cite-text-analysis commited on
Commit
4d64f39
·
verified ·
1 Parent(s): 7f4d1ce

Upload folder using huggingface_hub

Browse files
training_checkpoints/checkpoint-6720/added_tokens.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ {
2
+ "<pad>": 28996
3
+ }
training_checkpoints/checkpoint-6720/config.json ADDED
@@ -0,0 +1,38 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "distilbert/distilbert-base-cased",
3
+ "activation": "gelu",
4
+ "architectures": [
5
+ "DistilBertForSequenceClassification"
6
+ ],
7
+ "attention_dropout": 0.1,
8
+ "dim": 768,
9
+ "dropout": 0.1,
10
+ "hidden_dim": 3072,
11
+ "id2label": {
12
+ "0": "positive",
13
+ "1": "negative",
14
+ "2": "neutral",
15
+ "3": "rel_by_party"
16
+ },
17
+ "initializer_range": 0.02,
18
+ "label2id": {
19
+ "negative": 1,
20
+ "neutral": 2,
21
+ "positive": 0,
22
+ "rel_by_party": 3
23
+ },
24
+ "max_position_embeddings": 512,
25
+ "model_type": "distilbert",
26
+ "n_heads": 12,
27
+ "n_layers": 6,
28
+ "output_past": true,
29
+ "pad_token_id": 28996,
30
+ "problem_type": "single_label_classification",
31
+ "qa_dropout": 0.1,
32
+ "seq_classif_dropout": 0.2,
33
+ "sinusoidal_pos_embds": false,
34
+ "tie_weights_": true,
35
+ "torch_dtype": "float32",
36
+ "transformers_version": "4.40.2",
37
+ "vocab_size": 28997
38
+ }
training_checkpoints/checkpoint-6720/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:68271a1ef827109598a8cb79efd8402f7f1d5698f6ed37c2e4090afe7a983fca
3
+ size 263153912
training_checkpoints/checkpoint-6720/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f84355e8f35dcc029aefa1cbd13c8ecbd4b1dadfdb2958e66e4b7a55fe8bea69
3
+ size 526369978
training_checkpoints/checkpoint-6720/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d23c69272f07789ac9c35351f04b8cc9a74ee6f210b27375437077d43ce4b926
3
+ size 14244
training_checkpoints/checkpoint-6720/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fea6b06751fd544073018c634bf5756f794d308adb01dc2a1f8783d100304f96
3
+ size 1064
training_checkpoints/checkpoint-6720/special_tokens_map.json ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "cls_token": "[CLS]",
3
+ "mask_token": "[MASK]",
4
+ "pad_token": {
5
+ "content": "<pad>",
6
+ "lstrip": false,
7
+ "normalized": false,
8
+ "rstrip": false,
9
+ "single_word": false
10
+ },
11
+ "sep_token": "[SEP]",
12
+ "unk_token": "[UNK]"
13
+ }
training_checkpoints/checkpoint-6720/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
training_checkpoints/checkpoint-6720/tokenizer_config.json ADDED
@@ -0,0 +1,63 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "[PAD]",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "100": {
12
+ "content": "[UNK]",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "101": {
20
+ "content": "[CLS]",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "102": {
28
+ "content": "[SEP]",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "103": {
36
+ "content": "[MASK]",
37
+ "lstrip": false,
38
+ "normalized": false,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ },
43
+ "28996": {
44
+ "content": "<pad>",
45
+ "lstrip": false,
46
+ "normalized": false,
47
+ "rstrip": false,
48
+ "single_word": false,
49
+ "special": true
50
+ }
51
+ },
52
+ "clean_up_tokenization_spaces": true,
53
+ "cls_token": "[CLS]",
54
+ "do_lower_case": false,
55
+ "mask_token": "[MASK]",
56
+ "model_max_length": 512,
57
+ "pad_token": "<pad>",
58
+ "sep_token": "[SEP]",
59
+ "strip_accents": null,
60
+ "tokenize_chinese_chars": true,
61
+ "tokenizer_class": "DistilBertTokenizer",
62
+ "unk_token": "[UNK]"
63
+ }
training_checkpoints/checkpoint-6720/trainer_state.json ADDED
@@ -0,0 +1,772 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.6478205036835734,
3
+ "best_model_checkpoint": "case-analysis-distilbert-base-cased/checkpoint-3584",
4
+ "epoch": 30.0,
5
+ "eval_steps": 500,
6
+ "global_step": 6720,
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_accuracy": 0.7661469933184856,
14
+ "eval_f1_macro": 0.5270352134866867,
15
+ "eval_f1_micro": 0.7661469933184856,
16
+ "eval_f1_weighted": 0.7333313004442764,
17
+ "eval_loss": 0.7000867128372192,
18
+ "eval_macro_fpr": 0.1330263157894737,
19
+ "eval_macro_sensitivity": 0.5137440725676019,
20
+ "eval_macro_specificity": 0.8818942592408516,
21
+ "eval_precision": 0.7310834070149471,
22
+ "eval_precision_macro": 0.5790641299401408,
23
+ "eval_recall": 0.7661469933184856,
24
+ "eval_recall_macro": 0.5137440725676019,
25
+ "eval_runtime": 5.4998,
26
+ "eval_samples_per_second": 81.639,
27
+ "eval_steps_per_second": 10.364,
28
+ "eval_weighted_fpr": 0.09234828496042216,
29
+ "eval_weighted_sensitivity": 0.7661469933184856,
30
+ "eval_weighted_specificity": 0.761430043644921,
31
+ "step": 224
32
+ },
33
+ {
34
+ "epoch": 2.0,
35
+ "eval_accuracy": 0.7750556792873051,
36
+ "eval_f1_macro": 0.5486718521990608,
37
+ "eval_f1_micro": 0.775055679287305,
38
+ "eval_f1_weighted": 0.7492575717498114,
39
+ "eval_loss": 0.7387592792510986,
40
+ "eval_macro_fpr": 0.12078200823200373,
41
+ "eval_macro_sensitivity": 0.5464266787796199,
42
+ "eval_macro_specificity": 0.8914743027952345,
43
+ "eval_precision": 0.7315284458297402,
44
+ "eval_precision_macro": 0.5585311316366537,
45
+ "eval_recall": 0.7750556792873051,
46
+ "eval_recall_macro": 0.5464266787796199,
47
+ "eval_runtime": 23.5227,
48
+ "eval_samples_per_second": 19.088,
49
+ "eval_steps_per_second": 2.423,
50
+ "eval_weighted_fpr": 0.08820960698689956,
51
+ "eval_weighted_sensitivity": 0.7750556792873051,
52
+ "eval_weighted_specificity": 0.7908415318936333,
53
+ "step": 448
54
+ },
55
+ {
56
+ "epoch": 2.232142857142857,
57
+ "grad_norm": 22.91168785095215,
58
+ "learning_rate": 4.629464285714286e-05,
59
+ "loss": 0.7066,
60
+ "step": 500
61
+ },
62
+ {
63
+ "epoch": 3.0,
64
+ "eval_accuracy": 0.8017817371937639,
65
+ "eval_f1_macro": 0.576677451756589,
66
+ "eval_f1_micro": 0.8017817371937639,
67
+ "eval_f1_weighted": 0.7759761601356532,
68
+ "eval_loss": 0.7228724956512451,
69
+ "eval_macro_fpr": 0.10757194512338594,
70
+ "eval_macro_sensitivity": 0.5707859310800487,
71
+ "eval_macro_specificity": 0.9026901603449184,
72
+ "eval_precision": 0.7604834726234316,
73
+ "eval_precision_macro": 0.5932281353135314,
74
+ "eval_recall": 0.8017817371937639,
75
+ "eval_recall_macro": 0.5707859310800487,
76
+ "eval_runtime": 17.5142,
77
+ "eval_samples_per_second": 25.636,
78
+ "eval_steps_per_second": 3.254,
79
+ "eval_weighted_fpr": 0.07613344739093242,
80
+ "eval_weighted_sensitivity": 0.8017817371937639,
81
+ "eval_weighted_specificity": 0.8089789041859096,
82
+ "step": 672
83
+ },
84
+ {
85
+ "epoch": 4.0,
86
+ "eval_accuracy": 0.8062360801781737,
87
+ "eval_f1_macro": 0.6301119928151375,
88
+ "eval_f1_micro": 0.8062360801781738,
89
+ "eval_f1_weighted": 0.7933630533945198,
90
+ "eval_loss": 0.8331468105316162,
91
+ "eval_macro_fpr": 0.10175133159747105,
92
+ "eval_macro_sensitivity": 0.6114882070764424,
93
+ "eval_macro_specificity": 0.9069973738423197,
94
+ "eval_precision": 0.7896315051078991,
95
+ "eval_precision_macro": 0.6674639062867935,
96
+ "eval_recall": 0.8062360801781737,
97
+ "eval_recall_macro": 0.6114882070764424,
98
+ "eval_runtime": 17.9901,
99
+ "eval_samples_per_second": 24.958,
100
+ "eval_steps_per_second": 3.168,
101
+ "eval_weighted_fpr": 0.0741687979539642,
102
+ "eval_weighted_sensitivity": 0.8062360801781737,
103
+ "eval_weighted_specificity": 0.8217534151911048,
104
+ "step": 896
105
+ },
106
+ {
107
+ "epoch": 4.464285714285714,
108
+ "grad_norm": 0.23617026209831238,
109
+ "learning_rate": 4.258184523809524e-05,
110
+ "loss": 0.3654,
111
+ "step": 1000
112
+ },
113
+ {
114
+ "epoch": 5.0,
115
+ "eval_accuracy": 0.7683741648106904,
116
+ "eval_f1_macro": 0.5896043825978781,
117
+ "eval_f1_micro": 0.7683741648106904,
118
+ "eval_f1_weighted": 0.7610828501277613,
119
+ "eval_loss": 1.2299734354019165,
120
+ "eval_macro_fpr": 0.10662840460883276,
121
+ "eval_macro_sensitivity": 0.6130556645262528,
122
+ "eval_macro_specificity": 0.9056447032260023,
123
+ "eval_precision": 0.7698582566254396,
124
+ "eval_precision_macro": 0.6085394130060126,
125
+ "eval_recall": 0.7683741648106904,
126
+ "eval_recall_macro": 0.6130556645262528,
127
+ "eval_runtime": 19.5517,
128
+ "eval_samples_per_second": 22.965,
129
+ "eval_steps_per_second": 2.915,
130
+ "eval_weighted_fpr": 0.0913081650570676,
131
+ "eval_weighted_sensitivity": 0.7683741648106904,
132
+ "eval_weighted_specificity": 0.8542046480933186,
133
+ "step": 1120
134
+ },
135
+ {
136
+ "epoch": 6.0,
137
+ "eval_accuracy": 0.8129175946547884,
138
+ "eval_f1_macro": 0.6299577384875399,
139
+ "eval_f1_micro": 0.8129175946547884,
140
+ "eval_f1_weighted": 0.7972266629935387,
141
+ "eval_loss": 1.0698440074920654,
142
+ "eval_macro_fpr": 0.09572659519735909,
143
+ "eval_macro_sensitivity": 0.6152769461592991,
144
+ "eval_macro_specificity": 0.9133810019905357,
145
+ "eval_precision": 0.7940368363876591,
146
+ "eval_precision_macro": 0.6863588955891834,
147
+ "eval_recall": 0.8129175946547884,
148
+ "eval_recall_macro": 0.6152769461592991,
149
+ "eval_runtime": 21.7346,
150
+ "eval_samples_per_second": 20.658,
151
+ "eval_steps_per_second": 2.623,
152
+ "eval_weighted_fpr": 0.07124681933842239,
153
+ "eval_weighted_sensitivity": 0.8129175946547884,
154
+ "eval_weighted_specificity": 0.8406064133073544,
155
+ "step": 1344
156
+ },
157
+ {
158
+ "epoch": 6.696428571428571,
159
+ "grad_norm": 0.6412457823753357,
160
+ "learning_rate": 3.8869047619047625e-05,
161
+ "loss": 0.2047,
162
+ "step": 1500
163
+ },
164
+ {
165
+ "epoch": 7.0,
166
+ "eval_accuracy": 0.7884187082405345,
167
+ "eval_f1_macro": 0.6141215790585401,
168
+ "eval_f1_micro": 0.7884187082405345,
169
+ "eval_f1_weighted": 0.7892035783447391,
170
+ "eval_loss": 1.3299829959869385,
171
+ "eval_macro_fpr": 0.10442766929340637,
172
+ "eval_macro_sensitivity": 0.5958936400112871,
173
+ "eval_macro_specificity": 0.9076274830150074,
174
+ "eval_precision": 0.7959953084611945,
175
+ "eval_precision_macro": 0.6412251711254097,
176
+ "eval_recall": 0.7884187082405345,
177
+ "eval_recall_macro": 0.5958936400112871,
178
+ "eval_runtime": 19.4013,
179
+ "eval_samples_per_second": 23.143,
180
+ "eval_steps_per_second": 2.938,
181
+ "eval_weighted_fpr": 0.08210890233362143,
182
+ "eval_weighted_sensitivity": 0.7884187082405345,
183
+ "eval_weighted_specificity": 0.842091223819495,
184
+ "step": 1568
185
+ },
186
+ {
187
+ "epoch": 8.0,
188
+ "eval_accuracy": 0.8106904231625836,
189
+ "eval_f1_macro": 0.6163443598585857,
190
+ "eval_f1_micro": 0.8106904231625834,
191
+ "eval_f1_weighted": 0.7947158089605988,
192
+ "eval_loss": 1.3869845867156982,
193
+ "eval_macro_fpr": 0.09827643474073441,
194
+ "eval_macro_sensitivity": 0.6062894445247387,
195
+ "eval_macro_specificity": 0.9106173671247473,
196
+ "eval_precision": 0.7860704056527595,
197
+ "eval_precision_macro": 0.6466819479006202,
198
+ "eval_recall": 0.8106904231625836,
199
+ "eval_recall_macro": 0.6062894445247387,
200
+ "eval_runtime": 24.9242,
201
+ "eval_samples_per_second": 18.015,
202
+ "eval_steps_per_second": 2.287,
203
+ "eval_weighted_fpr": 0.07221750212404418,
204
+ "eval_weighted_sensitivity": 0.8106904231625836,
205
+ "eval_weighted_specificity": 0.8317790453364052,
206
+ "step": 1792
207
+ },
208
+ {
209
+ "epoch": 8.928571428571429,
210
+ "grad_norm": 0.009310582652688026,
211
+ "learning_rate": 3.5156250000000004e-05,
212
+ "loss": 0.0795,
213
+ "step": 2000
214
+ },
215
+ {
216
+ "epoch": 9.0,
217
+ "eval_accuracy": 0.7951002227171492,
218
+ "eval_f1_macro": 0.6036078586352096,
219
+ "eval_f1_micro": 0.7951002227171492,
220
+ "eval_f1_weighted": 0.7803481654390761,
221
+ "eval_loss": 1.5031198263168335,
222
+ "eval_macro_fpr": 0.10395014224349126,
223
+ "eval_macro_sensitivity": 0.596858890976538,
224
+ "eval_macro_specificity": 0.9067643380561508,
225
+ "eval_precision": 0.7719081858839341,
226
+ "eval_precision_macro": 0.6275149499476224,
227
+ "eval_recall": 0.7951002227171492,
228
+ "eval_recall_macro": 0.596858890976538,
229
+ "eval_runtime": 19.5528,
230
+ "eval_samples_per_second": 22.964,
231
+ "eval_steps_per_second": 2.915,
232
+ "eval_weighted_fpr": 0.07910576096302666,
233
+ "eval_weighted_sensitivity": 0.7951002227171492,
234
+ "eval_weighted_specificity": 0.8319571295074538,
235
+ "step": 2016
236
+ },
237
+ {
238
+ "epoch": 10.0,
239
+ "eval_accuracy": 0.7728285077951003,
240
+ "eval_f1_macro": 0.6196001194983027,
241
+ "eval_f1_micro": 0.7728285077951002,
242
+ "eval_f1_weighted": 0.7758603070332255,
243
+ "eval_loss": 1.6303768157958984,
244
+ "eval_macro_fpr": 0.10604441860615665,
245
+ "eval_macro_sensitivity": 0.6233043585984762,
246
+ "eval_macro_specificity": 0.9072269655447398,
247
+ "eval_precision": 0.7796278998362599,
248
+ "eval_precision_macro": 0.6171087696246884,
249
+ "eval_recall": 0.7728285077951003,
250
+ "eval_recall_macro": 0.6233043585984762,
251
+ "eval_runtime": 17.8727,
252
+ "eval_samples_per_second": 25.122,
253
+ "eval_steps_per_second": 3.189,
254
+ "eval_weighted_fpr": 0.08923884514435695,
255
+ "eval_weighted_sensitivity": 0.7728285077951003,
256
+ "eval_weighted_specificity": 0.8560793543838588,
257
+ "step": 2240
258
+ },
259
+ {
260
+ "epoch": 11.0,
261
+ "eval_accuracy": 0.8040089086859689,
262
+ "eval_f1_macro": 0.60971035256129,
263
+ "eval_f1_micro": 0.8040089086859689,
264
+ "eval_f1_weighted": 0.7884103774950618,
265
+ "eval_loss": 1.6553305387496948,
266
+ "eval_macro_fpr": 0.10029712404712406,
267
+ "eval_macro_sensitivity": 0.6047305900247076,
268
+ "eval_macro_specificity": 0.9093226487010472,
269
+ "eval_precision": 0.7801791198576906,
270
+ "eval_precision_macro": 0.6404694360494954,
271
+ "eval_recall": 0.8040089086859689,
272
+ "eval_recall_macro": 0.6047305900247076,
273
+ "eval_runtime": 18.7706,
274
+ "eval_samples_per_second": 23.92,
275
+ "eval_steps_per_second": 3.037,
276
+ "eval_weighted_fpr": 0.07514944491887275,
277
+ "eval_weighted_sensitivity": 0.8040089086859689,
278
+ "eval_weighted_specificity": 0.83328168611822,
279
+ "step": 2464
280
+ },
281
+ {
282
+ "epoch": 11.160714285714286,
283
+ "grad_norm": 0.003156836610287428,
284
+ "learning_rate": 3.143601190476191e-05,
285
+ "loss": 0.0309,
286
+ "step": 2500
287
+ },
288
+ {
289
+ "epoch": 12.0,
290
+ "eval_accuracy": 0.799554565701559,
291
+ "eval_f1_macro": 0.6072980420599631,
292
+ "eval_f1_micro": 0.799554565701559,
293
+ "eval_f1_weighted": 0.7860582530348104,
294
+ "eval_loss": 1.6668155193328857,
295
+ "eval_macro_fpr": 0.09988889279594411,
296
+ "eval_macro_sensitivity": 0.6084453437394614,
297
+ "eval_macro_specificity": 0.9106688993495651,
298
+ "eval_precision": 0.777569146196374,
299
+ "eval_precision_macro": 0.6247320881812615,
300
+ "eval_recall": 0.799554565701559,
301
+ "eval_recall_macro": 0.6084453437394614,
302
+ "eval_runtime": 17.8921,
303
+ "eval_samples_per_second": 25.095,
304
+ "eval_steps_per_second": 3.186,
305
+ "eval_weighted_fpr": 0.07712082262210797,
306
+ "eval_weighted_sensitivity": 0.799554565701559,
307
+ "eval_weighted_specificity": 0.8431210316967014,
308
+ "step": 2688
309
+ },
310
+ {
311
+ "epoch": 13.0,
312
+ "eval_accuracy": 0.8040089086859689,
313
+ "eval_f1_macro": 0.5912164538573286,
314
+ "eval_f1_micro": 0.8040089086859689,
315
+ "eval_f1_weighted": 0.7845547674905504,
316
+ "eval_loss": 1.7547632455825806,
317
+ "eval_macro_fpr": 0.10303078503610634,
318
+ "eval_macro_sensitivity": 0.5846762537939009,
319
+ "eval_macro_specificity": 0.9063991203807322,
320
+ "eval_precision": 0.7724328584664579,
321
+ "eval_precision_macro": 0.6059236646315017,
322
+ "eval_recall": 0.8040089086859689,
323
+ "eval_recall_macro": 0.5846762537939009,
324
+ "eval_runtime": 18.8418,
325
+ "eval_samples_per_second": 23.83,
326
+ "eval_steps_per_second": 3.025,
327
+ "eval_weighted_fpr": 0.07514944491887275,
328
+ "eval_weighted_sensitivity": 0.8040089086859689,
329
+ "eval_weighted_specificity": 0.8215875728369607,
330
+ "step": 2912
331
+ },
332
+ {
333
+ "epoch": 13.392857142857142,
334
+ "grad_norm": 0.007099219132214785,
335
+ "learning_rate": 2.771577380952381e-05,
336
+ "loss": 0.0225,
337
+ "step": 3000
338
+ },
339
+ {
340
+ "epoch": 14.0,
341
+ "eval_accuracy": 0.8106904231625836,
342
+ "eval_f1_macro": 0.59981866562268,
343
+ "eval_f1_micro": 0.8106904231625834,
344
+ "eval_f1_weighted": 0.7909009860593438,
345
+ "eval_loss": 1.669065237045288,
346
+ "eval_macro_fpr": 0.09738592200890205,
347
+ "eval_macro_sensitivity": 0.6043589425942367,
348
+ "eval_macro_specificity": 0.9110786308220206,
349
+ "eval_precision": 0.7736492429702276,
350
+ "eval_precision_macro": 0.5965496612268181,
351
+ "eval_recall": 0.8106904231625836,
352
+ "eval_recall_macro": 0.6043589425942367,
353
+ "eval_runtime": 18.7407,
354
+ "eval_samples_per_second": 23.959,
355
+ "eval_steps_per_second": 3.042,
356
+ "eval_weighted_fpr": 0.07221750212404418,
357
+ "eval_weighted_sensitivity": 0.8106904231625836,
358
+ "eval_weighted_specificity": 0.8336241001254988,
359
+ "step": 3136
360
+ },
361
+ {
362
+ "epoch": 15.0,
363
+ "eval_accuracy": 0.8040089086859689,
364
+ "eval_f1_macro": 0.6250733662143111,
365
+ "eval_f1_micro": 0.8040089086859689,
366
+ "eval_f1_weighted": 0.7938662667450653,
367
+ "eval_loss": 1.8750598430633545,
368
+ "eval_macro_fpr": 0.10070785070785071,
369
+ "eval_macro_sensitivity": 0.6081468213821154,
370
+ "eval_macro_specificity": 0.9090627111057672,
371
+ "eval_precision": 0.7897150103102598,
372
+ "eval_precision_macro": 0.6516313389003527,
373
+ "eval_recall": 0.8040089086859689,
374
+ "eval_recall_macro": 0.6081468213821154,
375
+ "eval_runtime": 19.5585,
376
+ "eval_samples_per_second": 22.957,
377
+ "eval_steps_per_second": 2.914,
378
+ "eval_weighted_fpr": 0.07514944491887275,
379
+ "eval_weighted_sensitivity": 0.8040089086859689,
380
+ "eval_weighted_specificity": 0.8322419357371,
381
+ "step": 3360
382
+ },
383
+ {
384
+ "epoch": 15.625,
385
+ "grad_norm": 0.000986579223535955,
386
+ "learning_rate": 2.3995535714285717e-05,
387
+ "loss": 0.0048,
388
+ "step": 3500
389
+ },
390
+ {
391
+ "epoch": 16.0,
392
+ "eval_accuracy": 0.8084632516703786,
393
+ "eval_f1_macro": 0.6478205036835734,
394
+ "eval_f1_micro": 0.8084632516703786,
395
+ "eval_f1_weighted": 0.8017713075570686,
396
+ "eval_loss": 1.8402190208435059,
397
+ "eval_macro_fpr": 0.09346528346528347,
398
+ "eval_macro_sensitivity": 0.6429079002608414,
399
+ "eval_macro_specificity": 0.9158084751306171,
400
+ "eval_precision": 0.7982655513136719,
401
+ "eval_precision_macro": 0.6608466128926487,
402
+ "eval_recall": 0.8084632516703786,
403
+ "eval_recall_macro": 0.6429079002608414,
404
+ "eval_runtime": 20.8557,
405
+ "eval_samples_per_second": 21.529,
406
+ "eval_steps_per_second": 2.733,
407
+ "eval_weighted_fpr": 0.07319148936170213,
408
+ "eval_weighted_sensitivity": 0.8084632516703786,
409
+ "eval_weighted_specificity": 0.8547706488520901,
410
+ "step": 3584
411
+ },
412
+ {
413
+ "epoch": 17.0,
414
+ "eval_accuracy": 0.7951002227171492,
415
+ "eval_f1_macro": 0.6250080665059693,
416
+ "eval_f1_micro": 0.7951002227171492,
417
+ "eval_f1_weighted": 0.7890988211833133,
418
+ "eval_loss": 1.9123594760894775,
419
+ "eval_macro_fpr": 0.10009209120981923,
420
+ "eval_macro_sensitivity": 0.6237465502171384,
421
+ "eval_macro_specificity": 0.9101728441693318,
422
+ "eval_precision": 0.7871349497482736,
423
+ "eval_precision_macro": 0.6331039915966387,
424
+ "eval_recall": 0.7951002227171492,
425
+ "eval_recall_macro": 0.6237465502171384,
426
+ "eval_runtime": 22.3885,
427
+ "eval_samples_per_second": 20.055,
428
+ "eval_steps_per_second": 2.546,
429
+ "eval_weighted_fpr": 0.07910576096302666,
430
+ "eval_weighted_sensitivity": 0.7951002227171492,
431
+ "eval_weighted_specificity": 0.8455911539601776,
432
+ "step": 3808
433
+ },
434
+ {
435
+ "epoch": 17.857142857142858,
436
+ "grad_norm": 0.0016844779020175338,
437
+ "learning_rate": 2.027529761904762e-05,
438
+ "loss": 0.0069,
439
+ "step": 4000
440
+ },
441
+ {
442
+ "epoch": 18.0,
443
+ "eval_accuracy": 0.7973273942093542,
444
+ "eval_f1_macro": 0.6062346830120622,
445
+ "eval_f1_micro": 0.7973273942093542,
446
+ "eval_f1_weighted": 0.7846722241151323,
447
+ "eval_loss": 1.8856843709945679,
448
+ "eval_macro_fpr": 0.1047864019411268,
449
+ "eval_macro_sensitivity": 0.5972288178170531,
450
+ "eval_macro_specificity": 0.9053198759071074,
451
+ "eval_precision": 0.7793901437600238,
452
+ "eval_precision_macro": 0.6267606864492677,
453
+ "eval_recall": 0.7973273942093542,
454
+ "eval_recall_macro": 0.5972288178170531,
455
+ "eval_runtime": 21.6525,
456
+ "eval_samples_per_second": 20.737,
457
+ "eval_steps_per_second": 2.632,
458
+ "eval_weighted_fpr": 0.07811158798283262,
459
+ "eval_weighted_sensitivity": 0.7973273942093542,
460
+ "eval_weighted_specificity": 0.8239521094190755,
461
+ "step": 4032
462
+ },
463
+ {
464
+ "epoch": 19.0,
465
+ "eval_accuracy": 0.8062360801781737,
466
+ "eval_f1_macro": 0.610672303898802,
467
+ "eval_f1_micro": 0.8062360801781738,
468
+ "eval_f1_weighted": 0.7895499021324558,
469
+ "eval_loss": 1.9492371082305908,
470
+ "eval_macro_fpr": 0.10059121982105523,
471
+ "eval_macro_sensitivity": 0.6015363147716088,
472
+ "eval_macro_specificity": 0.9085815747160192,
473
+ "eval_precision": 0.7813193080487297,
474
+ "eval_precision_macro": 0.6466954430951133,
475
+ "eval_recall": 0.8062360801781737,
476
+ "eval_recall_macro": 0.6015363147716088,
477
+ "eval_runtime": 18.7319,
478
+ "eval_samples_per_second": 23.97,
479
+ "eval_steps_per_second": 3.043,
480
+ "eval_weighted_fpr": 0.0741687979539642,
481
+ "eval_weighted_sensitivity": 0.8062360801781737,
482
+ "eval_weighted_specificity": 0.828090218685903,
483
+ "step": 4256
484
+ },
485
+ {
486
+ "epoch": 20.0,
487
+ "eval_accuracy": 0.8084632516703786,
488
+ "eval_f1_macro": 0.6144369196694778,
489
+ "eval_f1_micro": 0.8084632516703786,
490
+ "eval_f1_weighted": 0.7932213847166972,
491
+ "eval_loss": 1.899397850036621,
492
+ "eval_macro_fpr": 0.09880130869572286,
493
+ "eval_macro_sensitivity": 0.6066610919552096,
494
+ "eval_macro_specificity": 0.9101677843526085,
495
+ "eval_precision": 0.784920953393948,
496
+ "eval_precision_macro": 0.6417378517103144,
497
+ "eval_recall": 0.8084632516703786,
498
+ "eval_recall_macro": 0.6066610919552096,
499
+ "eval_runtime": 18.8499,
500
+ "eval_samples_per_second": 23.82,
501
+ "eval_steps_per_second": 3.024,
502
+ "eval_weighted_fpr": 0.07319148936170213,
503
+ "eval_weighted_sensitivity": 0.8084632516703786,
504
+ "eval_weighted_specificity": 0.8322078857400554,
505
+ "step": 4480
506
+ },
507
+ {
508
+ "epoch": 20.089285714285715,
509
+ "grad_norm": 0.0006921566091477871,
510
+ "learning_rate": 1.65625e-05,
511
+ "loss": 0.0034,
512
+ "step": 4500
513
+ },
514
+ {
515
+ "epoch": 21.0,
516
+ "eval_accuracy": 0.8040089086859689,
517
+ "eval_f1_macro": 0.6428743743483731,
518
+ "eval_f1_micro": 0.8040089086859689,
519
+ "eval_f1_weighted": 0.794199107090719,
520
+ "eval_loss": 1.981939435005188,
521
+ "eval_macro_fpr": 0.09763565524435089,
522
+ "eval_macro_sensitivity": 0.632509514862456,
523
+ "eval_macro_specificity": 0.9119739242182288,
524
+ "eval_precision": 0.7898236252959594,
525
+ "eval_precision_macro": 0.6747722186700768,
526
+ "eval_recall": 0.8040089086859689,
527
+ "eval_recall_macro": 0.632509514862456,
528
+ "eval_runtime": 18.6464,
529
+ "eval_samples_per_second": 24.08,
530
+ "eval_steps_per_second": 3.057,
531
+ "eval_weighted_fpr": 0.07514944491887275,
532
+ "eval_weighted_sensitivity": 0.8040089086859689,
533
+ "eval_weighted_specificity": 0.8438867881869458,
534
+ "step": 4704
535
+ },
536
+ {
537
+ "epoch": 22.0,
538
+ "eval_accuracy": 0.8062360801781737,
539
+ "eval_f1_macro": 0.6293220930410561,
540
+ "eval_f1_micro": 0.8062360801781738,
541
+ "eval_f1_weighted": 0.7929649161864832,
542
+ "eval_loss": 2.0180749893188477,
543
+ "eval_macro_fpr": 0.09769081016323067,
544
+ "eval_macro_sensitivity": 0.6204025836378777,
545
+ "eval_macro_specificity": 0.9117613753511575,
546
+ "eval_precision": 0.7879578221269433,
547
+ "eval_precision_macro": 0.6735832268090333,
548
+ "eval_recall": 0.8062360801781737,
549
+ "eval_recall_macro": 0.6204025836378777,
550
+ "eval_runtime": 21.6029,
551
+ "eval_samples_per_second": 20.784,
552
+ "eval_steps_per_second": 2.639,
553
+ "eval_weighted_fpr": 0.0741687979539642,
554
+ "eval_weighted_sensitivity": 0.8062360801781737,
555
+ "eval_weighted_specificity": 0.8408094212264563,
556
+ "step": 4928
557
+ },
558
+ {
559
+ "epoch": 22.321428571428573,
560
+ "grad_norm": 0.0017632667440921068,
561
+ "learning_rate": 1.2842261904761904e-05,
562
+ "loss": 0.0001,
563
+ "step": 5000
564
+ },
565
+ {
566
+ "epoch": 23.0,
567
+ "eval_accuracy": 0.8062360801781737,
568
+ "eval_f1_macro": 0.6293220930410561,
569
+ "eval_f1_micro": 0.8062360801781738,
570
+ "eval_f1_weighted": 0.7929649161864832,
571
+ "eval_loss": 2.0305473804473877,
572
+ "eval_macro_fpr": 0.09769081016323067,
573
+ "eval_macro_sensitivity": 0.6204025836378777,
574
+ "eval_macro_specificity": 0.9117613753511575,
575
+ "eval_precision": 0.7879578221269433,
576
+ "eval_precision_macro": 0.6735832268090333,
577
+ "eval_recall": 0.8062360801781737,
578
+ "eval_recall_macro": 0.6204025836378777,
579
+ "eval_runtime": 19.9323,
580
+ "eval_samples_per_second": 22.526,
581
+ "eval_steps_per_second": 2.86,
582
+ "eval_weighted_fpr": 0.0741687979539642,
583
+ "eval_weighted_sensitivity": 0.8062360801781737,
584
+ "eval_weighted_specificity": 0.8408094212264563,
585
+ "step": 5152
586
+ },
587
+ {
588
+ "epoch": 24.0,
589
+ "eval_accuracy": 0.8040089086859689,
590
+ "eval_f1_macro": 0.6091984969950299,
591
+ "eval_f1_micro": 0.8040089086859689,
592
+ "eval_f1_weighted": 0.7876978416949642,
593
+ "eval_loss": 2.0248725414276123,
594
+ "eval_macro_fpr": 0.10186804925610897,
595
+ "eval_macro_sensitivity": 0.6004222327751739,
596
+ "eval_macro_specificity": 0.9074012244038161,
597
+ "eval_precision": 0.7801376277130834,
598
+ "eval_precision_macro": 0.6448022472746214,
599
+ "eval_recall": 0.8040089086859689,
600
+ "eval_recall_macro": 0.6004222327751739,
601
+ "eval_runtime": 22.5428,
602
+ "eval_samples_per_second": 19.918,
603
+ "eval_steps_per_second": 2.529,
604
+ "eval_weighted_fpr": 0.07514944491887275,
605
+ "eval_weighted_sensitivity": 0.8040089086859689,
606
+ "eval_weighted_specificity": 0.8255959889292954,
607
+ "step": 5376
608
+ },
609
+ {
610
+ "epoch": 24.553571428571427,
611
+ "grad_norm": 0.00044994684867560863,
612
+ "learning_rate": 9.122023809523809e-06,
613
+ "loss": 0.0,
614
+ "step": 5500
615
+ },
616
+ {
617
+ "epoch": 25.0,
618
+ "eval_accuracy": 0.8017817371937639,
619
+ "eval_f1_macro": 0.6271799288322751,
620
+ "eval_f1_micro": 0.8017817371937639,
621
+ "eval_f1_weighted": 0.7907670941735037,
622
+ "eval_loss": 2.0139310359954834,
623
+ "eval_macro_fpr": 0.09838717278837039,
624
+ "eval_macro_sensitivity": 0.6226316079257256,
625
+ "eval_macro_specificity": 0.9113889022846089,
626
+ "eval_precision": 0.7848002776733288,
627
+ "eval_precision_macro": 0.6513528138528139,
628
+ "eval_recall": 0.8017817371937639,
629
+ "eval_recall_macro": 0.6226316079257256,
630
+ "eval_runtime": 17.4233,
631
+ "eval_samples_per_second": 25.77,
632
+ "eval_steps_per_second": 3.271,
633
+ "eval_weighted_fpr": 0.07613344739093242,
634
+ "eval_weighted_sensitivity": 0.8017817371937639,
635
+ "eval_weighted_specificity": 0.8437738719446717,
636
+ "step": 5600
637
+ },
638
+ {
639
+ "epoch": 26.0,
640
+ "eval_accuracy": 0.8040089086859689,
641
+ "eval_f1_macro": 0.6304811840155711,
642
+ "eval_f1_micro": 0.8040089086859689,
643
+ "eval_f1_weighted": 0.7925820064518235,
644
+ "eval_loss": 2.0075461864471436,
645
+ "eval_macro_fpr": 0.09613422453907529,
646
+ "eval_macro_sensitivity": 0.6280540471716942,
647
+ "eval_macro_specificity": 0.9131650391354523,
648
+ "eval_precision": 0.7867899740794329,
649
+ "eval_precision_macro": 0.6586347907326314,
650
+ "eval_recall": 0.8040089086859689,
651
+ "eval_recall_macro": 0.6280540471716942,
652
+ "eval_runtime": 19.888,
653
+ "eval_samples_per_second": 22.576,
654
+ "eval_steps_per_second": 2.866,
655
+ "eval_weighted_fpr": 0.07514944491887275,
656
+ "eval_weighted_sensitivity": 0.8040089086859689,
657
+ "eval_weighted_specificity": 0.8486512478558408,
658
+ "step": 5824
659
+ },
660
+ {
661
+ "epoch": 26.785714285714285,
662
+ "grad_norm": 0.00030680475174449384,
663
+ "learning_rate": 5.4017857142857145e-06,
664
+ "loss": 0.0026,
665
+ "step": 6000
666
+ },
667
+ {
668
+ "epoch": 27.0,
669
+ "eval_accuracy": 0.8040089086859689,
670
+ "eval_f1_macro": 0.6304811840155711,
671
+ "eval_f1_micro": 0.8040089086859689,
672
+ "eval_f1_weighted": 0.7925820064518235,
673
+ "eval_loss": 2.015469551086426,
674
+ "eval_macro_fpr": 0.09613422453907529,
675
+ "eval_macro_sensitivity": 0.6280540471716942,
676
+ "eval_macro_specificity": 0.9131650391354523,
677
+ "eval_precision": 0.7867899740794329,
678
+ "eval_precision_macro": 0.6586347907326314,
679
+ "eval_recall": 0.8040089086859689,
680
+ "eval_recall_macro": 0.6280540471716942,
681
+ "eval_runtime": 18.2129,
682
+ "eval_samples_per_second": 24.653,
683
+ "eval_steps_per_second": 3.13,
684
+ "eval_weighted_fpr": 0.07514944491887275,
685
+ "eval_weighted_sensitivity": 0.8040089086859689,
686
+ "eval_weighted_specificity": 0.8486512478558408,
687
+ "step": 6048
688
+ },
689
+ {
690
+ "epoch": 28.0,
691
+ "eval_accuracy": 0.8040089086859689,
692
+ "eval_f1_macro": 0.6282967821034426,
693
+ "eval_f1_micro": 0.8040089086859689,
694
+ "eval_f1_weighted": 0.7922964886726178,
695
+ "eval_loss": 2.019129753112793,
696
+ "eval_macro_fpr": 0.09701836451836453,
697
+ "eval_macro_sensitivity": 0.6237456899221605,
698
+ "eval_macro_specificity": 0.9125697107568691,
699
+ "eval_precision": 0.7864940838168837,
700
+ "eval_precision_macro": 0.6586431481702824,
701
+ "eval_recall": 0.8040089086859689,
702
+ "eval_recall_macro": 0.6237456899221605,
703
+ "eval_runtime": 20.937,
704
+ "eval_samples_per_second": 21.445,
705
+ "eval_steps_per_second": 2.722,
706
+ "eval_weighted_fpr": 0.07514944491887275,
707
+ "eval_weighted_sensitivity": 0.8040089086859689,
708
+ "eval_weighted_specificity": 0.8462699343415074,
709
+ "step": 6272
710
+ },
711
+ {
712
+ "epoch": 29.0,
713
+ "eval_accuracy": 0.8040089086859689,
714
+ "eval_f1_macro": 0.6282967821034426,
715
+ "eval_f1_micro": 0.8040089086859689,
716
+ "eval_f1_weighted": 0.7922964886726178,
717
+ "eval_loss": 2.0224666595458984,
718
+ "eval_macro_fpr": 0.09701836451836453,
719
+ "eval_macro_sensitivity": 0.6237456899221605,
720
+ "eval_macro_specificity": 0.9125697107568691,
721
+ "eval_precision": 0.7864940838168837,
722
+ "eval_precision_macro": 0.6586431481702824,
723
+ "eval_recall": 0.8040089086859689,
724
+ "eval_recall_macro": 0.6237456899221605,
725
+ "eval_runtime": 17.6785,
726
+ "eval_samples_per_second": 25.398,
727
+ "eval_steps_per_second": 3.224,
728
+ "eval_weighted_fpr": 0.07514944491887275,
729
+ "eval_weighted_sensitivity": 0.8040089086859689,
730
+ "eval_weighted_specificity": 0.8462699343415074,
731
+ "step": 6496
732
+ },
733
+ {
734
+ "epoch": 29.017857142857142,
735
+ "grad_norm": 0.00021328784350771457,
736
+ "learning_rate": 1.681547619047619e-06,
737
+ "loss": 0.0,
738
+ "step": 6500
739
+ },
740
+ {
741
+ "epoch": 30.0,
742
+ "eval_accuracy": 0.799554565701559,
743
+ "eval_f1_macro": 0.6244721903798025,
744
+ "eval_f1_micro": 0.799554565701559,
745
+ "eval_f1_weighted": 0.7887095002776057,
746
+ "eval_loss": 2.034322500228882,
747
+ "eval_macro_fpr": 0.09842820376592175,
748
+ "eval_macro_sensitivity": 0.6218151879916586,
749
+ "eval_macro_specificity": 0.9118590979466747,
750
+ "eval_precision": 0.782121296142182,
751
+ "eval_precision_macro": 0.6460922554232,
752
+ "eval_recall": 0.799554565701559,
753
+ "eval_recall_macro": 0.6218151879916586,
754
+ "eval_runtime": 19.6458,
755
+ "eval_samples_per_second": 22.855,
756
+ "eval_steps_per_second": 2.901,
757
+ "eval_weighted_fpr": 0.07712082262210797,
758
+ "eval_weighted_sensitivity": 0.799554565701559,
759
+ "eval_weighted_specificity": 0.8478818260851398,
760
+ "step": 6720
761
+ }
762
+ ],
763
+ "logging_steps": 500,
764
+ "max_steps": 6720,
765
+ "num_input_tokens_seen": 0,
766
+ "num_train_epochs": 30,
767
+ "save_steps": 500,
768
+ "total_flos": 7121701353553920.0,
769
+ "train_batch_size": 8,
770
+ "trial_name": null,
771
+ "trial_params": null
772
+ }
training_checkpoints/checkpoint-6720/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3168244efe16c0aeafe631cea29ca8c9c2bbb665a169c0d40c2e4d22eaca40d9
3
+ size 5048
training_checkpoints/checkpoint-6720/vocab.txt ADDED
The diff for this file is too large to render. See raw diff