mjtwins commited on
Commit
95821cd
1 Parent(s): e4c7d17

commit from mujeen

Browse files
README.md ADDED
@@ -0,0 +1,77 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ language:
3
+ - en
4
+ license: mit
5
+ tags:
6
+ - generated_from_trainer
7
+ datasets:
8
+ - glue
9
+ metrics:
10
+ - accuracy
11
+ model-index:
12
+ - name: roberta-base_mnli_bc
13
+ results:
14
+ - task:
15
+ name: Text Classification
16
+ type: text-classification
17
+ dataset:
18
+ name: GLUE MNLI
19
+ type: glue
20
+ args: mnli
21
+ metrics:
22
+ - name: Accuracy
23
+ type: accuracy
24
+ value: 0.9583768461882739
25
+ ---
26
+
27
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
28
+ should probably proofread and complete it, then remove this comment. -->
29
+
30
+ # roberta-base_mnli_bc
31
+
32
+ This model is a fine-tuned version of [roberta-base](https://huggingface.co/roberta-base) on the GLUE MNLI dataset.
33
+ It achieves the following results on the evaluation set:
34
+ - Loss: 0.2125
35
+ - Accuracy: 0.9584
36
+
37
+ ## Model description
38
+
39
+ More information needed
40
+
41
+ ## Intended uses & limitations
42
+
43
+ More information needed
44
+
45
+ ## Training and evaluation data
46
+
47
+ More information needed
48
+
49
+ ## Training procedure
50
+
51
+ ### Training hyperparameters
52
+
53
+ The following hyperparameters were used during training:
54
+ - learning_rate: 2e-05
55
+ - train_batch_size: 16
56
+ - eval_batch_size: 8
57
+ - seed: 42
58
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
59
+ - lr_scheduler_type: linear
60
+ - num_epochs: 3.0
61
+ - mixed_precision_training: Native AMP
62
+
63
+ ### Training results
64
+
65
+ | Training Loss | Epoch | Step | Validation Loss | Accuracy |
66
+ |:-------------:|:-----:|:-----:|:---------------:|:--------:|
67
+ | 0.2015 | 1.0 | 16363 | 0.1820 | 0.9470 |
68
+ | 0.1463 | 2.0 | 32726 | 0.1909 | 0.9559 |
69
+ | 0.0768 | 3.0 | 49089 | 0.2117 | 0.9585 |
70
+
71
+
72
+ ### Framework versions
73
+
74
+ - Transformers 4.13.0
75
+ - Pytorch 1.10.1+cu111
76
+ - Datasets 1.17.0
77
+ - Tokenizers 0.10.3
all_results.json ADDED
@@ -0,0 +1,14 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 3.0,
3
+ "eval_accuracy": 0.9583768461882739,
4
+ "eval_loss": 0.2125200629234314,
5
+ "eval_runtime": 18.5055,
6
+ "eval_samples": 6703,
7
+ "eval_samples_per_second": 362.217,
8
+ "eval_steps_per_second": 45.284,
9
+ "train_loss": 0.1589348805035952,
10
+ "train_runtime": 5162.793,
11
+ "train_samples": 261802,
12
+ "train_samples_per_second": 152.128,
13
+ "train_steps_per_second": 9.508
14
+ }
config.json ADDED
@@ -0,0 +1,39 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ "finetuning_task": "mnli",
11
+ "hidden_act": "gelu",
12
+ "hidden_dropout_prob": 0.1,
13
+ "hidden_size": 768,
14
+ "id2label": {
15
+ "0": "entailment",
16
+ "1": "neutral",
17
+ "2": "contradiction"
18
+ },
19
+ "initializer_range": 0.02,
20
+ "intermediate_size": 3072,
21
+ "label2id": {
22
+ "contradiction": 2,
23
+ "entailment": 0,
24
+ "neutral": 1
25
+ },
26
+ "layer_norm_eps": 1e-05,
27
+ "max_position_embeddings": 514,
28
+ "model_type": "roberta",
29
+ "num_attention_heads": 12,
30
+ "num_hidden_layers": 12,
31
+ "pad_token_id": 1,
32
+ "position_embedding_type": "absolute",
33
+ "problem_type": "single_label_classification",
34
+ "torch_dtype": "float32",
35
+ "transformers_version": "4.13.0",
36
+ "type_vocab_size": 1,
37
+ "use_cache": true,
38
+ "vocab_size": 50265
39
+ }
eval_results.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 3.0,
3
+ "eval_accuracy": 0.9583768461882739,
4
+ "eval_loss": 0.2125200629234314,
5
+ "eval_runtime": 18.5055,
6
+ "eval_samples": 6703,
7
+ "eval_samples_per_second": 362.217,
8
+ "eval_steps_per_second": 45.284
9
+ }
merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ababe75115c17ee4e75582d689eff69370198676a84d139a0bce7a34ed7830d6
3
+ size 498674093
special_tokens_map.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"bos_token": "<s>", "eos_token": "</s>", "unk_token": "<unk>", "sep_token": "</s>", "pad_token": "<pad>", "cls_token": "<s>", "mask_token": {"content": "<mask>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": false}}
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"unk_token": "<unk>", "bos_token": "<s>", "eos_token": "</s>", "add_prefix_space": false, "errors": "replace", "sep_token": "</s>", "cls_token": "<s>", "pad_token": "<pad>", "mask_token": "<mask>", "model_max_length": 512, "special_tokens_map_file": null, "name_or_path": "roberta-base", "tokenizer_class": "RobertaTokenizer"}
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 3.0,
3
+ "train_loss": 0.1589348805035952,
4
+ "train_runtime": 5162.793,
5
+ "train_samples": 261802,
6
+ "train_samples_per_second": 152.128,
7
+ "train_steps_per_second": 9.508
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,640 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.9584578601315002,
3
+ "best_model_checkpoint": "/hdd1/mujeen/retrieval_prf/output/roberta-base_mnli_bc/checkpoint-49089",
4
+ "epoch": 3.0,
5
+ "global_step": 49089,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.03,
12
+ "learning_rate": 1.9797918067184097e-05,
13
+ "loss": 0.3858,
14
+ "step": 500
15
+ },
16
+ {
17
+ "epoch": 0.06,
18
+ "learning_rate": 1.959420644136161e-05,
19
+ "loss": 0.284,
20
+ "step": 1000
21
+ },
22
+ {
23
+ "epoch": 0.09,
24
+ "learning_rate": 1.9390494815539124e-05,
25
+ "loss": 0.2536,
26
+ "step": 1500
27
+ },
28
+ {
29
+ "epoch": 0.12,
30
+ "learning_rate": 1.918719061296828e-05,
31
+ "loss": 0.2583,
32
+ "step": 2000
33
+ },
34
+ {
35
+ "epoch": 0.15,
36
+ "learning_rate": 1.89834789871458e-05,
37
+ "loss": 0.2522,
38
+ "step": 2500
39
+ },
40
+ {
41
+ "epoch": 0.18,
42
+ "learning_rate": 1.8779767361323312e-05,
43
+ "loss": 0.2526,
44
+ "step": 3000
45
+ },
46
+ {
47
+ "epoch": 0.21,
48
+ "learning_rate": 1.8576055735500826e-05,
49
+ "loss": 0.2457,
50
+ "step": 3500
51
+ },
52
+ {
53
+ "epoch": 0.24,
54
+ "learning_rate": 1.8372344109678343e-05,
55
+ "loss": 0.2335,
56
+ "step": 4000
57
+ },
58
+ {
59
+ "epoch": 0.28,
60
+ "learning_rate": 1.8168632483855853e-05,
61
+ "loss": 0.224,
62
+ "step": 4500
63
+ },
64
+ {
65
+ "epoch": 0.31,
66
+ "learning_rate": 1.796492085803337e-05,
67
+ "loss": 0.2234,
68
+ "step": 5000
69
+ },
70
+ {
71
+ "epoch": 0.34,
72
+ "learning_rate": 1.7761209232210884e-05,
73
+ "loss": 0.2335,
74
+ "step": 5500
75
+ },
76
+ {
77
+ "epoch": 0.37,
78
+ "learning_rate": 1.7557905029640045e-05,
79
+ "loss": 0.2341,
80
+ "step": 6000
81
+ },
82
+ {
83
+ "epoch": 0.4,
84
+ "learning_rate": 1.7354193403817555e-05,
85
+ "loss": 0.2235,
86
+ "step": 6500
87
+ },
88
+ {
89
+ "epoch": 0.43,
90
+ "learning_rate": 1.7150481777995072e-05,
91
+ "loss": 0.231,
92
+ "step": 7000
93
+ },
94
+ {
95
+ "epoch": 0.46,
96
+ "learning_rate": 1.6946770152172586e-05,
97
+ "loss": 0.2146,
98
+ "step": 7500
99
+ },
100
+ {
101
+ "epoch": 0.49,
102
+ "learning_rate": 1.67430585263501e-05,
103
+ "loss": 0.219,
104
+ "step": 8000
105
+ },
106
+ {
107
+ "epoch": 0.52,
108
+ "learning_rate": 1.653975432377926e-05,
109
+ "loss": 0.2263,
110
+ "step": 8500
111
+ },
112
+ {
113
+ "epoch": 0.55,
114
+ "learning_rate": 1.6336042697956774e-05,
115
+ "loss": 0.2197,
116
+ "step": 9000
117
+ },
118
+ {
119
+ "epoch": 0.58,
120
+ "learning_rate": 1.6132331072134288e-05,
121
+ "loss": 0.2072,
122
+ "step": 9500
123
+ },
124
+ {
125
+ "epoch": 0.61,
126
+ "learning_rate": 1.59286194463118e-05,
127
+ "loss": 0.2196,
128
+ "step": 10000
129
+ },
130
+ {
131
+ "epoch": 0.64,
132
+ "learning_rate": 1.5724907820489318e-05,
133
+ "loss": 0.2039,
134
+ "step": 10500
135
+ },
136
+ {
137
+ "epoch": 0.67,
138
+ "learning_rate": 1.5521603617918476e-05,
139
+ "loss": 0.201,
140
+ "step": 11000
141
+ },
142
+ {
143
+ "epoch": 0.7,
144
+ "learning_rate": 1.531789199209599e-05,
145
+ "loss": 0.2081,
146
+ "step": 11500
147
+ },
148
+ {
149
+ "epoch": 0.73,
150
+ "learning_rate": 1.5114180366273505e-05,
151
+ "loss": 0.2072,
152
+ "step": 12000
153
+ },
154
+ {
155
+ "epoch": 0.76,
156
+ "learning_rate": 1.4910468740451018e-05,
157
+ "loss": 0.1949,
158
+ "step": 12500
159
+ },
160
+ {
161
+ "epoch": 0.79,
162
+ "learning_rate": 1.4706757114628534e-05,
163
+ "loss": 0.2119,
164
+ "step": 13000
165
+ },
166
+ {
167
+ "epoch": 0.83,
168
+ "learning_rate": 1.4503452912057691e-05,
169
+ "loss": 0.209,
170
+ "step": 13500
171
+ },
172
+ {
173
+ "epoch": 0.86,
174
+ "learning_rate": 1.4299741286235206e-05,
175
+ "loss": 0.1992,
176
+ "step": 14000
177
+ },
178
+ {
179
+ "epoch": 0.89,
180
+ "learning_rate": 1.409602966041272e-05,
181
+ "loss": 0.1953,
182
+ "step": 14500
183
+ },
184
+ {
185
+ "epoch": 0.92,
186
+ "learning_rate": 1.3892318034590235e-05,
187
+ "loss": 0.2025,
188
+ "step": 15000
189
+ },
190
+ {
191
+ "epoch": 0.95,
192
+ "learning_rate": 1.368860640876775e-05,
193
+ "loss": 0.198,
194
+ "step": 15500
195
+ },
196
+ {
197
+ "epoch": 0.98,
198
+ "learning_rate": 1.3485302206196908e-05,
199
+ "loss": 0.2015,
200
+ "step": 16000
201
+ },
202
+ {
203
+ "epoch": 1.0,
204
+ "eval_accuracy": 0.9469515839808726,
205
+ "eval_loss": 0.18203075230121613,
206
+ "eval_runtime": 18.9219,
207
+ "eval_samples_per_second": 353.664,
208
+ "eval_steps_per_second": 44.234,
209
+ "step": 16363
210
+ },
211
+ {
212
+ "epoch": 1.01,
213
+ "learning_rate": 1.3281590580374423e-05,
214
+ "loss": 0.1932,
215
+ "step": 16500
216
+ },
217
+ {
218
+ "epoch": 1.04,
219
+ "learning_rate": 1.3077878954551937e-05,
220
+ "loss": 0.1541,
221
+ "step": 17000
222
+ },
223
+ {
224
+ "epoch": 1.07,
225
+ "learning_rate": 1.2874574751981098e-05,
226
+ "loss": 0.1556,
227
+ "step": 17500
228
+ },
229
+ {
230
+ "epoch": 1.1,
231
+ "learning_rate": 1.267086312615861e-05,
232
+ "loss": 0.1607,
233
+ "step": 18000
234
+ },
235
+ {
236
+ "epoch": 1.13,
237
+ "learning_rate": 1.2467151500336125e-05,
238
+ "loss": 0.1618,
239
+ "step": 18500
240
+ },
241
+ {
242
+ "epoch": 1.16,
243
+ "learning_rate": 1.2263439874513639e-05,
244
+ "loss": 0.1593,
245
+ "step": 19000
246
+ },
247
+ {
248
+ "epoch": 1.19,
249
+ "learning_rate": 1.2059728248691154e-05,
250
+ "loss": 0.1511,
251
+ "step": 19500
252
+ },
253
+ {
254
+ "epoch": 1.22,
255
+ "learning_rate": 1.185601662286867e-05,
256
+ "loss": 0.165,
257
+ "step": 20000
258
+ },
259
+ {
260
+ "epoch": 1.25,
261
+ "learning_rate": 1.1652304997046182e-05,
262
+ "loss": 0.157,
263
+ "step": 20500
264
+ },
265
+ {
266
+ "epoch": 1.28,
267
+ "learning_rate": 1.1448593371223697e-05,
268
+ "loss": 0.1385,
269
+ "step": 21000
270
+ },
271
+ {
272
+ "epoch": 1.31,
273
+ "learning_rate": 1.1245289168652856e-05,
274
+ "loss": 0.1646,
275
+ "step": 21500
276
+ },
277
+ {
278
+ "epoch": 1.34,
279
+ "learning_rate": 1.1041577542830371e-05,
280
+ "loss": 0.163,
281
+ "step": 22000
282
+ },
283
+ {
284
+ "epoch": 1.38,
285
+ "learning_rate": 1.0837865917007883e-05,
286
+ "loss": 0.156,
287
+ "step": 22500
288
+ },
289
+ {
290
+ "epoch": 1.41,
291
+ "learning_rate": 1.0634154291185399e-05,
292
+ "loss": 0.1556,
293
+ "step": 23000
294
+ },
295
+ {
296
+ "epoch": 1.44,
297
+ "learning_rate": 1.043085008861456e-05,
298
+ "loss": 0.1626,
299
+ "step": 23500
300
+ },
301
+ {
302
+ "epoch": 1.47,
303
+ "learning_rate": 1.0227545886043717e-05,
304
+ "loss": 0.1605,
305
+ "step": 24000
306
+ },
307
+ {
308
+ "epoch": 1.5,
309
+ "learning_rate": 1.0023834260221232e-05,
310
+ "loss": 0.1651,
311
+ "step": 24500
312
+ },
313
+ {
314
+ "epoch": 1.53,
315
+ "learning_rate": 9.820122634398746e-06,
316
+ "loss": 0.1438,
317
+ "step": 25000
318
+ },
319
+ {
320
+ "epoch": 1.56,
321
+ "learning_rate": 9.616411008576261e-06,
322
+ "loss": 0.1501,
323
+ "step": 25500
324
+ },
325
+ {
326
+ "epoch": 1.59,
327
+ "learning_rate": 9.412699382753775e-06,
328
+ "loss": 0.1562,
329
+ "step": 26000
330
+ },
331
+ {
332
+ "epoch": 1.62,
333
+ "learning_rate": 9.209395180182934e-06,
334
+ "loss": 0.158,
335
+ "step": 26500
336
+ },
337
+ {
338
+ "epoch": 1.65,
339
+ "learning_rate": 9.005683554360448e-06,
340
+ "loss": 0.149,
341
+ "step": 27000
342
+ },
343
+ {
344
+ "epoch": 1.68,
345
+ "learning_rate": 8.801971928537963e-06,
346
+ "loss": 0.1483,
347
+ "step": 27500
348
+ },
349
+ {
350
+ "epoch": 1.71,
351
+ "learning_rate": 8.598260302715477e-06,
352
+ "loss": 0.163,
353
+ "step": 28000
354
+ },
355
+ {
356
+ "epoch": 1.74,
357
+ "learning_rate": 8.39454867689299e-06,
358
+ "loss": 0.1507,
359
+ "step": 28500
360
+ },
361
+ {
362
+ "epoch": 1.77,
363
+ "learning_rate": 8.19124447432215e-06,
364
+ "loss": 0.148,
365
+ "step": 29000
366
+ },
367
+ {
368
+ "epoch": 1.8,
369
+ "learning_rate": 7.987532848499665e-06,
370
+ "loss": 0.1571,
371
+ "step": 29500
372
+ },
373
+ {
374
+ "epoch": 1.83,
375
+ "learning_rate": 7.78382122267718e-06,
376
+ "loss": 0.1467,
377
+ "step": 30000
378
+ },
379
+ {
380
+ "epoch": 1.86,
381
+ "learning_rate": 7.580109596854694e-06,
382
+ "loss": 0.1417,
383
+ "step": 30500
384
+ },
385
+ {
386
+ "epoch": 1.89,
387
+ "learning_rate": 7.376397971032207e-06,
388
+ "loss": 0.1526,
389
+ "step": 31000
390
+ },
391
+ {
392
+ "epoch": 1.93,
393
+ "learning_rate": 7.172686345209722e-06,
394
+ "loss": 0.1327,
395
+ "step": 31500
396
+ },
397
+ {
398
+ "epoch": 1.96,
399
+ "learning_rate": 6.969382142638881e-06,
400
+ "loss": 0.151,
401
+ "step": 32000
402
+ },
403
+ {
404
+ "epoch": 1.99,
405
+ "learning_rate": 6.7656705168163946e-06,
406
+ "loss": 0.1463,
407
+ "step": 32500
408
+ },
409
+ {
410
+ "epoch": 2.0,
411
+ "eval_accuracy": 0.9559175134488942,
412
+ "eval_loss": 0.19088995456695557,
413
+ "eval_runtime": 18.8218,
414
+ "eval_samples_per_second": 355.545,
415
+ "eval_steps_per_second": 44.47,
416
+ "step": 32726
417
+ },
418
+ {
419
+ "epoch": 2.02,
420
+ "learning_rate": 6.561958890993909e-06,
421
+ "loss": 0.1198,
422
+ "step": 33000
423
+ },
424
+ {
425
+ "epoch": 2.05,
426
+ "learning_rate": 6.358247265171424e-06,
427
+ "loss": 0.0851,
428
+ "step": 33500
429
+ },
430
+ {
431
+ "epoch": 2.08,
432
+ "learning_rate": 6.155350485852228e-06,
433
+ "loss": 0.1081,
434
+ "step": 34000
435
+ },
436
+ {
437
+ "epoch": 2.11,
438
+ "learning_rate": 5.9516388600297426e-06,
439
+ "loss": 0.1071,
440
+ "step": 34500
441
+ },
442
+ {
443
+ "epoch": 2.14,
444
+ "learning_rate": 5.747927234207257e-06,
445
+ "loss": 0.1046,
446
+ "step": 35000
447
+ },
448
+ {
449
+ "epoch": 2.17,
450
+ "learning_rate": 5.544215608384771e-06,
451
+ "loss": 0.0965,
452
+ "step": 35500
453
+ },
454
+ {
455
+ "epoch": 2.2,
456
+ "learning_rate": 5.340503982562285e-06,
457
+ "loss": 0.0906,
458
+ "step": 36000
459
+ },
460
+ {
461
+ "epoch": 2.23,
462
+ "learning_rate": 5.137199779991444e-06,
463
+ "loss": 0.0927,
464
+ "step": 36500
465
+ },
466
+ {
467
+ "epoch": 2.26,
468
+ "learning_rate": 4.933488154168959e-06,
469
+ "loss": 0.0947,
470
+ "step": 37000
471
+ },
472
+ {
473
+ "epoch": 2.29,
474
+ "learning_rate": 4.729776528346473e-06,
475
+ "loss": 0.1007,
476
+ "step": 37500
477
+ },
478
+ {
479
+ "epoch": 2.32,
480
+ "learning_rate": 4.526064902523987e-06,
481
+ "loss": 0.0981,
482
+ "step": 38000
483
+ },
484
+ {
485
+ "epoch": 2.35,
486
+ "learning_rate": 4.322353276701502e-06,
487
+ "loss": 0.105,
488
+ "step": 38500
489
+ },
490
+ {
491
+ "epoch": 2.38,
492
+ "learning_rate": 4.118641650879016e-06,
493
+ "loss": 0.1025,
494
+ "step": 39000
495
+ },
496
+ {
497
+ "epoch": 2.41,
498
+ "learning_rate": 3.915337448308175e-06,
499
+ "loss": 0.089,
500
+ "step": 39500
501
+ },
502
+ {
503
+ "epoch": 2.44,
504
+ "learning_rate": 3.7116258224856896e-06,
505
+ "loss": 0.1049,
506
+ "step": 40000
507
+ },
508
+ {
509
+ "epoch": 2.48,
510
+ "learning_rate": 3.507914196663204e-06,
511
+ "loss": 0.0982,
512
+ "step": 40500
513
+ },
514
+ {
515
+ "epoch": 2.51,
516
+ "learning_rate": 3.304202570840718e-06,
517
+ "loss": 0.1007,
518
+ "step": 41000
519
+ },
520
+ {
521
+ "epoch": 2.54,
522
+ "learning_rate": 3.1004909450182323e-06,
523
+ "loss": 0.098,
524
+ "step": 41500
525
+ },
526
+ {
527
+ "epoch": 2.57,
528
+ "learning_rate": 2.8967793191957468e-06,
529
+ "loss": 0.0889,
530
+ "step": 42000
531
+ },
532
+ {
533
+ "epoch": 2.6,
534
+ "learning_rate": 2.693067693373261e-06,
535
+ "loss": 0.0975,
536
+ "step": 42500
537
+ },
538
+ {
539
+ "epoch": 2.63,
540
+ "learning_rate": 2.4893560675507754e-06,
541
+ "loss": 0.1031,
542
+ "step": 43000
543
+ },
544
+ {
545
+ "epoch": 2.66,
546
+ "learning_rate": 2.28564444172829e-06,
547
+ "loss": 0.0882,
548
+ "step": 43500
549
+ },
550
+ {
551
+ "epoch": 2.69,
552
+ "learning_rate": 2.082747662409094e-06,
553
+ "loss": 0.0917,
554
+ "step": 44000
555
+ },
556
+ {
557
+ "epoch": 2.72,
558
+ "learning_rate": 1.879036036586608e-06,
559
+ "loss": 0.0935,
560
+ "step": 44500
561
+ },
562
+ {
563
+ "epoch": 2.75,
564
+ "learning_rate": 1.6753244107641225e-06,
565
+ "loss": 0.099,
566
+ "step": 45000
567
+ },
568
+ {
569
+ "epoch": 2.78,
570
+ "learning_rate": 1.4716127849416368e-06,
571
+ "loss": 0.0938,
572
+ "step": 45500
573
+ },
574
+ {
575
+ "epoch": 2.81,
576
+ "learning_rate": 1.2679011591191509e-06,
577
+ "loss": 0.0977,
578
+ "step": 46000
579
+ },
580
+ {
581
+ "epoch": 2.84,
582
+ "learning_rate": 1.0641895332966654e-06,
583
+ "loss": 0.0785,
584
+ "step": 46500
585
+ },
586
+ {
587
+ "epoch": 2.87,
588
+ "learning_rate": 8.604779074741797e-07,
589
+ "loss": 0.106,
590
+ "step": 47000
591
+ },
592
+ {
593
+ "epoch": 2.9,
594
+ "learning_rate": 6.571737049033389e-07,
595
+ "loss": 0.088,
596
+ "step": 47500
597
+ },
598
+ {
599
+ "epoch": 2.93,
600
+ "learning_rate": 4.5346207908085317e-07,
601
+ "loss": 0.087,
602
+ "step": 48000
603
+ },
604
+ {
605
+ "epoch": 2.96,
606
+ "learning_rate": 2.4975045325836747e-07,
607
+ "loss": 0.1096,
608
+ "step": 48500
609
+ },
610
+ {
611
+ "epoch": 2.99,
612
+ "learning_rate": 4.6038827435881766e-08,
613
+ "loss": 0.0768,
614
+ "step": 49000
615
+ },
616
+ {
617
+ "epoch": 3.0,
618
+ "eval_accuracy": 0.9584578601315002,
619
+ "eval_loss": 0.21173284947872162,
620
+ "eval_runtime": 18.7035,
621
+ "eval_samples_per_second": 357.794,
622
+ "eval_steps_per_second": 44.751,
623
+ "step": 49089
624
+ },
625
+ {
626
+ "epoch": 3.0,
627
+ "step": 49089,
628
+ "total_flos": 3.330874138659072e+16,
629
+ "train_loss": 0.1589348805035952,
630
+ "train_runtime": 5162.793,
631
+ "train_samples_per_second": 152.128,
632
+ "train_steps_per_second": 9.508
633
+ }
634
+ ],
635
+ "max_steps": 49089,
636
+ "num_train_epochs": 3,
637
+ "total_flos": 3.330874138659072e+16,
638
+ "trial_name": null,
639
+ "trial_params": null
640
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9bd94d35ba146b6ad3fb214b5d4aa59a0a9c8d5da7739cbcf3d07bea21c48e55
3
+ size 2991
vocab.json ADDED
The diff for this file is too large to render. See raw diff