Hugging Face's logo

Theoreticallyhugo
/
longformer-sep_tok

Token Classification
Transformers
Safetensors
longformer
Generated from Trainer
Eval Results
Inference Endpoints
Model card Files Community
Theoreticallyhugo commited on
Commit
5697a2f
·
verified ·
1 Parent(s): 4b74c45

Training in progress, epoch 13, checkpoint

Browse files
Files changed (12) hide show
  1. checkpoint-533/config.json +57 -0
  2. checkpoint-533/merges.txt +0 -0
  3. checkpoint-533/model.safetensors +3 -0
  4. checkpoint-533/optimizer.pt +3 -0
  5. checkpoint-533/rng_state.pth +3 -0
  6. checkpoint-533/scheduler.pt +3 -0
  7. checkpoint-533/special_tokens_map.json +15 -0
  8. checkpoint-533/tokenizer.json +0 -0
  9. checkpoint-533/tokenizer_config.json +57 -0
  10. checkpoint-533/trainer_state.json +612 -0
  11. checkpoint-533/training_args.bin +3 -0
  12. checkpoint-533/vocab.json +0 -0
checkpoint-533/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": "4.37.2",
55
+ "type_vocab_size": 1,
56
+ "vocab_size": 50265
57
+ }
checkpoint-533/merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-533/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e1873915d9c022f571ce576dd34d00f11e8be6afd298963e58140416ac1f54ae
3
+ size 592324828
checkpoint-533/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:39b3a2b1d1312dc34b24d9bb8e6d1724f8514e0464eb6e6c94d2b7983fbfdc69
3
+ size 1014657786
checkpoint-533/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:04834172ddf5ec4a4ab3f45302fc941de3f138924747420a0b413f26bf27ec45
3
+ size 14244
checkpoint-533/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:882fd16e71b72f8dc1638a8359688b90b6b590f02479a4aecbee8fa6c5c28dd2
3
+ size 1064
checkpoint-533/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-533/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-533/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-533/trainer_state.json ADDED
@@ -0,0 +1,612 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 13.0,
5
+ "eval_steps": 500,
6
+ "global_step": 533,
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": 0.476290832455216,
15
+ "precision": 0.564843099139128,
16
+ "recall": 0.4117408906882591,
17
+ "support": 4940.0
18
+ },
19
+ "eval_MajorClaim": {
20
+ "f1-score": 0.6887323943661973,
21
+ "precision": 0.5601374570446735,
22
+ "recall": 0.8939670932358318,
23
+ "support": 2188.0
24
+ },
25
+ "eval_O": {
26
+ "f1-score": 0.9920864118816337,
27
+ "precision": 0.9936889556724268,
28
+ "recall": 0.9904890286826931,
29
+ "support": 13353.0
30
+ },
31
+ "eval_Premise": {
32
+ "f1-score": 0.8866859078930858,
33
+ "precision": 0.8845214996557551,
34
+ "recall": 0.888860934649978,
35
+ "support": 15899.0
36
+ },
37
+ "eval_accuracy": 0.8616822429906542,
38
+ "eval_loss": 0.3469788730144501,
39
+ "eval_macro avg": {
40
+ "f1-score": 0.7609488866490333,
41
+ "precision": 0.7507977528779959,
42
+ "recall": 0.7962644868141905,
43
+ "support": 36380.0
44
+ },
45
+ "eval_runtime": 1.4365,
46
+ "eval_samples_per_second": 56.386,
47
+ "eval_steps_per_second": 7.657,
48
+ "eval_weighted avg": {
49
+ "f1-score": 0.8577397553229695,
50
+ "precision": 0.8616723918054371,
51
+ "recall": 0.8616822429906542,
52
+ "support": 36380.0
53
+ },
54
+ "step": 41
55
+ },
56
+ {
57
+ "epoch": 2.0,
58
+ "eval_Claim": {
59
+ "f1-score": 0.6444232602478551,
60
+ "precision": 0.609009009009009,
61
+ "recall": 0.6842105263157895,
62
+ "support": 4940.0
63
+ },
64
+ "eval_MajorClaim": {
65
+ "f1-score": 0.8019559902200488,
66
+ "precision": 0.8622502628811777,
67
+ "recall": 0.7495429616087751,
68
+ "support": 2188.0
69
+ },
70
+ "eval_O": {
71
+ "f1-score": 0.998463671450519,
72
+ "precision": 0.9991750412479377,
73
+ "recall": 0.9977533138620535,
74
+ "support": 13353.0
75
+ },
76
+ "eval_Premise": {
77
+ "f1-score": 0.9057885879401772,
78
+ "precision": 0.9146466589713993,
79
+ "recall": 0.8971004465689666,
80
+ "support": 15899.0
81
+ },
82
+ "eval_accuracy": 0.8962616822429906,
83
+ "eval_loss": 0.24824203550815582,
84
+ "eval_macro avg": {
85
+ "f1-score": 0.83765787746465,
86
+ "precision": 0.846270243027381,
87
+ "recall": 0.8321518120888962,
88
+ "support": 36380.0
89
+ },
90
+ "eval_runtime": 1.4437,
91
+ "eval_samples_per_second": 56.108,
92
+ "eval_steps_per_second": 7.62,
93
+ "eval_weighted avg": {
94
+ "f1-score": 0.898068960328904,
95
+ "precision": 0.901018681595891,
96
+ "recall": 0.8962616822429906,
97
+ "support": 36380.0
98
+ },
99
+ "step": 82
100
+ },
101
+ {
102
+ "epoch": 3.0,
103
+ "eval_Claim": {
104
+ "f1-score": 0.5137480141757302,
105
+ "precision": 0.6481652790625964,
106
+ "recall": 0.42550607287449393,
107
+ "support": 4940.0
108
+ },
109
+ "eval_MajorClaim": {
110
+ "f1-score": 0.8317541613316262,
111
+ "precision": 0.9458357600465929,
112
+ "recall": 0.7422303473491774,
113
+ "support": 2188.0
114
+ },
115
+ "eval_O": {
116
+ "f1-score": 0.9982381827042021,
117
+ "precision": 0.9993245271690183,
118
+ "recall": 0.997154197558601,
119
+ "support": 13353.0
120
+ },
121
+ "eval_Premise": {
122
+ "f1-score": 0.8998382115016914,
123
+ "precision": 0.8452144120247569,
124
+ "recall": 0.962010189320083,
125
+ "support": 15899.0
126
+ },
127
+ "eval_accuracy": 0.8888400219901045,
128
+ "eval_loss": 0.2560241222381592,
129
+ "eval_macro avg": {
130
+ "f1-score": 0.8108946424283126,
131
+ "precision": 0.8596349945757411,
132
+ "recall": 0.7817252017755888,
133
+ "support": 36380.0
134
+ },
135
+ "eval_runtime": 1.4504,
136
+ "eval_samples_per_second": 55.846,
137
+ "eval_steps_per_second": 7.584,
138
+ "eval_weighted avg": {
139
+ "f1-score": 0.8794336303830761,
140
+ "precision": 0.8810739271473523,
141
+ "recall": 0.8888400219901045,
142
+ "support": 36380.0
143
+ },
144
+ "step": 123
145
+ },
146
+ {
147
+ "epoch": 4.0,
148
+ "eval_Claim": {
149
+ "f1-score": 0.5716363636363636,
150
+ "precision": 0.7123867069486405,
151
+ "recall": 0.4773279352226721,
152
+ "support": 4940.0
153
+ },
154
+ "eval_MajorClaim": {
155
+ "f1-score": 0.8958780081010246,
156
+ "precision": 0.9357889497262319,
157
+ "recall": 0.8592321755027422,
158
+ "support": 2188.0
159
+ },
160
+ "eval_O": {
161
+ "f1-score": 0.9977489307421025,
162
+ "precision": 0.999699270731524,
163
+ "recall": 0.9958061858758331,
164
+ "support": 13353.0
165
+ },
166
+ "eval_Premise": {
167
+ "f1-score": 0.9059092664666211,
168
+ "precision": 0.8584459459459459,
169
+ "recall": 0.958928234480156,
170
+ "support": 15899.0
171
+ },
172
+ "eval_accuracy": 0.9010720175920836,
173
+ "eval_loss": 0.24002882838249207,
174
+ "eval_macro avg": {
175
+ "f1-score": 0.8427931422365279,
176
+ "precision": 0.8765802183380855,
177
+ "recall": 0.8228236327703509,
178
+ "support": 36380.0
179
+ },
180
+ "eval_runtime": 1.4479,
181
+ "eval_samples_per_second": 55.942,
182
+ "eval_steps_per_second": 7.597,
183
+ "eval_weighted avg": {
184
+ "f1-score": 0.8936244534865525,
185
+ "precision": 0.8951103081638239,
186
+ "recall": 0.9010720175920836,
187
+ "support": 36380.0
188
+ },
189
+ "step": 164
190
+ },
191
+ {
192
+ "epoch": 5.0,
193
+ "eval_Claim": {
194
+ "f1-score": 0.6379673804504604,
195
+ "precision": 0.7058679106309845,
196
+ "recall": 0.5819838056680162,
197
+ "support": 4940.0
198
+ },
199
+ "eval_MajorClaim": {
200
+ "f1-score": 0.8943466172381834,
201
+ "precision": 0.9069548872180451,
202
+ "recall": 0.8820840950639853,
203
+ "support": 2188.0
204
+ },
205
+ "eval_O": {
206
+ "f1-score": 0.9964329966582812,
207
+ "precision": 0.999171686746988,
208
+ "recall": 0.9937092788137497,
209
+ "support": 13353.0
210
+ },
211
+ "eval_Premise": {
212
+ "f1-score": 0.9112140984206354,
213
+ "precision": 0.8842535061246227,
214
+ "recall": 0.939870432102648,
215
+ "support": 15899.0
216
+ },
217
+ "eval_accuracy": 0.9075590984057175,
218
+ "eval_loss": 0.2713705599308014,
219
+ "eval_macro avg": {
220
+ "f1-score": 0.8599902731918901,
221
+ "precision": 0.8740619976801601,
222
+ "recall": 0.8494119029120998,
223
+ "support": 36380.0
224
+ },
225
+ "eval_runtime": 1.4528,
226
+ "eval_samples_per_second": 55.756,
227
+ "eval_steps_per_second": 7.572,
228
+ "eval_weighted avg": {
229
+ "f1-score": 0.9043747117402454,
230
+ "precision": 0.9035758878163292,
231
+ "recall": 0.9075590984057175,
232
+ "support": 36380.0
233
+ },
234
+ "step": 205
235
+ },
236
+ {
237
+ "epoch": 6.0,
238
+ "eval_Claim": {
239
+ "f1-score": 0.6678616617460986,
240
+ "precision": 0.6969630281690141,
241
+ "recall": 0.6410931174089068,
242
+ "support": 4940.0
243
+ },
244
+ "eval_MajorClaim": {
245
+ "f1-score": 0.8903225806451615,
246
+ "precision": 0.8673602080624188,
247
+ "recall": 0.9145338208409507,
248
+ "support": 2188.0
249
+ },
250
+ "eval_O": {
251
+ "f1-score": 0.9950327387672161,
252
+ "precision": 1.0,
253
+ "recall": 0.9901145809930353,
254
+ "support": 13353.0
255
+ },
256
+ "eval_Premise": {
257
+ "f1-score": 0.9114167727512653,
258
+ "precision": 0.8999877360804514,
259
+ "recall": 0.9231398201144726,
260
+ "support": 15899.0
261
+ },
262
+ "eval_accuracy": 0.9089059923034635,
263
+ "eval_loss": 0.265927255153656,
264
+ "eval_macro avg": {
265
+ "f1-score": 0.8661584384774353,
266
+ "precision": 0.8660777430779711,
267
+ "recall": 0.8672203348393414,
268
+ "support": 36380.0
269
+ },
270
+ "eval_runtime": 1.4493,
271
+ "eval_samples_per_second": 55.89,
272
+ "eval_steps_per_second": 7.59,
273
+ "eval_weighted avg": {
274
+ "f1-score": 0.9077666257891245,
275
+ "precision": 0.9071656544897909,
276
+ "recall": 0.9089059923034635,
277
+ "support": 36380.0
278
+ },
279
+ "step": 246
280
+ },
281
+ {
282
+ "epoch": 7.0,
283
+ "eval_Claim": {
284
+ "f1-score": 0.5876388218252052,
285
+ "precision": 0.7278708133971292,
286
+ "recall": 0.49271255060728747,
287
+ "support": 4940.0
288
+ },
289
+ "eval_MajorClaim": {
290
+ "f1-score": 0.888888888888889,
291
+ "precision": 0.8874715261958998,
292
+ "recall": 0.8903107861060329,
293
+ "support": 2188.0
294
+ },
295
+ "eval_O": {
296
+ "f1-score": 0.9968070320423725,
297
+ "precision": 1.0,
298
+ "recall": 0.9936343892758182,
299
+ "support": 13353.0
300
+ },
301
+ "eval_Premise": {
302
+ "f1-score": 0.9057122370936902,
303
+ "precision": 0.8625732658054971,
304
+ "recall": 0.9533932951757972,
305
+ "support": 15899.0
306
+ },
307
+ "eval_accuracy": 0.9018141836173722,
308
+ "eval_loss": 0.3055996596813202,
309
+ "eval_macro avg": {
310
+ "f1-score": 0.8447617449625392,
311
+ "precision": 0.8694789013496315,
312
+ "recall": 0.832512755291234,
313
+ "support": 36380.0
314
+ },
315
+ "eval_runtime": 1.4427,
316
+ "eval_samples_per_second": 56.144,
317
+ "eval_steps_per_second": 7.624,
318
+ "eval_weighted avg": {
319
+ "f1-score": 0.8949452398328692,
320
+ "precision": 0.8962210519664664,
321
+ "recall": 0.9018141836173722,
322
+ "support": 36380.0
323
+ },
324
+ "step": 287
325
+ },
326
+ {
327
+ "epoch": 8.0,
328
+ "eval_Claim": {
329
+ "f1-score": 0.6830078936435396,
330
+ "precision": 0.7013651877133106,
331
+ "recall": 0.665587044534413,
332
+ "support": 4940.0
333
+ },
334
+ "eval_MajorClaim": {
335
+ "f1-score": 0.9099371654642774,
336
+ "precision": 0.9269796111901375,
337
+ "recall": 0.893510054844607,
338
+ "support": 2188.0
339
+ },
340
+ "eval_O": {
341
+ "f1-score": 0.9968451889130925,
342
+ "precision": 0.9998493181646952,
343
+ "recall": 0.9938590578896128,
344
+ "support": 13353.0
345
+ },
346
+ "eval_Premise": {
347
+ "f1-score": 0.9144649011145954,
348
+ "precision": 0.9029429797670141,
349
+ "recall": 0.9262846719919492,
350
+ "support": 15899.0
351
+ },
352
+ "eval_accuracy": 0.9137163276525564,
353
+ "eval_loss": 0.31213706731796265,
354
+ "eval_macro avg": {
355
+ "f1-score": 0.8760637872838762,
356
+ "precision": 0.8827842742087894,
357
+ "recall": 0.8698102073151455,
358
+ "support": 36380.0
359
+ },
360
+ "eval_runtime": 1.4431,
361
+ "eval_samples_per_second": 56.13,
362
+ "eval_steps_per_second": 7.623,
363
+ "eval_weighted avg": {
364
+ "f1-score": 0.9130003513747224,
365
+ "precision": 0.9125853160350936,
366
+ "recall": 0.9137163276525564,
367
+ "support": 36380.0
368
+ },
369
+ "step": 328
370
+ },
371
+ {
372
+ "epoch": 9.0,
373
+ "eval_Claim": {
374
+ "f1-score": 0.6714174150722373,
375
+ "precision": 0.6483785822021116,
376
+ "recall": 0.6961538461538461,
377
+ "support": 4940.0
378
+ },
379
+ "eval_MajorClaim": {
380
+ "f1-score": 0.8994732221246707,
381
+ "precision": 0.8652871621621622,
382
+ "recall": 0.9364716636197441,
383
+ "support": 2188.0
384
+ },
385
+ "eval_O": {
386
+ "f1-score": 0.9986506746626687,
387
+ "precision": 0.9996248217903504,
388
+ "recall": 0.9976784243241219,
389
+ "support": 13353.0
390
+ },
391
+ "eval_Premise": {
392
+ "f1-score": 0.9039641943734017,
393
+ "precision": 0.9191860087120474,
394
+ "recall": 0.8892383168752752,
395
+ "support": 15899.0
396
+ },
397
+ "eval_accuracy": 0.9056624518966465,
398
+ "eval_loss": 0.3160102665424347,
399
+ "eval_macro avg": {
400
+ "f1-score": 0.8683763765582445,
401
+ "precision": 0.8581191437166679,
402
+ "recall": 0.8798855627432468,
403
+ "support": 36380.0
404
+ },
405
+ "eval_runtime": 1.4435,
406
+ "eval_samples_per_second": 56.114,
407
+ "eval_steps_per_second": 7.62,
408
+ "eval_weighted avg": {
409
+ "f1-score": 0.9068707703567609,
410
+ "precision": 0.9086961820991927,
411
+ "recall": 0.9056624518966465,
412
+ "support": 36380.0
413
+ },
414
+ "step": 369
415
+ },
416
+ {
417
+ "epoch": 10.0,
418
+ "eval_Claim": {
419
+ "f1-score": 0.6854596331226034,
420
+ "precision": 0.7022722446379274,
421
+ "recall": 0.6694331983805668,
422
+ "support": 4940.0
423
+ },
424
+ "eval_MajorClaim": {
425
+ "f1-score": 0.912037037037037,
426
+ "precision": 0.924015009380863,
427
+ "recall": 0.9003656307129799,
428
+ "support": 2188.0
429
+ },
430
+ "eval_O": {
431
+ "f1-score": 0.9982003599280145,
432
+ "precision": 0.999474435017644,
433
+ "recall": 0.9969295289448065,
434
+ "support": 13353.0
435
+ },
436
+ "eval_Premise": {
437
+ "f1-score": 0.9142252249447368,
438
+ "precision": 0.9051787916152898,
439
+ "recall": 0.9234543053022203,
440
+ "support": 15899.0
441
+ },
442
+ "eval_accuracy": 0.9145409565695437,
443
+ "eval_loss": 0.345079630613327,
444
+ "eval_macro avg": {
445
+ "f1-score": 0.8774805637580978,
446
+ "precision": 0.882735120162931,
447
+ "recall": 0.8725456658351434,
448
+ "support": 36380.0
449
+ },
450
+ "eval_runtime": 1.4444,
451
+ "eval_samples_per_second": 56.08,
452
+ "eval_steps_per_second": 7.616,
453
+ "eval_weighted avg": {
454
+ "f1-score": 0.9138522232594241,
455
+ "precision": 0.9133696939999666,
456
+ "recall": 0.9145409565695437,
457
+ "support": 36380.0
458
+ },
459
+ "step": 410
460
+ },
461
+ {
462
+ "epoch": 11.0,
463
+ "eval_Claim": {
464
+ "f1-score": 0.6640950792326938,
465
+ "precision": 0.6846517626827171,
466
+ "recall": 0.6447368421052632,
467
+ "support": 4940.0
468
+ },
469
+ "eval_MajorClaim": {
470
+ "f1-score": 0.9124319419237751,
471
+ "precision": 0.9058558558558558,
472
+ "recall": 0.9191042047531993,
473
+ "support": 2188.0
474
+ },
475
+ "eval_O": {
476
+ "f1-score": 0.9980882408066872,
477
+ "precision": 0.9991744220954668,
478
+ "recall": 0.9970044184827379,
479
+ "support": 13353.0
480
+ },
481
+ "eval_Premise": {
482
+ "f1-score": 0.9098899728828351,
483
+ "precision": 0.9018783984181908,
484
+ "recall": 0.9180451600729606,
485
+ "support": 15899.0
486
+ },
487
+ "eval_accuracy": 0.909978009895547,
488
+ "eval_loss": 0.3666386902332306,
489
+ "eval_macro avg": {
490
+ "f1-score": 0.8711263087114978,
491
+ "precision": 0.8728901097630577,
492
+ "recall": 0.8697226563535402,
493
+ "support": 36380.0
494
+ },
495
+ "eval_runtime": 1.4494,
496
+ "eval_samples_per_second": 55.887,
497
+ "eval_steps_per_second": 7.59,
498
+ "eval_weighted avg": {
499
+ "f1-score": 0.9090391352032605,
500
+ "precision": 0.9083324088773178,
501
+ "recall": 0.909978009895547,
502
+ "support": 36380.0
503
+ },
504
+ "step": 451
505
+ },
506
+ {
507
+ "epoch": 12.0,
508
+ "eval_Claim": {
509
+ "f1-score": 0.6594242259641498,
510
+ "precision": 0.7116060961313013,
511
+ "recall": 0.6143724696356275,
512
+ "support": 4940.0
513
+ },
514
+ "eval_MajorClaim": {
515
+ "f1-score": 0.897596882442087,
516
+ "precision": 0.8527354997943233,
517
+ "recall": 0.9474405850091407,
518
+ "support": 2188.0
519
+ },
520
+ "eval_O": {
521
+ "f1-score": 0.9980512666766601,
522
+ "precision": 0.9988748030905409,
523
+ "recall": 0.9972290870965326,
524
+ "support": 13353.0
525
+ },
526
+ "eval_Premise": {
527
+ "f1-score": 0.9140518417462483,
528
+ "precision": 0.9013636641594814,
529
+ "recall": 0.9271023334800931,
530
+ "support": 15899.0
531
+ },
532
+ "eval_accuracy": 0.9115997800989555,
533
+ "eval_loss": 0.4192342460155487,
534
+ "eval_macro avg": {
535
+ "f1-score": 0.8672810542072864,
536
+ "precision": 0.8661450157939118,
537
+ "recall": 0.8715361188053484,
538
+ "support": 36380.0
539
+ },
540
+ "eval_runtime": 1.4505,
541
+ "eval_samples_per_second": 55.842,
542
+ "eval_steps_per_second": 7.583,
543
+ "eval_weighted avg": {
544
+ "f1-score": 0.9093179343293907,
545
+ "precision": 0.9084627688449202,
546
+ "recall": 0.9115997800989555,
547
+ "support": 36380.0
548
+ },
549
+ "step": 492
550
+ },
551
+ {
552
+ "epoch": 12.2,
553
+ "learning_rate": 4.75609756097561e-06,
554
+ "loss": 0.1714,
555
+ "step": 500
556
+ },
557
+ {
558
+ "epoch": 13.0,
559
+ "eval_Claim": {
560
+ "f1-score": 0.6607543540976599,
561
+ "precision": 0.6997058157954288,
562
+ "recall": 0.625910931174089,
563
+ "support": 4940.0
564
+ },
565
+ "eval_MajorClaim": {
566
+ "f1-score": 0.9072817729534148,
567
+ "precision": 0.8979409131602507,
568
+ "recall": 0.916819012797075,
569
+ "support": 2188.0
570
+ },
571
+ "eval_O": {
572
+ "f1-score": 0.9972236812485931,
573
+ "precision": 0.9991729945116908,
574
+ "recall": 0.9952819591103123,
575
+ "support": 13353.0
576
+ },
577
+ "eval_Premise": {
578
+ "f1-score": 0.9105645784996134,
579
+ "precision": 0.8959576281504931,
580
+ "recall": 0.9256557016164538,
581
+ "support": 15899.0
582
+ },
583
+ "eval_accuracy": 0.909978009895547,
584
+ "eval_loss": 0.431348592042923,
585
+ "eval_macro avg": {
586
+ "f1-score": 0.8689560966998203,
587
+ "precision": 0.8731943379044659,
588
+ "recall": 0.8659169011744825,
589
+ "support": 36380.0
590
+ },
591
+ "eval_runtime": 1.4449,
592
+ "eval_samples_per_second": 56.059,
593
+ "eval_steps_per_second": 7.613,
594
+ "eval_weighted avg": {
595
+ "f1-score": 0.9082532456773593,
596
+ "precision": 0.9073125006515488,
597
+ "recall": 0.909978009895547,
598
+ "support": 36380.0
599
+ },
600
+ "step": 533
601
+ }
602
+ ],
603
+ "logging_steps": 500,
604
+ "max_steps": 656,
605
+ "num_input_tokens_seen": 0,
606
+ "num_train_epochs": 16,
607
+ "save_steps": 500,
608
+ "total_flos": 1863448390276200.0,
609
+ "train_batch_size": 8,
610
+ "trial_name": null,
611
+ "trial_params": null
612
+ }
checkpoint-533/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9131e825d1decae7bb99beb71561d875ec8f04a013ba8da62dbde2c9f130419a
3
+ size 4664
checkpoint-533/vocab.json ADDED
The diff for this file is too large to render. See raw diff