ },
599
+ {
600
+ "epoch": 0.13406940063091483,
601
+ "grad_norm": 34.76345443725586,
602
+ "learning_rate": 2.865930599369085e-05,
603
+ "loss": 0.8823,
604
+ "step": 85
605
+ },
606
+ {
607
+ "epoch": 0.13564668769716087,
608
+ "grad_norm": 38.68295669555664,
609
+ "learning_rate": 2.864353312302839e-05,
610
+ "loss": 0.7875,
611
+ "step": 86
612
+ },
613
+ {
614
+ "epoch": 0.13722397476340695,
615
+ "grad_norm": 47.18848419189453,
616
+ "learning_rate": 2.862776025236593e-05,
617
+ "loss": 1.383,
618
+ "step": 87
619
+ },
620
+ {
621
+ "epoch": 0.138801261829653,
622
+ "grad_norm": 44.14547348022461,
623
+ "learning_rate": 2.861198738170347e-05,
624
+ "loss": 0.7772,
625
+ "step": 88
626
+ },
627
+ {
628
+ "epoch": 0.14037854889589904,
629
+ "grad_norm": 37.833553314208984,
630
+ "learning_rate": 2.859621451104101e-05,
631
+ "loss": 1.2231,
632
+ "step": 89
633
+ },
634
+ {
635
+ "epoch": 0.14195583596214512,
636
+ "grad_norm": 47.564048767089844,
637
+ "learning_rate": 2.858044164037855e-05,
638
+ "loss": 0.9194,
639
+ "step": 90
640
+ },
641
+ {
642
+ "epoch": 0.14353312302839116,
643
+ "grad_norm": 19.151697158813477,
644
+ "learning_rate": 2.856466876971609e-05,
645
+ "loss": 0.4425,
646
+ "step": 91
647
+ },
648
+ {
649
+ "epoch": 0.14511041009463724,
650
+ "grad_norm": 19.90251350402832,
651
+ "learning_rate": 2.8548895899053628e-05,
652
+ "loss": 0.3942,
653
+ "step": 92
654
+ },
655
+ {
656
+ "epoch": 0.14668769716088328,
657
+ "grad_norm": 59.073631286621094,
658
+ "learning_rate": 2.8533123028391168e-05,
659
+ "loss": 1.4536,
660
+ "step": 93
661
+ },
662
+ {
663
+ "epoch": 0.14826498422712933,
664
+ "grad_norm": 52.932861328125,
665
+ "learning_rate": 2.8517350157728708e-05,
666
+ "loss": 0.994,
667
+ "step": 94
668
+ },
669
+ {
670
+ "epoch": 0.1498422712933754,
671
+ "grad_norm": 21.305513381958008,
672
+ "learning_rate": 2.8501577287066248e-05,
673
+ "loss": 0.707,
674
+ "step": 95
675
+ },
676
+ {
677
+ "epoch": 0.15141955835962145,
678
+ "grad_norm": 14.83249568939209,
679
+ "learning_rate": 2.8485804416403784e-05,
680
+ "loss": 0.4138,
681
+ "step": 96
682
+ },
683
+ {
684
+ "epoch": 0.1529968454258675,
685
+ "grad_norm": 19.941259384155273,
686
+ "learning_rate": 2.8470031545741324e-05,
687
+ "loss": 0.3036,
688
+ "step": 97
689
+ },
690
+ {
691
+ "epoch": 0.15457413249211358,
692
+ "grad_norm": 37.68672561645508,
693
+ "learning_rate": 2.8454258675078864e-05,
694
+ "loss": 0.5776,
695
+ "step": 98
696
+ },
697
+ {
698
+ "epoch": 0.15615141955835962,
699
+ "grad_norm": 54.10725784301758,
700
+ "learning_rate": 2.8438485804416403e-05,
701
+ "loss": 1.4085,
702
+ "step": 99
703
+ },
704
+ {
705
+ "epoch": 0.15772870662460567,
706
+ "grad_norm": 23.21993064880371,
707
+ "learning_rate": 2.8422712933753943e-05,
708
+ "loss": 0.2434,
709
+ "step": 100
710
+ },
711
+ {
712
+ "epoch": 0.15930599369085174,
713
+ "grad_norm": 37.54609680175781,
714
+ "learning_rate": 2.8406940063091483e-05,
715
+ "loss": 1.0403,
716
+ "step": 101
717
+ },
718
+ {
719
+ "epoch": 0.1608832807570978,
720
+ "grad_norm": 11.716938972473145,
721
+ "learning_rate": 2.8391167192429023e-05,
722
+ "loss": 0.391,
723
+ "step": 102
724
+ },
725
+ {
726
+ "epoch": 0.16246056782334384,
727
+ "grad_norm": 24.197599411010742,
728
+ "learning_rate": 2.8375394321766563e-05,
729
+ "loss": 0.6445,
730
+ "step": 103
731
+ },
732
+ {
733
+ "epoch": 0.1640378548895899,
734
+ "grad_norm": 30.646610260009766,
735
+ "learning_rate": 2.8359621451104102e-05,
736
+ "loss": 0.9609,
737
+ "step": 104
738
+ },
739
+ {
740
+ "epoch": 0.16561514195583596,
741
+ "grad_norm": 10.75748062133789,
742
+ "learning_rate": 2.8343848580441642e-05,
743
+ "loss": 0.6466,
744
+ "step": 105
745
+ },
746
+ {
747
+ "epoch": 0.167192429022082,
748
+ "grad_norm": 47.22863006591797,
749
+ "learning_rate": 2.8328075709779182e-05,
750
+ "loss": 1.3193,
751
+ "step": 106
752
+ },
753
+ {
754
+ "epoch": 0.16876971608832808,
755
+ "grad_norm": 22.453563690185547,
756
+ "learning_rate": 2.8312302839116722e-05,
757
+ "loss": 0.8988,
758
+ "step": 107
759
+ },
760
+ {
761
+ "epoch": 0.17034700315457413,
762
+ "grad_norm": 19.869998931884766,
763
+ "learning_rate": 2.829652996845426e-05,
764
+ "loss": 0.6431,
765
+ "step": 108
766
+ },
767
+ {
768
+ "epoch": 0.1719242902208202,
769
+ "grad_norm": 15.483345031738281,
770
+ "learning_rate": 2.8280757097791798e-05,
771
+ "loss": 0.5337,
772
+ "step": 109
773
+ },
774
+ {
775
+ "epoch": 0.17350157728706625,
776
+ "grad_norm": 31.25865364074707,
777
+ "learning_rate": 2.8264984227129338e-05,
778
+ "loss": 1.0998,
779
+ "step": 110
780
+ },
781
+ {
782
+ "epoch": 0.1750788643533123,
783
+ "grad_norm": 43.77351760864258,
784
+ "learning_rate": 2.8249211356466878e-05,
785
+ "loss": 1.8904,
786
+ "step": 111
787
+ },
788
+ {
789
+ "epoch": 0.17665615141955837,
790
+ "grad_norm": 23.07604217529297,
791
+ "learning_rate": 2.8233438485804417e-05,
792
+ "loss": 0.6687,
793
+ "step": 112
794
+ },
795
+ {
796
+ "epoch": 0.17823343848580442,
797
+ "grad_norm": 12.515803337097168,
798
+ "learning_rate": 2.8217665615141957e-05,
799
+ "loss": 0.5234,
800
+ "step": 113
801
+ },
802
+ {
803
+ "epoch": 0.17981072555205047,
804
+ "grad_norm": 28.48925018310547,
805
+ "learning_rate": 2.8201892744479497e-05,
806
+ "loss": 1.2136,
807
+ "step": 114
808
+ },
809
+ {
810
+ "epoch": 0.18138801261829654,
811
+ "grad_norm": 9.342551231384277,
812
+ "learning_rate": 2.8186119873817037e-05,
813
+ "loss": 0.1762,
814
+ "step": 115
815
+ },
816
+ {
817
+ "epoch": 0.1829652996845426,
818
+ "grad_norm": 16.196250915527344,
819
+ "learning_rate": 2.8170347003154577e-05,
820
+ "loss": 0.355,
821
+ "step": 116
822
+ },
823
+ {
824
+ "epoch": 0.18454258675078863,
825
+ "grad_norm": 36.83229064941406,
826
+ "learning_rate": 2.8154574132492116e-05,
827
+ "loss": 1.3007,
828
+ "step": 117
829
+ },
830
+ {
831
+ "epoch": 0.1861198738170347,
832
+ "grad_norm": 41.2702522277832,
833
+ "learning_rate": 2.8138801261829656e-05,
834
+ "loss": 1.4537,
835
+ "step": 118
836
+ },
837
+ {
838
+ "epoch": 0.18769716088328076,
839
+ "grad_norm": 56.45298385620117,
840
+ "learning_rate": 2.8123028391167193e-05,
841
+ "loss": 1.0358,
842
+ "step": 119
843
+ },
844
+ {
845
+ "epoch": 0.1892744479495268,
846
+ "grad_norm": 19.364944458007812,
847
+ "learning_rate": 2.8107255520504732e-05,
848
+ "loss": 0.8025,
849
+ "step": 120
850
+ },
851
+ {
852
+ "epoch": 0.19085173501577288,
853
+ "grad_norm": 12.787603378295898,
854
+ "learning_rate": 2.8091482649842272e-05,
855
+ "loss": 0.346,
856
+ "step": 121
857
+ },
858
+ {
859
+ "epoch": 0.19242902208201892,
860
+ "grad_norm": 17.909929275512695,
861
+ "learning_rate": 2.807570977917981e-05,
862
+ "loss": 0.3148,
863
+ "step": 122
864
+ },
865
+ {
866
+ "epoch": 0.19400630914826497,
867
+ "grad_norm": 10.402134895324707,
868
+ "learning_rate": 2.805993690851735e-05,
869
+ "loss": 0.1807,
870
+ "step": 123
871
+ },
872
+ {
873
+ "epoch": 0.19558359621451105,
874
+ "grad_norm": 60.755088806152344,
875
+ "learning_rate": 2.8044164037854888e-05,
876
+ "loss": 2.021,
877
+ "step": 124
878
+ },
879
+ {
880
+ "epoch": 0.1971608832807571,
881
+ "grad_norm": 69.40438842773438,
882
+ "learning_rate": 2.8028391167192428e-05,
883
+ "loss": 1.243,
884
+ "step": 125
885
+ },
886
+ {
887
+ "epoch": 0.19873817034700317,
888
+ "grad_norm": 8.44308853149414,
889
+ "learning_rate": 2.8012618296529968e-05,
890
+ "loss": 0.0773,
891
+ "step": 126
892
+ },
893
+ {
894
+ "epoch": 0.20031545741324921,
895
+ "grad_norm": 13.563224792480469,
896
+ "learning_rate": 2.7996845425867508e-05,
897
+ "loss": 0.1639,
898
+ "step": 127
899
+ },
900
+ {
901
+ "epoch": 0.20189274447949526,
902
+ "grad_norm": 26.898422241210938,
903
+ "learning_rate": 2.7981072555205047e-05,
904
+ "loss": 0.6918,
905
+ "step": 128
906
+ },
907
+ {
908
+ "epoch": 0.20347003154574134,
909
+ "grad_norm": 11.048733711242676,
910
+ "learning_rate": 2.7965299684542587e-05,
911
+ "loss": 0.3772,
912
+ "step": 129
913
+ },
914
+ {
915
+ "epoch": 0.20504731861198738,
916
+ "grad_norm": 34.6102294921875,
917
+ "learning_rate": 2.7949526813880127e-05,
918
+ "loss": 1.3842,
919
+ "step": 130
920
+ },
921
+ {
922
+ "epoch": 0.20662460567823343,
923
+ "grad_norm": 12.371031761169434,
924
+ "learning_rate": 2.7933753943217667e-05,
925
+ "loss": 0.3462,
926
+ "step": 131
927
+ },
928
+ {
929
+ "epoch": 0.2082018927444795,
930
+ "grad_norm": 50.64473342895508,
931
+ "learning_rate": 2.7917981072555207e-05,
932
+ "loss": 1.8944,
933
+ "step": 132
934
+ },
935
+ {
936
+ "epoch": 0.20977917981072555,
937
+ "grad_norm": 20.688491821289062,
938
+ "learning_rate": 2.7902208201892746e-05,
939
+ "loss": 0.4056,
940
+ "step": 133
941
+ },
942
+ {
943
+ "epoch": 0.2113564668769716,
944
+ "grad_norm": 55.98980712890625,
945
+ "learning_rate": 2.7886435331230286e-05,
946
+ "loss": 1.3478,
947
+ "step": 134
948
+ },
949
+ {
950
+ "epoch": 0.21293375394321767,
951
+ "grad_norm": 47.27955627441406,
952
+ "learning_rate": 2.7870662460567823e-05,
953
+ "loss": 1.3232,
954
+ "step": 135
955
+ },
956
+ {
957
+ "epoch": 0.21451104100946372,
958
+ "grad_norm": 15.445263862609863,
959
+ "learning_rate": 2.7854889589905362e-05,
960
+ "loss": 0.1573,
961
+ "step": 136
962
+ },
963
+ {
964
+ "epoch": 0.21608832807570977,
965
+ "grad_norm": 38.09133529663086,
966
+ "learning_rate": 2.7839116719242902e-05,
967
+ "loss": 0.8674,
968
+ "step": 137
969
+ },
970
+ {
971
+ "epoch": 0.21766561514195584,
972
+ "grad_norm": 52.17962646484375,
973
+ "learning_rate": 2.7823343848580442e-05,
974
+ "loss": 0.925,
975
+ "step": 138
976
+ },
977
+ {
978
+ "epoch": 0.2192429022082019,
979
+ "grad_norm": 28.15349769592285,
980
+ "learning_rate": 2.7807570977917982e-05,
981
+ "loss": 0.5467,
982
+ "step": 139
983
+ },
984
+ {
985
+ "epoch": 0.22082018927444794,
986
+ "grad_norm": 46.26422119140625,
987
+ "learning_rate": 2.779179810725552e-05,
988
+ "loss": 0.9296,
989
+ "step": 140
990
+ },
991
+ {
992
+ "epoch": 0.222397476340694,
993
+ "grad_norm": 50.83304214477539,
994
+ "learning_rate": 2.777602523659306e-05,
995
+ "loss": 1.2947,
996
+ "step": 141
997
+ },
998
+ {
999
+ "epoch": 0.22397476340694006,
1000
+ "grad_norm": 23.36612319946289,
1001
+ "learning_rate": 2.77602523659306e-05,
1002
+ "loss": 0.3856,
1003
+ "step": 142
1004
+ },
1005
+ {
1006
+ "epoch": 0.22555205047318613,
1007
+ "grad_norm": 31.2121639251709,
1008
+ "learning_rate": 2.774447949526814e-05,
1009
+ "loss": 0.9097,
1010
+ "step": 143
1011
+ },
1012
+ {
1013
+ "epoch": 0.22712933753943218,
1014
+ "grad_norm": 16.758663177490234,
1015
+ "learning_rate": 2.772870662460568e-05,
1016
+ "loss": 0.5638,
1017
+ "step": 144
1018
+ },
1019
+ {
1020
+ "epoch": 0.22870662460567823,
1021
+ "grad_norm": 46.832515716552734,
1022
+ "learning_rate": 2.771293375394322e-05,
1023
+ "loss": 2.1622,
1024
+ "step": 145
1025
+ },
1026
+ {
1027
+ "epoch": 0.2302839116719243,
1028
+ "grad_norm": 19.408634185791016,
1029
+ "learning_rate": 2.769716088328076e-05,
1030
+ "loss": 0.3337,
1031
+ "step": 146
1032
+ },
1033
+ {
1034
+ "epoch": 0.23186119873817035,
1035
+ "grad_norm": 38.688804626464844,
1036
+ "learning_rate": 2.7681388012618297e-05,
1037
+ "loss": 1.2306,
1038
+ "step": 147
1039
+ },
1040
+ {
1041
+ "epoch": 0.2334384858044164,
1042
+ "grad_norm": 20.09441375732422,
1043
+ "learning_rate": 2.7665615141955836e-05,
1044
+ "loss": 0.1702,
1045
+ "step": 148
1046
+ },
1047
+ {
1048
+ "epoch": 0.23501577287066247,
1049
+ "grad_norm": 34.48636245727539,
1050
+ "learning_rate": 2.7649842271293376e-05,
1051
+ "loss": 1.3417,
1052
+ "step": 149
1053
+ },
1054
+ {
1055
+ "epoch": 0.23659305993690852,
1056
+ "grad_norm": 31.512447357177734,
1057
+ "learning_rate": 2.7634069400630916e-05,
1058
+ "loss": 1.2554,
1059
+ "step": 150
1060
+ },
1061
+ {
1062
+ "epoch": 0.23817034700315456,
1063
+ "grad_norm": 16.80870819091797,
1064
+ "learning_rate": 2.7618296529968456e-05,
1065
+ "loss": 0.2686,
1066
+ "step": 151
1067
+ },
1068
+ {
1069
+ "epoch": 0.23974763406940064,
1070
+ "grad_norm": 18.114761352539062,
1071
+ "learning_rate": 2.7602523659305996e-05,
1072
+ "loss": 0.2751,
1073
+ "step": 152
1074
+ },
1075
+ {
1076
+ "epoch": 0.24132492113564669,
1077
+ "grad_norm": 14.865362167358398,
1078
+ "learning_rate": 2.7586750788643532e-05,
1079
+ "loss": 0.3965,
1080
+ "step": 153
1081
+ },
1082
+ {
1083
+ "epoch": 0.24290220820189273,
1084
+ "grad_norm": 10.992881774902344,
1085
+ "learning_rate": 2.7570977917981072e-05,
1086
+ "loss": 0.2587,
1087
+ "step": 154
1088
+ },
1089
+ {
1090
+ "epoch": 0.2444794952681388,
1091
+ "grad_norm": 12.7821683883667,
1092
+ "learning_rate": 2.755520504731861e-05,
1093
+ "loss": 0.2284,
1094
+ "step": 155
1095
+ },
1096
+ {
1097
+ "epoch": 0.24605678233438485,
1098
+ "grad_norm": 35.8667106628418,
1099
+ "learning_rate": 2.753943217665615e-05,
1100
+ "loss": 0.9525,
1101
+ "step": 156
1102
+ },
1103
+ {
1104
+ "epoch": 0.2476340694006309,
1105
+ "grad_norm": 40.34475326538086,
1106
+ "learning_rate": 2.752365930599369e-05,
1107
+ "loss": 0.7339,
1108
+ "step": 157
1109
+ },
1110
+ {
1111
+ "epoch": 0.24921135646687698,
1112
+ "grad_norm": 32.349266052246094,
1113
+ "learning_rate": 2.750788643533123e-05,
1114
+ "loss": 1.3183,
1115
+ "step": 158
1116
+ },
1117
+ {
1118
+ "epoch": 0.250788643533123,
1119
+ "grad_norm": 30.02971839904785,
1120
+ "learning_rate": 2.749211356466877e-05,
1121
+ "loss": 1.1126,
1122
+ "step": 159
1123
+ },
1124
+ {
1125
+ "epoch": 0.25236593059936907,
1126
+ "grad_norm": 34.16777801513672,
1127
+ "learning_rate": 2.7476340694006307e-05,
1128
+ "loss": 0.6192,
1129
+ "step": 160
1130
+ },
1131
+ {
1132
+ "epoch": 0.2539432176656151,
1133
+ "grad_norm": 47.89134979248047,
1134
+ "learning_rate": 2.7460567823343847e-05,
1135
+ "loss": 1.6407,
1136
+ "step": 161
1137
+ },
1138
+ {
1139
+ "epoch": 0.2555205047318612,
1140
+ "grad_norm": 23.737993240356445,
1141
+ "learning_rate": 2.7444794952681387e-05,
1142
+ "loss": 0.7058,
1143
+ "step": 162
1144
+ },
1145
+ {
1146
+ "epoch": 0.25709779179810727,
1147
+ "grad_norm": 37.49821853637695,
1148
+ "learning_rate": 2.7429022082018927e-05,
1149
+ "loss": 1.0032,
1150
+ "step": 163
1151
+ },
1152
+ {
1153
+ "epoch": 0.2586750788643533,
1154
+ "grad_norm": 49.00496292114258,
1155
+ "learning_rate": 2.7413249211356466e-05,
1156
+ "loss": 1.4291,
1157
+ "step": 164
1158
+ },
1159
+ {
1160
+ "epoch": 0.26025236593059936,
1161
+ "grad_norm": 10.890347480773926,
1162
+ "learning_rate": 2.7397476340694006e-05,
1163
+ "loss": 0.1571,
1164
+ "step": 165
1165
+ },
1166
+ {
1167
+ "epoch": 0.2618296529968454,
1168
+ "grad_norm": 43.78799057006836,
1169
+ "learning_rate": 2.7381703470031546e-05,
1170
+ "loss": 1.458,
1171
+ "step": 166
1172
+ },
1173
+ {
1174
+ "epoch": 0.2634069400630915,
1175
+ "grad_norm": 44.582908630371094,
1176
+ "learning_rate": 2.7365930599369086e-05,
1177
+ "loss": 1.9312,
1178
+ "step": 167
1179
+ },
1180
+ {
1181
+ "epoch": 0.26498422712933756,
1182
+ "grad_norm": 23.406883239746094,
1183
+ "learning_rate": 2.7350157728706626e-05,
1184
+ "loss": 0.3137,
1185
+ "step": 168
1186
+ },
1187
+ {
1188
+ "epoch": 0.2665615141955836,
1189
+ "grad_norm": 30.169294357299805,
1190
+ "learning_rate": 2.7334384858044165e-05,
1191
+ "loss": 0.581,
1192
+ "step": 169
1193
+ },
1194
+ {
1195
+ "epoch": 0.26813880126182965,
1196
+ "grad_norm": 34.44562530517578,
1197
+ "learning_rate": 2.7318611987381705e-05,
1198
+ "loss": 0.9645,
1199
+ "step": 170
1200
+ },
1201
+ {
1202
+ "epoch": 0.2697160883280757,
1203
+ "grad_norm": 54.72190856933594,
1204
+ "learning_rate": 2.7302839116719245e-05,
1205
+ "loss": 3.7047,
1206
+ "step": 171
1207
+ },
1208
+ {
1209
+ "epoch": 0.27129337539432175,
1210
+ "grad_norm": 33.93403625488281,
1211
+ "learning_rate": 2.7287066246056785e-05,
1212
+ "loss": 1.6295,
1213
+ "step": 172
1214
+ },
1215
+ {
1216
+ "epoch": 0.27287066246056785,
1217
+ "grad_norm": 21.585771560668945,
1218
+ "learning_rate": 2.727129337539432e-05,
1219
+ "loss": 0.5426,
1220
+ "step": 173
1221
+ },
1222
+ {
1223
+ "epoch": 0.2744479495268139,
1224
+ "grad_norm": 18.893335342407227,
1225
+ "learning_rate": 2.725552050473186e-05,
1226
+ "loss": 0.7974,
1227
+ "step": 174
1228
+ },
1229
+ {
1230
+ "epoch": 0.27602523659305994,
1231
+ "grad_norm": 10.669084548950195,
1232
+ "learning_rate": 2.72397476340694e-05,
1233
+ "loss": 0.2674,
1234
+ "step": 175
1235
+ },
1236
+ {
1237
+ "epoch": 0.277602523659306,
1238
+ "grad_norm": 22.88224983215332,
1239
+ "learning_rate": 2.722397476340694e-05,
1240
+ "loss": 0.6445,
1241
+ "step": 176
1242
+ },
1243
+ {
1244
+ "epoch": 0.27917981072555204,
1245
+ "grad_norm": 27.347850799560547,
1246
+ "learning_rate": 2.720820189274448e-05,
1247
+ "loss": 1.1371,
1248
+ "step": 177
1249
+ },
1250
+ {
1251
+ "epoch": 0.2807570977917981,
1252
+ "grad_norm": 38.01303482055664,
1253
+ "learning_rate": 2.719242902208202e-05,
1254
+ "loss": 1.8414,
1255
+ "step": 178
1256
+ },
1257
+ {
1258
+ "epoch": 0.2823343848580442,
1259
+ "grad_norm": 21.528642654418945,
1260
+ "learning_rate": 2.717665615141956e-05,
1261
+ "loss": 1.1501,
1262
+ "step": 179
1263
+ },
1264
+ {
1265
+ "epoch": 0.28391167192429023,
1266
+ "grad_norm": 13.079411506652832,
1267
+ "learning_rate": 2.71608832807571e-05,
1268
+ "loss": 0.2832,
1269
+ "step": 180
1270
+ },
1271
+ {
1272
+ "epoch": 0.2854889589905363,
1273
+ "grad_norm": 17.025938034057617,
1274
+ "learning_rate": 2.714511041009464e-05,
1275
+ "loss": 0.6787,
1276
+ "step": 181
1277
+ },
1278
+ {
1279
+ "epoch": 0.2870662460567823,
1280
+ "grad_norm": 39.62239074707031,
1281
+ "learning_rate": 2.712933753943218e-05,
1282
+ "loss": 1.3267,
1283
+ "step": 182
1284
+ },
1285
+ {
1286
+ "epoch": 0.2886435331230284,
1287
+ "grad_norm": 41.821807861328125,
1288
+ "learning_rate": 2.711356466876972e-05,
1289
+ "loss": 1.2281,
1290
+ "step": 183
1291
+ },
1292
+ {
1293
+ "epoch": 0.2902208201892745,
1294
+ "grad_norm": 18.662702560424805,
1295
+ "learning_rate": 2.709779179810726e-05,
1296
+ "loss": 0.5277,
1297
+ "step": 184
1298
+ },
1299
+ {
1300
+ "epoch": 0.2917981072555205,
1301
+ "grad_norm": 25.26817512512207,
1302
+ "learning_rate": 2.7082018927444795e-05,
1303
+ "loss": 0.5292,
1304
+ "step": 185
1305
+ },
1306
+ {
1307
+ "epoch": 0.29337539432176657,
1308
+ "grad_norm": 46.763648986816406,
1309
+ "learning_rate": 2.7066246056782335e-05,
1310
+ "loss": 1.3355,
1311
+ "step": 186
1312
+ },
1313
+ {
1314
+ "epoch": 0.2949526813880126,
1315
+ "grad_norm": 16.308727264404297,
1316
+ "learning_rate": 2.705047318611987e-05,
1317
+ "loss": 0.5661,
1318
+ "step": 187
1319
+ },
1320
+ {
1321
+ "epoch": 0.29652996845425866,
1322
+ "grad_norm": 22.644695281982422,
1323
+ "learning_rate": 2.703470031545741e-05,
1324
+ "loss": 1.1766,
1325
+ "step": 188
1326
+ },
1327
+ {
1328
+ "epoch": 0.2981072555205047,
1329
+ "grad_norm": 9.70143985748291,
1330
+ "learning_rate": 2.701892744479495e-05,
1331
+ "loss": 0.1691,
1332
+ "step": 189
1333
+ },
1334
+ {
1335
+ "epoch": 0.2996845425867508,
1336
+ "grad_norm": 21.577924728393555,
1337
+ "learning_rate": 2.700315457413249e-05,
1338
+ "loss": 0.8777,
1339
+ "step": 190
1340
+ },
1341
+ {
1342
+ "epoch": 0.30126182965299686,
1343
+ "grad_norm": 43.18935012817383,
1344
+ "learning_rate": 2.698738170347003e-05,
1345
+ "loss": 1.5301,
1346
+ "step": 191
1347
+ },
1348
+ {
1349
+ "epoch": 0.3028391167192429,
1350
+ "grad_norm": 41.677059173583984,
1351
+ "learning_rate": 2.697160883280757e-05,
1352
+ "loss": 1.32,
1353
+ "step": 192
1354
+ },
1355
+ {
1356
+ "epoch": 0.30441640378548895,
1357
+ "grad_norm": 33.50912857055664,
1358
+ "learning_rate": 2.695583596214511e-05,
1359
+ "loss": 0.5321,
1360
+ "step": 193
1361
+ },
1362
+ {
1363
+ "epoch": 0.305993690851735,
1364
+ "grad_norm": 54.38176345825195,
1365
+ "learning_rate": 2.694006309148265e-05,
1366
+ "loss": 0.8883,
1367
+ "step": 194
1368
+ },
1369
+ {
1370
+ "epoch": 0.30757097791798105,
1371
+ "grad_norm": 27.214263916015625,
1372
+ "learning_rate": 2.692429022082019e-05,
1373
+ "loss": 0.6177,
1374
+ "step": 195
1375
+ },
1376
+ {
1377
+ "epoch": 0.30914826498422715,
1378
+ "grad_norm": 58.25315856933594,
1379
+ "learning_rate": 2.690851735015773e-05,
1380
+ "loss": 0.4771,
1381
+ "step": 196
1382
+ },
1383
+ {
1384
+ "epoch": 0.3107255520504732,
1385
+ "grad_norm": 26.127994537353516,
1386
+ "learning_rate": 2.689274447949527e-05,
1387
+ "loss": 0.2347,
1388
+ "step": 197
1389
+ },
1390
+ {
1391
+ "epoch": 0.31230283911671924,
1392
+ "grad_norm": 52.52097702026367,
1393
+ "learning_rate": 2.6876971608832806e-05,
1394
+ "loss": 1.1984,
1395
+ "step": 198
1396
+ },
1397
+ {
1398
+ "epoch": 0.3138801261829653,
1399
+ "grad_norm": 13.21312427520752,
1400
+ "learning_rate": 2.6861198738170346e-05,
1401
+ "loss": 0.8052,
1402
+ "step": 199
1403
+ },
1404
+ {
1405
+ "epoch": 0.31545741324921134,
1406
+ "grad_norm": 15.413599967956543,
1407
+ "learning_rate": 2.6845425867507886e-05,
1408
+ "loss": 0.2896,
1409
+ "step": 200
1410
+ },
1411
+ {
1412
+ "epoch": 0.31703470031545744,
1413
+ "grad_norm": 37.33637619018555,
1414
+ "learning_rate": 2.6829652996845425e-05,
1415
+ "loss": 0.8401,
1416
+ "step": 201
1417
+ },
1418
+ {
1419
+ "epoch": 0.3186119873817035,
1420
+ "grad_norm": 23.99006462097168,
1421
+ "learning_rate": 2.6813880126182965e-05,
1422
+ "loss": 1.0187,
1423
+ "step": 202
1424
+ },
1425
+ {
1426
+ "epoch": 0.32018927444794953,
1427
+ "grad_norm": 39.50027847290039,
1428
+ "learning_rate": 2.6798107255520505e-05,
1429
+ "loss": 0.974,
1430
+ "step": 203
1431
+ },
1432
+ {
1433
+ "epoch": 0.3217665615141956,
1434
+ "grad_norm": 28.93987274169922,
1435
+ "learning_rate": 2.6782334384858045e-05,
1436
+ "loss": 0.2533,
1437
+ "step": 204
1438
+ },
1439
+ {
1440
+ "epoch": 0.32334384858044163,
1441
+ "grad_norm": 41.88859176635742,
1442
+ "learning_rate": 2.6766561514195585e-05,
1443
+ "loss": 2.1219,
1444
+ "step": 205
1445
+ },
1446
+ {
1447
+ "epoch": 0.3249211356466877,
1448
+ "grad_norm": 45.398529052734375,
1449
+ "learning_rate": 2.6750788643533124e-05,
1450
+ "loss": 1.1059,
1451
+ "step": 206
1452
+ },
1453
+ {
1454
+ "epoch": 0.3264984227129338,
1455
+ "grad_norm": 13.585319519042969,
1456
+ "learning_rate": 2.6735015772870664e-05,
1457
+ "loss": 0.587,
1458
+ "step": 207
1459
+ },
1460
+ {
1461
+ "epoch": 0.3280757097791798,
1462
+ "grad_norm": 29.599796295166016,
1463
+ "learning_rate": 2.6719242902208204e-05,
1464
+ "loss": 0.7408,
1465
+ "step": 208
1466
+ },
1467
+ {
1468
+ "epoch": 0.32965299684542587,
1469
+ "grad_norm": 32.71330261230469,
1470
+ "learning_rate": 2.6703470031545744e-05,
1471
+ "loss": 0.4655,
1472
+ "step": 209
1473
+ },
1474
+ {
1475
+ "epoch": 0.3312302839116719,
1476
+ "grad_norm": 34.84718322753906,
1477
+ "learning_rate": 2.6687697160883284e-05,
1478
+ "loss": 0.4384,
1479
+ "step": 210
1480
+ },
1481
+ {
1482
+ "epoch": 0.33280757097791797,
1483
+ "grad_norm": 47.05462646484375,
1484
+ "learning_rate": 2.667192429022082e-05,
1485
+ "loss": 1.4324,
1486
+ "step": 211
1487
+ },
1488
+ {
1489
+ "epoch": 0.334384858044164,
1490
+ "grad_norm": 49.042449951171875,
1491
+ "learning_rate": 2.665615141955836e-05,
1492
+ "loss": 1.2471,
1493
+ "step": 212
1494
+ },
1495
+ {
1496
+ "epoch": 0.3359621451104101,
1497
+ "grad_norm": 20.12432861328125,
1498
+ "learning_rate": 2.66403785488959e-05,
1499
+ "loss": 0.5112,
1500
+ "step": 213
1501
+ },
1502
+ {
1503
+ "epoch": 0.33753943217665616,
1504
+ "grad_norm": 25.5250186920166,
1505
+ "learning_rate": 2.662460567823344e-05,
1506
+ "loss": 0.4473,
1507
+ "step": 214
1508
+ },
1509
+ {
1510
+ "epoch": 0.3391167192429022,
1511
+ "grad_norm": 30.16222381591797,
1512
+ "learning_rate": 2.660883280757098e-05,
1513
+ "loss": 1.1406,
1514
+ "step": 215
1515
+ },
1516
+ {
1517
+ "epoch": 0.34069400630914826,
1518
+ "grad_norm": 36.36207580566406,
1519
+ "learning_rate": 2.659305993690852e-05,
1520
+ "loss": 1.4083,
1521
+ "step": 216
1522
+ },
1523
+ {
1524
+ "epoch": 0.3422712933753943,
1525
+ "grad_norm": 45.23368835449219,
1526
+ "learning_rate": 2.657728706624606e-05,
1527
+ "loss": 1.3055,
1528
+ "step": 217
1529
+ },
1530
+ {
1531
+ "epoch": 0.3438485804416404,
1532
+ "grad_norm": 14.205089569091797,
1533
+ "learning_rate": 2.65615141955836e-05,
1534
+ "loss": 0.9753,
1535
+ "step": 218
1536
+ },
1537
+ {
1538
+ "epoch": 0.34542586750788645,
1539
+ "grad_norm": 48.683414459228516,
1540
+ "learning_rate": 2.654574132492114e-05,
1541
+ "loss": 2.0407,
1542
+ "step": 219
1543
+ },
1544
+ {
1545
+ "epoch": 0.3470031545741325,
1546
+ "grad_norm": 32.59965515136719,
1547
+ "learning_rate": 2.6529968454258678e-05,
1548
+ "loss": 0.7369,
1549
+ "step": 220
1550
+ },
1551
+ {
1552
+ "epoch": 0.34858044164037855,
1553
+ "grad_norm": 35.51063537597656,
1554
+ "learning_rate": 2.6514195583596215e-05,
1555
+ "loss": 1.9212,
1556
+ "step": 221
1557
+ },
1558
+ {
1559
+ "epoch": 0.3501577287066246,
1560
+ "grad_norm": 31.5261173248291,
1561
+ "learning_rate": 2.6498422712933754e-05,
1562
+ "loss": 0.6756,
1563
+ "step": 222
1564
+ },
1565
+ {
1566
+ "epoch": 0.35173501577287064,
1567
+ "grad_norm": 31.000041961669922,
1568
+ "learning_rate": 2.6482649842271294e-05,
1569
+ "loss": 1.1533,
1570
+ "step": 223
1571
+ },
1572
+ {
1573
+ "epoch": 0.35331230283911674,
1574
+ "grad_norm": 27.37296485900879,
1575
+ "learning_rate": 2.646687697160883e-05,
1576
+ "loss": 0.931,
1577
+ "step": 224
1578
+ },
1579
+ {
1580
+ "epoch": 0.3548895899053628,
1581
+ "grad_norm": 12.282959938049316,
1582
+ "learning_rate": 2.645110410094637e-05,
1583
+ "loss": 0.3929,
1584
+ "step": 225
1585
+ },
1586
+ {
1587
+ "epoch": 0.35646687697160884,
1588
+ "grad_norm": 19.884578704833984,
1589
+ "learning_rate": 2.643533123028391e-05,
1590
+ "loss": 0.753,
1591
+ "step": 226
1592
+ },
1593
+ {
1594
+ "epoch": 0.3580441640378549,
1595
+ "grad_norm": 50.77869415283203,
1596
+ "learning_rate": 2.641955835962145e-05,
1597
+ "loss": 1.7115,
1598
+ "step": 227
1599
+ },
1600
+ {
1601
+ "epoch": 0.35962145110410093,
1602
+ "grad_norm": 47.47655487060547,
1603
+ "learning_rate": 2.640378548895899e-05,
1604
+ "loss": 1.5204,
1605
+ "step": 228
1606
+ },
1607
+ {
1608
+ "epoch": 0.361198738170347,
1609
+ "grad_norm": 33.199005126953125,
1610
+ "learning_rate": 2.638801261829653e-05,
1611
+ "loss": 0.7791,
1612
+ "step": 229
1613
+ },
1614
+ {
1615
+ "epoch": 0.3627760252365931,
1616
+ "grad_norm": 21.32021713256836,
1617
+ "learning_rate": 2.637223974763407e-05,
1618
+ "loss": 0.366,
1619
+ "step": 230
1620
+ },
1621
+ {
1622
+ "epoch": 0.3643533123028391,
1623
+ "grad_norm": 14.767241477966309,
1624
+ "learning_rate": 2.635646687697161e-05,
1625
+ "loss": 0.3419,
1626
+ "step": 231
1627
+ },
1628
+ {
1629
+ "epoch": 0.3659305993690852,
1630
+ "grad_norm": 33.6385383605957,
1631
+ "learning_rate": 2.634069400630915e-05,
1632
+ "loss": 1.5042,
1633
+ "step": 232
1634
+ },
1635
+ {
1636
+ "epoch": 0.3675078864353312,
1637
+ "grad_norm": 42.69963455200195,
1638
+ "learning_rate": 2.632492113564669e-05,
1639
+ "loss": 0.9933,
1640
+ "step": 233
1641
+ },
1642
+ {
1643
+ "epoch": 0.36908517350157727,
1644
+ "grad_norm": 37.23275375366211,
1645
+ "learning_rate": 2.630914826498423e-05,
1646
+ "loss": 1.9391,
1647
+ "step": 234
1648
+ },
1649
+ {
1650
+ "epoch": 0.37066246056782337,
1651
+ "grad_norm": 20.5907039642334,
1652
+ "learning_rate": 2.629337539432177e-05,
1653
+ "loss": 0.9843,
1654
+ "step": 235
1655
+ },
1656
+ {
1657
+ "epoch": 0.3722397476340694,
1658
+ "grad_norm": 19.181324005126953,
1659
+ "learning_rate": 2.6277602523659308e-05,
1660
+ "loss": 0.885,
1661
+ "step": 236
1662
+ },
1663
+ {
1664
+ "epoch": 0.37381703470031546,
1665
+ "grad_norm": 46.60091018676758,
1666
+ "learning_rate": 2.6261829652996845e-05,
1667
+ "loss": 1.6384,
1668
+ "step": 237
1669
+ },
1670
+ {
1671
+ "epoch": 0.3753943217665615,
1672
+ "grad_norm": 24.8387393951416,
1673
+ "learning_rate": 2.6246056782334384e-05,
1674
+ "loss": 0.5566,
1675
+ "step": 238
1676
+ },
1677
+ {
1678
+ "epoch": 0.37697160883280756,
1679
+ "grad_norm": 21.524572372436523,
1680
+ "learning_rate": 2.6230283911671924e-05,
1681
+ "loss": 1.0613,
1682
+ "step": 239
1683
+ },
1684
+ {
1685
+ "epoch": 0.3785488958990536,
1686
+ "grad_norm": 54.12159729003906,
1687
+ "learning_rate": 2.6214511041009464e-05,
1688
+ "loss": 1.2448,
1689
+ "step": 240
1690
+ },
1691
+ {
1692
+ "epoch": 0.3801261829652997,
1693
+ "grad_norm": 34.552642822265625,
1694
+ "learning_rate": 2.6198738170347004e-05,
1695
+ "loss": 1.9525,
1696
+ "step": 241
1697
+ },
1698
+ {
1699
+ "epoch": 0.38170347003154576,
1700
+ "grad_norm": 47.69414138793945,
1701
+ "learning_rate": 2.6182965299684544e-05,
1702
+ "loss": 1.7205,
1703
+ "step": 242
1704
+ },
1705
+ {
1706
+ "epoch": 0.3832807570977918,
1707
+ "grad_norm": 25.08366584777832,
1708
+ "learning_rate": 2.6167192429022083e-05,
1709
+ "loss": 0.7122,
1710
+ "step": 243
1711
+ },
1712
+ {
1713
+ "epoch": 0.38485804416403785,
1714
+ "grad_norm": 37.48938751220703,
1715
+ "learning_rate": 2.6151419558359623e-05,
1716
+ "loss": 0.8283,
1717
+ "step": 244
1718
+ },
1719
+ {
1720
+ "epoch": 0.3864353312302839,
1721
+ "grad_norm": 31.735260009765625,
1722
+ "learning_rate": 2.6135646687697163e-05,
1723
+ "loss": 1.7742,
1724
+ "step": 245
1725
+ },
1726
+ {
1727
+ "epoch": 0.38801261829652994,
1728
+ "grad_norm": 17.06745719909668,
1729
+ "learning_rate": 2.6119873817034703e-05,
1730
+ "loss": 0.435,
1731
+ "step": 246
1732
+ },
1733
+ {
1734
+ "epoch": 0.38958990536277605,
1735
+ "grad_norm": 15.29838752746582,
1736
+ "learning_rate": 2.6104100946372243e-05,
1737
+ "loss": 0.6465,
1738
+ "step": 247
1739
+ },
1740
+ {
1741
+ "epoch": 0.3911671924290221,
1742
+ "grad_norm": 28.84938621520996,
1743
+ "learning_rate": 2.6088328075709782e-05,
1744
+ "loss": 0.7146,
1745
+ "step": 248
1746
+ },
1747
+ {
1748
+ "epoch": 0.39274447949526814,
1749
+ "grad_norm": 53.88107681274414,
1750
+ "learning_rate": 2.607255520504732e-05,
1751
+ "loss": 1.0231,
1752
+ "step": 249
1753
+ },
1754
+ {
1755
+ "epoch": 0.3943217665615142,
1756
+ "grad_norm": 31.747798919677734,
1757
+ "learning_rate": 2.605678233438486e-05,
1758
+ "loss": 1.0606,
1759
+ "step": 250
1760
+ },
1761
+ {
1762
+ "epoch": 0.39589905362776023,
1763
+ "grad_norm": 50.53217697143555,
1764
+ "learning_rate": 2.60410094637224e-05,
1765
+ "loss": 1.688,
1766
+ "step": 251
1767
+ },
1768
+ {
1769
+ "epoch": 0.39747634069400634,
1770
+ "grad_norm": 28.37627601623535,
1771
+ "learning_rate": 2.6025236593059938e-05,
1772
+ "loss": 1.5603,
1773
+ "step": 252
1774
+ },
1775
+ {
1776
+ "epoch": 0.3990536277602524,
1777
+ "grad_norm": 17.82233428955078,
1778
+ "learning_rate": 2.6009463722397478e-05,
1779
+ "loss": 0.6615,
1780
+ "step": 253
1781
+ },
1782
+ {
1783
+ "epoch": 0.40063091482649843,
1784
+ "grad_norm": 13.466266632080078,
1785
+ "learning_rate": 2.5993690851735018e-05,
1786
+ "loss": 0.5135,
1787
+ "step": 254
1788
+ },
1789
+ {
1790
+ "epoch": 0.4022082018927445,
1791
+ "grad_norm": 484.76800537109375,
1792
+ "learning_rate": 2.5977917981072554e-05,
1793
+ "loss": 1.0092,
1794
+ "step": 255
1795
+ },
1796
+ {
1797
+ "epoch": 0.4037854889589905,
1798
+ "grad_norm": 37.61044692993164,
1799
+ "learning_rate": 2.5962145110410094e-05,
1800
+ "loss": 1.9792,
1801
+ "step": 256
1802
+ },
1803
+ {
1804
+ "epoch": 0.40536277602523657,
1805
+ "grad_norm": 41.90039825439453,
1806
+ "learning_rate": 2.5946372239747634e-05,
1807
+ "loss": 1.3666,
1808
+ "step": 257
1809
+ },
1810
+ {
1811
+ "epoch": 0.4069400630914827,
1812
+ "grad_norm": 41.69484329223633,
1813
+ "learning_rate": 2.5930599369085174e-05,
1814
+ "loss": 1.0425,
1815
+ "step": 258
1816
+ },
1817
+ {
1818
+ "epoch": 0.4085173501577287,
1819
+ "grad_norm": 30.779823303222656,
1820
+ "learning_rate": 2.5914826498422713e-05,
1821
+ "loss": 0.5538,
1822
+ "step": 259
1823
+ },
1824
+ {
1825
+ "epoch": 0.41009463722397477,
1826
+ "grad_norm": 37.00507354736328,
1827
+ "learning_rate": 2.5899053627760253e-05,
1828
+ "loss": 1.049,
1829
+ "step": 260
1830
+ },
1831
+ {
1832
+ "epoch": 0.4116719242902208,
1833
+ "grad_norm": 38.79997253417969,
1834
+ "learning_rate": 2.5883280757097793e-05,
1835
+ "loss": 0.5168,
1836
+ "step": 261
1837
+ },
1838
+ {
1839
+ "epoch": 0.41324921135646686,
1840
+ "grad_norm": 48.37226867675781,
1841
+ "learning_rate": 2.586750788643533e-05,
1842
+ "loss": 1.2198,
1843
+ "step": 262
1844
+ },
1845
+ {
1846
+ "epoch": 0.4148264984227129,
1847
+ "grad_norm": 29.85272979736328,
1848
+ "learning_rate": 2.585173501577287e-05,
1849
+ "loss": 0.8429,
1850
+ "step": 263
1851
+ },
1852
+ {
1853
+ "epoch": 0.416403785488959,
1854
+ "grad_norm": 19.446401596069336,
1855
+ "learning_rate": 2.583596214511041e-05,
1856
+ "loss": 0.4121,
1857
+ "step": 264
1858
+ },
1859
+ {
1860
+ "epoch": 0.41798107255520506,
1861
+ "grad_norm": 10.472084999084473,
1862
+ "learning_rate": 2.582018927444795e-05,
1863
+ "loss": 0.3234,
1864
+ "step": 265
1865
+ },
1866
+ {
1867
+ "epoch": 0.4195583596214511,
1868
+ "grad_norm": 33.94881820678711,
1869
+ "learning_rate": 2.580441640378549e-05,
1870
+ "loss": 1.6782,
1871
+ "step": 266
1872
+ },
1873
+ {
1874
+ "epoch": 0.42113564668769715,
1875
+ "grad_norm": 23.22264862060547,
1876
+ "learning_rate": 2.5788643533123028e-05,
1877
+ "loss": 1.3006,
1878
+ "step": 267
1879
+ },
1880
+ {
1881
+ "epoch": 0.4227129337539432,
1882
+ "grad_norm": 50.05253982543945,
1883
+ "learning_rate": 2.5772870662460568e-05,
1884
+ "loss": 2.5931,
1885
+ "step": 268
1886
+ },
1887
+ {
1888
+ "epoch": 0.4242902208201893,
1889
+ "grad_norm": 37.375465393066406,
1890
+ "learning_rate": 2.5757097791798108e-05,
1891
+ "loss": 1.0532,
1892
+ "step": 269
1893
+ },
1894
+ {
1895
+ "epoch": 0.42586750788643535,
1896
+ "grad_norm": 28.370405197143555,
1897
+ "learning_rate": 2.5741324921135648e-05,
1898
+ "loss": 0.5641,
1899
+ "step": 270
1900
+ },
1901
+ {
1902
+ "epoch": 0.4274447949526814,
1903
+ "grad_norm": 24.498916625976562,
1904
+ "learning_rate": 2.5725552050473187e-05,
1905
+ "loss": 0.9805,
1906
+ "step": 271
1907
+ },
1908
+ {
1909
+ "epoch": 0.42902208201892744,
1910
+ "grad_norm": 25.199342727661133,
1911
+ "learning_rate": 2.5709779179810727e-05,
1912
+ "loss": 0.6465,
1913
+ "step": 272
1914
+ },
1915
+ {
1916
+ "epoch": 0.4305993690851735,
1917
+ "grad_norm": 23.403337478637695,
1918
+ "learning_rate": 2.5694006309148267e-05,
1919
+ "loss": 0.7342,
1920
+ "step": 273
1921
+ },
1922
+ {
1923
+ "epoch": 0.43217665615141954,
1924
+ "grad_norm": 12.828851699829102,
1925
+ "learning_rate": 2.5678233438485807e-05,
1926
+ "loss": 0.5867,
1927
+ "step": 274
1928
+ },
1929
+ {
1930
+ "epoch": 0.43375394321766564,
1931
+ "grad_norm": 19.755699157714844,
1932
+ "learning_rate": 2.5662460567823343e-05,
1933
+ "loss": 0.6565,
1934
+ "step": 275
1935
+ },
1936
+ {
1937
+ "epoch": 0.4353312302839117,
1938
+ "grad_norm": 70.69967651367188,
1939
+ "learning_rate": 2.5646687697160883e-05,
1940
+ "loss": 1.3669,
1941
+ "step": 276
1942
+ },
1943
+ {
1944
+ "epoch": 0.43690851735015773,
1945
+ "grad_norm": 16.497039794921875,
1946
+ "learning_rate": 2.5630914826498423e-05,
1947
+ "loss": 0.3703,
1948
+ "step": 277
1949
+ },
1950
+ {
1951
+ "epoch": 0.4384858044164038,
1952
+ "grad_norm": 36.59125900268555,
1953
+ "learning_rate": 2.5615141955835963e-05,
1954
+ "loss": 1.2435,
1955
+ "step": 278
1956
+ },
1957
+ {
1958
+ "epoch": 0.4400630914826498,
1959
+ "grad_norm": 24.055252075195312,
1960
+ "learning_rate": 2.5599369085173502e-05,
1961
+ "loss": 0.5956,
1962
+ "step": 279
1963
+ },
1964
+ {
1965
+ "epoch": 0.4416403785488959,
1966
+ "grad_norm": 52.90360641479492,
1967
+ "learning_rate": 2.5583596214511042e-05,
1968
+ "loss": 1.8063,
1969
+ "step": 280
1970
+ },
1971
+ {
1972
+ "epoch": 0.443217665615142,
1973
+ "grad_norm": 33.3657112121582,
1974
+ "learning_rate": 2.5567823343848582e-05,
1975
+ "loss": 0.7502,
1976
+ "step": 281
1977
+ },
1978
+ {
1979
+ "epoch": 0.444794952681388,
1980
+ "grad_norm": 15.864405632019043,
1981
+ "learning_rate": 2.5552050473186122e-05,
1982
+ "loss": 0.6157,
1983
+ "step": 282
1984
+ },
1985
+ {
1986
+ "epoch": 0.44637223974763407,
1987
+ "grad_norm": 10.833452224731445,
1988
+ "learning_rate": 2.553627760252366e-05,
1989
+ "loss": 0.153,
1990
+ "step": 283
1991
+ },
1992
+ {
1993
+ "epoch": 0.4479495268138801,
1994
+ "grad_norm": 14.914109230041504,
1995
+ "learning_rate": 2.55205047318612e-05,
1996
+ "loss": 0.498,
1997
+ "step": 284
1998
+ },
1999
+ {
2000
+ "epoch": 0.44952681388012616,
2001
+ "grad_norm": 15.300338745117188,
2002
+ "learning_rate": 2.550473186119874e-05,
2003
+ "loss": 0.2256,
2004
+ "step": 285
2005
+ },
2006
+ {
2007
+ "epoch": 0.45110410094637227,
2008
+ "grad_norm": 26.07958221435547,
2009
+ "learning_rate": 2.548895899053628e-05,
2010
+ "loss": 0.5726,
2011
+ "step": 286
2012
+ },
2013
+ {
2014
+ "epoch": 0.4526813880126183,
2015
+ "grad_norm": 32.02494430541992,
2016
+ "learning_rate": 2.547318611987382e-05,
2017
+ "loss": 2.3079,
2018
+ "step": 287
2019
+ },
2020
+ {
2021
+ "epoch": 0.45425867507886436,
2022
+ "grad_norm": 19.21434211730957,
2023
+ "learning_rate": 2.5457413249211357e-05,
2024
+ "loss": 0.6546,
2025
+ "step": 288
2026
+ },
2027
+ {
2028
+ "epoch": 0.4558359621451104,
2029
+ "grad_norm": 32.230712890625,
2030
+ "learning_rate": 2.5441640378548894e-05,
2031
+ "loss": 1.0985,
2032
+ "step": 289
2033
+ },
2034
+ {
2035
+ "epoch": 0.45741324921135645,
2036
+ "grad_norm": 10.254904747009277,
2037
+ "learning_rate": 2.5425867507886433e-05,
2038
+ "loss": 0.2305,
2039
+ "step": 290
2040
+ },
2041
+ {
2042
+ "epoch": 0.4589905362776025,
2043
+ "grad_norm": 41.209861755371094,
2044
+ "learning_rate": 2.5410094637223973e-05,
2045
+ "loss": 1.3266,
2046
+ "step": 291
2047
+ },
2048
+ {
2049
+ "epoch": 0.4605678233438486,
2050
+ "grad_norm": 38.17776107788086,
2051
+ "learning_rate": 2.5394321766561513e-05,
2052
+ "loss": 1.1599,
2053
+ "step": 292
2054
+ },
2055
+ {
2056
+ "epoch": 0.46214511041009465,
2057
+ "grad_norm": 31.44940757751465,
2058
+ "learning_rate": 2.5378548895899053e-05,
2059
+ "loss": 0.7327,
2060
+ "step": 293
2061
+ },
2062
+ {
2063
+ "epoch": 0.4637223974763407,
2064
+ "grad_norm": 65.65186309814453,
2065
+ "learning_rate": 2.5362776025236593e-05,
2066
+ "loss": 0.915,
2067
+ "step": 294
2068
+ },
2069
+ {
2070
+ "epoch": 0.46529968454258674,
2071
+ "grad_norm": 45.166690826416016,
2072
+ "learning_rate": 2.5347003154574132e-05,
2073
+ "loss": 0.9291,
2074
+ "step": 295
2075
+ },
2076
+ {
2077
+ "epoch": 0.4668769716088328,
2078
+ "grad_norm": 40.72503662109375,
2079
+ "learning_rate": 2.5331230283911672e-05,
2080
+ "loss": 0.6219,
2081
+ "step": 296
2082
+ },
2083
+ {
2084
+ "epoch": 0.46845425867507884,
2085
+ "grad_norm": 11.175806045532227,
2086
+ "learning_rate": 2.5315457413249212e-05,
2087
+ "loss": 0.288,
2088
+ "step": 297
2089
+ },
2090
+ {
2091
+ "epoch": 0.47003154574132494,
2092
+ "grad_norm": 55.50349426269531,
2093
+ "learning_rate": 2.5299684542586752e-05,
2094
+ "loss": 0.9238,
2095
+ "step": 298
2096
+ },
2097
+ {
2098
+ "epoch": 0.471608832807571,
2099
+ "grad_norm": 60.939979553222656,
2100
+ "learning_rate": 2.528391167192429e-05,
2101
+ "loss": 1.5635,
2102
+ "step": 299
2103
+ },
2104
+ {
2105
+ "epoch": 0.47318611987381703,
2106
+ "grad_norm": 46.01141357421875,
2107
+ "learning_rate": 2.526813880126183e-05,
2108
+ "loss": 0.7667,
2109
+ "step": 300
2110
+ },
2111
+ {
2112
+ "epoch": 0.4747634069400631,
2113
+ "grad_norm": 45.75855255126953,
2114
+ "learning_rate": 2.5252365930599368e-05,
2115
+ "loss": 1.1335,
2116
+ "step": 301
2117
+ },
2118
+ {
2119
+ "epoch": 0.47634069400630913,
2120
+ "grad_norm": 40.72056579589844,
2121
+ "learning_rate": 2.5236593059936908e-05,
2122
+ "loss": 0.7474,
2123
+ "step": 302
2124
+ },
2125
+ {
2126
+ "epoch": 0.47791798107255523,
2127
+ "grad_norm": 35.266719818115234,
2128
+ "learning_rate": 2.5220820189274447e-05,
2129
+ "loss": 1.0378,
2130
+ "step": 303
2131
+ },
2132
+ {
2133
+ "epoch": 0.4794952681388013,
2134
+ "grad_norm": 57.24043273925781,
2135
+ "learning_rate": 2.5205047318611987e-05,
2136
+ "loss": 1.8198,
2137
+ "step": 304
2138
+ },
2139
+ {
2140
+ "epoch": 0.4810725552050473,
2141
+ "grad_norm": 16.190948486328125,
2142
+ "learning_rate": 2.5189274447949527e-05,
2143
+ "loss": 0.9752,
2144
+ "step": 305
2145
+ },
2146
+ {
2147
+ "epoch": 0.48264984227129337,
2148
+ "grad_norm": 27.43843650817871,
2149
+ "learning_rate": 2.5173501577287067e-05,
2150
+ "loss": 1.1502,
2151
+ "step": 306
2152
+ },
2153
+ {
2154
+ "epoch": 0.4842271293375394,
2155
+ "grad_norm": 12.933015823364258,
2156
+ "learning_rate": 2.5157728706624607e-05,
2157
+ "loss": 0.2031,
2158
+ "step": 307
2159
+ },
2160
+ {
2161
+ "epoch": 0.48580441640378547,
2162
+ "grad_norm": 22.37410545349121,
2163
+ "learning_rate": 2.5141955835962146e-05,
2164
+ "loss": 0.711,
2165
+ "step": 308
2166
+ },
2167
+ {
2168
+ "epoch": 0.48738170347003157,
2169
+ "grad_norm": 41.96189880371094,
2170
+ "learning_rate": 2.5126182965299686e-05,
2171
+ "loss": 1.4064,
2172
+ "step": 309
2173
+ },
2174
+ {
2175
+ "epoch": 0.4889589905362776,
2176
+ "grad_norm": 33.80104446411133,
2177
+ "learning_rate": 2.5110410094637226e-05,
2178
+ "loss": 1.4332,
2179
+ "step": 310
2180
+ },
2181
+ {
2182
+ "epoch": 0.49053627760252366,
2183
+ "grad_norm": 47.28125762939453,
2184
+ "learning_rate": 2.5094637223974766e-05,
2185
+ "loss": 1.5073,
2186
+ "step": 311
2187
+ },
2188
+ {
2189
+ "epoch": 0.4921135646687697,
2190
+ "grad_norm": 22.467430114746094,
2191
+ "learning_rate": 2.5078864353312306e-05,
2192
+ "loss": 1.2673,
2193
+ "step": 312
2194
+ },
2195
+ {
2196
+ "epoch": 0.49369085173501576,
2197
+ "grad_norm": 20.2239933013916,
2198
+ "learning_rate": 2.5063091482649842e-05,
2199
+ "loss": 0.6753,
2200
+ "step": 313
2201
+ },
2202
+ {
2203
+ "epoch": 0.4952681388012618,
2204
+ "grad_norm": 39.26515579223633,
2205
+ "learning_rate": 2.5047318611987382e-05,
2206
+ "loss": 1.047,
2207
+ "step": 314
2208
+ },
2209
+ {
2210
+ "epoch": 0.4968454258675079,
2211
+ "grad_norm": 12.582530975341797,
2212
+ "learning_rate": 2.503154574132492e-05,
2213
+ "loss": 0.2427,
2214
+ "step": 315
2215
+ },
2216
+ {
2217
+ "epoch": 0.49842271293375395,
2218
+ "grad_norm": 23.123321533203125,
2219
+ "learning_rate": 2.501577287066246e-05,
2220
+ "loss": 0.692,
2221
+ "step": 316
2222
+ },
2223
+ {
2224
+ "epoch": 0.5,
2225
+ "grad_norm": 26.053495407104492,
2226
+ "learning_rate": 2.5e-05,
2227
+ "loss": 0.4189,
2228
+ "step": 317
2229
+ },
2230
+ {
2231
+ "epoch": 0.501577287066246,
2232
+ "grad_norm": 16.502792358398438,
2233
+ "learning_rate": 2.498422712933754e-05,
2234
+ "loss": 0.91,
2235
+ "step": 318
2236
+ },
2237
+ {
2238
+ "epoch": 0.5031545741324921,
2239
+ "grad_norm": 38.79825210571289,
2240
+ "learning_rate": 2.496845425867508e-05,
2241
+ "loss": 0.4931,
2242
+ "step": 319
2243
+ },
2244
+ {
2245
+ "epoch": 0.5047318611987381,
2246
+ "grad_norm": 44.205421447753906,
2247
+ "learning_rate": 2.495268138801262e-05,
2248
+ "loss": 0.6371,
2249
+ "step": 320
2250
+ },
2251
+ {
2252
+ "epoch": 0.5063091482649842,
2253
+ "grad_norm": 53.30904006958008,
2254
+ "learning_rate": 2.493690851735016e-05,
2255
+ "loss": 2.8242,
2256
+ "step": 321
2257
+ },
2258
+ {
2259
+ "epoch": 0.5078864353312302,
2260
+ "grad_norm": 30.18499183654785,
2261
+ "learning_rate": 2.49211356466877e-05,
2262
+ "loss": 1.0873,
2263
+ "step": 322
2264
+ },
2265
+ {
2266
+ "epoch": 0.5094637223974764,
2267
+ "grad_norm": 26.88262367248535,
2268
+ "learning_rate": 2.4905362776025237e-05,
2269
+ "loss": 0.8978,
2270
+ "step": 323
2271
+ },
2272
+ {
2273
+ "epoch": 0.5110410094637224,
2274
+ "grad_norm": 17.68815040588379,
2275
+ "learning_rate": 2.4889589905362776e-05,
2276
+ "loss": 0.5498,
2277
+ "step": 324
2278
+ },
2279
+ {
2280
+ "epoch": 0.5126182965299685,
2281
+ "grad_norm": 17.32225227355957,
2282
+ "learning_rate": 2.4873817034700316e-05,
2283
+ "loss": 0.6921,
2284
+ "step": 325
2285
+ },
2286
+ {
2287
+ "epoch": 0.5141955835962145,
2288
+ "grad_norm": 18.670616149902344,
2289
+ "learning_rate": 2.4858044164037853e-05,
2290
+ "loss": 0.5688,
2291
+ "step": 326
2292
+ },
2293
+ {
2294
+ "epoch": 0.5157728706624606,
2295
+ "grad_norm": 39.22971725463867,
2296
+ "learning_rate": 2.4842271293375392e-05,
2297
+ "loss": 1.5304,
2298
+ "step": 327
2299
+ },
2300
+ {
2301
+ "epoch": 0.5173501577287066,
2302
+ "grad_norm": 41.706138610839844,
2303
+ "learning_rate": 2.4826498422712932e-05,
2304
+ "loss": 1.015,
2305
+ "step": 328
2306
+ },
2307
+ {
2308
+ "epoch": 0.5189274447949527,
2309
+ "grad_norm": 40.609073638916016,
2310
+ "learning_rate": 2.4810725552050472e-05,
2311
+ "loss": 1.0275,
2312
+ "step": 329
2313
+ },
2314
+ {
2315
+ "epoch": 0.5205047318611987,
2316
+ "grad_norm": 47.36741638183594,
2317
+ "learning_rate": 2.4794952681388012e-05,
2318
+ "loss": 0.9339,
2319
+ "step": 330
2320
+ },
2321
+ {
2322
+ "epoch": 0.5220820189274448,
2323
+ "grad_norm": 36.778499603271484,
2324
+ "learning_rate": 2.477917981072555e-05,
2325
+ "loss": 1.7758,
2326
+ "step": 331
2327
+ },
2328
+ {
2329
+ "epoch": 0.5236593059936908,
2330
+ "grad_norm": 36.07444381713867,
2331
+ "learning_rate": 2.476340694006309e-05,
2332
+ "loss": 1.3567,
2333
+ "step": 332
2334
+ },
2335
+ {
2336
+ "epoch": 0.5252365930599369,
2337
+ "grad_norm": 33.40023422241211,
2338
+ "learning_rate": 2.474763406940063e-05,
2339
+ "loss": 0.798,
2340
+ "step": 333
2341
+ },
2342
+ {
2343
+ "epoch": 0.526813880126183,
2344
+ "grad_norm": 37.33317947387695,
2345
+ "learning_rate": 2.473186119873817e-05,
2346
+ "loss": 1.2016,
2347
+ "step": 334
2348
+ },
2349
+ {
2350
+ "epoch": 0.5283911671924291,
2351
+ "grad_norm": 21.062698364257812,
2352
+ "learning_rate": 2.471608832807571e-05,
2353
+ "loss": 0.9024,
2354
+ "step": 335
2355
+ },
2356
+ {
2357
+ "epoch": 0.5299684542586751,
2358
+ "grad_norm": 39.730003356933594,
2359
+ "learning_rate": 2.470031545741325e-05,
2360
+ "loss": 1.4152,
2361
+ "step": 336
2362
+ },
2363
+ {
2364
+ "epoch": 0.5315457413249212,
2365
+ "grad_norm": 36.75265121459961,
2366
+ "learning_rate": 2.468454258675079e-05,
2367
+ "loss": 1.6933,
2368
+ "step": 337
2369
+ },
2370
+ {
2371
+ "epoch": 0.5331230283911672,
2372
+ "grad_norm": 42.32905197143555,
2373
+ "learning_rate": 2.466876971608833e-05,
2374
+ "loss": 0.6659,
2375
+ "step": 338
2376
+ },
2377
+ {
2378
+ "epoch": 0.5347003154574133,
2379
+ "grad_norm": 9.659038543701172,
2380
+ "learning_rate": 2.4652996845425867e-05,
2381
+ "loss": 0.1977,
2382
+ "step": 339
2383
+ },
2384
+ {
2385
+ "epoch": 0.5362776025236593,
2386
+ "grad_norm": 23.717344284057617,
2387
+ "learning_rate": 2.4637223974763406e-05,
2388
+ "loss": 0.7252,
2389
+ "step": 340
2390
+ },
2391
+ {
2392
+ "epoch": 0.5378548895899053,
2393
+ "grad_norm": 10.212286949157715,
2394
+ "learning_rate": 2.4621451104100946e-05,
2395
+ "loss": 0.2439,
2396
+ "step": 341
2397
+ },
2398
+ {
2399
+ "epoch": 0.5394321766561514,
2400
+ "grad_norm": 26.001909255981445,
2401
+ "learning_rate": 2.4605678233438486e-05,
2402
+ "loss": 0.3295,
2403
+ "step": 342
2404
+ },
2405
+ {
2406
+ "epoch": 0.5410094637223974,
2407
+ "grad_norm": 39.621437072753906,
2408
+ "learning_rate": 2.4589905362776026e-05,
2409
+ "loss": 1.5258,
2410
+ "step": 343
2411
+ },
2412
+ {
2413
+ "epoch": 0.5425867507886435,
2414
+ "grad_norm": 27.942232131958008,
2415
+ "learning_rate": 2.4574132492113566e-05,
2416
+ "loss": 0.88,
2417
+ "step": 344
2418
+ },
2419
+ {
2420
+ "epoch": 0.5441640378548895,
2421
+ "grad_norm": 18.207975387573242,
2422
+ "learning_rate": 2.4558359621451105e-05,
2423
+ "loss": 0.5134,
2424
+ "step": 345
2425
+ },
2426
+ {
2427
+ "epoch": 0.5457413249211357,
2428
+ "grad_norm": 31.57217788696289,
2429
+ "learning_rate": 2.4542586750788645e-05,
2430
+ "loss": 0.7175,
2431
+ "step": 346
2432
+ },
2433
+ {
2434
+ "epoch": 0.5473186119873817,
2435
+ "grad_norm": 18.8095703125,
2436
+ "learning_rate": 2.4526813880126185e-05,
2437
+ "loss": 0.5689,
2438
+ "step": 347
2439
+ },
2440
+ {
2441
+ "epoch": 0.5488958990536278,
2442
+ "grad_norm": 33.10816192626953,
2443
+ "learning_rate": 2.4511041009463725e-05,
2444
+ "loss": 2.1179,
2445
+ "step": 348
2446
+ },
2447
+ {
2448
+ "epoch": 0.5504731861198738,
2449
+ "grad_norm": 30.903202056884766,
2450
+ "learning_rate": 2.4495268138801265e-05,
2451
+ "loss": 0.5791,
2452
+ "step": 349
2453
+ },
2454
+ {
2455
+ "epoch": 0.5520504731861199,
2456
+ "grad_norm": 25.809715270996094,
2457
+ "learning_rate": 2.4479495268138804e-05,
2458
+ "loss": 0.4853,
2459
+ "step": 350
2460
+ },
2461
+ {
2462
+ "epoch": 0.5536277602523659,
2463
+ "grad_norm": 60.03340530395508,
2464
+ "learning_rate": 2.446372239747634e-05,
2465
+ "loss": 1.7992,
2466
+ "step": 351
2467
+ },
2468
+ {
2469
+ "epoch": 0.555205047318612,
2470
+ "grad_norm": 37.23116683959961,
2471
+ "learning_rate": 2.444794952681388e-05,
2472
+ "loss": 1.0519,
2473
+ "step": 352
2474
+ },
2475
+ {
2476
+ "epoch": 0.556782334384858,
2477
+ "grad_norm": 9.769363403320312,
2478
+ "learning_rate": 2.443217665615142e-05,
2479
+ "loss": 0.1357,
2480
+ "step": 353
2481
+ },
2482
+ {
2483
+ "epoch": 0.5583596214511041,
2484
+ "grad_norm": 24.00067901611328,
2485
+ "learning_rate": 2.441640378548896e-05,
2486
+ "loss": 0.9584,
2487
+ "step": 354
2488
+ },
2489
+ {
2490
+ "epoch": 0.5599369085173501,
2491
+ "grad_norm": 38.65916442871094,
2492
+ "learning_rate": 2.44006309148265e-05,
2493
+ "loss": 0.9789,
2494
+ "step": 355
2495
+ },
2496
+ {
2497
+ "epoch": 0.5615141955835962,
2498
+ "grad_norm": 15.097743034362793,
2499
+ "learning_rate": 2.438485804416404e-05,
2500
+ "loss": 0.3946,
2501
+ "step": 356
2502
+ },
2503
+ {
2504
+ "epoch": 0.5630914826498423,
2505
+ "grad_norm": 25.418912887573242,
2506
+ "learning_rate": 2.4369085173501576e-05,
2507
+ "loss": 1.1448,
2508
+ "step": 357
2509
+ },
2510
+ {
2511
+ "epoch": 0.5646687697160884,
2512
+ "grad_norm": 14.713468551635742,
2513
+ "learning_rate": 2.4353312302839116e-05,
2514
+ "loss": 0.4717,
2515
+ "step": 358
2516
+ },
2517
+ {
2518
+ "epoch": 0.5662460567823344,
2519
+ "grad_norm": 54.493621826171875,
2520
+ "learning_rate": 2.4337539432176656e-05,
2521
+ "loss": 1.5193,
2522
+ "step": 359
2523
+ },
2524
+ {
2525
+ "epoch": 0.5678233438485805,
2526
+ "grad_norm": 38.58003616333008,
2527
+ "learning_rate": 2.4321766561514196e-05,
2528
+ "loss": 1.5989,
2529
+ "step": 360
2530
+ },
2531
+ {
2532
+ "epoch": 0.5694006309148265,
2533
+ "grad_norm": 47.13187789916992,
2534
+ "learning_rate": 2.4305993690851735e-05,
2535
+ "loss": 0.748,
2536
+ "step": 361
2537
+ },
2538
+ {
2539
+ "epoch": 0.5709779179810726,
2540
+ "grad_norm": 39.339996337890625,
2541
+ "learning_rate": 2.4290220820189275e-05,
2542
+ "loss": 1.3707,
2543
+ "step": 362
2544
+ },
2545
+ {
2546
+ "epoch": 0.5725552050473186,
2547
+ "grad_norm": 30.47089195251465,
2548
+ "learning_rate": 2.4274447949526815e-05,
2549
+ "loss": 0.706,
2550
+ "step": 363
2551
+ },
2552
+ {
2553
+ "epoch": 0.5741324921135647,
2554
+ "grad_norm": 45.00666046142578,
2555
+ "learning_rate": 2.425867507886435e-05,
2556
+ "loss": 1.0671,
2557
+ "step": 364
2558
+ },
2559
+ {
2560
+ "epoch": 0.5757097791798107,
2561
+ "grad_norm": 23.93462562561035,
2562
+ "learning_rate": 2.424290220820189e-05,
2563
+ "loss": 0.6378,
2564
+ "step": 365
2565
+ },
2566
+ {
2567
+ "epoch": 0.5772870662460567,
2568
+ "grad_norm": 13.337026596069336,
2569
+ "learning_rate": 2.422712933753943e-05,
2570
+ "loss": 0.4105,
2571
+ "step": 366
2572
+ },
2573
+ {
2574
+ "epoch": 0.5788643533123028,
2575
+ "grad_norm": 38.95447540283203,
2576
+ "learning_rate": 2.421135646687697e-05,
2577
+ "loss": 1.1968,
2578
+ "step": 367
2579
+ },
2580
+ {
2581
+ "epoch": 0.580441640378549,
2582
+ "grad_norm": 11.434366226196289,
2583
+ "learning_rate": 2.419558359621451e-05,
2584
+ "loss": 0.2125,
2585
+ "step": 368
2586
+ },
2587
+ {
2588
+ "epoch": 0.582018927444795,
2589
+ "grad_norm": 41.746978759765625,
2590
+ "learning_rate": 2.417981072555205e-05,
2591
+ "loss": 0.6557,
2592
+ "step": 369
2593
+ },
2594
+ {
2595
+ "epoch": 0.583596214511041,
2596
+ "grad_norm": 11.216280937194824,
2597
+ "learning_rate": 2.416403785488959e-05,
2598
+ "loss": 0.405,
2599
+ "step": 370
2600
+ },
2601
+ {
2602
+ "epoch": 0.5851735015772871,
2603
+ "grad_norm": 24.09681510925293,
2604
+ "learning_rate": 2.414826498422713e-05,
2605
+ "loss": 0.8603,
2606
+ "step": 371
2607
+ },
2608
+ {
2609
+ "epoch": 0.5867507886435331,
2610
+ "grad_norm": 44.140377044677734,
2611
+ "learning_rate": 2.413249211356467e-05,
2612
+ "loss": 0.8732,
2613
+ "step": 372
2614
+ },
2615
+ {
2616
+ "epoch": 0.5883280757097792,
2617
+ "grad_norm": 45.55453109741211,
2618
+ "learning_rate": 2.411671924290221e-05,
2619
+ "loss": 0.5431,
2620
+ "step": 373
2621
+ },
2622
+ {
2623
+ "epoch": 0.5899053627760252,
2624
+ "grad_norm": 45.09754180908203,
2625
+ "learning_rate": 2.410094637223975e-05,
2626
+ "loss": 1.2394,
2627
+ "step": 374
2628
+ },
2629
+ {
2630
+ "epoch": 0.5914826498422713,
2631
+ "grad_norm": 23.333829879760742,
2632
+ "learning_rate": 2.408517350157729e-05,
2633
+ "loss": 0.4794,
2634
+ "step": 375
2635
+ },
2636
+ {
2637
+ "epoch": 0.5930599369085173,
2638
+ "grad_norm": 22.66351318359375,
2639
+ "learning_rate": 2.406940063091483e-05,
2640
+ "loss": 0.4481,
2641
+ "step": 376
2642
+ },
2643
+ {
2644
+ "epoch": 0.5946372239747634,
2645
+ "grad_norm": 37.45612335205078,
2646
+ "learning_rate": 2.4053627760252365e-05,
2647
+ "loss": 1.1188,
2648
+ "step": 377
2649
+ },
2650
+ {
2651
+ "epoch": 0.5962145110410094,
2652
+ "grad_norm": 25.51999282836914,
2653
+ "learning_rate": 2.4037854889589905e-05,
2654
+ "loss": 0.6079,
2655
+ "step": 378
2656
+ },
2657
+ {
2658
+ "epoch": 0.5977917981072555,
2659
+ "grad_norm": 32.34231185913086,
2660
+ "learning_rate": 2.4022082018927445e-05,
2661
+ "loss": 1.113,
2662
+ "step": 379
2663
+ },
2664
+ {
2665
+ "epoch": 0.5993690851735016,
2666
+ "grad_norm": 13.737470626831055,
2667
+ "learning_rate": 2.4006309148264985e-05,
2668
+ "loss": 0.2727,
2669
+ "step": 380
2670
+ },
2671
+ {
2672
+ "epoch": 0.6009463722397477,
2673
+ "grad_norm": 20.944168090820312,
2674
+ "learning_rate": 2.3990536277602524e-05,
2675
+ "loss": 0.3578,
2676
+ "step": 381
2677
+ },
2678
+ {
2679
+ "epoch": 0.6025236593059937,
2680
+ "grad_norm": 47.74925231933594,
2681
+ "learning_rate": 2.3974763406940064e-05,
2682
+ "loss": 2.4277,
2683
+ "step": 382
2684
+ },
2685
+ {
2686
+ "epoch": 0.6041009463722398,
2687
+ "grad_norm": 41.04169464111328,
2688
+ "learning_rate": 2.3958990536277604e-05,
2689
+ "loss": 1.3982,
2690
+ "step": 383
2691
+ },
2692
+ {
2693
+ "epoch": 0.6056782334384858,
2694
+ "grad_norm": 14.79502010345459,
2695
+ "learning_rate": 2.3943217665615144e-05,
2696
+ "loss": 0.2372,
2697
+ "step": 384
2698
+ },
2699
+ {
2700
+ "epoch": 0.6072555205047319,
2701
+ "grad_norm": 36.966896057128906,
2702
+ "learning_rate": 2.3927444794952684e-05,
2703
+ "loss": 1.1239,
2704
+ "step": 385
2705
+ },
2706
+ {
2707
+ "epoch": 0.6088328075709779,
2708
+ "grad_norm": 54.31798553466797,
2709
+ "learning_rate": 2.3911671924290223e-05,
2710
+ "loss": 1.4241,
2711
+ "step": 386
2712
+ },
2713
+ {
2714
+ "epoch": 0.610410094637224,
2715
+ "grad_norm": 37.95793533325195,
2716
+ "learning_rate": 2.3895899053627763e-05,
2717
+ "loss": 0.7553,
2718
+ "step": 387
2719
+ },
2720
+ {
2721
+ "epoch": 0.61198738170347,
2722
+ "grad_norm": 33.62470626831055,
2723
+ "learning_rate": 2.3880126182965303e-05,
2724
+ "loss": 0.8521,
2725
+ "step": 388
2726
+ },
2727
+ {
2728
+ "epoch": 0.613564668769716,
2729
+ "grad_norm": 44.78007125854492,
2730
+ "learning_rate": 2.3864353312302843e-05,
2731
+ "loss": 0.9662,
2732
+ "step": 389
2733
+ },
2734
+ {
2735
+ "epoch": 0.6151419558359621,
2736
+ "grad_norm": 17.080902099609375,
2737
+ "learning_rate": 2.384858044164038e-05,
2738
+ "loss": 0.6368,
2739
+ "step": 390
2740
+ },
2741
+ {
2742
+ "epoch": 0.6167192429022083,
2743
+ "grad_norm": 15.024806022644043,
2744
+ "learning_rate": 2.3832807570977916e-05,
2745
+ "loss": 0.6649,
2746
+ "step": 391
2747
+ },
2748
+ {
2749
+ "epoch": 0.6182965299684543,
2750
+ "grad_norm": 16.2637882232666,
2751
+ "learning_rate": 2.3817034700315455e-05,
2752
+ "loss": 0.5265,
2753
+ "step": 392
2754
+ },
2755
+ {
2756
+ "epoch": 0.6198738170347003,
2757
+ "grad_norm": 35.20927047729492,
2758
+ "learning_rate": 2.3801261829652995e-05,
2759
+ "loss": 1.2348,
2760
+ "step": 393
2761
+ },
2762
+ {
2763
+ "epoch": 0.6214511041009464,
2764
+ "grad_norm": 24.13702392578125,
2765
+ "learning_rate": 2.3785488958990535e-05,
2766
+ "loss": 0.7744,
2767
+ "step": 394
2768
+ },
2769
+ {
2770
+ "epoch": 0.6230283911671924,
2771
+ "grad_norm": 28.485437393188477,
2772
+ "learning_rate": 2.3769716088328075e-05,
2773
+ "loss": 0.4917,
2774
+ "step": 395
2775
+ },
2776
+ {
2777
+ "epoch": 0.6246056782334385,
2778
+ "grad_norm": 26.22724151611328,
2779
+ "learning_rate": 2.3753943217665615e-05,
2780
+ "loss": 0.9761,
2781
+ "step": 396
2782
+ },
2783
+ {
2784
+ "epoch": 0.6261829652996845,
2785
+ "grad_norm": 53.20966339111328,
2786
+ "learning_rate": 2.3738170347003154e-05,
2787
+ "loss": 1.4422,
2788
+ "step": 397
2789
+ },
2790
+ {
2791
+ "epoch": 0.6277602523659306,
2792
+ "grad_norm": 17.00950050354004,
2793
+ "learning_rate": 2.3722397476340694e-05,
2794
+ "loss": 0.6759,
2795
+ "step": 398
2796
+ },
2797
+ {
2798
+ "epoch": 0.6293375394321766,
2799
+ "grad_norm": 12.88286304473877,
2800
+ "learning_rate": 2.3706624605678234e-05,
2801
+ "loss": 0.317,
2802
+ "step": 399
2803
+ },
2804
+ {
2805
+ "epoch": 0.6309148264984227,
2806
+ "grad_norm": 7.572042942047119,
2807
+ "learning_rate": 2.3690851735015774e-05,
2808
+ "loss": 0.173,
2809
+ "step": 400
2810
+ },
2811
+ {
2812
+ "epoch": 0.6324921135646687,
2813
+ "grad_norm": 12.461848258972168,
2814
+ "learning_rate": 2.3675078864353314e-05,
2815
+ "loss": 0.1698,
2816
+ "step": 401
2817
+ },
2818
+ {
2819
+ "epoch": 0.6340694006309149,
2820
+ "grad_norm": 56.61540222167969,
2821
+ "learning_rate": 2.3659305993690853e-05,
2822
+ "loss": 0.6604,
2823
+ "step": 402
2824
+ },
2825
+ {
2826
+ "epoch": 0.6356466876971609,
2827
+ "grad_norm": 43.43484115600586,
2828
+ "learning_rate": 2.364353312302839e-05,
2829
+ "loss": 0.4951,
2830
+ "step": 403
2831
+ },
2832
+ {
2833
+ "epoch": 0.637223974763407,
2834
+ "grad_norm": 32.40031433105469,
2835
+ "learning_rate": 2.362776025236593e-05,
2836
+ "loss": 0.8648,
2837
+ "step": 404
2838
+ },
2839
+ {
2840
+ "epoch": 0.638801261829653,
2841
+ "grad_norm": 46.143062591552734,
2842
+ "learning_rate": 2.361198738170347e-05,
2843
+ "loss": 1.2006,
2844
+ "step": 405
2845
+ },
2846
+ {
2847
+ "epoch": 0.6403785488958991,
2848
+ "grad_norm": 36.997711181640625,
2849
+ "learning_rate": 2.359621451104101e-05,
2850
+ "loss": 0.6304,
2851
+ "step": 406
2852
+ },
2853
+ {
2854
+ "epoch": 0.6419558359621451,
2855
+ "grad_norm": 36.918148040771484,
2856
+ "learning_rate": 2.358044164037855e-05,
2857
+ "loss": 1.0931,
2858
+ "step": 407
2859
+ },
2860
+ {
2861
+ "epoch": 0.6435331230283912,
2862
+ "grad_norm": 32.16216278076172,
2863
+ "learning_rate": 2.356466876971609e-05,
2864
+ "loss": 0.6256,
2865
+ "step": 408
2866
+ },
2867
+ {
2868
+ "epoch": 0.6451104100946372,
2869
+ "grad_norm": 39.54144287109375,
2870
+ "learning_rate": 2.354889589905363e-05,
2871
+ "loss": 0.9816,
2872
+ "step": 409
2873
+ },
2874
+ {
2875
+ "epoch": 0.6466876971608833,
2876
+ "grad_norm": 46.52165222167969,
2877
+ "learning_rate": 2.353312302839117e-05,
2878
+ "loss": 1.3758,
2879
+ "step": 410
2880
+ },
2881
+ {
2882
+ "epoch": 0.6482649842271293,
2883
+ "grad_norm": 25.173683166503906,
2884
+ "learning_rate": 2.3517350157728708e-05,
2885
+ "loss": 0.8422,
2886
+ "step": 411
2887
+ },
2888
+ {
2889
+ "epoch": 0.6498422712933754,
2890
+ "grad_norm": 33.97991943359375,
2891
+ "learning_rate": 2.3501577287066248e-05,
2892
+ "loss": 0.932,
2893
+ "step": 412
2894
+ },
2895
+ {
2896
+ "epoch": 0.6514195583596214,
2897
+ "grad_norm": 35.27119827270508,
2898
+ "learning_rate": 2.3485804416403788e-05,
2899
+ "loss": 0.4381,
2900
+ "step": 413
2901
+ },
2902
+ {
2903
+ "epoch": 0.6529968454258676,
2904
+ "grad_norm": 40.9283447265625,
2905
+ "learning_rate": 2.3470031545741328e-05,
2906
+ "loss": 1.7348,
2907
+ "step": 414
2908
+ },
2909
+ {
2910
+ "epoch": 0.6545741324921136,
2911
+ "grad_norm": 29.246122360229492,
2912
+ "learning_rate": 2.3454258675078864e-05,
2913
+ "loss": 0.8435,
2914
+ "step": 415
2915
+ },
2916
+ {
2917
+ "epoch": 0.6561514195583596,
2918
+ "grad_norm": 58.37071228027344,
2919
+ "learning_rate": 2.3438485804416404e-05,
2920
+ "loss": 1.7078,
2921
+ "step": 416
2922
+ },
2923
+ {
2924
+ "epoch": 0.6577287066246057,
2925
+ "grad_norm": 27.45276641845703,
2926
+ "learning_rate": 2.3422712933753944e-05,
2927
+ "loss": 1.0732,
2928
+ "step": 417
2929
+ },
2930
+ {
2931
+ "epoch": 0.6593059936908517,
2932
+ "grad_norm": 31.824932098388672,
2933
+ "learning_rate": 2.3406940063091483e-05,
2934
+ "loss": 0.89,
2935
+ "step": 418
2936
+ },
2937
+ {
2938
+ "epoch": 0.6608832807570978,
2939
+ "grad_norm": 23.44261360168457,
2940
+ "learning_rate": 2.3391167192429023e-05,
2941
+ "loss": 0.4742,
2942
+ "step": 419
2943
+ },
2944
+ {
2945
+ "epoch": 0.6624605678233438,
2946
+ "grad_norm": 28.967531204223633,
2947
+ "learning_rate": 2.3375394321766563e-05,
2948
+ "loss": 1.0092,
2949
+ "step": 420
2950
+ },
2951
+ {
2952
+ "epoch": 0.6640378548895899,
2953
+ "grad_norm": 17.737171173095703,
2954
+ "learning_rate": 2.3359621451104103e-05,
2955
+ "loss": 0.6408,
2956
+ "step": 421
2957
+ },
2958
+ {
2959
+ "epoch": 0.6656151419558359,
2960
+ "grad_norm": 14.883234977722168,
2961
+ "learning_rate": 2.3343848580441643e-05,
2962
+ "loss": 0.6469,
2963
+ "step": 422
2964
+ },
2965
+ {
2966
+ "epoch": 0.667192429022082,
2967
+ "grad_norm": 24.25818634033203,
2968
+ "learning_rate": 2.3328075709779182e-05,
2969
+ "loss": 0.5914,
2970
+ "step": 423
2971
+ },
2972
+ {
2973
+ "epoch": 0.668769716088328,
2974
+ "grad_norm": 47.20992660522461,
2975
+ "learning_rate": 2.3312302839116722e-05,
2976
+ "loss": 1.0264,
2977
+ "step": 424
2978
+ },
2979
+ {
2980
+ "epoch": 0.6703470031545742,
2981
+ "grad_norm": 52.51894760131836,
2982
+ "learning_rate": 2.329652996845426e-05,
2983
+ "loss": 0.889,
2984
+ "step": 425
2985
+ },
2986
+ {
2987
+ "epoch": 0.6719242902208202,
2988
+ "grad_norm": 21.03269386291504,
2989
+ "learning_rate": 2.32807570977918e-05,
2990
+ "loss": 0.4338,
2991
+ "step": 426
2992
+ },
2993
+ {
2994
+ "epoch": 0.6735015772870663,
2995
+ "grad_norm": 46.70388412475586,
2996
+ "learning_rate": 2.3264984227129338e-05,
2997
+ "loss": 0.5682,
2998
+ "step": 427
2999
+ },
3000
+ {
3001
+ "epoch": 0.6750788643533123,
3002
+ "grad_norm": 38.72499084472656,
3003
+ "learning_rate": 2.3249211356466875e-05,
3004
+ "loss": 1.0972,
3005
+ "step": 428
3006
+ },
3007
+ {
3008
+ "epoch": 0.6766561514195584,
3009
+ "grad_norm": 32.51113510131836,
3010
+ "learning_rate": 2.3233438485804414e-05,
3011
+ "loss": 0.816,
3012
+ "step": 429
3013
+ },
3014
+ {
3015
+ "epoch": 0.6782334384858044,
3016
+ "grad_norm": 41.63493728637695,
3017
+ "learning_rate": 2.3217665615141954e-05,
3018
+ "loss": 1.4058,
3019
+ "step": 430
3020
+ },
3021
+ {
3022
+ "epoch": 0.6798107255520505,
3023
+ "grad_norm": 26.526561737060547,
3024
+ "learning_rate": 2.3201892744479494e-05,
3025
+ "loss": 0.7838,
3026
+ "step": 431
3027
+ },
3028
+ {
3029
+ "epoch": 0.6813880126182965,
3030
+ "grad_norm": 23.691131591796875,
3031
+ "learning_rate": 2.3186119873817034e-05,
3032
+ "loss": 0.8224,
3033
+ "step": 432
3034
+ },
3035
+ {
3036
+ "epoch": 0.6829652996845426,
3037
+ "grad_norm": 25.56239891052246,
3038
+ "learning_rate": 2.3170347003154574e-05,
3039
+ "loss": 0.6592,
3040
+ "step": 433
3041
+ },
3042
+ {
3043
+ "epoch": 0.6845425867507886,
3044
+ "grad_norm": 24.03537368774414,
3045
+ "learning_rate": 2.3154574132492113e-05,
3046
+ "loss": 0.8962,
3047
+ "step": 434
3048
+ },
3049
+ {
3050
+ "epoch": 0.6861198738170347,
3051
+ "grad_norm": 9.816405296325684,
3052
+ "learning_rate": 2.3138801261829653e-05,
3053
+ "loss": 0.3473,
3054
+ "step": 435
3055
+ },
3056
+ {
3057
+ "epoch": 0.6876971608832808,
3058
+ "grad_norm": 27.27027130126953,
3059
+ "learning_rate": 2.3123028391167193e-05,
3060
+ "loss": 0.7573,
3061
+ "step": 436
3062
+ },
3063
+ {
3064
+ "epoch": 0.6892744479495269,
3065
+ "grad_norm": 39.65195083618164,
3066
+ "learning_rate": 2.3107255520504733e-05,
3067
+ "loss": 0.5212,
3068
+ "step": 437
3069
+ },
3070
+ {
3071
+ "epoch": 0.6908517350157729,
3072
+ "grad_norm": 24.415267944335938,
3073
+ "learning_rate": 2.3091482649842273e-05,
3074
+ "loss": 0.9492,
3075
+ "step": 438
3076
+ },
3077
+ {
3078
+ "epoch": 0.692429022082019,
3079
+ "grad_norm": 22.194141387939453,
3080
+ "learning_rate": 2.3075709779179812e-05,
3081
+ "loss": 0.5267,
3082
+ "step": 439
3083
+ },
3084
+ {
3085
+ "epoch": 0.694006309148265,
3086
+ "grad_norm": 25.846446990966797,
3087
+ "learning_rate": 2.3059936908517352e-05,
3088
+ "loss": 0.5913,
3089
+ "step": 440
3090
+ },
3091
+ {
3092
+ "epoch": 0.695583596214511,
3093
+ "grad_norm": 41.77247619628906,
3094
+ "learning_rate": 2.304416403785489e-05,
3095
+ "loss": 1.1313,
3096
+ "step": 441
3097
+ },
3098
+ {
3099
+ "epoch": 0.6971608832807571,
3100
+ "grad_norm": 55.44232177734375,
3101
+ "learning_rate": 2.302839116719243e-05,
3102
+ "loss": 1.6034,
3103
+ "step": 442
3104
+ },
3105
+ {
3106
+ "epoch": 0.6987381703470031,
3107
+ "grad_norm": 17.483383178710938,
3108
+ "learning_rate": 2.3012618296529968e-05,
3109
+ "loss": 0.7665,
3110
+ "step": 443
3111
+ },
3112
+ {
3113
+ "epoch": 0.7003154574132492,
3114
+ "grad_norm": 34.683719635009766,
3115
+ "learning_rate": 2.2996845425867508e-05,
3116
+ "loss": 1.1795,
3117
+ "step": 444
3118
+ },
3119
+ {
3120
+ "epoch": 0.7018927444794952,
3121
+ "grad_norm": 28.64350700378418,
3122
+ "learning_rate": 2.2981072555205048e-05,
3123
+ "loss": 0.7955,
3124
+ "step": 445
3125
+ },
3126
+ {
3127
+ "epoch": 0.7034700315457413,
3128
+ "grad_norm": 30.405139923095703,
3129
+ "learning_rate": 2.2965299684542588e-05,
3130
+ "loss": 0.8045,
3131
+ "step": 446
3132
+ },
3133
+ {
3134
+ "epoch": 0.7050473186119873,
3135
+ "grad_norm": 27.87291717529297,
3136
+ "learning_rate": 2.2949526813880127e-05,
3137
+ "loss": 0.3252,
3138
+ "step": 447
3139
+ },
3140
+ {
3141
+ "epoch": 0.7066246056782335,
3142
+ "grad_norm": 25.16892433166504,
3143
+ "learning_rate": 2.2933753943217667e-05,
3144
+ "loss": 1.1456,
3145
+ "step": 448
3146
+ },
3147
+ {
3148
+ "epoch": 0.7082018927444795,
3149
+ "grad_norm": 18.941091537475586,
3150
+ "learning_rate": 2.2917981072555207e-05,
3151
+ "loss": 0.4488,
3152
+ "step": 449
3153
+ },
3154
+ {
3155
+ "epoch": 0.7097791798107256,
3156
+ "grad_norm": 11.795750617980957,
3157
+ "learning_rate": 2.2902208201892747e-05,
3158
+ "loss": 0.2912,
3159
+ "step": 450
3160
+ },
3161
+ {
3162
+ "epoch": 0.7113564668769716,
3163
+ "grad_norm": 75.78955078125,
3164
+ "learning_rate": 2.2886435331230287e-05,
3165
+ "loss": 1.4565,
3166
+ "step": 451
3167
+ },
3168
+ {
3169
+ "epoch": 0.7129337539432177,
3170
+ "grad_norm": 29.61087417602539,
3171
+ "learning_rate": 2.2870662460567826e-05,
3172
+ "loss": 1.0244,
3173
+ "step": 452
3174
+ },
3175
+ {
3176
+ "epoch": 0.7145110410094637,
3177
+ "grad_norm": 31.563507080078125,
3178
+ "learning_rate": 2.2854889589905366e-05,
3179
+ "loss": 0.5423,
3180
+ "step": 453
3181
+ },
3182
+ {
3183
+ "epoch": 0.7160883280757098,
3184
+ "grad_norm": 13.91758918762207,
3185
+ "learning_rate": 2.2839116719242903e-05,
3186
+ "loss": 0.2068,
3187
+ "step": 454
3188
+ },
3189
+ {
3190
+ "epoch": 0.7176656151419558,
3191
+ "grad_norm": 44.59634017944336,
3192
+ "learning_rate": 2.2823343848580442e-05,
3193
+ "loss": 1.2257,
3194
+ "step": 455
3195
+ },
3196
+ {
3197
+ "epoch": 0.7192429022082019,
3198
+ "grad_norm": 18.660573959350586,
3199
+ "learning_rate": 2.2807570977917982e-05,
3200
+ "loss": 0.7588,
3201
+ "step": 456
3202
+ },
3203
+ {
3204
+ "epoch": 0.7208201892744479,
3205
+ "grad_norm": 48.425846099853516,
3206
+ "learning_rate": 2.2791798107255522e-05,
3207
+ "loss": 0.9525,
3208
+ "step": 457
3209
+ },
3210
+ {
3211
+ "epoch": 0.722397476340694,
3212
+ "grad_norm": 32.65686798095703,
3213
+ "learning_rate": 2.2776025236593062e-05,
3214
+ "loss": 1.0382,
3215
+ "step": 458
3216
+ },
3217
+ {
3218
+ "epoch": 0.7239747634069401,
3219
+ "grad_norm": 30.510356903076172,
3220
+ "learning_rate": 2.2760252365930598e-05,
3221
+ "loss": 1.2827,
3222
+ "step": 459
3223
+ },
3224
+ {
3225
+ "epoch": 0.7255520504731862,
3226
+ "grad_norm": 38.7618293762207,
3227
+ "learning_rate": 2.2744479495268138e-05,
3228
+ "loss": 0.785,
3229
+ "step": 460
3230
+ },
3231
+ {
3232
+ "epoch": 0.7271293375394322,
3233
+ "grad_norm": 21.054779052734375,
3234
+ "learning_rate": 2.2728706624605678e-05,
3235
+ "loss": 0.8533,
3236
+ "step": 461
3237
+ },
3238
+ {
3239
+ "epoch": 0.7287066246056783,
3240
+ "grad_norm": 20.868669509887695,
3241
+ "learning_rate": 2.2712933753943218e-05,
3242
+ "loss": 0.2325,
3243
+ "step": 462
3244
+ },
3245
+ {
3246
+ "epoch": 0.7302839116719243,
3247
+ "grad_norm": 8.442056655883789,
3248
+ "learning_rate": 2.2697160883280757e-05,
3249
+ "loss": 0.3217,
3250
+ "step": 463
3251
+ },
3252
+ {
3253
+ "epoch": 0.7318611987381703,
3254
+ "grad_norm": 30.54999542236328,
3255
+ "learning_rate": 2.2681388012618297e-05,
3256
+ "loss": 1.2833,
3257
+ "step": 464
3258
+ },
3259
+ {
3260
+ "epoch": 0.7334384858044164,
3261
+ "grad_norm": 40.20530700683594,
3262
+ "learning_rate": 2.2665615141955837e-05,
3263
+ "loss": 0.8746,
3264
+ "step": 465
3265
+ },
3266
+ {
3267
+ "epoch": 0.7350157728706624,
3268
+ "grad_norm": 133.62002563476562,
3269
+ "learning_rate": 2.2649842271293373e-05,
3270
+ "loss": 1.1047,
3271
+ "step": 466
3272
+ },
3273
+ {
3274
+ "epoch": 0.7365930599369085,
3275
+ "grad_norm": 84.88129425048828,
3276
+ "learning_rate": 2.2634069400630913e-05,
3277
+ "loss": 1.9077,
3278
+ "step": 467
3279
+ },
3280
+ {
3281
+ "epoch": 0.7381703470031545,
3282
+ "grad_norm": 52.1624641418457,
3283
+ "learning_rate": 2.2618296529968453e-05,
3284
+ "loss": 1.59,
3285
+ "step": 468
3286
+ },
3287
+ {
3288
+ "epoch": 0.7397476340694006,
3289
+ "grad_norm": 38.160274505615234,
3290
+ "learning_rate": 2.2602523659305993e-05,
3291
+ "loss": 0.9177,
3292
+ "step": 469
3293
+ },
3294
+ {
3295
+ "epoch": 0.7413249211356467,
3296
+ "grad_norm": 19.550050735473633,
3297
+ "learning_rate": 2.2586750788643533e-05,
3298
+ "loss": 0.2459,
3299
+ "step": 470
3300
+ },
3301
+ {
3302
+ "epoch": 0.7429022082018928,
3303
+ "grad_norm": 44.474613189697266,
3304
+ "learning_rate": 2.2570977917981072e-05,
3305
+ "loss": 0.5768,
3306
+ "step": 471
3307
+ },
3308
+ {
3309
+ "epoch": 0.7444794952681388,
3310
+ "grad_norm": 25.13191032409668,
3311
+ "learning_rate": 2.2555205047318612e-05,
3312
+ "loss": 1.3645,
3313
+ "step": 472
3314
+ },
3315
+ {
3316
+ "epoch": 0.7460567823343849,
3317
+ "grad_norm": 22.508243560791016,
3318
+ "learning_rate": 2.2539432176656152e-05,
3319
+ "loss": 0.4959,
3320
+ "step": 473
3321
+ },
3322
+ {
3323
+ "epoch": 0.7476340694006309,
3324
+ "grad_norm": 27.79073715209961,
3325
+ "learning_rate": 2.2523659305993692e-05,
3326
+ "loss": 0.2589,
3327
+ "step": 474
3328
+ },
3329
+ {
3330
+ "epoch": 0.749211356466877,
3331
+ "grad_norm": 60.65824508666992,
3332
+ "learning_rate": 2.250788643533123e-05,
3333
+ "loss": 1.135,
3334
+ "step": 475
3335
+ },
3336
+ {
3337
+ "epoch": 0.750788643533123,
3338
+ "grad_norm": 42.320220947265625,
3339
+ "learning_rate": 2.249211356466877e-05,
3340
+ "loss": 1.0794,
3341
+ "step": 476
3342
+ },
3343
+ {
3344
+ "epoch": 0.7523659305993691,
3345
+ "grad_norm": 30.03775405883789,
3346
+ "learning_rate": 2.247634069400631e-05,
3347
+ "loss": 1.2921,
3348
+ "step": 477
3349
+ },
3350
+ {
3351
+ "epoch": 0.7539432176656151,
3352
+ "grad_norm": 35.67447280883789,
3353
+ "learning_rate": 2.246056782334385e-05,
3354
+ "loss": 1.2011,
3355
+ "step": 478
3356
+ },
3357
+ {
3358
+ "epoch": 0.7555205047318612,
3359
+ "grad_norm": 14.670977592468262,
3360
+ "learning_rate": 2.2444794952681387e-05,
3361
+ "loss": 0.68,
3362
+ "step": 479
3363
+ },
3364
+ {
3365
+ "epoch": 0.7570977917981072,
3366
+ "grad_norm": 40.297821044921875,
3367
+ "learning_rate": 2.2429022082018927e-05,
3368
+ "loss": 1.3634,
3369
+ "step": 480
3370
+ },
3371
+ {
3372
+ "epoch": 0.7586750788643533,
3373
+ "grad_norm": 33.14999771118164,
3374
+ "learning_rate": 2.2413249211356467e-05,
3375
+ "loss": 0.4286,
3376
+ "step": 481
3377
+ },
3378
+ {
3379
+ "epoch": 0.7602523659305994,
3380
+ "grad_norm": 52.47095489501953,
3381
+ "learning_rate": 2.2397476340694007e-05,
3382
+ "loss": 2.3665,
3383
+ "step": 482
3384
+ },
3385
+ {
3386
+ "epoch": 0.7618296529968455,
3387
+ "grad_norm": 8.408492088317871,
3388
+ "learning_rate": 2.2381703470031547e-05,
3389
+ "loss": 0.3237,
3390
+ "step": 483
3391
+ },
3392
+ {
3393
+ "epoch": 0.7634069400630915,
3394
+ "grad_norm": 56.395423889160156,
3395
+ "learning_rate": 2.2365930599369086e-05,
3396
+ "loss": 1.8043,
3397
+ "step": 484
3398
+ },
3399
+ {
3400
+ "epoch": 0.7649842271293376,
3401
+ "grad_norm": 56.55027770996094,
3402
+ "learning_rate": 2.2350157728706626e-05,
3403
+ "loss": 1.2955,
3404
+ "step": 485
3405
+ },
3406
+ {
3407
+ "epoch": 0.7665615141955836,
3408
+ "grad_norm": 16.443565368652344,
3409
+ "learning_rate": 2.2334384858044166e-05,
3410
+ "loss": 0.416,
3411
+ "step": 486
3412
+ },
3413
+ {
3414
+ "epoch": 0.7681388012618297,
3415
+ "grad_norm": 19.663917541503906,
3416
+ "learning_rate": 2.2318611987381706e-05,
3417
+ "loss": 0.5654,
3418
+ "step": 487
3419
+ },
3420
+ {
3421
+ "epoch": 0.7697160883280757,
3422
+ "grad_norm": 21.52880859375,
3423
+ "learning_rate": 2.2302839116719245e-05,
3424
+ "loss": 0.6703,
3425
+ "step": 488
3426
+ },
3427
+ {
3428
+ "epoch": 0.7712933753943217,
3429
+ "grad_norm": 41.78302764892578,
3430
+ "learning_rate": 2.2287066246056785e-05,
3431
+ "loss": 1.8075,
3432
+ "step": 489
3433
+ },
3434
+ {
3435
+ "epoch": 0.7728706624605678,
3436
+ "grad_norm": 13.108506202697754,
3437
+ "learning_rate": 2.2271293375394325e-05,
3438
+ "loss": 0.4115,
3439
+ "step": 490
3440
+ },
3441
+ {
3442
+ "epoch": 0.7744479495268138,
3443
+ "grad_norm": 37.963130950927734,
3444
+ "learning_rate": 2.2255520504731865e-05,
3445
+ "loss": 0.8782,
3446
+ "step": 491
3447
+ },
3448
+ {
3449
+ "epoch": 0.7760252365930599,
3450
+ "grad_norm": 26.62042999267578,
3451
+ "learning_rate": 2.22397476340694e-05,
3452
+ "loss": 0.659,
3453
+ "step": 492
3454
+ },
3455
+ {
3456
+ "epoch": 0.777602523659306,
3457
+ "grad_norm": 46.10493087768555,
3458
+ "learning_rate": 2.2223974763406938e-05,
3459
+ "loss": 1.2059,
3460
+ "step": 493
3461
+ },
3462
+ {
3463
+ "epoch": 0.7791798107255521,
3464
+ "grad_norm": 36.810768127441406,
3465
+ "learning_rate": 2.2208201892744477e-05,
3466
+ "loss": 1.2543,
3467
+ "step": 494
3468
+ },
3469
+ {
3470
+ "epoch": 0.7807570977917981,
3471
+ "grad_norm": 57.31834030151367,
3472
+ "learning_rate": 2.2192429022082017e-05,
3473
+ "loss": 1.7419,
3474
+ "step": 495
3475
+ },
3476
+ {
3477
+ "epoch": 0.7823343848580442,
3478
+ "grad_norm": 9.108223915100098,
3479
+ "learning_rate": 2.2176656151419557e-05,
3480
+ "loss": 0.2551,
3481
+ "step": 496
3482
+ },
3483
+ {
3484
+ "epoch": 0.7839116719242902,
3485
+ "grad_norm": 11.215947151184082,
3486
+ "learning_rate": 2.2160883280757097e-05,
3487
+ "loss": 0.4398,
3488
+ "step": 497
3489
+ },
3490
+ {
3491
+ "epoch": 0.7854889589905363,
3492
+ "grad_norm": 18.126249313354492,
3493
+ "learning_rate": 2.2145110410094637e-05,
3494
+ "loss": 0.7329,
3495
+ "step": 498
3496
+ },
3497
+ {
3498
+ "epoch": 0.7870662460567823,
3499
+ "grad_norm": 30.252431869506836,
3500
+ "learning_rate": 2.2129337539432176e-05,
3501
+ "loss": 1.1767,
3502
+ "step": 499
3503
+ },
3504
+ {
3505
+ "epoch": 0.7886435331230284,
3506
+ "grad_norm": 22.32280921936035,
3507
+ "learning_rate": 2.2113564668769716e-05,
3508
+ "loss": 0.6799,
3509
+ "step": 500
3510
+ }
3511
+ ],
3512
+ "logging_steps": 1,
3513
+ "max_steps": 1902,
3514
+ "num_input_tokens_seen": 0,
3515
+ "num_train_epochs": 3,
3516
+ "save_steps": 500,
3517
+ "stateful_callbacks": {
3518
+ "TrainerControl": {
3519
+ "args": {
3520
+ "should_epoch_stop": false,
3521
+ "should_evaluate": false,
3522
+ "should_log": false,
3523
+ "should_save": true,
3524
+ "should_training_stop": false
3525
+ },
3526
+ "attributes": {}
3527
+ }
3528
+ },
3529
+ "total_flos": 783890270208000.0,
3530
+ "train_batch_size": 8,
3531
+ "trial_name": null,
3532
+ "trial_params": null
3533
+ }
checkpoint-500/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4c46477fc32a2baa50a9a31f5fc414e1b1586156a0273be927ce2a5de7f607f3
3
+ size 5304
checkpoint-500/vocab.txt ADDED
The diff for this file is too large to render. See raw diff
 
config.json ADDED
@@ -0,0 +1,29 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "derhan/IndoBert-QA",
3
+ "architectures": [
4
+ "BertForQuestionAnswering"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.1,
7
+ "bos_token_id": 0,
8
+ "classifier_dropout": null,
9
+ "eos_token_ids": 0,
10
+ "gradient_checkpointing": false,
11
+ "hidden_act": "gelu",
12
+ "hidden_dropout_prob": 0.1,
13
+ "hidden_size": 768,
14
+ "initializer_range": 0.02,
15
+ "intermediate_size": 3072,
16
+ "layer_norm_eps": 1e-12,
17
+ "max_position_embeddings": 512,
18
+ "model_type": "bert",
19
+ "num_attention_heads": 12,
20
+ "num_hidden_layers": 12,
21
+ "output_past": true,
22
+ "pad_token_id": 0,
23
+ "position_embedding_type": "absolute",
24
+ "torch_dtype": "float32",
25
+ "transformers_version": "4.47.1",
26
+ "type_vocab_size": 2,
27
+ "use_cache": true,
28
+ "vocab_size": 31923
29
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d5b8499ed6aefc2d805cc18cd56df708dde1898d0c86166f20c35b590ffafac0
3
+ size 439899960
special_tokens_map.json ADDED
@@ -0,0 +1,37 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "cls_token": {
3
+ "content": "[CLS]",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "mask_token": {
10
+ "content": "[MASK]",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": {
17
+ "content": "[PAD]",
18
+ "lstrip": false,
19
+ "normalized": false,
20
+ "rstrip": false,
21
+ "single_word": false
22
+ },
23
+ "sep_token": {
24
+ "content": "[SEP]",
25
+ "lstrip": false,
26
+ "normalized": false,
27
+ "rstrip": false,
28
+ "single_word": false
29
+ },
30
+ "unk_token": {
31
+ "content": "[UNK]",
32
+ "lstrip": false,
33
+ "normalized": false,
34
+ "rstrip": false,
35
+ "single_word": false
36
+ }
37
+ }
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1,65 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "[PAD]",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "1": {
12
+ "content": "[UNK]",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "2": {
20
+ "content": "[MASK]",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "3": {
28
+ "content": "[CLS]",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "4": {
36
+ "content": "[SEP]",
37
+ "lstrip": false,
38
+ "normalized": false,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ }
43
+ },
44
+ "clean_up_tokenization_spaces": false,
45
+ "cls_token": "[CLS]",
46
+ "do_basic_tokenize": true,
47
+ "do_lower_case": true,
48
+ "extra_special_tokens": {},
49
+ "mask_token": "[MASK]",
50
+ "max_length": 384,
51
+ "model_max_length": 1000000000000000019884624838656,
52
+ "never_split": null,
53
+ "pad_to_multiple_of": null,
54
+ "pad_token": "[PAD]",
55
+ "pad_token_type_id": 0,
56
+ "padding_side": "right",
57
+ "sep_token": "[SEP]",
58
+ "stride": 128,
59
+ "strip_accents": null,
60
+ "tokenize_chinese_chars": true,
61
+ "tokenizer_class": "BertTokenizer",
62
+ "truncation_side": "right",
63
+ "truncation_strategy": "only_second",
64
+ "unk_token": "[UNK]"
65
+ }
vocab.txt ADDED
The diff for this file is too large to render. See raw diff