Theoreticallyhugo commited on
Commit
b5f8376
1 Parent(s): abddbf4

Training in progress, epoch 36, checkpoint

Browse files
checkpoint-2916/config.json ADDED
@@ -0,0 +1,53 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "allenai/longformer-base-4096",
3
+ "architectures": [
4
+ "LongformerForTokenClassification"
5
+ ],
6
+ "attention_mode": "longformer",
7
+ "attention_probs_dropout_prob": 0.1,
8
+ "attention_window": [
9
+ 700,
10
+ 700,
11
+ 700,
12
+ 700,
13
+ 700,
14
+ 700,
15
+ 700,
16
+ 700,
17
+ 700,
18
+ 700,
19
+ 700,
20
+ 700
21
+ ],
22
+ "bos_token_id": 0,
23
+ "eos_token_id": 2,
24
+ "gradient_checkpointing": false,
25
+ "hidden_act": "gelu",
26
+ "hidden_dropout_prob": 0.1,
27
+ "hidden_size": 768,
28
+ "id2label": {
29
+ "0": "O",
30
+ "1": "B",
31
+ "2": "I"
32
+ },
33
+ "ignore_attention_mask": false,
34
+ "initializer_range": 0.02,
35
+ "intermediate_size": 3072,
36
+ "label2id": {
37
+ "B": 1,
38
+ "I": 2,
39
+ "O": 0
40
+ },
41
+ "layer_norm_eps": 1e-05,
42
+ "max_position_embeddings": 4098,
43
+ "model_type": "longformer",
44
+ "num_attention_heads": 12,
45
+ "num_hidden_layers": 12,
46
+ "onnx_export": false,
47
+ "pad_token_id": 1,
48
+ "sep_token_id": 2,
49
+ "torch_dtype": "float32",
50
+ "transformers_version": "4.38.2",
51
+ "type_vocab_size": 1,
52
+ "vocab_size": 50265
53
+ }
checkpoint-2916/merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-2916/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:72ccd9affbd399d5f7785085e7e6bb70322408f609747a186c94de5a2f007d8a
3
+ size 592318676
checkpoint-2916/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7cb2a2ea6a92646fdf1185b7895b8c7eff88f8e3375c9cefb20bac85e0b8cfc8
3
+ size 1014645498
checkpoint-2916/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1092929a34f62aeea56e320b108dc708bfb48a5a1e0d610f0606ba11e8f55b81
3
+ size 14244
checkpoint-2916/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e0c1547281309162f40e30318d0efec0382975b6389d151874d4f0376d4ea3cf
3
+ size 1064
checkpoint-2916/special_tokens_map.json ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "<s>",
3
+ "cls_token": "<s>",
4
+ "eos_token": "</s>",
5
+ "mask_token": {
6
+ "content": "<mask>",
7
+ "lstrip": true,
8
+ "normalized": false,
9
+ "rstrip": false,
10
+ "single_word": false
11
+ },
12
+ "pad_token": "<pad>",
13
+ "sep_token": "</s>",
14
+ "unk_token": "<unk>"
15
+ }
checkpoint-2916/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-2916/tokenizer_config.json ADDED
@@ -0,0 +1,57 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": true,
3
+ "added_tokens_decoder": {
4
+ "0": {
5
+ "content": "<s>",
6
+ "lstrip": false,
7
+ "normalized": true,
8
+ "rstrip": false,
9
+ "single_word": false,
10
+ "special": true
11
+ },
12
+ "1": {
13
+ "content": "<pad>",
14
+ "lstrip": false,
15
+ "normalized": true,
16
+ "rstrip": false,
17
+ "single_word": false,
18
+ "special": true
19
+ },
20
+ "2": {
21
+ "content": "</s>",
22
+ "lstrip": false,
23
+ "normalized": true,
24
+ "rstrip": false,
25
+ "single_word": false,
26
+ "special": true
27
+ },
28
+ "3": {
29
+ "content": "<unk>",
30
+ "lstrip": false,
31
+ "normalized": true,
32
+ "rstrip": false,
33
+ "single_word": false,
34
+ "special": true
35
+ },
36
+ "50264": {
37
+ "content": "<mask>",
38
+ "lstrip": true,
39
+ "normalized": false,
40
+ "rstrip": false,
41
+ "single_word": false,
42
+ "special": true
43
+ }
44
+ },
45
+ "bos_token": "<s>",
46
+ "clean_up_tokenization_spaces": true,
47
+ "cls_token": "<s>",
48
+ "eos_token": "</s>",
49
+ "errors": "replace",
50
+ "mask_token": "<mask>",
51
+ "model_max_length": 4096,
52
+ "pad_token": "<pad>",
53
+ "sep_token": "</s>",
54
+ "tokenizer_class": "LongformerTokenizer",
55
+ "trim_offsets": true,
56
+ "unk_token": "<unk>"
57
+ }
checkpoint-2916/trainer_state.json ADDED
@@ -0,0 +1,1460 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 36.0,
5
+ "eval_steps": 500,
6
+ "global_step": 2916,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 1.0,
13
+ "eval_B": {
14
+ "f1-score": 0.8266764382557715,
15
+ "precision": 0.7866108786610879,
16
+ "recall": 0.871042471042471,
17
+ "support": 1295.0
18
+ },
19
+ "eval_I": {
20
+ "f1-score": 0.9410658617447671,
21
+ "precision": 0.9344077040239768,
22
+ "recall": 0.9478195863443808,
23
+ "support": 20065.0
24
+ },
25
+ "eval_O": {
26
+ "f1-score": 0.8562443302086484,
27
+ "precision": 0.8789421405512788,
28
+ "recall": 0.8346893055064262,
29
+ "support": 8481.0
30
+ },
31
+ "eval_accuracy": 0.912335377500754,
32
+ "eval_loss": 0.2450982630252838,
33
+ "eval_macro avg": {
34
+ "f1-score": 0.8746622100697291,
35
+ "precision": 0.8666535744121145,
36
+ "recall": 0.8845171209644261,
37
+ "support": 29841.0
38
+ },
39
+ "eval_runtime": 4.6269,
40
+ "eval_samples_per_second": 17.29,
41
+ "eval_steps_per_second": 2.161,
42
+ "eval_weighted avg": {
43
+ "f1-score": 0.9119949287205363,
44
+ "precision": 0.9122301519092053,
45
+ "recall": 0.912335377500754,
46
+ "support": 29841.0
47
+ },
48
+ "step": 81
49
+ },
50
+ {
51
+ "epoch": 2.0,
52
+ "eval_B": {
53
+ "f1-score": 0.8833395452851286,
54
+ "precision": 0.8537463976945245,
55
+ "recall": 0.915057915057915,
56
+ "support": 1295.0
57
+ },
58
+ "eval_I": {
59
+ "f1-score": 0.9549518227873249,
60
+ "precision": 0.9516903430183636,
61
+ "recall": 0.958235733864939,
62
+ "support": 20065.0
63
+ },
64
+ "eval_O": {
65
+ "f1-score": 0.88924750463212,
66
+ "precision": 0.9016969696969697,
67
+ "recall": 0.877137130055418,
68
+ "support": 8481.0
69
+ },
70
+ "eval_accuracy": 0.9333132267685399,
71
+ "eval_loss": 0.20076116919517517,
72
+ "eval_macro avg": {
73
+ "f1-score": 0.9091796242348579,
74
+ "precision": 0.9023779034699526,
75
+ "recall": 0.916810259659424,
76
+ "support": 29841.0
77
+ },
78
+ "eval_runtime": 4.6524,
79
+ "eval_samples_per_second": 17.195,
80
+ "eval_steps_per_second": 2.149,
81
+ "eval_weighted avg": {
82
+ "f1-score": 0.9331705077630417,
83
+ "precision": 0.9332314707173981,
84
+ "recall": 0.9333132267685399,
85
+ "support": 29841.0
86
+ },
87
+ "step": 162
88
+ },
89
+ {
90
+ "epoch": 3.0,
91
+ "eval_B": {
92
+ "f1-score": 0.8927335640138409,
93
+ "precision": 0.8889739663093414,
94
+ "recall": 0.8965250965250965,
95
+ "support": 1295.0
96
+ },
97
+ "eval_I": {
98
+ "f1-score": 0.9590235396687009,
99
+ "precision": 0.9586653386454184,
100
+ "recall": 0.9593820084724645,
101
+ "support": 20065.0
102
+ },
103
+ "eval_O": {
104
+ "f1-score": 0.9006849315068494,
105
+ "precision": 0.9020697811945594,
106
+ "recall": 0.8993043273198915,
107
+ "support": 8481.0
108
+ },
109
+ "eval_accuracy": 0.9395797727958178,
110
+ "eval_loss": 0.19863653182983398,
111
+ "eval_macro avg": {
112
+ "f1-score": 0.9174806783964637,
113
+ "precision": 0.9165696953831063,
114
+ "recall": 0.9184038107724842,
115
+ "support": 29841.0
116
+ },
117
+ "eval_runtime": 4.655,
118
+ "eval_samples_per_second": 17.186,
119
+ "eval_steps_per_second": 2.148,
120
+ "eval_weighted avg": {
121
+ "f1-score": 0.9395665759512081,
122
+ "precision": 0.9395561516236713,
123
+ "recall": 0.9395797727958178,
124
+ "support": 29841.0
125
+ },
126
+ "step": 243
127
+ },
128
+ {
129
+ "epoch": 4.0,
130
+ "eval_B": {
131
+ "f1-score": 0.8999616711383672,
132
+ "precision": 0.893455098934551,
133
+ "recall": 0.9065637065637066,
134
+ "support": 1295.0
135
+ },
136
+ "eval_I": {
137
+ "f1-score": 0.9612657113512985,
138
+ "precision": 0.9436362763454799,
139
+ "recall": 0.9795664091701969,
140
+ "support": 20065.0
141
+ },
142
+ "eval_O": {
143
+ "f1-score": 0.8996847765622102,
144
+ "precision": 0.9454403741231489,
145
+ "recall": 0.8581535196321188,
146
+ "support": 8481.0
147
+ },
148
+ "eval_accuracy": 0.9418920277470594,
149
+ "eval_loss": 0.19743724167346954,
150
+ "eval_macro avg": {
151
+ "f1-score": 0.920304053017292,
152
+ "precision": 0.9275105831343932,
153
+ "recall": 0.9147612117886741,
154
+ "support": 29841.0
155
+ },
156
+ "eval_runtime": 4.6615,
157
+ "eval_samples_per_second": 17.162,
158
+ "eval_steps_per_second": 2.145,
159
+ "eval_weighted avg": {
160
+ "f1-score": 0.9411036309913238,
161
+ "precision": 0.9419713163409645,
162
+ "recall": 0.9418920277470594,
163
+ "support": 29841.0
164
+ },
165
+ "step": 324
166
+ },
167
+ {
168
+ "epoch": 5.0,
169
+ "eval_B": {
170
+ "f1-score": 0.8953846153846153,
171
+ "precision": 0.8919540229885058,
172
+ "recall": 0.8988416988416988,
173
+ "support": 1295.0
174
+ },
175
+ "eval_I": {
176
+ "f1-score": 0.9602426713742261,
177
+ "precision": 0.9581224570804803,
178
+ "recall": 0.9623722900573137,
179
+ "support": 20065.0
180
+ },
181
+ "eval_O": {
182
+ "f1-score": 0.9022119433078337,
183
+ "precision": 0.9075399665950847,
184
+ "recall": 0.8969461148449476,
185
+ "support": 8481.0
186
+ },
187
+ "eval_accuracy": 0.9410207432726785,
188
+ "eval_loss": 0.2340461015701294,
189
+ "eval_macro avg": {
190
+ "f1-score": 0.9192797433555584,
191
+ "precision": 0.9192054822213569,
192
+ "recall": 0.9193867012479867,
193
+ "support": 29841.0
194
+ },
195
+ "eval_runtime": 4.6719,
196
+ "eval_samples_per_second": 17.124,
197
+ "eval_steps_per_second": 2.14,
198
+ "eval_weighted avg": {
199
+ "f1-score": 0.9409353496612266,
200
+ "precision": 0.9408751053176122,
201
+ "recall": 0.9410207432726785,
202
+ "support": 29841.0
203
+ },
204
+ "step": 405
205
+ },
206
+ {
207
+ "epoch": 6.0,
208
+ "eval_B": {
209
+ "f1-score": 0.8929831438651511,
210
+ "precision": 0.9068471337579618,
211
+ "recall": 0.8795366795366796,
212
+ "support": 1295.0
213
+ },
214
+ "eval_I": {
215
+ "f1-score": 0.9522293798155866,
216
+ "precision": 0.9655225409836066,
217
+ "recall": 0.9392972838275604,
218
+ "support": 20065.0
219
+ },
220
+ "eval_O": {
221
+ "f1-score": 0.8885216003647556,
222
+ "precision": 0.8599007170435742,
223
+ "recall": 0.919113312109421,
224
+ "support": 8481.0
225
+ },
226
+ "eval_accuracy": 0.930967460875976,
227
+ "eval_loss": 0.21774999797344208,
228
+ "eval_macro avg": {
229
+ "f1-score": 0.9112447080151643,
230
+ "precision": 0.9107567972617142,
231
+ "recall": 0.9126490918245537,
232
+ "support": 29841.0
233
+ },
234
+ "eval_runtime": 4.647,
235
+ "eval_samples_per_second": 17.215,
236
+ "eval_steps_per_second": 2.152,
237
+ "eval_weighted avg": {
238
+ "f1-score": 0.9315521386682287,
239
+ "precision": 0.9329578366776978,
240
+ "recall": 0.930967460875976,
241
+ "support": 29841.0
242
+ },
243
+ "step": 486
244
+ },
245
+ {
246
+ "epoch": 6.17,
247
+ "grad_norm": 9.310327529907227,
248
+ "learning_rate": 1.7530864197530865e-05,
249
+ "loss": 0.1627,
250
+ "step": 500
251
+ },
252
+ {
253
+ "epoch": 7.0,
254
+ "eval_B": {
255
+ "f1-score": 0.8936490041337843,
256
+ "precision": 0.8704245973645681,
257
+ "recall": 0.9181467181467181,
258
+ "support": 1295.0
259
+ },
260
+ "eval_I": {
261
+ "f1-score": 0.9599271850630982,
262
+ "precision": 0.9477800446905664,
263
+ "recall": 0.9723897333665587,
264
+ "support": 20065.0
265
+ },
266
+ "eval_O": {
267
+ "f1-score": 0.8979841172877214,
268
+ "precision": 0.9316770186335404,
269
+ "recall": 0.8666430845419172,
270
+ "support": 8481.0
271
+ },
272
+ "eval_accuracy": 0.9399819040916859,
273
+ "eval_loss": 0.27376389503479004,
274
+ "eval_macro avg": {
275
+ "f1-score": 0.9171867688282013,
276
+ "precision": 0.9166272202295582,
277
+ "recall": 0.9190598453517312,
278
+ "support": 29841.0
279
+ },
280
+ "eval_runtime": 4.6671,
281
+ "eval_samples_per_second": 17.141,
282
+ "eval_steps_per_second": 2.143,
283
+ "eval_weighted avg": {
284
+ "f1-score": 0.9394463230910989,
285
+ "precision": 0.9398464945991887,
286
+ "recall": 0.9399819040916859,
287
+ "support": 29841.0
288
+ },
289
+ "step": 567
290
+ },
291
+ {
292
+ "epoch": 8.0,
293
+ "eval_B": {
294
+ "f1-score": 0.8948545861297539,
295
+ "precision": 0.8651766402307137,
296
+ "recall": 0.9266409266409267,
297
+ "support": 1295.0
298
+ },
299
+ "eval_I": {
300
+ "f1-score": 0.9590139880661254,
301
+ "precision": 0.9414702069429106,
302
+ "recall": 0.9772240219287316,
303
+ "support": 20065.0
304
+ },
305
+ "eval_O": {
306
+ "f1-score": 0.8927241122423641,
307
+ "precision": 0.9427035531663825,
308
+ "recall": 0.8477773847423653,
309
+ "support": 8481.0
310
+ },
311
+ "eval_accuracy": 0.9382393351429241,
312
+ "eval_loss": 0.3103167414665222,
313
+ "eval_macro avg": {
314
+ "f1-score": 0.9155308954794145,
315
+ "precision": 0.916450133446669,
316
+ "recall": 0.9172141111040079,
317
+ "support": 29841.0
318
+ },
319
+ "eval_runtime": 4.6339,
320
+ "eval_samples_per_second": 17.264,
321
+ "eval_steps_per_second": 2.158,
322
+ "eval_weighted avg": {
323
+ "f1-score": 0.9373896838414373,
324
+ "precision": 0.9385098450391195,
325
+ "recall": 0.9382393351429241,
326
+ "support": 29841.0
327
+ },
328
+ "step": 648
329
+ },
330
+ {
331
+ "epoch": 9.0,
332
+ "eval_B": {
333
+ "f1-score": 0.9011235955056179,
334
+ "precision": 0.8749090909090909,
335
+ "recall": 0.9289575289575289,
336
+ "support": 1295.0
337
+ },
338
+ "eval_I": {
339
+ "f1-score": 0.9615119937082186,
340
+ "precision": 0.9485040973670175,
341
+ "recall": 0.9748816346872664,
342
+ "support": 20065.0
343
+ },
344
+ "eval_O": {
345
+ "f1-score": 0.9013722126929674,
346
+ "precision": 0.9380339155935229,
347
+ "recall": 0.8674684589081476,
348
+ "support": 8481.0
349
+ },
350
+ "eval_accuracy": 0.9423611809255722,
351
+ "eval_loss": 0.32883673906326294,
352
+ "eval_macro avg": {
353
+ "f1-score": 0.9213359339689347,
354
+ "precision": 0.9204823679565438,
355
+ "recall": 0.9237692075176476,
356
+ "support": 29841.0
357
+ },
358
+ "eval_runtime": 4.6714,
359
+ "eval_samples_per_second": 17.126,
360
+ "eval_steps_per_second": 2.141,
361
+ "eval_weighted avg": {
362
+ "f1-score": 0.9417992341337168,
363
+ "precision": 0.9423346276781992,
364
+ "recall": 0.9423611809255722,
365
+ "support": 29841.0
366
+ },
367
+ "step": 729
368
+ },
369
+ {
370
+ "epoch": 10.0,
371
+ "eval_B": {
372
+ "f1-score": 0.8935361216730039,
373
+ "precision": 0.8801498127340824,
374
+ "recall": 0.9073359073359073,
375
+ "support": 1295.0
376
+ },
377
+ "eval_I": {
378
+ "f1-score": 0.9632332861646931,
379
+ "precision": 0.9596359319351009,
380
+ "recall": 0.9668577124345876,
381
+ "support": 20065.0
382
+ },
383
+ "eval_O": {
384
+ "f1-score": 0.9095462405342556,
385
+ "precision": 0.9200241254523522,
386
+ "recall": 0.8993043273198915,
387
+ "support": 8481.0
388
+ },
389
+ "eval_accuracy": 0.9450755671726819,
390
+ "eval_loss": 0.3041447699069977,
391
+ "eval_macro avg": {
392
+ "f1-score": 0.9221052161239841,
393
+ "precision": 0.9199366233738452,
394
+ "recall": 0.9244993156967954,
395
+ "support": 29841.0
396
+ },
397
+ "eval_runtime": 4.6532,
398
+ "eval_samples_per_second": 17.193,
399
+ "eval_steps_per_second": 2.149,
400
+ "eval_weighted avg": {
401
+ "f1-score": 0.9449504651463465,
402
+ "precision": 0.9449285744355027,
403
+ "recall": 0.9450755671726819,
404
+ "support": 29841.0
405
+ },
406
+ "step": 810
407
+ },
408
+ {
409
+ "epoch": 11.0,
410
+ "eval_B": {
411
+ "f1-score": 0.8924122310305775,
412
+ "precision": 0.8729689807976366,
413
+ "recall": 0.9127413127413128,
414
+ "support": 1295.0
415
+ },
416
+ "eval_I": {
417
+ "f1-score": 0.9588723786993913,
418
+ "precision": 0.9483330083837005,
419
+ "recall": 0.9696486419137802,
420
+ "support": 20065.0
421
+ },
422
+ "eval_O": {
423
+ "f1-score": 0.8956965718453683,
424
+ "precision": 0.9243507715468574,
425
+ "recall": 0.8687654757693668,
426
+ "support": 8481.0
427
+ },
428
+ "eval_accuracy": 0.9385074226735028,
429
+ "eval_loss": 0.3531416654586792,
430
+ "eval_macro avg": {
431
+ "f1-score": 0.9156603938584458,
432
+ "precision": 0.9152175869093981,
433
+ "recall": 0.9170518101414866,
434
+ "support": 29841.0
435
+ },
436
+ "eval_runtime": 4.6508,
437
+ "eval_samples_per_second": 17.201,
438
+ "eval_steps_per_second": 2.15,
439
+ "eval_weighted avg": {
440
+ "f1-score": 0.938033267772811,
441
+ "precision": 0.9382465579853486,
442
+ "recall": 0.9385074226735028,
443
+ "support": 29841.0
444
+ },
445
+ "step": 891
446
+ },
447
+ {
448
+ "epoch": 12.0,
449
+ "eval_B": {
450
+ "f1-score": 0.8970199924556771,
451
+ "precision": 0.8768436578171092,
452
+ "recall": 0.9181467181467181,
453
+ "support": 1295.0
454
+ },
455
+ "eval_I": {
456
+ "f1-score": 0.9595004897159647,
457
+ "precision": 0.9431046931407943,
458
+ "recall": 0.9764764515325193,
459
+ "support": 20065.0
460
+ },
461
+ "eval_O": {
462
+ "f1-score": 0.8953122104873078,
463
+ "precision": 0.9400778210116731,
464
+ "recall": 0.8546162009197029,
465
+ "support": 8481.0
466
+ },
467
+ "eval_accuracy": 0.9393116852652391,
468
+ "eval_loss": 0.4237186312675476,
469
+ "eval_macro avg": {
470
+ "f1-score": 0.9172775642196499,
471
+ "precision": 0.9200087239898588,
472
+ "recall": 0.9164131235329801,
473
+ "support": 29841.0
474
+ },
475
+ "eval_runtime": 4.6388,
476
+ "eval_samples_per_second": 17.246,
477
+ "eval_steps_per_second": 2.156,
478
+ "eval_weighted avg": {
479
+ "f1-score": 0.9385463313402296,
480
+ "precision": 0.9393689288141548,
481
+ "recall": 0.9393116852652391,
482
+ "support": 29841.0
483
+ },
484
+ "step": 972
485
+ },
486
+ {
487
+ "epoch": 12.35,
488
+ "grad_norm": 0.5223535895347595,
489
+ "learning_rate": 1.506172839506173e-05,
490
+ "loss": 0.0196,
491
+ "step": 1000
492
+ },
493
+ {
494
+ "epoch": 13.0,
495
+ "eval_B": {
496
+ "f1-score": 0.8891411275066211,
497
+ "precision": 0.8716617210682492,
498
+ "recall": 0.9073359073359073,
499
+ "support": 1295.0
500
+ },
501
+ "eval_I": {
502
+ "f1-score": 0.9610692322951795,
503
+ "precision": 0.9563736860287223,
504
+ "recall": 0.9658111138798904,
505
+ "support": 20065.0
506
+ },
507
+ "eval_O": {
508
+ "f1-score": 0.9043145233678416,
509
+ "precision": 0.918104495747266,
510
+ "recall": 0.8909326730338404,
511
+ "support": 8481.0
512
+ },
513
+ "eval_accuracy": 0.9419925605710264,
514
+ "eval_loss": 0.4308999180793762,
515
+ "eval_macro avg": {
516
+ "f1-score": 0.9181749610565474,
517
+ "precision": 0.9153799676147458,
518
+ "recall": 0.9213598980832126,
519
+ "support": 29841.0
520
+ },
521
+ "eval_runtime": 4.6555,
522
+ "eval_samples_per_second": 17.184,
523
+ "eval_steps_per_second": 2.148,
524
+ "eval_weighted avg": {
525
+ "f1-score": 0.9418177466843106,
526
+ "precision": 0.9418211242043584,
527
+ "recall": 0.9419925605710264,
528
+ "support": 29841.0
529
+ },
530
+ "step": 1053
531
+ },
532
+ {
533
+ "epoch": 14.0,
534
+ "eval_B": {
535
+ "f1-score": 0.9025601834161253,
536
+ "precision": 0.8933434190620272,
537
+ "recall": 0.911969111969112,
538
+ "support": 1295.0
539
+ },
540
+ "eval_I": {
541
+ "f1-score": 0.960806215404473,
542
+ "precision": 0.9570764513895758,
543
+ "recall": 0.9645651632195366,
544
+ "support": 20065.0
545
+ },
546
+ "eval_O": {
547
+ "f1-score": 0.9030873763261413,
548
+ "precision": 0.9131011208870676,
549
+ "recall": 0.8932908855087843,
550
+ "support": 8481.0
551
+ },
552
+ "eval_accuracy": 0.9420260715123487,
553
+ "eval_loss": 0.4198700487613678,
554
+ "eval_macro avg": {
555
+ "f1-score": 0.9221512583822467,
556
+ "precision": 0.9211736637795568,
557
+ "recall": 0.9232750535658111,
558
+ "support": 29841.0
559
+ },
560
+ "eval_runtime": 4.6834,
561
+ "eval_samples_per_second": 17.082,
562
+ "eval_steps_per_second": 2.135,
563
+ "eval_weighted avg": {
564
+ "f1-score": 0.9418744743217934,
565
+ "precision": 0.9418125843993292,
566
+ "recall": 0.9420260715123487,
567
+ "support": 29841.0
568
+ },
569
+ "step": 1134
570
+ },
571
+ {
572
+ "epoch": 15.0,
573
+ "eval_B": {
574
+ "f1-score": 0.8935681470137825,
575
+ "precision": 0.8861047835990888,
576
+ "recall": 0.9011583011583012,
577
+ "support": 1295.0
578
+ },
579
+ "eval_I": {
580
+ "f1-score": 0.9633398694432723,
581
+ "precision": 0.9595530063291139,
582
+ "recall": 0.9671567405930726,
583
+ "support": 20065.0
584
+ },
585
+ "eval_O": {
586
+ "f1-score": 0.9100768726535964,
587
+ "precision": 0.92,
588
+ "recall": 0.9003655229336163,
589
+ "support": 8481.0
590
+ },
591
+ "eval_accuracy": 0.9453101437619382,
592
+ "eval_loss": 0.39179423451423645,
593
+ "eval_macro avg": {
594
+ "f1-score": 0.9223282963702171,
595
+ "precision": 0.9218859299760677,
596
+ "recall": 0.9228935215616634,
597
+ "support": 29841.0
598
+ },
599
+ "eval_runtime": 4.6645,
600
+ "eval_samples_per_second": 17.151,
601
+ "eval_steps_per_second": 2.144,
602
+ "eval_weighted avg": {
603
+ "f1-score": 0.945174330208011,
604
+ "precision": 0.9451243847979119,
605
+ "recall": 0.9453101437619382,
606
+ "support": 29841.0
607
+ },
608
+ "step": 1215
609
+ },
610
+ {
611
+ "epoch": 16.0,
612
+ "eval_B": {
613
+ "f1-score": 0.9022614028363357,
614
+ "precision": 0.895738203957382,
615
+ "recall": 0.9088803088803089,
616
+ "support": 1295.0
617
+ },
618
+ "eval_I": {
619
+ "f1-score": 0.9632045864531594,
620
+ "precision": 0.9552494853445741,
621
+ "recall": 0.9712932967854473,
622
+ "support": 20065.0
623
+ },
624
+ "eval_O": {
625
+ "f1-score": 0.9083463808262073,
626
+ "precision": 0.9282461538461538,
627
+ "recall": 0.8892819243013795,
628
+ "support": 8481.0
629
+ },
630
+ "eval_accuracy": 0.9452766328206159,
631
+ "eval_loss": 0.44052162766456604,
632
+ "eval_macro avg": {
633
+ "f1-score": 0.9246041233719008,
634
+ "precision": 0.92641128104937,
635
+ "recall": 0.9231518433223785,
636
+ "support": 29841.0
637
+ },
638
+ "eval_runtime": 4.6376,
639
+ "eval_samples_per_second": 17.25,
640
+ "eval_steps_per_second": 2.156,
641
+ "eval_weighted avg": {
642
+ "f1-score": 0.9449688080038458,
643
+ "precision": 0.944992377210312,
644
+ "recall": 0.9452766328206159,
645
+ "support": 29841.0
646
+ },
647
+ "step": 1296
648
+ },
649
+ {
650
+ "epoch": 17.0,
651
+ "eval_B": {
652
+ "f1-score": 0.8923649906890131,
653
+ "precision": 0.8618705035971223,
654
+ "recall": 0.9250965250965251,
655
+ "support": 1295.0
656
+ },
657
+ "eval_I": {
658
+ "f1-score": 0.9599605522682445,
659
+ "precision": 0.9498902171261283,
660
+ "recall": 0.97024669823075,
661
+ "support": 20065.0
662
+ },
663
+ "eval_O": {
664
+ "f1-score": 0.9001642635517431,
665
+ "precision": 0.9298642533936652,
666
+ "recall": 0.8723027944817828,
667
+ "support": 8481.0
668
+ },
669
+ "eval_accuracy": 0.9404510572701987,
670
+ "eval_loss": 0.4657233655452728,
671
+ "eval_macro avg": {
672
+ "f1-score": 0.9174966021696669,
673
+ "precision": 0.9138749913723053,
674
+ "recall": 0.9225486726030193,
675
+ "support": 29841.0
676
+ },
677
+ "eval_runtime": 4.6396,
678
+ "eval_samples_per_second": 17.243,
679
+ "eval_steps_per_second": 2.155,
680
+ "eval_weighted avg": {
681
+ "f1-score": 0.940032648483192,
682
+ "precision": 0.9403789498282803,
683
+ "recall": 0.9404510572701987,
684
+ "support": 29841.0
685
+ },
686
+ "step": 1377
687
+ },
688
+ {
689
+ "epoch": 18.0,
690
+ "eval_B": {
691
+ "f1-score": 0.9014302280633939,
692
+ "precision": 0.9024767801857585,
693
+ "recall": 0.9003861003861003,
694
+ "support": 1295.0
695
+ },
696
+ "eval_I": {
697
+ "f1-score": 0.9638021281301453,
698
+ "precision": 0.9506496024820632,
699
+ "recall": 0.9773236979815599,
700
+ "support": 20065.0
701
+ },
702
+ "eval_O": {
703
+ "f1-score": 0.9083038653822705,
704
+ "precision": 0.9404115641964399,
705
+ "recall": 0.87831623629289,
706
+ "support": 8481.0
707
+ },
708
+ "eval_accuracy": 0.9458463188230958,
709
+ "eval_loss": 0.4731404185295105,
710
+ "eval_macro avg": {
711
+ "f1-score": 0.9245120738586032,
712
+ "precision": 0.9311793156214204,
713
+ "recall": 0.9186753448868501,
714
+ "support": 29841.0
715
+ },
716
+ "eval_runtime": 4.6405,
717
+ "eval_samples_per_second": 17.24,
718
+ "eval_steps_per_second": 2.155,
719
+ "eval_weighted avg": {
720
+ "f1-score": 0.9453224398840687,
721
+ "precision": 0.9456493475450943,
722
+ "recall": 0.9458463188230958,
723
+ "support": 29841.0
724
+ },
725
+ "step": 1458
726
+ },
727
+ {
728
+ "epoch": 18.52,
729
+ "grad_norm": 1.548362135887146,
730
+ "learning_rate": 1.2592592592592593e-05,
731
+ "loss": 0.0038,
732
+ "step": 1500
733
+ },
734
+ {
735
+ "epoch": 19.0,
736
+ "eval_B": {
737
+ "f1-score": 0.8891402714932127,
738
+ "precision": 0.868828297715549,
739
+ "recall": 0.9104247104247104,
740
+ "support": 1295.0
741
+ },
742
+ "eval_I": {
743
+ "f1-score": 0.9603099696984749,
744
+ "precision": 0.9571718572065158,
745
+ "recall": 0.9634687266384251,
746
+ "support": 20065.0
747
+ },
748
+ "eval_O": {
749
+ "f1-score": 0.9027910305343512,
750
+ "precision": 0.913358271992277,
751
+ "recall": 0.8924655111425539,
752
+ "support": 8481.0
753
+ },
754
+ "eval_accuracy": 0.9409872323313562,
755
+ "eval_loss": 0.49517783522605896,
756
+ "eval_macro avg": {
757
+ "f1-score": 0.9174137572420129,
758
+ "precision": 0.9131194756381139,
759
+ "recall": 0.9221196494018965,
760
+ "support": 29841.0
761
+ },
762
+ "eval_runtime": 4.6377,
763
+ "eval_samples_per_second": 17.25,
764
+ "eval_steps_per_second": 2.156,
765
+ "eval_weighted avg": {
766
+ "f1-score": 0.9408741973642117,
767
+ "precision": 0.9408859443435835,
768
+ "recall": 0.9409872323313562,
769
+ "support": 29841.0
770
+ },
771
+ "step": 1539
772
+ },
773
+ {
774
+ "epoch": 20.0,
775
+ "eval_B": {
776
+ "f1-score": 0.899884925201381,
777
+ "precision": 0.8940548780487805,
778
+ "recall": 0.9057915057915058,
779
+ "support": 1295.0
780
+ },
781
+ "eval_I": {
782
+ "f1-score": 0.9634953846153845,
783
+ "precision": 0.951896887159533,
784
+ "recall": 0.9753800149514079,
785
+ "support": 20065.0
786
+ },
787
+ "eval_O": {
788
+ "f1-score": 0.9073556231003039,
789
+ "precision": 0.9365039528171665,
790
+ "recall": 0.8799669850253508,
791
+ "support": 8481.0
792
+ },
793
+ "eval_accuracy": 0.9452431218792936,
794
+ "eval_loss": 0.45907479524612427,
795
+ "eval_macro avg": {
796
+ "f1-score": 0.9235786443056897,
797
+ "precision": 0.9274852393418267,
798
+ "recall": 0.9203795019227549,
799
+ "support": 29841.0
800
+ },
801
+ "eval_runtime": 4.6356,
802
+ "eval_samples_per_second": 17.258,
803
+ "eval_steps_per_second": 2.157,
804
+ "eval_weighted avg": {
805
+ "f1-score": 0.9447796290324438,
806
+ "precision": 0.9450119678218422,
807
+ "recall": 0.9452431218792936,
808
+ "support": 29841.0
809
+ },
810
+ "step": 1620
811
+ },
812
+ {
813
+ "epoch": 21.0,
814
+ "eval_B": {
815
+ "f1-score": 0.8944337811900192,
816
+ "precision": 0.8893129770992366,
817
+ "recall": 0.8996138996138996,
818
+ "support": 1295.0
819
+ },
820
+ "eval_I": {
821
+ "f1-score": 0.9624703322784811,
822
+ "precision": 0.9549624687239366,
823
+ "recall": 0.9700971841515076,
824
+ "support": 20065.0
825
+ },
826
+ "eval_O": {
827
+ "f1-score": 0.9066089361958025,
828
+ "precision": 0.9251350024545901,
829
+ "recall": 0.8888102818063908,
830
+ "support": 8481.0
831
+ },
832
+ "eval_accuracy": 0.9439361951677223,
833
+ "eval_loss": 0.4614206850528717,
834
+ "eval_macro avg": {
835
+ "f1-score": 0.9211710165547675,
836
+ "precision": 0.9231368160925877,
837
+ "recall": 0.9195071218572659,
838
+ "support": 29841.0
839
+ },
840
+ "eval_runtime": 4.6636,
841
+ "eval_samples_per_second": 17.154,
842
+ "eval_steps_per_second": 2.144,
843
+ "eval_weighted avg": {
844
+ "f1-score": 0.9436416122678664,
845
+ "precision": 0.9436363458364894,
846
+ "recall": 0.9439361951677223,
847
+ "support": 29841.0
848
+ },
849
+ "step": 1701
850
+ },
851
+ {
852
+ "epoch": 22.0,
853
+ "eval_B": {
854
+ "f1-score": 0.9011072928598702,
855
+ "precision": 0.8912386706948641,
856
+ "recall": 0.9111969111969112,
857
+ "support": 1295.0
858
+ },
859
+ "eval_I": {
860
+ "f1-score": 0.9635141147970535,
861
+ "precision": 0.9558585511795576,
862
+ "recall": 0.9712932967854473,
863
+ "support": 20065.0
864
+ },
865
+ "eval_O": {
866
+ "f1-score": 0.9095068938527304,
867
+ "precision": 0.9292568897637795,
868
+ "recall": 0.8905789411625987,
869
+ "support": 8481.0
870
+ },
871
+ "eval_accuracy": 0.9457457859991287,
872
+ "eval_loss": 0.4724768102169037,
873
+ "eval_macro avg": {
874
+ "f1-score": 0.9247094338365515,
875
+ "precision": 0.9254513705460671,
876
+ "recall": 0.9243563830483191,
877
+ "support": 29841.0
878
+ },
879
+ "eval_runtime": 4.6333,
880
+ "eval_samples_per_second": 17.266,
881
+ "eval_steps_per_second": 2.158,
882
+ "eval_weighted avg": {
883
+ "f1-score": 0.9454566745223489,
884
+ "precision": 0.9454939040264833,
885
+ "recall": 0.9457457859991287,
886
+ "support": 29841.0
887
+ },
888
+ "step": 1782
889
+ },
890
+ {
891
+ "epoch": 23.0,
892
+ "eval_B": {
893
+ "f1-score": 0.8956916099773242,
894
+ "precision": 0.8771280532938565,
895
+ "recall": 0.915057915057915,
896
+ "support": 1295.0
897
+ },
898
+ "eval_I": {
899
+ "f1-score": 0.9624542808091362,
900
+ "precision": 0.9609957269204015,
901
+ "recall": 0.9639172688761525,
902
+ "support": 20065.0
903
+ },
904
+ "eval_O": {
905
+ "f1-score": 0.9084001187295934,
906
+ "precision": 0.914753706360593,
907
+ "recall": 0.9021341822898243,
908
+ "support": 8481.0
909
+ },
910
+ "eval_accuracy": 0.9442377936396233,
911
+ "eval_loss": 0.47468113899230957,
912
+ "eval_macro avg": {
913
+ "f1-score": 0.9221820031720179,
914
+ "precision": 0.9176258288582838,
915
+ "recall": 0.927036455407964,
916
+ "support": 29841.0
917
+ },
918
+ "eval_runtime": 4.6476,
919
+ "eval_samples_per_second": 17.213,
920
+ "eval_steps_per_second": 2.152,
921
+ "eval_weighted avg": {
922
+ "f1-score": 0.9441944702356367,
923
+ "precision": 0.9442138759866489,
924
+ "recall": 0.9442377936396233,
925
+ "support": 29841.0
926
+ },
927
+ "step": 1863
928
+ },
929
+ {
930
+ "epoch": 24.0,
931
+ "eval_B": {
932
+ "f1-score": 0.9008390541571321,
933
+ "precision": 0.8899773926149209,
934
+ "recall": 0.911969111969112,
935
+ "support": 1295.0
936
+ },
937
+ "eval_I": {
938
+ "f1-score": 0.9639726399486381,
939
+ "precision": 0.9553151918559123,
940
+ "recall": 0.9727884375778719,
941
+ "support": 20065.0
942
+ },
943
+ "eval_O": {
944
+ "f1-score": 0.9101008271448409,
945
+ "precision": 0.9325661964860184,
946
+ "recall": 0.8886923711826435,
947
+ "support": 8481.0
948
+ },
949
+ "eval_accuracy": 0.9462484501189639,
950
+ "eval_loss": 0.4712923467159271,
951
+ "eval_macro avg": {
952
+ "f1-score": 0.9249708404168704,
953
+ "precision": 0.9259529269856172,
954
+ "recall": 0.9244833069098758,
955
+ "support": 29841.0
956
+ },
957
+ "eval_runtime": 4.6509,
958
+ "eval_samples_per_second": 17.201,
959
+ "eval_steps_per_second": 2.15,
960
+ "eval_weighted avg": {
961
+ "f1-score": 0.9459221443892064,
962
+ "precision": 0.9460143413566277,
963
+ "recall": 0.9462484501189639,
964
+ "support": 29841.0
965
+ },
966
+ "step": 1944
967
+ },
968
+ {
969
+ "epoch": 24.69,
970
+ "grad_norm": 0.04072243347764015,
971
+ "learning_rate": 1.0123456790123458e-05,
972
+ "loss": 0.0014,
973
+ "step": 2000
974
+ },
975
+ {
976
+ "epoch": 25.0,
977
+ "eval_B": {
978
+ "f1-score": 0.8943768996960487,
979
+ "precision": 0.8803290949887809,
980
+ "recall": 0.9088803088803089,
981
+ "support": 1295.0
982
+ },
983
+ "eval_I": {
984
+ "f1-score": 0.9629187495336865,
985
+ "precision": 0.9610305798252582,
986
+ "recall": 0.9648143533516073,
987
+ "support": 20065.0
988
+ },
989
+ "eval_O": {
990
+ "f1-score": 0.9089721512974288,
991
+ "precision": 0.9155502392344498,
992
+ "recall": 0.9024879141610659,
993
+ "support": 8481.0
994
+ },
995
+ "eval_accuracy": 0.9446734358768137,
996
+ "eval_loss": 0.48565673828125,
997
+ "eval_macro avg": {
998
+ "f1-score": 0.922089266842388,
999
+ "precision": 0.9189699713494962,
1000
+ "recall": 0.925394192130994,
1001
+ "support": 29841.0
1002
+ },
1003
+ "eval_runtime": 4.6314,
1004
+ "eval_samples_per_second": 17.274,
1005
+ "eval_steps_per_second": 2.159,
1006
+ "eval_weighted avg": {
1007
+ "f1-score": 0.9446122988389564,
1008
+ "precision": 0.944602605179171,
1009
+ "recall": 0.9446734358768137,
1010
+ "support": 29841.0
1011
+ },
1012
+ "step": 2025
1013
+ },
1014
+ {
1015
+ "epoch": 26.0,
1016
+ "eval_B": {
1017
+ "f1-score": 0.897679726131609,
1018
+ "precision": 0.8845577211394303,
1019
+ "recall": 0.9111969111969112,
1020
+ "support": 1295.0
1021
+ },
1022
+ "eval_I": {
1023
+ "f1-score": 0.962990450744644,
1024
+ "precision": 0.9560839023431743,
1025
+ "recall": 0.9699975080986792,
1026
+ "support": 20065.0
1027
+ },
1028
+ "eval_O": {
1029
+ "f1-score": 0.9077024833142927,
1030
+ "precision": 0.9261349693251534,
1031
+ "recall": 0.8899893880438627,
1032
+ "support": 8481.0
1033
+ },
1034
+ "eval_accuracy": 0.9447069468181362,
1035
+ "eval_loss": 0.4857862889766693,
1036
+ "eval_macro avg": {
1037
+ "f1-score": 0.9227908867301818,
1038
+ "precision": 0.9222588642692527,
1039
+ "recall": 0.9237279357798177,
1040
+ "support": 29841.0
1041
+ },
1042
+ "eval_runtime": 4.7085,
1043
+ "eval_samples_per_second": 16.991,
1044
+ "eval_steps_per_second": 2.124,
1045
+ "eval_weighted avg": {
1046
+ "f1-score": 0.94444299455515,
1047
+ "precision": 0.9444682290887698,
1048
+ "recall": 0.9447069468181362,
1049
+ "support": 29841.0
1050
+ },
1051
+ "step": 2106
1052
+ },
1053
+ {
1054
+ "epoch": 27.0,
1055
+ "eval_B": {
1056
+ "f1-score": 0.8986280487804879,
1057
+ "precision": 0.8871331828442438,
1058
+ "recall": 0.9104247104247104,
1059
+ "support": 1295.0
1060
+ },
1061
+ "eval_I": {
1062
+ "f1-score": 0.9620497586334943,
1063
+ "precision": 0.9557796360059027,
1064
+ "recall": 0.9684026912534264,
1065
+ "support": 20065.0
1066
+ },
1067
+ "eval_O": {
1068
+ "f1-score": 0.9053591790193843,
1069
+ "precision": 0.9219017355169885,
1070
+ "recall": 0.8893998349251268,
1071
+ "support": 8481.0
1072
+ },
1073
+ "eval_accuracy": 0.9434335310478872,
1074
+ "eval_loss": 0.4996240735054016,
1075
+ "eval_macro avg": {
1076
+ "f1-score": 0.9220123288111223,
1077
+ "precision": 0.9216048514557116,
1078
+ "recall": 0.9227424122010879,
1079
+ "support": 29841.0
1080
+ },
1081
+ "eval_runtime": 4.6631,
1082
+ "eval_samples_per_second": 17.156,
1083
+ "eval_steps_per_second": 2.145,
1084
+ "eval_weighted avg": {
1085
+ "f1-score": 0.9431856481825407,
1086
+ "precision": 0.9431722960745722,
1087
+ "recall": 0.9434335310478872,
1088
+ "support": 29841.0
1089
+ },
1090
+ "step": 2187
1091
+ },
1092
+ {
1093
+ "epoch": 28.0,
1094
+ "eval_B": {
1095
+ "f1-score": 0.8919434898816342,
1096
+ "precision": 0.8821752265861027,
1097
+ "recall": 0.901930501930502,
1098
+ "support": 1295.0
1099
+ },
1100
+ "eval_I": {
1101
+ "f1-score": 0.9604717405427016,
1102
+ "precision": 0.9630706017938568,
1103
+ "recall": 0.9578868676800398,
1104
+ "support": 20065.0
1105
+ },
1106
+ "eval_O": {
1107
+ "f1-score": 0.9045243823719267,
1108
+ "precision": 0.9003504672897197,
1109
+ "recall": 0.9087371772196675,
1110
+ "support": 8481.0
1111
+ },
1112
+ "eval_accuracy": 0.9414898964511913,
1113
+ "eval_loss": 0.4974842071533203,
1114
+ "eval_macro avg": {
1115
+ "f1-score": 0.9189798709320876,
1116
+ "precision": 0.9151987652232264,
1117
+ "recall": 0.9228515156100698,
1118
+ "support": 29841.0
1119
+ },
1120
+ "eval_runtime": 4.6318,
1121
+ "eval_samples_per_second": 17.272,
1122
+ "eval_steps_per_second": 2.159,
1123
+ "eval_weighted avg": {
1124
+ "f1-score": 0.941597251442054,
1125
+ "precision": 0.9417345550251954,
1126
+ "recall": 0.9414898964511913,
1127
+ "support": 29841.0
1128
+ },
1129
+ "step": 2268
1130
+ },
1131
+ {
1132
+ "epoch": 29.0,
1133
+ "eval_B": {
1134
+ "f1-score": 0.9026415094339624,
1135
+ "precision": 0.8826568265682657,
1136
+ "recall": 0.9235521235521236,
1137
+ "support": 1295.0
1138
+ },
1139
+ "eval_I": {
1140
+ "f1-score": 0.963459587838005,
1141
+ "precision": 0.9565730005895068,
1142
+ "recall": 0.9704460503364066,
1143
+ "support": 20065.0
1144
+ },
1145
+ "eval_O": {
1146
+ "f1-score": 0.9098789958461262,
1147
+ "precision": 0.929520295202952,
1148
+ "recall": 0.8910505836575876,
1149
+ "support": 8481.0
1150
+ },
1151
+ "eval_accuracy": 0.9458463188230958,
1152
+ "eval_loss": 0.4864692687988281,
1153
+ "eval_macro avg": {
1154
+ "f1-score": 0.9253266977060312,
1155
+ "precision": 0.9229167074535748,
1156
+ "recall": 0.9283495858487059,
1157
+ "support": 29841.0
1158
+ },
1159
+ "eval_runtime": 4.6138,
1160
+ "eval_samples_per_second": 17.339,
1161
+ "eval_steps_per_second": 2.167,
1162
+ "eval_weighted avg": {
1163
+ "f1-score": 0.9455923443737659,
1164
+ "precision": 0.9456767357277099,
1165
+ "recall": 0.9458463188230958,
1166
+ "support": 29841.0
1167
+ },
1168
+ "step": 2349
1169
+ },
1170
+ {
1171
+ "epoch": 30.0,
1172
+ "eval_B": {
1173
+ "f1-score": 0.8974650018917897,
1174
+ "precision": 0.8798219584569733,
1175
+ "recall": 0.9158301158301159,
1176
+ "support": 1295.0
1177
+ },
1178
+ "eval_I": {
1179
+ "f1-score": 0.9618674357267865,
1180
+ "precision": 0.9515704252828716,
1181
+ "recall": 0.9723897333665587,
1182
+ "support": 20065.0
1183
+ },
1184
+ "eval_O": {
1185
+ "f1-score": 0.9045537340619307,
1186
+ "precision": 0.9324070597070973,
1187
+ "recall": 0.87831623629289,
1188
+ "support": 8481.0
1189
+ },
1190
+ "eval_accuracy": 0.9431989544586308,
1191
+ "eval_loss": 0.5332404375076294,
1192
+ "eval_macro avg": {
1193
+ "f1-score": 0.921295390560169,
1194
+ "precision": 0.9212664811489807,
1195
+ "recall": 0.9221786951631882,
1196
+ "support": 29841.0
1197
+ },
1198
+ "eval_runtime": 4.6482,
1199
+ "eval_samples_per_second": 17.211,
1200
+ "eval_steps_per_second": 2.151,
1201
+ "eval_weighted avg": {
1202
+ "f1-score": 0.9427836699134436,
1203
+ "precision": 0.9430104317173852,
1204
+ "recall": 0.9431989544586308,
1205
+ "support": 29841.0
1206
+ },
1207
+ "step": 2430
1208
+ },
1209
+ {
1210
+ "epoch": 30.86,
1211
+ "grad_norm": 0.05446575954556465,
1212
+ "learning_rate": 7.654320987654322e-06,
1213
+ "loss": 0.0008,
1214
+ "step": 2500
1215
+ },
1216
+ {
1217
+ "epoch": 31.0,
1218
+ "eval_B": {
1219
+ "f1-score": 0.9002674818494459,
1220
+ "precision": 0.8910741301059002,
1221
+ "recall": 0.9096525096525097,
1222
+ "support": 1295.0
1223
+ },
1224
+ "eval_I": {
1225
+ "f1-score": 0.9637251998815517,
1226
+ "precision": 0.9544454763184906,
1227
+ "recall": 0.9731871417891852,
1228
+ "support": 20065.0
1229
+ },
1230
+ "eval_O": {
1231
+ "f1-score": 0.9096185236684602,
1232
+ "precision": 0.9333746898263028,
1233
+ "recall": 0.8870416224501828,
1234
+ "support": 8481.0
1235
+ },
1236
+ "eval_accuracy": 0.9459468516470627,
1237
+ "eval_loss": 0.512050986289978,
1238
+ "eval_macro avg": {
1239
+ "f1-score": 0.9245370684664859,
1240
+ "precision": 0.9262980987502312,
1241
+ "recall": 0.9232937579639593,
1242
+ "support": 29841.0
1243
+ },
1244
+ "eval_runtime": 4.6554,
1245
+ "eval_samples_per_second": 17.184,
1246
+ "eval_steps_per_second": 2.148,
1247
+ "eval_weighted avg": {
1248
+ "f1-score": 0.9455938884035581,
1249
+ "precision": 0.945706920855016,
1250
+ "recall": 0.9459468516470627,
1251
+ "support": 29841.0
1252
+ },
1253
+ "step": 2511
1254
+ },
1255
+ {
1256
+ "epoch": 32.0,
1257
+ "eval_B": {
1258
+ "f1-score": 0.8938120702826584,
1259
+ "precision": 0.8843537414965986,
1260
+ "recall": 0.9034749034749034,
1261
+ "support": 1295.0
1262
+ },
1263
+ "eval_I": {
1264
+ "f1-score": 0.9597342259079783,
1265
+ "precision": 0.962041163803896,
1266
+ "recall": 0.9574383254423124,
1267
+ "support": 20065.0
1268
+ },
1269
+ "eval_O": {
1270
+ "f1-score": 0.903112155020552,
1271
+ "precision": 0.8995204117440636,
1272
+ "recall": 0.9067326966159651,
1273
+ "support": 8481.0
1274
+ },
1275
+ "eval_accuracy": 0.9406856338594551,
1276
+ "eval_loss": 0.5488592982292175,
1277
+ "eval_macro avg": {
1278
+ "f1-score": 0.9188861504037296,
1279
+ "precision": 0.9153051056815195,
1280
+ "recall": 0.9225486418443937,
1281
+ "support": 29841.0
1282
+ },
1283
+ "eval_runtime": 4.6459,
1284
+ "eval_samples_per_second": 17.22,
1285
+ "eval_steps_per_second": 2.152,
1286
+ "eval_weighted avg": {
1287
+ "f1-score": 0.9407810750507332,
1288
+ "precision": 0.9409009972509189,
1289
+ "recall": 0.9406856338594551,
1290
+ "support": 29841.0
1291
+ },
1292
+ "step": 2592
1293
+ },
1294
+ {
1295
+ "epoch": 33.0,
1296
+ "eval_B": {
1297
+ "f1-score": 0.9014084507042254,
1298
+ "precision": 0.8888888888888888,
1299
+ "recall": 0.9142857142857143,
1300
+ "support": 1295.0
1301
+ },
1302
+ "eval_I": {
1303
+ "f1-score": 0.963593310640817,
1304
+ "precision": 0.9538994969966303,
1305
+ "recall": 0.9734861699476701,
1306
+ "support": 20065.0
1307
+ },
1308
+ "eval_O": {
1309
+ "f1-score": 0.9089808030036942,
1310
+ "precision": 0.9343874501992032,
1311
+ "recall": 0.8849192312227332,
1312
+ "support": 8481.0
1313
+ },
1314
+ "eval_accuracy": 0.9457457859991287,
1315
+ "eval_loss": 0.518332839012146,
1316
+ "eval_macro avg": {
1317
+ "f1-score": 0.9246608547829123,
1318
+ "precision": 0.9257252786949075,
1319
+ "recall": 0.9242303718187058,
1320
+ "support": 29841.0
1321
+ },
1322
+ "eval_runtime": 4.6621,
1323
+ "eval_samples_per_second": 17.16,
1324
+ "eval_steps_per_second": 2.145,
1325
+ "eval_weighted avg": {
1326
+ "f1-score": 0.945373476490208,
1327
+ "precision": 0.9455328066582199,
1328
+ "recall": 0.9457457859991287,
1329
+ "support": 29841.0
1330
+ },
1331
+ "step": 2673
1332
+ },
1333
+ {
1334
+ "epoch": 34.0,
1335
+ "eval_B": {
1336
+ "f1-score": 0.9016641452344931,
1337
+ "precision": 0.8836174944403261,
1338
+ "recall": 0.9204633204633205,
1339
+ "support": 1295.0
1340
+ },
1341
+ "eval_I": {
1342
+ "f1-score": 0.9637216189536032,
1343
+ "precision": 0.9545343436812516,
1344
+ "recall": 0.9730874657363568,
1345
+ "support": 20065.0
1346
+ },
1347
+ "eval_O": {
1348
+ "f1-score": 0.9093110546070954,
1349
+ "precision": 0.934428269254697,
1350
+ "recall": 0.8855087843414692,
1351
+ "support": 8481.0
1352
+ },
1353
+ "eval_accuracy": 0.9459133407057404,
1354
+ "eval_loss": 0.5136498808860779,
1355
+ "eval_macro avg": {
1356
+ "f1-score": 0.9248989395983972,
1357
+ "precision": 0.9241933691254248,
1358
+ "recall": 0.9263531901803822,
1359
+ "support": 29841.0
1360
+ },
1361
+ "eval_runtime": 4.6341,
1362
+ "eval_samples_per_second": 17.263,
1363
+ "eval_steps_per_second": 2.158,
1364
+ "eval_weighted avg": {
1365
+ "f1-score": 0.9455647064946046,
1366
+ "precision": 0.9457425157606522,
1367
+ "recall": 0.9459133407057404,
1368
+ "support": 29841.0
1369
+ },
1370
+ "step": 2754
1371
+ },
1372
+ {
1373
+ "epoch": 35.0,
1374
+ "eval_B": {
1375
+ "f1-score": 0.9019157088122605,
1376
+ "precision": 0.8950570342205323,
1377
+ "recall": 0.9088803088803089,
1378
+ "support": 1295.0
1379
+ },
1380
+ "eval_I": {
1381
+ "f1-score": 0.9609330316066944,
1382
+ "precision": 0.9589537423069288,
1383
+ "recall": 0.9629205083478695,
1384
+ "support": 20065.0
1385
+ },
1386
+ "eval_O": {
1387
+ "f1-score": 0.90479862388042,
1388
+ "precision": 0.9103604678920983,
1389
+ "recall": 0.8993043273198915,
1390
+ "support": 8481.0
1391
+ },
1392
+ "eval_accuracy": 0.9424952246908616,
1393
+ "eval_loss": 0.5342629551887512,
1394
+ "eval_macro avg": {
1395
+ "f1-score": 0.922549121433125,
1396
+ "precision": 0.9214570814731865,
1397
+ "recall": 0.9237017148493566,
1398
+ "support": 29841.0
1399
+ },
1400
+ "eval_runtime": 4.6532,
1401
+ "eval_samples_per_second": 17.193,
1402
+ "eval_steps_per_second": 2.149,
1403
+ "eval_weighted avg": {
1404
+ "f1-score": 0.9424181244338341,
1405
+ "precision": 0.9423703236117087,
1406
+ "recall": 0.9424952246908616,
1407
+ "support": 29841.0
1408
+ },
1409
+ "step": 2835
1410
+ },
1411
+ {
1412
+ "epoch": 36.0,
1413
+ "eval_B": {
1414
+ "f1-score": 0.8977099236641222,
1415
+ "precision": 0.8875471698113208,
1416
+ "recall": 0.9081081081081082,
1417
+ "support": 1295.0
1418
+ },
1419
+ "eval_I": {
1420
+ "f1-score": 0.9619323959030825,
1421
+ "precision": 0.9573501826438938,
1422
+ "recall": 0.9665586842761027,
1423
+ "support": 20065.0
1424
+ },
1425
+ "eval_O": {
1426
+ "f1-score": 0.9061473206284726,
1427
+ "precision": 0.9183821748607411,
1428
+ "recall": 0.894234170498762,
1429
+ "support": 8481.0
1430
+ },
1431
+ "eval_accuracy": 0.9434670419892095,
1432
+ "eval_loss": 0.5332928895950317,
1433
+ "eval_macro avg": {
1434
+ "f1-score": 0.9219298800652257,
1435
+ "precision": 0.9210931757719852,
1436
+ "recall": 0.9229669876276576,
1437
+ "support": 29841.0
1438
+ },
1439
+ "eval_runtime": 4.6471,
1440
+ "eval_samples_per_second": 17.215,
1441
+ "eval_steps_per_second": 2.152,
1442
+ "eval_weighted avg": {
1443
+ "f1-score": 0.943290885063854,
1444
+ "precision": 0.9432460113484581,
1445
+ "recall": 0.9434670419892095,
1446
+ "support": 29841.0
1447
+ },
1448
+ "step": 2916
1449
+ }
1450
+ ],
1451
+ "logging_steps": 500,
1452
+ "max_steps": 4050,
1453
+ "num_input_tokens_seen": 0,
1454
+ "num_train_epochs": 50,
1455
+ "save_steps": 500,
1456
+ "total_flos": 5176319498481600.0,
1457
+ "train_batch_size": 4,
1458
+ "trial_name": null,
1459
+ "trial_params": null
1460
+ }
checkpoint-2916/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d229d7bc9127d222c880edfd3fa9418a9ccede656fe27f7f0aa04990c9605be4
3
+ size 4856
checkpoint-2916/vocab.json ADDED
The diff for this file is too large to render. See raw diff