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
5b4036c
·
verified ·
1 Parent(s): 9a8b2ec

Training in progress, epoch 12, checkpoint

Browse files
Files changed (12) hide show
  1. checkpoint-492/config.json +61 -0
  2. checkpoint-492/merges.txt +0 -0
  3. checkpoint-492/model.safetensors +3 -0
  4. checkpoint-492/optimizer.pt +3 -0
  5. checkpoint-492/rng_state.pth +3 -0
  6. checkpoint-492/scheduler.pt +3 -0
  7. checkpoint-492/special_tokens_map.json +15 -0
  8. checkpoint-492/tokenizer.json +0 -0
  9. checkpoint-492/tokenizer_config.json +57 -0
  10. checkpoint-492/trainer_state.json +777 -0
  11. checkpoint-492/training_args.bin +3 -0
  12. checkpoint-492/vocab.json +0 -0
checkpoint-492/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": "4.37.2",
59
+ "type_vocab_size": 1,
60
+ "vocab_size": 50265
61
+ }
checkpoint-492/merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-492/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c105cfd77f76a0ed82093cb4a84ea938929f97b1b57b67f84058683eb29cc998
3
+ size 592330980
checkpoint-492/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:77a7f245b3805ea480c62b85da4a37a54143c52941086d7c7b7da9c4cf9fb43e
3
+ size 1014670074
checkpoint-492/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b6e53e706c63ba0c275c9b6332ceadb76c337b40858f3ddadb0ab10622a334eb
3
+ size 14244
checkpoint-492/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4e40ce765a2db50b61afa20f73544a6db3531a7c20a21e8ebd94c0459daeb642
3
+ size 1064
checkpoint-492/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-492/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-492/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-492/trainer_state.json ADDED
@@ -0,0 +1,777 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 12.0,
5
+ "eval_steps": 500,
6
+ "global_step": 492,
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": 339.0
18
+ },
19
+ "eval_B-MajorClaim": {
20
+ "f1-score": 0.0,
21
+ "precision": 0.0,
22
+ "recall": 0.0,
23
+ "support": 160.0
24
+ },
25
+ "eval_B-Premise": {
26
+ "f1-score": 0.22426470588235295,
27
+ "precision": 0.8299319727891157,
28
+ "recall": 0.12964930924548354,
29
+ "support": 941.0
30
+ },
31
+ "eval_I-Claim": {
32
+ "f1-score": 0.38247663551401867,
33
+ "precision": 0.423873640600725,
34
+ "recall": 0.348446147296722,
35
+ "support": 4698.0
36
+ },
37
+ "eval_I-MajorClaim": {
38
+ "f1-score": 0.3634627625716104,
39
+ "precision": 0.5125673249551167,
40
+ "recall": 0.28155818540433925,
41
+ "support": 2028.0
42
+ },
43
+ "eval_I-Premise": {
44
+ "f1-score": 0.8298935191271198,
45
+ "precision": 0.8110747093209996,
46
+ "recall": 0.8496063521970257,
47
+ "support": 14861.0
48
+ },
49
+ "eval_O": {
50
+ "f1-score": 0.8121805609242796,
51
+ "precision": 0.7380952380952381,
52
+ "recall": 0.9027976701995608,
53
+ "support": 10473.0
54
+ },
55
+ "eval_accuracy": 0.7286865671641791,
56
+ "eval_loss": 0.763540506362915,
57
+ "eval_macro avg": {
58
+ "f1-score": 0.3731825977170545,
59
+ "precision": 0.47364898368017067,
60
+ "recall": 0.3588653806204473,
61
+ "support": 33500.0
62
+ },
63
+ "eval_runtime": 1.4539,
64
+ "eval_samples_per_second": 55.714,
65
+ "eval_steps_per_second": 7.566,
66
+ "eval_weighted avg": {
67
+ "f1-score": 0.7040007584084623,
68
+ "precision": 0.7043362259324342,
69
+ "recall": 0.7286865671641791,
70
+ "support": 33500.0
71
+ },
72
+ "step": 41
73
+ },
74
+ {
75
+ "epoch": 2.0,
76
+ "eval_B-Claim": {
77
+ "f1-score": 0.0,
78
+ "precision": 0.0,
79
+ "recall": 0.0,
80
+ "support": 339.0
81
+ },
82
+ "eval_B-MajorClaim": {
83
+ "f1-score": 0.0,
84
+ "precision": 0.0,
85
+ "recall": 0.0,
86
+ "support": 160.0
87
+ },
88
+ "eval_B-Premise": {
89
+ "f1-score": 0.6669691470054446,
90
+ "precision": 0.5819477434679335,
91
+ "recall": 0.7810839532412327,
92
+ "support": 941.0
93
+ },
94
+ "eval_I-Claim": {
95
+ "f1-score": 0.48473144236572113,
96
+ "precision": 0.559799275160301,
97
+ "recall": 0.4274159216687952,
98
+ "support": 4698.0
99
+ },
100
+ "eval_I-MajorClaim": {
101
+ "f1-score": 0.6472849591542528,
102
+ "precision": 0.6312089971883786,
103
+ "recall": 0.6642011834319527,
104
+ "support": 2028.0
105
+ },
106
+ "eval_I-Premise": {
107
+ "f1-score": 0.8676829461540899,
108
+ "precision": 0.8200934579439252,
109
+ "recall": 0.9211358589596932,
110
+ "support": 14861.0
111
+ },
112
+ "eval_O": {
113
+ "f1-score": 0.8814110459673844,
114
+ "precision": 0.9105252442996743,
115
+ "recall": 0.8541010216747827,
116
+ "support": 10473.0
117
+ },
118
+ "eval_accuracy": 0.7977313432835821,
119
+ "eval_loss": 0.558718204498291,
120
+ "eval_macro avg": {
121
+ "f1-score": 0.506868505806699,
122
+ "precision": 0.5005106740086018,
123
+ "recall": 0.5211339912823509,
124
+ "support": 33500.0
125
+ },
126
+ "eval_runtime": 1.4491,
127
+ "eval_samples_per_second": 55.895,
128
+ "eval_steps_per_second": 7.591,
129
+ "eval_weighted avg": {
130
+ "f1-score": 0.7863652038192075,
131
+ "precision": 0.7815218337211824,
132
+ "recall": 0.7977313432835821,
133
+ "support": 33500.0
134
+ },
135
+ "step": 82
136
+ },
137
+ {
138
+ "epoch": 3.0,
139
+ "eval_B-Claim": {
140
+ "f1-score": 0.29174664107485604,
141
+ "precision": 0.4175824175824176,
142
+ "recall": 0.22418879056047197,
143
+ "support": 339.0
144
+ },
145
+ "eval_B-MajorClaim": {
146
+ "f1-score": 0.1694915254237288,
147
+ "precision": 0.8823529411764706,
148
+ "recall": 0.09375,
149
+ "support": 160.0
150
+ },
151
+ "eval_B-Premise": {
152
+ "f1-score": 0.731958762886598,
153
+ "precision": 0.6546521374685667,
154
+ "recall": 0.8299681190223167,
155
+ "support": 941.0
156
+ },
157
+ "eval_I-Claim": {
158
+ "f1-score": 0.45169213973799127,
159
+ "precision": 0.629277566539924,
160
+ "recall": 0.35227756492124307,
161
+ "support": 4698.0
162
+ },
163
+ "eval_I-MajorClaim": {
164
+ "f1-score": 0.7669586983729662,
165
+ "precision": 0.7788510421962379,
166
+ "recall": 0.7554240631163708,
167
+ "support": 2028.0
168
+ },
169
+ "eval_I-Premise": {
170
+ "f1-score": 0.8619200734956362,
171
+ "precision": 0.7908845678318535,
172
+ "recall": 0.9469753044882578,
173
+ "support": 14861.0
174
+ },
175
+ "eval_O": {
176
+ "f1-score": 0.8982664685487864,
177
+ "precision": 0.9332098384274982,
178
+ "recall": 0.8658455074954645,
179
+ "support": 10473.0
180
+ },
181
+ "eval_accuracy": 0.8119402985074626,
182
+ "eval_loss": 0.5103888511657715,
183
+ "eval_macro avg": {
184
+ "f1-score": 0.5960049013629376,
185
+ "precision": 0.7266872158889954,
186
+ "recall": 0.5812041928005893,
187
+ "support": 33500.0
188
+ },
189
+ "eval_runtime": 1.4566,
190
+ "eval_samples_per_second": 55.611,
191
+ "eval_steps_per_second": 7.552,
192
+ "eval_weighted avg": {
193
+ "f1-score": 0.7972768597429257,
194
+ "precision": 0.804819781883149,
195
+ "recall": 0.8119402985074626,
196
+ "support": 33500.0
197
+ },
198
+ "step": 123
199
+ },
200
+ {
201
+ "epoch": 4.0,
202
+ "eval_B-Claim": {
203
+ "f1-score": 0.5957446808510639,
204
+ "precision": 0.6144200626959248,
205
+ "recall": 0.5781710914454278,
206
+ "support": 339.0
207
+ },
208
+ "eval_B-MajorClaim": {
209
+ "f1-score": 0.7210884353741497,
210
+ "precision": 0.7910447761194029,
211
+ "recall": 0.6625,
212
+ "support": 160.0
213
+ },
214
+ "eval_B-Premise": {
215
+ "f1-score": 0.7832738398776133,
216
+ "precision": 0.7529411764705882,
217
+ "recall": 0.8161530286928799,
218
+ "support": 941.0
219
+ },
220
+ "eval_I-Claim": {
221
+ "f1-score": 0.6127179923436835,
222
+ "precision": 0.6121971950701233,
223
+ "recall": 0.6132396764580673,
224
+ "support": 4698.0
225
+ },
226
+ "eval_I-MajorClaim": {
227
+ "f1-score": 0.8112728112728113,
228
+ "precision": 0.7631464580617123,
229
+ "recall": 0.8658777120315582,
230
+ "support": 2028.0
231
+ },
232
+ "eval_I-Premise": {
233
+ "f1-score": 0.8727813651577132,
234
+ "precision": 0.8770213344204376,
235
+ "recall": 0.8685821950070655,
236
+ "support": 14861.0
237
+ },
238
+ "eval_O": {
239
+ "f1-score": 0.9022382671480144,
240
+ "precision": 0.9097262667443214,
241
+ "recall": 0.8948725293612145,
242
+ "support": 10473.0
243
+ },
244
+ "eval_accuracy": 0.8354328358208956,
245
+ "eval_loss": 0.4638007879257202,
246
+ "eval_macro avg": {
247
+ "f1-score": 0.7570167702892928,
248
+ "precision": 0.7600710385117873,
249
+ "recall": 0.7570566047137447,
250
+ "support": 33500.0
251
+ },
252
+ "eval_runtime": 1.4488,
253
+ "eval_samples_per_second": 55.909,
254
+ "eval_steps_per_second": 7.593,
255
+ "eval_weighted avg": {
256
+ "f1-score": 0.8357536689114934,
257
+ "precision": 0.8366601162482339,
258
+ "recall": 0.8354328358208956,
259
+ "support": 33500.0
260
+ },
261
+ "step": 164
262
+ },
263
+ {
264
+ "epoch": 5.0,
265
+ "eval_B-Claim": {
266
+ "f1-score": 0.5777777777777778,
267
+ "precision": 0.6254295532646048,
268
+ "recall": 0.5368731563421829,
269
+ "support": 339.0
270
+ },
271
+ "eval_B-MajorClaim": {
272
+ "f1-score": 0.7753846153846152,
273
+ "precision": 0.7636363636363637,
274
+ "recall": 0.7875,
275
+ "support": 160.0
276
+ },
277
+ "eval_B-Premise": {
278
+ "f1-score": 0.7803692905733722,
279
+ "precision": 0.7188898836168308,
280
+ "recall": 0.8533475026567482,
281
+ "support": 941.0
282
+ },
283
+ "eval_I-Claim": {
284
+ "f1-score": 0.5902196060674665,
285
+ "precision": 0.6303191489361702,
286
+ "recall": 0.5549169859514687,
287
+ "support": 4698.0
288
+ },
289
+ "eval_I-MajorClaim": {
290
+ "f1-score": 0.8467858869018849,
291
+ "precision": 0.8303317535545024,
292
+ "recall": 0.863905325443787,
293
+ "support": 2028.0
294
+ },
295
+ "eval_I-Premise": {
296
+ "f1-score": 0.8742343541944074,
297
+ "precision": 0.8650767507740957,
298
+ "recall": 0.8835879146759976,
299
+ "support": 14861.0
300
+ },
301
+ "eval_O": {
302
+ "f1-score": 0.9062216924910609,
303
+ "precision": 0.904970481812988,
304
+ "recall": 0.9074763678029218,
305
+ "support": 10473.0
306
+ },
307
+ "eval_accuracy": 0.8389552238805971,
308
+ "eval_loss": 0.49385663866996765,
309
+ "eval_macro avg": {
310
+ "f1-score": 0.7644276033415122,
311
+ "precision": 0.7626648479422222,
312
+ "recall": 0.7696581789818724,
313
+ "support": 33500.0
314
+ },
315
+ "eval_runtime": 1.4492,
316
+ "eval_samples_per_second": 55.893,
317
+ "eval_steps_per_second": 7.59,
318
+ "eval_weighted avg": {
319
+ "f1-score": 0.8366342005564686,
320
+ "precision": 0.8355072066989216,
321
+ "recall": 0.8389552238805971,
322
+ "support": 33500.0
323
+ },
324
+ "step": 205
325
+ },
326
+ {
327
+ "epoch": 6.0,
328
+ "eval_B-Claim": {
329
+ "f1-score": 0.5768621236133122,
330
+ "precision": 0.6232876712328768,
331
+ "recall": 0.5368731563421829,
332
+ "support": 339.0
333
+ },
334
+ "eval_B-MajorClaim": {
335
+ "f1-score": 0.7721518987341772,
336
+ "precision": 0.782051282051282,
337
+ "recall": 0.7625,
338
+ "support": 160.0
339
+ },
340
+ "eval_B-Premise": {
341
+ "f1-score": 0.7921882824236354,
342
+ "precision": 0.7490530303030303,
343
+ "recall": 0.8405951115834219,
344
+ "support": 941.0
345
+ },
346
+ "eval_I-Claim": {
347
+ "f1-score": 0.6107428571428571,
348
+ "precision": 0.6594274432379073,
349
+ "recall": 0.5687526607066837,
350
+ "support": 4698.0
351
+ },
352
+ "eval_I-MajorClaim": {
353
+ "f1-score": 0.8353552859618717,
354
+ "precision": 0.8388861263053208,
355
+ "recall": 0.8318540433925049,
356
+ "support": 2028.0
357
+ },
358
+ "eval_I-Premise": {
359
+ "f1-score": 0.8799789494457783,
360
+ "precision": 0.8607000386050702,
361
+ "recall": 0.9001413094677343,
362
+ "support": 14861.0
363
+ },
364
+ "eval_O": {
365
+ "f1-score": 0.912864263803681,
366
+ "precision": 0.9164661726494081,
367
+ "recall": 0.9092905566695312,
368
+ "support": 10473.0
369
+ },
370
+ "eval_accuracy": 0.8463880597014926,
371
+ "eval_loss": 0.49789243936538696,
372
+ "eval_macro avg": {
373
+ "f1-score": 0.7685919515893305,
374
+ "precision": 0.7756959663406994,
375
+ "recall": 0.7642866911660084,
376
+ "support": 33500.0
377
+ },
378
+ "eval_runtime": 1.4655,
379
+ "eval_samples_per_second": 55.27,
380
+ "eval_steps_per_second": 7.506,
381
+ "eval_weighted avg": {
382
+ "f1-score": 0.843752680680015,
383
+ "precision": 0.8426730244638024,
384
+ "recall": 0.8463880597014926,
385
+ "support": 33500.0
386
+ },
387
+ "step": 246
388
+ },
389
+ {
390
+ "epoch": 7.0,
391
+ "eval_B-Claim": {
392
+ "f1-score": 0.5829307568438002,
393
+ "precision": 0.6418439716312057,
394
+ "recall": 0.5339233038348082,
395
+ "support": 339.0
396
+ },
397
+ "eval_B-MajorClaim": {
398
+ "f1-score": 0.8083832335329341,
399
+ "precision": 0.7758620689655172,
400
+ "recall": 0.84375,
401
+ "support": 160.0
402
+ },
403
+ "eval_B-Premise": {
404
+ "f1-score": 0.7938931297709925,
405
+ "precision": 0.7203463203463204,
406
+ "recall": 0.8841657810839533,
407
+ "support": 941.0
408
+ },
409
+ "eval_I-Claim": {
410
+ "f1-score": 0.560838445807771,
411
+ "precision": 0.7018554062699937,
412
+ "recall": 0.4670072371221797,
413
+ "support": 4698.0
414
+ },
415
+ "eval_I-MajorClaim": {
416
+ "f1-score": 0.8591447209470888,
417
+ "precision": 0.8422548555187115,
418
+ "recall": 0.8767258382642998,
419
+ "support": 2028.0
420
+ },
421
+ "eval_I-Premise": {
422
+ "f1-score": 0.882233502538071,
423
+ "precision": 0.834623927006423,
424
+ "recall": 0.9356032568467801,
425
+ "support": 14861.0
426
+ },
427
+ "eval_O": {
428
+ "f1-score": 0.9121469754715138,
429
+ "precision": 0.9340538376863805,
430
+ "recall": 0.8912441516279958,
431
+ "support": 10473.0
432
+ },
433
+ "eval_accuracy": 0.8465074626865672,
434
+ "eval_loss": 0.5638343691825867,
435
+ "eval_macro avg": {
436
+ "f1-score": 0.7713672521303102,
437
+ "precision": 0.7786914839177931,
438
+ "recall": 0.7760599383971453,
439
+ "support": 33500.0
440
+ },
441
+ "eval_runtime": 1.4511,
442
+ "eval_samples_per_second": 55.818,
443
+ "eval_steps_per_second": 7.58,
444
+ "eval_weighted avg": {
445
+ "f1-score": 0.8392525416975183,
446
+ "precision": 0.8421095669207056,
447
+ "recall": 0.8465074626865672,
448
+ "support": 33500.0
449
+ },
450
+ "step": 287
451
+ },
452
+ {
453
+ "epoch": 8.0,
454
+ "eval_B-Claim": {
455
+ "f1-score": 0.623229461756374,
456
+ "precision": 0.5994550408719346,
457
+ "recall": 0.6489675516224189,
458
+ "support": 339.0
459
+ },
460
+ "eval_B-MajorClaim": {
461
+ "f1-score": 0.7987987987987989,
462
+ "precision": 0.7687861271676301,
463
+ "recall": 0.83125,
464
+ "support": 160.0
465
+ },
466
+ "eval_B-Premise": {
467
+ "f1-score": 0.7859614434008898,
468
+ "precision": 0.7347504621072088,
469
+ "recall": 0.844845908607864,
470
+ "support": 941.0
471
+ },
472
+ "eval_I-Claim": {
473
+ "f1-score": 0.620274914089347,
474
+ "precision": 0.6259211096662332,
475
+ "recall": 0.6147296722009365,
476
+ "support": 4698.0
477
+ },
478
+ "eval_I-MajorClaim": {
479
+ "f1-score": 0.8596871239470517,
480
+ "precision": 0.8396803008932769,
481
+ "recall": 0.8806706114398422,
482
+ "support": 2028.0
483
+ },
484
+ "eval_I-Premise": {
485
+ "f1-score": 0.8772429318678407,
486
+ "precision": 0.8634082763115021,
487
+ "recall": 0.8915281609582127,
488
+ "support": 14861.0
489
+ },
490
+ "eval_O": {
491
+ "f1-score": 0.9116210214655811,
492
+ "precision": 0.9433210784313726,
493
+ "recall": 0.8819822400458321,
494
+ "support": 10473.0
495
+ },
496
+ "eval_accuracy": 0.8450149253731343,
497
+ "eval_loss": 0.5492746829986572,
498
+ "eval_macro avg": {
499
+ "f1-score": 0.782402242189412,
500
+ "precision": 0.7679031993498798,
501
+ "recall": 0.7991391635535867,
502
+ "support": 33500.0
503
+ },
504
+ "eval_runtime": 1.4551,
505
+ "eval_samples_per_second": 55.667,
506
+ "eval_steps_per_second": 7.56,
507
+ "eval_weighted avg": {
508
+ "f1-score": 0.8453815974816967,
509
+ "precision": 0.8469129043625616,
510
+ "recall": 0.8450149253731343,
511
+ "support": 33500.0
512
+ },
513
+ "step": 328
514
+ },
515
+ {
516
+ "epoch": 9.0,
517
+ "eval_B-Claim": {
518
+ "f1-score": 0.5891472868217054,
519
+ "precision": 0.6209150326797386,
520
+ "recall": 0.56047197640118,
521
+ "support": 339.0
522
+ },
523
+ "eval_B-MajorClaim": {
524
+ "f1-score": 0.8050314465408804,
525
+ "precision": 0.810126582278481,
526
+ "recall": 0.8,
527
+ "support": 160.0
528
+ },
529
+ "eval_B-Premise": {
530
+ "f1-score": 0.7891566265060241,
531
+ "precision": 0.7478591817316841,
532
+ "recall": 0.8352816153028693,
533
+ "support": 941.0
534
+ },
535
+ "eval_I-Claim": {
536
+ "f1-score": 0.5999312793494446,
537
+ "precision": 0.6493925117778329,
538
+ "recall": 0.5574712643678161,
539
+ "support": 4698.0
540
+ },
541
+ "eval_I-MajorClaim": {
542
+ "f1-score": 0.8532857870606215,
543
+ "precision": 0.8825079030558483,
544
+ "recall": 0.8259368836291914,
545
+ "support": 2028.0
546
+ },
547
+ "eval_I-Premise": {
548
+ "f1-score": 0.8768981374268038,
549
+ "precision": 0.8624886112195757,
550
+ "recall": 0.8917973218491353,
551
+ "support": 14861.0
552
+ },
553
+ "eval_O": {
554
+ "f1-score": 0.9065734133547564,
555
+ "precision": 0.8974550898203593,
556
+ "recall": 0.9158789267640599,
557
+ "support": 10473.0
558
+ },
559
+ "eval_accuracy": 0.8430746268656717,
560
+ "eval_loss": 0.5794806480407715,
561
+ "eval_macro avg": {
562
+ "f1-score": 0.7742891395800339,
563
+ "precision": 0.7815349875090741,
564
+ "recall": 0.7695482840448931,
565
+ "support": 33500.0
566
+ },
567
+ "eval_runtime": 1.4526,
568
+ "eval_samples_per_second": 55.763,
569
+ "eval_steps_per_second": 7.573,
570
+ "eval_weighted avg": {
571
+ "f1-score": 0.840184765735164,
572
+ "precision": 0.8388330863882018,
573
+ "recall": 0.8430746268656717,
574
+ "support": 33500.0
575
+ },
576
+ "step": 369
577
+ },
578
+ {
579
+ "epoch": 10.0,
580
+ "eval_B-Claim": {
581
+ "f1-score": 0.5950155763239875,
582
+ "precision": 0.6303630363036303,
583
+ "recall": 0.5634218289085545,
584
+ "support": 339.0
585
+ },
586
+ "eval_B-MajorClaim": {
587
+ "f1-score": 0.8123076923076924,
588
+ "precision": 0.8,
589
+ "recall": 0.825,
590
+ "support": 160.0
591
+ },
592
+ "eval_B-Premise": {
593
+ "f1-score": 0.7884143348060874,
594
+ "precision": 0.7326642335766423,
595
+ "recall": 0.8533475026567482,
596
+ "support": 941.0
597
+ },
598
+ "eval_I-Claim": {
599
+ "f1-score": 0.6070517281373231,
600
+ "precision": 0.666921508664628,
601
+ "recall": 0.5570455512984248,
602
+ "support": 4698.0
603
+ },
604
+ "eval_I-MajorClaim": {
605
+ "f1-score": 0.8758397611346106,
606
+ "precision": 0.8839779005524862,
607
+ "recall": 0.8678500986193294,
608
+ "support": 2028.0
609
+ },
610
+ "eval_I-Premise": {
611
+ "f1-score": 0.8810970656094628,
612
+ "precision": 0.851998491704374,
613
+ "recall": 0.9122535495592491,
614
+ "support": 14861.0
615
+ },
616
+ "eval_O": {
617
+ "f1-score": 0.9168205227869012,
618
+ "precision": 0.9333267385498071,
619
+ "recall": 0.900887997708393,
620
+ "support": 10473.0
621
+ },
622
+ "eval_accuracy": 0.8505970149253731,
623
+ "eval_loss": 0.5959911346435547,
624
+ "eval_macro avg": {
625
+ "f1-score": 0.7823638115865806,
626
+ "precision": 0.7856074156216526,
627
+ "recall": 0.7828295041072427,
628
+ "support": 33500.0
629
+ },
630
+ "eval_runtime": 1.4562,
631
+ "eval_samples_per_second": 55.623,
632
+ "eval_steps_per_second": 7.554,
633
+ "eval_weighted avg": {
634
+ "f1-score": 0.8476881875144826,
635
+ "precision": 0.8475616436173209,
636
+ "recall": 0.8505970149253731,
637
+ "support": 33500.0
638
+ },
639
+ "step": 410
640
+ },
641
+ {
642
+ "epoch": 11.0,
643
+ "eval_B-Claim": {
644
+ "f1-score": 0.6244477172312225,
645
+ "precision": 0.6235294117647059,
646
+ "recall": 0.6253687315634219,
647
+ "support": 339.0
648
+ },
649
+ "eval_B-MajorClaim": {
650
+ "f1-score": 0.8168168168168167,
651
+ "precision": 0.7861271676300579,
652
+ "recall": 0.85,
653
+ "support": 160.0
654
+ },
655
+ "eval_B-Premise": {
656
+ "f1-score": 0.7952917093142272,
657
+ "precision": 0.7670286278381047,
658
+ "recall": 0.8257173219978746,
659
+ "support": 941.0
660
+ },
661
+ "eval_I-Claim": {
662
+ "f1-score": 0.639956686518679,
663
+ "precision": 0.6513114392770554,
664
+ "recall": 0.6289910600255428,
665
+ "support": 4698.0
666
+ },
667
+ "eval_I-MajorClaim": {
668
+ "f1-score": 0.8596819457436857,
669
+ "precision": 0.8176156583629893,
670
+ "recall": 0.9063116370808678,
671
+ "support": 2028.0
672
+ },
673
+ "eval_I-Premise": {
674
+ "f1-score": 0.8823608047590401,
675
+ "precision": 0.8765022242878959,
676
+ "recall": 0.8882982302671422,
677
+ "support": 14861.0
678
+ },
679
+ "eval_O": {
680
+ "f1-score": 0.9133537206931702,
681
+ "precision": 0.9289099526066351,
682
+ "recall": 0.8983099398453165,
683
+ "support": 10473.0
684
+ },
685
+ "eval_accuracy": 0.8515522388059702,
686
+ "eval_loss": 0.5953634977340698,
687
+ "eval_macro avg": {
688
+ "f1-score": 0.7902727715824059,
689
+ "precision": 0.7787177831096349,
690
+ "recall": 0.8032852743971665,
691
+ "support": 33500.0
692
+ },
693
+ "eval_runtime": 1.4583,
694
+ "eval_samples_per_second": 55.544,
695
+ "eval_steps_per_second": 7.543,
696
+ "eval_weighted avg": {
697
+ "f1-score": 0.8513139373395039,
698
+ "precision": 0.8516743266232883,
699
+ "recall": 0.8515522388059702,
700
+ "support": 33500.0
701
+ },
702
+ "step": 451
703
+ },
704
+ {
705
+ "epoch": 12.0,
706
+ "eval_B-Claim": {
707
+ "f1-score": 0.6188466947960619,
708
+ "precision": 0.5913978494623656,
709
+ "recall": 0.6489675516224189,
710
+ "support": 339.0
711
+ },
712
+ "eval_B-MajorClaim": {
713
+ "f1-score": 0.8119402985074626,
714
+ "precision": 0.7771428571428571,
715
+ "recall": 0.85,
716
+ "support": 160.0
717
+ },
718
+ "eval_B-Premise": {
719
+ "f1-score": 0.7870702924576707,
720
+ "precision": 0.7609126984126984,
721
+ "recall": 0.8150903294367694,
722
+ "support": 941.0
723
+ },
724
+ "eval_I-Claim": {
725
+ "f1-score": 0.6262368337057134,
726
+ "precision": 0.6260370134014039,
727
+ "recall": 0.6264367816091954,
728
+ "support": 4698.0
729
+ },
730
+ "eval_I-MajorClaim": {
731
+ "f1-score": 0.8551000953288845,
732
+ "precision": 0.827490774907749,
733
+ "recall": 0.8846153846153846,
734
+ "support": 2028.0
735
+ },
736
+ "eval_I-Premise": {
737
+ "f1-score": 0.8772554685662445,
738
+ "precision": 0.8761073825503356,
739
+ "recall": 0.8784065675257385,
740
+ "support": 14861.0
741
+ },
742
+ "eval_O": {
743
+ "f1-score": 0.9119085670008232,
744
+ "precision": 0.9252161949685535,
745
+ "recall": 0.8989783252172252,
746
+ "support": 10473.0
747
+ },
748
+ "eval_accuracy": 0.8456417910447761,
749
+ "eval_loss": 0.6605897545814514,
750
+ "eval_macro avg": {
751
+ "f1-score": 0.784051178623266,
752
+ "precision": 0.7691863958351375,
753
+ "recall": 0.8003564200038189,
754
+ "support": 33500.0
755
+ },
756
+ "eval_runtime": 1.4573,
757
+ "eval_samples_per_second": 55.582,
758
+ "eval_steps_per_second": 7.548,
759
+ "eval_weighted avg": {
760
+ "f1-score": 0.8460850209295142,
761
+ "precision": 0.8468579038739062,
762
+ "recall": 0.8456417910447761,
763
+ "support": 33500.0
764
+ },
765
+ "step": 492
766
+ }
767
+ ],
768
+ "logging_steps": 500,
769
+ "max_steps": 656,
770
+ "num_input_tokens_seen": 0,
771
+ "num_train_epochs": 16,
772
+ "save_steps": 500,
773
+ "total_flos": 1720131088788000.0,
774
+ "train_batch_size": 8,
775
+ "trial_name": null,
776
+ "trial_params": null
777
+ }
checkpoint-492/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:920ab155975bf66f694757fefb246b73b0d1489bdf4f03eda383fa028f5d083a
3
+ size 4664
checkpoint-492/vocab.json ADDED
The diff for this file is too large to render. See raw diff