Hugging Face's logo

xshubhamx
/
legal-bert-base-uncased

Text Classification
Transformers
TensorBoard
Safetensors
bert
Generated from Trainer
Inference Endpoints
Model card Files Metrics Community
xshubhamx commited on
Commit
3766d53
•
1 Parent(s): 444f25e

Upload folder using huggingface_hub

Browse files
Files changed (7) hide show
  1. training_checkpoints/checkpoint-9645/config.json +63 -0
  2. training_checkpoints/checkpoint-9645/model.safetensors +3 -0
  3. training_checkpoints/checkpoint-9645/optimizer.pt +3 -0
  4. training_checkpoints/checkpoint-9645/rng_state.pth +3 -0
  5. training_checkpoints/checkpoint-9645/scheduler.pt +3 -0
  6. training_checkpoints/checkpoint-9645/trainer_state.json +484 -0
  7. training_checkpoints/checkpoint-9645/training_args.bin +3 -0
training_checkpoints/checkpoint-9645/config.json ADDED
@@ -0,0 +1,63 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "nlpaueb/legal-bert-base-uncased",
3
+ "architectures": [
4
+ "BertForSequenceClassification"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.1,
7
+ "bos_token_id": 0,
8
+ "classifier_dropout": null,
9
+ "eos_token_ids": 0,
10
+ "hidden_act": "gelu",
11
+ "hidden_dropout_prob": 0.1,
12
+ "hidden_size": 768,
13
+ "id2label": {
14
+ "0": "Issue",
15
+ "1": "Court Discourse",
16
+ "2": "Conclusion",
17
+ "3": "Precedent Analysis",
18
+ "4": "Section Analysis",
19
+ "5": "Argument by Petitioner",
20
+ "6": "Fact",
21
+ "7": "Argument by Respondent",
22
+ "8": "Ratio",
23
+ "9": "Appellant",
24
+ "10": "Respondent",
25
+ "11": "Argument by Appellant",
26
+ "12": "Petitioner",
27
+ "13": "Judge",
28
+ "14": "Argument by Defendant"
29
+ },
30
+ "initializer_range": 0.02,
31
+ "intermediate_size": 3072,
32
+ "label2id": {
33
+ "Appellant": 9,
34
+ "Argument by Appellant": 11,
35
+ "Argument by Defendant": 14,
36
+ "Argument by Petitioner": 5,
37
+ "Argument by Respondent": 7,
38
+ "Conclusion": 2,
39
+ "Court Discourse": 1,
40
+ "Fact": 6,
41
+ "Issue": 0,
42
+ "Judge": 13,
43
+ "Petitioner": 12,
44
+ "Precedent Analysis": 3,
45
+ "Ratio": 8,
46
+ "Respondent": 10,
47
+ "Section Analysis": 4
48
+ },
49
+ "layer_norm_eps": 1e-12,
50
+ "max_position_embeddings": 512,
51
+ "model_type": "bert",
52
+ "num_attention_heads": 12,
53
+ "num_hidden_layers": 12,
54
+ "output_past": true,
55
+ "pad_token_id": 0,
56
+ "position_embedding_type": "absolute",
57
+ "problem_type": "single_label_classification",
58
+ "torch_dtype": "float32",
59
+ "transformers_version": "4.38.2",
60
+ "type_vocab_size": 2,
61
+ "use_cache": true,
62
+ "vocab_size": 30522
63
+ }
training_checkpoints/checkpoint-9645/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b8d15abcec7eb5845e054fc3e83e9a47ece3adb5435f7e1245b9837f2c34871d
3
+ size 437998636
training_checkpoints/checkpoint-9645/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6cecfd1027a6966fa39c9643ddee6377cf16e0134a5254814736eb175cb803a4
3
+ size 876118266
training_checkpoints/checkpoint-9645/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3478af8c1ca7cb1ba6156bbbfde9f004b14ce219647297da45bc089d8408a0ce
3
+ size 14244
training_checkpoints/checkpoint-9645/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b2ea26d5863b134f939a91355b486dc3e0d6944d3b9f245a66a619e2a9272a31
3
+ size 1064
training_checkpoints/checkpoint-9645/trainer_state.json ADDED
@@ -0,0 +1,484 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.7839844036299938,
3
+ "best_model_checkpoint": "legal-bert-base-uncased/checkpoint-3858",
4
+ "epoch": 15.0,
5
+ "eval_steps": 500,
6
+ "global_step": 9645,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.78,
13
+ "grad_norm": 6.735908031463623,
14
+ "learning_rate": 4.742353551062727e-05,
15
+ "loss": 1.1096,
16
+ "step": 500
17
+ },
18
+ {
19
+ "epoch": 1.0,
20
+ "eval_accuracy": 0.7978311386522076,
21
+ "eval_f1_macro": 0.613418966848975,
22
+ "eval_f1_micro": 0.7978311386522076,
23
+ "eval_f1_weighted": 0.7839880886224406,
24
+ "eval_loss": 0.6747979521751404,
25
+ "eval_macro_fpr": 0.018830502815301575,
26
+ "eval_macro_sensitivity": 0.6339578261148517,
27
+ "eval_macro_specificity": 0.9845345883924012,
28
+ "eval_precision": 0.7854860044786733,
29
+ "eval_precision_macro": 0.6239440349375213,
30
+ "eval_recall": 0.7978311386522076,
31
+ "eval_recall_macro": 0.6339578261148517,
32
+ "eval_runtime": 30.9921,
33
+ "eval_samples_per_second": 41.656,
34
+ "eval_steps_per_second": 5.227,
35
+ "eval_weighted_fpr": 0.017778080512226686,
36
+ "eval_weighted_sensitivity": 0.7978311386522076,
37
+ "eval_weighted_specificity": 0.9701876872338092,
38
+ "step": 643
39
+ },
40
+ {
41
+ "epoch": 1.56,
42
+ "grad_norm": 6.824150562286377,
43
+ "learning_rate": 4.48315189217211e-05,
44
+ "loss": 0.6187,
45
+ "step": 1000
46
+ },
47
+ {
48
+ "epoch": 2.0,
49
+ "eval_accuracy": 0.8109992254066615,
50
+ "eval_f1_macro": 0.7268228216676939,
51
+ "eval_f1_micro": 0.8109992254066615,
52
+ "eval_f1_weighted": 0.8089875323563979,
53
+ "eval_loss": 0.6449354887008667,
54
+ "eval_macro_fpr": 0.016938802987367406,
55
+ "eval_macro_sensitivity": 0.7327307314578203,
56
+ "eval_macro_specificity": 0.9857686909763926,
57
+ "eval_precision": 0.8196084470842065,
58
+ "eval_precision_macro": 0.7805936449528967,
59
+ "eval_recall": 0.8109992254066615,
60
+ "eval_recall_macro": 0.7327307314578203,
61
+ "eval_runtime": 58.3071,
62
+ "eval_samples_per_second": 22.141,
63
+ "eval_steps_per_second": 2.778,
64
+ "eval_weighted_fpr": 0.016373641121997046,
65
+ "eval_weighted_sensitivity": 0.8109992254066615,
66
+ "eval_weighted_specificity": 0.9755311392392266,
67
+ "step": 1286
68
+ },
69
+ {
70
+ "epoch": 2.33,
71
+ "grad_norm": 20.75279998779297,
72
+ "learning_rate": 4.223950233281493e-05,
73
+ "loss": 0.4747,
74
+ "step": 1500
75
+ },
76
+ {
77
+ "epoch": 3.0,
78
+ "eval_accuracy": 0.814872192099148,
79
+ "eval_f1_macro": 0.7369957890904685,
80
+ "eval_f1_micro": 0.814872192099148,
81
+ "eval_f1_weighted": 0.8124596447054482,
82
+ "eval_loss": 0.8151384592056274,
83
+ "eval_macro_fpr": 0.016571707879599946,
84
+ "eval_macro_sensitivity": 0.739039840019417,
85
+ "eval_macro_specificity": 0.9860629698102752,
86
+ "eval_precision": 0.8192276160546627,
87
+ "eval_precision_macro": 0.7659183928188651,
88
+ "eval_recall": 0.814872192099148,
89
+ "eval_recall_macro": 0.739039840019417,
90
+ "eval_runtime": 55.9094,
91
+ "eval_samples_per_second": 23.091,
92
+ "eval_steps_per_second": 2.898,
93
+ "eval_weighted_fpr": 0.015968463954032203,
94
+ "eval_weighted_sensitivity": 0.814872192099148,
95
+ "eval_weighted_specificity": 0.9760723550549795,
96
+ "step": 1929
97
+ },
98
+ {
99
+ "epoch": 3.11,
100
+ "grad_norm": 10.737870216369629,
101
+ "learning_rate": 3.9652669777086575e-05,
102
+ "loss": 0.3866,
103
+ "step": 2000
104
+ },
105
+ {
106
+ "epoch": 3.89,
107
+ "grad_norm": 13.75886344909668,
108
+ "learning_rate": 3.7060653188180404e-05,
109
+ "loss": 0.2645,
110
+ "step": 2500
111
+ },
112
+ {
113
+ "epoch": 4.0,
114
+ "eval_accuracy": 0.8218435321456236,
115
+ "eval_f1_macro": 0.738478964900184,
116
+ "eval_f1_micro": 0.8218435321456236,
117
+ "eval_f1_weighted": 0.8189030195253059,
118
+ "eval_loss": 0.9344585537910461,
119
+ "eval_macro_fpr": 0.015763618193954002,
120
+ "eval_macro_sensitivity": 0.7412892486668318,
121
+ "eval_macro_specificity": 0.986618111597957,
122
+ "eval_precision": 0.8198244063808544,
123
+ "eval_precision_macro": 0.7446229748697999,
124
+ "eval_recall": 0.8218435321456236,
125
+ "eval_recall_macro": 0.7412892486668318,
126
+ "eval_runtime": 75.578,
127
+ "eval_samples_per_second": 17.082,
128
+ "eval_steps_per_second": 2.143,
129
+ "eval_weighted_fpr": 0.015247944842216919,
130
+ "eval_weighted_sensitivity": 0.8218435321456236,
131
+ "eval_weighted_specificity": 0.9774281418237326,
132
+ "step": 2572
133
+ },
134
+ {
135
+ "epoch": 4.67,
136
+ "grad_norm": 8.32991886138916,
137
+ "learning_rate": 3.4468636599274234e-05,
138
+ "loss": 0.1901,
139
+ "step": 3000
140
+ },
141
+ {
142
+ "epoch": 5.0,
143
+ "eval_accuracy": 0.8195197521301317,
144
+ "eval_f1_macro": 0.7594855831156184,
145
+ "eval_f1_micro": 0.8195197521301317,
146
+ "eval_f1_weighted": 0.81658937168797,
147
+ "eval_loss": 1.0929232835769653,
148
+ "eval_macro_fpr": 0.01614879224826684,
149
+ "eval_macro_sensitivity": 0.743232314276478,
150
+ "eval_macro_specificity": 0.9863035723232563,
151
+ "eval_precision": 0.8241924882578779,
152
+ "eval_precision_macro": 0.8264084869739744,
153
+ "eval_recall": 0.8195197521301317,
154
+ "eval_recall_macro": 0.743232314276478,
155
+ "eval_runtime": 65.7326,
156
+ "eval_samples_per_second": 19.64,
157
+ "eval_steps_per_second": 2.465,
158
+ "eval_weighted_fpr": 0.01548687271518777,
159
+ "eval_weighted_sensitivity": 0.8195197521301317,
160
+ "eval_weighted_specificity": 0.9750338327187125,
161
+ "step": 3215
162
+ },
163
+ {
164
+ "epoch": 5.44,
165
+ "grad_norm": 50.74209976196289,
166
+ "learning_rate": 3.187662001036807e-05,
167
+ "loss": 0.1131,
168
+ "step": 3500
169
+ },
170
+ {
171
+ "epoch": 6.0,
172
+ "eval_accuracy": 0.820294345468629,
173
+ "eval_f1_macro": 0.7839844036299938,
174
+ "eval_f1_micro": 0.8202943454686291,
175
+ "eval_f1_weighted": 0.8197212343639901,
176
+ "eval_loss": 1.1536158323287964,
177
+ "eval_macro_fpr": 0.01593810671400651,
178
+ "eval_macro_sensitivity": 0.7786487516932519,
179
+ "eval_macro_specificity": 0.9864617261648754,
180
+ "eval_precision": 0.821246410485655,
181
+ "eval_precision_macro": 0.7968460348821496,
182
+ "eval_recall": 0.820294345468629,
183
+ "eval_recall_macro": 0.7786487516932519,
184
+ "eval_runtime": 54.9389,
185
+ "eval_samples_per_second": 23.499,
186
+ "eval_steps_per_second": 2.949,
187
+ "eval_weighted_fpr": 0.015407092575375215,
188
+ "eval_weighted_sensitivity": 0.820294345468629,
189
+ "eval_weighted_specificity": 0.9766315470045015,
190
+ "step": 3858
191
+ },
192
+ {
193
+ "epoch": 6.22,
194
+ "grad_norm": 0.02055634744465351,
195
+ "learning_rate": 2.928978745463971e-05,
196
+ "loss": 0.0975,
197
+ "step": 4000
198
+ },
199
+ {
200
+ "epoch": 7.0,
201
+ "grad_norm": 32.151573181152344,
202
+ "learning_rate": 2.669777086573354e-05,
203
+ "loss": 0.063,
204
+ "step": 4500
205
+ },
206
+ {
207
+ "epoch": 7.0,
208
+ "eval_accuracy": 0.8117738187451587,
209
+ "eval_f1_macro": 0.7495479890901372,
210
+ "eval_f1_micro": 0.8117738187451587,
211
+ "eval_f1_weighted": 0.8135965211162124,
212
+ "eval_loss": 1.3217942714691162,
213
+ "eval_macro_fpr": 0.01664162427116977,
214
+ "eval_macro_sensitivity": 0.7526435450791772,
215
+ "eval_macro_specificity": 0.9859379584321952,
216
+ "eval_precision": 0.818420676521013,
217
+ "eval_precision_macro": 0.7517839435012339,
218
+ "eval_recall": 0.8117738187451587,
219
+ "eval_recall_macro": 0.7526435450791772,
220
+ "eval_runtime": 59.9636,
221
+ "eval_samples_per_second": 21.53,
222
+ "eval_steps_per_second": 2.702,
223
+ "eval_weighted_fpr": 0.016292323164599398,
224
+ "eval_weighted_sensitivity": 0.8117738187451587,
225
+ "eval_weighted_specificity": 0.9772955577377688,
226
+ "step": 4501
227
+ },
228
+ {
229
+ "epoch": 7.78,
230
+ "grad_norm": 0.0045046876184642315,
231
+ "learning_rate": 2.4105754276827372e-05,
232
+ "loss": 0.0264,
233
+ "step": 5000
234
+ },
235
+ {
236
+ "epoch": 8.0,
237
+ "eval_accuracy": 0.82571649883811,
238
+ "eval_f1_macro": 0.7730278469427766,
239
+ "eval_f1_micro": 0.82571649883811,
240
+ "eval_f1_weighted": 0.8247069887539016,
241
+ "eval_loss": 1.3862823247909546,
242
+ "eval_macro_fpr": 0.015458317393184977,
243
+ "eval_macro_sensitivity": 0.7767515241074096,
244
+ "eval_macro_specificity": 0.9868314420976501,
245
+ "eval_precision": 0.8261549954416628,
246
+ "eval_precision_macro": 0.7783699125562222,
247
+ "eval_recall": 0.82571649883811,
248
+ "eval_recall_macro": 0.7767515241074096,
249
+ "eval_runtime": 58.9779,
250
+ "eval_samples_per_second": 21.89,
251
+ "eval_steps_per_second": 2.747,
252
+ "eval_weighted_fpr": 0.01485246550927454,
253
+ "eval_weighted_sensitivity": 0.82571649883811,
254
+ "eval_weighted_specificity": 0.97675513262664,
255
+ "step": 5144
256
+ },
257
+ {
258
+ "epoch": 8.55,
259
+ "grad_norm": 0.007681012153625488,
260
+ "learning_rate": 2.1513737687921205e-05,
261
+ "loss": 0.03,
262
+ "step": 5500
263
+ },
264
+ {
265
+ "epoch": 9.0,
266
+ "eval_accuracy": 0.8079008520526724,
267
+ "eval_f1_macro": 0.7594758314568825,
268
+ "eval_f1_micro": 0.8079008520526724,
269
+ "eval_f1_weighted": 0.8095900464774608,
270
+ "eval_loss": 1.5542314052581787,
271
+ "eval_macro_fpr": 0.017245749289239998,
272
+ "eval_macro_sensitivity": 0.7652642279270354,
273
+ "eval_macro_specificity": 0.9854886624123184,
274
+ "eval_precision": 0.8166769511091512,
275
+ "eval_precision_macro": 0.7639132572736053,
276
+ "eval_recall": 0.8079008520526724,
277
+ "eval_recall_macro": 0.7652642279270354,
278
+ "eval_runtime": 70.914,
279
+ "eval_samples_per_second": 18.205,
280
+ "eval_steps_per_second": 2.284,
281
+ "eval_weighted_fpr": 0.0167003367003367,
282
+ "eval_weighted_sensitivity": 0.8079008520526724,
283
+ "eval_weighted_specificity": 0.9744290841320995,
284
+ "step": 5787
285
+ },
286
+ {
287
+ "epoch": 9.33,
288
+ "grad_norm": 0.0017093586502596736,
289
+ "learning_rate": 1.8926905132192844e-05,
290
+ "loss": 0.0149,
291
+ "step": 6000
292
+ },
293
+ {
294
+ "epoch": 10.0,
295
+ "eval_accuracy": 0.8140975987606507,
296
+ "eval_f1_macro": 0.7412163056128125,
297
+ "eval_f1_micro": 0.8140975987606507,
298
+ "eval_f1_weighted": 0.8126789471820407,
299
+ "eval_loss": 1.583540678024292,
300
+ "eval_macro_fpr": 0.016780932672955,
301
+ "eval_macro_sensitivity": 0.7360563323248509,
302
+ "eval_macro_specificity": 0.9858047486397965,
303
+ "eval_precision": 0.8155262118469742,
304
+ "eval_precision_macro": 0.7545069126135399,
305
+ "eval_recall": 0.8140975987606507,
306
+ "eval_recall_macro": 0.7360563323248509,
307
+ "eval_runtime": 70.4882,
308
+ "eval_samples_per_second": 18.315,
309
+ "eval_steps_per_second": 2.298,
310
+ "eval_weighted_fpr": 0.016049217600641968,
311
+ "eval_weighted_sensitivity": 0.8140975987606507,
312
+ "eval_weighted_specificity": 0.9729736308362953,
313
+ "step": 6430
314
+ },
315
+ {
316
+ "epoch": 10.11,
317
+ "grad_norm": 0.004950440023094416,
318
+ "learning_rate": 1.6334888543286677e-05,
319
+ "loss": 0.0092,
320
+ "step": 6500
321
+ },
322
+ {
323
+ "epoch": 10.89,
324
+ "grad_norm": 0.004135847091674805,
325
+ "learning_rate": 1.374287195438051e-05,
326
+ "loss": 0.005,
327
+ "step": 7000
328
+ },
329
+ {
330
+ "epoch": 11.0,
331
+ "eval_accuracy": 0.8241673121611154,
332
+ "eval_f1_macro": 0.7681252828650814,
333
+ "eval_f1_micro": 0.8241673121611154,
334
+ "eval_f1_weighted": 0.8226392641142015,
335
+ "eval_loss": 1.5324584245681763,
336
+ "eval_macro_fpr": 0.015637185816662357,
337
+ "eval_macro_sensitivity": 0.7811775190571517,
338
+ "eval_macro_specificity": 0.9866649056306106,
339
+ "eval_precision": 0.8249899677721785,
340
+ "eval_precision_macro": 0.780545510593793,
341
+ "eval_recall": 0.8241673121611154,
342
+ "eval_recall_macro": 0.7811775190571517,
343
+ "eval_runtime": 67.2418,
344
+ "eval_samples_per_second": 19.199,
345
+ "eval_steps_per_second": 2.409,
346
+ "eval_weighted_fpr": 0.015010249289162203,
347
+ "eval_weighted_sensitivity": 0.8241673121611154,
348
+ "eval_weighted_specificity": 0.9758062722980435,
349
+ "step": 7073
350
+ },
351
+ {
352
+ "epoch": 11.66,
353
+ "grad_norm": 0.0040180375799536705,
354
+ "learning_rate": 1.115085536547434e-05,
355
+ "loss": 0.003,
356
+ "step": 7500
357
+ },
358
+ {
359
+ "epoch": 12.0,
360
+ "eval_accuracy": 0.8288148721920992,
361
+ "eval_f1_macro": 0.7625723813018138,
362
+ "eval_f1_micro": 0.8288148721920992,
363
+ "eval_f1_weighted": 0.8276081900441157,
364
+ "eval_loss": 1.5713844299316406,
365
+ "eval_macro_fpr": 0.015190042544950861,
366
+ "eval_macro_sensitivity": 0.7678635910180709,
367
+ "eval_macro_specificity": 0.9870237726059469,
368
+ "eval_precision": 0.8299388397158192,
369
+ "eval_precision_macro": 0.7700952459098459,
370
+ "eval_recall": 0.8288148721920992,
371
+ "eval_recall_macro": 0.7678635910180709,
372
+ "eval_runtime": 61.0459,
373
+ "eval_samples_per_second": 21.148,
374
+ "eval_steps_per_second": 2.654,
375
+ "eval_weighted_fpr": 0.014538517202815604,
376
+ "eval_weighted_sensitivity": 0.8288148721920992,
377
+ "eval_weighted_specificity": 0.9765417168971026,
378
+ "step": 7716
379
+ },
380
+ {
381
+ "epoch": 12.44,
382
+ "grad_norm": 0.004260718822479248,
383
+ "learning_rate": 8.55883877656817e-06,
384
+ "loss": 0.0033,
385
+ "step": 8000
386
+ },
387
+ {
388
+ "epoch": 13.0,
389
+ "eval_accuracy": 0.8249419054996127,
390
+ "eval_f1_macro": 0.7607533001144507,
391
+ "eval_f1_micro": 0.8249419054996127,
392
+ "eval_f1_weighted": 0.8224643561391238,
393
+ "eval_loss": 1.551088571548462,
394
+ "eval_macro_fpr": 0.015560628647913471,
395
+ "eval_macro_sensitivity": 0.7597662805173221,
396
+ "eval_macro_specificity": 0.9867279232222673,
397
+ "eval_precision": 0.821925563604887,
398
+ "eval_precision_macro": 0.767644957538908,
399
+ "eval_recall": 0.8249419054996127,
400
+ "eval_recall_macro": 0.7597662805173221,
401
+ "eval_runtime": 66.776,
402
+ "eval_samples_per_second": 19.333,
403
+ "eval_steps_per_second": 2.426,
404
+ "eval_weighted_fpr": 0.014931289640591967,
405
+ "eval_weighted_sensitivity": 0.8249419054996127,
406
+ "eval_weighted_specificity": 0.9759769428343987,
407
+ "step": 8359
408
+ },
409
+ {
410
+ "epoch": 13.22,
411
+ "grad_norm": 0.001955605112016201,
412
+ "learning_rate": 5.9668221876620015e-06,
413
+ "loss": 0.0043,
414
+ "step": 8500
415
+ },
416
+ {
417
+ "epoch": 14.0,
418
+ "grad_norm": 0.0009153374703601003,
419
+ "learning_rate": 3.379989631933644e-06,
420
+ "loss": 0.0018,
421
+ "step": 9000
422
+ },
423
+ {
424
+ "epoch": 14.0,
425
+ "eval_accuracy": 0.8249419054996127,
426
+ "eval_f1_macro": 0.7571714820627711,
427
+ "eval_f1_micro": 0.8249419054996127,
428
+ "eval_f1_weighted": 0.8224001833855348,
429
+ "eval_loss": 1.5509757995605469,
430
+ "eval_macro_fpr": 0.015514116711070718,
431
+ "eval_macro_sensitivity": 0.7554168858318507,
432
+ "eval_macro_specificity": 0.9867787829147775,
433
+ "eval_precision": 0.8225003050604829,
434
+ "eval_precision_macro": 0.768615439122566,
435
+ "eval_recall": 0.8249419054996127,
436
+ "eval_recall_macro": 0.7554168858318507,
437
+ "eval_runtime": 56.3702,
438
+ "eval_samples_per_second": 22.902,
439
+ "eval_steps_per_second": 2.874,
440
+ "eval_weighted_fpr": 0.014931289640591967,
441
+ "eval_weighted_sensitivity": 0.8249419054996127,
442
+ "eval_weighted_specificity": 0.97673983822205,
443
+ "step": 9002
444
+ },
445
+ {
446
+ "epoch": 14.77,
447
+ "grad_norm": 0.0028363633900880814,
448
+ "learning_rate": 7.879730430274754e-07,
449
+ "loss": 0.0008,
450
+ "step": 9500
451
+ },
452
+ {
453
+ "epoch": 15.0,
454
+ "eval_accuracy": 0.8241673121611154,
455
+ "eval_f1_macro": 0.7566047769559027,
456
+ "eval_f1_micro": 0.8241673121611154,
457
+ "eval_f1_weighted": 0.8221253561257641,
458
+ "eval_loss": 1.5468684434890747,
459
+ "eval_macro_fpr": 0.015578325462564525,
460
+ "eval_macro_sensitivity": 0.7547799189157873,
461
+ "eval_macro_specificity": 0.9867164858554927,
462
+ "eval_precision": 0.8220097803512278,
463
+ "eval_precision_macro": 0.7660327191463564,
464
+ "eval_recall": 0.8241673121611154,
465
+ "eval_recall_macro": 0.7547799189157873,
466
+ "eval_runtime": 62.3696,
467
+ "eval_samples_per_second": 20.699,
468
+ "eval_steps_per_second": 2.597,
469
+ "eval_weighted_fpr": 0.015010249289162203,
470
+ "eval_weighted_sensitivity": 0.8241673121611154,
471
+ "eval_weighted_specificity": 0.9765799756712747,
472
+ "step": 9645
473
+ }
474
+ ],
475
+ "logging_steps": 500,
476
+ "max_steps": 9645,
477
+ "num_input_tokens_seen": 0,
478
+ "num_train_epochs": 15,
479
+ "save_steps": 500,
480
+ "total_flos": 2.028033591671808e+16,
481
+ "train_batch_size": 8,
482
+ "trial_name": null,
483
+ "trial_params": null
484
+ }
training_checkpoints/checkpoint-9645/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:21938ff48cdf92422f693d69544854f8291bee6a4c7091c49d48bd0e59616530
3
+ size 4920