josecannete commited on
Commit
d85d6bc
1 Parent(s): 12ba864

adding model finetuned on QA (TAR)

Browse files
.gitattributes CHANGED
@@ -25,3 +25,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
25
  *.zip filter=lfs diff=lfs merge=lfs -text
26
  *.zstandard filter=lfs diff=lfs merge=lfs -text
27
  *tfevents* filter=lfs diff=lfs merge=lfs -text
 
 
25
  *.zip filter=lfs diff=lfs merge=lfs -text
26
  *.zstandard filter=lfs diff=lfs merge=lfs -text
27
  *tfevents* filter=lfs diff=lfs merge=lfs -text
28
+ eval_nbest_predictions.json filter=lfs diff=lfs merge=lfs -text
all_results.json ADDED
@@ -0,0 +1,11 @@
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 4.0,
3
+ "eval_exact_match": 53.074739829706715,
4
+ "eval_f1": 69.66985602287733,
5
+ "eval_samples": 10614,
6
+ "train_loss": 1.5664932693461966,
7
+ "train_runtime": 16788.5182,
8
+ "train_samples": 87728,
9
+ "train_samples_per_second": 20.902,
10
+ "train_steps_per_second": 1.306
11
+ }
config.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "CenIA/distillbert-base-spanish-uncased",
3
+ "activation": "gelu",
4
+ "architectures": [
5
+ "DistilBertForQuestionAnswering"
6
+ ],
7
+ "attention_dropout": 0.1,
8
+ "dim": 768,
9
+ "dropout": 0.1,
10
+ "hidden_dim": 3072,
11
+ "initializer_range": 0.02,
12
+ "max_position_embeddings": 512,
13
+ "model_type": "distilbert",
14
+ "n_heads": 12,
15
+ "n_layers": 6,
16
+ "pad_token_id": 0,
17
+ "qa_dropout": 0.1,
18
+ "seq_classif_dropout": 0.2,
19
+ "sinusoidal_pos_embds": true,
20
+ "tie_weights_": true,
21
+ "torch_dtype": "float32",
22
+ "transformers_version": "4.15.0",
23
+ "vocab_size": 31002
24
+ }
eval_nbest_predictions.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:28e0e8c87780f37dd642d68394b76eec7dcd63b615a484230cfbb31c9e058ee3
3
+ size 45105557
eval_predictions.json ADDED
The diff for this file is too large to render. See raw diff
 
eval_results.json ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 4.0,
3
+ "eval_exact_match": 53.074739829706715,
4
+ "eval_f1": 69.66985602287733,
5
+ "eval_samples": 10614
6
+ }
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:03a1b2d8ccda1dd5a4036d8fe7e54db64c9da6f3695be255ab3c16fd62e3e548
3
+ size 266971637
special_tokens_map.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]"}
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
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": false, "do_basic_tokenize": true, "never_split": null, "model_max_length": 512, "special_tokens_map_file": null, "name_or_path": "CenIA/distillbert-base-spanish-uncased", "tokenizer_class": "DistilBertTokenizer"}
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 4.0,
3
+ "train_loss": 1.5664932693461966,
4
+ "train_runtime": 16788.5182,
5
+ "train_samples": 87728,
6
+ "train_samples_per_second": 20.902,
7
+ "train_steps_per_second": 1.306
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,721 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 69.66985602287733,
3
+ "best_model_checkpoint": "/data/jcanete/all_results/tar/distillbeto/epochs_4_bs_16_lr_5e-5/checkpoint-16800",
4
+ "epoch": 4.0,
5
+ "global_step": 21932,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.05,
12
+ "eval_exact_match": 26.310312204351938,
13
+ "eval_f1": 40.606556876208344,
14
+ "step": 300
15
+ },
16
+ {
17
+ "epoch": 0.09,
18
+ "learning_rate": 4.886695239832209e-05,
19
+ "loss": 3.3838,
20
+ "step": 500
21
+ },
22
+ {
23
+ "epoch": 0.11,
24
+ "eval_exact_match": 35.042573320719015,
25
+ "eval_f1": 51.1874637199927,
26
+ "step": 600
27
+ },
28
+ {
29
+ "epoch": 0.16,
30
+ "eval_exact_match": 37.59697256385998,
31
+ "eval_f1": 55.70379412517118,
32
+ "step": 900
33
+ },
34
+ {
35
+ "epoch": 0.18,
36
+ "learning_rate": 4.772706547510487e-05,
37
+ "loss": 2.6487,
38
+ "step": 1000
39
+ },
40
+ {
41
+ "epoch": 0.22,
42
+ "eval_exact_match": 40.30274361400189,
43
+ "eval_f1": 57.845372827145646,
44
+ "step": 1200
45
+ },
46
+ {
47
+ "epoch": 0.27,
48
+ "learning_rate": 4.658717855188766e-05,
49
+ "loss": 2.4073,
50
+ "step": 1500
51
+ },
52
+ {
53
+ "epoch": 0.27,
54
+ "eval_exact_match": 41.5042573320719,
55
+ "eval_f1": 59.344242314747255,
56
+ "step": 1500
57
+ },
58
+ {
59
+ "epoch": 0.33,
60
+ "eval_exact_match": 43.528855250709555,
61
+ "eval_f1": 60.57412156026134,
62
+ "step": 1800
63
+ },
64
+ {
65
+ "epoch": 0.36,
66
+ "learning_rate": 4.544729162867044e-05,
67
+ "loss": 2.3006,
68
+ "step": 2000
69
+ },
70
+ {
71
+ "epoch": 0.38,
72
+ "eval_exact_match": 43.85998107852412,
73
+ "eval_f1": 60.74218457977161,
74
+ "step": 2100
75
+ },
76
+ {
77
+ "epoch": 0.44,
78
+ "eval_exact_match": 45.32639545884579,
79
+ "eval_f1": 62.60100745933504,
80
+ "step": 2400
81
+ },
82
+ {
83
+ "epoch": 0.46,
84
+ "learning_rate": 4.4309684479299654e-05,
85
+ "loss": 2.2202,
86
+ "step": 2500
87
+ },
88
+ {
89
+ "epoch": 0.49,
90
+ "eval_exact_match": 45.87511825922422,
91
+ "eval_f1": 62.66596281773377,
92
+ "step": 2700
93
+ },
94
+ {
95
+ "epoch": 0.55,
96
+ "learning_rate": 4.3169797556082435e-05,
97
+ "loss": 2.154,
98
+ "step": 3000
99
+ },
100
+ {
101
+ "epoch": 0.55,
102
+ "eval_exact_match": 45.53453169347209,
103
+ "eval_f1": 62.82975014751129,
104
+ "step": 3000
105
+ },
106
+ {
107
+ "epoch": 0.6,
108
+ "eval_exact_match": 46.395458845789975,
109
+ "eval_f1": 63.47874761462123,
110
+ "step": 3300
111
+ },
112
+ {
113
+ "epoch": 0.64,
114
+ "learning_rate": 4.202991063286522e-05,
115
+ "loss": 2.1242,
116
+ "step": 3500
117
+ },
118
+ {
119
+ "epoch": 0.66,
120
+ "eval_exact_match": 47.44560075685904,
121
+ "eval_f1": 64.43571994761247,
122
+ "step": 3600
123
+ },
124
+ {
125
+ "epoch": 0.71,
126
+ "eval_exact_match": 47.62535477767266,
127
+ "eval_f1": 65.01568330950501,
128
+ "step": 3900
129
+ },
130
+ {
131
+ "epoch": 0.73,
132
+ "learning_rate": 4.0890023709648004e-05,
133
+ "loss": 2.0919,
134
+ "step": 4000
135
+ },
136
+ {
137
+ "epoch": 0.77,
138
+ "eval_exact_match": 48.58088930936613,
139
+ "eval_f1": 65.844186710796,
140
+ "step": 4200
141
+ },
142
+ {
143
+ "epoch": 0.82,
144
+ "learning_rate": 3.9750136786430785e-05,
145
+ "loss": 2.0242,
146
+ "step": 4500
147
+ },
148
+ {
149
+ "epoch": 0.82,
150
+ "eval_exact_match": 48.618732261116364,
151
+ "eval_f1": 65.93288102675687,
152
+ "step": 4500
153
+ },
154
+ {
155
+ "epoch": 0.88,
156
+ "eval_exact_match": 47.62535477767266,
157
+ "eval_f1": 65.07422958403791,
158
+ "step": 4800
159
+ },
160
+ {
161
+ "epoch": 0.91,
162
+ "learning_rate": 3.861252963706001e-05,
163
+ "loss": 2.0193,
164
+ "step": 5000
165
+ },
166
+ {
167
+ "epoch": 0.93,
168
+ "eval_exact_match": 49.09176915799432,
169
+ "eval_f1": 66.2146996747605,
170
+ "step": 5100
171
+ },
172
+ {
173
+ "epoch": 0.98,
174
+ "eval_exact_match": 49.26206244087039,
175
+ "eval_f1": 67.13615087977189,
176
+ "step": 5400
177
+ },
178
+ {
179
+ "epoch": 1.0,
180
+ "learning_rate": 3.7474922487689224e-05,
181
+ "loss": 1.9651,
182
+ "step": 5500
183
+ },
184
+ {
185
+ "epoch": 1.04,
186
+ "eval_exact_match": 49.92431409649953,
187
+ "eval_f1": 66.56942877421614,
188
+ "step": 5700
189
+ },
190
+ {
191
+ "epoch": 1.09,
192
+ "learning_rate": 3.633503556447201e-05,
193
+ "loss": 1.6733,
194
+ "step": 6000
195
+ },
196
+ {
197
+ "epoch": 1.09,
198
+ "eval_exact_match": 50.08514664143803,
199
+ "eval_f1": 66.7344480216334,
200
+ "step": 6000
201
+ },
202
+ {
203
+ "epoch": 1.15,
204
+ "eval_exact_match": 49.85808893093661,
205
+ "eval_f1": 67.17831265861523,
206
+ "step": 6300
207
+ },
208
+ {
209
+ "epoch": 1.19,
210
+ "learning_rate": 3.5195148641254786e-05,
211
+ "loss": 1.7062,
212
+ "step": 6500
213
+ },
214
+ {
215
+ "epoch": 1.2,
216
+ "eval_exact_match": 50.07568590350047,
217
+ "eval_f1": 67.34676690188185,
218
+ "step": 6600
219
+ },
220
+ {
221
+ "epoch": 1.26,
222
+ "eval_exact_match": 49.92431409649953,
223
+ "eval_f1": 66.86831954211718,
224
+ "step": 6900
225
+ },
226
+ {
227
+ "epoch": 1.28,
228
+ "learning_rate": 3.4055261718037574e-05,
229
+ "loss": 1.6685,
230
+ "step": 7000
231
+ },
232
+ {
233
+ "epoch": 1.31,
234
+ "eval_exact_match": 50.3689687795648,
235
+ "eval_f1": 67.58117044489109,
236
+ "step": 7200
237
+ },
238
+ {
239
+ "epoch": 1.37,
240
+ "learning_rate": 3.2915374794820355e-05,
241
+ "loss": 1.683,
242
+ "step": 7500
243
+ },
244
+ {
245
+ "epoch": 1.37,
246
+ "eval_exact_match": 51.116367076631974,
247
+ "eval_f1": 67.65638675710437,
248
+ "step": 7500
249
+ },
250
+ {
251
+ "epoch": 1.42,
252
+ "eval_exact_match": 50.879848628193,
253
+ "eval_f1": 67.74322775380313,
254
+ "step": 7800
255
+ },
256
+ {
257
+ "epoch": 1.46,
258
+ "learning_rate": 3.177776764544958e-05,
259
+ "loss": 1.6271,
260
+ "step": 8000
261
+ },
262
+ {
263
+ "epoch": 1.48,
264
+ "eval_exact_match": 51.14474929044466,
265
+ "eval_f1": 67.41847561216078,
266
+ "step": 8100
267
+ },
268
+ {
269
+ "epoch": 1.53,
270
+ "eval_exact_match": 50.90823084200568,
271
+ "eval_f1": 67.83244584175503,
272
+ "step": 8400
273
+ },
274
+ {
275
+ "epoch": 1.55,
276
+ "learning_rate": 3.063788072223235e-05,
277
+ "loss": 1.6509,
278
+ "step": 8500
279
+ },
280
+ {
281
+ "epoch": 1.59,
282
+ "eval_exact_match": 51.485335856196784,
283
+ "eval_f1": 68.03877774930996,
284
+ "step": 8700
285
+ },
286
+ {
287
+ "epoch": 1.64,
288
+ "learning_rate": 2.949799379901514e-05,
289
+ "loss": 1.6591,
290
+ "step": 9000
291
+ },
292
+ {
293
+ "epoch": 1.64,
294
+ "eval_exact_match": 51.390728476821195,
295
+ "eval_f1": 68.12396413437372,
296
+ "step": 9000
297
+ },
298
+ {
299
+ "epoch": 1.7,
300
+ "eval_exact_match": 51.91106906338695,
301
+ "eval_f1": 68.28274234884077,
302
+ "step": 9300
303
+ },
304
+ {
305
+ "epoch": 1.73,
306
+ "learning_rate": 2.8358106875797924e-05,
307
+ "loss": 1.6597,
308
+ "step": 9500
309
+ },
310
+ {
311
+ "epoch": 1.75,
312
+ "eval_exact_match": 52.147587511825925,
313
+ "eval_f1": 68.44926769664524,
314
+ "step": 9600
315
+ },
316
+ {
317
+ "epoch": 1.81,
318
+ "eval_exact_match": 51.920529801324506,
319
+ "eval_f1": 68.25478495844789,
320
+ "step": 9900
321
+ },
322
+ {
323
+ "epoch": 1.82,
324
+ "learning_rate": 2.7218219952580705e-05,
325
+ "loss": 1.6526,
326
+ "step": 10000
327
+ },
328
+ {
329
+ "epoch": 1.86,
330
+ "eval_exact_match": 51.88268684957427,
331
+ "eval_f1": 68.59555473771334,
332
+ "step": 10200
333
+ },
334
+ {
335
+ "epoch": 1.92,
336
+ "learning_rate": 2.607833302936349e-05,
337
+ "loss": 1.6083,
338
+ "step": 10500
339
+ },
340
+ {
341
+ "epoch": 1.92,
342
+ "eval_exact_match": 52.223273415326396,
343
+ "eval_f1": 68.3894855383881,
344
+ "step": 10500
345
+ },
346
+ {
347
+ "epoch": 1.97,
348
+ "eval_exact_match": 52.04351939451277,
349
+ "eval_f1": 68.85931291757578,
350
+ "step": 10800
351
+ },
352
+ {
353
+ "epoch": 2.01,
354
+ "learning_rate": 2.493844610614627e-05,
355
+ "loss": 1.5584,
356
+ "step": 11000
357
+ },
358
+ {
359
+ "epoch": 2.02,
360
+ "eval_exact_match": 51.41911069063387,
361
+ "eval_f1": 68.45020620158613,
362
+ "step": 11100
363
+ },
364
+ {
365
+ "epoch": 2.08,
366
+ "eval_exact_match": 51.42857142857143,
367
+ "eval_f1": 68.42914280017077,
368
+ "step": 11400
369
+ },
370
+ {
371
+ "epoch": 2.1,
372
+ "learning_rate": 2.3798559182929054e-05,
373
+ "loss": 1.2913,
374
+ "step": 11500
375
+ },
376
+ {
377
+ "epoch": 2.13,
378
+ "eval_exact_match": 51.788079470198674,
379
+ "eval_f1": 68.62435982099952,
380
+ "step": 11700
381
+ },
382
+ {
383
+ "epoch": 2.19,
384
+ "learning_rate": 2.265867225971184e-05,
385
+ "loss": 1.2903,
386
+ "step": 12000
387
+ },
388
+ {
389
+ "epoch": 2.19,
390
+ "eval_exact_match": 51.83538315988647,
391
+ "eval_f1": 68.70784934153268,
392
+ "step": 12000
393
+ },
394
+ {
395
+ "epoch": 2.24,
396
+ "eval_exact_match": 52.29895931882687,
397
+ "eval_f1": 68.91609842633096,
398
+ "step": 12300
399
+ },
400
+ {
401
+ "epoch": 2.28,
402
+ "learning_rate": 2.1521065110341055e-05,
403
+ "loss": 1.3203,
404
+ "step": 12500
405
+ },
406
+ {
407
+ "epoch": 2.3,
408
+ "eval_exact_match": 51.9678334910123,
409
+ "eval_f1": 68.89606884437242,
410
+ "step": 12600
411
+ },
412
+ {
413
+ "epoch": 2.35,
414
+ "eval_exact_match": 52.21381267738884,
415
+ "eval_f1": 68.5166390817109,
416
+ "step": 12900
417
+ },
418
+ {
419
+ "epoch": 2.37,
420
+ "learning_rate": 2.0383457960970274e-05,
421
+ "loss": 1.2923,
422
+ "step": 13000
423
+ },
424
+ {
425
+ "epoch": 2.41,
426
+ "eval_exact_match": 52.11920529801324,
427
+ "eval_f1": 68.80154847964492,
428
+ "step": 13200
429
+ },
430
+ {
431
+ "epoch": 2.46,
432
+ "learning_rate": 1.9243571037753055e-05,
433
+ "loss": 1.3007,
434
+ "step": 13500
435
+ },
436
+ {
437
+ "epoch": 2.46,
438
+ "eval_exact_match": 51.93945127719962,
439
+ "eval_f1": 68.56070274331668,
440
+ "step": 13500
441
+ },
442
+ {
443
+ "epoch": 2.52,
444
+ "eval_exact_match": 52.450331125827816,
445
+ "eval_f1": 68.92160672281842,
446
+ "step": 13800
447
+ },
448
+ {
449
+ "epoch": 2.55,
450
+ "learning_rate": 1.8103684114535836e-05,
451
+ "loss": 1.3022,
452
+ "step": 14000
453
+ },
454
+ {
455
+ "epoch": 2.57,
456
+ "eval_exact_match": 52.081362346263006,
457
+ "eval_f1": 68.734639590144,
458
+ "step": 14100
459
+ },
460
+ {
461
+ "epoch": 2.63,
462
+ "eval_exact_match": 52.56385998107852,
463
+ "eval_f1": 69.14264904059371,
464
+ "step": 14400
465
+ },
466
+ {
467
+ "epoch": 2.64,
468
+ "learning_rate": 1.6963797191318624e-05,
469
+ "loss": 1.2735,
470
+ "step": 14500
471
+ },
472
+ {
473
+ "epoch": 2.68,
474
+ "eval_exact_match": 52.65846736045412,
475
+ "eval_f1": 69.27808981447744,
476
+ "step": 14700
477
+ },
478
+ {
479
+ "epoch": 2.74,
480
+ "learning_rate": 1.5823910268101405e-05,
481
+ "loss": 1.3057,
482
+ "step": 15000
483
+ },
484
+ {
485
+ "epoch": 2.74,
486
+ "eval_exact_match": 52.37464522232734,
487
+ "eval_f1": 69.28041873050952,
488
+ "step": 15000
489
+ },
490
+ {
491
+ "epoch": 2.79,
492
+ "eval_exact_match": 52.639545884579,
493
+ "eval_f1": 69.10864509041865,
494
+ "step": 15300
495
+ },
496
+ {
497
+ "epoch": 2.83,
498
+ "learning_rate": 1.4684023344884188e-05,
499
+ "loss": 1.2689,
500
+ "step": 15500
501
+ },
502
+ {
503
+ "epoch": 2.85,
504
+ "eval_exact_match": 52.157048249763484,
505
+ "eval_f1": 68.73617410296806,
506
+ "step": 15600
507
+ },
508
+ {
509
+ "epoch": 2.9,
510
+ "eval_exact_match": 52.58278145695364,
511
+ "eval_f1": 69.30018384216147,
512
+ "step": 15900
513
+ },
514
+ {
515
+ "epoch": 2.92,
516
+ "learning_rate": 1.3544136421666972e-05,
517
+ "loss": 1.2489,
518
+ "step": 16000
519
+ },
520
+ {
521
+ "epoch": 2.95,
522
+ "eval_exact_match": 52.639545884579,
523
+ "eval_f1": 69.01439480212419,
524
+ "step": 16200
525
+ },
526
+ {
527
+ "epoch": 3.01,
528
+ "learning_rate": 1.2406529272296188e-05,
529
+ "loss": 1.2788,
530
+ "step": 16500
531
+ },
532
+ {
533
+ "epoch": 3.01,
534
+ "eval_exact_match": 52.61116367076632,
535
+ "eval_f1": 69.30278936364587,
536
+ "step": 16500
537
+ },
538
+ {
539
+ "epoch": 3.06,
540
+ "eval_exact_match": 53.074739829706715,
541
+ "eval_f1": 69.66985602287733,
542
+ "step": 16800
543
+ },
544
+ {
545
+ "epoch": 3.1,
546
+ "learning_rate": 1.1266642349078972e-05,
547
+ "loss": 1.0115,
548
+ "step": 17000
549
+ },
550
+ {
551
+ "epoch": 3.12,
552
+ "eval_exact_match": 52.12866603595081,
553
+ "eval_f1": 68.93516419404564,
554
+ "step": 17100
555
+ },
556
+ {
557
+ "epoch": 3.17,
558
+ "eval_exact_match": 52.544938505203405,
559
+ "eval_f1": 68.96323841158566,
560
+ "step": 17400
561
+ },
562
+ {
563
+ "epoch": 3.19,
564
+ "learning_rate": 1.0126755425861755e-05,
565
+ "loss": 1.0276,
566
+ "step": 17500
567
+ },
568
+ {
569
+ "epoch": 3.23,
570
+ "eval_exact_match": 52.49763481551561,
571
+ "eval_f1": 69.07220845517458,
572
+ "step": 17700
573
+ },
574
+ {
575
+ "epoch": 3.28,
576
+ "learning_rate": 8.986868502644538e-06,
577
+ "loss": 1.0218,
578
+ "step": 18000
579
+ },
580
+ {
581
+ "epoch": 3.28,
582
+ "eval_exact_match": 52.29895931882687,
583
+ "eval_f1": 69.02811306482906,
584
+ "step": 18000
585
+ },
586
+ {
587
+ "epoch": 3.34,
588
+ "eval_exact_match": 52.639545884579,
589
+ "eval_f1": 69.2185250484871,
590
+ "step": 18300
591
+ },
592
+ {
593
+ "epoch": 3.37,
594
+ "learning_rate": 7.846981579427322e-06,
595
+ "loss": 1.0386,
596
+ "step": 18500
597
+ },
598
+ {
599
+ "epoch": 3.39,
600
+ "eval_exact_match": 52.459791863765375,
601
+ "eval_f1": 69.36180052659964,
602
+ "step": 18600
603
+ },
604
+ {
605
+ "epoch": 3.45,
606
+ "eval_exact_match": 52.57332071901608,
607
+ "eval_f1": 69.30319952546826,
608
+ "step": 18900
609
+ },
610
+ {
611
+ "epoch": 3.47,
612
+ "learning_rate": 6.707094656210104e-06,
613
+ "loss": 1.0086,
614
+ "step": 19000
615
+ },
616
+ {
617
+ "epoch": 3.5,
618
+ "eval_exact_match": 52.47871333964049,
619
+ "eval_f1": 69.30902776116545,
620
+ "step": 19200
621
+ },
622
+ {
623
+ "epoch": 3.56,
624
+ "learning_rate": 5.5694875068393215e-06,
625
+ "loss": 1.0143,
626
+ "step": 19500
627
+ },
628
+ {
629
+ "epoch": 3.56,
630
+ "eval_exact_match": 52.28949858088931,
631
+ "eval_f1": 69.21311662285441,
632
+ "step": 19500
633
+ },
634
+ {
635
+ "epoch": 3.61,
636
+ "eval_exact_match": 52.61116367076632,
637
+ "eval_f1": 69.32649279145429,
638
+ "step": 19800
639
+ },
640
+ {
641
+ "epoch": 3.65,
642
+ "learning_rate": 4.429600583622105e-06,
643
+ "loss": 1.0071,
644
+ "step": 20000
645
+ },
646
+ {
647
+ "epoch": 3.67,
648
+ "eval_exact_match": 52.544938505203405,
649
+ "eval_f1": 68.97807013082176,
650
+ "step": 20100
651
+ },
652
+ {
653
+ "epoch": 3.72,
654
+ "eval_exact_match": 52.5922421948912,
655
+ "eval_f1": 69.18859635059812,
656
+ "step": 20400
657
+ },
658
+ {
659
+ "epoch": 3.74,
660
+ "learning_rate": 3.289713660404888e-06,
661
+ "loss": 1.0236,
662
+ "step": 20500
663
+ },
664
+ {
665
+ "epoch": 3.78,
666
+ "eval_exact_match": 52.55439924314096,
667
+ "eval_f1": 69.08256750898333,
668
+ "step": 20700
669
+ },
670
+ {
671
+ "epoch": 3.83,
672
+ "learning_rate": 2.149826737187671e-06,
673
+ "loss": 1.0015,
674
+ "step": 21000
675
+ },
676
+ {
677
+ "epoch": 3.83,
678
+ "eval_exact_match": 52.70577105014191,
679
+ "eval_f1": 69.22107041576008,
680
+ "step": 21000
681
+ },
682
+ {
683
+ "epoch": 3.88,
684
+ "eval_exact_match": 52.73415326395459,
685
+ "eval_f1": 69.36938667664833,
686
+ "step": 21300
687
+ },
688
+ {
689
+ "epoch": 3.92,
690
+ "learning_rate": 1.0099398139704543e-06,
691
+ "loss": 1.0189,
692
+ "step": 21500
693
+ },
694
+ {
695
+ "epoch": 3.94,
696
+ "eval_exact_match": 52.63008514664144,
697
+ "eval_f1": 69.28051735265512,
698
+ "step": 21600
699
+ },
700
+ {
701
+ "epoch": 3.99,
702
+ "eval_exact_match": 52.71523178807947,
703
+ "eval_f1": 69.35772325822661,
704
+ "step": 21900
705
+ },
706
+ {
707
+ "epoch": 4.0,
708
+ "step": 21932,
709
+ "total_flos": 2.986644942216499e+16,
710
+ "train_loss": 1.5664932693461966,
711
+ "train_runtime": 16788.5182,
712
+ "train_samples_per_second": 20.902,
713
+ "train_steps_per_second": 1.306
714
+ }
715
+ ],
716
+ "max_steps": 21932,
717
+ "num_train_epochs": 4,
718
+ "total_flos": 2.986644942216499e+16,
719
+ "trial_name": null,
720
+ "trial_params": null
721
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6f3d40efbf110ab4662f579cbb5209f59612d699158341f8e98b866b04a85750
3
+ size 2991
vocab.txt ADDED
The diff for this file is too large to render. See raw diff