ChrisZeng commited on
Commit
1b4e2e8
1 Parent(s): e74679b

Training in progress, epoch 16

Browse files
checkpoint-3570/config.json ADDED
@@ -0,0 +1,41 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "ynie/electra-large-discriminator-snli_mnli_fever_anli_R1_R2_R3-nli",
3
+ "architectures": [
4
+ "ElectraForSequenceClassification"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.1,
7
+ "classifier_dropout": null,
8
+ "embedding_size": 1024,
9
+ "hidden_act": "gelu",
10
+ "hidden_dropout_prob": 0.1,
11
+ "hidden_size": 1024,
12
+ "id2label": {
13
+ "0": "entailment",
14
+ "1": "neutral",
15
+ "2": "contradiction"
16
+ },
17
+ "initializer_range": 0.02,
18
+ "intermediate_size": 4096,
19
+ "label2id": {
20
+ "contradiction": 2,
21
+ "entailment": 0,
22
+ "neutral": 1
23
+ },
24
+ "layer_norm_eps": 1e-12,
25
+ "max_position_embeddings": 512,
26
+ "model_type": "electra",
27
+ "num_attention_heads": 16,
28
+ "num_hidden_layers": 24,
29
+ "pad_token_id": 0,
30
+ "position_embedding_type": "absolute",
31
+ "problem_type": "multi_label_classification",
32
+ "summary_activation": "gelu",
33
+ "summary_last_dropout": 0.1,
34
+ "summary_type": "first",
35
+ "summary_use_proj": true,
36
+ "torch_dtype": "float32",
37
+ "transformers_version": "4.17.0",
38
+ "type_vocab_size": 2,
39
+ "use_cache": true,
40
+ "vocab_size": 30522
41
+ }
checkpoint-3570/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9dd7fb035b4e7c7421ac03f7ad269e3c1a1e69003f1810bf4aa580fd2a553c94
3
+ size 2681386461
checkpoint-3570/pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5165e9058f7e1bfb43a04739d7edcf26e7ca672cf3054aad242610adbb516841
3
+ size 1340745837
checkpoint-3570/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:22b1bab5c4b5f18060e247e863f6bd3be2f91054fbf430131a8522169e675da5
3
+ size 14503
checkpoint-3570/scaler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c5dc14e956b9498009a2afbe211f1386376196aff8b4f3dac9422e72a8f6bf6f
3
+ size 559
checkpoint-3570/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7753f8d09d981e31f48bf0ba861ecb5a5eda5f00f3dc414ba16338a8742491a5
3
+ size 623
checkpoint-3570/special_tokens_map.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]"}
checkpoint-3570/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-3570/tokenizer_config.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"do_lower_case": true, "unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]", "tokenize_chinese_chars": true, "strip_accents": null, "normalization": true, "model_max_length": 512, "special_tokens_map_file": null, "name_or_path": "google/electra-large-discriminator", "tokenizer_class": "ElectraTokenizer"}
checkpoint-3570/trainer_state.json ADDED
@@ -0,0 +1,288 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.3247086703777313,
3
+ "best_model_checkpoint": "outputs/electra-nli-efl/checkpoint-1050",
4
+ "epoch": 16.995555555555555,
5
+ "global_step": 3570,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 1.0,
12
+ "learning_rate": 9.499999999999999e-07,
13
+ "loss": 0.4175,
14
+ "step": 210
15
+ },
16
+ {
17
+ "epoch": 1.0,
18
+ "eval_accuracy": 0.7316666666666667,
19
+ "eval_f1": 0.7305281480923569,
20
+ "eval_loss": 0.4019607603549957,
21
+ "eval_runtime": 11.5986,
22
+ "eval_samples_per_second": 258.651,
23
+ "eval_steps_per_second": 32.331,
24
+ "step": 210
25
+ },
26
+ {
27
+ "epoch": 2.0,
28
+ "learning_rate": 9e-07,
29
+ "loss": 0.3061,
30
+ "step": 420
31
+ },
32
+ {
33
+ "epoch": 2.0,
34
+ "eval_accuracy": 0.768,
35
+ "eval_f1": 0.7674936528439713,
36
+ "eval_loss": 0.3519645631313324,
37
+ "eval_runtime": 11.6033,
38
+ "eval_samples_per_second": 258.548,
39
+ "eval_steps_per_second": 32.319,
40
+ "step": 420
41
+ },
42
+ {
43
+ "epoch": 3.0,
44
+ "learning_rate": 8.499999999999999e-07,
45
+ "loss": 0.2588,
46
+ "step": 630
47
+ },
48
+ {
49
+ "epoch": 3.0,
50
+ "eval_accuracy": 0.79,
51
+ "eval_f1": 0.7887799932409598,
52
+ "eval_loss": 0.32531991600990295,
53
+ "eval_runtime": 11.5951,
54
+ "eval_samples_per_second": 258.729,
55
+ "eval_steps_per_second": 32.341,
56
+ "step": 630
57
+ },
58
+ {
59
+ "epoch": 4.0,
60
+ "learning_rate": 8e-07,
61
+ "loss": 0.234,
62
+ "step": 840
63
+ },
64
+ {
65
+ "epoch": 4.0,
66
+ "eval_accuracy": 0.788,
67
+ "eval_f1": 0.7877041067016072,
68
+ "eval_loss": 0.33730262517929077,
69
+ "eval_runtime": 11.6935,
70
+ "eval_samples_per_second": 256.553,
71
+ "eval_steps_per_second": 32.069,
72
+ "step": 840
73
+ },
74
+ {
75
+ "epoch": 5.0,
76
+ "learning_rate": 7.5e-07,
77
+ "loss": 0.2116,
78
+ "step": 1050
79
+ },
80
+ {
81
+ "epoch": 5.0,
82
+ "eval_accuracy": 0.804,
83
+ "eval_f1": 0.8032759681298146,
84
+ "eval_loss": 0.3247086703777313,
85
+ "eval_runtime": 11.5938,
86
+ "eval_samples_per_second": 258.758,
87
+ "eval_steps_per_second": 32.345,
88
+ "step": 1050
89
+ },
90
+ {
91
+ "epoch": 6.0,
92
+ "learning_rate": 7e-07,
93
+ "loss": 0.1974,
94
+ "step": 1260
95
+ },
96
+ {
97
+ "epoch": 6.0,
98
+ "eval_accuracy": 0.793,
99
+ "eval_f1": 0.7928414316247181,
100
+ "eval_loss": 0.34000951051712036,
101
+ "eval_runtime": 11.6038,
102
+ "eval_samples_per_second": 258.537,
103
+ "eval_steps_per_second": 32.317,
104
+ "step": 1260
105
+ },
106
+ {
107
+ "epoch": 7.0,
108
+ "learning_rate": 6.5e-07,
109
+ "loss": 0.1807,
110
+ "step": 1470
111
+ },
112
+ {
113
+ "epoch": 7.0,
114
+ "eval_accuracy": 0.7973333333333333,
115
+ "eval_f1": 0.7969281824938732,
116
+ "eval_loss": 0.35112178325653076,
117
+ "eval_runtime": 11.589,
118
+ "eval_samples_per_second": 258.866,
119
+ "eval_steps_per_second": 32.358,
120
+ "step": 1470
121
+ },
122
+ {
123
+ "epoch": 8.0,
124
+ "learning_rate": 6e-07,
125
+ "loss": 0.1715,
126
+ "step": 1680
127
+ },
128
+ {
129
+ "epoch": 8.0,
130
+ "eval_accuracy": 0.7993333333333333,
131
+ "eval_f1": 0.79885693716362,
132
+ "eval_loss": 0.34963640570640564,
133
+ "eval_runtime": 11.6019,
134
+ "eval_samples_per_second": 258.579,
135
+ "eval_steps_per_second": 32.322,
136
+ "step": 1680
137
+ },
138
+ {
139
+ "epoch": 9.0,
140
+ "learning_rate": 5.5e-07,
141
+ "loss": 0.1577,
142
+ "step": 1890
143
+ },
144
+ {
145
+ "epoch": 9.0,
146
+ "eval_accuracy": 0.8043333333333333,
147
+ "eval_f1": 0.803166306363764,
148
+ "eval_loss": 0.3507426679134369,
149
+ "eval_runtime": 11.6051,
150
+ "eval_samples_per_second": 258.507,
151
+ "eval_steps_per_second": 32.313,
152
+ "step": 1890
153
+ },
154
+ {
155
+ "epoch": 10.0,
156
+ "learning_rate": 5e-07,
157
+ "loss": 0.1469,
158
+ "step": 2100
159
+ },
160
+ {
161
+ "epoch": 10.0,
162
+ "eval_accuracy": 0.798,
163
+ "eval_f1": 0.7970242731280681,
164
+ "eval_loss": 0.3604212701320648,
165
+ "eval_runtime": 11.7325,
166
+ "eval_samples_per_second": 255.699,
167
+ "eval_steps_per_second": 31.962,
168
+ "step": 2100
169
+ },
170
+ {
171
+ "epoch": 11.0,
172
+ "learning_rate": 4.5e-07,
173
+ "loss": 0.1394,
174
+ "step": 2310
175
+ },
176
+ {
177
+ "epoch": 11.0,
178
+ "eval_accuracy": 0.7966666666666666,
179
+ "eval_f1": 0.7957033854293423,
180
+ "eval_loss": 0.37343233823776245,
181
+ "eval_runtime": 11.6069,
182
+ "eval_samples_per_second": 258.468,
183
+ "eval_steps_per_second": 32.308,
184
+ "step": 2310
185
+ },
186
+ {
187
+ "epoch": 12.0,
188
+ "learning_rate": 4e-07,
189
+ "loss": 0.1322,
190
+ "step": 2520
191
+ },
192
+ {
193
+ "epoch": 12.0,
194
+ "eval_accuracy": 0.7913333333333333,
195
+ "eval_f1": 0.7905794192426068,
196
+ "eval_loss": 0.39285749197006226,
197
+ "eval_runtime": 11.6585,
198
+ "eval_samples_per_second": 257.322,
199
+ "eval_steps_per_second": 32.165,
200
+ "step": 2520
201
+ },
202
+ {
203
+ "epoch": 13.0,
204
+ "learning_rate": 3.5e-07,
205
+ "loss": 0.1231,
206
+ "step": 2730
207
+ },
208
+ {
209
+ "epoch": 13.0,
210
+ "eval_accuracy": 0.795,
211
+ "eval_f1": 0.7941480244179493,
212
+ "eval_loss": 0.39544808864593506,
213
+ "eval_runtime": 11.5711,
214
+ "eval_samples_per_second": 259.267,
215
+ "eval_steps_per_second": 32.408,
216
+ "step": 2730
217
+ },
218
+ {
219
+ "epoch": 14.0,
220
+ "learning_rate": 3e-07,
221
+ "loss": 0.1189,
222
+ "step": 2940
223
+ },
224
+ {
225
+ "epoch": 14.0,
226
+ "eval_accuracy": 0.7976666666666666,
227
+ "eval_f1": 0.7963303903578043,
228
+ "eval_loss": 0.39935171604156494,
229
+ "eval_runtime": 11.7059,
230
+ "eval_samples_per_second": 256.282,
231
+ "eval_steps_per_second": 32.035,
232
+ "step": 2940
233
+ },
234
+ {
235
+ "epoch": 15.0,
236
+ "learning_rate": 2.5e-07,
237
+ "loss": 0.1143,
238
+ "step": 3150
239
+ },
240
+ {
241
+ "epoch": 15.0,
242
+ "eval_accuracy": 0.7993333333333333,
243
+ "eval_f1": 0.798040794417606,
244
+ "eval_loss": 0.39949560165405273,
245
+ "eval_runtime": 11.6007,
246
+ "eval_samples_per_second": 258.605,
247
+ "eval_steps_per_second": 32.326,
248
+ "step": 3150
249
+ },
250
+ {
251
+ "epoch": 16.0,
252
+ "learning_rate": 2e-07,
253
+ "loss": 0.1083,
254
+ "step": 3360
255
+ },
256
+ {
257
+ "epoch": 16.0,
258
+ "eval_accuracy": 0.7926666666666666,
259
+ "eval_f1": 0.7918316804069658,
260
+ "eval_loss": 0.41252416372299194,
261
+ "eval_runtime": 11.6447,
262
+ "eval_samples_per_second": 257.627,
263
+ "eval_steps_per_second": 32.203,
264
+ "step": 3360
265
+ },
266
+ {
267
+ "epoch": 17.0,
268
+ "learning_rate": 1.5e-07,
269
+ "loss": 0.1079,
270
+ "step": 3570
271
+ },
272
+ {
273
+ "epoch": 17.0,
274
+ "eval_accuracy": 0.7993333333333333,
275
+ "eval_f1": 0.79793006993007,
276
+ "eval_loss": 0.4035763442516327,
277
+ "eval_runtime": 11.5923,
278
+ "eval_samples_per_second": 258.793,
279
+ "eval_steps_per_second": 32.349,
280
+ "step": 3570
281
+ }
282
+ ],
283
+ "max_steps": 4200,
284
+ "num_train_epochs": 20,
285
+ "total_flos": 7.071798658380902e+16,
286
+ "trial_name": null,
287
+ "trial_params": null
288
+ }
checkpoint-3570/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c104103a95a8e00f67dc451eba8bec78c0cbf934c84f4059200155cf9a45b752
3
+ size 3119
checkpoint-3570/vocab.txt ADDED
The diff for this file is too large to render. See raw diff
 
pytorch_model.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:29e2db159a37348d307c7da158c095941848e612e864942bdd8c87653f17b2f2
3
  size 1340745837
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5165e9058f7e1bfb43a04739d7edcf26e7ca672cf3054aad242610adbb516841
3
  size 1340745837