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
2db1750
•
1 Parent(s): edd5974

Training in progress, epoch 17, checkpoint

Browse files
Files changed (12) hide show
  1. checkpoint-1377/config.json +57 -0
  2. checkpoint-1377/merges.txt +0 -0
  3. checkpoint-1377/model.safetensors +3 -0
  4. checkpoint-1377/optimizer.pt +3 -0
  5. checkpoint-1377/rng_state.pth +3 -0
  6. checkpoint-1377/scheduler.pt +3 -0
  7. checkpoint-1377/special_tokens_map.json +15 -0
  8. checkpoint-1377/tokenizer.json +0 -0
  9. checkpoint-1377/tokenizer_config.json +57 -0
  10. checkpoint-1377/trainer_state.json +800 -0
  11. checkpoint-1377/training_args.bin +3 -0
  12. checkpoint-1377/vocab.json +0 -0
checkpoint-1377/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-1377/merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-1377/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:005f50aa569c01dfaa3ac8b0657ecc85176103d751aafb02c106938ea6e57db6
3
+ size 592324828
checkpoint-1377/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:565ffafac0716876782da6a509e2afb9f47ef86fedf07562d3017eeb1bdb061a
3
+ size 1014657786
checkpoint-1377/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c1cd922b677a4c8500b488971373471495569ff9033c60b0a4e52298773e09e9
3
+ size 14244
checkpoint-1377/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:032f609362cfbe4a4e5efe180a3007cc3751761c4055f83fe66b2c2d7b6093ac
3
+ size 1064
checkpoint-1377/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-1377/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-1377/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-1377/trainer_state.json ADDED
@@ -0,0 +1,800 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 17.0,
5
+ "eval_steps": 500,
6
+ "global_step": 1377,
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": 0.46243291592128805,
15
+ "precision": 0.6588785046728972,
16
+ "recall": 0.3562241616903996,
17
+ "support": 4354.0
18
+ },
19
+ "eval_MajorClaim": {
20
+ "f1-score": 0.8342621259029929,
21
+ "precision": 0.8096955128205128,
22
+ "recall": 0.8603661132396765,
23
+ "support": 2349.0
24
+ },
25
+ "eval_O": {
26
+ "f1-score": 0.9993622448979592,
27
+ "precision": 0.9987253027405991,
28
+ "recall": 1.0,
29
+ "support": 12536.0
30
+ },
31
+ "eval_Premise": {
32
+ "f1-score": 0.8941052999825083,
33
+ "precision": 0.8401157057392676,
34
+ "recall": 0.955510692388216,
35
+ "support": 13374.0
36
+ },
37
+ "eval_accuracy": 0.8857510808573268,
38
+ "eval_loss": 0.2754688858985901,
39
+ "eval_macro avg": {
40
+ "f1-score": 0.7975406466761871,
41
+ "precision": 0.8268537564933192,
42
+ "recall": 0.793025241829573,
43
+ "support": 32613.0
44
+ },
45
+ "eval_runtime": 4.8781,
46
+ "eval_samples_per_second": 16.605,
47
+ "eval_steps_per_second": 2.255,
48
+ "eval_weighted avg": {
49
+ "f1-score": 0.8726239240080429,
50
+ "precision": 0.8746959682542021,
51
+ "recall": 0.8857510808573268,
52
+ "support": 32613.0
53
+ },
54
+ "step": 81
55
+ },
56
+ {
57
+ "epoch": 2.0,
58
+ "eval_Claim": {
59
+ "f1-score": 0.6446312571646924,
60
+ "precision": 0.5518482172064115,
61
+ "recall": 0.77491961414791,
62
+ "support": 4354.0
63
+ },
64
+ "eval_MajorClaim": {
65
+ "f1-score": 0.8912341158733577,
66
+ "precision": 0.9019180470793374,
67
+ "recall": 0.8808003405704555,
68
+ "support": 2349.0
69
+ },
70
+ "eval_O": {
71
+ "f1-score": 0.9998005504806733,
72
+ "precision": 0.9999202106439001,
73
+ "recall": 0.9996809189534142,
74
+ "support": 12536.0
75
+ },
76
+ "eval_Premise": {
77
+ "f1-score": 0.8690409646250897,
78
+ "precision": 0.9324023303632625,
79
+ "recall": 0.8137430835950351,
80
+ "support": 13374.0
81
+ },
82
+ "eval_accuracy": 0.8848618649004998,
83
+ "eval_loss": 0.25178349018096924,
84
+ "eval_macro avg": {
85
+ "f1-score": 0.8511767220359533,
86
+ "precision": 0.8465222013232279,
87
+ "recall": 0.8672859893167036,
88
+ "support": 32613.0
89
+ },
90
+ "eval_runtime": 4.907,
91
+ "eval_samples_per_second": 16.507,
92
+ "eval_steps_per_second": 2.242,
93
+ "eval_weighted avg": {
94
+ "f1-score": 0.8909418634778542,
95
+ "precision": 0.9053537287957648,
96
+ "recall": 0.8848618649004998,
97
+ "support": 32613.0
98
+ },
99
+ "step": 162
100
+ },
101
+ {
102
+ "epoch": 3.0,
103
+ "eval_Claim": {
104
+ "f1-score": 0.6598764019226366,
105
+ "precision": 0.6576186131386861,
106
+ "recall": 0.6621497473587505,
107
+ "support": 4354.0
108
+ },
109
+ "eval_MajorClaim": {
110
+ "f1-score": 0.9062119366626066,
111
+ "precision": 0.8661233993015134,
112
+ "recall": 0.9501915708812261,
113
+ "support": 2349.0
114
+ },
115
+ "eval_O": {
116
+ "f1-score": 0.9998404722022812,
117
+ "precision": 0.9997607273887382,
118
+ "recall": 0.9999202297383536,
119
+ "support": 12536.0
120
+ },
121
+ "eval_Premise": {
122
+ "f1-score": 0.9007852612503775,
123
+ "precision": 0.9097148086014946,
124
+ "recall": 0.8920293106026619,
125
+ "support": 13374.0
126
+ },
127
+ "eval_accuracy": 0.9070002759635728,
128
+ "eval_loss": 0.23778581619262695,
129
+ "eval_macro avg": {
130
+ "f1-score": 0.8666785180094754,
131
+ "precision": 0.8583043871076081,
132
+ "recall": 0.876072714645248,
133
+ "support": 32613.0
134
+ },
135
+ "eval_runtime": 4.9168,
136
+ "eval_samples_per_second": 16.474,
137
+ "eval_steps_per_second": 2.237,
138
+ "eval_weighted avg": {
139
+ "f1-score": 0.907089072967282,
140
+ "precision": 0.9075314026721463,
141
+ "recall": 0.9070002759635728,
142
+ "support": 32613.0
143
+ },
144
+ "step": 243
145
+ },
146
+ {
147
+ "epoch": 4.0,
148
+ "eval_Claim": {
149
+ "f1-score": 0.6604639922667954,
150
+ "precision": 0.696838347781744,
151
+ "recall": 0.6276986678915939,
152
+ "support": 4354.0
153
+ },
154
+ "eval_MajorClaim": {
155
+ "f1-score": 0.9140214420853479,
156
+ "precision": 0.9028239202657807,
157
+ "recall": 0.9255002128565347,
158
+ "support": 2349.0
159
+ },
160
+ "eval_O": {
161
+ "f1-score": 0.9994417862838917,
162
+ "precision": 0.9991230867346939,
163
+ "recall": 0.9997606892150607,
164
+ "support": 12536.0
165
+ },
166
+ "eval_Premise": {
167
+ "f1-score": 0.9084203149780549,
168
+ "precision": 0.8963534463934785,
169
+ "recall": 0.9208165096455809,
170
+ "support": 13374.0
171
+ },
172
+ "eval_accuracy": 0.912366234323736,
173
+ "eval_loss": 0.2598011791706085,
174
+ "eval_macro avg": {
175
+ "f1-score": 0.8705868839035225,
176
+ "precision": 0.8737847002939243,
177
+ "recall": 0.8684440199021926,
178
+ "support": 32613.0
179
+ },
180
+ "eval_runtime": 4.8872,
181
+ "eval_samples_per_second": 16.574,
182
+ "eval_steps_per_second": 2.251,
183
+ "eval_weighted avg": {
184
+ "f1-score": 0.9107077581074873,
185
+ "precision": 0.9096864919608297,
186
+ "recall": 0.912366234323736,
187
+ "support": 32613.0
188
+ },
189
+ "step": 324
190
+ },
191
+ {
192
+ "epoch": 5.0,
193
+ "eval_Claim": {
194
+ "f1-score": 0.6731001206272618,
195
+ "precision": 0.6440713536201469,
196
+ "recall": 0.7048690858980248,
197
+ "support": 4354.0
198
+ },
199
+ "eval_MajorClaim": {
200
+ "f1-score": 0.8871466886734063,
201
+ "precision": 0.8606885508406725,
202
+ "recall": 0.9152830991911451,
203
+ "support": 2349.0
204
+ },
205
+ "eval_O": {
206
+ "f1-score": 0.9998005345673595,
207
+ "precision": 1.0,
208
+ "recall": 0.9996011486917677,
209
+ "support": 12536.0
210
+ },
211
+ "eval_Premise": {
212
+ "f1-score": 0.9039056236399038,
213
+ "precision": 0.9234729698104376,
214
+ "recall": 0.8851502916105877,
215
+ "support": 13374.0
216
+ },
217
+ "eval_accuracy": 0.9072455769171802,
218
+ "eval_loss": 0.3166182339191437,
219
+ "eval_macro avg": {
220
+ "f1-score": 0.8659882418769828,
221
+ "precision": 0.8570582185678143,
222
+ "recall": 0.8762259063478812,
223
+ "support": 32613.0
224
+ },
225
+ "eval_runtime": 4.9018,
226
+ "eval_samples_per_second": 16.525,
227
+ "eval_steps_per_second": 2.244,
228
+ "eval_weighted avg": {
229
+ "f1-score": 0.9087455557232214,
230
+ "precision": 0.9110652677714914,
231
+ "recall": 0.9072455769171802,
232
+ "support": 32613.0
233
+ },
234
+ "step": 405
235
+ },
236
+ {
237
+ "epoch": 6.0,
238
+ "eval_Claim": {
239
+ "f1-score": 0.634920634920635,
240
+ "precision": 0.7171775592828225,
241
+ "recall": 0.5695911805236564,
242
+ "support": 4354.0
243
+ },
244
+ "eval_MajorClaim": {
245
+ "f1-score": 0.9061357171881348,
246
+ "precision": 0.8666925767586475,
247
+ "recall": 0.9493401447424435,
248
+ "support": 2349.0
249
+ },
250
+ "eval_O": {
251
+ "f1-score": 0.9998404722022812,
252
+ "precision": 0.9997607273887382,
253
+ "recall": 0.9999202297383536,
254
+ "support": 12536.0
255
+ },
256
+ "eval_Premise": {
257
+ "f1-score": 0.9096214165876431,
258
+ "precision": 0.8879236684705212,
259
+ "recall": 0.9324061612083147,
260
+ "support": 13374.0
261
+ },
262
+ "eval_accuracy": 0.9111397295556987,
263
+ "eval_loss": 0.32181796431541443,
264
+ "eval_macro avg": {
265
+ "f1-score": 0.8626295602246735,
266
+ "precision": 0.8678886329751823,
267
+ "recall": 0.8628144290531921,
268
+ "support": 32613.0
269
+ },
270
+ "eval_runtime": 4.8856,
271
+ "eval_samples_per_second": 16.579,
272
+ "eval_steps_per_second": 2.252,
273
+ "eval_weighted avg": {
274
+ "f1-score": 0.907375409471386,
275
+ "precision": 0.9065876667768511,
276
+ "recall": 0.9111397295556987,
277
+ "support": 32613.0
278
+ },
279
+ "step": 486
280
+ },
281
+ {
282
+ "epoch": 6.17,
283
+ "grad_norm": 7.612313747406006,
284
+ "learning_rate": 1.7530864197530865e-05,
285
+ "loss": 0.2088,
286
+ "step": 500
287
+ },
288
+ {
289
+ "epoch": 7.0,
290
+ "eval_Claim": {
291
+ "f1-score": 0.6538414409998775,
292
+ "precision": 0.7008142894667717,
293
+ "recall": 0.6127698667891593,
294
+ "support": 4354.0
295
+ },
296
+ "eval_MajorClaim": {
297
+ "f1-score": 0.8866711022399646,
298
+ "precision": 0.9254629629629629,
299
+ "recall": 0.8510004257130694,
300
+ "support": 2349.0
301
+ },
302
+ "eval_O": {
303
+ "f1-score": 0.999880349379811,
304
+ "precision": 0.9998404722022812,
305
+ "recall": 0.9999202297383536,
306
+ "support": 12536.0
307
+ },
308
+ "eval_Premise": {
309
+ "f1-score": 0.9127824473310774,
310
+ "precision": 0.8890070167977886,
311
+ "recall": 0.9378645132346344,
312
+ "support": 13374.0
313
+ },
314
+ "eval_accuracy": 0.9120596081317266,
315
+ "eval_loss": 0.4053190350532532,
316
+ "eval_macro avg": {
317
+ "f1-score": 0.8632938349876826,
318
+ "precision": 0.8787811853574512,
319
+ "recall": 0.8503887588688042,
320
+ "support": 32613.0
321
+ },
322
+ "eval_runtime": 4.9061,
323
+ "eval_samples_per_second": 16.51,
324
+ "eval_steps_per_second": 2.242,
325
+ "eval_weighted avg": {
326
+ "f1-score": 0.909811074225195,
327
+ "precision": 0.9091110268457285,
328
+ "recall": 0.9120596081317266,
329
+ "support": 32613.0
330
+ },
331
+ "step": 567
332
+ },
333
+ {
334
+ "epoch": 8.0,
335
+ "eval_Claim": {
336
+ "f1-score": 0.668355545764952,
337
+ "precision": 0.6421164021164021,
338
+ "recall": 0.6968305006890216,
339
+ "support": 4354.0
340
+ },
341
+ "eval_MajorClaim": {
342
+ "f1-score": 0.8960263537162857,
343
+ "precision": 0.8676236044657097,
344
+ "recall": 0.9263516389953171,
345
+ "support": 2349.0
346
+ },
347
+ "eval_O": {
348
+ "f1-score": 1.0,
349
+ "precision": 1.0,
350
+ "recall": 1.0,
351
+ "support": 12536.0
352
+ },
353
+ "eval_Premise": {
354
+ "f1-score": 0.9012129071630178,
355
+ "precision": 0.9198069137340392,
356
+ "recall": 0.8833557649170032,
357
+ "support": 13374.0
358
+ },
359
+ "eval_accuracy": 0.9063870235795541,
360
+ "eval_loss": 0.4514281749725342,
361
+ "eval_macro avg": {
362
+ "f1-score": 0.8663987016610639,
363
+ "precision": 0.8573867300790377,
364
+ "recall": 0.8766344761503355,
365
+ "support": 32613.0
366
+ },
367
+ "eval_runtime": 4.8816,
368
+ "eval_samples_per_second": 16.593,
369
+ "eval_steps_per_second": 2.253,
370
+ "eval_weighted avg": {
371
+ "f1-score": 0.9077241398073883,
372
+ "precision": 0.9098003963445191,
373
+ "recall": 0.9063870235795541,
374
+ "support": 32613.0
375
+ },
376
+ "step": 648
377
+ },
378
+ {
379
+ "epoch": 9.0,
380
+ "eval_Claim": {
381
+ "f1-score": 0.6913841807909605,
382
+ "precision": 0.7090777402221149,
383
+ "recall": 0.674552135966927,
384
+ "support": 4354.0
385
+ },
386
+ "eval_MajorClaim": {
387
+ "f1-score": 0.9055304497525284,
388
+ "precision": 0.9155787641427328,
389
+ "recall": 0.8957002979991485,
390
+ "support": 2349.0
391
+ },
392
+ "eval_O": {
393
+ "f1-score": 0.9998404340194671,
394
+ "precision": 1.0,
395
+ "recall": 0.9996809189534142,
396
+ "support": 12536.0
397
+ },
398
+ "eval_Premise": {
399
+ "f1-score": 0.9162317231167869,
400
+ "precision": 0.9072648632798181,
401
+ "recall": 0.9253775983251085,
402
+ "support": 13374.0
403
+ },
404
+ "eval_accuracy": 0.9183147824487168,
405
+ "eval_loss": 0.4471658766269684,
406
+ "eval_macro avg": {
407
+ "f1-score": 0.8782466969199356,
408
+ "precision": 0.8829803419111664,
409
+ "recall": 0.8738277378111496,
410
+ "support": 32613.0
411
+ },
412
+ "eval_runtime": 4.8813,
413
+ "eval_samples_per_second": 16.594,
414
+ "eval_steps_per_second": 2.253,
415
+ "eval_weighted avg": {
416
+ "f1-score": 0.9175807345372852,
417
+ "precision": 0.9170508471898524,
418
+ "recall": 0.9183147824487168,
419
+ "support": 32613.0
420
+ },
421
+ "step": 729
422
+ },
423
+ {
424
+ "epoch": 10.0,
425
+ "eval_Claim": {
426
+ "f1-score": 0.6829727187206021,
427
+ "precision": 0.6997590361445784,
428
+ "recall": 0.6669728984841525,
429
+ "support": 4354.0
430
+ },
431
+ "eval_MajorClaim": {
432
+ "f1-score": 0.908567906307787,
433
+ "precision": 0.8780778395552026,
434
+ "recall": 0.9412515964240102,
435
+ "support": 2349.0
436
+ },
437
+ "eval_O": {
438
+ "f1-score": 0.9999202233745512,
439
+ "precision": 1.0,
440
+ "recall": 0.9998404594767071,
441
+ "support": 12536.0
442
+ },
443
+ "eval_Premise": {
444
+ "f1-score": 0.9131976852716073,
445
+ "precision": 0.9119379613749907,
446
+ "recall": 0.914460894272469,
447
+ "support": 13374.0
448
+ },
449
+ "eval_accuracy": 0.9161683991046515,
450
+ "eval_loss": 0.45130759477615356,
451
+ "eval_macro avg": {
452
+ "f1-score": 0.8761646334186369,
453
+ "precision": 0.8724437092686929,
454
+ "recall": 0.8806314621643347,
455
+ "support": 32613.0
456
+ },
457
+ "eval_runtime": 4.8986,
458
+ "eval_samples_per_second": 16.535,
459
+ "eval_steps_per_second": 2.246,
460
+ "eval_weighted avg": {
461
+ "f1-score": 0.9154630053129839,
462
+ "precision": 0.9150220459300829,
463
+ "recall": 0.9161683991046515,
464
+ "support": 32613.0
465
+ },
466
+ "step": 810
467
+ },
468
+ {
469
+ "epoch": 11.0,
470
+ "eval_Claim": {
471
+ "f1-score": 0.6209771986970684,
472
+ "precision": 0.717554953327311,
473
+ "recall": 0.5473128158015618,
474
+ "support": 4354.0
475
+ },
476
+ "eval_MajorClaim": {
477
+ "f1-score": 0.8710538016896399,
478
+ "precision": 0.9115867845509539,
479
+ "recall": 0.8339719029374202,
480
+ "support": 2349.0
481
+ },
482
+ "eval_O": {
483
+ "f1-score": 1.0,
484
+ "precision": 1.0,
485
+ "recall": 1.0,
486
+ "support": 12536.0
487
+ },
488
+ "eval_Premise": {
489
+ "f1-score": 0.9144776812837283,
490
+ "precision": 0.8758814267132197,
491
+ "recall": 0.9566322715717063,
492
+ "support": 13374.0
493
+ },
494
+ "eval_accuracy": 0.9098212369300586,
495
+ "eval_loss": 0.5217961668968201,
496
+ "eval_macro avg": {
497
+ "f1-score": 0.8516271704176092,
498
+ "precision": 0.8762557911478711,
499
+ "recall": 0.8344792475776721,
500
+ "support": 32613.0
501
+ },
502
+ "eval_runtime": 4.8915,
503
+ "eval_samples_per_second": 16.559,
504
+ "eval_steps_per_second": 2.249,
505
+ "eval_weighted avg": {
506
+ "f1-score": 0.9050398495319223,
507
+ "precision": 0.9050252912813879,
508
+ "recall": 0.9098212369300586,
509
+ "support": 32613.0
510
+ },
511
+ "step": 891
512
+ },
513
+ {
514
+ "epoch": 12.0,
515
+ "eval_Claim": {
516
+ "f1-score": 0.6767220902612826,
517
+ "precision": 0.7006886374815543,
518
+ "recall": 0.6543408360128617,
519
+ "support": 4354.0
520
+ },
521
+ "eval_MajorClaim": {
522
+ "f1-score": 0.9093525179856115,
523
+ "precision": 0.8791732909379968,
524
+ "recall": 0.9416773094934014,
525
+ "support": 2349.0
526
+ },
527
+ "eval_O": {
528
+ "f1-score": 0.9999202233745512,
529
+ "precision": 1.0,
530
+ "recall": 0.9998404594767071,
531
+ "support": 12536.0
532
+ },
533
+ "eval_Premise": {
534
+ "f1-score": 0.9131033456142309,
535
+ "precision": 0.9089427280136326,
536
+ "recall": 0.9173022282039779,
537
+ "support": 13374.0
538
+ },
539
+ "eval_accuracy": 0.9156777971974366,
540
+ "eval_loss": 0.48604533076286316,
541
+ "eval_macro avg": {
542
+ "f1-score": 0.8747745443089191,
543
+ "precision": 0.872201164108296,
544
+ "recall": 0.878290208296737,
545
+ "support": 32613.0
546
+ },
547
+ "eval_runtime": 4.8984,
548
+ "eval_samples_per_second": 16.536,
549
+ "eval_steps_per_second": 2.246,
550
+ "eval_weighted avg": {
551
+ "f1-score": 0.9146463407295841,
552
+ "precision": 0.9139967630227935,
553
+ "recall": 0.9156777971974366,
554
+ "support": 32613.0
555
+ },
556
+ "step": 972
557
+ },
558
+ {
559
+ "epoch": 12.35,
560
+ "grad_norm": 1.4397014379501343,
561
+ "learning_rate": 1.506172839506173e-05,
562
+ "loss": 0.0253,
563
+ "step": 1000
564
+ },
565
+ {
566
+ "epoch": 13.0,
567
+ "eval_Claim": {
568
+ "f1-score": 0.6805793134907059,
569
+ "precision": 0.6758776896942242,
570
+ "recall": 0.6853468075333027,
571
+ "support": 4354.0
572
+ },
573
+ "eval_MajorClaim": {
574
+ "f1-score": 0.903767491926803,
575
+ "precision": 0.914198606271777,
576
+ "recall": 0.8935717326521925,
577
+ "support": 2349.0
578
+ },
579
+ "eval_O": {
580
+ "f1-score": 0.9998005982053838,
581
+ "precision": 0.9996809952946806,
582
+ "recall": 0.9999202297383536,
583
+ "support": 12536.0
584
+ },
585
+ "eval_Premise": {
586
+ "f1-score": 0.9103489546321576,
587
+ "precision": 0.9107236399012197,
588
+ "recall": 0.9099745775385075,
589
+ "support": 13374.0
590
+ },
591
+ "eval_accuracy": 0.9133781007573667,
592
+ "eval_loss": 0.5282083749771118,
593
+ "eval_macro avg": {
594
+ "f1-score": 0.8736240895637626,
595
+ "precision": 0.8751202327904755,
596
+ "recall": 0.872203336865589,
597
+ "support": 32613.0
598
+ },
599
+ "eval_runtime": 4.89,
600
+ "eval_samples_per_second": 16.564,
601
+ "eval_steps_per_second": 2.249,
602
+ "eval_weighted avg": {
603
+ "f1-score": 0.9135835215352088,
604
+ "precision": 0.9138148255025323,
605
+ "recall": 0.9133781007573667,
606
+ "support": 32613.0
607
+ },
608
+ "step": 1053
609
+ },
610
+ {
611
+ "epoch": 14.0,
612
+ "eval_Claim": {
613
+ "f1-score": 0.6549993880797944,
614
+ "precision": 0.7010741419963322,
615
+ "recall": 0.6146072576940744,
616
+ "support": 4354.0
617
+ },
618
+ "eval_MajorClaim": {
619
+ "f1-score": 0.9030425963488844,
620
+ "precision": 0.8624564122433166,
621
+ "recall": 0.9476372924648787,
622
+ "support": 2349.0
623
+ },
624
+ "eval_O": {
625
+ "f1-score": 0.9999601132782897,
626
+ "precision": 1.0,
627
+ "recall": 0.9999202297383536,
628
+ "support": 12536.0
629
+ },
630
+ "eval_Premise": {
631
+ "f1-score": 0.9098839358320396,
632
+ "precision": 0.8997076023391812,
633
+ "recall": 0.9202931060266188,
634
+ "support": 13374.0
635
+ },
636
+ "eval_accuracy": 0.9120596081317266,
637
+ "eval_loss": 0.5426023602485657,
638
+ "eval_macro avg": {
639
+ "f1-score": 0.866971508384752,
640
+ "precision": 0.8658095391447076,
641
+ "recall": 0.8706144714809814,
642
+ "support": 32613.0
643
+ },
644
+ "eval_runtime": 4.8976,
645
+ "eval_samples_per_second": 16.539,
646
+ "eval_steps_per_second": 2.246,
647
+ "eval_weighted avg": {
648
+ "f1-score": 0.9099868804586297,
649
+ "precision": 0.9090570140832118,
650
+ "recall": 0.9120596081317266,
651
+ "support": 32613.0
652
+ },
653
+ "step": 1134
654
+ },
655
+ {
656
+ "epoch": 15.0,
657
+ "eval_Claim": {
658
+ "f1-score": 0.6663435909861886,
659
+ "precision": 0.7051282051282052,
660
+ "recall": 0.6316031235645384,
661
+ "support": 4354.0
662
+ },
663
+ "eval_MajorClaim": {
664
+ "f1-score": 0.8880537974683544,
665
+ "precision": 0.8293313631326191,
666
+ "recall": 0.955725840783312,
667
+ "support": 2349.0
668
+ },
669
+ "eval_O": {
670
+ "f1-score": 0.9999601132782897,
671
+ "precision": 1.0,
672
+ "recall": 0.9999202297383536,
673
+ "support": 12536.0
674
+ },
675
+ "eval_Premise": {
676
+ "f1-score": 0.9116781523561184,
677
+ "precision": 0.9083958132284166,
678
+ "recall": 0.9149842978914311,
679
+ "support": 13374.0
680
+ },
681
+ "eval_accuracy": 0.9127341857541471,
682
+ "eval_loss": 0.6165347099304199,
683
+ "eval_macro avg": {
684
+ "f1-score": 0.8665089135222377,
685
+ "precision": 0.8607138453723102,
686
+ "recall": 0.8755583729944088,
687
+ "support": 32613.0
688
+ },
689
+ "eval_runtime": 4.8787,
690
+ "eval_samples_per_second": 16.603,
691
+ "eval_steps_per_second": 2.255,
692
+ "eval_weighted avg": {
693
+ "f1-score": 0.9111575738225369,
694
+ "precision": 0.910775248619985,
695
+ "recall": 0.9127341857541471,
696
+ "support": 32613.0
697
+ },
698
+ "step": 1215
699
+ },
700
+ {
701
+ "epoch": 16.0,
702
+ "eval_Claim": {
703
+ "f1-score": 0.674565560821485,
704
+ "precision": 0.6630434782608695,
705
+ "recall": 0.6864951768488746,
706
+ "support": 4354.0
707
+ },
708
+ "eval_MajorClaim": {
709
+ "f1-score": 0.9101030927835051,
710
+ "precision": 0.8824470211915234,
711
+ "recall": 0.9395487441464453,
712
+ "support": 2349.0
713
+ },
714
+ "eval_O": {
715
+ "f1-score": 0.9999601132782897,
716
+ "precision": 1.0,
717
+ "recall": 0.9999202297383536,
718
+ "support": 12536.0
719
+ },
720
+ "eval_Premise": {
721
+ "f1-score": 0.9041334190523012,
722
+ "precision": 0.9146836024179356,
723
+ "recall": 0.8938238372962465,
724
+ "support": 13374.0
725
+ },
726
+ "eval_accuracy": 0.9102198509796707,
727
+ "eval_loss": 0.6390220522880554,
728
+ "eval_macro avg": {
729
+ "f1-score": 0.8721905464838953,
730
+ "precision": 0.8650435254675821,
731
+ "recall": 0.8799469970074799,
732
+ "support": 32613.0
733
+ },
734
+ "eval_runtime": 4.9309,
735
+ "eval_samples_per_second": 16.427,
736
+ "eval_steps_per_second": 2.231,
737
+ "eval_weighted avg": {
738
+ "f1-score": 0.9107494233350908,
739
+ "precision": 0.9115609681986994,
740
+ "recall": 0.9102198509796707,
741
+ "support": 32613.0
742
+ },
743
+ "step": 1296
744
+ },
745
+ {
746
+ "epoch": 17.0,
747
+ "eval_Claim": {
748
+ "f1-score": 0.6808510638297871,
749
+ "precision": 0.7280334728033473,
750
+ "recall": 0.6394120349104272,
751
+ "support": 4354.0
752
+ },
753
+ "eval_MajorClaim": {
754
+ "f1-score": 0.9149623250807319,
755
+ "precision": 0.9255226480836237,
756
+ "recall": 0.9046402724563644,
757
+ "support": 2349.0
758
+ },
759
+ "eval_O": {
760
+ "f1-score": 0.9999601132782897,
761
+ "precision": 1.0,
762
+ "recall": 0.9999202297383536,
763
+ "support": 12536.0
764
+ },
765
+ "eval_Premise": {
766
+ "f1-score": 0.9178984340699547,
767
+ "precision": 0.8986960882647944,
768
+ "recall": 0.9379392851802004,
769
+ "support": 13374.0
770
+ },
771
+ "eval_accuracy": 0.9195106245975532,
772
+ "eval_loss": 0.5812303423881531,
773
+ "eval_macro avg": {
774
+ "f1-score": 0.8784179840646908,
775
+ "precision": 0.8880630522879414,
776
+ "recall": 0.8704779555713364,
777
+ "support": 32613.0
778
+ },
779
+ "eval_runtime": 4.8888,
780
+ "eval_samples_per_second": 16.568,
781
+ "eval_steps_per_second": 2.25,
782
+ "eval_weighted avg": {
783
+ "f1-score": 0.9175833462373209,
784
+ "precision": 0.91678385690944,
785
+ "recall": 0.9195106245975532,
786
+ "support": 32613.0
787
+ },
788
+ "step": 1377
789
+ }
790
+ ],
791
+ "logging_steps": 500,
792
+ "max_steps": 4050,
793
+ "num_input_tokens_seen": 0,
794
+ "num_train_epochs": 50,
795
+ "save_steps": 500,
796
+ "total_flos": 2436817125745800.0,
797
+ "train_batch_size": 4,
798
+ "trial_name": null,
799
+ "trial_params": null
800
+ }
checkpoint-1377/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3f40e108617f6f8082cc0f1a47e89121d861cfb2e77ef6b602a909d9b6ba8d15
3
+ size 4856
checkpoint-1377/vocab.json ADDED
The diff for this file is too large to render. See raw diff