Hugging Face's logo

Theoreticallyhugo
/
longformer-simple

Token Classification
Transformers
Safetensors
longformer
Generated from Trainer
Eval Results
Inference Endpoints
Model card Files Community
Theoreticallyhugo commited on
Commit
d543d5e
•
1 Parent(s): 341a393

Training in progress, epoch 8, checkpoint

Browse files
Files changed (12) hide show
  1. checkpoint-648/config.json +57 -0
  2. checkpoint-648/merges.txt +0 -0
  3. checkpoint-648/model.safetensors +3 -0
  4. checkpoint-648/optimizer.pt +3 -0
  5. checkpoint-648/rng_state.pth +3 -0
  6. checkpoint-648/scheduler.pt +3 -0
  7. checkpoint-648/special_tokens_map.json +15 -0
  8. checkpoint-648/tokenizer.json +0 -0
  9. checkpoint-648/tokenizer_config.json +57 -0
  10. checkpoint-648/trainer_state.json +388 -0
  11. checkpoint-648/training_args.bin +3 -0
  12. checkpoint-648/vocab.json +0 -0
checkpoint-648/config.json ADDED
@@ -0,0 +1,57 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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": "X_placeholder_X",
31
+ "2": "MajorClaim",
32
+ "3": "Claim",
33
+ "4": "Premise"
34
+ },
35
+ "ignore_attention_mask": false,
36
+ "initializer_range": 0.02,
37
+ "intermediate_size": 3072,
38
+ "label2id": {
39
+ "Claim": 3,
40
+ "MajorClaim": 2,
41
+ "O": 0,
42
+ "Premise": 4,
43
+ "X_placeholder_X": 1
44
+ },
45
+ "layer_norm_eps": 1e-05,
46
+ "max_position_embeddings": 4098,
47
+ "model_type": "longformer",
48
+ "num_attention_heads": 12,
49
+ "num_hidden_layers": 12,
50
+ "onnx_export": false,
51
+ "pad_token_id": 1,
52
+ "sep_token_id": 2,
53
+ "torch_dtype": "float32",
54
+ "transformers_version": "4.38.2",
55
+ "type_vocab_size": 1,
56
+ "vocab_size": 50265
57
+ }
checkpoint-648/merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-648/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:60c9ae57744dc6358c24fa2e14d4fc6e36499d544ca81303879d2c68cf04c3a0
3
+ size 592324828
checkpoint-648/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1f678887af8ae33b6144c8b383ac63501414d4822ec65b790f08d63e4a57193d
3
+ size 1014657786
checkpoint-648/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e241cb4a4c96fc25c055fcd71173708b81cd84562a71b4d7503d9ffa5397d596
3
+ size 14244
checkpoint-648/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:17d9e668ea6c9b485e63f3e5ef87c1cfb58934d5fcbfc76edc86a350fba51ded
3
+ size 1064
checkpoint-648/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-648/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-648/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-648/trainer_state.json ADDED
@@ -0,0 +1,388 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 8.0,
5
+ "eval_steps": 500,
6
+ "global_step": 648,
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_Claim": {
14
+ "f1-score": 0.44734338904742493,
15
+ "precision": 0.5477537437603993,
16
+ "recall": 0.3780431786862655,
17
+ "support": 4354.0
18
+ },
19
+ "eval_MajorClaim": {
20
+ "f1-score": 0.6504965622612681,
21
+ "precision": 0.5898856944925528,
22
+ "recall": 0.7249893571732652,
23
+ "support": 2349.0
24
+ },
25
+ "eval_O": {
26
+ "f1-score": 0.8859784283513097,
27
+ "precision": 0.8689081979599547,
28
+ "recall": 0.9037328094302554,
29
+ "support": 10180.0
30
+ },
31
+ "eval_Premise": {
32
+ "f1-score": 0.8720120805863504,
33
+ "precision": 0.859258183929738,
34
+ "recall": 0.8851502916105877,
35
+ "support": 13374.0
36
+ },
37
+ "eval_accuracy": 0.8059953068711373,
38
+ "eval_loss": 0.47828415036201477,
39
+ "eval_macro avg": {
40
+ "f1-score": 0.7139576150615883,
41
+ "precision": 0.7164514550356611,
42
+ "recall": 0.7229789092250934,
43
+ "support": 30257.0
44
+ },
45
+ "eval_runtime": 4.8422,
46
+ "eval_samples_per_second": 16.728,
47
+ "eval_steps_per_second": 2.272,
48
+ "eval_weighted avg": {
49
+ "f1-score": 0.7984036588902531,
50
+ "precision": 0.7967665566250599,
51
+ "recall": 0.8059953068711373,
52
+ "support": 30257.0
53
+ },
54
+ "step": 81
55
+ },
56
+ {
57
+ "epoch": 2.0,
58
+ "eval_Claim": {
59
+ "f1-score": 0.5787561000934482,
60
+ "precision": 0.5281409891984082,
61
+ "recall": 0.6401010564997703,
62
+ "support": 4354.0
63
+ },
64
+ "eval_MajorClaim": {
65
+ "f1-score": 0.7702650787552824,
66
+ "precision": 0.7017850892544627,
67
+ "recall": 0.8535547041294168,
68
+ "support": 2349.0
69
+ },
70
+ "eval_O": {
71
+ "f1-score": 0.9095337508698677,
72
+ "precision": 0.9206077681626081,
73
+ "recall": 0.8987229862475442,
74
+ "support": 10180.0
75
+ },
76
+ "eval_Premise": {
77
+ "f1-score": 0.8736648538675224,
78
+ "precision": 0.9162905211325401,
79
+ "recall": 0.8348287722446538,
80
+ "support": 13374.0
81
+ },
82
+ "eval_accuracy": 0.8297584030141786,
83
+ "eval_loss": 0.4241589605808258,
84
+ "eval_macro avg": {
85
+ "f1-score": 0.7830549458965301,
86
+ "precision": 0.7667060919370048,
87
+ "recall": 0.8068018797803463,
88
+ "support": 30257.0
89
+ },
90
+ "eval_runtime": 4.8636,
91
+ "eval_samples_per_second": 16.654,
92
+ "eval_steps_per_second": 2.262,
93
+ "eval_weighted avg": {
94
+ "f1-score": 0.8352680063880269,
95
+ "precision": 0.8452350051608073,
96
+ "recall": 0.8297584030141786,
97
+ "support": 30257.0
98
+ },
99
+ "step": 162
100
+ },
101
+ {
102
+ "epoch": 3.0,
103
+ "eval_Claim": {
104
+ "f1-score": 0.5728465083303793,
105
+ "precision": 0.5899245558530056,
106
+ "recall": 0.5567294441892513,
107
+ "support": 4354.0
108
+ },
109
+ "eval_MajorClaim": {
110
+ "f1-score": 0.7862876979822088,
111
+ "precision": 0.8017699115044248,
112
+ "recall": 0.7713920817369093,
113
+ "support": 2349.0
114
+ },
115
+ "eval_O": {
116
+ "f1-score": 0.914532821534425,
117
+ "precision": 0.9048168445341794,
118
+ "recall": 0.9244597249508841,
119
+ "support": 10180.0
120
+ },
121
+ "eval_Premise": {
122
+ "f1-score": 0.8880533114924986,
123
+ "precision": 0.8843330614665975,
124
+ "recall": 0.8918049947659639,
125
+ "support": 13374.0
126
+ },
127
+ "eval_accuracy": 0.8452258981392735,
128
+ "eval_loss": 0.41067537665367126,
129
+ "eval_macro avg": {
130
+ "f1-score": 0.7904300848348779,
131
+ "precision": 0.7952110933395518,
132
+ "recall": 0.7860965614107521,
133
+ "support": 30257.0
134
+ },
135
+ "eval_runtime": 4.8684,
136
+ "eval_samples_per_second": 16.638,
137
+ "eval_steps_per_second": 2.259,
138
+ "eval_weighted avg": {
139
+ "f1-score": 0.8437033615676307,
140
+ "precision": 0.8424495118392472,
141
+ "recall": 0.8452258981392735,
142
+ "support": 30257.0
143
+ },
144
+ "step": 243
145
+ },
146
+ {
147
+ "epoch": 4.0,
148
+ "eval_Claim": {
149
+ "f1-score": 0.5865384615384616,
150
+ "precision": 0.5714596949891068,
151
+ "recall": 0.6024345429490124,
152
+ "support": 4354.0
153
+ },
154
+ "eval_MajorClaim": {
155
+ "f1-score": 0.7783630470016208,
156
+ "precision": 0.7425589485890993,
157
+ "recall": 0.8177948063005535,
158
+ "support": 2349.0
159
+ },
160
+ "eval_O": {
161
+ "f1-score": 0.9108591535529275,
162
+ "precision": 0.9063411787589963,
163
+ "recall": 0.9154223968565816,
164
+ "support": 10180.0
165
+ },
166
+ "eval_Premise": {
167
+ "f1-score": 0.8868256151612411,
168
+ "precision": 0.9067823097358962,
169
+ "recall": 0.8677284282937042,
170
+ "support": 13374.0
171
+ },
172
+ "eval_accuracy": 0.8417225765938461,
173
+ "eval_loss": 0.4545813202857971,
174
+ "eval_macro avg": {
175
+ "f1-score": 0.7906465693135627,
176
+ "precision": 0.7817855330182747,
177
+ "recall": 0.8008450435999629,
178
+ "support": 30257.0
179
+ },
180
+ "eval_runtime": 4.9112,
181
+ "eval_samples_per_second": 16.493,
182
+ "eval_steps_per_second": 2.24,
183
+ "eval_weighted avg": {
184
+ "f1-score": 0.8432797441676475,
185
+ "precision": 0.8456313015960876,
186
+ "recall": 0.8417225765938461,
187
+ "support": 30257.0
188
+ },
189
+ "step": 324
190
+ },
191
+ {
192
+ "epoch": 5.0,
193
+ "eval_Claim": {
194
+ "f1-score": 0.597399608790703,
195
+ "precision": 0.5985704403965875,
196
+ "recall": 0.5962333486449242,
197
+ "support": 4354.0
198
+ },
199
+ "eval_MajorClaim": {
200
+ "f1-score": 0.7933532168726033,
201
+ "precision": 0.7940298507462686,
202
+ "recall": 0.7926777352064709,
203
+ "support": 2349.0
204
+ },
205
+ "eval_O": {
206
+ "f1-score": 0.9101112629318758,
207
+ "precision": 0.9042862684251358,
208
+ "recall": 0.9160117878192534,
209
+ "support": 10180.0
210
+ },
211
+ "eval_Premise": {
212
+ "f1-score": 0.8923677591320345,
213
+ "precision": 0.8961019377214808,
214
+ "recall": 0.8886645730521908,
215
+ "support": 13374.0
216
+ },
217
+ "eval_accuracy": 0.848332617245596,
218
+ "eval_loss": 0.48704993724823,
219
+ "eval_macro avg": {
220
+ "f1-score": 0.7983079619318041,
221
+ "precision": 0.7982471243223681,
222
+ "recall": 0.7983968611807098,
223
+ "support": 30257.0
224
+ },
225
+ "eval_runtime": 4.8769,
226
+ "eval_samples_per_second": 16.609,
227
+ "eval_steps_per_second": 2.256,
228
+ "eval_weighted avg": {
229
+ "f1-score": 0.8482045037639816,
230
+ "precision": 0.8481162489521332,
231
+ "recall": 0.848332617245596,
232
+ "support": 30257.0
233
+ },
234
+ "step": 405
235
+ },
236
+ {
237
+ "epoch": 6.0,
238
+ "eval_Claim": {
239
+ "f1-score": 0.6060809300245921,
240
+ "precision": 0.5903745644599303,
241
+ "recall": 0.6226458429030777,
242
+ "support": 4354.0
243
+ },
244
+ "eval_MajorClaim": {
245
+ "f1-score": 0.7858529221479404,
246
+ "precision": 0.7963286713286714,
247
+ "recall": 0.7756492124308216,
248
+ "support": 2349.0
249
+ },
250
+ "eval_O": {
251
+ "f1-score": 0.9135615470059578,
252
+ "precision": 0.9398504051527115,
253
+ "recall": 0.8887033398821218,
254
+ "support": 10180.0
255
+ },
256
+ "eval_Premise": {
257
+ "f1-score": 0.895557603686636,
258
+ "precision": 0.88328121591157,
259
+ "recall": 0.908180050844923,
260
+ "support": 13374.0
261
+ },
262
+ "eval_accuracy": 0.8502495290346036,
263
+ "eval_loss": 0.5423789024353027,
264
+ "eval_macro avg": {
265
+ "f1-score": 0.8002632507162816,
266
+ "precision": 0.8024587142132208,
267
+ "recall": 0.7987946115152361,
268
+ "support": 30257.0
269
+ },
270
+ "eval_runtime": 4.867,
271
+ "eval_samples_per_second": 16.643,
272
+ "eval_steps_per_second": 2.26,
273
+ "eval_weighted avg": {
274
+ "f1-score": 0.8514422719925407,
275
+ "precision": 0.8534139871324166,
276
+ "recall": 0.8502495290346036,
277
+ "support": 30257.0
278
+ },
279
+ "step": 486
280
+ },
281
+ {
282
+ "epoch": 6.17,
283
+ "grad_norm": 4.494631767272949,
284
+ "learning_rate": 1.7530864197530865e-05,
285
+ "loss": 0.35,
286
+ "step": 500
287
+ },
288
+ {
289
+ "epoch": 7.0,
290
+ "eval_Claim": {
291
+ "f1-score": 0.6040515653775324,
292
+ "precision": 0.6054453161052146,
293
+ "recall": 0.6026642168121268,
294
+ "support": 4354.0
295
+ },
296
+ "eval_MajorClaim": {
297
+ "f1-score": 0.8052777186635455,
298
+ "precision": 0.8051063829787234,
299
+ "recall": 0.8054491272882077,
300
+ "support": 2349.0
301
+ },
302
+ "eval_O": {
303
+ "f1-score": 0.9184720354599332,
304
+ "precision": 0.9315082331548641,
305
+ "recall": 0.9057956777996071,
306
+ "support": 10180.0
307
+ },
308
+ "eval_Premise": {
309
+ "f1-score": 0.8908606921029282,
310
+ "precision": 0.8810881965774463,
311
+ "recall": 0.9008524001794527,
312
+ "support": 13374.0
313
+ },
314
+ "eval_accuracy": 0.8521994910268698,
315
+ "eval_loss": 0.5764245986938477,
316
+ "eval_macro avg": {
317
+ "f1-score": 0.8046655029009848,
318
+ "precision": 0.8057870322040621,
319
+ "recall": 0.8036903555198486,
320
+ "support": 30257.0
321
+ },
322
+ "eval_runtime": 4.8398,
323
+ "eval_samples_per_second": 16.736,
324
+ "eval_steps_per_second": 2.273,
325
+ "eval_weighted avg": {
326
+ "f1-score": 0.8522343290465387,
327
+ "precision": 0.8524880574571969,
328
+ "recall": 0.8521994910268698,
329
+ "support": 30257.0
330
+ },
331
+ "step": 567
332
+ },
333
+ {
334
+ "epoch": 8.0,
335
+ "eval_Claim": {
336
+ "f1-score": 0.597346719254213,
337
+ "precision": 0.6227261400448543,
338
+ "recall": 0.5739549839228296,
339
+ "support": 4354.0
340
+ },
341
+ "eval_MajorClaim": {
342
+ "f1-score": 0.8047001620745543,
343
+ "precision": 0.7676845767298028,
344
+ "recall": 0.8454661558109834,
345
+ "support": 2349.0
346
+ },
347
+ "eval_O": {
348
+ "f1-score": 0.9099175618286286,
349
+ "precision": 0.9258769700050838,
350
+ "recall": 0.8944990176817289,
351
+ "support": 10180.0
352
+ },
353
+ "eval_Premise": {
354
+ "f1-score": 0.8942491542873952,
355
+ "precision": 0.8797569092750688,
356
+ "recall": 0.9092268580828473,
357
+ "support": 13374.0
358
+ },
359
+ "eval_accuracy": 0.8510757841160723,
360
+ "eval_loss": 0.5954988598823547,
361
+ "eval_macro avg": {
362
+ "f1-score": 0.8015533993611977,
363
+ "precision": 0.7990111490137024,
364
+ "recall": 0.8057867538745973,
365
+ "support": 30257.0
366
+ },
367
+ "eval_runtime": 4.8609,
368
+ "eval_samples_per_second": 16.663,
369
+ "eval_steps_per_second": 2.263,
370
+ "eval_weighted avg": {
371
+ "f1-score": 0.8498442431569896,
372
+ "precision": 0.8495864475589161,
373
+ "recall": 0.8510757841160723,
374
+ "support": 30257.0
375
+ },
376
+ "step": 648
377
+ }
378
+ ],
379
+ "logging_steps": 500,
380
+ "max_steps": 4050,
381
+ "num_input_tokens_seen": 0,
382
+ "num_train_epochs": 50,
383
+ "save_steps": 500,
384
+ "total_flos": 1146737470939200.0,
385
+ "train_batch_size": 4,
386
+ "trial_name": null,
387
+ "trial_params": null
388
+ }
checkpoint-648/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0ce83aa27c1a93c128c1dae6a1646206ab5d85d935efad925ac4e4beb0be29a0
3
+ size 4856
checkpoint-648/vocab.json ADDED
The diff for this file is too large to render. See raw diff