Hugging Face's logo

Theoreticallyhugo
/
longformer-full_labels

Token Classification
Transformers
Safetensors
longformer
Generated from Trainer
Eval Results
Model card Files Community
Theoreticallyhugo commited on
Commit
4b1bf0c
·
verified ·
1 Parent(s): 4f564b3

Training in progress, epoch 12, checkpoint

Browse files
Files changed (12) hide show
  1. checkpoint-972/config.json +61 -0
  2. checkpoint-972/merges.txt +0 -0
  3. checkpoint-972/model.safetensors +3 -0
  4. checkpoint-972/optimizer.pt +3 -0
  5. checkpoint-972/rng_state.pth +3 -0
  6. checkpoint-972/scheduler.pt +3 -0
  7. checkpoint-972/special_tokens_map.json +15 -0
  8. checkpoint-972/tokenizer.json +0 -0
  9. checkpoint-972/tokenizer_config.json +57 -0
  10. checkpoint-972/trainer_state.json +784 -0
  11. checkpoint-972/training_args.bin +3 -0
  12. checkpoint-972/vocab.json +0 -0
checkpoint-972/config.json ADDED
@@ -0,0 +1,61 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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-MajorClaim",
31
+ "2": "I-MajorClaim",
32
+ "3": "B-Claim",
33
+ "4": "I-Claim",
34
+ "5": "B-Premise",
35
+ "6": "I-Premise"
36
+ },
37
+ "ignore_attention_mask": false,
38
+ "initializer_range": 0.02,
39
+ "intermediate_size": 3072,
40
+ "label2id": {
41
+ "B-Claim": 3,
42
+ "B-MajorClaim": 1,
43
+ "B-Premise": 5,
44
+ "I-Claim": 4,
45
+ "I-MajorClaim": 2,
46
+ "I-Premise": 6,
47
+ "O": 0
48
+ },
49
+ "layer_norm_eps": 1e-05,
50
+ "max_position_embeddings": 4098,
51
+ "model_type": "longformer",
52
+ "num_attention_heads": 12,
53
+ "num_hidden_layers": 12,
54
+ "onnx_export": false,
55
+ "pad_token_id": 1,
56
+ "sep_token_id": 2,
57
+ "torch_dtype": "float32",
58
+ "transformers_version": "4.38.2",
59
+ "type_vocab_size": 1,
60
+ "vocab_size": 50265
61
+ }
checkpoint-972/merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-972/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ef7356bd3639275af236c0144fac5f520c809e808daca7fbd2dfaa955b42b368
3
+ size 592330980
checkpoint-972/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f271a3c21c2d2fd7f3588b39c9611d1a26ac634a4f8d5a1525474a036392ddf2
3
+ size 1014670074
checkpoint-972/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1fb40ad83fdc19f8cb7eaa10ff457e54b691181f0739f0dc8a11e71c0ccd9a7b
3
+ size 14244
checkpoint-972/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:71d15ed2b51959ebacca01545575be50f99ed992ff90b76ac4f453d2b82e5b2e
3
+ size 1064
checkpoint-972/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-972/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-972/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-972/trainer_state.json ADDED
@@ -0,0 +1,784 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 12.0,
5
+ "eval_steps": 500,
6
+ "global_step": 972,
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-Claim": {
14
+ "f1-score": 0.0,
15
+ "precision": 0.0,
16
+ "recall": 0.0,
17
+ "support": 295.0
18
+ },
19
+ "eval_B-MajorClaim": {
20
+ "f1-score": 0.0,
21
+ "precision": 0.0,
22
+ "recall": 0.0,
23
+ "support": 156.0
24
+ },
25
+ "eval_B-Premise": {
26
+ "f1-score": 0.6820276497695853,
27
+ "precision": 0.6540404040404041,
28
+ "recall": 0.7125171939477304,
29
+ "support": 727.0
30
+ },
31
+ "eval_I-Claim": {
32
+ "f1-score": 0.43557936284183824,
33
+ "precision": 0.5235513385293121,
34
+ "recall": 0.3729181752353367,
35
+ "support": 4143.0
36
+ },
37
+ "eval_I-MajorClaim": {
38
+ "f1-score": 0.5835814547001276,
39
+ "precision": 0.5468314069350338,
40
+ "recall": 0.6256269949840402,
41
+ "support": 2193.0
42
+ },
43
+ "eval_I-Premise": {
44
+ "f1-score": 0.8638885700072705,
45
+ "precision": 0.8318349299926308,
46
+ "recall": 0.8985115020297699,
47
+ "support": 12563.0
48
+ },
49
+ "eval_O": {
50
+ "f1-score": 0.8838224593742419,
51
+ "precision": 0.8730234786775275,
52
+ "recall": 0.8948919449901768,
53
+ "support": 10180.0
54
+ },
55
+ "eval_accuracy": 0.7876854942657897,
56
+ "eval_loss": 0.5461820363998413,
57
+ "eval_macro avg": {
58
+ "f1-score": 0.4926999280990091,
59
+ "precision": 0.4898973654535584,
60
+ "recall": 0.500637973026722,
61
+ "support": 30257.0
62
+ },
63
+ "eval_runtime": 5.0655,
64
+ "eval_samples_per_second": 15.991,
65
+ "eval_steps_per_second": 2.172,
66
+ "eval_weighted avg": {
67
+ "f1-score": 0.7743853744001297,
68
+ "precision": 0.7661527277359783,
69
+ "recall": 0.7876854942657897,
70
+ "support": 30257.0
71
+ },
72
+ "step": 81
73
+ },
74
+ {
75
+ "epoch": 2.0,
76
+ "eval_B-Claim": {
77
+ "f1-score": 0.46397188049209137,
78
+ "precision": 0.48175182481751827,
79
+ "recall": 0.44745762711864406,
80
+ "support": 295.0
81
+ },
82
+ "eval_B-MajorClaim": {
83
+ "f1-score": 0.6402439024390244,
84
+ "precision": 0.6104651162790697,
85
+ "recall": 0.6730769230769231,
86
+ "support": 156.0
87
+ },
88
+ "eval_B-Premise": {
89
+ "f1-score": 0.7515299877600979,
90
+ "precision": 0.6769570011025359,
91
+ "recall": 0.844566712517194,
92
+ "support": 727.0
93
+ },
94
+ "eval_I-Claim": {
95
+ "f1-score": 0.5573419078242231,
96
+ "precision": 0.5500705218617772,
97
+ "recall": 0.5648081100651702,
98
+ "support": 4143.0
99
+ },
100
+ "eval_I-MajorClaim": {
101
+ "f1-score": 0.7704582651391162,
102
+ "precision": 0.6987012987012987,
103
+ "recall": 0.8586411308709531,
104
+ "support": 2193.0
105
+ },
106
+ "eval_I-Premise": {
107
+ "f1-score": 0.8736837792436597,
108
+ "precision": 0.9002026342451874,
109
+ "recall": 0.8486826394969355,
110
+ "support": 12563.0
111
+ },
112
+ "eval_O": {
113
+ "f1-score": 0.9114385688236163,
114
+ "precision": 0.9145485115221046,
115
+ "recall": 0.9083497053045186,
116
+ "support": 10180.0
117
+ },
118
+ "eval_accuracy": 0.8256932280133523,
119
+ "eval_loss": 0.4454804062843323,
120
+ "eval_macro avg": {
121
+ "f1-score": 0.7098097559602613,
122
+ "precision": 0.6903852726470703,
123
+ "recall": 0.7350832640643341,
124
+ "support": 30257.0
125
+ },
126
+ "eval_runtime": 5.0817,
127
+ "eval_samples_per_second": 15.94,
128
+ "eval_steps_per_second": 2.165,
129
+ "eval_weighted avg": {
130
+ "f1-score": 0.8274557458920446,
131
+ "precision": 0.8315444607100787,
132
+ "recall": 0.8256932280133523,
133
+ "support": 30257.0
134
+ },
135
+ "step": 162
136
+ },
137
+ {
138
+ "epoch": 3.0,
139
+ "eval_B-Claim": {
140
+ "f1-score": 0.5551724137931033,
141
+ "precision": 0.5649122807017544,
142
+ "recall": 0.5457627118644067,
143
+ "support": 295.0
144
+ },
145
+ "eval_B-MajorClaim": {
146
+ "f1-score": 0.7108433734939757,
147
+ "precision": 0.6704545454545454,
148
+ "recall": 0.7564102564102564,
149
+ "support": 156.0
150
+ },
151
+ "eval_B-Premise": {
152
+ "f1-score": 0.7860892388451444,
153
+ "precision": 0.7515683814303639,
154
+ "recall": 0.8239339752407153,
155
+ "support": 727.0
156
+ },
157
+ "eval_I-Claim": {
158
+ "f1-score": 0.5878616796047988,
159
+ "precision": 0.573296627666896,
160
+ "recall": 0.6031860970311369,
161
+ "support": 4143.0
162
+ },
163
+ "eval_I-MajorClaim": {
164
+ "f1-score": 0.7989093387866394,
165
+ "precision": 0.7961956521739131,
166
+ "recall": 0.801641586867305,
167
+ "support": 2193.0
168
+ },
169
+ "eval_I-Premise": {
170
+ "f1-score": 0.8824656529006482,
171
+ "precision": 0.9044793581815143,
172
+ "recall": 0.8614980498288626,
173
+ "support": 12563.0
174
+ },
175
+ "eval_O": {
176
+ "f1-score": 0.9169814976266588,
177
+ "precision": 0.9044525128989107,
178
+ "recall": 0.9298624754420433,
179
+ "support": 10180.0
180
+ },
181
+ "eval_accuracy": 0.8402683676504611,
182
+ "eval_loss": 0.41143855452537537,
183
+ "eval_macro avg": {
184
+ "f1-score": 0.7483318850072812,
185
+ "precision": 0.7379084797868425,
186
+ "recall": 0.7603278789549609,
187
+ "support": 30257.0
188
+ },
189
+ "eval_runtime": 5.0654,
190
+ "eval_samples_per_second": 15.991,
191
+ "eval_steps_per_second": 2.172,
192
+ "eval_weighted avg": {
193
+ "f1-score": 0.8412916372002002,
194
+ "precision": 0.8430827906594975,
195
+ "recall": 0.8402683676504611,
196
+ "support": 30257.0
197
+ },
198
+ "step": 243
199
+ },
200
+ {
201
+ "epoch": 4.0,
202
+ "eval_B-Claim": {
203
+ "f1-score": 0.5518394648829431,
204
+ "precision": 0.5445544554455446,
205
+ "recall": 0.559322033898305,
206
+ "support": 295.0
207
+ },
208
+ "eval_B-MajorClaim": {
209
+ "f1-score": 0.7048192771084337,
210
+ "precision": 0.6647727272727273,
211
+ "recall": 0.75,
212
+ "support": 156.0
213
+ },
214
+ "eval_B-Premise": {
215
+ "f1-score": 0.792225201072386,
216
+ "precision": 0.7725490196078432,
217
+ "recall": 0.81292984869326,
218
+ "support": 727.0
219
+ },
220
+ "eval_I-Claim": {
221
+ "f1-score": 0.577999737153371,
222
+ "precision": 0.6344489324870167,
223
+ "recall": 0.5307748008689356,
224
+ "support": 4143.0
225
+ },
226
+ "eval_I-MajorClaim": {
227
+ "f1-score": 0.800089746466233,
228
+ "precision": 0.7875441696113075,
229
+ "recall": 0.8130414956680346,
230
+ "support": 2193.0
231
+ },
232
+ "eval_I-Premise": {
233
+ "f1-score": 0.8919449901768174,
234
+ "precision": 0.8642131979695431,
235
+ "recall": 0.9215155615696887,
236
+ "support": 12563.0
237
+ },
238
+ "eval_O": {
239
+ "f1-score": 0.9180246175312703,
240
+ "precision": 0.9316273895013655,
241
+ "recall": 0.9048133595284872,
242
+ "support": 10180.0
243
+ },
244
+ "eval_accuracy": 0.8475063621641273,
245
+ "eval_loss": 0.476072758436203,
246
+ "eval_macro avg": {
247
+ "f1-score": 0.7481347191987793,
248
+ "precision": 0.7428156988421926,
249
+ "recall": 0.7560567286038158,
250
+ "support": 30257.0
251
+ },
252
+ "eval_runtime": 5.0796,
253
+ "eval_samples_per_second": 15.946,
254
+ "eval_steps_per_second": 2.166,
255
+ "eval_weighted avg": {
256
+ "f1-score": 0.8443975745657389,
257
+ "precision": 0.8435295888400192,
258
+ "recall": 0.8475063621641273,
259
+ "support": 30257.0
260
+ },
261
+ "step": 324
262
+ },
263
+ {
264
+ "epoch": 5.0,
265
+ "eval_B-Claim": {
266
+ "f1-score": 0.5932203389830508,
267
+ "precision": 0.5932203389830508,
268
+ "recall": 0.5932203389830508,
269
+ "support": 295.0
270
+ },
271
+ "eval_B-MajorClaim": {
272
+ "f1-score": 0.7398119122257053,
273
+ "precision": 0.7239263803680982,
274
+ "recall": 0.7564102564102564,
275
+ "support": 156.0
276
+ },
277
+ "eval_B-Premise": {
278
+ "f1-score": 0.7774172615184944,
279
+ "precision": 0.7358722358722358,
280
+ "recall": 0.8239339752407153,
281
+ "support": 727.0
282
+ },
283
+ "eval_I-Claim": {
284
+ "f1-score": 0.5827523883294604,
285
+ "precision": 0.6264224257563141,
286
+ "recall": 0.5447743181269611,
287
+ "support": 4143.0
288
+ },
289
+ "eval_I-MajorClaim": {
290
+ "f1-score": 0.79782711384034,
291
+ "precision": 0.8275355218030377,
292
+ "recall": 0.7701778385772914,
293
+ "support": 2193.0
294
+ },
295
+ "eval_I-Premise": {
296
+ "f1-score": 0.8862478777589133,
297
+ "precision": 0.8792698213726104,
298
+ "recall": 0.8933375786038367,
299
+ "support": 12563.0
300
+ },
301
+ "eval_O": {
302
+ "f1-score": 0.9135231488172665,
303
+ "precision": 0.8963789354259242,
304
+ "recall": 0.931335952848723,
305
+ "support": 10180.0
306
+ },
307
+ "eval_accuracy": 0.8441682916349935,
308
+ "eval_loss": 0.526358962059021,
309
+ "eval_macro avg": {
310
+ "f1-score": 0.7558285773533188,
311
+ "precision": 0.7546608085116101,
312
+ "recall": 0.7590271798272621,
313
+ "support": 30257.0
314
+ },
315
+ "eval_runtime": 5.076,
316
+ "eval_samples_per_second": 15.958,
317
+ "eval_steps_per_second": 2.167,
318
+ "eval_weighted avg": {
319
+ "f1-score": 0.841232400970915,
320
+ "precision": 0.8396195745969204,
321
+ "recall": 0.8441682916349935,
322
+ "support": 30257.0
323
+ },
324
+ "step": 405
325
+ },
326
+ {
327
+ "epoch": 6.0,
328
+ "eval_B-Claim": {
329
+ "f1-score": 0.5809682804674458,
330
+ "precision": 0.5723684210526315,
331
+ "recall": 0.5898305084745763,
332
+ "support": 295.0
333
+ },
334
+ "eval_B-MajorClaim": {
335
+ "f1-score": 0.7402597402597402,
336
+ "precision": 0.75,
337
+ "recall": 0.7307692307692307,
338
+ "support": 156.0
339
+ },
340
+ "eval_B-Premise": {
341
+ "f1-score": 0.7956298200514138,
342
+ "precision": 0.7466827503015682,
343
+ "recall": 0.8514442916093535,
344
+ "support": 727.0
345
+ },
346
+ "eval_I-Claim": {
347
+ "f1-score": 0.580742605412209,
348
+ "precision": 0.6067859021567596,
349
+ "recall": 0.556842867487328,
350
+ "support": 4143.0
351
+ },
352
+ "eval_I-MajorClaim": {
353
+ "f1-score": 0.798828125,
354
+ "precision": 0.859695218076721,
355
+ "recall": 0.7460100319197447,
356
+ "support": 2193.0
357
+ },
358
+ "eval_I-Premise": {
359
+ "f1-score": 0.8905630763194036,
360
+ "precision": 0.8781922303049063,
361
+ "recall": 0.9032874313460161,
362
+ "support": 12563.0
363
+ },
364
+ "eval_O": {
365
+ "f1-score": 0.9183922046285018,
366
+ "precision": 0.9110681488641856,
367
+ "recall": 0.9258349705304518,
368
+ "support": 10180.0
369
+ },
370
+ "eval_accuracy": 0.8468453580989523,
371
+ "eval_loss": 0.5279645919799805,
372
+ "eval_macro avg": {
373
+ "f1-score": 0.7579119788769592,
374
+ "precision": 0.7606846672509676,
375
+ "recall": 0.7577170474481001,
376
+ "support": 30257.0
377
+ },
378
+ "eval_runtime": 5.0684,
379
+ "eval_samples_per_second": 15.981,
380
+ "eval_steps_per_second": 2.17,
381
+ "eval_weighted avg": {
382
+ "f1-score": 0.8447801270668736,
383
+ "precision": 0.8439473640584135,
384
+ "recall": 0.8468453580989523,
385
+ "support": 30257.0
386
+ },
387
+ "step": 486
388
+ },
389
+ {
390
+ "epoch": 6.17,
391
+ "grad_norm": 3.7896041870117188,
392
+ "learning_rate": 1.7530864197530865e-05,
393
+ "loss": 0.4007,
394
+ "step": 500
395
+ },
396
+ {
397
+ "epoch": 7.0,
398
+ "eval_B-Claim": {
399
+ "f1-score": 0.5654450261780105,
400
+ "precision": 0.5827338129496403,
401
+ "recall": 0.5491525423728814,
402
+ "support": 295.0
403
+ },
404
+ "eval_B-MajorClaim": {
405
+ "f1-score": 0.7454545454545455,
406
+ "precision": 0.7068965517241379,
407
+ "recall": 0.7884615384615384,
408
+ "support": 156.0
409
+ },
410
+ "eval_B-Premise": {
411
+ "f1-score": 0.8080155138978669,
412
+ "precision": 0.7621951219512195,
413
+ "recall": 0.859697386519945,
414
+ "support": 727.0
415
+ },
416
+ "eval_I-Claim": {
417
+ "f1-score": 0.5279187817258884,
418
+ "precision": 0.6115702479338843,
419
+ "recall": 0.4643977793869177,
420
+ "support": 4143.0
421
+ },
422
+ "eval_I-MajorClaim": {
423
+ "f1-score": 0.8239801667793555,
424
+ "precision": 0.8146167557932263,
425
+ "recall": 0.833561331509348,
426
+ "support": 2193.0
427
+ },
428
+ "eval_I-Premise": {
429
+ "f1-score": 0.8874376008707728,
430
+ "precision": 0.8396306818181818,
431
+ "recall": 0.9410172729443604,
432
+ "support": 12563.0
433
+ },
434
+ "eval_O": {
435
+ "f1-score": 0.916984006092917,
436
+ "precision": 0.9490278507619548,
437
+ "recall": 0.8870333988212181,
438
+ "support": 10180.0
439
+ },
440
+ "eval_accuracy": 0.8432428859437485,
441
+ "eval_loss": 0.673554003238678,
442
+ "eval_macro avg": {
443
+ "f1-score": 0.7536050915713366,
444
+ "precision": 0.7523815747046064,
445
+ "recall": 0.7604744642880299,
446
+ "support": 30257.0
447
+ },
448
+ "eval_runtime": 5.0724,
449
+ "eval_samples_per_second": 15.969,
450
+ "eval_steps_per_second": 2.169,
451
+ "eval_weighted avg": {
452
+ "f1-score": 0.8377716313785262,
453
+ "precision": 0.8383472270757442,
454
+ "recall": 0.8432428859437485,
455
+ "support": 30257.0
456
+ },
457
+ "step": 567
458
+ },
459
+ {
460
+ "epoch": 8.0,
461
+ "eval_B-Claim": {
462
+ "f1-score": 0.5714285714285714,
463
+ "precision": 0.554140127388535,
464
+ "recall": 0.5898305084745763,
465
+ "support": 295.0
466
+ },
467
+ "eval_B-MajorClaim": {
468
+ "f1-score": 0.7103825136612022,
469
+ "precision": 0.6190476190476191,
470
+ "recall": 0.8333333333333334,
471
+ "support": 156.0
472
+ },
473
+ "eval_B-Premise": {
474
+ "f1-score": 0.7976111479761113,
475
+ "precision": 0.7705128205128206,
476
+ "recall": 0.8266850068775791,
477
+ "support": 727.0
478
+ },
479
+ "eval_I-Claim": {
480
+ "f1-score": 0.5769132977370963,
481
+ "precision": 0.6094547408004297,
482
+ "recall": 0.5476707699734492,
483
+ "support": 4143.0
484
+ },
485
+ "eval_I-MajorClaim": {
486
+ "f1-score": 0.7987654320987654,
487
+ "precision": 0.7277840269966255,
488
+ "recall": 0.8850889192886456,
489
+ "support": 2193.0
490
+ },
491
+ "eval_I-Premise": {
492
+ "f1-score": 0.8899450117831893,
493
+ "precision": 0.8784213382957277,
494
+ "recall": 0.9017750537292049,
495
+ "support": 12563.0
496
+ },
497
+ "eval_O": {
498
+ "f1-score": 0.9160536128187039,
499
+ "precision": 0.9404096834264432,
500
+ "recall": 0.8929273084479371,
501
+ "support": 10180.0
502
+ },
503
+ "eval_accuracy": 0.8439038900089235,
504
+ "eval_loss": 0.6914934515953064,
505
+ "eval_macro avg": {
506
+ "f1-score": 0.7515856553576629,
507
+ "precision": 0.7285386223526001,
508
+ "recall": 0.7824729857321036,
509
+ "support": 30257.0
510
+ },
511
+ "eval_runtime": 5.0936,
512
+ "eval_samples_per_second": 15.902,
513
+ "eval_steps_per_second": 2.16,
514
+ "eval_weighted avg": {
515
+ "f1-score": 0.843008353502502,
516
+ "precision": 0.8444388009152946,
517
+ "recall": 0.8439038900089235,
518
+ "support": 30257.0
519
+ },
520
+ "step": 648
521
+ },
522
+ {
523
+ "epoch": 9.0,
524
+ "eval_B-Claim": {
525
+ "f1-score": 0.5940246045694201,
526
+ "precision": 0.6167883211678832,
527
+ "recall": 0.5728813559322034,
528
+ "support": 295.0
529
+ },
530
+ "eval_B-MajorClaim": {
531
+ "f1-score": 0.7795527156549521,
532
+ "precision": 0.7770700636942676,
533
+ "recall": 0.782051282051282,
534
+ "support": 156.0
535
+ },
536
+ "eval_B-Premise": {
537
+ "f1-score": 0.8103896103896104,
538
+ "precision": 0.7675276752767528,
539
+ "recall": 0.8583218707015131,
540
+ "support": 727.0
541
+ },
542
+ "eval_I-Claim": {
543
+ "f1-score": 0.597254004576659,
544
+ "precision": 0.6309427880741337,
545
+ "recall": 0.5669804489500362,
546
+ "support": 4143.0
547
+ },
548
+ "eval_I-MajorClaim": {
549
+ "f1-score": 0.8170818505338079,
550
+ "precision": 0.8516320474777448,
551
+ "recall": 0.7852257181942545,
552
+ "support": 2193.0
553
+ },
554
+ "eval_I-Premise": {
555
+ "f1-score": 0.8963960457694404,
556
+ "precision": 0.8770086055898256,
557
+ "recall": 0.9166600334315053,
558
+ "support": 12563.0
559
+ },
560
+ "eval_O": {
561
+ "f1-score": 0.9201161588817246,
562
+ "precision": 0.9220676728815231,
563
+ "recall": 0.9181728880157171,
564
+ "support": 10180.0
565
+ },
566
+ "eval_accuracy": 0.8543147040354299,
567
+ "eval_loss": 0.6210296154022217,
568
+ "eval_macro avg": {
569
+ "f1-score": 0.7735449986250879,
570
+ "precision": 0.7775767391660188,
571
+ "recall": 0.7714705138966444,
572
+ "support": 30257.0
573
+ },
574
+ "eval_runtime": 5.1135,
575
+ "eval_samples_per_second": 15.84,
576
+ "eval_steps_per_second": 2.151,
577
+ "eval_weighted avg": {
578
+ "f1-score": 0.8520504540550145,
579
+ "precision": 0.8509535372864376,
580
+ "recall": 0.8543147040354299,
581
+ "support": 30257.0
582
+ },
583
+ "step": 729
584
+ },
585
+ {
586
+ "epoch": 10.0,
587
+ "eval_B-Claim": {
588
+ "f1-score": 0.6135338345864663,
589
+ "precision": 0.5513513513513514,
590
+ "recall": 0.6915254237288135,
591
+ "support": 295.0
592
+ },
593
+ "eval_B-MajorClaim": {
594
+ "f1-score": 0.7411764705882354,
595
+ "precision": 0.6847826086956522,
596
+ "recall": 0.8076923076923077,
597
+ "support": 156.0
598
+ },
599
+ "eval_B-Premise": {
600
+ "f1-score": 0.7975708502024291,
601
+ "precision": 0.7827814569536424,
602
+ "recall": 0.81292984869326,
603
+ "support": 727.0
604
+ },
605
+ "eval_I-Claim": {
606
+ "f1-score": 0.6205202634810763,
607
+ "precision": 0.5772746157041961,
608
+ "recall": 0.6707699734491914,
609
+ "support": 4143.0
610
+ },
611
+ "eval_I-MajorClaim": {
612
+ "f1-score": 0.7967443906731192,
613
+ "precision": 0.7696557586060349,
614
+ "recall": 0.8258093935248518,
615
+ "support": 2193.0
616
+ },
617
+ "eval_I-Premise": {
618
+ "f1-score": 0.8895712958099318,
619
+ "precision": 0.9033316921056951,
620
+ "recall": 0.876223831887288,
621
+ "support": 12563.0
622
+ },
623
+ "eval_O": {
624
+ "f1-score": 0.9100379266750948,
625
+ "precision": 0.9377800937988535,
626
+ "recall": 0.8838899803536345,
627
+ "support": 10180.0
628
+ },
629
+ "eval_accuracy": 0.8433420365535248,
630
+ "eval_loss": 0.7195751070976257,
631
+ "eval_macro avg": {
632
+ "f1-score": 0.7670221474309076,
633
+ "precision": 0.7438510824593465,
634
+ "recall": 0.7955486799041925,
635
+ "support": 30257.0
636
+ },
637
+ "eval_runtime": 5.1021,
638
+ "eval_samples_per_second": 15.876,
639
+ "eval_steps_per_second": 2.156,
640
+ "eval_weighted avg": {
641
+ "f1-score": 0.847222004887227,
642
+ "precision": 0.8531321039459353,
643
+ "recall": 0.8433420365535248,
644
+ "support": 30257.0
645
+ },
646
+ "step": 810
647
+ },
648
+ {
649
+ "epoch": 11.0,
650
+ "eval_B-Claim": {
651
+ "f1-score": 0.6165884194053208,
652
+ "precision": 0.5726744186046512,
653
+ "recall": 0.6677966101694915,
654
+ "support": 295.0
655
+ },
656
+ "eval_B-MajorClaim": {
657
+ "f1-score": 0.7208672086720868,
658
+ "precision": 0.6244131455399061,
659
+ "recall": 0.8525641025641025,
660
+ "support": 156.0
661
+ },
662
+ "eval_B-Premise": {
663
+ "f1-score": 0.8049792531120332,
664
+ "precision": 0.8094575799721836,
665
+ "recall": 0.8005502063273727,
666
+ "support": 727.0
667
+ },
668
+ "eval_I-Claim": {
669
+ "f1-score": 0.6097560975609757,
670
+ "precision": 0.587642713230356,
671
+ "recall": 0.6335988414192614,
672
+ "support": 4143.0
673
+ },
674
+ "eval_I-MajorClaim": {
675
+ "f1-score": 0.7834637768412603,
676
+ "precision": 0.6996415770609319,
677
+ "recall": 0.8901048791609667,
678
+ "support": 2193.0
679
+ },
680
+ "eval_I-Premise": {
681
+ "f1-score": 0.8909987071751778,
682
+ "precision": 0.9046681434079908,
683
+ "recall": 0.8777362095040994,
684
+ "support": 12563.0
685
+ },
686
+ "eval_O": {
687
+ "f1-score": 0.9085467917829064,
688
+ "precision": 0.9392763502884112,
689
+ "recall": 0.8797642436149312,
690
+ "support": 10180.0
691
+ },
692
+ "eval_accuracy": 0.8418547774068811,
693
+ "eval_loss": 0.7594641447067261,
694
+ "eval_macro avg": {
695
+ "f1-score": 0.7621714649356802,
696
+ "precision": 0.7339677040149185,
697
+ "recall": 0.8003021561086037,
698
+ "support": 30257.0
699
+ },
700
+ "eval_runtime": 5.074,
701
+ "eval_samples_per_second": 15.964,
702
+ "eval_steps_per_second": 2.168,
703
+ "eval_weighted avg": {
704
+ "f1-score": 0.8449795901341711,
705
+ "precision": 0.8510731379783514,
706
+ "recall": 0.8418547774068811,
707
+ "support": 30257.0
708
+ },
709
+ "step": 891
710
+ },
711
+ {
712
+ "epoch": 12.0,
713
+ "eval_B-Claim": {
714
+ "f1-score": 0.6103448275862069,
715
+ "precision": 0.6210526315789474,
716
+ "recall": 0.6,
717
+ "support": 295.0
718
+ },
719
+ "eval_B-MajorClaim": {
720
+ "f1-score": 0.7815384615384615,
721
+ "precision": 0.7514792899408284,
722
+ "recall": 0.8141025641025641,
723
+ "support": 156.0
724
+ },
725
+ "eval_B-Premise": {
726
+ "f1-score": 0.8209255533199196,
727
+ "precision": 0.8010471204188482,
728
+ "recall": 0.8418156808803301,
729
+ "support": 727.0
730
+ },
731
+ "eval_I-Claim": {
732
+ "f1-score": 0.6016401590457257,
733
+ "precision": 0.6199743918053777,
734
+ "recall": 0.5843591600289645,
735
+ "support": 4143.0
736
+ },
737
+ "eval_I-MajorClaim": {
738
+ "f1-score": 0.8330076004343105,
739
+ "precision": 0.7951907131011609,
740
+ "recall": 0.8746010031919744,
741
+ "support": 2193.0
742
+ },
743
+ "eval_I-Premise": {
744
+ "f1-score": 0.8897350468031097,
745
+ "precision": 0.886710411890268,
746
+ "recall": 0.8927803868502746,
747
+ "support": 12563.0
748
+ },
749
+ "eval_O": {
750
+ "f1-score": 0.911568656495334,
751
+ "precision": 0.9164102054998511,
752
+ "recall": 0.906777996070727,
753
+ "support": 10180.0
754
+ },
755
+ "eval_accuracy": 0.8494563241563936,
756
+ "eval_loss": 0.8027455806732178,
757
+ "eval_macro avg": {
758
+ "f1-score": 0.7783943293175811,
759
+ "precision": 0.7702663948907544,
760
+ "recall": 0.7877766844464048,
761
+ "support": 30257.0
762
+ },
763
+ "eval_runtime": 5.087,
764
+ "eval_samples_per_second": 15.923,
765
+ "eval_steps_per_second": 2.162,
766
+ "eval_weighted avg": {
767
+ "f1-score": 0.848586302812254,
768
+ "precision": 0.8482006969540109,
769
+ "recall": 0.8494563241563936,
770
+ "support": 30257.0
771
+ },
772
+ "step": 972
773
+ }
774
+ ],
775
+ "logging_steps": 500,
776
+ "max_steps": 4050,
777
+ "num_input_tokens_seen": 0,
778
+ "num_train_epochs": 50,
779
+ "save_steps": 500,
780
+ "total_flos": 1720131088788000.0,
781
+ "train_batch_size": 4,
782
+ "trial_name": null,
783
+ "trial_params": null
784
+ }
checkpoint-972/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7cf5d4f8fefb4e78449403d6f60739ea086b5ecbb9bc2500c2c1407621dbdb5a
3
+ size 4856
checkpoint-972/vocab.json ADDED
The diff for this file is too large to render. See raw diff