Hugging Face's logo

bldeaw
/
pii-ner

Safetensors
bert
Model card Files
xet
 Community
bldeaw commited on
Commit
6f7ca21
·
verified ·
1 Parent(s): 9cf94c9

Upload folder using huggingface_hub

Browse files
Files changed (11) hide show
  1. config.json +107 -37
  2. model.safetensors +2 -2
  3. optimizer.pt +3 -0
  4. rng_state.pth +3 -0
  5. scheduler.pt +3 -0
  6. special_tokens_map.json +5 -18
  7. tokenizer.json +0 -0
  8. tokenizer_config.json +19 -48
  9. trainer_state.json +1924 -0
  10. training_args.bin +2 -2
  11. vocab.txt +0 -0
config.json CHANGED
@@ -1,62 +1,132 @@
1
  {
2
  "architectures": [
3
- "CamembertForTokenClassification"
4
  ],
5
  "attention_probs_dropout_prob": 0.1,
6
- "bos_token_id": 5,
7
  "classifier_dropout": null,
 
8
  "dtype": "float32",
9
- "eos_token_id": 6,
10
- "gradient_checkpointing": false,
11
  "hidden_act": "gelu",
12
  "hidden_dropout_prob": 0.1,
13
  "hidden_size": 768,
14
  "id2label": {
15
  "0": "O",
16
- "1": "B-PER",
17
- "2": "I-PER",
18
- "3": "B-LOC",
19
- "4": "I-LOC",
20
- "5": "B-ORG",
21
- "6": "I-ORG",
22
- "7": "B-ID",
23
- "8": "I-ID",
24
- "9": "B-EMAIL",
25
- "10": "I-EMAIL",
26
- "11": "B-PHONE",
27
- "12": "I-PHONE",
28
- "13": "B-BANK",
29
- "14": "I-BANK"
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
30
  },
31
  "initializer_range": 0.02,
32
  "intermediate_size": 3072,
33
  "label2id": {
34
- "B-BANK": 13,
35
- "B-EMAIL": 9,
36
- "B-ID": 7,
37
- "B-LOC": 3,
38
- "B-ORG": 5,
39
- "B-PER": 1,
40
- "B-PHONE": 11,
41
- "I-BANK": 14,
42
- "I-EMAIL": 10,
43
- "I-ID": 8,
44
- "I-LOC": 4,
45
- "I-ORG": 6,
46
- "I-PER": 2,
47
- "I-PHONE": 12,
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
48
  "O": 0
49
  },
50
  "layer_norm_eps": 1e-12,
51
  "max_position_embeddings": 512,
52
- "model_type": "camembert",
53
- "num_attention_head": 12,
54
  "num_attention_heads": 12,
55
  "num_hidden_layers": 12,
56
- "pad_token_id": 1,
 
 
 
 
 
57
  "position_embedding_type": "absolute",
58
  "transformers_version": "4.57.1",
59
- "type_vocab_size": 1,
60
  "use_cache": true,
61
- "vocab_size": 25005
62
  }
 
1
  {
2
  "architectures": [
3
+ "BertForTokenClassification"
4
  ],
5
  "attention_probs_dropout_prob": 0.1,
 
6
  "classifier_dropout": null,
7
+ "directionality": "bidi",
8
  "dtype": "float32",
 
 
9
  "hidden_act": "gelu",
10
  "hidden_dropout_prob": 0.1,
11
  "hidden_size": 768,
12
  "id2label": {
13
  "0": "O",
14
+ "1": "B-FULL_NAME",
15
+ "2": "I-FULL_NAME",
16
+ "3": "B-DATE_OF_BIRTH",
17
+ "4": "I-DATE_OF_BIRTH",
18
+ "5": "B-AGE",
19
+ "6": "I-AGE",
20
+ "7": "B-GENDER",
21
+ "8": "I-GENDER",
22
+ "9": "B-NATIONALITY",
23
+ "10": "I-NATIONALITY",
24
+ "11": "B-ETHNICITY",
25
+ "12": "I-ETHNICITY",
26
+ "13": "B-RELIGION",
27
+ "14": "I-RELIGION",
28
+ "15": "B-MARITAL_STATUS",
29
+ "16": "I-MARITAL_STATUS",
30
+ "17": "B-PHONE",
31
+ "18": "I-PHONE",
32
+ "19": "B-HOME_PHONE",
33
+ "20": "I-HOME_PHONE",
34
+ "21": "B-EMAIL",
35
+ "22": "I-EMAIL",
36
+ "23": "B-ADDRESS",
37
+ "24": "I-ADDRESS",
38
+ "25": "B-NATIONAL_ID",
39
+ "26": "I-NATIONAL_ID",
40
+ "27": "B-PASSPORT_NUMBER",
41
+ "28": "I-PASSPORT_NUMBER",
42
+ "29": "B-DRIVER_LICENSE",
43
+ "30": "I-DRIVER_LICENSE",
44
+ "31": "B-SOCIAL_SECURITY_NUMBER",
45
+ "32": "I-SOCIAL_SECURITY_NUMBER",
46
+ "33": "B-TAX_ID",
47
+ "34": "I-TAX_ID",
48
+ "35": "B-BANK_ACCOUNT",
49
+ "36": "I-BANK_ACCOUNT",
50
+ "37": "B-CREDIT_CARD",
51
+ "38": "I-CREDIT_CARD",
52
+ "39": "B-INCOME",
53
+ "40": "I-INCOME",
54
+ "41": "B-OCCUPATION",
55
+ "42": "I-OCCUPATION",
56
+ "43": "B-WORKPLACE",
57
+ "44": "I-WORKPLACE",
58
+ "45": "B-JOB_TITLE",
59
+ "46": "I-JOB_TITLE",
60
+ "47": "B-EDUCATION_LEVEL",
61
+ "48": "I-EDUCATION_LEVEL"
62
  },
63
  "initializer_range": 0.02,
64
  "intermediate_size": 3072,
65
  "label2id": {
66
+ "B-ADDRESS": 23,
67
+ "B-AGE": 5,
68
+ "B-BANK_ACCOUNT": 35,
69
+ "B-CREDIT_CARD": 37,
70
+ "B-DATE_OF_BIRTH": 3,
71
+ "B-DRIVER_LICENSE": 29,
72
+ "B-EDUCATION_LEVEL": 47,
73
+ "B-EMAIL": 21,
74
+ "B-ETHNICITY": 11,
75
+ "B-FULL_NAME": 1,
76
+ "B-GENDER": 7,
77
+ "B-HOME_PHONE": 19,
78
+ "B-INCOME": 39,
79
+ "B-JOB_TITLE": 45,
80
+ "B-MARITAL_STATUS": 15,
81
+ "B-NATIONALITY": 9,
82
+ "B-NATIONAL_ID": 25,
83
+ "B-OCCUPATION": 41,
84
+ "B-PASSPORT_NUMBER": 27,
85
+ "B-PHONE": 17,
86
+ "B-RELIGION": 13,
87
+ "B-SOCIAL_SECURITY_NUMBER": 31,
88
+ "B-TAX_ID": 33,
89
+ "B-WORKPLACE": 43,
90
+ "I-ADDRESS": 24,
91
+ "I-AGE": 6,
92
+ "I-BANK_ACCOUNT": 36,
93
+ "I-CREDIT_CARD": 38,
94
+ "I-DATE_OF_BIRTH": 4,
95
+ "I-DRIVER_LICENSE": 30,
96
+ "I-EDUCATION_LEVEL": 48,
97
+ "I-EMAIL": 22,
98
+ "I-ETHNICITY": 12,
99
+ "I-FULL_NAME": 2,
100
+ "I-GENDER": 8,
101
+ "I-HOME_PHONE": 20,
102
+ "I-INCOME": 40,
103
+ "I-JOB_TITLE": 46,
104
+ "I-MARITAL_STATUS": 16,
105
+ "I-NATIONALITY": 10,
106
+ "I-NATIONAL_ID": 26,
107
+ "I-OCCUPATION": 42,
108
+ "I-PASSPORT_NUMBER": 28,
109
+ "I-PHONE": 18,
110
+ "I-RELIGION": 14,
111
+ "I-SOCIAL_SECURITY_NUMBER": 32,
112
+ "I-TAX_ID": 34,
113
+ "I-WORKPLACE": 44,
114
  "O": 0
115
  },
116
  "layer_norm_eps": 1e-12,
117
  "max_position_embeddings": 512,
118
+ "model_type": "bert",
 
119
  "num_attention_heads": 12,
120
  "num_hidden_layers": 12,
121
+ "pad_token_id": 0,
122
+ "pooler_fc_size": 768,
123
+ "pooler_num_attention_heads": 12,
124
+ "pooler_num_fc_layers": 3,
125
+ "pooler_size_per_head": 128,
126
+ "pooler_type": "first_token_transform",
127
  "position_embedding_type": "absolute",
128
  "transformers_version": "4.57.1",
129
+ "type_vocab_size": 2,
130
  "use_cache": true,
131
+ "vocab_size": 119547
132
  }
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:31cc2a2dfddd0418c62c8e58dfea03ede9c691c5b1655b70902fb750f0758689
3
- size 418685332
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cf7d6b037d8aa5acff6e727bc8a3f02562a249e2c38dee49874a393c46759d8d
3
+ size 709225476
optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:55f56104cfbbc16de04a7264826fdfc612d3b98613c2ea008e9227b5996125be
3
+ size 1418573963
rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:630de98c43b9d6559c43c666d01d6c0c6cf635648be0be86f5e3dd2729180c4f
3
+ size 14645
scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:09f593c71f7f11e8c50230097f6aa42e615c25105785c6d902e51ffaf525e38c
3
+ size 1465
special_tokens_map.json CHANGED
@@ -1,20 +1,7 @@
1
  {
2
- "additional_special_tokens": [
3
- "<s>NOTUSED",
4
- "</s>NOTUSED",
5
- "<_>"
6
- ],
7
- "bos_token": "<s>",
8
- "cls_token": "<s>",
9
- "eos_token": "</s>",
10
- "mask_token": {
11
- "content": "<mask>",
12
- "lstrip": true,
13
- "normalized": false,
14
- "rstrip": false,
15
- "single_word": false
16
- },
17
- "pad_token": "<pad>",
18
- "sep_token": "</s>",
19
- "unk_token": "<unk>"
20
  }
 
1
  {
2
+ "cls_token": "[CLS]",
3
+ "mask_token": "[MASK]",
4
+ "pad_token": "[PAD]",
5
+ "sep_token": "[SEP]",
6
+ "unk_token": "[UNK]"
 
 
 
 
 
 
 
 
 
 
 
 
 
7
  }
tokenizer.json CHANGED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json CHANGED
@@ -1,85 +1,56 @@
1
  {
2
  "added_tokens_decoder": {
3
  "0": {
4
- "content": "<s>NOTUSED",
5
  "lstrip": false,
6
  "normalized": false,
7
  "rstrip": false,
8
  "single_word": false,
9
  "special": true
10
  },
11
- "1": {
12
- "content": "<pad>",
13
  "lstrip": false,
14
  "normalized": false,
15
  "rstrip": false,
16
  "single_word": false,
17
  "special": true
18
  },
19
- "2": {
20
- "content": "</s>NOTUSED",
21
  "lstrip": false,
22
  "normalized": false,
23
  "rstrip": false,
24
  "single_word": false,
25
  "special": true
26
  },
27
- "3": {
28
- "content": "<unk>",
29
  "lstrip": false,
30
  "normalized": false,
31
  "rstrip": false,
32
  "single_word": false,
33
  "special": true
34
  },
35
- "5": {
36
- "content": "<s>",
37
  "lstrip": false,
38
  "normalized": false,
39
  "rstrip": false,
40
  "single_word": false,
41
  "special": true
42
- },
43
- "6": {
44
- "content": "</s>",
45
- "lstrip": false,
46
- "normalized": false,
47
- "rstrip": false,
48
- "single_word": false,
49
- "special": true
50
- },
51
- "8": {
52
- "content": "<_>",
53
- "lstrip": false,
54
- "normalized": false,
55
- "rstrip": false,
56
- "single_word": false,
57
- "special": false
58
- },
59
- "25004": {
60
- "content": "<mask>",
61
- "lstrip": true,
62
- "normalized": false,
63
- "rstrip": false,
64
- "single_word": false,
65
- "special": true
66
  }
67
  },
68
- "additional_special_tokens": [
69
- "<s>NOTUSED",
70
- "</s>NOTUSED",
71
- "<_>"
72
- ],
73
- "bos_token": "<s>",
74
  "clean_up_tokenization_spaces": false,
75
- "cls_token": "<s>",
76
- "eos_token": "</s>",
77
  "extra_special_tokens": {},
78
- "mask_token": "<mask>",
79
- "model_max_length": 1000000000000000019884624838656,
80
- "pad_token": "<pad>",
81
- "sep_token": "</s>",
82
- "sp_model_kwargs": {},
83
- "tokenizer_class": "CamembertTokenizer",
84
- "unk_token": "<unk>"
 
85
  }
 
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
+ "100": {
12
+ "content": "[UNK]",
13
  "lstrip": false,
14
  "normalized": false,
15
  "rstrip": false,
16
  "single_word": false,
17
  "special": true
18
  },
19
+ "101": {
20
+ "content": "[CLS]",
21
  "lstrip": false,
22
  "normalized": false,
23
  "rstrip": false,
24
  "single_word": false,
25
  "special": true
26
  },
27
+ "102": {
28
+ "content": "[SEP]",
29
  "lstrip": false,
30
  "normalized": false,
31
  "rstrip": false,
32
  "single_word": false,
33
  "special": true
34
  },
35
+ "103": {
36
+ "content": "[MASK]",
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_lower_case": false,
47
  "extra_special_tokens": {},
48
+ "mask_token": "[MASK]",
49
+ "model_max_length": 512,
50
+ "pad_token": "[PAD]",
51
+ "sep_token": "[SEP]",
52
+ "strip_accents": null,
53
+ "tokenize_chinese_chars": true,
54
+ "tokenizer_class": "BertTokenizer",
55
+ "unk_token": "[UNK]"
56
  }
trainer_state.json ADDED
@@ -0,0 +1,1924 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 3.0,
6
+ "eval_steps": 500,
7
+ "global_step": 13527,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.011088933244621867,
14
+ "grad_norm": 8.14461898803711,
15
+ "learning_rate": 1.9927552302801804e-05,
16
+ "loss": 0.902,
17
+ "step": 50
18
+ },
19
+ {
20
+ "epoch": 0.022177866489243733,
21
+ "grad_norm": 6.75203800201416,
22
+ "learning_rate": 1.9853626081170995e-05,
23
+ "loss": 0.448,
24
+ "step": 100
25
+ },
26
+ {
27
+ "epoch": 0.0332667997338656,
28
+ "grad_norm": 7.353200912475586,
29
+ "learning_rate": 1.977969985954018e-05,
30
+ "loss": 0.3174,
31
+ "step": 150
32
+ },
33
+ {
34
+ "epoch": 0.04435573297848747,
35
+ "grad_norm": 2.7292733192443848,
36
+ "learning_rate": 1.970577363790937e-05,
37
+ "loss": 0.281,
38
+ "step": 200
39
+ },
40
+ {
41
+ "epoch": 0.05544466622310934,
42
+ "grad_norm": 3.5009043216705322,
43
+ "learning_rate": 1.9631847416278556e-05,
44
+ "loss": 0.2798,
45
+ "step": 250
46
+ },
47
+ {
48
+ "epoch": 0.0665335994677312,
49
+ "grad_norm": 5.440974712371826,
50
+ "learning_rate": 1.9557921194647743e-05,
51
+ "loss": 0.2409,
52
+ "step": 300
53
+ },
54
+ {
55
+ "epoch": 0.07762253271235307,
56
+ "grad_norm": 4.327282905578613,
57
+ "learning_rate": 1.948399497301693e-05,
58
+ "loss": 0.203,
59
+ "step": 350
60
+ },
61
+ {
62
+ "epoch": 0.08871146595697493,
63
+ "grad_norm": 1.8929752111434937,
64
+ "learning_rate": 1.9410068751386117e-05,
65
+ "loss": 0.2169,
66
+ "step": 400
67
+ },
68
+ {
69
+ "epoch": 0.0998003992015968,
70
+ "grad_norm": 2.9351725578308105,
71
+ "learning_rate": 1.9336142529755308e-05,
72
+ "loss": 0.2197,
73
+ "step": 450
74
+ },
75
+ {
76
+ "epoch": 0.11088933244621868,
77
+ "grad_norm": 5.145724296569824,
78
+ "learning_rate": 1.926221630812449e-05,
79
+ "loss": 0.2073,
80
+ "step": 500
81
+ },
82
+ {
83
+ "epoch": 0.12197826569084054,
84
+ "grad_norm": 4.723082542419434,
85
+ "learning_rate": 1.9188290086493682e-05,
86
+ "loss": 0.2188,
87
+ "step": 550
88
+ },
89
+ {
90
+ "epoch": 0.1330671989354624,
91
+ "grad_norm": 2.782973289489746,
92
+ "learning_rate": 1.911436386486287e-05,
93
+ "loss": 0.2062,
94
+ "step": 600
95
+ },
96
+ {
97
+ "epoch": 0.14415613218008427,
98
+ "grad_norm": 3.5191564559936523,
99
+ "learning_rate": 1.9040437643232056e-05,
100
+ "loss": 0.213,
101
+ "step": 650
102
+ },
103
+ {
104
+ "epoch": 0.15524506542470615,
105
+ "grad_norm": 1.734594702720642,
106
+ "learning_rate": 1.8966511421601243e-05,
107
+ "loss": 0.2064,
108
+ "step": 700
109
+ },
110
+ {
111
+ "epoch": 0.16633399866932802,
112
+ "grad_norm": 3.7910728454589844,
113
+ "learning_rate": 1.889258519997043e-05,
114
+ "loss": 0.1943,
115
+ "step": 750
116
+ },
117
+ {
118
+ "epoch": 0.17742293191394987,
119
+ "grad_norm": 3.688796281814575,
120
+ "learning_rate": 1.881865897833962e-05,
121
+ "loss": 0.1928,
122
+ "step": 800
123
+ },
124
+ {
125
+ "epoch": 0.18851186515857174,
126
+ "grad_norm": 2.2231035232543945,
127
+ "learning_rate": 1.8744732756708804e-05,
128
+ "loss": 0.1825,
129
+ "step": 850
130
+ },
131
+ {
132
+ "epoch": 0.1996007984031936,
133
+ "grad_norm": 4.24806022644043,
134
+ "learning_rate": 1.8670806535077994e-05,
135
+ "loss": 0.1578,
136
+ "step": 900
137
+ },
138
+ {
139
+ "epoch": 0.2106897316478155,
140
+ "grad_norm": 3.1724870204925537,
141
+ "learning_rate": 1.859688031344718e-05,
142
+ "loss": 0.1645,
143
+ "step": 950
144
+ },
145
+ {
146
+ "epoch": 0.22177866489243736,
147
+ "grad_norm": 2.4956796169281006,
148
+ "learning_rate": 1.852295409181637e-05,
149
+ "loss": 0.1796,
150
+ "step": 1000
151
+ },
152
+ {
153
+ "epoch": 0.2328675981370592,
154
+ "grad_norm": 1.245104432106018,
155
+ "learning_rate": 1.8449027870185556e-05,
156
+ "loss": 0.1669,
157
+ "step": 1050
158
+ },
159
+ {
160
+ "epoch": 0.24395653138168108,
161
+ "grad_norm": 2.2229013442993164,
162
+ "learning_rate": 1.8375101648554743e-05,
163
+ "loss": 0.1743,
164
+ "step": 1100
165
+ },
166
+ {
167
+ "epoch": 0.25504546462630295,
168
+ "grad_norm": 3.6590652465820312,
169
+ "learning_rate": 1.8301175426923933e-05,
170
+ "loss": 0.1654,
171
+ "step": 1150
172
+ },
173
+ {
174
+ "epoch": 0.2661343978709248,
175
+ "grad_norm": 2.119847297668457,
176
+ "learning_rate": 1.8227249205293117e-05,
177
+ "loss": 0.1603,
178
+ "step": 1200
179
+ },
180
+ {
181
+ "epoch": 0.2772233311155467,
182
+ "grad_norm": 3.36673903465271,
183
+ "learning_rate": 1.8153322983662307e-05,
184
+ "loss": 0.1593,
185
+ "step": 1250
186
+ },
187
+ {
188
+ "epoch": 0.28831226436016855,
189
+ "grad_norm": 2.931980609893799,
190
+ "learning_rate": 1.8079396762031494e-05,
191
+ "loss": 0.1585,
192
+ "step": 1300
193
+ },
194
+ {
195
+ "epoch": 0.2994011976047904,
196
+ "grad_norm": 1.6280852556228638,
197
+ "learning_rate": 1.800547054040068e-05,
198
+ "loss": 0.1676,
199
+ "step": 1350
200
+ },
201
+ {
202
+ "epoch": 0.3104901308494123,
203
+ "grad_norm": 5.353924751281738,
204
+ "learning_rate": 1.793154431876987e-05,
205
+ "loss": 0.155,
206
+ "step": 1400
207
+ },
208
+ {
209
+ "epoch": 0.32157906409403414,
210
+ "grad_norm": 1.9066994190216064,
211
+ "learning_rate": 1.7857618097139056e-05,
212
+ "loss": 0.168,
213
+ "step": 1450
214
+ },
215
+ {
216
+ "epoch": 0.33266799733865604,
217
+ "grad_norm": 3.3019158840179443,
218
+ "learning_rate": 1.7783691875508246e-05,
219
+ "loss": 0.144,
220
+ "step": 1500
221
+ },
222
+ {
223
+ "epoch": 0.3437569305832779,
224
+ "grad_norm": 2.2679409980773926,
225
+ "learning_rate": 1.770976565387743e-05,
226
+ "loss": 0.1616,
227
+ "step": 1550
228
+ },
229
+ {
230
+ "epoch": 0.35484586382789973,
231
+ "grad_norm": 3.734102964401245,
232
+ "learning_rate": 1.763583943224662e-05,
233
+ "loss": 0.1331,
234
+ "step": 1600
235
+ },
236
+ {
237
+ "epoch": 0.36593479707252163,
238
+ "grad_norm": 2.348113775253296,
239
+ "learning_rate": 1.7561913210615807e-05,
240
+ "loss": 0.1477,
241
+ "step": 1650
242
+ },
243
+ {
244
+ "epoch": 0.3770237303171435,
245
+ "grad_norm": 3.4904134273529053,
246
+ "learning_rate": 1.7487986988984994e-05,
247
+ "loss": 0.1448,
248
+ "step": 1700
249
+ },
250
+ {
251
+ "epoch": 0.3881126635617654,
252
+ "grad_norm": 1.9997397661209106,
253
+ "learning_rate": 1.741406076735418e-05,
254
+ "loss": 0.1461,
255
+ "step": 1750
256
+ },
257
+ {
258
+ "epoch": 0.3992015968063872,
259
+ "grad_norm": 2.7343485355377197,
260
+ "learning_rate": 1.7340134545723372e-05,
261
+ "loss": 0.149,
262
+ "step": 1800
263
+ },
264
+ {
265
+ "epoch": 0.4102905300510091,
266
+ "grad_norm": 3.363621950149536,
267
+ "learning_rate": 1.726620832409256e-05,
268
+ "loss": 0.1601,
269
+ "step": 1850
270
+ },
271
+ {
272
+ "epoch": 0.421379463295631,
273
+ "grad_norm": 1.4055149555206299,
274
+ "learning_rate": 1.7192282102461742e-05,
275
+ "loss": 0.1475,
276
+ "step": 1900
277
+ },
278
+ {
279
+ "epoch": 0.4324683965402528,
280
+ "grad_norm": 4.3813066482543945,
281
+ "learning_rate": 1.7118355880830933e-05,
282
+ "loss": 0.1434,
283
+ "step": 1950
284
+ },
285
+ {
286
+ "epoch": 0.4435573297848747,
287
+ "grad_norm": 2.2316620349884033,
288
+ "learning_rate": 1.704442965920012e-05,
289
+ "loss": 0.1452,
290
+ "step": 2000
291
+ },
292
+ {
293
+ "epoch": 0.45464626302949657,
294
+ "grad_norm": 1.3528106212615967,
295
+ "learning_rate": 1.6970503437569307e-05,
296
+ "loss": 0.138,
297
+ "step": 2050
298
+ },
299
+ {
300
+ "epoch": 0.4657351962741184,
301
+ "grad_norm": 2.043443441390991,
302
+ "learning_rate": 1.6896577215938494e-05,
303
+ "loss": 0.1586,
304
+ "step": 2100
305
+ },
306
+ {
307
+ "epoch": 0.4768241295187403,
308
+ "grad_norm": 1.920172929763794,
309
+ "learning_rate": 1.6822650994307685e-05,
310
+ "loss": 0.1267,
311
+ "step": 2150
312
+ },
313
+ {
314
+ "epoch": 0.48791306276336216,
315
+ "grad_norm": 2.8047444820404053,
316
+ "learning_rate": 1.6748724772676868e-05,
317
+ "loss": 0.1417,
318
+ "step": 2200
319
+ },
320
+ {
321
+ "epoch": 0.499001996007984,
322
+ "grad_norm": 3.8060503005981445,
323
+ "learning_rate": 1.6674798551046055e-05,
324
+ "loss": 0.1345,
325
+ "step": 2250
326
+ },
327
+ {
328
+ "epoch": 0.5100909292526059,
329
+ "grad_norm": 2.5561251640319824,
330
+ "learning_rate": 1.6600872329415246e-05,
331
+ "loss": 0.1266,
332
+ "step": 2300
333
+ },
334
+ {
335
+ "epoch": 0.5211798624972278,
336
+ "grad_norm": 4.142666339874268,
337
+ "learning_rate": 1.6526946107784433e-05,
338
+ "loss": 0.1407,
339
+ "step": 2350
340
+ },
341
+ {
342
+ "epoch": 0.5322687957418496,
343
+ "grad_norm": 1.7247148752212524,
344
+ "learning_rate": 1.645301988615362e-05,
345
+ "loss": 0.1339,
346
+ "step": 2400
347
+ },
348
+ {
349
+ "epoch": 0.5433577289864715,
350
+ "grad_norm": 4.735507965087891,
351
+ "learning_rate": 1.6379093664522807e-05,
352
+ "loss": 0.1355,
353
+ "step": 2450
354
+ },
355
+ {
356
+ "epoch": 0.5544466622310934,
357
+ "grad_norm": 4.915506362915039,
358
+ "learning_rate": 1.6305167442891997e-05,
359
+ "loss": 0.1373,
360
+ "step": 2500
361
+ },
362
+ {
363
+ "epoch": 0.5655355954757152,
364
+ "grad_norm": 2.5551493167877197,
365
+ "learning_rate": 1.623124122126118e-05,
366
+ "loss": 0.1344,
367
+ "step": 2550
368
+ },
369
+ {
370
+ "epoch": 0.5766245287203371,
371
+ "grad_norm": 1.054781436920166,
372
+ "learning_rate": 1.615731499963037e-05,
373
+ "loss": 0.126,
374
+ "step": 2600
375
+ },
376
+ {
377
+ "epoch": 0.587713461964959,
378
+ "grad_norm": 2.835658311843872,
379
+ "learning_rate": 1.608338877799956e-05,
380
+ "loss": 0.1312,
381
+ "step": 2650
382
+ },
383
+ {
384
+ "epoch": 0.5988023952095808,
385
+ "grad_norm": 3.4936070442199707,
386
+ "learning_rate": 1.6009462556368746e-05,
387
+ "loss": 0.1078,
388
+ "step": 2700
389
+ },
390
+ {
391
+ "epoch": 0.6098913284542027,
392
+ "grad_norm": 2.4390571117401123,
393
+ "learning_rate": 1.5935536334737933e-05,
394
+ "loss": 0.1405,
395
+ "step": 2750
396
+ },
397
+ {
398
+ "epoch": 0.6209802616988246,
399
+ "grad_norm": 2.4469242095947266,
400
+ "learning_rate": 1.586161011310712e-05,
401
+ "loss": 0.1246,
402
+ "step": 2800
403
+ },
404
+ {
405
+ "epoch": 0.6320691949434465,
406
+ "grad_norm": 3.6933956146240234,
407
+ "learning_rate": 1.578768389147631e-05,
408
+ "loss": 0.1297,
409
+ "step": 2850
410
+ },
411
+ {
412
+ "epoch": 0.6431581281880683,
413
+ "grad_norm": 2.2289130687713623,
414
+ "learning_rate": 1.5713757669845494e-05,
415
+ "loss": 0.1265,
416
+ "step": 2900
417
+ },
418
+ {
419
+ "epoch": 0.6542470614326902,
420
+ "grad_norm": 1.5156584978103638,
421
+ "learning_rate": 1.5639831448214684e-05,
422
+ "loss": 0.1311,
423
+ "step": 2950
424
+ },
425
+ {
426
+ "epoch": 0.6653359946773121,
427
+ "grad_norm": 1.9378397464752197,
428
+ "learning_rate": 1.556590522658387e-05,
429
+ "loss": 0.1275,
430
+ "step": 3000
431
+ },
432
+ {
433
+ "epoch": 0.6764249279219339,
434
+ "grad_norm": 1.9141340255737305,
435
+ "learning_rate": 1.549197900495306e-05,
436
+ "loss": 0.1141,
437
+ "step": 3050
438
+ },
439
+ {
440
+ "epoch": 0.6875138611665558,
441
+ "grad_norm": 1.9459818601608276,
442
+ "learning_rate": 1.5418052783322245e-05,
443
+ "loss": 0.1318,
444
+ "step": 3100
445
+ },
446
+ {
447
+ "epoch": 0.6986027944111777,
448
+ "grad_norm": 7.391958236694336,
449
+ "learning_rate": 1.5344126561691432e-05,
450
+ "loss": 0.1203,
451
+ "step": 3150
452
+ },
453
+ {
454
+ "epoch": 0.7096917276557995,
455
+ "grad_norm": 2.0345418453216553,
456
+ "learning_rate": 1.5270200340060623e-05,
457
+ "loss": 0.1273,
458
+ "step": 3200
459
+ },
460
+ {
461
+ "epoch": 0.7207806609004214,
462
+ "grad_norm": 4.687112808227539,
463
+ "learning_rate": 1.5196274118429808e-05,
464
+ "loss": 0.1018,
465
+ "step": 3250
466
+ },
467
+ {
468
+ "epoch": 0.7318695941450433,
469
+ "grad_norm": 2.794402837753296,
470
+ "learning_rate": 1.5122347896798997e-05,
471
+ "loss": 0.1164,
472
+ "step": 3300
473
+ },
474
+ {
475
+ "epoch": 0.7429585273896652,
476
+ "grad_norm": 2.7743284702301025,
477
+ "learning_rate": 1.5048421675168184e-05,
478
+ "loss": 0.1164,
479
+ "step": 3350
480
+ },
481
+ {
482
+ "epoch": 0.754047460634287,
483
+ "grad_norm": 3.1311397552490234,
484
+ "learning_rate": 1.497449545353737e-05,
485
+ "loss": 0.1245,
486
+ "step": 3400
487
+ },
488
+ {
489
+ "epoch": 0.7651363938789089,
490
+ "grad_norm": 3.217726230621338,
491
+ "learning_rate": 1.4900569231906558e-05,
492
+ "loss": 0.1045,
493
+ "step": 3450
494
+ },
495
+ {
496
+ "epoch": 0.7762253271235308,
497
+ "grad_norm": 1.691542387008667,
498
+ "learning_rate": 1.4826643010275745e-05,
499
+ "loss": 0.134,
500
+ "step": 3500
501
+ },
502
+ {
503
+ "epoch": 0.7873142603681526,
504
+ "grad_norm": 3.6782033443450928,
505
+ "learning_rate": 1.4752716788644934e-05,
506
+ "loss": 0.1276,
507
+ "step": 3550
508
+ },
509
+ {
510
+ "epoch": 0.7984031936127745,
511
+ "grad_norm": 1.4673993587493896,
512
+ "learning_rate": 1.4678790567014121e-05,
513
+ "loss": 0.1059,
514
+ "step": 3600
515
+ },
516
+ {
517
+ "epoch": 0.8094921268573964,
518
+ "grad_norm": 2.9146125316619873,
519
+ "learning_rate": 1.460486434538331e-05,
520
+ "loss": 0.1165,
521
+ "step": 3650
522
+ },
523
+ {
524
+ "epoch": 0.8205810601020181,
525
+ "grad_norm": 2.6374094486236572,
526
+ "learning_rate": 1.4530938123752497e-05,
527
+ "loss": 0.1245,
528
+ "step": 3700
529
+ },
530
+ {
531
+ "epoch": 0.83166999334664,
532
+ "grad_norm": 0.7516615986824036,
533
+ "learning_rate": 1.4457011902121686e-05,
534
+ "loss": 0.1178,
535
+ "step": 3750
536
+ },
537
+ {
538
+ "epoch": 0.842758926591262,
539
+ "grad_norm": 2.7078309059143066,
540
+ "learning_rate": 1.4383085680490871e-05,
541
+ "loss": 0.1216,
542
+ "step": 3800
543
+ },
544
+ {
545
+ "epoch": 0.8538478598358837,
546
+ "grad_norm": 1.5147875547409058,
547
+ "learning_rate": 1.4309159458860058e-05,
548
+ "loss": 0.1178,
549
+ "step": 3850
550
+ },
551
+ {
552
+ "epoch": 0.8649367930805056,
553
+ "grad_norm": 1.8173058032989502,
554
+ "learning_rate": 1.4235233237229247e-05,
555
+ "loss": 0.1177,
556
+ "step": 3900
557
+ },
558
+ {
559
+ "epoch": 0.8760257263251275,
560
+ "grad_norm": 2.6974475383758545,
561
+ "learning_rate": 1.4161307015598434e-05,
562
+ "loss": 0.1093,
563
+ "step": 3950
564
+ },
565
+ {
566
+ "epoch": 0.8871146595697494,
567
+ "grad_norm": 2.176143169403076,
568
+ "learning_rate": 1.4087380793967623e-05,
569
+ "loss": 0.1058,
570
+ "step": 4000
571
+ },
572
+ {
573
+ "epoch": 0.8982035928143712,
574
+ "grad_norm": 2.174964427947998,
575
+ "learning_rate": 1.4013454572336808e-05,
576
+ "loss": 0.1164,
577
+ "step": 4050
578
+ },
579
+ {
580
+ "epoch": 0.9092925260589931,
581
+ "grad_norm": 2.4584383964538574,
582
+ "learning_rate": 1.3939528350705997e-05,
583
+ "loss": 0.1188,
584
+ "step": 4100
585
+ },
586
+ {
587
+ "epoch": 0.920381459303615,
588
+ "grad_norm": 1.6693115234375,
589
+ "learning_rate": 1.3865602129075184e-05,
590
+ "loss": 0.112,
591
+ "step": 4150
592
+ },
593
+ {
594
+ "epoch": 0.9314703925482368,
595
+ "grad_norm": 2.2045669555664062,
596
+ "learning_rate": 1.379167590744437e-05,
597
+ "loss": 0.1234,
598
+ "step": 4200
599
+ },
600
+ {
601
+ "epoch": 0.9425593257928587,
602
+ "grad_norm": 2.9538867473602295,
603
+ "learning_rate": 1.371774968581356e-05,
604
+ "loss": 0.1019,
605
+ "step": 4250
606
+ },
607
+ {
608
+ "epoch": 0.9536482590374806,
609
+ "grad_norm": 1.7292181253433228,
610
+ "learning_rate": 1.3643823464182747e-05,
611
+ "loss": 0.118,
612
+ "step": 4300
613
+ },
614
+ {
615
+ "epoch": 0.9647371922821024,
616
+ "grad_norm": 1.2080588340759277,
617
+ "learning_rate": 1.3569897242551935e-05,
618
+ "loss": 0.1096,
619
+ "step": 4350
620
+ },
621
+ {
622
+ "epoch": 0.9758261255267243,
623
+ "grad_norm": 1.0146644115447998,
624
+ "learning_rate": 1.349597102092112e-05,
625
+ "loss": 0.1211,
626
+ "step": 4400
627
+ },
628
+ {
629
+ "epoch": 0.9869150587713462,
630
+ "grad_norm": 3.40978741645813,
631
+ "learning_rate": 1.342204479929031e-05,
632
+ "loss": 0.114,
633
+ "step": 4450
634
+ },
635
+ {
636
+ "epoch": 0.998003992015968,
637
+ "grad_norm": 2.2283411026000977,
638
+ "learning_rate": 1.3348118577659497e-05,
639
+ "loss": 0.1024,
640
+ "step": 4500
641
+ },
642
+ {
643
+ "epoch": 1.00909292526059,
644
+ "grad_norm": 3.0386204719543457,
645
+ "learning_rate": 1.3274192356028684e-05,
646
+ "loss": 0.1037,
647
+ "step": 4550
648
+ },
649
+ {
650
+ "epoch": 1.0201818585052118,
651
+ "grad_norm": 2.2395098209381104,
652
+ "learning_rate": 1.3200266134397872e-05,
653
+ "loss": 0.0962,
654
+ "step": 4600
655
+ },
656
+ {
657
+ "epoch": 1.0312707917498336,
658
+ "grad_norm": 0.779739499092102,
659
+ "learning_rate": 1.312633991276706e-05,
660
+ "loss": 0.0897,
661
+ "step": 4650
662
+ },
663
+ {
664
+ "epoch": 1.0423597249944556,
665
+ "grad_norm": 2.3607208728790283,
666
+ "learning_rate": 1.3052413691136248e-05,
667
+ "loss": 0.0913,
668
+ "step": 4700
669
+ },
670
+ {
671
+ "epoch": 1.0534486582390774,
672
+ "grad_norm": 1.5109652280807495,
673
+ "learning_rate": 1.2978487469505434e-05,
674
+ "loss": 0.0987,
675
+ "step": 4750
676
+ },
677
+ {
678
+ "epoch": 1.0645375914836992,
679
+ "grad_norm": 4.1896467208862305,
680
+ "learning_rate": 1.2904561247874622e-05,
681
+ "loss": 0.1044,
682
+ "step": 4800
683
+ },
684
+ {
685
+ "epoch": 1.0756265247283212,
686
+ "grad_norm": 3.4481632709503174,
687
+ "learning_rate": 1.283063502624381e-05,
688
+ "loss": 0.1065,
689
+ "step": 4850
690
+ },
691
+ {
692
+ "epoch": 1.086715457972943,
693
+ "grad_norm": 1.4650681018829346,
694
+ "learning_rate": 1.2756708804612998e-05,
695
+ "loss": 0.0854,
696
+ "step": 4900
697
+ },
698
+ {
699
+ "epoch": 1.0978043912175648,
700
+ "grad_norm": 2.925077438354492,
701
+ "learning_rate": 1.2682782582982185e-05,
702
+ "loss": 0.1055,
703
+ "step": 4950
704
+ },
705
+ {
706
+ "epoch": 1.1088933244621868,
707
+ "grad_norm": 1.6256483793258667,
708
+ "learning_rate": 1.2608856361351372e-05,
709
+ "loss": 0.1016,
710
+ "step": 5000
711
+ },
712
+ {
713
+ "epoch": 1.1199822577068086,
714
+ "grad_norm": 2.7225170135498047,
715
+ "learning_rate": 1.2534930139720561e-05,
716
+ "loss": 0.0999,
717
+ "step": 5050
718
+ },
719
+ {
720
+ "epoch": 1.1310711909514304,
721
+ "grad_norm": 2.22430419921875,
722
+ "learning_rate": 1.2461003918089746e-05,
723
+ "loss": 0.0886,
724
+ "step": 5100
725
+ },
726
+ {
727
+ "epoch": 1.1421601241960524,
728
+ "grad_norm": 2.4130799770355225,
729
+ "learning_rate": 1.2387077696458935e-05,
730
+ "loss": 0.1012,
731
+ "step": 5150
732
+ },
733
+ {
734
+ "epoch": 1.1532490574406742,
735
+ "grad_norm": 1.5306169986724854,
736
+ "learning_rate": 1.2313151474828122e-05,
737
+ "loss": 0.0988,
738
+ "step": 5200
739
+ },
740
+ {
741
+ "epoch": 1.1643379906852962,
742
+ "grad_norm": 1.8464442491531372,
743
+ "learning_rate": 1.2239225253197311e-05,
744
+ "loss": 0.0928,
745
+ "step": 5250
746
+ },
747
+ {
748
+ "epoch": 1.175426923929918,
749
+ "grad_norm": 2.2706775665283203,
750
+ "learning_rate": 1.2165299031566498e-05,
751
+ "loss": 0.094,
752
+ "step": 5300
753
+ },
754
+ {
755
+ "epoch": 1.1865158571745398,
756
+ "grad_norm": 1.517385482788086,
757
+ "learning_rate": 1.2091372809935685e-05,
758
+ "loss": 0.1032,
759
+ "step": 5350
760
+ },
761
+ {
762
+ "epoch": 1.1976047904191618,
763
+ "grad_norm": 1.9841318130493164,
764
+ "learning_rate": 1.2017446588304874e-05,
765
+ "loss": 0.0974,
766
+ "step": 5400
767
+ },
768
+ {
769
+ "epoch": 1.2086937236637836,
770
+ "grad_norm": 3.8151936531066895,
771
+ "learning_rate": 1.1943520366674059e-05,
772
+ "loss": 0.0921,
773
+ "step": 5450
774
+ },
775
+ {
776
+ "epoch": 1.2197826569084054,
777
+ "grad_norm": 1.1752043962478638,
778
+ "learning_rate": 1.1869594145043248e-05,
779
+ "loss": 0.0918,
780
+ "step": 5500
781
+ },
782
+ {
783
+ "epoch": 1.2308715901530274,
784
+ "grad_norm": 3.7480061054229736,
785
+ "learning_rate": 1.1795667923412435e-05,
786
+ "loss": 0.0935,
787
+ "step": 5550
788
+ },
789
+ {
790
+ "epoch": 1.2419605233976492,
791
+ "grad_norm": 2.0084166526794434,
792
+ "learning_rate": 1.1721741701781624e-05,
793
+ "loss": 0.1018,
794
+ "step": 5600
795
+ },
796
+ {
797
+ "epoch": 1.253049456642271,
798
+ "grad_norm": 3.461491823196411,
799
+ "learning_rate": 1.164781548015081e-05,
800
+ "loss": 0.1044,
801
+ "step": 5650
802
+ },
803
+ {
804
+ "epoch": 1.264138389886893,
805
+ "grad_norm": 1.8434693813323975,
806
+ "learning_rate": 1.1573889258519996e-05,
807
+ "loss": 0.0939,
808
+ "step": 5700
809
+ },
810
+ {
811
+ "epoch": 1.2752273231315148,
812
+ "grad_norm": 2.570587158203125,
813
+ "learning_rate": 1.1499963036889185e-05,
814
+ "loss": 0.0984,
815
+ "step": 5750
816
+ },
817
+ {
818
+ "epoch": 1.2863162563761366,
819
+ "grad_norm": 1.5073240995407104,
820
+ "learning_rate": 1.1426036815258372e-05,
821
+ "loss": 0.0889,
822
+ "step": 5800
823
+ },
824
+ {
825
+ "epoch": 1.2974051896207586,
826
+ "grad_norm": 2.8674674034118652,
827
+ "learning_rate": 1.135211059362756e-05,
828
+ "loss": 0.1007,
829
+ "step": 5850
830
+ },
831
+ {
832
+ "epoch": 1.3084941228653804,
833
+ "grad_norm": 5.509634971618652,
834
+ "learning_rate": 1.1278184371996748e-05,
835
+ "loss": 0.0916,
836
+ "step": 5900
837
+ },
838
+ {
839
+ "epoch": 1.3195830561100022,
840
+ "grad_norm": 1.8779860734939575,
841
+ "learning_rate": 1.1204258150365936e-05,
842
+ "loss": 0.1009,
843
+ "step": 5950
844
+ },
845
+ {
846
+ "epoch": 1.3306719893546242,
847
+ "grad_norm": 1.6348674297332764,
848
+ "learning_rate": 1.1130331928735124e-05,
849
+ "loss": 0.1009,
850
+ "step": 6000
851
+ },
852
+ {
853
+ "epoch": 1.341760922599246,
854
+ "grad_norm": 2.8110697269439697,
855
+ "learning_rate": 1.1056405707104312e-05,
856
+ "loss": 0.11,
857
+ "step": 6050
858
+ },
859
+ {
860
+ "epoch": 1.3528498558438677,
861
+ "grad_norm": 1.2461209297180176,
862
+ "learning_rate": 1.0982479485473498e-05,
863
+ "loss": 0.101,
864
+ "step": 6100
865
+ },
866
+ {
867
+ "epoch": 1.3639387890884898,
868
+ "grad_norm": 3.5219790935516357,
869
+ "learning_rate": 1.0908553263842685e-05,
870
+ "loss": 0.0935,
871
+ "step": 6150
872
+ },
873
+ {
874
+ "epoch": 1.3750277223331115,
875
+ "grad_norm": 1.9491550922393799,
876
+ "learning_rate": 1.0834627042211873e-05,
877
+ "loss": 0.0987,
878
+ "step": 6200
879
+ },
880
+ {
881
+ "epoch": 1.3861166555777333,
882
+ "grad_norm": 1.0792142152786255,
883
+ "learning_rate": 1.076070082058106e-05,
884
+ "loss": 0.0887,
885
+ "step": 6250
886
+ },
887
+ {
888
+ "epoch": 1.3972055888223553,
889
+ "grad_norm": 2.6937804222106934,
890
+ "learning_rate": 1.068677459895025e-05,
891
+ "loss": 0.0844,
892
+ "step": 6300
893
+ },
894
+ {
895
+ "epoch": 1.4082945220669771,
896
+ "grad_norm": 1.8635441064834595,
897
+ "learning_rate": 1.0612848377319436e-05,
898
+ "loss": 0.0961,
899
+ "step": 6350
900
+ },
901
+ {
902
+ "epoch": 1.419383455311599,
903
+ "grad_norm": 2.886840581893921,
904
+ "learning_rate": 1.0538922155688625e-05,
905
+ "loss": 0.1027,
906
+ "step": 6400
907
+ },
908
+ {
909
+ "epoch": 1.430472388556221,
910
+ "grad_norm": 1.2699713706970215,
911
+ "learning_rate": 1.046499593405781e-05,
912
+ "loss": 0.084,
913
+ "step": 6450
914
+ },
915
+ {
916
+ "epoch": 1.4415613218008427,
917
+ "grad_norm": 3.5690715312957764,
918
+ "learning_rate": 1.0391069712426998e-05,
919
+ "loss": 0.1007,
920
+ "step": 6500
921
+ },
922
+ {
923
+ "epoch": 1.4526502550454645,
924
+ "grad_norm": 2.6432716846466064,
925
+ "learning_rate": 1.0317143490796186e-05,
926
+ "loss": 0.086,
927
+ "step": 6550
928
+ },
929
+ {
930
+ "epoch": 1.4637391882900865,
931
+ "grad_norm": 1.3745814561843872,
932
+ "learning_rate": 1.0243217269165373e-05,
933
+ "loss": 0.1001,
934
+ "step": 6600
935
+ },
936
+ {
937
+ "epoch": 1.4748281215347083,
938
+ "grad_norm": 2.1724355220794678,
939
+ "learning_rate": 1.0169291047534562e-05,
940
+ "loss": 0.0761,
941
+ "step": 6650
942
+ },
943
+ {
944
+ "epoch": 1.4859170547793301,
945
+ "grad_norm": 2.703815460205078,
946
+ "learning_rate": 1.0095364825903749e-05,
947
+ "loss": 0.0921,
948
+ "step": 6700
949
+ },
950
+ {
951
+ "epoch": 1.4970059880239521,
952
+ "grad_norm": 5.101301670074463,
953
+ "learning_rate": 1.0021438604272938e-05,
954
+ "loss": 0.1016,
955
+ "step": 6750
956
+ },
957
+ {
958
+ "epoch": 1.508094921268574,
959
+ "grad_norm": 2.663522243499756,
960
+ "learning_rate": 9.947512382642123e-06,
961
+ "loss": 0.0872,
962
+ "step": 6800
963
+ },
964
+ {
965
+ "epoch": 1.5191838545131957,
966
+ "grad_norm": 1.159928321838379,
967
+ "learning_rate": 9.873586161011312e-06,
968
+ "loss": 0.0897,
969
+ "step": 6850
970
+ },
971
+ {
972
+ "epoch": 1.5302727877578177,
973
+ "grad_norm": 1.6328009366989136,
974
+ "learning_rate": 9.799659939380499e-06,
975
+ "loss": 0.0823,
976
+ "step": 6900
977
+ },
978
+ {
979
+ "epoch": 1.5413617210024395,
980
+ "grad_norm": 2.2539258003234863,
981
+ "learning_rate": 9.725733717749686e-06,
982
+ "loss": 0.0974,
983
+ "step": 6950
984
+ },
985
+ {
986
+ "epoch": 1.5524506542470613,
987
+ "grad_norm": 1.0964053869247437,
988
+ "learning_rate": 9.651807496118875e-06,
989
+ "loss": 0.0906,
990
+ "step": 7000
991
+ },
992
+ {
993
+ "epoch": 1.5635395874916833,
994
+ "grad_norm": 1.5801488161087036,
995
+ "learning_rate": 9.577881274488062e-06,
996
+ "loss": 0.1,
997
+ "step": 7050
998
+ },
999
+ {
1000
+ "epoch": 1.5746285207363053,
1001
+ "grad_norm": 1.7104038000106812,
1002
+ "learning_rate": 9.503955052857249e-06,
1003
+ "loss": 0.0935,
1004
+ "step": 7100
1005
+ },
1006
+ {
1007
+ "epoch": 1.585717453980927,
1008
+ "grad_norm": 2.5046310424804688,
1009
+ "learning_rate": 9.430028831226436e-06,
1010
+ "loss": 0.0908,
1011
+ "step": 7150
1012
+ },
1013
+ {
1014
+ "epoch": 1.596806387225549,
1015
+ "grad_norm": 1.518405795097351,
1016
+ "learning_rate": 9.356102609595625e-06,
1017
+ "loss": 0.092,
1018
+ "step": 7200
1019
+ },
1020
+ {
1021
+ "epoch": 1.607895320470171,
1022
+ "grad_norm": 1.8131279945373535,
1023
+ "learning_rate": 9.282176387964812e-06,
1024
+ "loss": 0.0837,
1025
+ "step": 7250
1026
+ },
1027
+ {
1028
+ "epoch": 1.6189842537147925,
1029
+ "grad_norm": 1.3197726011276245,
1030
+ "learning_rate": 9.208250166333999e-06,
1031
+ "loss": 0.0984,
1032
+ "step": 7300
1033
+ },
1034
+ {
1035
+ "epoch": 1.6300731869594145,
1036
+ "grad_norm": 2.3708226680755615,
1037
+ "learning_rate": 9.134323944703188e-06,
1038
+ "loss": 0.0861,
1039
+ "step": 7350
1040
+ },
1041
+ {
1042
+ "epoch": 1.6411621202040365,
1043
+ "grad_norm": 3.385481119155884,
1044
+ "learning_rate": 9.060397723072375e-06,
1045
+ "loss": 0.0889,
1046
+ "step": 7400
1047
+ },
1048
+ {
1049
+ "epoch": 1.652251053448658,
1050
+ "grad_norm": 1.1704720258712769,
1051
+ "learning_rate": 8.986471501441562e-06,
1052
+ "loss": 0.1005,
1053
+ "step": 7450
1054
+ },
1055
+ {
1056
+ "epoch": 1.66333998669328,
1057
+ "grad_norm": 2.1572084426879883,
1058
+ "learning_rate": 8.912545279810749e-06,
1059
+ "loss": 0.0884,
1060
+ "step": 7500
1061
+ },
1062
+ {
1063
+ "epoch": 1.674428919937902,
1064
+ "grad_norm": 3.3791210651397705,
1065
+ "learning_rate": 8.838619058179938e-06,
1066
+ "loss": 0.09,
1067
+ "step": 7550
1068
+ },
1069
+ {
1070
+ "epoch": 1.685517853182524,
1071
+ "grad_norm": 2.125641345977783,
1072
+ "learning_rate": 8.764692836549125e-06,
1073
+ "loss": 0.0856,
1074
+ "step": 7600
1075
+ },
1076
+ {
1077
+ "epoch": 1.6966067864271457,
1078
+ "grad_norm": 2.459035873413086,
1079
+ "learning_rate": 8.690766614918312e-06,
1080
+ "loss": 0.0845,
1081
+ "step": 7650
1082
+ },
1083
+ {
1084
+ "epoch": 1.7076957196717677,
1085
+ "grad_norm": 0.977972686290741,
1086
+ "learning_rate": 8.6168403932875e-06,
1087
+ "loss": 0.0973,
1088
+ "step": 7700
1089
+ },
1090
+ {
1091
+ "epoch": 1.7187846529163895,
1092
+ "grad_norm": 2.432441473007202,
1093
+ "learning_rate": 8.542914171656688e-06,
1094
+ "loss": 0.0984,
1095
+ "step": 7750
1096
+ },
1097
+ {
1098
+ "epoch": 1.7298735861610113,
1099
+ "grad_norm": 1.9185181856155396,
1100
+ "learning_rate": 8.468987950025875e-06,
1101
+ "loss": 0.0878,
1102
+ "step": 7800
1103
+ },
1104
+ {
1105
+ "epoch": 1.7409625194056333,
1106
+ "grad_norm": 1.8849973678588867,
1107
+ "learning_rate": 8.395061728395062e-06,
1108
+ "loss": 0.0822,
1109
+ "step": 7850
1110
+ },
1111
+ {
1112
+ "epoch": 1.752051452650255,
1113
+ "grad_norm": 3.4242374897003174,
1114
+ "learning_rate": 8.32113550676425e-06,
1115
+ "loss": 0.088,
1116
+ "step": 7900
1117
+ },
1118
+ {
1119
+ "epoch": 1.7631403858948769,
1120
+ "grad_norm": 3.122772216796875,
1121
+ "learning_rate": 8.247209285133437e-06,
1122
+ "loss": 0.0858,
1123
+ "step": 7950
1124
+ },
1125
+ {
1126
+ "epoch": 1.7742293191394989,
1127
+ "grad_norm": 2.070228338241577,
1128
+ "learning_rate": 8.173283063502625e-06,
1129
+ "loss": 0.0863,
1130
+ "step": 8000
1131
+ },
1132
+ {
1133
+ "epoch": 1.7853182523841207,
1134
+ "grad_norm": 1.955519199371338,
1135
+ "learning_rate": 8.099356841871813e-06,
1136
+ "loss": 0.0919,
1137
+ "step": 8050
1138
+ },
1139
+ {
1140
+ "epoch": 1.7964071856287425,
1141
+ "grad_norm": 3.2033774852752686,
1142
+ "learning_rate": 8.025430620241e-06,
1143
+ "loss": 0.0906,
1144
+ "step": 8100
1145
+ },
1146
+ {
1147
+ "epoch": 1.8074961188733645,
1148
+ "grad_norm": 2.1565141677856445,
1149
+ "learning_rate": 7.951504398610187e-06,
1150
+ "loss": 0.0989,
1151
+ "step": 8150
1152
+ },
1153
+ {
1154
+ "epoch": 1.8185850521179863,
1155
+ "grad_norm": 2.1477930545806885,
1156
+ "learning_rate": 7.877578176979376e-06,
1157
+ "loss": 0.0812,
1158
+ "step": 8200
1159
+ },
1160
+ {
1161
+ "epoch": 1.829673985362608,
1162
+ "grad_norm": 1.8666229248046875,
1163
+ "learning_rate": 7.803651955348563e-06,
1164
+ "loss": 0.088,
1165
+ "step": 8250
1166
+ },
1167
+ {
1168
+ "epoch": 1.84076291860723,
1169
+ "grad_norm": 1.3660918474197388,
1170
+ "learning_rate": 7.72972573371775e-06,
1171
+ "loss": 0.0936,
1172
+ "step": 8300
1173
+ },
1174
+ {
1175
+ "epoch": 1.8518518518518519,
1176
+ "grad_norm": 2.6135261058807373,
1177
+ "learning_rate": 7.655799512086937e-06,
1178
+ "loss": 0.0839,
1179
+ "step": 8350
1180
+ },
1181
+ {
1182
+ "epoch": 1.8629407850964737,
1183
+ "grad_norm": 3.5965628623962402,
1184
+ "learning_rate": 7.581873290456125e-06,
1185
+ "loss": 0.1014,
1186
+ "step": 8400
1187
+ },
1188
+ {
1189
+ "epoch": 1.8740297183410957,
1190
+ "grad_norm": 4.500585079193115,
1191
+ "learning_rate": 7.507947068825313e-06,
1192
+ "loss": 0.0819,
1193
+ "step": 8450
1194
+ },
1195
+ {
1196
+ "epoch": 1.8851186515857175,
1197
+ "grad_norm": 2.053957223892212,
1198
+ "learning_rate": 7.434020847194501e-06,
1199
+ "loss": 0.087,
1200
+ "step": 8500
1201
+ },
1202
+ {
1203
+ "epoch": 1.8962075848303392,
1204
+ "grad_norm": 1.6143734455108643,
1205
+ "learning_rate": 7.360094625563688e-06,
1206
+ "loss": 0.0908,
1207
+ "step": 8550
1208
+ },
1209
+ {
1210
+ "epoch": 1.9072965180749613,
1211
+ "grad_norm": 2.1243669986724854,
1212
+ "learning_rate": 7.286168403932875e-06,
1213
+ "loss": 0.095,
1214
+ "step": 8600
1215
+ },
1216
+ {
1217
+ "epoch": 1.918385451319583,
1218
+ "grad_norm": 1.8068082332611084,
1219
+ "learning_rate": 7.212242182302063e-06,
1220
+ "loss": 0.0853,
1221
+ "step": 8650
1222
+ },
1223
+ {
1224
+ "epoch": 1.9294743845642048,
1225
+ "grad_norm": 1.7410554885864258,
1226
+ "learning_rate": 7.13831596067125e-06,
1227
+ "loss": 0.0887,
1228
+ "step": 8700
1229
+ },
1230
+ {
1231
+ "epoch": 1.9405633178088268,
1232
+ "grad_norm": 3.2952396869659424,
1233
+ "learning_rate": 7.064389739040438e-06,
1234
+ "loss": 0.093,
1235
+ "step": 8750
1236
+ },
1237
+ {
1238
+ "epoch": 1.9516522510534486,
1239
+ "grad_norm": 1.382659673690796,
1240
+ "learning_rate": 6.990463517409626e-06,
1241
+ "loss": 0.0988,
1242
+ "step": 8800
1243
+ },
1244
+ {
1245
+ "epoch": 1.9627411842980704,
1246
+ "grad_norm": 1.8536710739135742,
1247
+ "learning_rate": 6.916537295778814e-06,
1248
+ "loss": 0.0855,
1249
+ "step": 8850
1250
+ },
1251
+ {
1252
+ "epoch": 1.9738301175426924,
1253
+ "grad_norm": 1.3422974348068237,
1254
+ "learning_rate": 6.842611074148001e-06,
1255
+ "loss": 0.0913,
1256
+ "step": 8900
1257
+ },
1258
+ {
1259
+ "epoch": 1.9849190507873142,
1260
+ "grad_norm": 0.9699559211730957,
1261
+ "learning_rate": 6.768684852517189e-06,
1262
+ "loss": 0.0827,
1263
+ "step": 8950
1264
+ },
1265
+ {
1266
+ "epoch": 1.996007984031936,
1267
+ "grad_norm": 2.4076015949249268,
1268
+ "learning_rate": 6.694758630886376e-06,
1269
+ "loss": 0.0738,
1270
+ "step": 9000
1271
+ },
1272
+ {
1273
+ "epoch": 2.007096917276558,
1274
+ "grad_norm": 2.115738868713379,
1275
+ "learning_rate": 6.620832409255563e-06,
1276
+ "loss": 0.0799,
1277
+ "step": 9050
1278
+ },
1279
+ {
1280
+ "epoch": 2.01818585052118,
1281
+ "grad_norm": 0.871468186378479,
1282
+ "learning_rate": 6.546906187624751e-06,
1283
+ "loss": 0.0636,
1284
+ "step": 9100
1285
+ },
1286
+ {
1287
+ "epoch": 2.0292747837658016,
1288
+ "grad_norm": 3.783653497695923,
1289
+ "learning_rate": 6.472979965993939e-06,
1290
+ "loss": 0.0689,
1291
+ "step": 9150
1292
+ },
1293
+ {
1294
+ "epoch": 2.0403637170104236,
1295
+ "grad_norm": 2.4445955753326416,
1296
+ "learning_rate": 6.399053744363127e-06,
1297
+ "loss": 0.0792,
1298
+ "step": 9200
1299
+ },
1300
+ {
1301
+ "epoch": 2.0514526502550456,
1302
+ "grad_norm": 1.8684531450271606,
1303
+ "learning_rate": 6.325127522732314e-06,
1304
+ "loss": 0.0661,
1305
+ "step": 9250
1306
+ },
1307
+ {
1308
+ "epoch": 2.062541583499667,
1309
+ "grad_norm": 1.0839345455169678,
1310
+ "learning_rate": 6.2512013011015016e-06,
1311
+ "loss": 0.0775,
1312
+ "step": 9300
1313
+ },
1314
+ {
1315
+ "epoch": 2.073630516744289,
1316
+ "grad_norm": 0.9116957187652588,
1317
+ "learning_rate": 6.1772750794706895e-06,
1318
+ "loss": 0.0772,
1319
+ "step": 9350
1320
+ },
1321
+ {
1322
+ "epoch": 2.0847194499889112,
1323
+ "grad_norm": 1.6374276876449585,
1324
+ "learning_rate": 6.103348857839876e-06,
1325
+ "loss": 0.0649,
1326
+ "step": 9400
1327
+ },
1328
+ {
1329
+ "epoch": 2.095808383233533,
1330
+ "grad_norm": 2.419456720352173,
1331
+ "learning_rate": 6.029422636209064e-06,
1332
+ "loss": 0.0688,
1333
+ "step": 9450
1334
+ },
1335
+ {
1336
+ "epoch": 2.106897316478155,
1337
+ "grad_norm": 3.2619240283966064,
1338
+ "learning_rate": 5.9554964145782515e-06,
1339
+ "loss": 0.0773,
1340
+ "step": 9500
1341
+ },
1342
+ {
1343
+ "epoch": 2.117986249722777,
1344
+ "grad_norm": 3.0861144065856934,
1345
+ "learning_rate": 5.8815701929474386e-06,
1346
+ "loss": 0.074,
1347
+ "step": 9550
1348
+ },
1349
+ {
1350
+ "epoch": 2.1290751829673984,
1351
+ "grad_norm": 1.3020782470703125,
1352
+ "learning_rate": 5.8076439713166265e-06,
1353
+ "loss": 0.0702,
1354
+ "step": 9600
1355
+ },
1356
+ {
1357
+ "epoch": 2.1401641162120204,
1358
+ "grad_norm": 2.4096367359161377,
1359
+ "learning_rate": 5.733717749685814e-06,
1360
+ "loss": 0.0713,
1361
+ "step": 9650
1362
+ },
1363
+ {
1364
+ "epoch": 2.1512530494566424,
1365
+ "grad_norm": 1.2149136066436768,
1366
+ "learning_rate": 5.659791528055002e-06,
1367
+ "loss": 0.0664,
1368
+ "step": 9700
1369
+ },
1370
+ {
1371
+ "epoch": 2.162341982701264,
1372
+ "grad_norm": 3.8028881549835205,
1373
+ "learning_rate": 5.585865306424189e-06,
1374
+ "loss": 0.0736,
1375
+ "step": 9750
1376
+ },
1377
+ {
1378
+ "epoch": 2.173430915945886,
1379
+ "grad_norm": 2.2303028106689453,
1380
+ "learning_rate": 5.511939084793376e-06,
1381
+ "loss": 0.0748,
1382
+ "step": 9800
1383
+ },
1384
+ {
1385
+ "epoch": 2.184519849190508,
1386
+ "grad_norm": 2.6183419227600098,
1387
+ "learning_rate": 5.438012863162564e-06,
1388
+ "loss": 0.0768,
1389
+ "step": 9850
1390
+ },
1391
+ {
1392
+ "epoch": 2.1956087824351296,
1393
+ "grad_norm": 1.402911901473999,
1394
+ "learning_rate": 5.364086641531751e-06,
1395
+ "loss": 0.0663,
1396
+ "step": 9900
1397
+ },
1398
+ {
1399
+ "epoch": 2.2066977156797516,
1400
+ "grad_norm": 2.14359450340271,
1401
+ "learning_rate": 5.290160419900939e-06,
1402
+ "loss": 0.0654,
1403
+ "step": 9950
1404
+ },
1405
+ {
1406
+ "epoch": 2.2177866489243736,
1407
+ "grad_norm": 1.4012452363967896,
1408
+ "learning_rate": 5.216234198270127e-06,
1409
+ "loss": 0.0803,
1410
+ "step": 10000
1411
+ },
1412
+ {
1413
+ "epoch": 2.228875582168995,
1414
+ "grad_norm": 0.9938985705375671,
1415
+ "learning_rate": 5.142307976639315e-06,
1416
+ "loss": 0.0743,
1417
+ "step": 10050
1418
+ },
1419
+ {
1420
+ "epoch": 2.239964515413617,
1421
+ "grad_norm": 1.9487541913986206,
1422
+ "learning_rate": 5.068381755008502e-06,
1423
+ "loss": 0.0748,
1424
+ "step": 10100
1425
+ },
1426
+ {
1427
+ "epoch": 2.251053448658239,
1428
+ "grad_norm": 6.113137245178223,
1429
+ "learning_rate": 4.994455533377689e-06,
1430
+ "loss": 0.0816,
1431
+ "step": 10150
1432
+ },
1433
+ {
1434
+ "epoch": 2.2621423819028608,
1435
+ "grad_norm": 1.868437647819519,
1436
+ "learning_rate": 4.920529311746877e-06,
1437
+ "loss": 0.0752,
1438
+ "step": 10200
1439
+ },
1440
+ {
1441
+ "epoch": 2.273231315147483,
1442
+ "grad_norm": 1.6830744743347168,
1443
+ "learning_rate": 4.846603090116064e-06,
1444
+ "loss": 0.0706,
1445
+ "step": 10250
1446
+ },
1447
+ {
1448
+ "epoch": 2.284320248392105,
1449
+ "grad_norm": 1.8964732885360718,
1450
+ "learning_rate": 4.772676868485252e-06,
1451
+ "loss": 0.0687,
1452
+ "step": 10300
1453
+ },
1454
+ {
1455
+ "epoch": 2.2954091816367264,
1456
+ "grad_norm": 2.1506104469299316,
1457
+ "learning_rate": 4.69875064685444e-06,
1458
+ "loss": 0.0838,
1459
+ "step": 10350
1460
+ },
1461
+ {
1462
+ "epoch": 2.3064981148813484,
1463
+ "grad_norm": 2.777594804763794,
1464
+ "learning_rate": 4.624824425223627e-06,
1465
+ "loss": 0.0684,
1466
+ "step": 10400
1467
+ },
1468
+ {
1469
+ "epoch": 2.3175870481259704,
1470
+ "grad_norm": 1.3911370038986206,
1471
+ "learning_rate": 4.550898203592815e-06,
1472
+ "loss": 0.0679,
1473
+ "step": 10450
1474
+ },
1475
+ {
1476
+ "epoch": 2.3286759813705924,
1477
+ "grad_norm": 1.7069896459579468,
1478
+ "learning_rate": 4.476971981962002e-06,
1479
+ "loss": 0.0683,
1480
+ "step": 10500
1481
+ },
1482
+ {
1483
+ "epoch": 2.339764914615214,
1484
+ "grad_norm": 3.0551908016204834,
1485
+ "learning_rate": 4.40304576033119e-06,
1486
+ "loss": 0.0719,
1487
+ "step": 10550
1488
+ },
1489
+ {
1490
+ "epoch": 2.350853847859836,
1491
+ "grad_norm": 1.223029375076294,
1492
+ "learning_rate": 4.329119538700378e-06,
1493
+ "loss": 0.0692,
1494
+ "step": 10600
1495
+ },
1496
+ {
1497
+ "epoch": 2.3619427811044575,
1498
+ "grad_norm": 3.9574992656707764,
1499
+ "learning_rate": 4.255193317069565e-06,
1500
+ "loss": 0.0716,
1501
+ "step": 10650
1502
+ },
1503
+ {
1504
+ "epoch": 2.3730317143490796,
1505
+ "grad_norm": 1.701469898223877,
1506
+ "learning_rate": 4.181267095438753e-06,
1507
+ "loss": 0.0857,
1508
+ "step": 10700
1509
+ },
1510
+ {
1511
+ "epoch": 2.3841206475937016,
1512
+ "grad_norm": 1.5294601917266846,
1513
+ "learning_rate": 4.10734087380794e-06,
1514
+ "loss": 0.0714,
1515
+ "step": 10750
1516
+ },
1517
+ {
1518
+ "epoch": 2.3952095808383236,
1519
+ "grad_norm": 2.6890389919281006,
1520
+ "learning_rate": 4.033414652177128e-06,
1521
+ "loss": 0.0649,
1522
+ "step": 10800
1523
+ },
1524
+ {
1525
+ "epoch": 2.406298514082945,
1526
+ "grad_norm": 1.7154722213745117,
1527
+ "learning_rate": 3.959488430546315e-06,
1528
+ "loss": 0.0736,
1529
+ "step": 10850
1530
+ },
1531
+ {
1532
+ "epoch": 2.417387447327567,
1533
+ "grad_norm": 1.9531513452529907,
1534
+ "learning_rate": 3.885562208915503e-06,
1535
+ "loss": 0.0764,
1536
+ "step": 10900
1537
+ },
1538
+ {
1539
+ "epoch": 2.4284763805721887,
1540
+ "grad_norm": 1.5990239381790161,
1541
+ "learning_rate": 3.81163598728469e-06,
1542
+ "loss": 0.0617,
1543
+ "step": 10950
1544
+ },
1545
+ {
1546
+ "epoch": 2.4395653138168107,
1547
+ "grad_norm": 1.2892380952835083,
1548
+ "learning_rate": 3.737709765653878e-06,
1549
+ "loss": 0.0644,
1550
+ "step": 11000
1551
+ },
1552
+ {
1553
+ "epoch": 2.4506542470614328,
1554
+ "grad_norm": 2.5764858722686768,
1555
+ "learning_rate": 3.663783544023065e-06,
1556
+ "loss": 0.0606,
1557
+ "step": 11050
1558
+ },
1559
+ {
1560
+ "epoch": 2.4617431803060548,
1561
+ "grad_norm": 2.1783535480499268,
1562
+ "learning_rate": 3.589857322392253e-06,
1563
+ "loss": 0.0816,
1564
+ "step": 11100
1565
+ },
1566
+ {
1567
+ "epoch": 2.4728321135506763,
1568
+ "grad_norm": 2.3354318141937256,
1569
+ "learning_rate": 3.5159311007614405e-06,
1570
+ "loss": 0.0817,
1571
+ "step": 11150
1572
+ },
1573
+ {
1574
+ "epoch": 2.4839210467952983,
1575
+ "grad_norm": 3.374755382537842,
1576
+ "learning_rate": 3.4420048791306276e-06,
1577
+ "loss": 0.0737,
1578
+ "step": 11200
1579
+ },
1580
+ {
1581
+ "epoch": 2.49500998003992,
1582
+ "grad_norm": 3.841001272201538,
1583
+ "learning_rate": 3.3680786574998155e-06,
1584
+ "loss": 0.068,
1585
+ "step": 11250
1586
+ },
1587
+ {
1588
+ "epoch": 2.506098913284542,
1589
+ "grad_norm": 1.0251437425613403,
1590
+ "learning_rate": 3.294152435869003e-06,
1591
+ "loss": 0.0776,
1592
+ "step": 11300
1593
+ },
1594
+ {
1595
+ "epoch": 2.517187846529164,
1596
+ "grad_norm": 2.3336868286132812,
1597
+ "learning_rate": 3.220226214238191e-06,
1598
+ "loss": 0.074,
1599
+ "step": 11350
1600
+ },
1601
+ {
1602
+ "epoch": 2.528276779773786,
1603
+ "grad_norm": 1.9687585830688477,
1604
+ "learning_rate": 3.146299992607378e-06,
1605
+ "loss": 0.0634,
1606
+ "step": 11400
1607
+ },
1608
+ {
1609
+ "epoch": 2.5393657130184075,
1610
+ "grad_norm": 3.6472342014312744,
1611
+ "learning_rate": 3.0723737709765654e-06,
1612
+ "loss": 0.0713,
1613
+ "step": 11450
1614
+ },
1615
+ {
1616
+ "epoch": 2.5504546462630295,
1617
+ "grad_norm": 1.97185218334198,
1618
+ "learning_rate": 2.9984475493457533e-06,
1619
+ "loss": 0.0692,
1620
+ "step": 11500
1621
+ },
1622
+ {
1623
+ "epoch": 2.561543579507651,
1624
+ "grad_norm": 8.52320671081543,
1625
+ "learning_rate": 2.9245213277149408e-06,
1626
+ "loss": 0.0679,
1627
+ "step": 11550
1628
+ },
1629
+ {
1630
+ "epoch": 2.572632512752273,
1631
+ "grad_norm": 4.320559501647949,
1632
+ "learning_rate": 2.850595106084128e-06,
1633
+ "loss": 0.0775,
1634
+ "step": 11600
1635
+ },
1636
+ {
1637
+ "epoch": 2.583721445996895,
1638
+ "grad_norm": 0.5261452794075012,
1639
+ "learning_rate": 2.7766688844533158e-06,
1640
+ "loss": 0.0577,
1641
+ "step": 11650
1642
+ },
1643
+ {
1644
+ "epoch": 2.594810379241517,
1645
+ "grad_norm": 1.5301717519760132,
1646
+ "learning_rate": 2.7027426628225032e-06,
1647
+ "loss": 0.0694,
1648
+ "step": 11700
1649
+ },
1650
+ {
1651
+ "epoch": 2.6058993124861387,
1652
+ "grad_norm": 1.3441177606582642,
1653
+ "learning_rate": 2.628816441191691e-06,
1654
+ "loss": 0.0733,
1655
+ "step": 11750
1656
+ },
1657
+ {
1658
+ "epoch": 2.6169882457307607,
1659
+ "grad_norm": 1.2103335857391357,
1660
+ "learning_rate": 2.554890219560878e-06,
1661
+ "loss": 0.0764,
1662
+ "step": 11800
1663
+ },
1664
+ {
1665
+ "epoch": 2.6280771789753827,
1666
+ "grad_norm": 2.5863022804260254,
1667
+ "learning_rate": 2.480963997930066e-06,
1668
+ "loss": 0.0764,
1669
+ "step": 11850
1670
+ },
1671
+ {
1672
+ "epoch": 2.6391661122200043,
1673
+ "grad_norm": 2.4411420822143555,
1674
+ "learning_rate": 2.4070377762992536e-06,
1675
+ "loss": 0.0763,
1676
+ "step": 11900
1677
+ },
1678
+ {
1679
+ "epoch": 2.6502550454646263,
1680
+ "grad_norm": 2.061426877975464,
1681
+ "learning_rate": 2.333111554668441e-06,
1682
+ "loss": 0.075,
1683
+ "step": 11950
1684
+ },
1685
+ {
1686
+ "epoch": 2.6613439787092483,
1687
+ "grad_norm": 0.7995760440826416,
1688
+ "learning_rate": 2.2591853330376285e-06,
1689
+ "loss": 0.0622,
1690
+ "step": 12000
1691
+ },
1692
+ {
1693
+ "epoch": 2.67243291195387,
1694
+ "grad_norm": 2.9890496730804443,
1695
+ "learning_rate": 2.185259111406816e-06,
1696
+ "loss": 0.0766,
1697
+ "step": 12050
1698
+ },
1699
+ {
1700
+ "epoch": 2.683521845198492,
1701
+ "grad_norm": 2.9018654823303223,
1702
+ "learning_rate": 2.111332889776004e-06,
1703
+ "loss": 0.0677,
1704
+ "step": 12100
1705
+ },
1706
+ {
1707
+ "epoch": 2.694610778443114,
1708
+ "grad_norm": 0.40557801723480225,
1709
+ "learning_rate": 2.0374066681451914e-06,
1710
+ "loss": 0.0647,
1711
+ "step": 12150
1712
+ },
1713
+ {
1714
+ "epoch": 2.7056997116877355,
1715
+ "grad_norm": 4.465837001800537,
1716
+ "learning_rate": 1.963480446514379e-06,
1717
+ "loss": 0.0649,
1718
+ "step": 12200
1719
+ },
1720
+ {
1721
+ "epoch": 2.7167886449323575,
1722
+ "grad_norm": 2.300358533859253,
1723
+ "learning_rate": 1.8895542248835664e-06,
1724
+ "loss": 0.0797,
1725
+ "step": 12250
1726
+ },
1727
+ {
1728
+ "epoch": 2.7278775781769795,
1729
+ "grad_norm": 1.4945405721664429,
1730
+ "learning_rate": 1.815628003252754e-06,
1731
+ "loss": 0.0651,
1732
+ "step": 12300
1733
+ },
1734
+ {
1735
+ "epoch": 2.738966511421601,
1736
+ "grad_norm": 1.3923050165176392,
1737
+ "learning_rate": 1.7417017816219413e-06,
1738
+ "loss": 0.0857,
1739
+ "step": 12350
1740
+ },
1741
+ {
1742
+ "epoch": 2.750055444666223,
1743
+ "grad_norm": 1.2946971654891968,
1744
+ "learning_rate": 1.667775559991129e-06,
1745
+ "loss": 0.0634,
1746
+ "step": 12400
1747
+ },
1748
+ {
1749
+ "epoch": 2.761144377910845,
1750
+ "grad_norm": 2.017392158508301,
1751
+ "learning_rate": 1.5938493383603165e-06,
1752
+ "loss": 0.0585,
1753
+ "step": 12450
1754
+ },
1755
+ {
1756
+ "epoch": 2.7722333111554667,
1757
+ "grad_norm": 2.730674982070923,
1758
+ "learning_rate": 1.5199231167295042e-06,
1759
+ "loss": 0.0647,
1760
+ "step": 12500
1761
+ },
1762
+ {
1763
+ "epoch": 2.7833222444000887,
1764
+ "grad_norm": 0.6496580243110657,
1765
+ "learning_rate": 1.4459968950986917e-06,
1766
+ "loss": 0.0611,
1767
+ "step": 12550
1768
+ },
1769
+ {
1770
+ "epoch": 2.7944111776447107,
1771
+ "grad_norm": 3.458625078201294,
1772
+ "learning_rate": 1.372070673467879e-06,
1773
+ "loss": 0.0784,
1774
+ "step": 12600
1775
+ },
1776
+ {
1777
+ "epoch": 2.8055001108893327,
1778
+ "grad_norm": 8.372861862182617,
1779
+ "learning_rate": 1.2981444518370667e-06,
1780
+ "loss": 0.0668,
1781
+ "step": 12650
1782
+ },
1783
+ {
1784
+ "epoch": 2.8165890441339543,
1785
+ "grad_norm": 3.9860599040985107,
1786
+ "learning_rate": 1.2242182302062543e-06,
1787
+ "loss": 0.067,
1788
+ "step": 12700
1789
+ },
1790
+ {
1791
+ "epoch": 2.8276779773785763,
1792
+ "grad_norm": 1.4425193071365356,
1793
+ "learning_rate": 1.1502920085754418e-06,
1794
+ "loss": 0.0706,
1795
+ "step": 12750
1796
+ },
1797
+ {
1798
+ "epoch": 2.838766910623198,
1799
+ "grad_norm": 2.859701156616211,
1800
+ "learning_rate": 1.0763657869446293e-06,
1801
+ "loss": 0.0679,
1802
+ "step": 12800
1803
+ },
1804
+ {
1805
+ "epoch": 2.84985584386782,
1806
+ "grad_norm": 2.2282726764678955,
1807
+ "learning_rate": 1.002439565313817e-06,
1808
+ "loss": 0.0724,
1809
+ "step": 12850
1810
+ },
1811
+ {
1812
+ "epoch": 2.860944777112442,
1813
+ "grad_norm": 2.407137870788574,
1814
+ "learning_rate": 9.285133436830045e-07,
1815
+ "loss": 0.0627,
1816
+ "step": 12900
1817
+ },
1818
+ {
1819
+ "epoch": 2.872033710357064,
1820
+ "grad_norm": 3.5280649662017822,
1821
+ "learning_rate": 8.54587122052192e-07,
1822
+ "loss": 0.0674,
1823
+ "step": 12950
1824
+ },
1825
+ {
1826
+ "epoch": 2.8831226436016855,
1827
+ "grad_norm": 1.6543505191802979,
1828
+ "learning_rate": 7.806609004213796e-07,
1829
+ "loss": 0.0765,
1830
+ "step": 13000
1831
+ },
1832
+ {
1833
+ "epoch": 2.8942115768463075,
1834
+ "grad_norm": 0.9745455384254456,
1835
+ "learning_rate": 7.067346787905671e-07,
1836
+ "loss": 0.0602,
1837
+ "step": 13050
1838
+ },
1839
+ {
1840
+ "epoch": 2.905300510090929,
1841
+ "grad_norm": 2.4250497817993164,
1842
+ "learning_rate": 6.328084571597546e-07,
1843
+ "loss": 0.0735,
1844
+ "step": 13100
1845
+ },
1846
+ {
1847
+ "epoch": 2.916389443335551,
1848
+ "grad_norm": 2.5135929584503174,
1849
+ "learning_rate": 5.588822355289421e-07,
1850
+ "loss": 0.0654,
1851
+ "step": 13150
1852
+ },
1853
+ {
1854
+ "epoch": 2.927478376580173,
1855
+ "grad_norm": 1.6418380737304688,
1856
+ "learning_rate": 4.849560138981297e-07,
1857
+ "loss": 0.0667,
1858
+ "step": 13200
1859
+ },
1860
+ {
1861
+ "epoch": 2.938567309824795,
1862
+ "grad_norm": 4.118490695953369,
1863
+ "learning_rate": 4.1102979226731723e-07,
1864
+ "loss": 0.0725,
1865
+ "step": 13250
1866
+ },
1867
+ {
1868
+ "epoch": 2.9496562430694167,
1869
+ "grad_norm": 0.931894063949585,
1870
+ "learning_rate": 3.371035706365048e-07,
1871
+ "loss": 0.0548,
1872
+ "step": 13300
1873
+ },
1874
+ {
1875
+ "epoch": 2.9607451763140387,
1876
+ "grad_norm": 0.6899275183677673,
1877
+ "learning_rate": 2.6317734900569235e-07,
1878
+ "loss": 0.0826,
1879
+ "step": 13350
1880
+ },
1881
+ {
1882
+ "epoch": 2.9718341095586602,
1883
+ "grad_norm": 1.263277292251587,
1884
+ "learning_rate": 1.892511273748799e-07,
1885
+ "loss": 0.0683,
1886
+ "step": 13400
1887
+ },
1888
+ {
1889
+ "epoch": 2.9829230428032822,
1890
+ "grad_norm": 2.3304920196533203,
1891
+ "learning_rate": 1.1532490574406743e-07,
1892
+ "loss": 0.0651,
1893
+ "step": 13450
1894
+ },
1895
+ {
1896
+ "epoch": 2.9940119760479043,
1897
+ "grad_norm": 1.8537914752960205,
1898
+ "learning_rate": 4.139868411325497e-08,
1899
+ "loss": 0.0648,
1900
+ "step": 13500
1901
+ }
1902
+ ],
1903
+ "logging_steps": 50,
1904
+ "max_steps": 13527,
1905
+ "num_input_tokens_seen": 0,
1906
+ "num_train_epochs": 3,
1907
+ "save_steps": 500,
1908
+ "stateful_callbacks": {
1909
+ "TrainerControl": {
1910
+ "args": {
1911
+ "should_epoch_stop": false,
1912
+ "should_evaluate": false,
1913
+ "should_log": false,
1914
+ "should_save": true,
1915
+ "should_training_stop": true
1916
+ },
1917
+ "attributes": {}
1918
+ }
1919
+ },
1920
+ "total_flos": 1.414150781306112e+16,
1921
+ "train_batch_size": 16,
1922
+ "trial_name": null,
1923
+ "trial_params": null
1924
+ }
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:41b854bd197cde1be486a48879107c11b7986db9a09d96c8e3a41249dfd6a0a1
3
- size 5841
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:182df7cd75d32df85eab8dcf2961196c90e2fa3f6087a2784f43b4a9ccd1160d
3
+ size 5777
vocab.txt ADDED
The diff for this file is too large to render. See raw diff