Hugging Face's logo

Theoreticallyhugo
/
longformer-full_labels

Token Classification
Transformers
Safetensors
longformer
Generated from Trainer
Eval Results
Inference Endpoints
Model card Files Community
Theoreticallyhugo commited on
Commit
1663da5
•
1 Parent(s): 0999200

Training in progress, epoch 13, checkpoint

Browse files
Files changed (12) hide show
  1. checkpoint-533/config.json +61 -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 +846 -0
  11. checkpoint-533/training_args.bin +3 -0
  12. checkpoint-533/vocab.json +0 -0
checkpoint-533/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-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 sha256:5320294299a3711fa3951e5c22d23d096a249c8fdfb70f42cb07e98d29df3c0b
3
+ size 592330980
checkpoint-533/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4a19073bf31678729208423432357c73d5594a453a846d4da2ecdde9e0f01640
3
+ size 1014670074
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 +1,846 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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_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": 0.1227621483375959,
27
+ "precision": 0.8727272727272727,
28
+ "recall": 0.06602475928473177,
29
+ "support": 727.0
30
+ },
31
+ "eval_I-Claim": {
32
+ "f1-score": 0.35051546391752586,
33
+ "precision": 0.5111008325624422,
34
+ "recall": 0.26671494086410813,
35
+ "support": 4143.0
36
+ },
37
+ "eval_I-MajorClaim": {
38
+ "f1-score": 0.2476060191518468,
39
+ "precision": 0.4952120383036936,
40
+ "recall": 0.16507067943456452,
41
+ "support": 2193.0
42
+ },
43
+ "eval_I-Premise": {
44
+ "f1-score": 0.8531195999852935,
45
+ "precision": 0.7927029242962558,
46
+ "recall": 0.9235055321181247,
47
+ "support": 12563.0
48
+ },
49
+ "eval_O": {
50
+ "f1-score": 0.8310506279263117,
51
+ "precision": 0.7493095557484416,
52
+ "recall": 0.9328094302554027,
53
+ "support": 10180.0
54
+ },
55
+ "eval_accuracy": 0.7473642462901147,
56
+ "eval_loss": 0.6794929504394531,
57
+ "eval_macro avg": {
58
+ "f1-score": 0.34357912275979624,
59
+ "precision": 0.48872180337687227,
60
+ "recall": 0.3363036202795617,
61
+ "support": 30257.0
62
+ },
63
+ "eval_runtime": 1.4284,
64
+ "eval_samples_per_second": 56.707,
65
+ "eval_steps_per_second": 7.701,
66
+ "eval_weighted avg": {
67
+ "f1-score": 0.7027223642713039,
68
+ "precision": 0.7080894203665903,
69
+ "recall": 0.7473642462901147,
70
+ "support": 30257.0
71
+ },
72
+ "step": 41
73
+ },
74
+ {
75
+ "epoch": 2.0,
76
+ "eval_B-Claim": {
77
+ "f1-score": 0.1337386018237082,
78
+ "precision": 0.6470588235294118,
79
+ "recall": 0.07457627118644068,
80
+ "support": 295.0
81
+ },
82
+ "eval_B-MajorClaim": {
83
+ "f1-score": 0.0,
84
+ "precision": 0.0,
85
+ "recall": 0.0,
86
+ "support": 156.0
87
+ },
88
+ "eval_B-Premise": {
89
+ "f1-score": 0.7234567901234568,
90
+ "precision": 0.6562150055991042,
91
+ "recall": 0.8060522696011004,
92
+ "support": 727.0
93
+ },
94
+ "eval_I-Claim": {
95
+ "f1-score": 0.5708538220831472,
96
+ "precision": 0.5315296566077004,
97
+ "recall": 0.6164615013275404,
98
+ "support": 4143.0
99
+ },
100
+ "eval_I-MajorClaim": {
101
+ "f1-score": 0.6949243098842386,
102
+ "precision": 0.5701344243132671,
103
+ "recall": 0.8896488828089375,
104
+ "support": 2193.0
105
+ },
106
+ "eval_I-Premise": {
107
+ "f1-score": 0.861198871056068,
108
+ "precision": 0.914638511095204,
109
+ "recall": 0.8136591578444639,
110
+ "support": 12563.0
111
+ },
112
+ "eval_O": {
113
+ "f1-score": 0.9030685830805192,
114
+ "precision": 0.9145764077767704,
115
+ "recall": 0.8918467583497053,
116
+ "support": 10180.0
117
+ },
118
+ "eval_accuracy": 0.8068876623591235,
119
+ "eval_loss": 0.511056661605835,
120
+ "eval_macro avg": {
121
+ "f1-score": 0.5553201397215911,
122
+ "precision": 0.6048789755602082,
123
+ "recall": 0.5846064058740269,
124
+ "support": 30257.0
125
+ },
126
+ "eval_runtime": 1.4332,
127
+ "eval_samples_per_second": 56.519,
128
+ "eval_steps_per_second": 7.675,
129
+ "eval_weighted avg": {
130
+ "f1-score": 0.8086360829977005,
131
+ "precision": 0.8236564850419075,
132
+ "recall": 0.8068876623591235,
133
+ "support": 30257.0
134
+ },
135
+ "step": 82
136
+ },
137
+ {
138
+ "epoch": 3.0,
139
+ "eval_B-Claim": {
140
+ "f1-score": 0.47176079734219273,
141
+ "precision": 0.46254071661237783,
142
+ "recall": 0.48135593220338985,
143
+ "support": 295.0
144
+ },
145
+ "eval_B-MajorClaim": {
146
+ "f1-score": 0.07272727272727274,
147
+ "precision": 0.6666666666666666,
148
+ "recall": 0.038461538461538464,
149
+ "support": 156.0
150
+ },
151
+ "eval_B-Premise": {
152
+ "f1-score": 0.7764857881136951,
153
+ "precision": 0.7320341047503045,
154
+ "recall": 0.8266850068775791,
155
+ "support": 727.0
156
+ },
157
+ "eval_I-Claim": {
158
+ "f1-score": 0.597796762589928,
159
+ "precision": 0.5594361455922575,
160
+ "recall": 0.6418054549843109,
161
+ "support": 4143.0
162
+ },
163
+ "eval_I-MajorClaim": {
164
+ "f1-score": 0.730778274159417,
165
+ "precision": 0.7543689320388349,
166
+ "recall": 0.7086183310533516,
167
+ "support": 2193.0
168
+ },
169
+ "eval_I-Premise": {
170
+ "f1-score": 0.8887291317197858,
171
+ "precision": 0.8918637274549098,
172
+ "recall": 0.8856164928759054,
173
+ "support": 12563.0
174
+ },
175
+ "eval_O": {
176
+ "f1-score": 0.9178492904257447,
177
+ "precision": 0.934092758340114,
178
+ "recall": 0.9021611001964637,
179
+ "support": 10180.0
180
+ },
181
+ "eval_accuracy": 0.835244736755131,
182
+ "eval_loss": 0.4231853783130646,
183
+ "eval_macro avg": {
184
+ "f1-score": 0.6365896167254336,
185
+ "precision": 0.7144290073507807,
186
+ "recall": 0.6406719795217912,
187
+ "support": 30257.0
188
+ },
189
+ "eval_runtime": 1.4362,
190
+ "eval_samples_per_second": 56.399,
191
+ "eval_steps_per_second": 7.659,
192
+ "eval_weighted avg": {
193
+ "f1-score": 0.836272553745959,
194
+ "precision": 0.841400720911604,
195
+ "recall": 0.835244736755131,
196
+ "support": 30257.0
197
+ },
198
+ "step": 123
199
+ },
200
+ {
201
+ "epoch": 4.0,
202
+ "eval_B-Claim": {
203
+ "f1-score": 0.5212947189097104,
204
+ "precision": 0.523972602739726,
205
+ "recall": 0.5186440677966102,
206
+ "support": 295.0
207
+ },
208
+ "eval_B-MajorClaim": {
209
+ "f1-score": 0.6433566433566433,
210
+ "precision": 0.7076923076923077,
211
+ "recall": 0.5897435897435898,
212
+ "support": 156.0
213
+ },
214
+ "eval_B-Premise": {
215
+ "f1-score": 0.7744020685197156,
216
+ "precision": 0.7304878048780488,
217
+ "recall": 0.8239339752407153,
218
+ "support": 727.0
219
+ },
220
+ "eval_I-Claim": {
221
+ "f1-score": 0.5872821016449792,
222
+ "precision": 0.5975518361229079,
223
+ "recall": 0.5773594013999517,
224
+ "support": 4143.0
225
+ },
226
+ "eval_I-MajorClaim": {
227
+ "f1-score": 0.7907801418439716,
228
+ "precision": 0.7692971108236308,
229
+ "recall": 0.8134974920200638,
230
+ "support": 2193.0
231
+ },
232
+ "eval_I-Premise": {
233
+ "f1-score": 0.8883939418482223,
234
+ "precision": 0.9042123485285631,
235
+ "recall": 0.873119477831728,
236
+ "support": 12563.0
237
+ },
238
+ "eval_O": {
239
+ "f1-score": 0.9123517500723171,
240
+ "precision": 0.8958530581329294,
241
+ "recall": 0.9294695481335953,
242
+ "support": 10180.0
243
+ },
244
+ "eval_accuracy": 0.8411607231384473,
245
+ "eval_loss": 0.42953914403915405,
246
+ "eval_macro avg": {
247
+ "f1-score": 0.7311230523136514,
248
+ "precision": 0.7327238669883019,
249
+ "recall": 0.7322525074523221,
250
+ "support": 30257.0
251
+ },
252
+ "eval_runtime": 1.4336,
253
+ "eval_samples_per_second": 56.501,
254
+ "eval_steps_per_second": 7.673,
255
+ "eval_weighted avg": {
256
+ "f1-score": 0.8405678153025818,
257
+ "precision": 0.8407365647422267,
258
+ "recall": 0.8411607231384473,
259
+ "support": 30257.0
260
+ },
261
+ "step": 164
262
+ },
263
+ {
264
+ "epoch": 5.0,
265
+ "eval_B-Claim": {
266
+ "f1-score": 0.5539452495974235,
267
+ "precision": 0.5276073619631901,
268
+ "recall": 0.5830508474576271,
269
+ "support": 295.0
270
+ },
271
+ "eval_B-MajorClaim": {
272
+ "f1-score": 0.6887417218543046,
273
+ "precision": 0.7123287671232876,
274
+ "recall": 0.6666666666666666,
275
+ "support": 156.0
276
+ },
277
+ "eval_B-Premise": {
278
+ "f1-score": 0.8093083387201035,
279
+ "precision": 0.7634146341463415,
280
+ "recall": 0.8610729023383769,
281
+ "support": 727.0
282
+ },
283
+ "eval_I-Claim": {
284
+ "f1-score": 0.5725826193390453,
285
+ "precision": 0.5808294015396076,
286
+ "recall": 0.5645667390779628,
287
+ "support": 4143.0
288
+ },
289
+ "eval_I-MajorClaim": {
290
+ "f1-score": 0.7844767844767844,
291
+ "precision": 0.7949438202247191,
292
+ "recall": 0.774281805745554,
293
+ "support": 2193.0
294
+ },
295
+ "eval_I-Premise": {
296
+ "f1-score": 0.8950967891503419,
297
+ "precision": 0.8697251839615557,
298
+ "recall": 0.9219931545013134,
299
+ "support": 12563.0
300
+ },
301
+ "eval_O": {
302
+ "f1-score": 0.9191415785191211,
303
+ "precision": 0.9528679881906369,
304
+ "recall": 0.887721021611002,
305
+ "support": 10180.0
306
+ },
307
+ "eval_accuracy": 0.8447301450903923,
308
+ "eval_loss": 0.456537127494812,
309
+ "eval_macro avg": {
310
+ "f1-score": 0.7461847259510178,
311
+ "precision": 0.7431024510213342,
312
+ "recall": 0.7513361624855005,
313
+ "support": 30257.0
314
+ },
315
+ "eval_runtime": 1.4256,
316
+ "eval_samples_per_second": 56.819,
317
+ "eval_steps_per_second": 7.716,
318
+ "eval_weighted avg": {
319
+ "f1-score": 0.8445567746699831,
320
+ "precision": 0.8460194835144224,
321
+ "recall": 0.8447301450903923,
322
+ "support": 30257.0
323
+ },
324
+ "step": 205
325
+ },
326
+ {
327
+ "epoch": 6.0,
328
+ "eval_B-Claim": {
329
+ "f1-score": 0.5251798561151079,
330
+ "precision": 0.5593869731800766,
331
+ "recall": 0.49491525423728816,
332
+ "support": 295.0
333
+ },
334
+ "eval_B-MajorClaim": {
335
+ "f1-score": 0.7146974063400577,
336
+ "precision": 0.6492146596858639,
337
+ "recall": 0.7948717948717948,
338
+ "support": 156.0
339
+ },
340
+ "eval_B-Premise": {
341
+ "f1-score": 0.7888040712468193,
342
+ "precision": 0.7337278106508875,
343
+ "recall": 0.8528198074277854,
344
+ "support": 727.0
345
+ },
346
+ "eval_I-Claim": {
347
+ "f1-score": 0.5050387030816416,
348
+ "precision": 0.6394230769230769,
349
+ "recall": 0.41733043688148685,
350
+ "support": 4143.0
351
+ },
352
+ "eval_I-MajorClaim": {
353
+ "f1-score": 0.8131487889273357,
354
+ "precision": 0.7733443027560675,
355
+ "recall": 0.8572731418148655,
356
+ "support": 2193.0
357
+ },
358
+ "eval_I-Premise": {
359
+ "f1-score": 0.8898820701569895,
360
+ "precision": 0.8424233805020266,
361
+ "recall": 0.9430072434927963,
362
+ "support": 12563.0
363
+ },
364
+ "eval_O": {
365
+ "f1-score": 0.9201684886169893,
366
+ "precision": 0.9398688793280066,
367
+ "recall": 0.9012770137524558,
368
+ "support": 10180.0
369
+ },
370
+ "eval_accuracy": 0.8434742373665598,
371
+ "eval_loss": 0.4916144609451294,
372
+ "eval_macro avg": {
373
+ "f1-score": 0.7367027692121344,
374
+ "precision": 0.7339127261465722,
375
+ "recall": 0.7516420989254959,
376
+ "support": 30257.0
377
+ },
378
+ "eval_runtime": 1.4434,
379
+ "eval_samples_per_second": 56.118,
380
+ "eval_steps_per_second": 7.621,
381
+ "eval_weighted avg": {
382
+ "f1-score": 0.8349273131912464,
383
+ "precision": 0.8360386273187937,
384
+ "recall": 0.8434742373665598,
385
+ "support": 30257.0
386
+ },
387
+ "step": 246
388
+ },
389
+ {
390
+ "epoch": 7.0,
391
+ "eval_B-Claim": {
392
+ "f1-score": 0.5845648604269295,
393
+ "precision": 0.5668789808917197,
394
+ "recall": 0.6033898305084746,
395
+ "support": 295.0
396
+ },
397
+ "eval_B-MajorClaim": {
398
+ "f1-score": 0.7261904761904762,
399
+ "precision": 0.6777777777777778,
400
+ "recall": 0.782051282051282,
401
+ "support": 156.0
402
+ },
403
+ "eval_B-Premise": {
404
+ "f1-score": 0.8031088082901554,
405
+ "precision": 0.758873929008568,
406
+ "recall": 0.8528198074277854,
407
+ "support": 727.0
408
+ },
409
+ "eval_I-Claim": {
410
+ "f1-score": 0.5772882643372579,
411
+ "precision": 0.6070287539936102,
412
+ "recall": 0.5503258508327299,
413
+ "support": 4143.0
414
+ },
415
+ "eval_I-MajorClaim": {
416
+ "f1-score": 0.8086917562724014,
417
+ "precision": 0.7948040510788199,
418
+ "recall": 0.8230734154126766,
419
+ "support": 2193.0
420
+ },
421
+ "eval_I-Premise": {
422
+ "f1-score": 0.8942490402140614,
423
+ "precision": 0.8718995765275257,
424
+ "recall": 0.9177744169386293,
425
+ "support": 12563.0
426
+ },
427
+ "eval_O": {
428
+ "f1-score": 0.9187421383647798,
429
+ "precision": 0.9417225373904075,
430
+ "recall": 0.8968565815324165,
431
+ "support": 10180.0
432
+ },
433
+ "eval_accuracy": 0.8482334666358198,
434
+ "eval_loss": 0.5145291686058044,
435
+ "eval_macro avg": {
436
+ "f1-score": 0.7589764777280089,
437
+ "precision": 0.745569372381204,
438
+ "recall": 0.7751844549577135,
439
+ "support": 30257.0
440
+ },
441
+ "eval_runtime": 1.4296,
442
+ "eval_samples_per_second": 56.659,
443
+ "eval_steps_per_second": 7.694,
444
+ "eval_weighted avg": {
445
+ "f1-score": 0.8468124537513947,
446
+ "precision": 0.8468453317066018,
447
+ "recall": 0.8482334666358198,
448
+ "support": 30257.0
449
+ },
450
+ "step": 287
451
+ },
452
+ {
453
+ "epoch": 8.0,
454
+ "eval_B-Claim": {
455
+ "f1-score": 0.5750394944707741,
456
+ "precision": 0.5384615384615384,
457
+ "recall": 0.6169491525423729,
458
+ "support": 295.0
459
+ },
460
+ "eval_B-MajorClaim": {
461
+ "f1-score": 0.7195121951219512,
462
+ "precision": 0.686046511627907,
463
+ "recall": 0.7564102564102564,
464
+ "support": 156.0
465
+ },
466
+ "eval_B-Premise": {
467
+ "f1-score": 0.7879185817465528,
468
+ "precision": 0.7537688442211056,
469
+ "recall": 0.8253094910591472,
470
+ "support": 727.0
471
+ },
472
+ "eval_I-Claim": {
473
+ "f1-score": 0.5927414588012767,
474
+ "precision": 0.5808619091751622,
475
+ "recall": 0.6051170649287956,
476
+ "support": 4143.0
477
+ },
478
+ "eval_I-MajorClaim": {
479
+ "f1-score": 0.7924528301886793,
480
+ "precision": 0.8207132388861749,
481
+ "recall": 0.7660738714090287,
482
+ "support": 2193.0
483
+ },
484
+ "eval_I-Premise": {
485
+ "f1-score": 0.8916528468767275,
486
+ "precision": 0.8846666144323435,
487
+ "recall": 0.8987502984955823,
488
+ "support": 12563.0
489
+ },
490
+ "eval_O": {
491
+ "f1-score": 0.9119720069982503,
492
+ "precision": 0.9284478371501272,
493
+ "recall": 0.8960707269155206,
494
+ "support": 10180.0
495
+ },
496
+ "eval_accuracy": 0.8427801830981261,
497
+ "eval_loss": 0.538555920124054,
498
+ "eval_macro avg": {
499
+ "f1-score": 0.7530413448863159,
500
+ "precision": 0.7418523562791942,
501
+ "recall": 0.7663829802515292,
502
+ "support": 30257.0
503
+ },
504
+ "eval_runtime": 1.4347,
505
+ "eval_samples_per_second": 56.458,
506
+ "eval_steps_per_second": 7.667,
507
+ "eval_weighted avg": {
508
+ "f1-score": 0.8439033621710555,
509
+ "precision": 0.8456178749149208,
510
+ "recall": 0.8427801830981261,
511
+ "support": 30257.0
512
+ },
513
+ "step": 328
514
+ },
515
+ {
516
+ "epoch": 9.0,
517
+ "eval_B-Claim": {
518
+ "f1-score": 0.5622895622895623,
519
+ "precision": 0.5585284280936454,
520
+ "recall": 0.5661016949152542,
521
+ "support": 295.0
522
+ },
523
+ "eval_B-MajorClaim": {
524
+ "f1-score": 0.7289156626506025,
525
+ "precision": 0.6875,
526
+ "recall": 0.7756410256410257,
527
+ "support": 156.0
528
+ },
529
+ "eval_B-Premise": {
530
+ "f1-score": 0.7851186658114176,
531
+ "precision": 0.7355769230769231,
532
+ "recall": 0.8418156808803301,
533
+ "support": 727.0
534
+ },
535
+ "eval_I-Claim": {
536
+ "f1-score": 0.587080745341615,
537
+ "precision": 0.604811876119785,
538
+ "recall": 0.570359642770939,
539
+ "support": 4143.0
540
+ },
541
+ "eval_I-MajorClaim": {
542
+ "f1-score": 0.8209806157354618,
543
+ "precision": 0.8211678832116789,
544
+ "recall": 0.8207934336525308,
545
+ "support": 2193.0
546
+ },
547
+ "eval_I-Premise": {
548
+ "f1-score": 0.8914104344376116,
549
+ "precision": 0.888625217528872,
550
+ "recall": 0.8942131656451484,
551
+ "support": 12563.0
552
+ },
553
+ "eval_O": {
554
+ "f1-score": 0.917259306488793,
555
+ "precision": 0.91595650896268,
556
+ "recall": 0.918565815324165,
557
+ "support": 10180.0
558
+ },
559
+ "eval_accuracy": 0.8476716131804211,
560
+ "eval_loss": 0.5474137663841248,
561
+ "eval_macro avg": {
562
+ "f1-score": 0.7561507132507234,
563
+ "precision": 0.7445952624276548,
564
+ "recall": 0.7696414941184847,
565
+ "support": 30257.0
566
+ },
567
+ "eval_runtime": 1.4423,
568
+ "eval_samples_per_second": 56.16,
569
+ "eval_steps_per_second": 7.627,
570
+ "eval_weighted avg": {
571
+ "f1-score": 0.8467310896799954,
572
+ "precision": 0.8461371897103921,
573
+ "recall": 0.8476716131804211,
574
+ "support": 30257.0
575
+ },
576
+ "step": 369
577
+ },
578
+ {
579
+ "epoch": 10.0,
580
+ "eval_B-Claim": {
581
+ "f1-score": 0.5833333333333334,
582
+ "precision": 0.5531914893617021,
583
+ "recall": 0.6169491525423729,
584
+ "support": 295.0
585
+ },
586
+ "eval_B-MajorClaim": {
587
+ "f1-score": 0.7380281690140845,
588
+ "precision": 0.6582914572864321,
589
+ "recall": 0.8397435897435898,
590
+ "support": 156.0
591
+ },
592
+ "eval_B-Premise": {
593
+ "f1-score": 0.782495101241019,
594
+ "precision": 0.7450248756218906,
595
+ "recall": 0.8239339752407153,
596
+ "support": 727.0
597
+ },
598
+ "eval_I-Claim": {
599
+ "f1-score": 0.5871350941488689,
600
+ "precision": 0.6161803713527851,
601
+ "recall": 0.5607048032826454,
602
+ "support": 4143.0
603
+ },
604
+ "eval_I-MajorClaim": {
605
+ "f1-score": 0.8225702638918686,
606
+ "precision": 0.776742301458671,
607
+ "recall": 0.8741450068399452,
608
+ "support": 2193.0
609
+ },
610
+ "eval_I-Premise": {
611
+ "f1-score": 0.888669171445521,
612
+ "precision": 0.8831158622858041,
613
+ "recall": 0.8942927644670858,
614
+ "support": 12563.0
615
+ },
616
+ "eval_O": {
617
+ "f1-score": 0.9133780094316207,
618
+ "precision": 0.9232313095835424,
619
+ "recall": 0.9037328094302554,
620
+ "support": 10180.0
621
+ },
622
+ "eval_accuracy": 0.8456555507816373,
623
+ "eval_loss": 0.5961998701095581,
624
+ "eval_macro avg": {
625
+ "f1-score": 0.7593727346437593,
626
+ "precision": 0.7365396667072609,
627
+ "recall": 0.7876431573638013,
628
+ "support": 30257.0
629
+ },
630
+ "eval_runtime": 1.4347,
631
+ "eval_samples_per_second": 56.458,
632
+ "eval_steps_per_second": 7.667,
633
+ "eval_weighted avg": {
634
+ "f1-score": 0.844598799846626,
635
+ "precision": 0.8446583764532496,
636
+ "recall": 0.8456555507816373,
637
+ "support": 30257.0
638
+ },
639
+ "step": 410
640
+ },
641
+ {
642
+ "epoch": 11.0,
643
+ "eval_B-Claim": {
644
+ "f1-score": 0.577922077922078,
645
+ "precision": 0.5545171339563862,
646
+ "recall": 0.6033898305084746,
647
+ "support": 295.0
648
+ },
649
+ "eval_B-MajorClaim": {
650
+ "f1-score": 0.7282913165266107,
651
+ "precision": 0.6467661691542289,
652
+ "recall": 0.8333333333333334,
653
+ "support": 156.0
654
+ },
655
+ "eval_B-Premise": {
656
+ "f1-score": 0.7925283522348232,
657
+ "precision": 0.7694300518134715,
658
+ "recall": 0.8170563961485557,
659
+ "support": 727.0
660
+ },
661
+ "eval_I-Claim": {
662
+ "f1-score": 0.6015424164524421,
663
+ "precision": 0.6102831594634873,
664
+ "recall": 0.5930485155684286,
665
+ "support": 4143.0
666
+ },
667
+ "eval_I-MajorClaim": {
668
+ "f1-score": 0.8221789040502492,
669
+ "precision": 0.783003300330033,
670
+ "recall": 0.8654810761513908,
671
+ "support": 2193.0
672
+ },
673
+ "eval_I-Premise": {
674
+ "f1-score": 0.8906343447230972,
675
+ "precision": 0.892235181338872,
676
+ "recall": 0.8890392422192152,
677
+ "support": 12563.0
678
+ },
679
+ "eval_O": {
680
+ "f1-score": 0.917472118959108,
681
+ "precision": 0.9259629814907454,
682
+ "recall": 0.9091355599214146,
683
+ "support": 10180.0
684
+ },
685
+ "eval_accuracy": 0.8487622698879598,
686
+ "eval_loss": 0.5956980586051941,
687
+ "eval_macro avg": {
688
+ "f1-score": 0.7615099329812013,
689
+ "precision": 0.7403139967924607,
690
+ "recall": 0.7872119934072589,
691
+ "support": 30257.0
692
+ },
693
+ "eval_runtime": 1.4315,
694
+ "eval_samples_per_second": 56.583,
695
+ "eval_steps_per_second": 7.684,
696
+ "eval_weighted avg": {
697
+ "f1-score": 0.8488747259194414,
698
+ "precision": 0.8495500818518233,
699
+ "recall": 0.8487622698879598,
700
+ "support": 30257.0
701
+ },
702
+ "step": 451
703
+ },
704
+ {
705
+ "epoch": 12.0,
706
+ "eval_B-Claim": {
707
+ "f1-score": 0.5705024311183144,
708
+ "precision": 0.546583850931677,
709
+ "recall": 0.5966101694915255,
710
+ "support": 295.0
711
+ },
712
+ "eval_B-MajorClaim": {
713
+ "f1-score": 0.7277936962750716,
714
+ "precision": 0.6580310880829016,
715
+ "recall": 0.8141025641025641,
716
+ "support": 156.0
717
+ },
718
+ "eval_B-Premise": {
719
+ "f1-score": 0.7944845699277742,
720
+ "precision": 0.7600502512562815,
721
+ "recall": 0.8321870701513068,
722
+ "support": 727.0
723
+ },
724
+ "eval_I-Claim": {
725
+ "f1-score": 0.5785911255953873,
726
+ "precision": 0.6018252933507171,
727
+ "recall": 0.5570842384745354,
728
+ "support": 4143.0
729
+ },
730
+ "eval_I-MajorClaim": {
731
+ "f1-score": 0.8082901554404146,
732
+ "precision": 0.7675276752767528,
733
+ "recall": 0.853625170998632,
734
+ "support": 2193.0
735
+ },
736
+ "eval_I-Premise": {
737
+ "f1-score": 0.8910157942416007,
738
+ "precision": 0.8818805551224076,
739
+ "recall": 0.9003422749343309,
740
+ "support": 12563.0
741
+ },
742
+ "eval_O": {
743
+ "f1-score": 0.9105163287725957,
744
+ "precision": 0.9259597806215722,
745
+ "recall": 0.8955795677799607,
746
+ "support": 10180.0
747
+ },
748
+ "eval_accuracy": 0.8433089863502661,
749
+ "eval_loss": 0.6362738013267517,
750
+ "eval_macro avg": {
751
+ "f1-score": 0.7544563001958797,
752
+ "precision": 0.73455121352033,
753
+ "recall": 0.7785044365618364,
754
+ "support": 30257.0
755
+ },
756
+ "eval_runtime": 1.4338,
757
+ "eval_samples_per_second": 56.492,
758
+ "eval_steps_per_second": 7.672,
759
+ "eval_weighted avg": {
760
+ "f1-score": 0.8425156264772145,
761
+ "precision": 0.8427252530453597,
762
+ "recall": 0.8433089863502661,
763
+ "support": 30257.0
764
+ },
765
+ "step": 492
766
+ },
767
+ {
768
+ "epoch": 12.2,
769
+ "learning_rate": 4.75609756097561e-06,
770
+ "loss": 0.3164,
771
+ "step": 500
772
+ },
773
+ {
774
+ "epoch": 13.0,
775
+ "eval_B-Claim": {
776
+ "f1-score": 0.5771144278606964,
777
+ "precision": 0.564935064935065,
778
+ "recall": 0.5898305084745763,
779
+ "support": 295.0
780
+ },
781
+ "eval_B-MajorClaim": {
782
+ "f1-score": 0.7350427350427349,
783
+ "precision": 0.6615384615384615,
784
+ "recall": 0.8269230769230769,
785
+ "support": 156.0
786
+ },
787
+ "eval_B-Premise": {
788
+ "f1-score": 0.7949901120632828,
789
+ "precision": 0.7632911392405063,
790
+ "recall": 0.8294360385144429,
791
+ "support": 727.0
792
+ },
793
+ "eval_I-Claim": {
794
+ "f1-score": 0.5925018377848567,
795
+ "precision": 0.6016421995521274,
796
+ "recall": 0.5836350470673425,
797
+ "support": 4143.0
798
+ },
799
+ "eval_I-MajorClaim": {
800
+ "f1-score": 0.8274802458296752,
801
+ "precision": 0.7977147693609818,
802
+ "recall": 0.8595531235750113,
803
+ "support": 2193.0
804
+ },
805
+ "eval_I-Premise": {
806
+ "f1-score": 0.8883437524728971,
807
+ "precision": 0.8831720556997876,
808
+ "recall": 0.8935763750696489,
809
+ "support": 12563.0
810
+ },
811
+ "eval_O": {
812
+ "f1-score": 0.9159642910578025,
813
+ "precision": 0.9303008813696687,
814
+ "recall": 0.9020628683693517,
815
+ "support": 10180.0
816
+ },
817
+ "eval_accuracy": 0.8466801070826585,
818
+ "eval_loss": 0.6392464637756348,
819
+ "eval_macro avg": {
820
+ "f1-score": 0.7616339145874208,
821
+ "precision": 0.743227795956657,
822
+ "recall": 0.7835738625704929,
823
+ "support": 30257.0
824
+ },
825
+ "eval_runtime": 1.4326,
826
+ "eval_samples_per_second": 56.541,
827
+ "eval_steps_per_second": 7.678,
828
+ "eval_weighted avg": {
829
+ "f1-score": 0.8466487613672259,
830
+ "precision": 0.8471598021098077,
831
+ "recall": 0.8466801070826585,
832
+ "support": 30257.0
833
+ },
834
+ "step": 533
835
+ }
836
+ ],
837
+ "logging_steps": 500,
838
+ "max_steps": 656,
839
+ "num_input_tokens_seen": 0,
840
+ "num_train_epochs": 16,
841
+ "save_steps": 500,
842
+ "total_flos": 1863475346187000.0,
843
+ "train_batch_size": 8,
844
+ "trial_name": null,
845
+ "trial_params": null
846
+ }
checkpoint-533/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2f1dc302eca5129fec8b02daa92ee1b99cab9ac1f7862cc767c288e09ad4b9f2
3
+ size 4664
checkpoint-533/vocab.json ADDED
The diff for this file is too large to render. See raw diff