Hugging Face's logo

CaffeineThief
/
cysecbert-ttp-bert-25per_ood_data

Safetensors
bert
Model card Files
xet
 Community
CaffeineThief commited on
Commit
bdf964e
·
verified ·
1 Parent(s): 539b6c1

Upload saved model files

Browse files
This view is limited to 50 files because it contains too many changes.   See raw diff
Files changed (50) hide show
  1. checkpoint-1080/config.json +130 -0
  2. checkpoint-1080/model.safetensors +3 -0
  3. checkpoint-1080/optimizer.pt +3 -0
  4. checkpoint-1080/rng_state.pth +3 -0
  5. checkpoint-1080/scaler.pt +3 -0
  6. checkpoint-1080/scheduler.pt +3 -0
  7. checkpoint-1080/special_tokens_map.json +37 -0
  8. checkpoint-1080/tokenizer.json +0 -0
  9. checkpoint-1080/tokenizer_config.json +56 -0
  10. checkpoint-1080/trainer_state.json +0 -0
  11. checkpoint-1080/training_args.bin +3 -0
  12. checkpoint-1080/vocab.txt +0 -0
  13. checkpoint-120/config.json +130 -0
  14. checkpoint-120/model.safetensors +3 -0
  15. checkpoint-120/optimizer.pt +3 -0
  16. checkpoint-120/rng_state.pth +3 -0
  17. checkpoint-120/scaler.pt +3 -0
  18. checkpoint-120/scheduler.pt +3 -0
  19. checkpoint-120/special_tokens_map.json +37 -0
  20. checkpoint-120/tokenizer.json +0 -0
  21. checkpoint-120/tokenizer_config.json +56 -0
  22. checkpoint-120/trainer_state.json +886 -0
  23. checkpoint-120/training_args.bin +3 -0
  24. checkpoint-120/vocab.txt +0 -0
  25. checkpoint-1200/config.json +130 -0
  26. checkpoint-1200/model.safetensors +3 -0
  27. checkpoint-1200/optimizer.pt +3 -0
  28. checkpoint-1200/rng_state.pth +3 -0
  29. checkpoint-1200/scaler.pt +3 -0
  30. checkpoint-1200/scheduler.pt +3 -0
  31. checkpoint-1200/special_tokens_map.json +37 -0
  32. checkpoint-1200/tokenizer.json +0 -0
  33. checkpoint-1200/tokenizer_config.json +56 -0
  34. checkpoint-1200/trainer_state.json +0 -0
  35. checkpoint-1200/training_args.bin +3 -0
  36. checkpoint-1200/vocab.txt +0 -0
  37. checkpoint-240/config.json +130 -0
  38. checkpoint-240/model.safetensors +3 -0
  39. checkpoint-240/optimizer.pt +3 -0
  40. checkpoint-240/rng_state.pth +3 -0
  41. checkpoint-240/scaler.pt +3 -0
  42. checkpoint-240/scheduler.pt +3 -0
  43. checkpoint-240/special_tokens_map.json +37 -0
  44. checkpoint-240/tokenizer.json +0 -0
  45. checkpoint-240/tokenizer_config.json +56 -0
  46. checkpoint-240/trainer_state.json +1738 -0
  47. checkpoint-240/training_args.bin +3 -0
  48. checkpoint-240/vocab.txt +0 -0
  49. checkpoint-360/config.json +130 -0
  50. checkpoint-360/model.safetensors +3 -0
checkpoint-1080/config.json ADDED
@@ -0,0 +1,130 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "BertForSequenceClassification"
4
+ ],
5
+ "attention_probs_dropout_prob": 0.1,
6
+ "classifier_dropout": null,
7
+ "gradient_checkpointing": false,
8
+ "hidden_act": "gelu",
9
+ "hidden_dropout_prob": 0.1,
10
+ "hidden_size": 768,
11
+ "id2label": {
12
+ "0": "T1003.001",
13
+ "1": "T1005",
14
+ "2": "T1012",
15
+ "3": "T1016",
16
+ "4": "T1021.001",
17
+ "5": "T1027",
18
+ "6": "T1033",
19
+ "7": "T1036.005",
20
+ "8": "T1041",
21
+ "9": "T1047",
22
+ "10": "T1053.005",
23
+ "11": "T1055",
24
+ "12": "T1056.001",
25
+ "13": "T1057",
26
+ "14": "T1059.003",
27
+ "15": "T1068",
28
+ "16": "T1070.004",
29
+ "17": "T1071.001",
30
+ "18": "T1072",
31
+ "19": "T1074.001",
32
+ "20": "T1078",
33
+ "21": "T1082",
34
+ "22": "T1083",
35
+ "23": "T1090",
36
+ "24": "T1095",
37
+ "25": "T1105",
38
+ "26": "T1106",
39
+ "27": "T1110",
40
+ "28": "T1112",
41
+ "29": "T1113",
42
+ "30": "T1140",
43
+ "31": "T1190",
44
+ "32": "T1204.002",
45
+ "33": "T1210",
46
+ "34": "T1218.011",
47
+ "35": "T1219",
48
+ "36": "T1484.001",
49
+ "37": "T1518.001",
50
+ "38": "T1543.003",
51
+ "39": "T1547.001",
52
+ "40": "T1548.002",
53
+ "41": "T1552.001",
54
+ "42": "T1557.001",
55
+ "43": "T1562.001",
56
+ "44": "T1564.001",
57
+ "45": "T1566.001",
58
+ "46": "T1569.002",
59
+ "47": "T1570",
60
+ "48": "T1573.001",
61
+ "49": "T1574.002"
62
+ },
63
+ "initializer_range": 0.02,
64
+ "intermediate_size": 3072,
65
+ "label2id": {
66
+ "T1003.001": 0,
67
+ "T1005": 1,
68
+ "T1012": 2,
69
+ "T1016": 3,
70
+ "T1021.001": 4,
71
+ "T1027": 5,
72
+ "T1033": 6,
73
+ "T1036.005": 7,
74
+ "T1041": 8,
75
+ "T1047": 9,
76
+ "T1053.005": 10,
77
+ "T1055": 11,
78
+ "T1056.001": 12,
79
+ "T1057": 13,
80
+ "T1059.003": 14,
81
+ "T1068": 15,
82
+ "T1070.004": 16,
83
+ "T1071.001": 17,
84
+ "T1072": 18,
85
+ "T1074.001": 19,
86
+ "T1078": 20,
87
+ "T1082": 21,
88
+ "T1083": 22,
89
+ "T1090": 23,
90
+ "T1095": 24,
91
+ "T1105": 25,
92
+ "T1106": 26,
93
+ "T1110": 27,
94
+ "T1112": 28,
95
+ "T1113": 29,
96
+ "T1140": 30,
97
+ "T1190": 31,
98
+ "T1204.002": 32,
99
+ "T1210": 33,
100
+ "T1218.011": 34,
101
+ "T1219": 35,
102
+ "T1484.001": 36,
103
+ "T1518.001": 37,
104
+ "T1543.003": 38,
105
+ "T1547.001": 39,
106
+ "T1548.002": 40,
107
+ "T1552.001": 41,
108
+ "T1557.001": 42,
109
+ "T1562.001": 43,
110
+ "T1564.001": 44,
111
+ "T1566.001": 45,
112
+ "T1569.002": 46,
113
+ "T1570": 47,
114
+ "T1573.001": 48,
115
+ "T1574.002": 49
116
+ },
117
+ "layer_norm_eps": 1e-12,
118
+ "max_position_embeddings": 512,
119
+ "model_type": "bert",
120
+ "num_attention_heads": 12,
121
+ "num_hidden_layers": 12,
122
+ "pad_token_id": 0,
123
+ "position_embedding_type": "absolute",
124
+ "problem_type": "multi_label_classification",
125
+ "torch_dtype": "float32",
126
+ "transformers_version": "4.55.2",
127
+ "type_vocab_size": 2,
128
+ "use_cache": true,
129
+ "vocab_size": 30522
130
+ }
checkpoint-1080/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1ee15a5639bc3ad9397a2db7b850d0885be735d85fd524e828edcdce9f3316e5
3
+ size 438106296
checkpoint-1080/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b3a9367b42d0899dd2cb9527d5fc098414b990cca7366393f5567ebff7490d32
3
+ size 876333626
checkpoint-1080/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a187c1f1cc6167177af92085f8231de15123eada48166c081f0e5a02659e76e5
3
+ size 14244
checkpoint-1080/scaler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b30172cf14f5dbe00280d63e36224a9f28dc7a0e8b38a74ceb5eb284e84da363
3
+ size 988
checkpoint-1080/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c2f4aee3d9e964f2973aaa9e847fd07b362f160f9383e94bb7d499346d0da15c
3
+ size 1064
checkpoint-1080/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-1080/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-1080/tokenizer_config.json ADDED
@@ -0,0 +1,56 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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": true,
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
+ }
checkpoint-1080/trainer_state.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-1080/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:84b38cc83b3a41fa6c531ffc2fce557168d7eea296dc27400ca0cad58c63e8db
3
+ size 5368
checkpoint-1080/vocab.txt ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-120/config.json ADDED
@@ -0,0 +1,130 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "BertForSequenceClassification"
4
+ ],
5
+ "attention_probs_dropout_prob": 0.1,
6
+ "classifier_dropout": null,
7
+ "gradient_checkpointing": false,
8
+ "hidden_act": "gelu",
9
+ "hidden_dropout_prob": 0.1,
10
+ "hidden_size": 768,
11
+ "id2label": {
12
+ "0": "T1003.001",
13
+ "1": "T1005",
14
+ "2": "T1012",
15
+ "3": "T1016",
16
+ "4": "T1021.001",
17
+ "5": "T1027",
18
+ "6": "T1033",
19
+ "7": "T1036.005",
20
+ "8": "T1041",
21
+ "9": "T1047",
22
+ "10": "T1053.005",
23
+ "11": "T1055",
24
+ "12": "T1056.001",
25
+ "13": "T1057",
26
+ "14": "T1059.003",
27
+ "15": "T1068",
28
+ "16": "T1070.004",
29
+ "17": "T1071.001",
30
+ "18": "T1072",
31
+ "19": "T1074.001",
32
+ "20": "T1078",
33
+ "21": "T1082",
34
+ "22": "T1083",
35
+ "23": "T1090",
36
+ "24": "T1095",
37
+ "25": "T1105",
38
+ "26": "T1106",
39
+ "27": "T1110",
40
+ "28": "T1112",
41
+ "29": "T1113",
42
+ "30": "T1140",
43
+ "31": "T1190",
44
+ "32": "T1204.002",
45
+ "33": "T1210",
46
+ "34": "T1218.011",
47
+ "35": "T1219",
48
+ "36": "T1484.001",
49
+ "37": "T1518.001",
50
+ "38": "T1543.003",
51
+ "39": "T1547.001",
52
+ "40": "T1548.002",
53
+ "41": "T1552.001",
54
+ "42": "T1557.001",
55
+ "43": "T1562.001",
56
+ "44": "T1564.001",
57
+ "45": "T1566.001",
58
+ "46": "T1569.002",
59
+ "47": "T1570",
60
+ "48": "T1573.001",
61
+ "49": "T1574.002"
62
+ },
63
+ "initializer_range": 0.02,
64
+ "intermediate_size": 3072,
65
+ "label2id": {
66
+ "T1003.001": 0,
67
+ "T1005": 1,
68
+ "T1012": 2,
69
+ "T1016": 3,
70
+ "T1021.001": 4,
71
+ "T1027": 5,
72
+ "T1033": 6,
73
+ "T1036.005": 7,
74
+ "T1041": 8,
75
+ "T1047": 9,
76
+ "T1053.005": 10,
77
+ "T1055": 11,
78
+ "T1056.001": 12,
79
+ "T1057": 13,
80
+ "T1059.003": 14,
81
+ "T1068": 15,
82
+ "T1070.004": 16,
83
+ "T1071.001": 17,
84
+ "T1072": 18,
85
+ "T1074.001": 19,
86
+ "T1078": 20,
87
+ "T1082": 21,
88
+ "T1083": 22,
89
+ "T1090": 23,
90
+ "T1095": 24,
91
+ "T1105": 25,
92
+ "T1106": 26,
93
+ "T1110": 27,
94
+ "T1112": 28,
95
+ "T1113": 29,
96
+ "T1140": 30,
97
+ "T1190": 31,
98
+ "T1204.002": 32,
99
+ "T1210": 33,
100
+ "T1218.011": 34,
101
+ "T1219": 35,
102
+ "T1484.001": 36,
103
+ "T1518.001": 37,
104
+ "T1543.003": 38,
105
+ "T1547.001": 39,
106
+ "T1548.002": 40,
107
+ "T1552.001": 41,
108
+ "T1557.001": 42,
109
+ "T1562.001": 43,
110
+ "T1564.001": 44,
111
+ "T1566.001": 45,
112
+ "T1569.002": 46,
113
+ "T1570": 47,
114
+ "T1573.001": 48,
115
+ "T1574.002": 49
116
+ },
117
+ "layer_norm_eps": 1e-12,
118
+ "max_position_embeddings": 512,
119
+ "model_type": "bert",
120
+ "num_attention_heads": 12,
121
+ "num_hidden_layers": 12,
122
+ "pad_token_id": 0,
123
+ "position_embedding_type": "absolute",
124
+ "problem_type": "multi_label_classification",
125
+ "torch_dtype": "float32",
126
+ "transformers_version": "4.55.2",
127
+ "type_vocab_size": 2,
128
+ "use_cache": true,
129
+ "vocab_size": 30522
130
+ }
checkpoint-120/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4cb502270d6da305df71869cf64ed6983b3ccd4feb3543c5483f09396f74604f
3
+ size 438106296
checkpoint-120/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:94126d351c7d66bbe377317cb216a37464dcbb8e635e7fbaf9237933e911f052
3
+ size 876333626
checkpoint-120/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:494b7643d7945d85c2b10891c07f21bc273175bf95d2c62f8679871d1f3567f5
3
+ size 14244
checkpoint-120/scaler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b30172cf14f5dbe00280d63e36224a9f28dc7a0e8b38a74ceb5eb284e84da363
3
+ size 988
checkpoint-120/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:876801210d38ecf8e854b61b6c8fcaf3dff92f0793b73fbb68442c6b95bac0a8
3
+ size 1064
checkpoint-120/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-120/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-120/tokenizer_config.json ADDED
@@ -0,0 +1,56 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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": true,
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
+ }
checkpoint-120/trainer_state.json ADDED
@@ -0,0 +1,886 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 120,
3
+ "best_metric": 0.0,
4
+ "best_model_checkpoint": "./cysecbert-ttp-bert-25per_ood_data/checkpoint-120",
5
+ "epoch": 1.0,
6
+ "eval_steps": 500,
7
+ "global_step": 120,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.008333333333333333,
14
+ "grad_norm": 44649.65625,
15
+ "learning_rate": 0.0,
16
+ "loss": 0.688,
17
+ "step": 1
18
+ },
19
+ {
20
+ "epoch": 0.016666666666666666,
21
+ "grad_norm": 49860.90234375,
22
+ "learning_rate": 5.000000000000001e-07,
23
+ "loss": 0.6867,
24
+ "step": 2
25
+ },
26
+ {
27
+ "epoch": 0.025,
28
+ "grad_norm": 44687.42578125,
29
+ "learning_rate": 1.0000000000000002e-06,
30
+ "loss": 0.6866,
31
+ "step": 3
32
+ },
33
+ {
34
+ "epoch": 0.03333333333333333,
35
+ "grad_norm": 51534.46875,
36
+ "learning_rate": 1.5e-06,
37
+ "loss": 0.6857,
38
+ "step": 4
39
+ },
40
+ {
41
+ "epoch": 0.041666666666666664,
42
+ "grad_norm": 44336.78125,
43
+ "learning_rate": 2.0000000000000003e-06,
44
+ "loss": 0.6824,
45
+ "step": 5
46
+ },
47
+ {
48
+ "epoch": 0.05,
49
+ "grad_norm": 46192.3984375,
50
+ "learning_rate": 2.5e-06,
51
+ "loss": 0.6799,
52
+ "step": 6
53
+ },
54
+ {
55
+ "epoch": 0.058333333333333334,
56
+ "grad_norm": 43591.48828125,
57
+ "learning_rate": 3e-06,
58
+ "loss": 0.6762,
59
+ "step": 7
60
+ },
61
+ {
62
+ "epoch": 0.06666666666666667,
63
+ "grad_norm": 48831.93359375,
64
+ "learning_rate": 3.5000000000000004e-06,
65
+ "loss": 0.6767,
66
+ "step": 8
67
+ },
68
+ {
69
+ "epoch": 0.075,
70
+ "grad_norm": 119280.2734375,
71
+ "learning_rate": 4.000000000000001e-06,
72
+ "loss": 0.673,
73
+ "step": 9
74
+ },
75
+ {
76
+ "epoch": 0.08333333333333333,
77
+ "grad_norm": 46532.3671875,
78
+ "learning_rate": 4.5e-06,
79
+ "loss": 0.6693,
80
+ "step": 10
81
+ },
82
+ {
83
+ "epoch": 0.09166666666666666,
84
+ "grad_norm": 44995.81640625,
85
+ "learning_rate": 5e-06,
86
+ "loss": 0.6663,
87
+ "step": 11
88
+ },
89
+ {
90
+ "epoch": 0.1,
91
+ "grad_norm": 45732.64453125,
92
+ "learning_rate": 5.500000000000001e-06,
93
+ "loss": 0.663,
94
+ "step": 12
95
+ },
96
+ {
97
+ "epoch": 0.10833333333333334,
98
+ "grad_norm": 45771.0234375,
99
+ "learning_rate": 6e-06,
100
+ "loss": 0.6549,
101
+ "step": 13
102
+ },
103
+ {
104
+ "epoch": 0.11666666666666667,
105
+ "grad_norm": 45296.47265625,
106
+ "learning_rate": 6.5000000000000004e-06,
107
+ "loss": 0.6525,
108
+ "step": 14
109
+ },
110
+ {
111
+ "epoch": 0.125,
112
+ "grad_norm": 52268.5390625,
113
+ "learning_rate": 7.000000000000001e-06,
114
+ "loss": 0.6485,
115
+ "step": 15
116
+ },
117
+ {
118
+ "epoch": 0.13333333333333333,
119
+ "grad_norm": 50110.92578125,
120
+ "learning_rate": 7.5e-06,
121
+ "loss": 0.6391,
122
+ "step": 16
123
+ },
124
+ {
125
+ "epoch": 0.14166666666666666,
126
+ "grad_norm": 51549.265625,
127
+ "learning_rate": 8.000000000000001e-06,
128
+ "loss": 0.6349,
129
+ "step": 17
130
+ },
131
+ {
132
+ "epoch": 0.15,
133
+ "grad_norm": 106632.3203125,
134
+ "learning_rate": 8.500000000000002e-06,
135
+ "loss": 0.6323,
136
+ "step": 18
137
+ },
138
+ {
139
+ "epoch": 0.15833333333333333,
140
+ "grad_norm": 57003.57421875,
141
+ "learning_rate": 9e-06,
142
+ "loss": 0.6226,
143
+ "step": 19
144
+ },
145
+ {
146
+ "epoch": 0.16666666666666666,
147
+ "grad_norm": 59664.578125,
148
+ "learning_rate": 9.5e-06,
149
+ "loss": 0.613,
150
+ "step": 20
151
+ },
152
+ {
153
+ "epoch": 0.175,
154
+ "grad_norm": 76325.9140625,
155
+ "learning_rate": 1e-05,
156
+ "loss": 0.5995,
157
+ "step": 21
158
+ },
159
+ {
160
+ "epoch": 0.18333333333333332,
161
+ "grad_norm": 66561.0234375,
162
+ "learning_rate": 1.05e-05,
163
+ "loss": 0.5864,
164
+ "step": 22
165
+ },
166
+ {
167
+ "epoch": 0.19166666666666668,
168
+ "grad_norm": 52762.265625,
169
+ "learning_rate": 1.1000000000000001e-05,
170
+ "loss": 0.5768,
171
+ "step": 23
172
+ },
173
+ {
174
+ "epoch": 0.2,
175
+ "grad_norm": 50716.7890625,
176
+ "learning_rate": 1.1500000000000002e-05,
177
+ "loss": 0.5654,
178
+ "step": 24
179
+ },
180
+ {
181
+ "epoch": 0.20833333333333334,
182
+ "grad_norm": 53293.3046875,
183
+ "learning_rate": 1.2e-05,
184
+ "loss": 0.5525,
185
+ "step": 25
186
+ },
187
+ {
188
+ "epoch": 0.21666666666666667,
189
+ "grad_norm": 54021.37109375,
190
+ "learning_rate": 1.25e-05,
191
+ "loss": 0.5529,
192
+ "step": 26
193
+ },
194
+ {
195
+ "epoch": 0.225,
196
+ "grad_norm": 54463.3515625,
197
+ "learning_rate": 1.3000000000000001e-05,
198
+ "loss": 0.5384,
199
+ "step": 27
200
+ },
201
+ {
202
+ "epoch": 0.23333333333333334,
203
+ "grad_norm": 77561.1484375,
204
+ "learning_rate": 1.3500000000000001e-05,
205
+ "loss": 0.5316,
206
+ "step": 28
207
+ },
208
+ {
209
+ "epoch": 0.24166666666666667,
210
+ "grad_norm": 52105.3515625,
211
+ "learning_rate": 1.4000000000000001e-05,
212
+ "loss": 0.5191,
213
+ "step": 29
214
+ },
215
+ {
216
+ "epoch": 0.25,
217
+ "grad_norm": 46957.71484375,
218
+ "learning_rate": 1.45e-05,
219
+ "loss": 0.507,
220
+ "step": 30
221
+ },
222
+ {
223
+ "epoch": 0.25833333333333336,
224
+ "grad_norm": 46143.05078125,
225
+ "learning_rate": 1.5e-05,
226
+ "loss": 0.4961,
227
+ "step": 31
228
+ },
229
+ {
230
+ "epoch": 0.26666666666666666,
231
+ "grad_norm": 44479.36328125,
232
+ "learning_rate": 1.55e-05,
233
+ "loss": 0.4846,
234
+ "step": 32
235
+ },
236
+ {
237
+ "epoch": 0.275,
238
+ "grad_norm": 49707.58203125,
239
+ "learning_rate": 1.6000000000000003e-05,
240
+ "loss": 0.4793,
241
+ "step": 33
242
+ },
243
+ {
244
+ "epoch": 0.2833333333333333,
245
+ "grad_norm": 46631.30078125,
246
+ "learning_rate": 1.65e-05,
247
+ "loss": 0.4686,
248
+ "step": 34
249
+ },
250
+ {
251
+ "epoch": 0.2916666666666667,
252
+ "grad_norm": 43869.328125,
253
+ "learning_rate": 1.7000000000000003e-05,
254
+ "loss": 0.465,
255
+ "step": 35
256
+ },
257
+ {
258
+ "epoch": 0.3,
259
+ "grad_norm": 42714.52734375,
260
+ "learning_rate": 1.75e-05,
261
+ "loss": 0.4569,
262
+ "step": 36
263
+ },
264
+ {
265
+ "epoch": 0.30833333333333335,
266
+ "grad_norm": 42391.1953125,
267
+ "learning_rate": 1.8e-05,
268
+ "loss": 0.4444,
269
+ "step": 37
270
+ },
271
+ {
272
+ "epoch": 0.31666666666666665,
273
+ "grad_norm": 42775.91796875,
274
+ "learning_rate": 1.85e-05,
275
+ "loss": 0.4346,
276
+ "step": 38
277
+ },
278
+ {
279
+ "epoch": 0.325,
280
+ "grad_norm": 45365.82421875,
281
+ "learning_rate": 1.9e-05,
282
+ "loss": 0.4334,
283
+ "step": 39
284
+ },
285
+ {
286
+ "epoch": 0.3333333333333333,
287
+ "grad_norm": 43367.6875,
288
+ "learning_rate": 1.9500000000000003e-05,
289
+ "loss": 0.4211,
290
+ "step": 40
291
+ },
292
+ {
293
+ "epoch": 0.3416666666666667,
294
+ "grad_norm": 40098.22265625,
295
+ "learning_rate": 2e-05,
296
+ "loss": 0.4069,
297
+ "step": 41
298
+ },
299
+ {
300
+ "epoch": 0.35,
301
+ "grad_norm": 39089.37109375,
302
+ "learning_rate": 2.05e-05,
303
+ "loss": 0.4048,
304
+ "step": 42
305
+ },
306
+ {
307
+ "epoch": 0.35833333333333334,
308
+ "grad_norm": 39849.1328125,
309
+ "learning_rate": 2.1e-05,
310
+ "loss": 0.3931,
311
+ "step": 43
312
+ },
313
+ {
314
+ "epoch": 0.36666666666666664,
315
+ "grad_norm": 39146.015625,
316
+ "learning_rate": 2.15e-05,
317
+ "loss": 0.3877,
318
+ "step": 44
319
+ },
320
+ {
321
+ "epoch": 0.375,
322
+ "grad_norm": 38992.43359375,
323
+ "learning_rate": 2.2000000000000003e-05,
324
+ "loss": 0.3791,
325
+ "step": 45
326
+ },
327
+ {
328
+ "epoch": 0.38333333333333336,
329
+ "grad_norm": 37866.18359375,
330
+ "learning_rate": 2.25e-05,
331
+ "loss": 0.3726,
332
+ "step": 46
333
+ },
334
+ {
335
+ "epoch": 0.39166666666666666,
336
+ "grad_norm": 37756.671875,
337
+ "learning_rate": 2.3000000000000003e-05,
338
+ "loss": 0.364,
339
+ "step": 47
340
+ },
341
+ {
342
+ "epoch": 0.4,
343
+ "grad_norm": 37182.94921875,
344
+ "learning_rate": 2.35e-05,
345
+ "loss": 0.3605,
346
+ "step": 48
347
+ },
348
+ {
349
+ "epoch": 0.4083333333333333,
350
+ "grad_norm": 36981.40234375,
351
+ "learning_rate": 2.4e-05,
352
+ "loss": 0.3522,
353
+ "step": 49
354
+ },
355
+ {
356
+ "epoch": 0.4166666666666667,
357
+ "grad_norm": 36842.3984375,
358
+ "learning_rate": 2.45e-05,
359
+ "loss": 0.3421,
360
+ "step": 50
361
+ },
362
+ {
363
+ "epoch": 0.425,
364
+ "grad_norm": 35325.8828125,
365
+ "learning_rate": 2.5e-05,
366
+ "loss": 0.3442,
367
+ "step": 51
368
+ },
369
+ {
370
+ "epoch": 0.43333333333333335,
371
+ "grad_norm": 35363.078125,
372
+ "learning_rate": 2.5500000000000003e-05,
373
+ "loss": 0.3318,
374
+ "step": 52
375
+ },
376
+ {
377
+ "epoch": 0.44166666666666665,
378
+ "grad_norm": 35645.59765625,
379
+ "learning_rate": 2.6000000000000002e-05,
380
+ "loss": 0.3277,
381
+ "step": 53
382
+ },
383
+ {
384
+ "epoch": 0.45,
385
+ "grad_norm": 34691.13671875,
386
+ "learning_rate": 2.6500000000000004e-05,
387
+ "loss": 0.3171,
388
+ "step": 54
389
+ },
390
+ {
391
+ "epoch": 0.4583333333333333,
392
+ "grad_norm": 34211.484375,
393
+ "learning_rate": 2.7000000000000002e-05,
394
+ "loss": 0.3084,
395
+ "step": 55
396
+ },
397
+ {
398
+ "epoch": 0.4666666666666667,
399
+ "grad_norm": 33361.140625,
400
+ "learning_rate": 2.7500000000000004e-05,
401
+ "loss": 0.3076,
402
+ "step": 56
403
+ },
404
+ {
405
+ "epoch": 0.475,
406
+ "grad_norm": 32967.97265625,
407
+ "learning_rate": 2.8000000000000003e-05,
408
+ "loss": 0.298,
409
+ "step": 57
410
+ },
411
+ {
412
+ "epoch": 0.48333333333333334,
413
+ "grad_norm": 32798.56640625,
414
+ "learning_rate": 2.8499999999999998e-05,
415
+ "loss": 0.2929,
416
+ "step": 58
417
+ },
418
+ {
419
+ "epoch": 0.49166666666666664,
420
+ "grad_norm": 31929.83203125,
421
+ "learning_rate": 2.9e-05,
422
+ "loss": 0.2907,
423
+ "step": 59
424
+ },
425
+ {
426
+ "epoch": 0.5,
427
+ "grad_norm": 31776.216796875,
428
+ "learning_rate": 2.95e-05,
429
+ "loss": 0.2797,
430
+ "step": 60
431
+ },
432
+ {
433
+ "epoch": 0.5083333333333333,
434
+ "grad_norm": 31287.02734375,
435
+ "learning_rate": 3e-05,
436
+ "loss": 0.2724,
437
+ "step": 61
438
+ },
439
+ {
440
+ "epoch": 0.5166666666666667,
441
+ "grad_norm": 30368.2421875,
442
+ "learning_rate": 3.05e-05,
443
+ "loss": 0.2658,
444
+ "step": 62
445
+ },
446
+ {
447
+ "epoch": 0.525,
448
+ "grad_norm": 29610.740234375,
449
+ "learning_rate": 3.1e-05,
450
+ "loss": 0.2606,
451
+ "step": 63
452
+ },
453
+ {
454
+ "epoch": 0.5333333333333333,
455
+ "grad_norm": 29145.74609375,
456
+ "learning_rate": 3.15e-05,
457
+ "loss": 0.2556,
458
+ "step": 64
459
+ },
460
+ {
461
+ "epoch": 0.5416666666666666,
462
+ "grad_norm": 28216.568359375,
463
+ "learning_rate": 3.2000000000000005e-05,
464
+ "loss": 0.253,
465
+ "step": 65
466
+ },
467
+ {
468
+ "epoch": 0.55,
469
+ "grad_norm": 28172.978515625,
470
+ "learning_rate": 3.2500000000000004e-05,
471
+ "loss": 0.2418,
472
+ "step": 66
473
+ },
474
+ {
475
+ "epoch": 0.5583333333333333,
476
+ "grad_norm": 26664.701171875,
477
+ "learning_rate": 3.3e-05,
478
+ "loss": 0.2427,
479
+ "step": 67
480
+ },
481
+ {
482
+ "epoch": 0.5666666666666667,
483
+ "grad_norm": 26883.462890625,
484
+ "learning_rate": 3.35e-05,
485
+ "loss": 0.2346,
486
+ "step": 68
487
+ },
488
+ {
489
+ "epoch": 0.575,
490
+ "grad_norm": 26251.072265625,
491
+ "learning_rate": 3.4000000000000007e-05,
492
+ "loss": 0.2292,
493
+ "step": 69
494
+ },
495
+ {
496
+ "epoch": 0.5833333333333334,
497
+ "grad_norm": 25158.37109375,
498
+ "learning_rate": 3.45e-05,
499
+ "loss": 0.2277,
500
+ "step": 70
501
+ },
502
+ {
503
+ "epoch": 0.5916666666666667,
504
+ "grad_norm": 25251.408203125,
505
+ "learning_rate": 3.5e-05,
506
+ "loss": 0.2119,
507
+ "step": 71
508
+ },
509
+ {
510
+ "epoch": 0.6,
511
+ "grad_norm": 55416.34765625,
512
+ "learning_rate": 3.55e-05,
513
+ "loss": 0.2059,
514
+ "step": 72
515
+ },
516
+ {
517
+ "epoch": 0.6083333333333333,
518
+ "grad_norm": 23741.2421875,
519
+ "learning_rate": 3.6e-05,
520
+ "loss": 0.2014,
521
+ "step": 73
522
+ },
523
+ {
524
+ "epoch": 0.6166666666666667,
525
+ "grad_norm": 22732.470703125,
526
+ "learning_rate": 3.65e-05,
527
+ "loss": 0.2058,
528
+ "step": 74
529
+ },
530
+ {
531
+ "epoch": 0.625,
532
+ "grad_norm": 22756.88671875,
533
+ "learning_rate": 3.7e-05,
534
+ "loss": 0.1922,
535
+ "step": 75
536
+ },
537
+ {
538
+ "epoch": 0.6333333333333333,
539
+ "grad_norm": 22566.482421875,
540
+ "learning_rate": 3.7500000000000003e-05,
541
+ "loss": 0.1892,
542
+ "step": 76
543
+ },
544
+ {
545
+ "epoch": 0.6416666666666667,
546
+ "grad_norm": 21044.404296875,
547
+ "learning_rate": 3.8e-05,
548
+ "loss": 0.1895,
549
+ "step": 77
550
+ },
551
+ {
552
+ "epoch": 0.65,
553
+ "grad_norm": 19801.0390625,
554
+ "learning_rate": 3.85e-05,
555
+ "loss": 0.2051,
556
+ "step": 78
557
+ },
558
+ {
559
+ "epoch": 0.6583333333333333,
560
+ "grad_norm": 20280.55078125,
561
+ "learning_rate": 3.9000000000000006e-05,
562
+ "loss": 0.1739,
563
+ "step": 79
564
+ },
565
+ {
566
+ "epoch": 0.6666666666666666,
567
+ "grad_norm": 21561.10546875,
568
+ "learning_rate": 3.9500000000000005e-05,
569
+ "loss": 0.1748,
570
+ "step": 80
571
+ },
572
+ {
573
+ "epoch": 0.675,
574
+ "grad_norm": 18942.365234375,
575
+ "learning_rate": 4e-05,
576
+ "loss": 0.1607,
577
+ "step": 81
578
+ },
579
+ {
580
+ "epoch": 0.6833333333333333,
581
+ "grad_norm": 18100.796875,
582
+ "learning_rate": 4.05e-05,
583
+ "loss": 0.1581,
584
+ "step": 82
585
+ },
586
+ {
587
+ "epoch": 0.6916666666666667,
588
+ "grad_norm": 17815.296875,
589
+ "learning_rate": 4.1e-05,
590
+ "loss": 0.1603,
591
+ "step": 83
592
+ },
593
+ {
594
+ "epoch": 0.7,
595
+ "grad_norm": 17038.046875,
596
+ "learning_rate": 4.15e-05,
597
+ "loss": 0.1543,
598
+ "step": 84
599
+ },
600
+ {
601
+ "epoch": 0.7083333333333334,
602
+ "grad_norm": 16171.5625,
603
+ "learning_rate": 4.2e-05,
604
+ "loss": 0.1622,
605
+ "step": 85
606
+ },
607
+ {
608
+ "epoch": 0.7166666666666667,
609
+ "grad_norm": 15270.400390625,
610
+ "learning_rate": 4.25e-05,
611
+ "loss": 0.1621,
612
+ "step": 86
613
+ },
614
+ {
615
+ "epoch": 0.725,
616
+ "grad_norm": 15798.998046875,
617
+ "learning_rate": 4.3e-05,
618
+ "loss": 0.147,
619
+ "step": 87
620
+ },
621
+ {
622
+ "epoch": 0.7333333333333333,
623
+ "grad_norm": 14164.265625,
624
+ "learning_rate": 4.35e-05,
625
+ "loss": 0.1485,
626
+ "step": 88
627
+ },
628
+ {
629
+ "epoch": 0.7416666666666667,
630
+ "grad_norm": 15282.1884765625,
631
+ "learning_rate": 4.4000000000000006e-05,
632
+ "loss": 0.1455,
633
+ "step": 89
634
+ },
635
+ {
636
+ "epoch": 0.75,
637
+ "grad_norm": 14389.99609375,
638
+ "learning_rate": 4.4500000000000004e-05,
639
+ "loss": 0.1358,
640
+ "step": 90
641
+ },
642
+ {
643
+ "epoch": 0.7583333333333333,
644
+ "grad_norm": 13489.986328125,
645
+ "learning_rate": 4.5e-05,
646
+ "loss": 0.1308,
647
+ "step": 91
648
+ },
649
+ {
650
+ "epoch": 0.7666666666666667,
651
+ "grad_norm": 12742.55078125,
652
+ "learning_rate": 4.55e-05,
653
+ "loss": 0.1409,
654
+ "step": 92
655
+ },
656
+ {
657
+ "epoch": 0.775,
658
+ "grad_norm": 11714.9228515625,
659
+ "learning_rate": 4.600000000000001e-05,
660
+ "loss": 0.1316,
661
+ "step": 93
662
+ },
663
+ {
664
+ "epoch": 0.7833333333333333,
665
+ "grad_norm": 11921.5185546875,
666
+ "learning_rate": 4.6500000000000005e-05,
667
+ "loss": 0.1226,
668
+ "step": 94
669
+ },
670
+ {
671
+ "epoch": 0.7916666666666666,
672
+ "grad_norm": 11369.017578125,
673
+ "learning_rate": 4.7e-05,
674
+ "loss": 0.1302,
675
+ "step": 95
676
+ },
677
+ {
678
+ "epoch": 0.8,
679
+ "grad_norm": 10840.740234375,
680
+ "learning_rate": 4.75e-05,
681
+ "loss": 0.1276,
682
+ "step": 96
683
+ },
684
+ {
685
+ "epoch": 0.8083333333333333,
686
+ "grad_norm": 10310.0458984375,
687
+ "learning_rate": 4.8e-05,
688
+ "loss": 0.1496,
689
+ "step": 97
690
+ },
691
+ {
692
+ "epoch": 0.8166666666666667,
693
+ "grad_norm": 10777.8271484375,
694
+ "learning_rate": 4.85e-05,
695
+ "loss": 0.1249,
696
+ "step": 98
697
+ },
698
+ {
699
+ "epoch": 0.825,
700
+ "grad_norm": 10753.0048828125,
701
+ "learning_rate": 4.9e-05,
702
+ "loss": 0.1336,
703
+ "step": 99
704
+ },
705
+ {
706
+ "epoch": 0.8333333333333334,
707
+ "grad_norm": 9117.2509765625,
708
+ "learning_rate": 4.9500000000000004e-05,
709
+ "loss": 0.121,
710
+ "step": 100
711
+ },
712
+ {
713
+ "epoch": 0.8416666666666667,
714
+ "grad_norm": 10000.8408203125,
715
+ "learning_rate": 5e-05,
716
+ "loss": 0.1071,
717
+ "step": 101
718
+ },
719
+ {
720
+ "epoch": 0.85,
721
+ "grad_norm": 9281.32421875,
722
+ "learning_rate": 4.995454545454546e-05,
723
+ "loss": 0.1183,
724
+ "step": 102
725
+ },
726
+ {
727
+ "epoch": 0.8583333333333333,
728
+ "grad_norm": 9341.63671875,
729
+ "learning_rate": 4.990909090909091e-05,
730
+ "loss": 0.1177,
731
+ "step": 103
732
+ },
733
+ {
734
+ "epoch": 0.8666666666666667,
735
+ "grad_norm": 8904.6572265625,
736
+ "learning_rate": 4.986363636363637e-05,
737
+ "loss": 0.1116,
738
+ "step": 104
739
+ },
740
+ {
741
+ "epoch": 0.875,
742
+ "grad_norm": 8144.94091796875,
743
+ "learning_rate": 4.981818181818182e-05,
744
+ "loss": 0.1083,
745
+ "step": 105
746
+ },
747
+ {
748
+ "epoch": 0.8833333333333333,
749
+ "grad_norm": 8514.19140625,
750
+ "learning_rate": 4.9772727272727275e-05,
751
+ "loss": 0.1208,
752
+ "step": 106
753
+ },
754
+ {
755
+ "epoch": 0.8916666666666667,
756
+ "grad_norm": 9342.4384765625,
757
+ "learning_rate": 4.9727272727272725e-05,
758
+ "loss": 0.108,
759
+ "step": 107
760
+ },
761
+ {
762
+ "epoch": 0.9,
763
+ "grad_norm": 7503.6064453125,
764
+ "learning_rate": 4.968181818181818e-05,
765
+ "loss": 0.1121,
766
+ "step": 108
767
+ },
768
+ {
769
+ "epoch": 0.9083333333333333,
770
+ "grad_norm": 7928.822265625,
771
+ "learning_rate": 4.963636363636364e-05,
772
+ "loss": 0.1123,
773
+ "step": 109
774
+ },
775
+ {
776
+ "epoch": 0.9166666666666666,
777
+ "grad_norm": 7986.73388671875,
778
+ "learning_rate": 4.9590909090909096e-05,
779
+ "loss": 0.1028,
780
+ "step": 110
781
+ },
782
+ {
783
+ "epoch": 0.925,
784
+ "grad_norm": 7877.8583984375,
785
+ "learning_rate": 4.9545454545454553e-05,
786
+ "loss": 0.1077,
787
+ "step": 111
788
+ },
789
+ {
790
+ "epoch": 0.9333333333333333,
791
+ "grad_norm": 8335.400390625,
792
+ "learning_rate": 4.9500000000000004e-05,
793
+ "loss": 0.1126,
794
+ "step": 112
795
+ },
796
+ {
797
+ "epoch": 0.9416666666666667,
798
+ "grad_norm": 7083.4833984375,
799
+ "learning_rate": 4.945454545454546e-05,
800
+ "loss": 0.1098,
801
+ "step": 113
802
+ },
803
+ {
804
+ "epoch": 0.95,
805
+ "grad_norm": 9419.419921875,
806
+ "learning_rate": 4.940909090909091e-05,
807
+ "loss": 0.1058,
808
+ "step": 114
809
+ },
810
+ {
811
+ "epoch": 0.9583333333333334,
812
+ "grad_norm": 7879.18994140625,
813
+ "learning_rate": 4.936363636363637e-05,
814
+ "loss": 0.1081,
815
+ "step": 115
816
+ },
817
+ {
818
+ "epoch": 0.9666666666666667,
819
+ "grad_norm": 8113.90283203125,
820
+ "learning_rate": 4.931818181818182e-05,
821
+ "loss": 0.1075,
822
+ "step": 116
823
+ },
824
+ {
825
+ "epoch": 0.975,
826
+ "grad_norm": 7590.12744140625,
827
+ "learning_rate": 4.9272727272727276e-05,
828
+ "loss": 0.1041,
829
+ "step": 117
830
+ },
831
+ {
832
+ "epoch": 0.9833333333333333,
833
+ "grad_norm": 7461.189453125,
834
+ "learning_rate": 4.9227272727272726e-05,
835
+ "loss": 0.1064,
836
+ "step": 118
837
+ },
838
+ {
839
+ "epoch": 0.9916666666666667,
840
+ "grad_norm": 7857.06298828125,
841
+ "learning_rate": 4.9181818181818183e-05,
842
+ "loss": 0.1102,
843
+ "step": 119
844
+ },
845
+ {
846
+ "epoch": 1.0,
847
+ "grad_norm": 34157.33203125,
848
+ "learning_rate": 4.913636363636364e-05,
849
+ "loss": 0.1032,
850
+ "step": 120
851
+ },
852
+ {
853
+ "epoch": 1.0,
854
+ "eval_f1_macro": 0.0,
855
+ "eval_f1_micro": 0.0,
856
+ "eval_loss": 0.10901560634374619,
857
+ "eval_precision": 0.0,
858
+ "eval_recall": 0.0,
859
+ "eval_runtime": 0.8962,
860
+ "eval_samples_per_second": 354.843,
861
+ "eval_steps_per_second": 15.622,
862
+ "step": 120
863
+ }
864
+ ],
865
+ "logging_steps": 1,
866
+ "max_steps": 1200,
867
+ "num_input_tokens_seen": 0,
868
+ "num_train_epochs": 10,
869
+ "save_steps": 500,
870
+ "stateful_callbacks": {
871
+ "TrainerControl": {
872
+ "args": {
873
+ "should_epoch_stop": false,
874
+ "should_evaluate": false,
875
+ "should_log": false,
876
+ "should_save": true,
877
+ "should_training_stop": false
878
+ },
879
+ "attributes": {}
880
+ }
881
+ },
882
+ "total_flos": 752032250861568.0,
883
+ "train_batch_size": 24,
884
+ "trial_name": null,
885
+ "trial_params": null
886
+ }
checkpoint-120/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:84b38cc83b3a41fa6c531ffc2fce557168d7eea296dc27400ca0cad58c63e8db
3
+ size 5368
checkpoint-120/vocab.txt ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-1200/config.json ADDED
@@ -0,0 +1,130 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "BertForSequenceClassification"
4
+ ],
5
+ "attention_probs_dropout_prob": 0.1,
6
+ "classifier_dropout": null,
7
+ "gradient_checkpointing": false,
8
+ "hidden_act": "gelu",
9
+ "hidden_dropout_prob": 0.1,
10
+ "hidden_size": 768,
11
+ "id2label": {
12
+ "0": "T1003.001",
13
+ "1": "T1005",
14
+ "2": "T1012",
15
+ "3": "T1016",
16
+ "4": "T1021.001",
17
+ "5": "T1027",
18
+ "6": "T1033",
19
+ "7": "T1036.005",
20
+ "8": "T1041",
21
+ "9": "T1047",
22
+ "10": "T1053.005",
23
+ "11": "T1055",
24
+ "12": "T1056.001",
25
+ "13": "T1057",
26
+ "14": "T1059.003",
27
+ "15": "T1068",
28
+ "16": "T1070.004",
29
+ "17": "T1071.001",
30
+ "18": "T1072",
31
+ "19": "T1074.001",
32
+ "20": "T1078",
33
+ "21": "T1082",
34
+ "22": "T1083",
35
+ "23": "T1090",
36
+ "24": "T1095",
37
+ "25": "T1105",
38
+ "26": "T1106",
39
+ "27": "T1110",
40
+ "28": "T1112",
41
+ "29": "T1113",
42
+ "30": "T1140",
43
+ "31": "T1190",
44
+ "32": "T1204.002",
45
+ "33": "T1210",
46
+ "34": "T1218.011",
47
+ "35": "T1219",
48
+ "36": "T1484.001",
49
+ "37": "T1518.001",
50
+ "38": "T1543.003",
51
+ "39": "T1547.001",
52
+ "40": "T1548.002",
53
+ "41": "T1552.001",
54
+ "42": "T1557.001",
55
+ "43": "T1562.001",
56
+ "44": "T1564.001",
57
+ "45": "T1566.001",
58
+ "46": "T1569.002",
59
+ "47": "T1570",
60
+ "48": "T1573.001",
61
+ "49": "T1574.002"
62
+ },
63
+ "initializer_range": 0.02,
64
+ "intermediate_size": 3072,
65
+ "label2id": {
66
+ "T1003.001": 0,
67
+ "T1005": 1,
68
+ "T1012": 2,
69
+ "T1016": 3,
70
+ "T1021.001": 4,
71
+ "T1027": 5,
72
+ "T1033": 6,
73
+ "T1036.005": 7,
74
+ "T1041": 8,
75
+ "T1047": 9,
76
+ "T1053.005": 10,
77
+ "T1055": 11,
78
+ "T1056.001": 12,
79
+ "T1057": 13,
80
+ "T1059.003": 14,
81
+ "T1068": 15,
82
+ "T1070.004": 16,
83
+ "T1071.001": 17,
84
+ "T1072": 18,
85
+ "T1074.001": 19,
86
+ "T1078": 20,
87
+ "T1082": 21,
88
+ "T1083": 22,
89
+ "T1090": 23,
90
+ "T1095": 24,
91
+ "T1105": 25,
92
+ "T1106": 26,
93
+ "T1110": 27,
94
+ "T1112": 28,
95
+ "T1113": 29,
96
+ "T1140": 30,
97
+ "T1190": 31,
98
+ "T1204.002": 32,
99
+ "T1210": 33,
100
+ "T1218.011": 34,
101
+ "T1219": 35,
102
+ "T1484.001": 36,
103
+ "T1518.001": 37,
104
+ "T1543.003": 38,
105
+ "T1547.001": 39,
106
+ "T1548.002": 40,
107
+ "T1552.001": 41,
108
+ "T1557.001": 42,
109
+ "T1562.001": 43,
110
+ "T1564.001": 44,
111
+ "T1566.001": 45,
112
+ "T1569.002": 46,
113
+ "T1570": 47,
114
+ "T1573.001": 48,
115
+ "T1574.002": 49
116
+ },
117
+ "layer_norm_eps": 1e-12,
118
+ "max_position_embeddings": 512,
119
+ "model_type": "bert",
120
+ "num_attention_heads": 12,
121
+ "num_hidden_layers": 12,
122
+ "pad_token_id": 0,
123
+ "position_embedding_type": "absolute",
124
+ "problem_type": "multi_label_classification",
125
+ "torch_dtype": "float32",
126
+ "transformers_version": "4.55.2",
127
+ "type_vocab_size": 2,
128
+ "use_cache": true,
129
+ "vocab_size": 30522
130
+ }
checkpoint-1200/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7229b8963da97a1776c7170a00a3c184cce51d021a41d592c0055040c2d21178
3
+ size 438106296
checkpoint-1200/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ef8ab7f0a022c79892bb5d087b213b924435cd07b4cddfc09a9a9cf8aa143dff
3
+ size 876333626
checkpoint-1200/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:069b7fdd7340f3dd03f095241f597c2895cd7257195dd50765f9ff08b0ecc7cc
3
+ size 14244
checkpoint-1200/scaler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b30172cf14f5dbe00280d63e36224a9f28dc7a0e8b38a74ceb5eb284e84da363
3
+ size 988
checkpoint-1200/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:91e4eb7cd9d12f59e34fbe1b67158cadecdeb47748a3bd1a542758468e96c52c
3
+ size 1064
checkpoint-1200/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-1200/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-1200/tokenizer_config.json ADDED
@@ -0,0 +1,56 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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": true,
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
+ }
checkpoint-1200/trainer_state.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-1200/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:84b38cc83b3a41fa6c531ffc2fce557168d7eea296dc27400ca0cad58c63e8db
3
+ size 5368
checkpoint-1200/vocab.txt ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-240/config.json ADDED
@@ -0,0 +1,130 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "BertForSequenceClassification"
4
+ ],
5
+ "attention_probs_dropout_prob": 0.1,
6
+ "classifier_dropout": null,
7
+ "gradient_checkpointing": false,
8
+ "hidden_act": "gelu",
9
+ "hidden_dropout_prob": 0.1,
10
+ "hidden_size": 768,
11
+ "id2label": {
12
+ "0": "T1003.001",
13
+ "1": "T1005",
14
+ "2": "T1012",
15
+ "3": "T1016",
16
+ "4": "T1021.001",
17
+ "5": "T1027",
18
+ "6": "T1033",
19
+ "7": "T1036.005",
20
+ "8": "T1041",
21
+ "9": "T1047",
22
+ "10": "T1053.005",
23
+ "11": "T1055",
24
+ "12": "T1056.001",
25
+ "13": "T1057",
26
+ "14": "T1059.003",
27
+ "15": "T1068",
28
+ "16": "T1070.004",
29
+ "17": "T1071.001",
30
+ "18": "T1072",
31
+ "19": "T1074.001",
32
+ "20": "T1078",
33
+ "21": "T1082",
34
+ "22": "T1083",
35
+ "23": "T1090",
36
+ "24": "T1095",
37
+ "25": "T1105",
38
+ "26": "T1106",
39
+ "27": "T1110",
40
+ "28": "T1112",
41
+ "29": "T1113",
42
+ "30": "T1140",
43
+ "31": "T1190",
44
+ "32": "T1204.002",
45
+ "33": "T1210",
46
+ "34": "T1218.011",
47
+ "35": "T1219",
48
+ "36": "T1484.001",
49
+ "37": "T1518.001",
50
+ "38": "T1543.003",
51
+ "39": "T1547.001",
52
+ "40": "T1548.002",
53
+ "41": "T1552.001",
54
+ "42": "T1557.001",
55
+ "43": "T1562.001",
56
+ "44": "T1564.001",
57
+ "45": "T1566.001",
58
+ "46": "T1569.002",
59
+ "47": "T1570",
60
+ "48": "T1573.001",
61
+ "49": "T1574.002"
62
+ },
63
+ "initializer_range": 0.02,
64
+ "intermediate_size": 3072,
65
+ "label2id": {
66
+ "T1003.001": 0,
67
+ "T1005": 1,
68
+ "T1012": 2,
69
+ "T1016": 3,
70
+ "T1021.001": 4,
71
+ "T1027": 5,
72
+ "T1033": 6,
73
+ "T1036.005": 7,
74
+ "T1041": 8,
75
+ "T1047": 9,
76
+ "T1053.005": 10,
77
+ "T1055": 11,
78
+ "T1056.001": 12,
79
+ "T1057": 13,
80
+ "T1059.003": 14,
81
+ "T1068": 15,
82
+ "T1070.004": 16,
83
+ "T1071.001": 17,
84
+ "T1072": 18,
85
+ "T1074.001": 19,
86
+ "T1078": 20,
87
+ "T1082": 21,
88
+ "T1083": 22,
89
+ "T1090": 23,
90
+ "T1095": 24,
91
+ "T1105": 25,
92
+ "T1106": 26,
93
+ "T1110": 27,
94
+ "T1112": 28,
95
+ "T1113": 29,
96
+ "T1140": 30,
97
+ "T1190": 31,
98
+ "T1204.002": 32,
99
+ "T1210": 33,
100
+ "T1218.011": 34,
101
+ "T1219": 35,
102
+ "T1484.001": 36,
103
+ "T1518.001": 37,
104
+ "T1543.003": 38,
105
+ "T1547.001": 39,
106
+ "T1548.002": 40,
107
+ "T1552.001": 41,
108
+ "T1557.001": 42,
109
+ "T1562.001": 43,
110
+ "T1564.001": 44,
111
+ "T1566.001": 45,
112
+ "T1569.002": 46,
113
+ "T1570": 47,
114
+ "T1573.001": 48,
115
+ "T1574.002": 49
116
+ },
117
+ "layer_norm_eps": 1e-12,
118
+ "max_position_embeddings": 512,
119
+ "model_type": "bert",
120
+ "num_attention_heads": 12,
121
+ "num_hidden_layers": 12,
122
+ "pad_token_id": 0,
123
+ "position_embedding_type": "absolute",
124
+ "problem_type": "multi_label_classification",
125
+ "torch_dtype": "float32",
126
+ "transformers_version": "4.55.2",
127
+ "type_vocab_size": 2,
128
+ "use_cache": true,
129
+ "vocab_size": 30522
130
+ }
checkpoint-240/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bf74ab022e235173a68d36b734861b403cf22859564320933ae58472de49041d
3
+ size 438106296
checkpoint-240/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dc46728cc46ae60f0dceaeaeb54ff7169c8d3ed10edf9d005f35054a9fda735e
3
+ size 876333626
checkpoint-240/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:81c4210d647a3f04240efbbbd2e120a06d6a5c91fad404654cec3a50df6b0471
3
+ size 14244
checkpoint-240/scaler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b30172cf14f5dbe00280d63e36224a9f28dc7a0e8b38a74ceb5eb284e84da363
3
+ size 988
checkpoint-240/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:55ec425a1e959fc72e5949a5e4d58589408fdf368647f2529a4265a4f313d111
3
+ size 1064
checkpoint-240/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-240/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-240/tokenizer_config.json ADDED
@@ -0,0 +1,56 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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": true,
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
+ }
checkpoint-240/trainer_state.json ADDED
@@ -0,0 +1,1738 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 240,
3
+ "best_metric": 0.24436090225563908,
4
+ "best_model_checkpoint": "./cysecbert-ttp-bert-25per_ood_data/checkpoint-240",
5
+ "epoch": 2.0,
6
+ "eval_steps": 500,
7
+ "global_step": 240,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.008333333333333333,
14
+ "grad_norm": 44649.65625,
15
+ "learning_rate": 0.0,
16
+ "loss": 0.688,
17
+ "step": 1
18
+ },
19
+ {
20
+ "epoch": 0.016666666666666666,
21
+ "grad_norm": 49860.90234375,
22
+ "learning_rate": 5.000000000000001e-07,
23
+ "loss": 0.6867,
24
+ "step": 2
25
+ },
26
+ {
27
+ "epoch": 0.025,
28
+ "grad_norm": 44687.42578125,
29
+ "learning_rate": 1.0000000000000002e-06,
30
+ "loss": 0.6866,
31
+ "step": 3
32
+ },
33
+ {
34
+ "epoch": 0.03333333333333333,
35
+ "grad_norm": 51534.46875,
36
+ "learning_rate": 1.5e-06,
37
+ "loss": 0.6857,
38
+ "step": 4
39
+ },
40
+ {
41
+ "epoch": 0.041666666666666664,
42
+ "grad_norm": 44336.78125,
43
+ "learning_rate": 2.0000000000000003e-06,
44
+ "loss": 0.6824,
45
+ "step": 5
46
+ },
47
+ {
48
+ "epoch": 0.05,
49
+ "grad_norm": 46192.3984375,
50
+ "learning_rate": 2.5e-06,
51
+ "loss": 0.6799,
52
+ "step": 6
53
+ },
54
+ {
55
+ "epoch": 0.058333333333333334,
56
+ "grad_norm": 43591.48828125,
57
+ "learning_rate": 3e-06,
58
+ "loss": 0.6762,
59
+ "step": 7
60
+ },
61
+ {
62
+ "epoch": 0.06666666666666667,
63
+ "grad_norm": 48831.93359375,
64
+ "learning_rate": 3.5000000000000004e-06,
65
+ "loss": 0.6767,
66
+ "step": 8
67
+ },
68
+ {
69
+ "epoch": 0.075,
70
+ "grad_norm": 119280.2734375,
71
+ "learning_rate": 4.000000000000001e-06,
72
+ "loss": 0.673,
73
+ "step": 9
74
+ },
75
+ {
76
+ "epoch": 0.08333333333333333,
77
+ "grad_norm": 46532.3671875,
78
+ "learning_rate": 4.5e-06,
79
+ "loss": 0.6693,
80
+ "step": 10
81
+ },
82
+ {
83
+ "epoch": 0.09166666666666666,
84
+ "grad_norm": 44995.81640625,
85
+ "learning_rate": 5e-06,
86
+ "loss": 0.6663,
87
+ "step": 11
88
+ },
89
+ {
90
+ "epoch": 0.1,
91
+ "grad_norm": 45732.64453125,
92
+ "learning_rate": 5.500000000000001e-06,
93
+ "loss": 0.663,
94
+ "step": 12
95
+ },
96
+ {
97
+ "epoch": 0.10833333333333334,
98
+ "grad_norm": 45771.0234375,
99
+ "learning_rate": 6e-06,
100
+ "loss": 0.6549,
101
+ "step": 13
102
+ },
103
+ {
104
+ "epoch": 0.11666666666666667,
105
+ "grad_norm": 45296.47265625,
106
+ "learning_rate": 6.5000000000000004e-06,
107
+ "loss": 0.6525,
108
+ "step": 14
109
+ },
110
+ {
111
+ "epoch": 0.125,
112
+ "grad_norm": 52268.5390625,
113
+ "learning_rate": 7.000000000000001e-06,
114
+ "loss": 0.6485,
115
+ "step": 15
116
+ },
117
+ {
118
+ "epoch": 0.13333333333333333,
119
+ "grad_norm": 50110.92578125,
120
+ "learning_rate": 7.5e-06,
121
+ "loss": 0.6391,
122
+ "step": 16
123
+ },
124
+ {
125
+ "epoch": 0.14166666666666666,
126
+ "grad_norm": 51549.265625,
127
+ "learning_rate": 8.000000000000001e-06,
128
+ "loss": 0.6349,
129
+ "step": 17
130
+ },
131
+ {
132
+ "epoch": 0.15,
133
+ "grad_norm": 106632.3203125,
134
+ "learning_rate": 8.500000000000002e-06,
135
+ "loss": 0.6323,
136
+ "step": 18
137
+ },
138
+ {
139
+ "epoch": 0.15833333333333333,
140
+ "grad_norm": 57003.57421875,
141
+ "learning_rate": 9e-06,
142
+ "loss": 0.6226,
143
+ "step": 19
144
+ },
145
+ {
146
+ "epoch": 0.16666666666666666,
147
+ "grad_norm": 59664.578125,
148
+ "learning_rate": 9.5e-06,
149
+ "loss": 0.613,
150
+ "step": 20
151
+ },
152
+ {
153
+ "epoch": 0.175,
154
+ "grad_norm": 76325.9140625,
155
+ "learning_rate": 1e-05,
156
+ "loss": 0.5995,
157
+ "step": 21
158
+ },
159
+ {
160
+ "epoch": 0.18333333333333332,
161
+ "grad_norm": 66561.0234375,
162
+ "learning_rate": 1.05e-05,
163
+ "loss": 0.5864,
164
+ "step": 22
165
+ },
166
+ {
167
+ "epoch": 0.19166666666666668,
168
+ "grad_norm": 52762.265625,
169
+ "learning_rate": 1.1000000000000001e-05,
170
+ "loss": 0.5768,
171
+ "step": 23
172
+ },
173
+ {
174
+ "epoch": 0.2,
175
+ "grad_norm": 50716.7890625,
176
+ "learning_rate": 1.1500000000000002e-05,
177
+ "loss": 0.5654,
178
+ "step": 24
179
+ },
180
+ {
181
+ "epoch": 0.20833333333333334,
182
+ "grad_norm": 53293.3046875,
183
+ "learning_rate": 1.2e-05,
184
+ "loss": 0.5525,
185
+ "step": 25
186
+ },
187
+ {
188
+ "epoch": 0.21666666666666667,
189
+ "grad_norm": 54021.37109375,
190
+ "learning_rate": 1.25e-05,
191
+ "loss": 0.5529,
192
+ "step": 26
193
+ },
194
+ {
195
+ "epoch": 0.225,
196
+ "grad_norm": 54463.3515625,
197
+ "learning_rate": 1.3000000000000001e-05,
198
+ "loss": 0.5384,
199
+ "step": 27
200
+ },
201
+ {
202
+ "epoch": 0.23333333333333334,
203
+ "grad_norm": 77561.1484375,
204
+ "learning_rate": 1.3500000000000001e-05,
205
+ "loss": 0.5316,
206
+ "step": 28
207
+ },
208
+ {
209
+ "epoch": 0.24166666666666667,
210
+ "grad_norm": 52105.3515625,
211
+ "learning_rate": 1.4000000000000001e-05,
212
+ "loss": 0.5191,
213
+ "step": 29
214
+ },
215
+ {
216
+ "epoch": 0.25,
217
+ "grad_norm": 46957.71484375,
218
+ "learning_rate": 1.45e-05,
219
+ "loss": 0.507,
220
+ "step": 30
221
+ },
222
+ {
223
+ "epoch": 0.25833333333333336,
224
+ "grad_norm": 46143.05078125,
225
+ "learning_rate": 1.5e-05,
226
+ "loss": 0.4961,
227
+ "step": 31
228
+ },
229
+ {
230
+ "epoch": 0.26666666666666666,
231
+ "grad_norm": 44479.36328125,
232
+ "learning_rate": 1.55e-05,
233
+ "loss": 0.4846,
234
+ "step": 32
235
+ },
236
+ {
237
+ "epoch": 0.275,
238
+ "grad_norm": 49707.58203125,
239
+ "learning_rate": 1.6000000000000003e-05,
240
+ "loss": 0.4793,
241
+ "step": 33
242
+ },
243
+ {
244
+ "epoch": 0.2833333333333333,
245
+ "grad_norm": 46631.30078125,
246
+ "learning_rate": 1.65e-05,
247
+ "loss": 0.4686,
248
+ "step": 34
249
+ },
250
+ {
251
+ "epoch": 0.2916666666666667,
252
+ "grad_norm": 43869.328125,
253
+ "learning_rate": 1.7000000000000003e-05,
254
+ "loss": 0.465,
255
+ "step": 35
256
+ },
257
+ {
258
+ "epoch": 0.3,
259
+ "grad_norm": 42714.52734375,
260
+ "learning_rate": 1.75e-05,
261
+ "loss": 0.4569,
262
+ "step": 36
263
+ },
264
+ {
265
+ "epoch": 0.30833333333333335,
266
+ "grad_norm": 42391.1953125,
267
+ "learning_rate": 1.8e-05,
268
+ "loss": 0.4444,
269
+ "step": 37
270
+ },
271
+ {
272
+ "epoch": 0.31666666666666665,
273
+ "grad_norm": 42775.91796875,
274
+ "learning_rate": 1.85e-05,
275
+ "loss": 0.4346,
276
+ "step": 38
277
+ },
278
+ {
279
+ "epoch": 0.325,
280
+ "grad_norm": 45365.82421875,
281
+ "learning_rate": 1.9e-05,
282
+ "loss": 0.4334,
283
+ "step": 39
284
+ },
285
+ {
286
+ "epoch": 0.3333333333333333,
287
+ "grad_norm": 43367.6875,
288
+ "learning_rate": 1.9500000000000003e-05,
289
+ "loss": 0.4211,
290
+ "step": 40
291
+ },
292
+ {
293
+ "epoch": 0.3416666666666667,
294
+ "grad_norm": 40098.22265625,
295
+ "learning_rate": 2e-05,
296
+ "loss": 0.4069,
297
+ "step": 41
298
+ },
299
+ {
300
+ "epoch": 0.35,
301
+ "grad_norm": 39089.37109375,
302
+ "learning_rate": 2.05e-05,
303
+ "loss": 0.4048,
304
+ "step": 42
305
+ },
306
+ {
307
+ "epoch": 0.35833333333333334,
308
+ "grad_norm": 39849.1328125,
309
+ "learning_rate": 2.1e-05,
310
+ "loss": 0.3931,
311
+ "step": 43
312
+ },
313
+ {
314
+ "epoch": 0.36666666666666664,
315
+ "grad_norm": 39146.015625,
316
+ "learning_rate": 2.15e-05,
317
+ "loss": 0.3877,
318
+ "step": 44
319
+ },
320
+ {
321
+ "epoch": 0.375,
322
+ "grad_norm": 38992.43359375,
323
+ "learning_rate": 2.2000000000000003e-05,
324
+ "loss": 0.3791,
325
+ "step": 45
326
+ },
327
+ {
328
+ "epoch": 0.38333333333333336,
329
+ "grad_norm": 37866.18359375,
330
+ "learning_rate": 2.25e-05,
331
+ "loss": 0.3726,
332
+ "step": 46
333
+ },
334
+ {
335
+ "epoch": 0.39166666666666666,
336
+ "grad_norm": 37756.671875,
337
+ "learning_rate": 2.3000000000000003e-05,
338
+ "loss": 0.364,
339
+ "step": 47
340
+ },
341
+ {
342
+ "epoch": 0.4,
343
+ "grad_norm": 37182.94921875,
344
+ "learning_rate": 2.35e-05,
345
+ "loss": 0.3605,
346
+ "step": 48
347
+ },
348
+ {
349
+ "epoch": 0.4083333333333333,
350
+ "grad_norm": 36981.40234375,
351
+ "learning_rate": 2.4e-05,
352
+ "loss": 0.3522,
353
+ "step": 49
354
+ },
355
+ {
356
+ "epoch": 0.4166666666666667,
357
+ "grad_norm": 36842.3984375,
358
+ "learning_rate": 2.45e-05,
359
+ "loss": 0.3421,
360
+ "step": 50
361
+ },
362
+ {
363
+ "epoch": 0.425,
364
+ "grad_norm": 35325.8828125,
365
+ "learning_rate": 2.5e-05,
366
+ "loss": 0.3442,
367
+ "step": 51
368
+ },
369
+ {
370
+ "epoch": 0.43333333333333335,
371
+ "grad_norm": 35363.078125,
372
+ "learning_rate": 2.5500000000000003e-05,
373
+ "loss": 0.3318,
374
+ "step": 52
375
+ },
376
+ {
377
+ "epoch": 0.44166666666666665,
378
+ "grad_norm": 35645.59765625,
379
+ "learning_rate": 2.6000000000000002e-05,
380
+ "loss": 0.3277,
381
+ "step": 53
382
+ },
383
+ {
384
+ "epoch": 0.45,
385
+ "grad_norm": 34691.13671875,
386
+ "learning_rate": 2.6500000000000004e-05,
387
+ "loss": 0.3171,
388
+ "step": 54
389
+ },
390
+ {
391
+ "epoch": 0.4583333333333333,
392
+ "grad_norm": 34211.484375,
393
+ "learning_rate": 2.7000000000000002e-05,
394
+ "loss": 0.3084,
395
+ "step": 55
396
+ },
397
+ {
398
+ "epoch": 0.4666666666666667,
399
+ "grad_norm": 33361.140625,
400
+ "learning_rate": 2.7500000000000004e-05,
401
+ "loss": 0.3076,
402
+ "step": 56
403
+ },
404
+ {
405
+ "epoch": 0.475,
406
+ "grad_norm": 32967.97265625,
407
+ "learning_rate": 2.8000000000000003e-05,
408
+ "loss": 0.298,
409
+ "step": 57
410
+ },
411
+ {
412
+ "epoch": 0.48333333333333334,
413
+ "grad_norm": 32798.56640625,
414
+ "learning_rate": 2.8499999999999998e-05,
415
+ "loss": 0.2929,
416
+ "step": 58
417
+ },
418
+ {
419
+ "epoch": 0.49166666666666664,
420
+ "grad_norm": 31929.83203125,
421
+ "learning_rate": 2.9e-05,
422
+ "loss": 0.2907,
423
+ "step": 59
424
+ },
425
+ {
426
+ "epoch": 0.5,
427
+ "grad_norm": 31776.216796875,
428
+ "learning_rate": 2.95e-05,
429
+ "loss": 0.2797,
430
+ "step": 60
431
+ },
432
+ {
433
+ "epoch": 0.5083333333333333,
434
+ "grad_norm": 31287.02734375,
435
+ "learning_rate": 3e-05,
436
+ "loss": 0.2724,
437
+ "step": 61
438
+ },
439
+ {
440
+ "epoch": 0.5166666666666667,
441
+ "grad_norm": 30368.2421875,
442
+ "learning_rate": 3.05e-05,
443
+ "loss": 0.2658,
444
+ "step": 62
445
+ },
446
+ {
447
+ "epoch": 0.525,
448
+ "grad_norm": 29610.740234375,
449
+ "learning_rate": 3.1e-05,
450
+ "loss": 0.2606,
451
+ "step": 63
452
+ },
453
+ {
454
+ "epoch": 0.5333333333333333,
455
+ "grad_norm": 29145.74609375,
456
+ "learning_rate": 3.15e-05,
457
+ "loss": 0.2556,
458
+ "step": 64
459
+ },
460
+ {
461
+ "epoch": 0.5416666666666666,
462
+ "grad_norm": 28216.568359375,
463
+ "learning_rate": 3.2000000000000005e-05,
464
+ "loss": 0.253,
465
+ "step": 65
466
+ },
467
+ {
468
+ "epoch": 0.55,
469
+ "grad_norm": 28172.978515625,
470
+ "learning_rate": 3.2500000000000004e-05,
471
+ "loss": 0.2418,
472
+ "step": 66
473
+ },
474
+ {
475
+ "epoch": 0.5583333333333333,
476
+ "grad_norm": 26664.701171875,
477
+ "learning_rate": 3.3e-05,
478
+ "loss": 0.2427,
479
+ "step": 67
480
+ },
481
+ {
482
+ "epoch": 0.5666666666666667,
483
+ "grad_norm": 26883.462890625,
484
+ "learning_rate": 3.35e-05,
485
+ "loss": 0.2346,
486
+ "step": 68
487
+ },
488
+ {
489
+ "epoch": 0.575,
490
+ "grad_norm": 26251.072265625,
491
+ "learning_rate": 3.4000000000000007e-05,
492
+ "loss": 0.2292,
493
+ "step": 69
494
+ },
495
+ {
496
+ "epoch": 0.5833333333333334,
497
+ "grad_norm": 25158.37109375,
498
+ "learning_rate": 3.45e-05,
499
+ "loss": 0.2277,
500
+ "step": 70
501
+ },
502
+ {
503
+ "epoch": 0.5916666666666667,
504
+ "grad_norm": 25251.408203125,
505
+ "learning_rate": 3.5e-05,
506
+ "loss": 0.2119,
507
+ "step": 71
508
+ },
509
+ {
510
+ "epoch": 0.6,
511
+ "grad_norm": 55416.34765625,
512
+ "learning_rate": 3.55e-05,
513
+ "loss": 0.2059,
514
+ "step": 72
515
+ },
516
+ {
517
+ "epoch": 0.6083333333333333,
518
+ "grad_norm": 23741.2421875,
519
+ "learning_rate": 3.6e-05,
520
+ "loss": 0.2014,
521
+ "step": 73
522
+ },
523
+ {
524
+ "epoch": 0.6166666666666667,
525
+ "grad_norm": 22732.470703125,
526
+ "learning_rate": 3.65e-05,
527
+ "loss": 0.2058,
528
+ "step": 74
529
+ },
530
+ {
531
+ "epoch": 0.625,
532
+ "grad_norm": 22756.88671875,
533
+ "learning_rate": 3.7e-05,
534
+ "loss": 0.1922,
535
+ "step": 75
536
+ },
537
+ {
538
+ "epoch": 0.6333333333333333,
539
+ "grad_norm": 22566.482421875,
540
+ "learning_rate": 3.7500000000000003e-05,
541
+ "loss": 0.1892,
542
+ "step": 76
543
+ },
544
+ {
545
+ "epoch": 0.6416666666666667,
546
+ "grad_norm": 21044.404296875,
547
+ "learning_rate": 3.8e-05,
548
+ "loss": 0.1895,
549
+ "step": 77
550
+ },
551
+ {
552
+ "epoch": 0.65,
553
+ "grad_norm": 19801.0390625,
554
+ "learning_rate": 3.85e-05,
555
+ "loss": 0.2051,
556
+ "step": 78
557
+ },
558
+ {
559
+ "epoch": 0.6583333333333333,
560
+ "grad_norm": 20280.55078125,
561
+ "learning_rate": 3.9000000000000006e-05,
562
+ "loss": 0.1739,
563
+ "step": 79
564
+ },
565
+ {
566
+ "epoch": 0.6666666666666666,
567
+ "grad_norm": 21561.10546875,
568
+ "learning_rate": 3.9500000000000005e-05,
569
+ "loss": 0.1748,
570
+ "step": 80
571
+ },
572
+ {
573
+ "epoch": 0.675,
574
+ "grad_norm": 18942.365234375,
575
+ "learning_rate": 4e-05,
576
+ "loss": 0.1607,
577
+ "step": 81
578
+ },
579
+ {
580
+ "epoch": 0.6833333333333333,
581
+ "grad_norm": 18100.796875,
582
+ "learning_rate": 4.05e-05,
583
+ "loss": 0.1581,
584
+ "step": 82
585
+ },
586
+ {
587
+ "epoch": 0.6916666666666667,
588
+ "grad_norm": 17815.296875,
589
+ "learning_rate": 4.1e-05,
590
+ "loss": 0.1603,
591
+ "step": 83
592
+ },
593
+ {
594
+ "epoch": 0.7,
595
+ "grad_norm": 17038.046875,
596
+ "learning_rate": 4.15e-05,
597
+ "loss": 0.1543,
598
+ "step": 84
599
+ },
600
+ {
601
+ "epoch": 0.7083333333333334,
602
+ "grad_norm": 16171.5625,
603
+ "learning_rate": 4.2e-05,
604
+ "loss": 0.1622,
605
+ "step": 85
606
+ },
607
+ {
608
+ "epoch": 0.7166666666666667,
609
+ "grad_norm": 15270.400390625,
610
+ "learning_rate": 4.25e-05,
611
+ "loss": 0.1621,
612
+ "step": 86
613
+ },
614
+ {
615
+ "epoch": 0.725,
616
+ "grad_norm": 15798.998046875,
617
+ "learning_rate": 4.3e-05,
618
+ "loss": 0.147,
619
+ "step": 87
620
+ },
621
+ {
622
+ "epoch": 0.7333333333333333,
623
+ "grad_norm": 14164.265625,
624
+ "learning_rate": 4.35e-05,
625
+ "loss": 0.1485,
626
+ "step": 88
627
+ },
628
+ {
629
+ "epoch": 0.7416666666666667,
630
+ "grad_norm": 15282.1884765625,
631
+ "learning_rate": 4.4000000000000006e-05,
632
+ "loss": 0.1455,
633
+ "step": 89
634
+ },
635
+ {
636
+ "epoch": 0.75,
637
+ "grad_norm": 14389.99609375,
638
+ "learning_rate": 4.4500000000000004e-05,
639
+ "loss": 0.1358,
640
+ "step": 90
641
+ },
642
+ {
643
+ "epoch": 0.7583333333333333,
644
+ "grad_norm": 13489.986328125,
645
+ "learning_rate": 4.5e-05,
646
+ "loss": 0.1308,
647
+ "step": 91
648
+ },
649
+ {
650
+ "epoch": 0.7666666666666667,
651
+ "grad_norm": 12742.55078125,
652
+ "learning_rate": 4.55e-05,
653
+ "loss": 0.1409,
654
+ "step": 92
655
+ },
656
+ {
657
+ "epoch": 0.775,
658
+ "grad_norm": 11714.9228515625,
659
+ "learning_rate": 4.600000000000001e-05,
660
+ "loss": 0.1316,
661
+ "step": 93
662
+ },
663
+ {
664
+ "epoch": 0.7833333333333333,
665
+ "grad_norm": 11921.5185546875,
666
+ "learning_rate": 4.6500000000000005e-05,
667
+ "loss": 0.1226,
668
+ "step": 94
669
+ },
670
+ {
671
+ "epoch": 0.7916666666666666,
672
+ "grad_norm": 11369.017578125,
673
+ "learning_rate": 4.7e-05,
674
+ "loss": 0.1302,
675
+ "step": 95
676
+ },
677
+ {
678
+ "epoch": 0.8,
679
+ "grad_norm": 10840.740234375,
680
+ "learning_rate": 4.75e-05,
681
+ "loss": 0.1276,
682
+ "step": 96
683
+ },
684
+ {
685
+ "epoch": 0.8083333333333333,
686
+ "grad_norm": 10310.0458984375,
687
+ "learning_rate": 4.8e-05,
688
+ "loss": 0.1496,
689
+ "step": 97
690
+ },
691
+ {
692
+ "epoch": 0.8166666666666667,
693
+ "grad_norm": 10777.8271484375,
694
+ "learning_rate": 4.85e-05,
695
+ "loss": 0.1249,
696
+ "step": 98
697
+ },
698
+ {
699
+ "epoch": 0.825,
700
+ "grad_norm": 10753.0048828125,
701
+ "learning_rate": 4.9e-05,
702
+ "loss": 0.1336,
703
+ "step": 99
704
+ },
705
+ {
706
+ "epoch": 0.8333333333333334,
707
+ "grad_norm": 9117.2509765625,
708
+ "learning_rate": 4.9500000000000004e-05,
709
+ "loss": 0.121,
710
+ "step": 100
711
+ },
712
+ {
713
+ "epoch": 0.8416666666666667,
714
+ "grad_norm": 10000.8408203125,
715
+ "learning_rate": 5e-05,
716
+ "loss": 0.1071,
717
+ "step": 101
718
+ },
719
+ {
720
+ "epoch": 0.85,
721
+ "grad_norm": 9281.32421875,
722
+ "learning_rate": 4.995454545454546e-05,
723
+ "loss": 0.1183,
724
+ "step": 102
725
+ },
726
+ {
727
+ "epoch": 0.8583333333333333,
728
+ "grad_norm": 9341.63671875,
729
+ "learning_rate": 4.990909090909091e-05,
730
+ "loss": 0.1177,
731
+ "step": 103
732
+ },
733
+ {
734
+ "epoch": 0.8666666666666667,
735
+ "grad_norm": 8904.6572265625,
736
+ "learning_rate": 4.986363636363637e-05,
737
+ "loss": 0.1116,
738
+ "step": 104
739
+ },
740
+ {
741
+ "epoch": 0.875,
742
+ "grad_norm": 8144.94091796875,
743
+ "learning_rate": 4.981818181818182e-05,
744
+ "loss": 0.1083,
745
+ "step": 105
746
+ },
747
+ {
748
+ "epoch": 0.8833333333333333,
749
+ "grad_norm": 8514.19140625,
750
+ "learning_rate": 4.9772727272727275e-05,
751
+ "loss": 0.1208,
752
+ "step": 106
753
+ },
754
+ {
755
+ "epoch": 0.8916666666666667,
756
+ "grad_norm": 9342.4384765625,
757
+ "learning_rate": 4.9727272727272725e-05,
758
+ "loss": 0.108,
759
+ "step": 107
760
+ },
761
+ {
762
+ "epoch": 0.9,
763
+ "grad_norm": 7503.6064453125,
764
+ "learning_rate": 4.968181818181818e-05,
765
+ "loss": 0.1121,
766
+ "step": 108
767
+ },
768
+ {
769
+ "epoch": 0.9083333333333333,
770
+ "grad_norm": 7928.822265625,
771
+ "learning_rate": 4.963636363636364e-05,
772
+ "loss": 0.1123,
773
+ "step": 109
774
+ },
775
+ {
776
+ "epoch": 0.9166666666666666,
777
+ "grad_norm": 7986.73388671875,
778
+ "learning_rate": 4.9590909090909096e-05,
779
+ "loss": 0.1028,
780
+ "step": 110
781
+ },
782
+ {
783
+ "epoch": 0.925,
784
+ "grad_norm": 7877.8583984375,
785
+ "learning_rate": 4.9545454545454553e-05,
786
+ "loss": 0.1077,
787
+ "step": 111
788
+ },
789
+ {
790
+ "epoch": 0.9333333333333333,
791
+ "grad_norm": 8335.400390625,
792
+ "learning_rate": 4.9500000000000004e-05,
793
+ "loss": 0.1126,
794
+ "step": 112
795
+ },
796
+ {
797
+ "epoch": 0.9416666666666667,
798
+ "grad_norm": 7083.4833984375,
799
+ "learning_rate": 4.945454545454546e-05,
800
+ "loss": 0.1098,
801
+ "step": 113
802
+ },
803
+ {
804
+ "epoch": 0.95,
805
+ "grad_norm": 9419.419921875,
806
+ "learning_rate": 4.940909090909091e-05,
807
+ "loss": 0.1058,
808
+ "step": 114
809
+ },
810
+ {
811
+ "epoch": 0.9583333333333334,
812
+ "grad_norm": 7879.18994140625,
813
+ "learning_rate": 4.936363636363637e-05,
814
+ "loss": 0.1081,
815
+ "step": 115
816
+ },
817
+ {
818
+ "epoch": 0.9666666666666667,
819
+ "grad_norm": 8113.90283203125,
820
+ "learning_rate": 4.931818181818182e-05,
821
+ "loss": 0.1075,
822
+ "step": 116
823
+ },
824
+ {
825
+ "epoch": 0.975,
826
+ "grad_norm": 7590.12744140625,
827
+ "learning_rate": 4.9272727272727276e-05,
828
+ "loss": 0.1041,
829
+ "step": 117
830
+ },
831
+ {
832
+ "epoch": 0.9833333333333333,
833
+ "grad_norm": 7461.189453125,
834
+ "learning_rate": 4.9227272727272726e-05,
835
+ "loss": 0.1064,
836
+ "step": 118
837
+ },
838
+ {
839
+ "epoch": 0.9916666666666667,
840
+ "grad_norm": 7857.06298828125,
841
+ "learning_rate": 4.9181818181818183e-05,
842
+ "loss": 0.1102,
843
+ "step": 119
844
+ },
845
+ {
846
+ "epoch": 1.0,
847
+ "grad_norm": 34157.33203125,
848
+ "learning_rate": 4.913636363636364e-05,
849
+ "loss": 0.1032,
850
+ "step": 120
851
+ },
852
+ {
853
+ "epoch": 1.0,
854
+ "eval_f1_macro": 0.0,
855
+ "eval_f1_micro": 0.0,
856
+ "eval_loss": 0.10901560634374619,
857
+ "eval_precision": 0.0,
858
+ "eval_recall": 0.0,
859
+ "eval_runtime": 0.8962,
860
+ "eval_samples_per_second": 354.843,
861
+ "eval_steps_per_second": 15.622,
862
+ "step": 120
863
+ },
864
+ {
865
+ "epoch": 1.0083333333333333,
866
+ "grad_norm": 7351.72998046875,
867
+ "learning_rate": 4.909090909090909e-05,
868
+ "loss": 0.1147,
869
+ "step": 121
870
+ },
871
+ {
872
+ "epoch": 1.0166666666666666,
873
+ "grad_norm": 8307.1962890625,
874
+ "learning_rate": 4.904545454545455e-05,
875
+ "loss": 0.1115,
876
+ "step": 122
877
+ },
878
+ {
879
+ "epoch": 1.025,
880
+ "grad_norm": 8783.328125,
881
+ "learning_rate": 4.9e-05,
882
+ "loss": 0.0979,
883
+ "step": 123
884
+ },
885
+ {
886
+ "epoch": 1.0333333333333334,
887
+ "grad_norm": 7471.53857421875,
888
+ "learning_rate": 4.8954545454545456e-05,
889
+ "loss": 0.1015,
890
+ "step": 124
891
+ },
892
+ {
893
+ "epoch": 1.0416666666666667,
894
+ "grad_norm": 8831.017578125,
895
+ "learning_rate": 4.890909090909091e-05,
896
+ "loss": 0.1177,
897
+ "step": 125
898
+ },
899
+ {
900
+ "epoch": 1.05,
901
+ "grad_norm": 8687.0224609375,
902
+ "learning_rate": 4.886363636363637e-05,
903
+ "loss": 0.1101,
904
+ "step": 126
905
+ },
906
+ {
907
+ "epoch": 1.0583333333333333,
908
+ "grad_norm": 7138.138671875,
909
+ "learning_rate": 4.881818181818182e-05,
910
+ "loss": 0.111,
911
+ "step": 127
912
+ },
913
+ {
914
+ "epoch": 1.0666666666666667,
915
+ "grad_norm": 7717.380859375,
916
+ "learning_rate": 4.877272727272728e-05,
917
+ "loss": 0.0995,
918
+ "step": 128
919
+ },
920
+ {
921
+ "epoch": 1.075,
922
+ "grad_norm": 6875.6474609375,
923
+ "learning_rate": 4.872727272727273e-05,
924
+ "loss": 0.1094,
925
+ "step": 129
926
+ },
927
+ {
928
+ "epoch": 1.0833333333333333,
929
+ "grad_norm": 8128.5595703125,
930
+ "learning_rate": 4.8681818181818185e-05,
931
+ "loss": 0.116,
932
+ "step": 130
933
+ },
934
+ {
935
+ "epoch": 1.0916666666666666,
936
+ "grad_norm": 8500.302734375,
937
+ "learning_rate": 4.863636363636364e-05,
938
+ "loss": 0.1195,
939
+ "step": 131
940
+ },
941
+ {
942
+ "epoch": 1.1,
943
+ "grad_norm": 6833.93701171875,
944
+ "learning_rate": 4.859090909090909e-05,
945
+ "loss": 0.1136,
946
+ "step": 132
947
+ },
948
+ {
949
+ "epoch": 1.1083333333333334,
950
+ "grad_norm": 8277.92578125,
951
+ "learning_rate": 4.854545454545455e-05,
952
+ "loss": 0.113,
953
+ "step": 133
954
+ },
955
+ {
956
+ "epoch": 1.1166666666666667,
957
+ "grad_norm": 6291.90380859375,
958
+ "learning_rate": 4.85e-05,
959
+ "loss": 0.0971,
960
+ "step": 134
961
+ },
962
+ {
963
+ "epoch": 1.125,
964
+ "grad_norm": 7217.54345703125,
965
+ "learning_rate": 4.845454545454546e-05,
966
+ "loss": 0.1095,
967
+ "step": 135
968
+ },
969
+ {
970
+ "epoch": 1.1333333333333333,
971
+ "grad_norm": 9153.27734375,
972
+ "learning_rate": 4.840909090909091e-05,
973
+ "loss": 0.1238,
974
+ "step": 136
975
+ },
976
+ {
977
+ "epoch": 1.1416666666666666,
978
+ "grad_norm": 7221.31884765625,
979
+ "learning_rate": 4.8363636363636364e-05,
980
+ "loss": 0.1049,
981
+ "step": 137
982
+ },
983
+ {
984
+ "epoch": 1.15,
985
+ "grad_norm": 7583.30126953125,
986
+ "learning_rate": 4.8318181818181815e-05,
987
+ "loss": 0.0962,
988
+ "step": 138
989
+ },
990
+ {
991
+ "epoch": 1.1583333333333332,
992
+ "grad_norm": 7487.25927734375,
993
+ "learning_rate": 4.827272727272727e-05,
994
+ "loss": 0.1068,
995
+ "step": 139
996
+ },
997
+ {
998
+ "epoch": 1.1666666666666667,
999
+ "grad_norm": 8195.296875,
1000
+ "learning_rate": 4.822727272727273e-05,
1001
+ "loss": 0.1065,
1002
+ "step": 140
1003
+ },
1004
+ {
1005
+ "epoch": 1.175,
1006
+ "grad_norm": 7319.97119140625,
1007
+ "learning_rate": 4.8181818181818186e-05,
1008
+ "loss": 0.1163,
1009
+ "step": 141
1010
+ },
1011
+ {
1012
+ "epoch": 1.1833333333333333,
1013
+ "grad_norm": 7353.1728515625,
1014
+ "learning_rate": 4.813636363636364e-05,
1015
+ "loss": 0.1033,
1016
+ "step": 142
1017
+ },
1018
+ {
1019
+ "epoch": 1.1916666666666667,
1020
+ "grad_norm": 6305.76025390625,
1021
+ "learning_rate": 4.8090909090909094e-05,
1022
+ "loss": 0.107,
1023
+ "step": 143
1024
+ },
1025
+ {
1026
+ "epoch": 1.2,
1027
+ "grad_norm": 7193.05322265625,
1028
+ "learning_rate": 4.804545454545455e-05,
1029
+ "loss": 0.1054,
1030
+ "step": 144
1031
+ },
1032
+ {
1033
+ "epoch": 1.2083333333333333,
1034
+ "grad_norm": 8265.4169921875,
1035
+ "learning_rate": 4.8e-05,
1036
+ "loss": 0.1064,
1037
+ "step": 145
1038
+ },
1039
+ {
1040
+ "epoch": 1.2166666666666668,
1041
+ "grad_norm": 7570.91357421875,
1042
+ "learning_rate": 4.795454545454546e-05,
1043
+ "loss": 0.0951,
1044
+ "step": 146
1045
+ },
1046
+ {
1047
+ "epoch": 1.225,
1048
+ "grad_norm": 7928.24755859375,
1049
+ "learning_rate": 4.790909090909091e-05,
1050
+ "loss": 0.1174,
1051
+ "step": 147
1052
+ },
1053
+ {
1054
+ "epoch": 1.2333333333333334,
1055
+ "grad_norm": 8941.95703125,
1056
+ "learning_rate": 4.7863636363636366e-05,
1057
+ "loss": 0.138,
1058
+ "step": 148
1059
+ },
1060
+ {
1061
+ "epoch": 1.2416666666666667,
1062
+ "grad_norm": 8456.47265625,
1063
+ "learning_rate": 4.781818181818182e-05,
1064
+ "loss": 0.1046,
1065
+ "step": 149
1066
+ },
1067
+ {
1068
+ "epoch": 1.25,
1069
+ "grad_norm": 7923.06982421875,
1070
+ "learning_rate": 4.777272727272727e-05,
1071
+ "loss": 0.1076,
1072
+ "step": 150
1073
+ },
1074
+ {
1075
+ "epoch": 1.2583333333333333,
1076
+ "grad_norm": 8985.126953125,
1077
+ "learning_rate": 4.772727272727273e-05,
1078
+ "loss": 0.1035,
1079
+ "step": 151
1080
+ },
1081
+ {
1082
+ "epoch": 1.2666666666666666,
1083
+ "grad_norm": 7761.240234375,
1084
+ "learning_rate": 4.768181818181818e-05,
1085
+ "loss": 0.1027,
1086
+ "step": 152
1087
+ },
1088
+ {
1089
+ "epoch": 1.275,
1090
+ "grad_norm": 8492.732421875,
1091
+ "learning_rate": 4.763636363636364e-05,
1092
+ "loss": 0.1081,
1093
+ "step": 153
1094
+ },
1095
+ {
1096
+ "epoch": 1.2833333333333332,
1097
+ "grad_norm": 7053.7744140625,
1098
+ "learning_rate": 4.759090909090909e-05,
1099
+ "loss": 0.0934,
1100
+ "step": 154
1101
+ },
1102
+ {
1103
+ "epoch": 1.2916666666666667,
1104
+ "grad_norm": 7600.5458984375,
1105
+ "learning_rate": 4.7545454545454545e-05,
1106
+ "loss": 0.0971,
1107
+ "step": 155
1108
+ },
1109
+ {
1110
+ "epoch": 1.3,
1111
+ "grad_norm": 6331.5673828125,
1112
+ "learning_rate": 4.75e-05,
1113
+ "loss": 0.1149,
1114
+ "step": 156
1115
+ },
1116
+ {
1117
+ "epoch": 1.3083333333333333,
1118
+ "grad_norm": 8746.0419921875,
1119
+ "learning_rate": 4.745454545454546e-05,
1120
+ "loss": 0.1057,
1121
+ "step": 157
1122
+ },
1123
+ {
1124
+ "epoch": 1.3166666666666667,
1125
+ "grad_norm": 7221.12646484375,
1126
+ "learning_rate": 4.740909090909092e-05,
1127
+ "loss": 0.1155,
1128
+ "step": 158
1129
+ },
1130
+ {
1131
+ "epoch": 1.325,
1132
+ "grad_norm": 6219.18408203125,
1133
+ "learning_rate": 4.736363636363637e-05,
1134
+ "loss": 0.1037,
1135
+ "step": 159
1136
+ },
1137
+ {
1138
+ "epoch": 1.3333333333333333,
1139
+ "grad_norm": 8318.43359375,
1140
+ "learning_rate": 4.7318181818181824e-05,
1141
+ "loss": 0.093,
1142
+ "step": 160
1143
+ },
1144
+ {
1145
+ "epoch": 1.3416666666666668,
1146
+ "grad_norm": 6739.9990234375,
1147
+ "learning_rate": 4.7272727272727275e-05,
1148
+ "loss": 0.0845,
1149
+ "step": 161
1150
+ },
1151
+ {
1152
+ "epoch": 1.35,
1153
+ "grad_norm": 6488.94482421875,
1154
+ "learning_rate": 4.722727272727273e-05,
1155
+ "loss": 0.0939,
1156
+ "step": 162
1157
+ },
1158
+ {
1159
+ "epoch": 1.3583333333333334,
1160
+ "grad_norm": 7930.939453125,
1161
+ "learning_rate": 4.718181818181818e-05,
1162
+ "loss": 0.1202,
1163
+ "step": 163
1164
+ },
1165
+ {
1166
+ "epoch": 1.3666666666666667,
1167
+ "grad_norm": 8253.9482421875,
1168
+ "learning_rate": 4.713636363636364e-05,
1169
+ "loss": 0.1138,
1170
+ "step": 164
1171
+ },
1172
+ {
1173
+ "epoch": 1.375,
1174
+ "grad_norm": 7846.85009765625,
1175
+ "learning_rate": 4.709090909090909e-05,
1176
+ "loss": 0.116,
1177
+ "step": 165
1178
+ },
1179
+ {
1180
+ "epoch": 1.3833333333333333,
1181
+ "grad_norm": 6417.66357421875,
1182
+ "learning_rate": 4.704545454545455e-05,
1183
+ "loss": 0.1054,
1184
+ "step": 166
1185
+ },
1186
+ {
1187
+ "epoch": 1.3916666666666666,
1188
+ "grad_norm": 6214.9521484375,
1189
+ "learning_rate": 4.7e-05,
1190
+ "loss": 0.1045,
1191
+ "step": 167
1192
+ },
1193
+ {
1194
+ "epoch": 1.4,
1195
+ "grad_norm": 6744.1416015625,
1196
+ "learning_rate": 4.6954545454545454e-05,
1197
+ "loss": 0.1029,
1198
+ "step": 168
1199
+ },
1200
+ {
1201
+ "epoch": 1.4083333333333332,
1202
+ "grad_norm": 8765.5869140625,
1203
+ "learning_rate": 4.690909090909091e-05,
1204
+ "loss": 0.1084,
1205
+ "step": 169
1206
+ },
1207
+ {
1208
+ "epoch": 1.4166666666666667,
1209
+ "grad_norm": 9230.552734375,
1210
+ "learning_rate": 4.686363636363636e-05,
1211
+ "loss": 0.1464,
1212
+ "step": 170
1213
+ },
1214
+ {
1215
+ "epoch": 1.425,
1216
+ "grad_norm": 7388.099609375,
1217
+ "learning_rate": 4.681818181818182e-05,
1218
+ "loss": 0.1047,
1219
+ "step": 171
1220
+ },
1221
+ {
1222
+ "epoch": 1.4333333333333333,
1223
+ "grad_norm": 6448.68408203125,
1224
+ "learning_rate": 4.6772727272727276e-05,
1225
+ "loss": 0.0948,
1226
+ "step": 172
1227
+ },
1228
+ {
1229
+ "epoch": 1.4416666666666667,
1230
+ "grad_norm": 7005.68701171875,
1231
+ "learning_rate": 4.672727272727273e-05,
1232
+ "loss": 0.1133,
1233
+ "step": 173
1234
+ },
1235
+ {
1236
+ "epoch": 1.45,
1237
+ "grad_norm": 8248.9482421875,
1238
+ "learning_rate": 4.6681818181818184e-05,
1239
+ "loss": 0.118,
1240
+ "step": 174
1241
+ },
1242
+ {
1243
+ "epoch": 1.4583333333333333,
1244
+ "grad_norm": 7844.08544921875,
1245
+ "learning_rate": 4.663636363636364e-05,
1246
+ "loss": 0.0957,
1247
+ "step": 175
1248
+ },
1249
+ {
1250
+ "epoch": 1.4666666666666668,
1251
+ "grad_norm": 8574.84765625,
1252
+ "learning_rate": 4.659090909090909e-05,
1253
+ "loss": 0.1243,
1254
+ "step": 176
1255
+ },
1256
+ {
1257
+ "epoch": 1.475,
1258
+ "grad_norm": 8588.7138671875,
1259
+ "learning_rate": 4.654545454545455e-05,
1260
+ "loss": 0.1032,
1261
+ "step": 177
1262
+ },
1263
+ {
1264
+ "epoch": 1.4833333333333334,
1265
+ "grad_norm": 8352.0380859375,
1266
+ "learning_rate": 4.6500000000000005e-05,
1267
+ "loss": 0.1259,
1268
+ "step": 178
1269
+ },
1270
+ {
1271
+ "epoch": 1.4916666666666667,
1272
+ "grad_norm": 7560.94140625,
1273
+ "learning_rate": 4.6454545454545456e-05,
1274
+ "loss": 0.0995,
1275
+ "step": 179
1276
+ },
1277
+ {
1278
+ "epoch": 1.5,
1279
+ "grad_norm": 8052.83984375,
1280
+ "learning_rate": 4.640909090909091e-05,
1281
+ "loss": 0.0995,
1282
+ "step": 180
1283
+ },
1284
+ {
1285
+ "epoch": 1.5083333333333333,
1286
+ "grad_norm": 7812.76953125,
1287
+ "learning_rate": 4.636363636363636e-05,
1288
+ "loss": 0.1075,
1289
+ "step": 181
1290
+ },
1291
+ {
1292
+ "epoch": 1.5166666666666666,
1293
+ "grad_norm": 8199.2724609375,
1294
+ "learning_rate": 4.631818181818182e-05,
1295
+ "loss": 0.1172,
1296
+ "step": 182
1297
+ },
1298
+ {
1299
+ "epoch": 1.525,
1300
+ "grad_norm": 25283.591796875,
1301
+ "learning_rate": 4.627272727272727e-05,
1302
+ "loss": 0.1395,
1303
+ "step": 183
1304
+ },
1305
+ {
1306
+ "epoch": 1.5333333333333332,
1307
+ "grad_norm": 7990.4560546875,
1308
+ "learning_rate": 4.622727272727273e-05,
1309
+ "loss": 0.0965,
1310
+ "step": 184
1311
+ },
1312
+ {
1313
+ "epoch": 1.5416666666666665,
1314
+ "grad_norm": 8850.798828125,
1315
+ "learning_rate": 4.618181818181818e-05,
1316
+ "loss": 0.0971,
1317
+ "step": 185
1318
+ },
1319
+ {
1320
+ "epoch": 1.55,
1321
+ "grad_norm": 7082.34521484375,
1322
+ "learning_rate": 4.6136363636363635e-05,
1323
+ "loss": 0.1097,
1324
+ "step": 186
1325
+ },
1326
+ {
1327
+ "epoch": 1.5583333333333333,
1328
+ "grad_norm": 8550.869140625,
1329
+ "learning_rate": 4.609090909090909e-05,
1330
+ "loss": 0.106,
1331
+ "step": 187
1332
+ },
1333
+ {
1334
+ "epoch": 1.5666666666666667,
1335
+ "grad_norm": 5758.966796875,
1336
+ "learning_rate": 4.604545454545455e-05,
1337
+ "loss": 0.0996,
1338
+ "step": 188
1339
+ },
1340
+ {
1341
+ "epoch": 1.575,
1342
+ "grad_norm": 8699.7333984375,
1343
+ "learning_rate": 4.600000000000001e-05,
1344
+ "loss": 0.0974,
1345
+ "step": 189
1346
+ },
1347
+ {
1348
+ "epoch": 1.5833333333333335,
1349
+ "grad_norm": 5898.751953125,
1350
+ "learning_rate": 4.595454545454546e-05,
1351
+ "loss": 0.097,
1352
+ "step": 190
1353
+ },
1354
+ {
1355
+ "epoch": 1.5916666666666668,
1356
+ "grad_norm": 6824.419921875,
1357
+ "learning_rate": 4.5909090909090914e-05,
1358
+ "loss": 0.1012,
1359
+ "step": 191
1360
+ },
1361
+ {
1362
+ "epoch": 1.6,
1363
+ "grad_norm": 8745.8408203125,
1364
+ "learning_rate": 4.5863636363636365e-05,
1365
+ "loss": 0.1172,
1366
+ "step": 192
1367
+ },
1368
+ {
1369
+ "epoch": 1.6083333333333334,
1370
+ "grad_norm": 7022.79296875,
1371
+ "learning_rate": 4.581818181818182e-05,
1372
+ "loss": 0.0914,
1373
+ "step": 193
1374
+ },
1375
+ {
1376
+ "epoch": 1.6166666666666667,
1377
+ "grad_norm": 8727.6943359375,
1378
+ "learning_rate": 4.577272727272727e-05,
1379
+ "loss": 0.1124,
1380
+ "step": 194
1381
+ },
1382
+ {
1383
+ "epoch": 1.625,
1384
+ "grad_norm": 7952.49609375,
1385
+ "learning_rate": 4.572727272727273e-05,
1386
+ "loss": 0.113,
1387
+ "step": 195
1388
+ },
1389
+ {
1390
+ "epoch": 1.6333333333333333,
1391
+ "grad_norm": 6855.783203125,
1392
+ "learning_rate": 4.5681818181818186e-05,
1393
+ "loss": 0.0944,
1394
+ "step": 196
1395
+ },
1396
+ {
1397
+ "epoch": 1.6416666666666666,
1398
+ "grad_norm": 7499.74609375,
1399
+ "learning_rate": 4.563636363636364e-05,
1400
+ "loss": 0.1151,
1401
+ "step": 197
1402
+ },
1403
+ {
1404
+ "epoch": 1.65,
1405
+ "grad_norm": 6449.2529296875,
1406
+ "learning_rate": 4.5590909090909094e-05,
1407
+ "loss": 0.0972,
1408
+ "step": 198
1409
+ },
1410
+ {
1411
+ "epoch": 1.6583333333333332,
1412
+ "grad_norm": 7116.66748046875,
1413
+ "learning_rate": 4.5545454545454544e-05,
1414
+ "loss": 0.1049,
1415
+ "step": 199
1416
+ },
1417
+ {
1418
+ "epoch": 1.6666666666666665,
1419
+ "grad_norm": 8555.966796875,
1420
+ "learning_rate": 4.55e-05,
1421
+ "loss": 0.1139,
1422
+ "step": 200
1423
+ },
1424
+ {
1425
+ "epoch": 1.675,
1426
+ "grad_norm": 7615.99267578125,
1427
+ "learning_rate": 4.545454545454546e-05,
1428
+ "loss": 0.1016,
1429
+ "step": 201
1430
+ },
1431
+ {
1432
+ "epoch": 1.6833333333333333,
1433
+ "grad_norm": 8245.642578125,
1434
+ "learning_rate": 4.540909090909091e-05,
1435
+ "loss": 0.098,
1436
+ "step": 202
1437
+ },
1438
+ {
1439
+ "epoch": 1.6916666666666667,
1440
+ "grad_norm": 6856.2958984375,
1441
+ "learning_rate": 4.5363636363636366e-05,
1442
+ "loss": 0.1067,
1443
+ "step": 203
1444
+ },
1445
+ {
1446
+ "epoch": 1.7,
1447
+ "grad_norm": 7053.49560546875,
1448
+ "learning_rate": 4.531818181818182e-05,
1449
+ "loss": 0.1097,
1450
+ "step": 204
1451
+ },
1452
+ {
1453
+ "epoch": 1.7083333333333335,
1454
+ "grad_norm": 8732.923828125,
1455
+ "learning_rate": 4.5272727272727274e-05,
1456
+ "loss": 0.098,
1457
+ "step": 205
1458
+ },
1459
+ {
1460
+ "epoch": 1.7166666666666668,
1461
+ "grad_norm": 8398.2392578125,
1462
+ "learning_rate": 4.522727272727273e-05,
1463
+ "loss": 0.1119,
1464
+ "step": 206
1465
+ },
1466
+ {
1467
+ "epoch": 1.725,
1468
+ "grad_norm": 7186.2236328125,
1469
+ "learning_rate": 4.518181818181819e-05,
1470
+ "loss": 0.1104,
1471
+ "step": 207
1472
+ },
1473
+ {
1474
+ "epoch": 1.7333333333333334,
1475
+ "grad_norm": 7283.79345703125,
1476
+ "learning_rate": 4.513636363636364e-05,
1477
+ "loss": 0.1042,
1478
+ "step": 208
1479
+ },
1480
+ {
1481
+ "epoch": 1.7416666666666667,
1482
+ "grad_norm": 8617.919921875,
1483
+ "learning_rate": 4.5090909090909095e-05,
1484
+ "loss": 0.1182,
1485
+ "step": 209
1486
+ },
1487
+ {
1488
+ "epoch": 1.75,
1489
+ "grad_norm": 8389.6796875,
1490
+ "learning_rate": 4.5045454545454546e-05,
1491
+ "loss": 0.1064,
1492
+ "step": 210
1493
+ },
1494
+ {
1495
+ "epoch": 1.7583333333333333,
1496
+ "grad_norm": 7137.05224609375,
1497
+ "learning_rate": 4.5e-05,
1498
+ "loss": 0.1052,
1499
+ "step": 211
1500
+ },
1501
+ {
1502
+ "epoch": 1.7666666666666666,
1503
+ "grad_norm": 5784.734375,
1504
+ "learning_rate": 4.495454545454545e-05,
1505
+ "loss": 0.0978,
1506
+ "step": 212
1507
+ },
1508
+ {
1509
+ "epoch": 1.775,
1510
+ "grad_norm": 6999.89111328125,
1511
+ "learning_rate": 4.490909090909091e-05,
1512
+ "loss": 0.0943,
1513
+ "step": 213
1514
+ },
1515
+ {
1516
+ "epoch": 1.7833333333333332,
1517
+ "grad_norm": 6968.7197265625,
1518
+ "learning_rate": 4.486363636363636e-05,
1519
+ "loss": 0.1101,
1520
+ "step": 214
1521
+ },
1522
+ {
1523
+ "epoch": 1.7916666666666665,
1524
+ "grad_norm": 8037.56884765625,
1525
+ "learning_rate": 4.481818181818182e-05,
1526
+ "loss": 0.0989,
1527
+ "step": 215
1528
+ },
1529
+ {
1530
+ "epoch": 1.8,
1531
+ "grad_norm": 7295.52197265625,
1532
+ "learning_rate": 4.4772727272727275e-05,
1533
+ "loss": 0.115,
1534
+ "step": 216
1535
+ },
1536
+ {
1537
+ "epoch": 1.8083333333333333,
1538
+ "grad_norm": 6018.81298828125,
1539
+ "learning_rate": 4.472727272727273e-05,
1540
+ "loss": 0.093,
1541
+ "step": 217
1542
+ },
1543
+ {
1544
+ "epoch": 1.8166666666666667,
1545
+ "grad_norm": 6783.51318359375,
1546
+ "learning_rate": 4.468181818181818e-05,
1547
+ "loss": 0.1019,
1548
+ "step": 218
1549
+ },
1550
+ {
1551
+ "epoch": 1.825,
1552
+ "grad_norm": 7277.119140625,
1553
+ "learning_rate": 4.463636363636364e-05,
1554
+ "loss": 0.0984,
1555
+ "step": 219
1556
+ },
1557
+ {
1558
+ "epoch": 1.8333333333333335,
1559
+ "grad_norm": 6515.0615234375,
1560
+ "learning_rate": 4.45909090909091e-05,
1561
+ "loss": 0.1009,
1562
+ "step": 220
1563
+ },
1564
+ {
1565
+ "epoch": 1.8416666666666668,
1566
+ "grad_norm": 7060.08056640625,
1567
+ "learning_rate": 4.454545454545455e-05,
1568
+ "loss": 0.1068,
1569
+ "step": 221
1570
+ },
1571
+ {
1572
+ "epoch": 1.85,
1573
+ "grad_norm": 7915.4287109375,
1574
+ "learning_rate": 4.4500000000000004e-05,
1575
+ "loss": 0.109,
1576
+ "step": 222
1577
+ },
1578
+ {
1579
+ "epoch": 1.8583333333333334,
1580
+ "grad_norm": 7376.220703125,
1581
+ "learning_rate": 4.4454545454545455e-05,
1582
+ "loss": 0.093,
1583
+ "step": 223
1584
+ },
1585
+ {
1586
+ "epoch": 1.8666666666666667,
1587
+ "grad_norm": 7639.013671875,
1588
+ "learning_rate": 4.440909090909091e-05,
1589
+ "loss": 0.106,
1590
+ "step": 224
1591
+ },
1592
+ {
1593
+ "epoch": 1.875,
1594
+ "grad_norm": 7439.11474609375,
1595
+ "learning_rate": 4.436363636363637e-05,
1596
+ "loss": 0.1106,
1597
+ "step": 225
1598
+ },
1599
+ {
1600
+ "epoch": 1.8833333333333333,
1601
+ "grad_norm": 8460.6650390625,
1602
+ "learning_rate": 4.431818181818182e-05,
1603
+ "loss": 0.1079,
1604
+ "step": 226
1605
+ },
1606
+ {
1607
+ "epoch": 1.8916666666666666,
1608
+ "grad_norm": 7310.08740234375,
1609
+ "learning_rate": 4.4272727272727276e-05,
1610
+ "loss": 0.0968,
1611
+ "step": 227
1612
+ },
1613
+ {
1614
+ "epoch": 1.9,
1615
+ "grad_norm": 6232.28759765625,
1616
+ "learning_rate": 4.422727272727273e-05,
1617
+ "loss": 0.1067,
1618
+ "step": 228
1619
+ },
1620
+ {
1621
+ "epoch": 1.9083333333333332,
1622
+ "grad_norm": 7846.275390625,
1623
+ "learning_rate": 4.4181818181818184e-05,
1624
+ "loss": 0.1038,
1625
+ "step": 229
1626
+ },
1627
+ {
1628
+ "epoch": 1.9166666666666665,
1629
+ "grad_norm": 6419.7333984375,
1630
+ "learning_rate": 4.4136363636363634e-05,
1631
+ "loss": 0.0999,
1632
+ "step": 230
1633
+ },
1634
+ {
1635
+ "epoch": 1.925,
1636
+ "grad_norm": 7637.369140625,
1637
+ "learning_rate": 4.409090909090909e-05,
1638
+ "loss": 0.0905,
1639
+ "step": 231
1640
+ },
1641
+ {
1642
+ "epoch": 1.9333333333333333,
1643
+ "grad_norm": 6210.56396484375,
1644
+ "learning_rate": 4.404545454545455e-05,
1645
+ "loss": 0.0994,
1646
+ "step": 232
1647
+ },
1648
+ {
1649
+ "epoch": 1.9416666666666667,
1650
+ "grad_norm": 7951.947265625,
1651
+ "learning_rate": 4.4000000000000006e-05,
1652
+ "loss": 0.0981,
1653
+ "step": 233
1654
+ },
1655
+ {
1656
+ "epoch": 1.95,
1657
+ "grad_norm": 6833.03466796875,
1658
+ "learning_rate": 4.3954545454545456e-05,
1659
+ "loss": 0.1007,
1660
+ "step": 234
1661
+ },
1662
+ {
1663
+ "epoch": 1.9583333333333335,
1664
+ "grad_norm": 7926.4765625,
1665
+ "learning_rate": 4.390909090909091e-05,
1666
+ "loss": 0.0925,
1667
+ "step": 235
1668
+ },
1669
+ {
1670
+ "epoch": 1.9666666666666668,
1671
+ "grad_norm": 7166.40087890625,
1672
+ "learning_rate": 4.386363636363637e-05,
1673
+ "loss": 0.1111,
1674
+ "step": 236
1675
+ },
1676
+ {
1677
+ "epoch": 1.975,
1678
+ "grad_norm": 8137.634765625,
1679
+ "learning_rate": 4.381818181818182e-05,
1680
+ "loss": 0.1079,
1681
+ "step": 237
1682
+ },
1683
+ {
1684
+ "epoch": 1.9833333333333334,
1685
+ "grad_norm": 8614.6142578125,
1686
+ "learning_rate": 4.377272727272728e-05,
1687
+ "loss": 0.1006,
1688
+ "step": 238
1689
+ },
1690
+ {
1691
+ "epoch": 1.9916666666666667,
1692
+ "grad_norm": 10941.376953125,
1693
+ "learning_rate": 4.372727272727273e-05,
1694
+ "loss": 0.0937,
1695
+ "step": 239
1696
+ },
1697
+ {
1698
+ "epoch": 2.0,
1699
+ "grad_norm": 34603.39453125,
1700
+ "learning_rate": 4.3681818181818185e-05,
1701
+ "loss": 0.0831,
1702
+ "step": 240
1703
+ },
1704
+ {
1705
+ "epoch": 2.0,
1706
+ "eval_f1_macro": 0.025460502366217694,
1707
+ "eval_f1_micro": 0.24436090225563908,
1708
+ "eval_loss": 0.10331574827432632,
1709
+ "eval_precision": 0.48507462686567165,
1710
+ "eval_recall": 0.16331658291457288,
1711
+ "eval_runtime": 0.8993,
1712
+ "eval_samples_per_second": 353.616,
1713
+ "eval_steps_per_second": 15.568,
1714
+ "step": 240
1715
+ }
1716
+ ],
1717
+ "logging_steps": 1,
1718
+ "max_steps": 1200,
1719
+ "num_input_tokens_seen": 0,
1720
+ "num_train_epochs": 10,
1721
+ "save_steps": 500,
1722
+ "stateful_callbacks": {
1723
+ "TrainerControl": {
1724
+ "args": {
1725
+ "should_epoch_stop": false,
1726
+ "should_evaluate": false,
1727
+ "should_log": false,
1728
+ "should_save": true,
1729
+ "should_training_stop": false
1730
+ },
1731
+ "attributes": {}
1732
+ }
1733
+ },
1734
+ "total_flos": 1504064501723136.0,
1735
+ "train_batch_size": 24,
1736
+ "trial_name": null,
1737
+ "trial_params": null
1738
+ }
checkpoint-240/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:84b38cc83b3a41fa6c531ffc2fce557168d7eea296dc27400ca0cad58c63e8db
3
+ size 5368
checkpoint-240/vocab.txt ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-360/config.json ADDED
@@ -0,0 +1,130 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "BertForSequenceClassification"
4
+ ],
5
+ "attention_probs_dropout_prob": 0.1,
6
+ "classifier_dropout": null,
7
+ "gradient_checkpointing": false,
8
+ "hidden_act": "gelu",
9
+ "hidden_dropout_prob": 0.1,
10
+ "hidden_size": 768,
11
+ "id2label": {
12
+ "0": "T1003.001",
13
+ "1": "T1005",
14
+ "2": "T1012",
15
+ "3": "T1016",
16
+ "4": "T1021.001",
17
+ "5": "T1027",
18
+ "6": "T1033",
19
+ "7": "T1036.005",
20
+ "8": "T1041",
21
+ "9": "T1047",
22
+ "10": "T1053.005",
23
+ "11": "T1055",
24
+ "12": "T1056.001",
25
+ "13": "T1057",
26
+ "14": "T1059.003",
27
+ "15": "T1068",
28
+ "16": "T1070.004",
29
+ "17": "T1071.001",
30
+ "18": "T1072",
31
+ "19": "T1074.001",
32
+ "20": "T1078",
33
+ "21": "T1082",
34
+ "22": "T1083",
35
+ "23": "T1090",
36
+ "24": "T1095",
37
+ "25": "T1105",
38
+ "26": "T1106",
39
+ "27": "T1110",
40
+ "28": "T1112",
41
+ "29": "T1113",
42
+ "30": "T1140",
43
+ "31": "T1190",
44
+ "32": "T1204.002",
45
+ "33": "T1210",
46
+ "34": "T1218.011",
47
+ "35": "T1219",
48
+ "36": "T1484.001",
49
+ "37": "T1518.001",
50
+ "38": "T1543.003",
51
+ "39": "T1547.001",
52
+ "40": "T1548.002",
53
+ "41": "T1552.001",
54
+ "42": "T1557.001",
55
+ "43": "T1562.001",
56
+ "44": "T1564.001",
57
+ "45": "T1566.001",
58
+ "46": "T1569.002",
59
+ "47": "T1570",
60
+ "48": "T1573.001",
61
+ "49": "T1574.002"
62
+ },
63
+ "initializer_range": 0.02,
64
+ "intermediate_size": 3072,
65
+ "label2id": {
66
+ "T1003.001": 0,
67
+ "T1005": 1,
68
+ "T1012": 2,
69
+ "T1016": 3,
70
+ "T1021.001": 4,
71
+ "T1027": 5,
72
+ "T1033": 6,
73
+ "T1036.005": 7,
74
+ "T1041": 8,
75
+ "T1047": 9,
76
+ "T1053.005": 10,
77
+ "T1055": 11,
78
+ "T1056.001": 12,
79
+ "T1057": 13,
80
+ "T1059.003": 14,
81
+ "T1068": 15,
82
+ "T1070.004": 16,
83
+ "T1071.001": 17,
84
+ "T1072": 18,
85
+ "T1074.001": 19,
86
+ "T1078": 20,
87
+ "T1082": 21,
88
+ "T1083": 22,
89
+ "T1090": 23,
90
+ "T1095": 24,
91
+ "T1105": 25,
92
+ "T1106": 26,
93
+ "T1110": 27,
94
+ "T1112": 28,
95
+ "T1113": 29,
96
+ "T1140": 30,
97
+ "T1190": 31,
98
+ "T1204.002": 32,
99
+ "T1210": 33,
100
+ "T1218.011": 34,
101
+ "T1219": 35,
102
+ "T1484.001": 36,
103
+ "T1518.001": 37,
104
+ "T1543.003": 38,
105
+ "T1547.001": 39,
106
+ "T1548.002": 40,
107
+ "T1552.001": 41,
108
+ "T1557.001": 42,
109
+ "T1562.001": 43,
110
+ "T1564.001": 44,
111
+ "T1566.001": 45,
112
+ "T1569.002": 46,
113
+ "T1570": 47,
114
+ "T1573.001": 48,
115
+ "T1574.002": 49
116
+ },
117
+ "layer_norm_eps": 1e-12,
118
+ "max_position_embeddings": 512,
119
+ "model_type": "bert",
120
+ "num_attention_heads": 12,
121
+ "num_hidden_layers": 12,
122
+ "pad_token_id": 0,
123
+ "position_embedding_type": "absolute",
124
+ "problem_type": "multi_label_classification",
125
+ "torch_dtype": "float32",
126
+ "transformers_version": "4.55.2",
127
+ "type_vocab_size": 2,
128
+ "use_cache": true,
129
+ "vocab_size": 30522
130
+ }
checkpoint-360/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ade367c852e7f7fefe67f1f0b6de558c5883b4312ae236fd4672b57b25c5a494
3
+ size 438106296