abhi11nav commited on
Commit
a72f499
·
1 Parent(s): 18b4e48

Upload 12 files

Browse files
config.json ADDED
@@ -0,0 +1,75 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "codet5-base",
3
+ "architectures": [
4
+ "T5ForConditionalGeneration"
5
+ ],
6
+ "bos_token_id": 1,
7
+ "d_ff": 3072,
8
+ "d_kv": 64,
9
+ "d_model": 768,
10
+ "decoder_start_token_id": 0,
11
+ "dense_act_fn": "relu",
12
+ "do_sample": true,
13
+ "dropout_rate": 0.1,
14
+ "early_stopping": true,
15
+ "eos_token_id": 2,
16
+ "feed_forward_proj": "relu",
17
+ "gradient_checkpointing": false,
18
+ "id2label": {
19
+ "0": "LABEL_0"
20
+ },
21
+ "initializer_factor": 1.0,
22
+ "is_encoder_decoder": true,
23
+ "is_gated_act": false,
24
+ "label2id": {
25
+ "LABEL_0": 0
26
+ },
27
+ "layer_norm_epsilon": 1e-06,
28
+ "length_penalty": 2.0,
29
+ "model_type": "t5",
30
+ "n_positions": 512,
31
+ "no_repeat_ngram_size": 3,
32
+ "num_beams": 10,
33
+ "num_decoder_layers": 12,
34
+ "num_heads": 12,
35
+ "num_layers": 12,
36
+ "output_past": true,
37
+ "pad_token_id": 0,
38
+ "relative_attention_max_distance": 128,
39
+ "relative_attention_num_buckets": 32,
40
+ "task_specific_params": {
41
+ "summarization": {
42
+ "early_stopping": true,
43
+ "length_penalty": 2.0,
44
+ "max_length": 200,
45
+ "min_length": 30,
46
+ "no_repeat_ngram_size": 3,
47
+ "num_beams": 4,
48
+ "prefix": "summarize: "
49
+ },
50
+ "translation_en_to_de": {
51
+ "early_stopping": true,
52
+ "max_length": 300,
53
+ "num_beams": 4,
54
+ "prefix": "translate English to German: "
55
+ },
56
+ "translation_en_to_fr": {
57
+ "early_stopping": true,
58
+ "max_length": 300,
59
+ "num_beams": 4,
60
+ "prefix": "translate English to French: "
61
+ },
62
+ "translation_en_to_ro": {
63
+ "early_stopping": true,
64
+ "max_length": 300,
65
+ "num_beams": 4,
66
+ "prefix": "translate English to Romanian: "
67
+ }
68
+ },
69
+ "top_k": 60,
70
+ "top_p": 0.95,
71
+ "torch_dtype": "float32",
72
+ "transformers_version": "4.26.1",
73
+ "use_cache": true,
74
+ "vocab_size": 32100
75
+ }
generation_config.json ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 1,
4
+ "decoder_start_token_id": 0,
5
+ "do_sample": true,
6
+ "early_stopping": true,
7
+ "eos_token_id": 2,
8
+ "length_penalty": 2.0,
9
+ "no_repeat_ngram_size": 3,
10
+ "num_beams": 10,
11
+ "pad_token_id": 0,
12
+ "top_k": 60,
13
+ "top_p": 0.95,
14
+ "transformers_version": "4.26.1"
15
+ }
merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:db4e1c4411746e86ef67c2a88633536419ae7cb241f76ec973251f75406ecb32
3
+ size 1783208709
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a64015ca1ecf6e974a99f5a841645cc70886ec9223f9ee59e22b92a3a85bb0ef
3
+ size 891616913
rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f3b904159387259dd553572c6b2bf0cb1d274f38c40f72de8cb51e63c8c21579
3
+ size 14575
scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6b71e08ead81acd50c2c1e258310b23b1b8e1f17daac1d146037d775c9be01eb
3
+ size 627
special_tokens_map.json ADDED
@@ -0,0 +1,753 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "additional_special_tokens": [
3
+ {
4
+ "content": "<extra_id_99>",
5
+ "lstrip": true,
6
+ "normalized": true,
7
+ "rstrip": false,
8
+ "single_word": false
9
+ },
10
+ {
11
+ "content": "<extra_id_98>",
12
+ "lstrip": true,
13
+ "normalized": true,
14
+ "rstrip": false,
15
+ "single_word": false
16
+ },
17
+ {
18
+ "content": "<extra_id_97>",
19
+ "lstrip": true,
20
+ "normalized": true,
21
+ "rstrip": false,
22
+ "single_word": false
23
+ },
24
+ {
25
+ "content": "<extra_id_96>",
26
+ "lstrip": true,
27
+ "normalized": true,
28
+ "rstrip": false,
29
+ "single_word": false
30
+ },
31
+ {
32
+ "content": "<extra_id_95>",
33
+ "lstrip": true,
34
+ "normalized": true,
35
+ "rstrip": false,
36
+ "single_word": false
37
+ },
38
+ {
39
+ "content": "<extra_id_94>",
40
+ "lstrip": true,
41
+ "normalized": true,
42
+ "rstrip": false,
43
+ "single_word": false
44
+ },
45
+ {
46
+ "content": "<extra_id_93>",
47
+ "lstrip": true,
48
+ "normalized": true,
49
+ "rstrip": false,
50
+ "single_word": false
51
+ },
52
+ {
53
+ "content": "<extra_id_92>",
54
+ "lstrip": true,
55
+ "normalized": true,
56
+ "rstrip": false,
57
+ "single_word": false
58
+ },
59
+ {
60
+ "content": "<extra_id_91>",
61
+ "lstrip": true,
62
+ "normalized": true,
63
+ "rstrip": false,
64
+ "single_word": false
65
+ },
66
+ {
67
+ "content": "<extra_id_90>",
68
+ "lstrip": true,
69
+ "normalized": true,
70
+ "rstrip": false,
71
+ "single_word": false
72
+ },
73
+ {
74
+ "content": "<extra_id_89>",
75
+ "lstrip": true,
76
+ "normalized": true,
77
+ "rstrip": false,
78
+ "single_word": false
79
+ },
80
+ {
81
+ "content": "<extra_id_88>",
82
+ "lstrip": true,
83
+ "normalized": true,
84
+ "rstrip": false,
85
+ "single_word": false
86
+ },
87
+ {
88
+ "content": "<extra_id_87>",
89
+ "lstrip": true,
90
+ "normalized": true,
91
+ "rstrip": false,
92
+ "single_word": false
93
+ },
94
+ {
95
+ "content": "<extra_id_86>",
96
+ "lstrip": true,
97
+ "normalized": true,
98
+ "rstrip": false,
99
+ "single_word": false
100
+ },
101
+ {
102
+ "content": "<extra_id_85>",
103
+ "lstrip": true,
104
+ "normalized": true,
105
+ "rstrip": false,
106
+ "single_word": false
107
+ },
108
+ {
109
+ "content": "<extra_id_84>",
110
+ "lstrip": true,
111
+ "normalized": true,
112
+ "rstrip": false,
113
+ "single_word": false
114
+ },
115
+ {
116
+ "content": "<extra_id_83>",
117
+ "lstrip": true,
118
+ "normalized": true,
119
+ "rstrip": false,
120
+ "single_word": false
121
+ },
122
+ {
123
+ "content": "<extra_id_82>",
124
+ "lstrip": true,
125
+ "normalized": true,
126
+ "rstrip": false,
127
+ "single_word": false
128
+ },
129
+ {
130
+ "content": "<extra_id_81>",
131
+ "lstrip": true,
132
+ "normalized": true,
133
+ "rstrip": false,
134
+ "single_word": false
135
+ },
136
+ {
137
+ "content": "<extra_id_80>",
138
+ "lstrip": true,
139
+ "normalized": true,
140
+ "rstrip": false,
141
+ "single_word": false
142
+ },
143
+ {
144
+ "content": "<extra_id_79>",
145
+ "lstrip": true,
146
+ "normalized": true,
147
+ "rstrip": false,
148
+ "single_word": false
149
+ },
150
+ {
151
+ "content": "<extra_id_78>",
152
+ "lstrip": true,
153
+ "normalized": true,
154
+ "rstrip": false,
155
+ "single_word": false
156
+ },
157
+ {
158
+ "content": "<extra_id_77>",
159
+ "lstrip": true,
160
+ "normalized": true,
161
+ "rstrip": false,
162
+ "single_word": false
163
+ },
164
+ {
165
+ "content": "<extra_id_76>",
166
+ "lstrip": true,
167
+ "normalized": true,
168
+ "rstrip": false,
169
+ "single_word": false
170
+ },
171
+ {
172
+ "content": "<extra_id_75>",
173
+ "lstrip": true,
174
+ "normalized": true,
175
+ "rstrip": false,
176
+ "single_word": false
177
+ },
178
+ {
179
+ "content": "<extra_id_74>",
180
+ "lstrip": true,
181
+ "normalized": true,
182
+ "rstrip": false,
183
+ "single_word": false
184
+ },
185
+ {
186
+ "content": "<extra_id_73>",
187
+ "lstrip": true,
188
+ "normalized": true,
189
+ "rstrip": false,
190
+ "single_word": false
191
+ },
192
+ {
193
+ "content": "<extra_id_72>",
194
+ "lstrip": true,
195
+ "normalized": true,
196
+ "rstrip": false,
197
+ "single_word": false
198
+ },
199
+ {
200
+ "content": "<extra_id_71>",
201
+ "lstrip": true,
202
+ "normalized": true,
203
+ "rstrip": false,
204
+ "single_word": false
205
+ },
206
+ {
207
+ "content": "<extra_id_70>",
208
+ "lstrip": true,
209
+ "normalized": true,
210
+ "rstrip": false,
211
+ "single_word": false
212
+ },
213
+ {
214
+ "content": "<extra_id_69>",
215
+ "lstrip": true,
216
+ "normalized": true,
217
+ "rstrip": false,
218
+ "single_word": false
219
+ },
220
+ {
221
+ "content": "<extra_id_68>",
222
+ "lstrip": true,
223
+ "normalized": true,
224
+ "rstrip": false,
225
+ "single_word": false
226
+ },
227
+ {
228
+ "content": "<extra_id_67>",
229
+ "lstrip": true,
230
+ "normalized": true,
231
+ "rstrip": false,
232
+ "single_word": false
233
+ },
234
+ {
235
+ "content": "<extra_id_66>",
236
+ "lstrip": true,
237
+ "normalized": true,
238
+ "rstrip": false,
239
+ "single_word": false
240
+ },
241
+ {
242
+ "content": "<extra_id_65>",
243
+ "lstrip": true,
244
+ "normalized": true,
245
+ "rstrip": false,
246
+ "single_word": false
247
+ },
248
+ {
249
+ "content": "<extra_id_64>",
250
+ "lstrip": true,
251
+ "normalized": true,
252
+ "rstrip": false,
253
+ "single_word": false
254
+ },
255
+ {
256
+ "content": "<extra_id_63>",
257
+ "lstrip": true,
258
+ "normalized": true,
259
+ "rstrip": false,
260
+ "single_word": false
261
+ },
262
+ {
263
+ "content": "<extra_id_62>",
264
+ "lstrip": true,
265
+ "normalized": true,
266
+ "rstrip": false,
267
+ "single_word": false
268
+ },
269
+ {
270
+ "content": "<extra_id_61>",
271
+ "lstrip": true,
272
+ "normalized": true,
273
+ "rstrip": false,
274
+ "single_word": false
275
+ },
276
+ {
277
+ "content": "<extra_id_60>",
278
+ "lstrip": true,
279
+ "normalized": true,
280
+ "rstrip": false,
281
+ "single_word": false
282
+ },
283
+ {
284
+ "content": "<extra_id_59>",
285
+ "lstrip": true,
286
+ "normalized": true,
287
+ "rstrip": false,
288
+ "single_word": false
289
+ },
290
+ {
291
+ "content": "<extra_id_58>",
292
+ "lstrip": true,
293
+ "normalized": true,
294
+ "rstrip": false,
295
+ "single_word": false
296
+ },
297
+ {
298
+ "content": "<extra_id_57>",
299
+ "lstrip": true,
300
+ "normalized": true,
301
+ "rstrip": false,
302
+ "single_word": false
303
+ },
304
+ {
305
+ "content": "<extra_id_56>",
306
+ "lstrip": true,
307
+ "normalized": true,
308
+ "rstrip": false,
309
+ "single_word": false
310
+ },
311
+ {
312
+ "content": "<extra_id_55>",
313
+ "lstrip": true,
314
+ "normalized": true,
315
+ "rstrip": false,
316
+ "single_word": false
317
+ },
318
+ {
319
+ "content": "<extra_id_54>",
320
+ "lstrip": true,
321
+ "normalized": true,
322
+ "rstrip": false,
323
+ "single_word": false
324
+ },
325
+ {
326
+ "content": "<extra_id_53>",
327
+ "lstrip": true,
328
+ "normalized": true,
329
+ "rstrip": false,
330
+ "single_word": false
331
+ },
332
+ {
333
+ "content": "<extra_id_52>",
334
+ "lstrip": true,
335
+ "normalized": true,
336
+ "rstrip": false,
337
+ "single_word": false
338
+ },
339
+ {
340
+ "content": "<extra_id_51>",
341
+ "lstrip": true,
342
+ "normalized": true,
343
+ "rstrip": false,
344
+ "single_word": false
345
+ },
346
+ {
347
+ "content": "<extra_id_50>",
348
+ "lstrip": true,
349
+ "normalized": true,
350
+ "rstrip": false,
351
+ "single_word": false
352
+ },
353
+ {
354
+ "content": "<extra_id_49>",
355
+ "lstrip": true,
356
+ "normalized": true,
357
+ "rstrip": false,
358
+ "single_word": false
359
+ },
360
+ {
361
+ "content": "<extra_id_48>",
362
+ "lstrip": true,
363
+ "normalized": true,
364
+ "rstrip": false,
365
+ "single_word": false
366
+ },
367
+ {
368
+ "content": "<extra_id_47>",
369
+ "lstrip": true,
370
+ "normalized": true,
371
+ "rstrip": false,
372
+ "single_word": false
373
+ },
374
+ {
375
+ "content": "<extra_id_46>",
376
+ "lstrip": true,
377
+ "normalized": true,
378
+ "rstrip": false,
379
+ "single_word": false
380
+ },
381
+ {
382
+ "content": "<extra_id_45>",
383
+ "lstrip": true,
384
+ "normalized": true,
385
+ "rstrip": false,
386
+ "single_word": false
387
+ },
388
+ {
389
+ "content": "<extra_id_44>",
390
+ "lstrip": true,
391
+ "normalized": true,
392
+ "rstrip": false,
393
+ "single_word": false
394
+ },
395
+ {
396
+ "content": "<extra_id_43>",
397
+ "lstrip": true,
398
+ "normalized": true,
399
+ "rstrip": false,
400
+ "single_word": false
401
+ },
402
+ {
403
+ "content": "<extra_id_42>",
404
+ "lstrip": true,
405
+ "normalized": true,
406
+ "rstrip": false,
407
+ "single_word": false
408
+ },
409
+ {
410
+ "content": "<extra_id_41>",
411
+ "lstrip": true,
412
+ "normalized": true,
413
+ "rstrip": false,
414
+ "single_word": false
415
+ },
416
+ {
417
+ "content": "<extra_id_40>",
418
+ "lstrip": true,
419
+ "normalized": true,
420
+ "rstrip": false,
421
+ "single_word": false
422
+ },
423
+ {
424
+ "content": "<extra_id_39>",
425
+ "lstrip": true,
426
+ "normalized": true,
427
+ "rstrip": false,
428
+ "single_word": false
429
+ },
430
+ {
431
+ "content": "<extra_id_38>",
432
+ "lstrip": true,
433
+ "normalized": true,
434
+ "rstrip": false,
435
+ "single_word": false
436
+ },
437
+ {
438
+ "content": "<extra_id_37>",
439
+ "lstrip": true,
440
+ "normalized": true,
441
+ "rstrip": false,
442
+ "single_word": false
443
+ },
444
+ {
445
+ "content": "<extra_id_36>",
446
+ "lstrip": true,
447
+ "normalized": true,
448
+ "rstrip": false,
449
+ "single_word": false
450
+ },
451
+ {
452
+ "content": "<extra_id_35>",
453
+ "lstrip": true,
454
+ "normalized": true,
455
+ "rstrip": false,
456
+ "single_word": false
457
+ },
458
+ {
459
+ "content": "<extra_id_34>",
460
+ "lstrip": true,
461
+ "normalized": true,
462
+ "rstrip": false,
463
+ "single_word": false
464
+ },
465
+ {
466
+ "content": "<extra_id_33>",
467
+ "lstrip": true,
468
+ "normalized": true,
469
+ "rstrip": false,
470
+ "single_word": false
471
+ },
472
+ {
473
+ "content": "<extra_id_32>",
474
+ "lstrip": true,
475
+ "normalized": true,
476
+ "rstrip": false,
477
+ "single_word": false
478
+ },
479
+ {
480
+ "content": "<extra_id_31>",
481
+ "lstrip": true,
482
+ "normalized": true,
483
+ "rstrip": false,
484
+ "single_word": false
485
+ },
486
+ {
487
+ "content": "<extra_id_30>",
488
+ "lstrip": true,
489
+ "normalized": true,
490
+ "rstrip": false,
491
+ "single_word": false
492
+ },
493
+ {
494
+ "content": "<extra_id_29>",
495
+ "lstrip": true,
496
+ "normalized": true,
497
+ "rstrip": false,
498
+ "single_word": false
499
+ },
500
+ {
501
+ "content": "<extra_id_28>",
502
+ "lstrip": true,
503
+ "normalized": true,
504
+ "rstrip": false,
505
+ "single_word": false
506
+ },
507
+ {
508
+ "content": "<extra_id_27>",
509
+ "lstrip": true,
510
+ "normalized": true,
511
+ "rstrip": false,
512
+ "single_word": false
513
+ },
514
+ {
515
+ "content": "<extra_id_26>",
516
+ "lstrip": true,
517
+ "normalized": true,
518
+ "rstrip": false,
519
+ "single_word": false
520
+ },
521
+ {
522
+ "content": "<extra_id_25>",
523
+ "lstrip": true,
524
+ "normalized": true,
525
+ "rstrip": false,
526
+ "single_word": false
527
+ },
528
+ {
529
+ "content": "<extra_id_24>",
530
+ "lstrip": true,
531
+ "normalized": true,
532
+ "rstrip": false,
533
+ "single_word": false
534
+ },
535
+ {
536
+ "content": "<extra_id_23>",
537
+ "lstrip": true,
538
+ "normalized": true,
539
+ "rstrip": false,
540
+ "single_word": false
541
+ },
542
+ {
543
+ "content": "<extra_id_22>",
544
+ "lstrip": true,
545
+ "normalized": true,
546
+ "rstrip": false,
547
+ "single_word": false
548
+ },
549
+ {
550
+ "content": "<extra_id_21>",
551
+ "lstrip": true,
552
+ "normalized": true,
553
+ "rstrip": false,
554
+ "single_word": false
555
+ },
556
+ {
557
+ "content": "<extra_id_20>",
558
+ "lstrip": true,
559
+ "normalized": true,
560
+ "rstrip": false,
561
+ "single_word": false
562
+ },
563
+ {
564
+ "content": "<extra_id_19>",
565
+ "lstrip": true,
566
+ "normalized": true,
567
+ "rstrip": false,
568
+ "single_word": false
569
+ },
570
+ {
571
+ "content": "<extra_id_18>",
572
+ "lstrip": true,
573
+ "normalized": true,
574
+ "rstrip": false,
575
+ "single_word": false
576
+ },
577
+ {
578
+ "content": "<extra_id_17>",
579
+ "lstrip": true,
580
+ "normalized": true,
581
+ "rstrip": false,
582
+ "single_word": false
583
+ },
584
+ {
585
+ "content": "<extra_id_16>",
586
+ "lstrip": true,
587
+ "normalized": true,
588
+ "rstrip": false,
589
+ "single_word": false
590
+ },
591
+ {
592
+ "content": "<extra_id_15>",
593
+ "lstrip": true,
594
+ "normalized": true,
595
+ "rstrip": false,
596
+ "single_word": false
597
+ },
598
+ {
599
+ "content": "<extra_id_14>",
600
+ "lstrip": true,
601
+ "normalized": true,
602
+ "rstrip": false,
603
+ "single_word": false
604
+ },
605
+ {
606
+ "content": "<extra_id_13>",
607
+ "lstrip": true,
608
+ "normalized": true,
609
+ "rstrip": false,
610
+ "single_word": false
611
+ },
612
+ {
613
+ "content": "<extra_id_12>",
614
+ "lstrip": true,
615
+ "normalized": true,
616
+ "rstrip": false,
617
+ "single_word": false
618
+ },
619
+ {
620
+ "content": "<extra_id_11>",
621
+ "lstrip": true,
622
+ "normalized": true,
623
+ "rstrip": false,
624
+ "single_word": false
625
+ },
626
+ {
627
+ "content": "<extra_id_10>",
628
+ "lstrip": true,
629
+ "normalized": true,
630
+ "rstrip": false,
631
+ "single_word": false
632
+ },
633
+ {
634
+ "content": "<extra_id_9>",
635
+ "lstrip": true,
636
+ "normalized": true,
637
+ "rstrip": false,
638
+ "single_word": false
639
+ },
640
+ {
641
+ "content": "<extra_id_8>",
642
+ "lstrip": true,
643
+ "normalized": true,
644
+ "rstrip": false,
645
+ "single_word": false
646
+ },
647
+ {
648
+ "content": "<extra_id_7>",
649
+ "lstrip": true,
650
+ "normalized": true,
651
+ "rstrip": false,
652
+ "single_word": false
653
+ },
654
+ {
655
+ "content": "<extra_id_6>",
656
+ "lstrip": true,
657
+ "normalized": true,
658
+ "rstrip": false,
659
+ "single_word": false
660
+ },
661
+ {
662
+ "content": "<extra_id_5>",
663
+ "lstrip": true,
664
+ "normalized": true,
665
+ "rstrip": false,
666
+ "single_word": false
667
+ },
668
+ {
669
+ "content": "<extra_id_4>",
670
+ "lstrip": true,
671
+ "normalized": true,
672
+ "rstrip": false,
673
+ "single_word": false
674
+ },
675
+ {
676
+ "content": "<extra_id_3>",
677
+ "lstrip": true,
678
+ "normalized": true,
679
+ "rstrip": false,
680
+ "single_word": false
681
+ },
682
+ {
683
+ "content": "<extra_id_2>",
684
+ "lstrip": true,
685
+ "normalized": true,
686
+ "rstrip": false,
687
+ "single_word": false
688
+ },
689
+ {
690
+ "content": "<extra_id_1>",
691
+ "lstrip": true,
692
+ "normalized": true,
693
+ "rstrip": false,
694
+ "single_word": false
695
+ },
696
+ {
697
+ "content": "<extra_id_0>",
698
+ "lstrip": true,
699
+ "normalized": true,
700
+ "rstrip": false,
701
+ "single_word": false
702
+ }
703
+ ],
704
+ "bos_token": {
705
+ "content": "<s>",
706
+ "lstrip": false,
707
+ "normalized": true,
708
+ "rstrip": false,
709
+ "single_word": false
710
+ },
711
+ "cls_token": {
712
+ "content": "<s>",
713
+ "lstrip": false,
714
+ "normalized": true,
715
+ "rstrip": false,
716
+ "single_word": false
717
+ },
718
+ "eos_token": {
719
+ "content": "</s>",
720
+ "lstrip": false,
721
+ "normalized": true,
722
+ "rstrip": false,
723
+ "single_word": false
724
+ },
725
+ "mask_token": {
726
+ "content": "<mask>",
727
+ "lstrip": true,
728
+ "normalized": true,
729
+ "rstrip": false,
730
+ "single_word": false
731
+ },
732
+ "pad_token": {
733
+ "content": "<pad>",
734
+ "lstrip": false,
735
+ "normalized": true,
736
+ "rstrip": false,
737
+ "single_word": false
738
+ },
739
+ "sep_token": {
740
+ "content": "</s>",
741
+ "lstrip": false,
742
+ "normalized": true,
743
+ "rstrip": false,
744
+ "single_word": false
745
+ },
746
+ "unk_token": {
747
+ "content": "<unk>",
748
+ "lstrip": false,
749
+ "normalized": true,
750
+ "rstrip": false,
751
+ "single_word": false
752
+ }
753
+ }
tokenizer_config.json ADDED
@@ -0,0 +1,64 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": false,
3
+ "bos_token": {
4
+ "__type": "AddedToken",
5
+ "content": "<s>",
6
+ "lstrip": false,
7
+ "normalized": true,
8
+ "rstrip": false,
9
+ "single_word": false
10
+ },
11
+ "cls_token": {
12
+ "__type": "AddedToken",
13
+ "content": "<s>",
14
+ "lstrip": false,
15
+ "normalized": true,
16
+ "rstrip": false,
17
+ "single_word": false
18
+ },
19
+ "eos_token": {
20
+ "__type": "AddedToken",
21
+ "content": "</s>",
22
+ "lstrip": false,
23
+ "normalized": true,
24
+ "rstrip": false,
25
+ "single_word": false
26
+ },
27
+ "errors": "replace",
28
+ "mask_token": {
29
+ "__type": "AddedToken",
30
+ "content": "<mask>",
31
+ "lstrip": true,
32
+ "normalized": true,
33
+ "rstrip": false,
34
+ "single_word": false
35
+ },
36
+ "model_max_length": 512,
37
+ "name_or_path": "codet5-base",
38
+ "pad_token": {
39
+ "__type": "AddedToken",
40
+ "content": "<pad>",
41
+ "lstrip": false,
42
+ "normalized": true,
43
+ "rstrip": false,
44
+ "single_word": false
45
+ },
46
+ "sep_token": {
47
+ "__type": "AddedToken",
48
+ "content": "</s>",
49
+ "lstrip": false,
50
+ "normalized": true,
51
+ "rstrip": false,
52
+ "single_word": false
53
+ },
54
+ "special_tokens_map_file": "codet5-base/special_tokens_map.json",
55
+ "tokenizer_class": "RobertaTokenizer",
56
+ "unk_token": {
57
+ "__type": "AddedToken",
58
+ "content": "<unk>",
59
+ "lstrip": false,
60
+ "normalized": true,
61
+ "rstrip": false,
62
+ "single_word": false
63
+ }
64
+ }
trainer_state.json ADDED
@@ -0,0 +1,279 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 11.0,
5
+ "global_step": 10329,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.53,
12
+ "learning_rate": 5e-05,
13
+ "loss": 4.5299,
14
+ "step": 500
15
+ },
16
+ {
17
+ "epoch": 1.0,
18
+ "eval_bleu1_score": 0.1103,
19
+ "eval_bleu2_score": 0.0516,
20
+ "eval_bleu3_score": 0.0302,
21
+ "eval_bleu4_score": 0.019,
22
+ "eval_loss": 3.4503543376922607,
23
+ "eval_rougeL": 0.1058,
24
+ "eval_runtime": 174.17,
25
+ "eval_samples_per_second": 9.588,
26
+ "eval_steps_per_second": 0.603,
27
+ "step": 939
28
+ },
29
+ {
30
+ "epoch": 1.06,
31
+ "learning_rate": 4.745650625699461e-05,
32
+ "loss": 3.6998,
33
+ "step": 1000
34
+ },
35
+ {
36
+ "epoch": 1.6,
37
+ "learning_rate": 4.4913012513989216e-05,
38
+ "loss": 3.374,
39
+ "step": 1500
40
+ },
41
+ {
42
+ "epoch": 2.0,
43
+ "eval_bleu1_score": 0.1251,
44
+ "eval_bleu2_score": 0.0617,
45
+ "eval_bleu3_score": 0.0349,
46
+ "eval_bleu4_score": 0.0207,
47
+ "eval_loss": 3.306809425354004,
48
+ "eval_rougeL": 0.1227,
49
+ "eval_runtime": 174.9295,
50
+ "eval_samples_per_second": 9.547,
51
+ "eval_steps_per_second": 0.6,
52
+ "step": 1878
53
+ },
54
+ {
55
+ "epoch": 2.13,
56
+ "learning_rate": 4.2369518770983826e-05,
57
+ "loss": 3.2617,
58
+ "step": 2000
59
+ },
60
+ {
61
+ "epoch": 2.66,
62
+ "learning_rate": 3.982602502797843e-05,
63
+ "loss": 2.9885,
64
+ "step": 2500
65
+ },
66
+ {
67
+ "epoch": 3.0,
68
+ "eval_bleu1_score": 0.1377,
69
+ "eval_bleu2_score": 0.0693,
70
+ "eval_bleu3_score": 0.0386,
71
+ "eval_bleu4_score": 0.0223,
72
+ "eval_loss": 3.2763712406158447,
73
+ "eval_rougeL": 0.1372,
74
+ "eval_runtime": 174.431,
75
+ "eval_samples_per_second": 9.574,
76
+ "eval_steps_per_second": 0.602,
77
+ "step": 2817
78
+ },
79
+ {
80
+ "epoch": 3.19,
81
+ "learning_rate": 3.7282531284973047e-05,
82
+ "loss": 2.8648,
83
+ "step": 3000
84
+ },
85
+ {
86
+ "epoch": 3.73,
87
+ "learning_rate": 3.473903754196765e-05,
88
+ "loss": 2.6525,
89
+ "step": 3500
90
+ },
91
+ {
92
+ "epoch": 4.0,
93
+ "eval_bleu1_score": 0.1382,
94
+ "eval_bleu2_score": 0.073,
95
+ "eval_bleu3_score": 0.0432,
96
+ "eval_bleu4_score": 0.027,
97
+ "eval_loss": 3.3136885166168213,
98
+ "eval_rougeL": 0.1429,
99
+ "eval_runtime": 173.1129,
100
+ "eval_samples_per_second": 9.647,
101
+ "eval_steps_per_second": 0.607,
102
+ "step": 3756
103
+ },
104
+ {
105
+ "epoch": 4.26,
106
+ "learning_rate": 3.219554379896226e-05,
107
+ "loss": 2.5232,
108
+ "step": 4000
109
+ },
110
+ {
111
+ "epoch": 4.79,
112
+ "learning_rate": 2.9652050055956863e-05,
113
+ "loss": 2.3911,
114
+ "step": 4500
115
+ },
116
+ {
117
+ "epoch": 5.0,
118
+ "eval_bleu1_score": 0.1455,
119
+ "eval_bleu2_score": 0.076,
120
+ "eval_bleu3_score": 0.0451,
121
+ "eval_bleu4_score": 0.0285,
122
+ "eval_loss": 3.354043960571289,
123
+ "eval_rougeL": 0.1431,
124
+ "eval_runtime": 172.1208,
125
+ "eval_samples_per_second": 9.702,
126
+ "eval_steps_per_second": 0.61,
127
+ "step": 4695
128
+ },
129
+ {
130
+ "epoch": 5.32,
131
+ "learning_rate": 2.710855631295147e-05,
132
+ "loss": 2.2543,
133
+ "step": 5000
134
+ },
135
+ {
136
+ "epoch": 5.86,
137
+ "learning_rate": 2.456506256994608e-05,
138
+ "loss": 2.1732,
139
+ "step": 5500
140
+ },
141
+ {
142
+ "epoch": 6.0,
143
+ "eval_bleu1_score": 0.1462,
144
+ "eval_bleu2_score": 0.0784,
145
+ "eval_bleu3_score": 0.0486,
146
+ "eval_bleu4_score": 0.0326,
147
+ "eval_loss": 3.422563314437866,
148
+ "eval_rougeL": 0.1439,
149
+ "eval_runtime": 172.488,
150
+ "eval_samples_per_second": 9.682,
151
+ "eval_steps_per_second": 0.609,
152
+ "step": 5634
153
+ },
154
+ {
155
+ "epoch": 6.39,
156
+ "learning_rate": 2.2021568826940687e-05,
157
+ "loss": 2.021,
158
+ "step": 6000
159
+ },
160
+ {
161
+ "epoch": 6.92,
162
+ "learning_rate": 1.9478075083935294e-05,
163
+ "loss": 1.966,
164
+ "step": 6500
165
+ },
166
+ {
167
+ "epoch": 7.0,
168
+ "eval_bleu1_score": 0.1527,
169
+ "eval_bleu2_score": 0.0836,
170
+ "eval_bleu3_score": 0.0514,
171
+ "eval_bleu4_score": 0.035,
172
+ "eval_loss": 3.4865562915802,
173
+ "eval_rougeL": 0.1446,
174
+ "eval_runtime": 173.8607,
175
+ "eval_samples_per_second": 9.605,
176
+ "eval_steps_per_second": 0.604,
177
+ "step": 6573
178
+ },
179
+ {
180
+ "epoch": 7.45,
181
+ "learning_rate": 1.6934581340929904e-05,
182
+ "loss": 1.8372,
183
+ "step": 7000
184
+ },
185
+ {
186
+ "epoch": 7.99,
187
+ "learning_rate": 1.4391087597924511e-05,
188
+ "loss": 1.8303,
189
+ "step": 7500
190
+ },
191
+ {
192
+ "epoch": 8.0,
193
+ "eval_bleu1_score": 0.1517,
194
+ "eval_bleu2_score": 0.0854,
195
+ "eval_bleu3_score": 0.0551,
196
+ "eval_bleu4_score": 0.0387,
197
+ "eval_loss": 3.54579758644104,
198
+ "eval_rougeL": 0.1532,
199
+ "eval_runtime": 172.6178,
200
+ "eval_samples_per_second": 9.675,
201
+ "eval_steps_per_second": 0.608,
202
+ "step": 7512
203
+ },
204
+ {
205
+ "epoch": 8.52,
206
+ "learning_rate": 1.1847593854919118e-05,
207
+ "loss": 1.6837,
208
+ "step": 8000
209
+ },
210
+ {
211
+ "epoch": 9.0,
212
+ "eval_bleu1_score": 0.1575,
213
+ "eval_bleu2_score": 0.0881,
214
+ "eval_bleu3_score": 0.0575,
215
+ "eval_bleu4_score": 0.0414,
216
+ "eval_loss": 3.5976877212524414,
217
+ "eval_rougeL": 0.1527,
218
+ "eval_runtime": 173.4085,
219
+ "eval_samples_per_second": 9.63,
220
+ "eval_steps_per_second": 0.606,
221
+ "step": 8451
222
+ },
223
+ {
224
+ "epoch": 9.05,
225
+ "learning_rate": 9.304100111913726e-06,
226
+ "loss": 1.6791,
227
+ "step": 8500
228
+ },
229
+ {
230
+ "epoch": 9.58,
231
+ "learning_rate": 6.760606368908333e-06,
232
+ "loss": 1.591,
233
+ "step": 9000
234
+ },
235
+ {
236
+ "epoch": 10.0,
237
+ "eval_bleu1_score": 0.1533,
238
+ "eval_bleu2_score": 0.0887,
239
+ "eval_bleu3_score": 0.0595,
240
+ "eval_bleu4_score": 0.0437,
241
+ "eval_loss": 3.6393702030181885,
242
+ "eval_rougeL": 0.1505,
243
+ "eval_runtime": 172.9918,
244
+ "eval_samples_per_second": 9.654,
245
+ "eval_steps_per_second": 0.607,
246
+ "step": 9390
247
+ },
248
+ {
249
+ "epoch": 10.12,
250
+ "learning_rate": 4.217112625902941e-06,
251
+ "loss": 1.5695,
252
+ "step": 9500
253
+ },
254
+ {
255
+ "epoch": 10.65,
256
+ "learning_rate": 1.6736188828975481e-06,
257
+ "loss": 1.5271,
258
+ "step": 10000
259
+ },
260
+ {
261
+ "epoch": 11.0,
262
+ "eval_bleu1_score": 0.1599,
263
+ "eval_bleu2_score": 0.0937,
264
+ "eval_bleu3_score": 0.0638,
265
+ "eval_bleu4_score": 0.0478,
266
+ "eval_loss": 3.6564855575561523,
267
+ "eval_rougeL": 0.1548,
268
+ "eval_runtime": 172.8273,
269
+ "eval_samples_per_second": 9.663,
270
+ "eval_steps_per_second": 0.608,
271
+ "step": 10329
272
+ }
273
+ ],
274
+ "max_steps": 10329,
275
+ "num_train_epochs": 11,
276
+ "total_flos": 1.0063211828871168e+17,
277
+ "trial_name": null,
278
+ "trial_params": null
279
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:38217653e9e6de8ef2af04d565784b9fe1ae338be2f4a23819368c9ad8b5000b
3
+ size 3579
vocab.json ADDED
The diff for this file is too large to render. See raw diff