Hugging Face's logo

Theoreticallyhugo
/
longformer-spans

Token Classification
Transformers
Safetensors
longformer
Generated from Trainer
Eval Results
Inference Endpoints
Model card Files Community
Theoreticallyhugo commited on
Commit
305bf48
•
1 Parent(s): 98e8065

Training in progress, epoch 20, checkpoint

Browse files
Files changed (12) hide show
  1. checkpoint-1620/config.json +53 -0
  2. checkpoint-1620/merges.txt +0 -0
  3. checkpoint-1620/model.safetensors +3 -0
  4. checkpoint-1620/optimizer.pt +3 -0
  5. checkpoint-1620/rng_state.pth +3 -0
  6. checkpoint-1620/scheduler.pt +3 -0
  7. checkpoint-1620/special_tokens_map.json +15 -0
  8. checkpoint-1620/tokenizer.json +0 -0
  9. checkpoint-1620/tokenizer_config.json +57 -0
  10. checkpoint-1620/trainer_state.json +822 -0
  11. checkpoint-1620/training_args.bin +3 -0
  12. checkpoint-1620/vocab.json +0 -0
checkpoint-1620/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-1620/merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-1620/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6e95308ba1edab58536c99b771ec05a42b646b745527f5e670545125f2f04c39
3
+ size 592318676
checkpoint-1620/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:760608a3f6988b3619b7221803bd6e4819492cc7e82c9f0ad68b7d8af9831359
3
+ size 1014645498
checkpoint-1620/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9fcb986af28d68a2426af950db35feed25413b0fcefbd85f4e0c31a98393e213
3
+ size 14244
checkpoint-1620/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:62fa97b6a73c620ff6f69ca713c8a45c421ed3fdb8b0992517caf27356998de4
3
+ size 1064
checkpoint-1620/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-1620/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-1620/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-1620/trainer_state.json ADDED
@@ -0,0 +1,822 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 20.0,
5
+ "eval_steps": 500,
6
+ "global_step": 1620,
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": 0.8087666523420713,
15
+ "precision": 0.8189730200174065,
16
+ "recall": 0.7988115449915111,
17
+ "support": 1178.0
18
+ },
19
+ "eval_I": {
20
+ "f1-score": 0.9374597034171502,
21
+ "precision": 0.9521418826739427,
22
+ "recall": 0.9232234509762421,
23
+ "support": 18899.0
24
+ },
25
+ "eval_O": {
26
+ "f1-score": 0.8881362400419788,
27
+ "precision": 0.863303347862376,
28
+ "recall": 0.9144400785854617,
29
+ "support": 10180.0
30
+ },
31
+ "eval_accuracy": 0.9154245298608586,
32
+ "eval_loss": 0.2228638380765915,
33
+ "eval_macro avg": {
34
+ "f1-score": 0.8781208652670669,
35
+ "precision": 0.8781394168512416,
36
+ "recall": 0.8788250248510717,
37
+ "support": 30257.0
38
+ },
39
+ "eval_runtime": 4.7167,
40
+ "eval_samples_per_second": 17.173,
41
+ "eval_steps_per_second": 2.332,
42
+ "eval_weighted avg": {
43
+ "f1-score": 0.9158543469268937,
44
+ "precision": 0.9170673807540184,
45
+ "recall": 0.9154245298608586,
46
+ "support": 30257.0
47
+ },
48
+ "step": 81
49
+ },
50
+ {
51
+ "epoch": 2.0,
52
+ "eval_B": {
53
+ "f1-score": 0.8401515151515152,
54
+ "precision": 0.7585499316005472,
55
+ "recall": 0.9414261460101867,
56
+ "support": 1178.0
57
+ },
58
+ "eval_I": {
59
+ "f1-score": 0.9568235603470944,
60
+ "precision": 0.9510219016256338,
61
+ "recall": 0.9626964389650247,
62
+ "support": 18899.0
63
+ },
64
+ "eval_O": {
65
+ "f1-score": 0.9185648054827655,
66
+ "precision": 0.9430877483443708,
67
+ "recall": 0.8952848722986247,
68
+ "support": 10180.0
69
+ },
70
+ "eval_accuracy": 0.9391876260038999,
71
+ "eval_loss": 0.16692574322223663,
72
+ "eval_macro avg": {
73
+ "f1-score": 0.905179960327125,
74
+ "precision": 0.8842198605235172,
75
+ "recall": 0.9331358190912787,
76
+ "support": 30257.0
77
+ },
78
+ "eval_runtime": 4.7116,
79
+ "eval_samples_per_second": 17.192,
80
+ "eval_steps_per_second": 2.335,
81
+ "eval_weighted avg": {
82
+ "f1-score": 0.9394089523635119,
83
+ "precision": 0.9408589092241132,
84
+ "recall": 0.9391876260038999,
85
+ "support": 30257.0
86
+ },
87
+ "step": 162
88
+ },
89
+ {
90
+ "epoch": 3.0,
91
+ "eval_B": {
92
+ "f1-score": 0.8612244897959184,
93
+ "precision": 0.8294025157232704,
94
+ "recall": 0.8955857385398981,
95
+ "support": 1178.0
96
+ },
97
+ "eval_I": {
98
+ "f1-score": 0.951106371634231,
99
+ "precision": 0.9584654236741712,
100
+ "recall": 0.9438594634636753,
101
+ "support": 18899.0
102
+ },
103
+ "eval_O": {
104
+ "f1-score": 0.9106743212999903,
105
+ "precision": 0.9021592442645074,
106
+ "recall": 0.9193516699410609,
107
+ "support": 10180.0
108
+ },
109
+ "eval_accuracy": 0.9337343424662061,
110
+ "eval_loss": 0.1742253601551056,
111
+ "eval_macro avg": {
112
+ "f1-score": 0.9076683942433799,
113
+ "precision": 0.8966757278873163,
114
+ "recall": 0.9195989573148781,
115
+ "support": 30257.0
116
+ },
117
+ "eval_runtime": 4.7094,
118
+ "eval_samples_per_second": 17.2,
119
+ "eval_steps_per_second": 2.336,
120
+ "eval_weighted avg": {
121
+ "f1-score": 0.9340035812317422,
122
+ "precision": 0.9344963252190521,
123
+ "recall": 0.9337343424662061,
124
+ "support": 30257.0
125
+ },
126
+ "step": 243
127
+ },
128
+ {
129
+ "epoch": 4.0,
130
+ "eval_B": {
131
+ "f1-score": 0.8717948717948719,
132
+ "precision": 0.8254931714719271,
133
+ "recall": 0.9235993208828522,
134
+ "support": 1178.0
135
+ },
136
+ "eval_I": {
137
+ "f1-score": 0.9561271586371415,
138
+ "precision": 0.9374586871408959,
139
+ "recall": 0.9755542621302714,
140
+ "support": 18899.0
141
+ },
142
+ "eval_O": {
143
+ "f1-score": 0.9117828500925355,
144
+ "precision": 0.956427955133736,
145
+ "recall": 0.8711198428290766,
146
+ "support": 10180.0
147
+ },
148
+ "eval_accuracy": 0.9383944211256899,
149
+ "eval_loss": 0.18792153894901276,
150
+ "eval_macro avg": {
151
+ "f1-score": 0.9132349601748496,
152
+ "precision": 0.9064599379155197,
153
+ "recall": 0.9234244752807333,
154
+ "support": 30257.0
155
+ },
156
+ "eval_runtime": 4.7342,
157
+ "eval_samples_per_second": 17.11,
158
+ "eval_steps_per_second": 2.324,
159
+ "eval_weighted avg": {
160
+ "f1-score": 0.9379241479327001,
161
+ "precision": 0.9394817486046585,
162
+ "recall": 0.9383944211256899,
163
+ "support": 30257.0
164
+ },
165
+ "step": 324
166
+ },
167
+ {
168
+ "epoch": 5.0,
169
+ "eval_B": {
170
+ "f1-score": 0.8742949234488315,
171
+ "precision": 0.8320552147239264,
172
+ "recall": 0.9210526315789473,
173
+ "support": 1178.0
174
+ },
175
+ "eval_I": {
176
+ "f1-score": 0.9583957528450158,
177
+ "precision": 0.9521148825065274,
178
+ "recall": 0.9647600402137679,
179
+ "support": 18899.0
180
+ },
181
+ "eval_O": {
182
+ "f1-score": 0.9198818996146724,
183
+ "precision": 0.9375701315923697,
184
+ "recall": 0.9028487229862475,
185
+ "support": 10180.0
186
+ },
187
+ "eval_accuracy": 0.9422282447037049,
188
+ "eval_loss": 0.19990424811840057,
189
+ "eval_macro avg": {
190
+ "f1-score": 0.9175241919695066,
191
+ "precision": 0.9072467429409411,
192
+ "recall": 0.9295537982596542,
193
+ "support": 30257.0
194
+ },
195
+ "eval_runtime": 4.7101,
196
+ "eval_samples_per_second": 17.197,
197
+ "eval_steps_per_second": 2.335,
198
+ "eval_weighted avg": {
199
+ "f1-score": 0.9421634164298524,
200
+ "precision": 0.9425469857238316,
201
+ "recall": 0.9422282447037049,
202
+ "support": 30257.0
203
+ },
204
+ "step": 405
205
+ },
206
+ {
207
+ "epoch": 6.0,
208
+ "eval_B": {
209
+ "f1-score": 0.8808247422680412,
210
+ "precision": 0.8564554931836408,
211
+ "recall": 0.9066213921901528,
212
+ "support": 1178.0
213
+ },
214
+ "eval_I": {
215
+ "f1-score": 0.957406827435014,
216
+ "precision": 0.9446361435855178,
217
+ "recall": 0.9705275411397428,
218
+ "support": 18899.0
219
+ },
220
+ "eval_O": {
221
+ "f1-score": 0.9163000050574014,
222
+ "precision": 0.9443344105076619,
223
+ "recall": 0.8898821218074656,
224
+ "support": 10180.0
225
+ },
226
+ "eval_accuracy": 0.9409062365733549,
227
+ "eval_loss": 0.20615626871585846,
228
+ "eval_macro avg": {
229
+ "f1-score": 0.9181771915868189,
230
+ "precision": 0.9151420157589403,
231
+ "recall": 0.922343685045787,
232
+ "support": 30257.0
233
+ },
234
+ "eval_runtime": 4.6949,
235
+ "eval_samples_per_second": 17.253,
236
+ "eval_steps_per_second": 2.343,
237
+ "eval_weighted avg": {
238
+ "f1-score": 0.9405948120953971,
239
+ "precision": 0.941101475610967,
240
+ "recall": 0.9409062365733549,
241
+ "support": 30257.0
242
+ },
243
+ "step": 486
244
+ },
245
+ {
246
+ "epoch": 6.17,
247
+ "grad_norm": 2.4104158878326416,
248
+ "learning_rate": 1.7530864197530865e-05,
249
+ "loss": 0.1662,
250
+ "step": 500
251
+ },
252
+ {
253
+ "epoch": 7.0,
254
+ "eval_B": {
255
+ "f1-score": 0.8836820083682009,
256
+ "precision": 0.8712871287128713,
257
+ "recall": 0.8964346349745331,
258
+ "support": 1178.0
259
+ },
260
+ "eval_I": {
261
+ "f1-score": 0.9614174269469787,
262
+ "precision": 0.9609854091774159,
263
+ "recall": 0.9618498333245146,
264
+ "support": 18899.0
265
+ },
266
+ "eval_O": {
267
+ "f1-score": 0.9271751440248166,
268
+ "precision": 0.9295093296475466,
269
+ "recall": 0.924852652259332,
270
+ "support": 10180.0
271
+ },
272
+ "eval_accuracy": 0.9468552731599299,
273
+ "eval_loss": 0.22925782203674316,
274
+ "eval_macro avg": {
275
+ "f1-score": 0.9240915264466655,
276
+ "precision": 0.9205939558459445,
277
+ "recall": 0.9277123735194599,
278
+ "support": 30257.0
279
+ },
280
+ "eval_runtime": 4.7288,
281
+ "eval_samples_per_second": 17.129,
282
+ "eval_steps_per_second": 2.326,
283
+ "eval_weighted avg": {
284
+ "f1-score": 0.9468700903559945,
285
+ "precision": 0.9469030129054358,
286
+ "recall": 0.9468552731599299,
287
+ "support": 30257.0
288
+ },
289
+ "step": 567
290
+ },
291
+ {
292
+ "epoch": 8.0,
293
+ "eval_B": {
294
+ "f1-score": 0.8803630363036303,
295
+ "precision": 0.8563402889245586,
296
+ "recall": 0.9057724957555179,
297
+ "support": 1178.0
298
+ },
299
+ "eval_I": {
300
+ "f1-score": 0.9580651967507121,
301
+ "precision": 0.9550928116947994,
302
+ "recall": 0.9610561405365363,
303
+ "support": 18899.0
304
+ },
305
+ "eval_O": {
306
+ "f1-score": 0.9208882720333101,
307
+ "precision": 0.9294576746047628,
308
+ "recall": 0.912475442043222,
309
+ "support": 10180.0
310
+ },
311
+ "eval_accuracy": 0.9425587467362925,
312
+ "eval_loss": 0.22973769903182983,
313
+ "eval_macro avg": {
314
+ "f1-score": 0.9197721683625509,
315
+ "precision": 0.9136302584080402,
316
+ "recall": 0.9264346927784253,
317
+ "support": 30257.0
318
+ },
319
+ "eval_runtime": 4.7191,
320
+ "eval_samples_per_second": 17.164,
321
+ "eval_steps_per_second": 2.331,
322
+ "eval_weighted avg": {
323
+ "f1-score": 0.9425317916335554,
324
+ "precision": 0.942623096673485,
325
+ "recall": 0.9425587467362925,
326
+ "support": 30257.0
327
+ },
328
+ "step": 648
329
+ },
330
+ {
331
+ "epoch": 9.0,
332
+ "eval_B": {
333
+ "f1-score": 0.8885191347753745,
334
+ "precision": 0.8711256117455138,
335
+ "recall": 0.9066213921901528,
336
+ "support": 1178.0
337
+ },
338
+ "eval_I": {
339
+ "f1-score": 0.9620326487625066,
340
+ "precision": 0.9574445783763954,
341
+ "recall": 0.9666649029049156,
342
+ "support": 18899.0
343
+ },
344
+ "eval_O": {
345
+ "f1-score": 0.9275707898658718,
346
+ "precision": 0.9382914572864322,
347
+ "recall": 0.9170923379174852,
348
+ "support": 10180.0
349
+ },
350
+ "eval_accuracy": 0.9476484780381399,
351
+ "eval_loss": 0.24401217699050903,
352
+ "eval_macro avg": {
353
+ "f1-score": 0.9260408578012509,
354
+ "precision": 0.9222872158027805,
355
+ "recall": 0.9301262110041845,
356
+ "support": 30257.0
357
+ },
358
+ "eval_runtime": 4.701,
359
+ "eval_samples_per_second": 17.23,
360
+ "eval_steps_per_second": 2.34,
361
+ "eval_weighted avg": {
362
+ "f1-score": 0.94757580760031,
363
+ "precision": 0.9476398219436029,
364
+ "recall": 0.9476484780381399,
365
+ "support": 30257.0
366
+ },
367
+ "step": 729
368
+ },
369
+ {
370
+ "epoch": 10.0,
371
+ "eval_B": {
372
+ "f1-score": 0.875959595959596,
373
+ "precision": 0.8357748650732459,
374
+ "recall": 0.9202037351443124,
375
+ "support": 1178.0
376
+ },
377
+ "eval_I": {
378
+ "f1-score": 0.955652945166171,
379
+ "precision": 0.9424749163879599,
380
+ "recall": 0.9692047198264458,
381
+ "support": 18899.0
382
+ },
383
+ "eval_O": {
384
+ "f1-score": 0.9127632580563309,
385
+ "precision": 0.9441469816272966,
386
+ "recall": 0.8833988212180747,
387
+ "support": 10180.0
388
+ },
389
+ "eval_accuracy": 0.9384274713289487,
390
+ "eval_loss": 0.29471203684806824,
391
+ "eval_macro avg": {
392
+ "f1-score": 0.9147919330606994,
393
+ "precision": 0.9074655876961675,
394
+ "recall": 0.9242690920629443,
395
+ "support": 30257.0
396
+ },
397
+ "eval_runtime": 4.6985,
398
+ "eval_samples_per_second": 17.24,
399
+ "eval_steps_per_second": 2.341,
400
+ "eval_weighted avg": {
401
+ "f1-score": 0.9381199518045186,
402
+ "precision": 0.9388833165494999,
403
+ "recall": 0.9384274713289487,
404
+ "support": 30257.0
405
+ },
406
+ "step": 810
407
+ },
408
+ {
409
+ "epoch": 11.0,
410
+ "eval_B": {
411
+ "f1-score": 0.875515251442704,
412
+ "precision": 0.8509615384615384,
413
+ "recall": 0.9015280135823429,
414
+ "support": 1178.0
415
+ },
416
+ "eval_I": {
417
+ "f1-score": 0.9575287538250502,
418
+ "precision": 0.9547582723972855,
419
+ "recall": 0.96031536060109,
420
+ "support": 18899.0
421
+ },
422
+ "eval_O": {
423
+ "f1-score": 0.9195242814667988,
424
+ "precision": 0.9278,
425
+ "recall": 0.9113948919449901,
426
+ "support": 10180.0
427
+ },
428
+ "eval_accuracy": 0.9415672406385299,
429
+ "eval_loss": 0.30941879749298096,
430
+ "eval_macro avg": {
431
+ "f1-score": 0.9175227622448511,
432
+ "precision": 0.9111732702862746,
433
+ "recall": 0.9244127553761411,
434
+ "support": 30257.0
435
+ },
436
+ "eval_runtime": 4.7017,
437
+ "eval_samples_per_second": 17.228,
438
+ "eval_steps_per_second": 2.34,
439
+ "eval_weighted avg": {
440
+ "f1-score": 0.9415490653426031,
441
+ "precision": 0.9416470001105195,
442
+ "recall": 0.9415672406385299,
443
+ "support": 30257.0
444
+ },
445
+ "step": 891
446
+ },
447
+ {
448
+ "epoch": 12.0,
449
+ "eval_B": {
450
+ "f1-score": 0.8818181818181817,
451
+ "precision": 0.8590982286634461,
452
+ "recall": 0.9057724957555179,
453
+ "support": 1178.0
454
+ },
455
+ "eval_I": {
456
+ "f1-score": 0.9542487087386278,
457
+ "precision": 0.9457436856875585,
458
+ "recall": 0.9629080903751521,
459
+ "support": 18899.0
460
+ },
461
+ "eval_O": {
462
+ "f1-score": 0.910740239562973,
463
+ "precision": 0.9297042873222142,
464
+ "recall": 0.8925343811394892,
465
+ "support": 10180.0
466
+ },
467
+ "eval_accuracy": 0.9370063125888224,
468
+ "eval_loss": 0.28610122203826904,
469
+ "eval_macro avg": {
470
+ "f1-score": 0.9156023767065942,
471
+ "precision": 0.9115154005577396,
472
+ "recall": 0.9204049890900531,
473
+ "support": 30257.0
474
+ },
475
+ "eval_runtime": 4.6911,
476
+ "eval_samples_per_second": 17.267,
477
+ "eval_steps_per_second": 2.345,
478
+ "eval_weighted avg": {
479
+ "f1-score": 0.9367902899621314,
480
+ "precision": 0.9369738332985704,
481
+ "recall": 0.9370063125888224,
482
+ "support": 30257.0
483
+ },
484
+ "step": 972
485
+ },
486
+ {
487
+ "epoch": 12.35,
488
+ "grad_norm": 0.9797911643981934,
489
+ "learning_rate": 1.506172839506173e-05,
490
+ "loss": 0.0233,
491
+ "step": 1000
492
+ },
493
+ {
494
+ "epoch": 13.0,
495
+ "eval_B": {
496
+ "f1-score": 0.8875154511742892,
497
+ "precision": 0.8622898318654924,
498
+ "recall": 0.9142614601018676,
499
+ "support": 1178.0
500
+ },
501
+ "eval_I": {
502
+ "f1-score": 0.9602009576889865,
503
+ "precision": 0.9497877627083549,
504
+ "recall": 0.9708450182549341,
505
+ "support": 18899.0
506
+ },
507
+ "eval_O": {
508
+ "f1-score": 0.9229994967287367,
509
+ "precision": 0.9463364293085655,
510
+ "recall": 0.9007858546168959,
511
+ "support": 10180.0
512
+ },
513
+ "eval_accuracy": 0.9450705621839575,
514
+ "eval_loss": 0.3311145305633545,
515
+ "eval_macro avg": {
516
+ "f1-score": 0.9235719685306708,
517
+ "precision": 0.9194713412941375,
518
+ "recall": 0.9286307776578991,
519
+ "support": 30257.0
520
+ },
521
+ "eval_runtime": 4.7187,
522
+ "eval_samples_per_second": 17.166,
523
+ "eval_steps_per_second": 2.331,
524
+ "eval_weighted avg": {
525
+ "f1-score": 0.9448546114137558,
526
+ "precision": 0.9452199887538072,
527
+ "recall": 0.9450705621839575,
528
+ "support": 30257.0
529
+ },
530
+ "step": 1053
531
+ },
532
+ {
533
+ "epoch": 14.0,
534
+ "eval_B": {
535
+ "f1-score": 0.8758762886597938,
536
+ "precision": 0.8516439454691259,
537
+ "recall": 0.9015280135823429,
538
+ "support": 1178.0
539
+ },
540
+ "eval_I": {
541
+ "f1-score": 0.9588910891089109,
542
+ "precision": 0.9569456155143339,
543
+ "recall": 0.9608444891264089,
544
+ "support": 18899.0
545
+ },
546
+ "eval_O": {
547
+ "f1-score": 0.9223310576827941,
548
+ "precision": 0.9290412597169623,
549
+ "recall": 0.9157170923379175,
550
+ "support": 10180.0
551
+ },
552
+ "eval_accuracy": 0.9433519516145025,
553
+ "eval_loss": 0.32673054933547974,
554
+ "eval_macro avg": {
555
+ "f1-score": 0.9190328118171663,
556
+ "precision": 0.9125436069001407,
557
+ "recall": 0.9260298650155564,
558
+ "support": 30257.0
559
+ },
560
+ "eval_runtime": 4.7171,
561
+ "eval_samples_per_second": 17.171,
562
+ "eval_steps_per_second": 2.332,
563
+ "eval_weighted avg": {
564
+ "f1-score": 0.9433584006451857,
565
+ "precision": 0.943457440568685,
566
+ "recall": 0.9433519516145025,
567
+ "support": 30257.0
568
+ },
569
+ "step": 1134
570
+ },
571
+ {
572
+ "epoch": 15.0,
573
+ "eval_B": {
574
+ "f1-score": 0.8844750709363599,
575
+ "precision": 0.8463925523661753,
576
+ "recall": 0.9261460101867572,
577
+ "support": 1178.0
578
+ },
579
+ "eval_I": {
580
+ "f1-score": 0.9554282158321882,
581
+ "precision": 0.9367500508440105,
582
+ "recall": 0.974866395047357,
583
+ "support": 18899.0
584
+ },
585
+ "eval_O": {
586
+ "f1-score": 0.9105749486652978,
587
+ "precision": 0.9536559139784946,
588
+ "recall": 0.8712180746561886,
589
+ "support": 10180.0
590
+ },
591
+ "eval_accuracy": 0.9380969692963612,
592
+ "eval_loss": 0.4276476800441742,
593
+ "eval_macro avg": {
594
+ "f1-score": 0.9168260784779486,
595
+ "precision": 0.9122661723962269,
596
+ "recall": 0.924076826630101,
597
+ "support": 30257.0
598
+ },
599
+ "eval_runtime": 4.6973,
600
+ "eval_samples_per_second": 17.244,
601
+ "eval_steps_per_second": 2.342,
602
+ "eval_weighted avg": {
603
+ "f1-score": 0.9375748574540862,
604
+ "precision": 0.9389201454833387,
605
+ "recall": 0.9380969692963612,
606
+ "support": 30257.0
607
+ },
608
+ "step": 1215
609
+ },
610
+ {
611
+ "epoch": 16.0,
612
+ "eval_B": {
613
+ "f1-score": 0.8840336134453782,
614
+ "precision": 0.8752079866888519,
615
+ "recall": 0.8930390492359932,
616
+ "support": 1178.0
617
+ },
618
+ "eval_I": {
619
+ "f1-score": 0.9589266737513285,
620
+ "precision": 0.9629688917346994,
621
+ "recall": 0.9549182496428382,
622
+ "support": 18899.0
623
+ },
624
+ "eval_O": {
625
+ "f1-score": 0.923489801893237,
626
+ "precision": 0.9174907892185379,
627
+ "recall": 0.9295677799607073,
628
+ "support": 10180.0
629
+ },
630
+ "eval_accuracy": 0.9439799054764186,
631
+ "eval_loss": 0.3447544574737549,
632
+ "eval_macro avg": {
633
+ "f1-score": 0.9221500296966479,
634
+ "precision": 0.9185558892140296,
635
+ "recall": 0.9258416929465129,
636
+ "support": 30257.0
637
+ },
638
+ "eval_runtime": 4.7359,
639
+ "eval_samples_per_second": 17.103,
640
+ "eval_steps_per_second": 2.323,
641
+ "eval_weighted avg": {
642
+ "f1-score": 0.9440880783665984,
643
+ "precision": 0.9442509279657028,
644
+ "recall": 0.9439799054764186,
645
+ "support": 30257.0
646
+ },
647
+ "step": 1296
648
+ },
649
+ {
650
+ "epoch": 17.0,
651
+ "eval_B": {
652
+ "f1-score": 0.8751040799333889,
653
+ "precision": 0.8586601307189542,
654
+ "recall": 0.8921901528013583,
655
+ "support": 1178.0
656
+ },
657
+ "eval_I": {
658
+ "f1-score": 0.9534606521336053,
659
+ "precision": 0.954598493688342,
660
+ "recall": 0.9523255198687761,
661
+ "support": 18899.0
662
+ },
663
+ "eval_O": {
664
+ "f1-score": 0.9118326047448303,
665
+ "precision": 0.9118773946360154,
666
+ "recall": 0.9117878192534381,
667
+ "support": 10180.0
668
+ },
669
+ "eval_accuracy": 0.9363453085236474,
670
+ "eval_loss": 0.3998367488384247,
671
+ "eval_macro avg": {
672
+ "f1-score": 0.9134657789372748,
673
+ "precision": 0.9083786730144373,
674
+ "recall": 0.9187678306411908,
675
+ "support": 30257.0
676
+ },
677
+ "eval_runtime": 4.7114,
678
+ "eval_samples_per_second": 17.192,
679
+ "eval_steps_per_second": 2.335,
680
+ "eval_weighted avg": {
681
+ "f1-score": 0.9364041837306049,
682
+ "precision": 0.936489752572877,
683
+ "recall": 0.9363453085236474,
684
+ "support": 30257.0
685
+ },
686
+ "step": 1377
687
+ },
688
+ {
689
+ "epoch": 18.0,
690
+ "eval_B": {
691
+ "f1-score": 0.8807644370585792,
692
+ "precision": 0.8624898291293734,
693
+ "recall": 0.8998302207130731,
694
+ "support": 1178.0
695
+ },
696
+ "eval_I": {
697
+ "f1-score": 0.958089514470979,
698
+ "precision": 0.9545669415410474,
699
+ "recall": 0.961638181914387,
700
+ "support": 18899.0
701
+ },
702
+ "eval_O": {
703
+ "f1-score": 0.9206207546234321,
704
+ "precision": 0.9294223646010612,
705
+ "recall": 0.9119842829076621,
706
+ "support": 10180.0
707
+ },
708
+ "eval_accuracy": 0.9425256965330336,
709
+ "eval_loss": 0.3564019203186035,
710
+ "eval_macro avg": {
711
+ "f1-score": 0.9198249020509968,
712
+ "precision": 0.915493045090494,
713
+ "recall": 0.9244842285117074,
714
+ "support": 30257.0
715
+ },
716
+ "eval_runtime": 4.6986,
717
+ "eval_samples_per_second": 17.239,
718
+ "eval_steps_per_second": 2.341,
719
+ "eval_weighted avg": {
720
+ "f1-score": 0.9424726021386316,
721
+ "precision": 0.9425221706890129,
722
+ "recall": 0.9425256965330336,
723
+ "support": 30257.0
724
+ },
725
+ "step": 1458
726
+ },
727
+ {
728
+ "epoch": 18.52,
729
+ "grad_norm": 0.01765240728855133,
730
+ "learning_rate": 1.2592592592592593e-05,
731
+ "loss": 0.0067,
732
+ "step": 1500
733
+ },
734
+ {
735
+ "epoch": 19.0,
736
+ "eval_B": {
737
+ "f1-score": 0.8756097560975611,
738
+ "precision": 0.8400936037441498,
739
+ "recall": 0.9142614601018676,
740
+ "support": 1178.0
741
+ },
742
+ "eval_I": {
743
+ "f1-score": 0.9563048159235334,
744
+ "precision": 0.9492258770786633,
745
+ "recall": 0.9634901317530028,
746
+ "support": 18899.0
747
+ },
748
+ "eval_O": {
749
+ "f1-score": 0.9161826557180053,
750
+ "precision": 0.9343341503267973,
751
+ "recall": 0.8987229862475442,
752
+ "support": 10180.0
753
+ },
754
+ "eval_accuracy": 0.9397825296625574,
755
+ "eval_loss": 0.41976475715637207,
756
+ "eval_macro avg": {
757
+ "f1-score": 0.9160324092463666,
758
+ "precision": 0.9078845437165368,
759
+ "recall": 0.9254915260341382,
760
+ "support": 30257.0
761
+ },
762
+ "eval_runtime": 4.7401,
763
+ "eval_samples_per_second": 17.088,
764
+ "eval_steps_per_second": 2.321,
765
+ "eval_weighted avg": {
766
+ "f1-score": 0.9396639602085823,
767
+ "precision": 0.9399666776761432,
768
+ "recall": 0.9397825296625574,
769
+ "support": 30257.0
770
+ },
771
+ "step": 1539
772
+ },
773
+ {
774
+ "epoch": 20.0,
775
+ "eval_B": {
776
+ "f1-score": 0.8851099128992119,
777
+ "precision": 0.8653690186536902,
778
+ "recall": 0.9057724957555179,
779
+ "support": 1178.0
780
+ },
781
+ "eval_I": {
782
+ "f1-score": 0.9574706357267495,
783
+ "precision": 0.9427633603446507,
784
+ "recall": 0.972644055241018,
785
+ "support": 18899.0
786
+ },
787
+ "eval_O": {
788
+ "f1-score": 0.9156602050137015,
789
+ "precision": 0.9470921688011757,
790
+ "recall": 0.8862475442043222,
791
+ "support": 10180.0
792
+ },
793
+ "eval_accuracy": 0.9409723369798725,
794
+ "eval_loss": 0.4046960175037384,
795
+ "eval_macro avg": {
796
+ "f1-score": 0.9194135845465543,
797
+ "precision": 0.9184081825998388,
798
+ "recall": 0.921554698400286,
799
+ "support": 30257.0
800
+ },
801
+ "eval_runtime": 4.6993,
802
+ "eval_samples_per_second": 17.237,
803
+ "eval_steps_per_second": 2.341,
804
+ "eval_weighted avg": {
805
+ "f1-score": 0.9405862414989786,
806
+ "precision": 0.9412065878812694,
807
+ "recall": 0.9409723369798725,
808
+ "support": 30257.0
809
+ },
810
+ "step": 1620
811
+ }
812
+ ],
813
+ "logging_steps": 500,
814
+ "max_steps": 4050,
815
+ "num_input_tokens_seen": 0,
816
+ "num_train_epochs": 50,
817
+ "save_steps": 500,
818
+ "total_flos": 2866802206716000.0,
819
+ "train_batch_size": 4,
820
+ "trial_name": null,
821
+ "trial_params": null
822
+ }
checkpoint-1620/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a9f81281de988dbf2616f1c96e725c00f6e4415b289b670a7cb21136eb66e95f
3
+ size 4856
checkpoint-1620/vocab.json ADDED
The diff for this file is too large to render. See raw diff