Hugging Face's logo

xshubhamx
/
roberta-base

Text Classification
Transformers
TensorBoard
Safetensors
roberta
Generated from Trainer
text-embeddings-inference
Model card Files
xet
 Metrics Community
xshubhamx commited on
Commit
71837ea
·
verified ·
1 Parent(s): 912c7ab

Upload folder using huggingface_hub

Browse files
Files changed (7) hide show
  1. training_checkpoints/checkpoint-9002/config.json +62 -0
  2. training_checkpoints/checkpoint-9002/model.safetensors +3 -0
  3. training_checkpoints/checkpoint-9002/optimizer.pt +3 -0
  4. training_checkpoints/checkpoint-9002/rng_state.pth +3 -0
  5. training_checkpoints/checkpoint-9002/scheduler.pt +3 -0
  6. training_checkpoints/checkpoint-9002/trainer_state.json +455 -0
  7. training_checkpoints/checkpoint-9002/training_args.bin +3 -0
training_checkpoints/checkpoint-9002/config.json ADDED
@@ -0,0 +1,62 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "roberta-base",
3
+ "architectures": [
4
+ "RobertaForSequenceClassification"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.1,
7
+ "bos_token_id": 0,
8
+ "classifier_dropout": null,
9
+ "eos_token_id": 2,
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-05,
50
+ "max_position_embeddings": 514,
51
+ "model_type": "roberta",
52
+ "num_attention_heads": 12,
53
+ "num_hidden_layers": 12,
54
+ "pad_token_id": 1,
55
+ "position_embedding_type": "absolute",
56
+ "problem_type": "single_label_classification",
57
+ "torch_dtype": "float32",
58
+ "transformers_version": "4.38.2",
59
+ "type_vocab_size": 1,
60
+ "use_cache": true,
61
+ "vocab_size": 50265
62
+ }
training_checkpoints/checkpoint-9002/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:acec199143766381ec042756627dae57b9da2c6058c402fd546f906f378d20ac
3
+ size 498652812
training_checkpoints/checkpoint-9002/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6515d45d405c4c8ab578459d85881e9f3c7d8faf456b4dc2ca2f841de252d2ed
3
+ size 997425402
training_checkpoints/checkpoint-9002/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d151b334398b4f9c1223c0119151f11053e38d72f3d594a6923fefe7381b0ef6
3
+ size 14244
training_checkpoints/checkpoint-9002/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:19ee98d598424113be8afcd1541f43eff7d5ea46b98e52df48522705ec0ba3f4
3
+ size 1064
training_checkpoints/checkpoint-9002/trainer_state.json ADDED
@@ -0,0 +1,455 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.34083702910591834,
3
+ "best_model_checkpoint": "roberta-base/checkpoint-9002",
4
+ "epoch": 14.0,
5
+ "eval_steps": 500,
6
+ "global_step": 9002,
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": 6.534026145935059,
14
+ "learning_rate": 4.742871954380509e-05,
15
+ "loss": 2.2214,
16
+ "step": 500
17
+ },
18
+ {
19
+ "epoch": 1.0,
20
+ "eval_accuracy": 0.24554608830364058,
21
+ "eval_f1_macro": 0.02628524046434494,
22
+ "eval_f1_micro": 0.24554608830364058,
23
+ "eval_f1_weighted": 0.09681356964210704,
24
+ "eval_loss": 2.1982107162475586,
25
+ "eval_macro_fpr": 0.06666666666666667,
26
+ "eval_macro_sensitivity": 0.06666666666666667,
27
+ "eval_macro_specificity": 0.9333333333333333,
28
+ "eval_precision": 0.06029288148121926,
29
+ "eval_precision_macro": 0.016369739220242704,
30
+ "eval_recall": 0.24554608830364058,
31
+ "eval_recall_macro": 0.06666666666666667,
32
+ "eval_runtime": 29.4439,
33
+ "eval_samples_per_second": 43.846,
34
+ "eval_steps_per_second": 5.502,
35
+ "eval_weighted_fpr": 0.17997043606799704,
36
+ "eval_weighted_sensitivity": 0.24554608830364058,
37
+ "eval_weighted_specificity": 0.7544539116963594,
38
+ "step": 643
39
+ },
40
+ {
41
+ "epoch": 1.56,
42
+ "grad_norm": 6.595059871673584,
43
+ "learning_rate": 4.483670295489891e-05,
44
+ "loss": 2.1462,
45
+ "step": 1000
46
+ },
47
+ {
48
+ "epoch": 2.0,
49
+ "eval_accuracy": 0.3361735089078234,
50
+ "eval_f1_macro": 0.06552880630375763,
51
+ "eval_f1_micro": 0.3361735089078234,
52
+ "eval_f1_weighted": 0.2010480796875688,
53
+ "eval_loss": 1.9634978771209717,
54
+ "eval_macro_fpr": 0.09092506634165738,
55
+ "eval_macro_sensitivity": 0.11547020741602194,
56
+ "eval_macro_specificity": 0.9457066851291582,
57
+ "eval_precision": 0.1591168191834871,
58
+ "eval_precision_macro": 0.05308920125936356,
59
+ "eval_recall": 0.3361735089078234,
60
+ "eval_recall_macro": 0.11547020741602194,
61
+ "eval_runtime": 57.2389,
62
+ "eval_samples_per_second": 22.555,
63
+ "eval_steps_per_second": 2.83,
64
+ "eval_weighted_fpr": 0.12361171210154334,
65
+ "eval_weighted_sensitivity": 0.3361735089078234,
66
+ "eval_weighted_specificity": 0.84942676802955,
67
+ "step": 1286
68
+ },
69
+ {
70
+ "epoch": 2.33,
71
+ "grad_norm": 41.05898666381836,
72
+ "learning_rate": 4.224987039917055e-05,
73
+ "loss": 1.995,
74
+ "step": 1500
75
+ },
76
+ {
77
+ "epoch": 3.0,
78
+ "eval_accuracy": 0.3555383423702556,
79
+ "eval_f1_macro": 0.09312335075016234,
80
+ "eval_f1_micro": 0.3555383423702556,
81
+ "eval_f1_weighted": 0.2541975590745503,
82
+ "eval_loss": 1.932247281074524,
83
+ "eval_macro_fpr": 0.09537141294157879,
84
+ "eval_macro_sensitivity": 0.14351969127084646,
85
+ "eval_macro_specificity": 0.9497795632717652,
86
+ "eval_precision": 0.23412920314698393,
87
+ "eval_precision_macro": 0.09260367628400626,
88
+ "eval_recall": 0.3555383423702556,
89
+ "eval_recall_macro": 0.14351969127084646,
90
+ "eval_runtime": 59.8887,
91
+ "eval_samples_per_second": 21.557,
92
+ "eval_steps_per_second": 2.705,
93
+ "eval_weighted_fpr": 0.11463213006337834,
94
+ "eval_weighted_sensitivity": 0.3555383423702556,
95
+ "eval_weighted_specificity": 0.8911551067062222,
96
+ "step": 1929
97
+ },
98
+ {
99
+ "epoch": 3.11,
100
+ "grad_norm": 4.287881851196289,
101
+ "learning_rate": 3.96630378434422e-05,
102
+ "loss": 2.0064,
103
+ "step": 2000
104
+ },
105
+ {
106
+ "epoch": 3.89,
107
+ "grad_norm": 5.713494777679443,
108
+ "learning_rate": 3.7076205287713846e-05,
109
+ "loss": 1.8909,
110
+ "step": 2500
111
+ },
112
+ {
113
+ "epoch": 4.0,
114
+ "eval_accuracy": 0.3787761425251743,
115
+ "eval_f1_macro": 0.14496386130680605,
116
+ "eval_f1_micro": 0.3787761425251743,
117
+ "eval_f1_weighted": 0.29773714484044667,
118
+ "eval_loss": 1.8874640464782715,
119
+ "eval_macro_fpr": 0.08706956494211866,
120
+ "eval_macro_sensitivity": 0.2191470222337379,
121
+ "eval_macro_specificity": 0.9525422186332462,
122
+ "eval_precision": 0.28749388015966415,
123
+ "eval_precision_macro": 0.12703293957809556,
124
+ "eval_recall": 0.3787761425251743,
125
+ "eval_recall_macro": 0.2191470222337379,
126
+ "eval_runtime": 69.4305,
127
+ "eval_samples_per_second": 18.594,
128
+ "eval_steps_per_second": 2.333,
129
+ "eval_weighted_fpr": 0.10486401673640168,
130
+ "eval_weighted_sensitivity": 0.3787761425251743,
131
+ "eval_weighted_specificity": 0.9093571369735185,
132
+ "step": 2572
133
+ },
134
+ {
135
+ "epoch": 4.67,
136
+ "grad_norm": 18.91859245300293,
137
+ "learning_rate": 3.4484188698807675e-05,
138
+ "loss": 1.7621,
139
+ "step": 3000
140
+ },
141
+ {
142
+ "epoch": 5.0,
143
+ "eval_accuracy": 0.42835011618900076,
144
+ "eval_f1_macro": 0.1745447316852942,
145
+ "eval_f1_micro": 0.4283501161890007,
146
+ "eval_f1_weighted": 0.3593798293323463,
147
+ "eval_loss": 1.7168936729431152,
148
+ "eval_macro_fpr": 0.07431772683182448,
149
+ "eval_macro_sensitivity": 0.2270221309090987,
150
+ "eval_macro_specificity": 0.95583829077833,
151
+ "eval_precision": 0.374728097525222,
152
+ "eval_precision_macro": 0.18202607308331148,
153
+ "eval_recall": 0.42835011618900076,
154
+ "eval_recall_macro": 0.2270221309090987,
155
+ "eval_runtime": 58.7135,
156
+ "eval_samples_per_second": 21.988,
157
+ "eval_steps_per_second": 2.759,
158
+ "eval_weighted_fpr": 0.08702830188679245,
159
+ "eval_weighted_sensitivity": 0.42835011618900076,
160
+ "eval_weighted_specificity": 0.9092242454859507,
161
+ "step": 3215
162
+ },
163
+ {
164
+ "epoch": 5.44,
165
+ "grad_norm": 7.408999919891357,
166
+ "learning_rate": 3.1892172109901504e-05,
167
+ "loss": 1.6254,
168
+ "step": 3500
169
+ },
170
+ {
171
+ "epoch": 6.0,
172
+ "eval_accuracy": 0.4500387296669249,
173
+ "eval_f1_macro": 0.18588331301714964,
174
+ "eval_f1_micro": 0.4500387296669249,
175
+ "eval_f1_weighted": 0.39697994452590485,
176
+ "eval_loss": 1.8831233978271484,
177
+ "eval_macro_fpr": 0.07944022523045574,
178
+ "eval_macro_sensitivity": 0.2544695464493438,
179
+ "eval_macro_specificity": 0.9582711519830998,
180
+ "eval_precision": 0.3947542941359455,
181
+ "eval_precision_macro": 0.1877780146030796,
182
+ "eval_recall": 0.4500387296669249,
183
+ "eval_recall_macro": 0.2544695464493438,
184
+ "eval_runtime": 65.7966,
185
+ "eval_samples_per_second": 19.621,
186
+ "eval_steps_per_second": 2.462,
187
+ "eval_weighted_fpr": 0.08028041610131162,
188
+ "eval_weighted_sensitivity": 0.4500387296669249,
189
+ "eval_weighted_specificity": 0.92402855007957,
190
+ "step": 3858
191
+ },
192
+ {
193
+ "epoch": 6.22,
194
+ "grad_norm": 8.139199256896973,
195
+ "learning_rate": 2.9300155520995337e-05,
196
+ "loss": 1.4996,
197
+ "step": 4000
198
+ },
199
+ {
200
+ "epoch": 7.0,
201
+ "grad_norm": 40.22551727294922,
202
+ "learning_rate": 2.671332296526698e-05,
203
+ "loss": 1.4139,
204
+ "step": 4500
205
+ },
206
+ {
207
+ "epoch": 7.0,
208
+ "eval_accuracy": 0.46398140975987606,
209
+ "eval_f1_macro": 0.22845278849292336,
210
+ "eval_f1_micro": 0.4639814097598761,
211
+ "eval_f1_weighted": 0.4297793996222424,
212
+ "eval_loss": 1.662539005279541,
213
+ "eval_macro_fpr": 0.07171513504664777,
214
+ "eval_macro_sensitivity": 0.2949397340388095,
215
+ "eval_macro_specificity": 0.9601261188979237,
216
+ "eval_precision": 0.451602882474783,
217
+ "eval_precision_macro": 0.232765042661752,
218
+ "eval_recall": 0.46398140975987606,
219
+ "eval_recall_macro": 0.2949397340388095,
220
+ "eval_runtime": 60.1854,
221
+ "eval_samples_per_second": 21.45,
222
+ "eval_steps_per_second": 2.692,
223
+ "eval_weighted_fpr": 0.07622824410663141,
224
+ "eval_weighted_sensitivity": 0.46398140975987606,
225
+ "eval_weighted_specificity": 0.9379103737089811,
226
+ "step": 4501
227
+ },
228
+ {
229
+ "epoch": 7.78,
230
+ "grad_norm": 5.4409565925598145,
231
+ "learning_rate": 2.412130637636081e-05,
232
+ "loss": 1.2681,
233
+ "step": 5000
234
+ },
235
+ {
236
+ "epoch": 8.0,
237
+ "eval_accuracy": 0.5383423702556158,
238
+ "eval_f1_macro": 0.2703391292878622,
239
+ "eval_f1_micro": 0.5383423702556158,
240
+ "eval_f1_weighted": 0.48876940771692845,
241
+ "eval_loss": 1.4461586475372314,
242
+ "eval_macro_fpr": 0.056020043825060734,
243
+ "eval_macro_sensitivity": 0.3381295733874737,
244
+ "eval_macro_specificity": 0.9655641856187219,
245
+ "eval_precision": 0.5269280210096271,
246
+ "eval_precision_macro": 0.2765942082966614,
247
+ "eval_recall": 0.5383423702556158,
248
+ "eval_recall_macro": 0.3381295733874737,
249
+ "eval_runtime": 64.8556,
250
+ "eval_samples_per_second": 19.906,
251
+ "eval_steps_per_second": 2.498,
252
+ "eval_weighted_fpr": 0.057718380786364515,
253
+ "eval_weighted_sensitivity": 0.5383423702556158,
254
+ "eval_weighted_specificity": 0.9451204140252133,
255
+ "step": 5144
256
+ },
257
+ {
258
+ "epoch": 8.55,
259
+ "grad_norm": 6.948422908782959,
260
+ "learning_rate": 2.152928978745464e-05,
261
+ "loss": 1.2351,
262
+ "step": 5500
263
+ },
264
+ {
265
+ "epoch": 9.0,
266
+ "eval_accuracy": 0.5515104570100697,
267
+ "eval_f1_macro": 0.29219268534520737,
268
+ "eval_f1_micro": 0.5515104570100697,
269
+ "eval_f1_weighted": 0.5183378322557094,
270
+ "eval_loss": 1.3802897930145264,
271
+ "eval_macro_fpr": 0.05395959885799773,
272
+ "eval_macro_sensitivity": 0.33994734275998406,
273
+ "eval_macro_specificity": 0.966222282286659,
274
+ "eval_precision": 0.5182882506526222,
275
+ "eval_precision_macro": 0.2800150936557356,
276
+ "eval_recall": 0.5515104570100697,
277
+ "eval_recall_macro": 0.33994734275998406,
278
+ "eval_runtime": 95.1023,
279
+ "eval_samples_per_second": 13.575,
280
+ "eval_steps_per_second": 1.703,
281
+ "eval_weighted_fpr": 0.05489712714515976,
282
+ "eval_weighted_sensitivity": 0.5515104570100697,
283
+ "eval_weighted_specificity": 0.9418237772898147,
284
+ "step": 5787
285
+ },
286
+ {
287
+ "epoch": 9.33,
288
+ "grad_norm": 17.84939956665039,
289
+ "learning_rate": 1.8937273198548472e-05,
290
+ "loss": 1.1161,
291
+ "step": 6000
292
+ },
293
+ {
294
+ "epoch": 10.0,
295
+ "eval_accuracy": 0.5747482571649883,
296
+ "eval_f1_macro": 0.3016038026372148,
297
+ "eval_f1_micro": 0.5747482571649883,
298
+ "eval_f1_weighted": 0.5350722531862059,
299
+ "eval_loss": 1.3405604362487793,
300
+ "eval_macro_fpr": 0.050184494231821634,
301
+ "eval_macro_sensitivity": 0.35749011781425877,
302
+ "eval_macro_specificity": 0.968005386818716,
303
+ "eval_precision": 0.5342322193126818,
304
+ "eval_precision_macro": 0.2899738892872161,
305
+ "eval_recall": 0.5747482571649883,
306
+ "eval_recall_macro": 0.35749011781425877,
307
+ "eval_runtime": 59.1957,
308
+ "eval_samples_per_second": 21.809,
309
+ "eval_steps_per_second": 2.737,
310
+ "eval_weighted_fpr": 0.05019658041510469,
311
+ "eval_weighted_sensitivity": 0.5747482571649883,
312
+ "eval_weighted_specificity": 0.945332545115751,
313
+ "step": 6430
314
+ },
315
+ {
316
+ "epoch": 10.11,
317
+ "grad_norm": 34.46049499511719,
318
+ "learning_rate": 1.6345256609642305e-05,
319
+ "loss": 1.0384,
320
+ "step": 6500
321
+ },
322
+ {
323
+ "epoch": 10.89,
324
+ "grad_norm": 18.093345642089844,
325
+ "learning_rate": 1.3753240020736132e-05,
326
+ "loss": 0.9747,
327
+ "step": 7000
328
+ },
329
+ {
330
+ "epoch": 11.0,
331
+ "eval_accuracy": 0.6150271107668474,
332
+ "eval_f1_macro": 0.3261856936782915,
333
+ "eval_f1_micro": 0.6150271107668474,
334
+ "eval_f1_weighted": 0.5766051994823087,
335
+ "eval_loss": 1.380952000617981,
336
+ "eval_macro_fpr": 0.0444013730647484,
337
+ "eval_macro_sensitivity": 0.3747372380690592,
338
+ "eval_macro_specificity": 0.9705808122777527,
339
+ "eval_precision": 0.5606186818865333,
340
+ "eval_precision_macro": 0.3083540755636765,
341
+ "eval_recall": 0.6150271107668474,
342
+ "eval_recall_macro": 0.3747372380690592,
343
+ "eval_runtime": 54.8103,
344
+ "eval_samples_per_second": 23.554,
345
+ "eval_steps_per_second": 2.956,
346
+ "eval_weighted_fpr": 0.04279686558167571,
347
+ "eval_weighted_sensitivity": 0.6150271107668474,
348
+ "eval_weighted_specificity": 0.9436850733994432,
349
+ "step": 7073
350
+ },
351
+ {
352
+ "epoch": 11.66,
353
+ "grad_norm": 18.54058265686035,
354
+ "learning_rate": 1.1161223431829965e-05,
355
+ "loss": 0.9145,
356
+ "step": 7500
357
+ },
358
+ {
359
+ "epoch": 12.0,
360
+ "eval_accuracy": 0.6196746707978311,
361
+ "eval_f1_macro": 0.33300079962435747,
362
+ "eval_f1_micro": 0.6196746707978311,
363
+ "eval_f1_weighted": 0.5913426830324416,
364
+ "eval_loss": 1.273578405380249,
365
+ "eval_macro_fpr": 0.04205592092867989,
366
+ "eval_macro_sensitivity": 0.38075990989198766,
367
+ "eval_macro_specificity": 0.9713642071716353,
368
+ "eval_precision": 0.5857440172055212,
369
+ "eval_precision_macro": 0.3198823734674615,
370
+ "eval_recall": 0.6196746707978311,
371
+ "eval_recall_macro": 0.38075990989198766,
372
+ "eval_runtime": 53.8432,
373
+ "eval_samples_per_second": 23.977,
374
+ "eval_steps_per_second": 3.009,
375
+ "eval_weighted_fpr": 0.041998118210589346,
376
+ "eval_weighted_sensitivity": 0.6196746707978311,
377
+ "eval_weighted_specificity": 0.9507884367766963,
378
+ "step": 7716
379
+ },
380
+ {
381
+ "epoch": 12.44,
382
+ "grad_norm": 2.550917863845825,
383
+ "learning_rate": 8.574390876101608e-06,
384
+ "loss": 0.8879,
385
+ "step": 8000
386
+ },
387
+ {
388
+ "epoch": 13.0,
389
+ "eval_accuracy": 0.635166537567777,
390
+ "eval_f1_macro": 0.3368012141860189,
391
+ "eval_f1_micro": 0.635166537567777,
392
+ "eval_f1_weighted": 0.6018918369596469,
393
+ "eval_loss": 1.3463146686553955,
394
+ "eval_macro_fpr": 0.039790684997017586,
395
+ "eval_macro_sensitivity": 0.3876584042144534,
396
+ "eval_macro_specificity": 0.972524656182703,
397
+ "eval_precision": 0.591491914014134,
398
+ "eval_precision_macro": 0.3194926993208432,
399
+ "eval_recall": 0.635166537567777,
400
+ "eval_recall_macro": 0.3876584042144534,
401
+ "eval_runtime": 53.0494,
402
+ "eval_samples_per_second": 24.336,
403
+ "eval_steps_per_second": 3.054,
404
+ "eval_weighted_fpr": 0.039410927955819595,
405
+ "eval_weighted_sensitivity": 0.635166537567777,
406
+ "eval_weighted_specificity": 0.9527033051727678,
407
+ "step": 8359
408
+ },
409
+ {
410
+ "epoch": 13.22,
411
+ "grad_norm": 20.480079650878906,
412
+ "learning_rate": 5.9823742871954386e-06,
413
+ "loss": 0.818,
414
+ "step": 8500
415
+ },
416
+ {
417
+ "epoch": 14.0,
418
+ "grad_norm": 23.086130142211914,
419
+ "learning_rate": 3.390357698289269e-06,
420
+ "loss": 0.8034,
421
+ "step": 9000
422
+ },
423
+ {
424
+ "epoch": 14.0,
425
+ "eval_accuracy": 0.6312935708752905,
426
+ "eval_f1_macro": 0.34083702910591834,
427
+ "eval_f1_micro": 0.6312935708752905,
428
+ "eval_f1_weighted": 0.6028724695599472,
429
+ "eval_loss": 1.3203202486038208,
430
+ "eval_macro_fpr": 0.04021000066286082,
431
+ "eval_macro_sensitivity": 0.3887264350085824,
432
+ "eval_macro_specificity": 0.9722365595844839,
433
+ "eval_precision": 0.5912742157720344,
434
+ "eval_precision_macro": 0.3223323142886684,
435
+ "eval_recall": 0.6312935708752905,
436
+ "eval_recall_macro": 0.3887264350085824,
437
+ "eval_runtime": 53.2083,
438
+ "eval_samples_per_second": 24.263,
439
+ "eval_steps_per_second": 3.045,
440
+ "eval_weighted_fpr": 0.04004711425206125,
441
+ "eval_weighted_sensitivity": 0.6312935708752905,
442
+ "eval_weighted_specificity": 0.9522548228919672,
443
+ "step": 9002
444
+ }
445
+ ],
446
+ "logging_steps": 500,
447
+ "max_steps": 9645,
448
+ "num_input_tokens_seen": 0,
449
+ "num_train_epochs": 15,
450
+ "save_steps": 500,
451
+ "total_flos": 1.892831352227021e+16,
452
+ "train_batch_size": 8,
453
+ "trial_name": null,
454
+ "trial_params": null
455
+ }
training_checkpoints/checkpoint-9002/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6da235c4e1e89b590bb6c4dd543edc36c67b04a23baad1858c6c64a331ed4bd1
3
+ size 4920