TinyPixel commited on
Commit
f503c93
·
verified ·
1 Parent(s): 06ca39c

Upload folder using huggingface_hub

Browse files
adapter_config.json CHANGED
@@ -19,8 +19,8 @@
19
  "rank_pattern": {},
20
  "revision": null,
21
  "target_modules": [
22
- "down_proj",
23
  "gate_proj",
 
24
  "up_proj"
25
  ],
26
  "task_type": "CAUSAL_LM"
 
19
  "rank_pattern": {},
20
  "revision": null,
21
  "target_modules": [
 
22
  "gate_proj",
23
+ "down_proj",
24
  "up_proj"
25
  ],
26
  "task_type": "CAUSAL_LM"
adapter_model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:04bf91a4c947109f5b9766ab66e408333048b69fab74b08b3a80b02c1542dfd2
3
  size 58221088
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:030df8c30a66a1b2d7a1edd7a4267c77e80234033ccf2042f410b4db87c65af3
3
  size 58221088
optimizer.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:3ba738836b2b6b83f24fd038fdd8c51636338a423fe0293f316112aac7b58dce
3
  size 116508154
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:57a3b322fae3803f6115be766a996014f2cafbaf84c56b8ee45e201d1d44bf16
3
  size 116508154
rng_state.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:667f9a42015c0f8a731077598faf1119cbef4fce1f50c07b59697aeda4ee2f10
3
  size 14244
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ca428527922547719a3a8da12c376222f09b0f08bb0c2b20baeef54f220f95b8
3
  size 14244
scheduler.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:4f949432cf8f6431a77c86c6fe146a84509067c31252b0f2c6c27767ebbfcc8b
3
  size 1064
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4372b8bc129cb832d61f89de539110be4b375eeb5fe693144c4dd880623a9aaf
3
  size 1064
special_tokens_map.json CHANGED
@@ -1,20 +1,4 @@
1
  {
2
- "additional_special_tokens": [
3
- {
4
- "content": "<|im_end|>",
5
- "lstrip": false,
6
- "normalized": false,
7
- "rstrip": false,
8
- "single_word": false
9
- },
10
- {
11
- "content": "<|im_start|>",
12
- "lstrip": false,
13
- "normalized": false,
14
- "rstrip": false,
15
- "single_word": false
16
- }
17
- ],
18
  "bos_token": {
19
  "content": "<|endoftext|>",
20
  "lstrip": false,
@@ -29,7 +13,13 @@
29
  "rstrip": false,
30
  "single_word": false
31
  },
32
- "pad_token": "<|endoftext|>",
 
 
 
 
 
 
33
  "unk_token": {
34
  "content": "<|endoftext|>",
35
  "lstrip": false,
 
1
  {
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
2
  "bos_token": {
3
  "content": "<|endoftext|>",
4
  "lstrip": false,
 
13
  "rstrip": false,
14
  "single_word": false
15
  },
16
+ "pad_token": {
17
+ "content": "[PAD]",
18
+ "lstrip": false,
19
+ "normalized": false,
20
+ "rstrip": false,
21
+ "single_word": false
22
+ },
23
  "unk_token": {
24
  "content": "<|endoftext|>",
25
  "lstrip": false,
tokenizer.json CHANGED
@@ -230,16 +230,7 @@
230
  },
231
  {
232
  "id": 50277,
233
- "content": "<|im_end|>",
234
- "single_word": false,
235
- "lstrip": false,
236
- "rstrip": false,
237
- "normalized": false,
238
- "special": true
239
- },
240
- {
241
- "id": 50278,
242
- "content": "<|im_start|>",
243
  "single_word": false,
244
  "lstrip": false,
245
  "rstrip": false,
 
230
  },
231
  {
232
  "id": 50277,
233
+ "content": "[PAD]",
 
 
 
 
 
 
 
 
 
234
  "single_word": false,
235
  "lstrip": false,
236
  "rstrip": false,
tokenizer_config.json CHANGED
@@ -202,15 +202,7 @@
202
  "special": false
203
  },
204
  "50277": {
205
- "content": "<|im_end|>",
206
- "lstrip": false,
207
- "normalized": false,
208
- "rstrip": false,
209
- "single_word": false,
210
- "special": true
211
- },
212
- "50278": {
213
- "content": "<|im_start|>",
214
  "lstrip": false,
215
  "normalized": false,
216
  "rstrip": false,
@@ -218,15 +210,11 @@
218
  "special": true
219
  }
220
  },
221
- "additional_special_tokens": [
222
- "<|im_end|>",
223
- "<|im_start|>"
224
- ],
225
  "bos_token": "<|endoftext|>",
226
  "clean_up_tokenization_spaces": true,
227
  "eos_token": "<|endoftext|>",
228
  "model_max_length": 1000000000000000019884624838656,
229
- "pad_token": "<|endoftext|>",
230
  "tokenizer_class": "GPTNeoXTokenizer",
231
  "unk_token": "<|endoftext|>"
232
  }
 
202
  "special": false
203
  },
204
  "50277": {
205
+ "content": "[PAD]",
 
 
 
 
 
 
 
 
206
  "lstrip": false,
207
  "normalized": false,
208
  "rstrip": false,
 
210
  "special": true
211
  }
212
  },
 
 
 
 
213
  "bos_token": "<|endoftext|>",
214
  "clean_up_tokenization_spaces": true,
215
  "eos_token": "<|endoftext|>",
216
  "model_max_length": 1000000000000000019884624838656,
217
+ "pad_token": "[PAD]",
218
  "tokenizer_class": "GPTNeoXTokenizer",
219
  "unk_token": "<|endoftext|>"
220
  }
trainer_state.json CHANGED
@@ -1,620 +1,164 @@
1
  {
2
  "best_metric": null,
3
  "best_model_checkpoint": null,
4
- "epoch": 2.9972041006523766,
5
  "eval_steps": 500,
6
- "global_step": 201,
7
  "is_hyper_param_search": false,
8
  "is_local_process_zero": true,
9
  "is_world_process_zero": true,
10
  "log_history": [
11
  {
12
- "epoch": 0.03,
13
  "learning_rate": 2e-05,
14
- "loss": 2.3694,
15
  "step": 2
16
  },
17
  {
18
- "epoch": 0.06,
19
  "learning_rate": 2e-05,
20
- "loss": 1.5968,
21
  "step": 4
22
  },
23
  {
24
- "epoch": 0.09,
25
  "learning_rate": 2e-05,
26
- "loss": 1.49,
27
  "step": 6
28
  },
29
  {
30
- "epoch": 0.12,
31
  "learning_rate": 2e-05,
32
- "loss": 1.3725,
33
  "step": 8
34
  },
35
  {
36
- "epoch": 0.15,
37
  "learning_rate": 2e-05,
38
- "loss": 1.5102,
39
  "step": 10
40
  },
41
  {
42
- "epoch": 0.18,
43
  "learning_rate": 2e-05,
44
- "loss": 1.6967,
45
  "step": 12
46
  },
47
  {
48
- "epoch": 0.21,
49
  "learning_rate": 2e-05,
50
- "loss": 2.1444,
51
  "step": 14
52
  },
53
  {
54
- "epoch": 0.24,
55
  "learning_rate": 2e-05,
56
- "loss": 2.9943,
57
  "step": 16
58
  },
59
  {
60
- "epoch": 0.27,
61
  "learning_rate": 2e-05,
62
- "loss": 2.2151,
63
  "step": 18
64
  },
65
  {
66
- "epoch": 0.3,
67
  "learning_rate": 2e-05,
68
- "loss": 1.5057,
69
  "step": 20
70
  },
71
  {
72
- "epoch": 0.33,
73
  "learning_rate": 2e-05,
74
- "loss": 1.4797,
75
  "step": 22
76
  },
77
  {
78
- "epoch": 0.36,
79
  "learning_rate": 2e-05,
80
- "loss": 1.5333,
81
  "step": 24
82
  },
83
  {
84
- "epoch": 0.39,
85
  "learning_rate": 2e-05,
86
- "loss": 1.3528,
87
  "step": 26
88
  },
89
  {
90
- "epoch": 0.42,
91
  "learning_rate": 2e-05,
92
- "loss": 1.8362,
93
  "step": 28
94
  },
95
  {
96
- "epoch": 0.45,
97
  "learning_rate": 2e-05,
98
- "loss": 2.1474,
99
  "step": 30
100
  },
101
  {
102
- "epoch": 0.48,
103
  "learning_rate": 2e-05,
104
- "loss": 2.6363,
105
  "step": 32
106
  },
107
  {
108
- "epoch": 0.51,
109
  "learning_rate": 2e-05,
110
- "loss": 1.8868,
111
  "step": 34
112
  },
113
  {
114
- "epoch": 0.54,
115
  "learning_rate": 2e-05,
116
- "loss": 1.697,
117
  "step": 36
118
  },
119
  {
120
- "epoch": 0.57,
121
  "learning_rate": 2e-05,
122
- "loss": 1.3203,
123
  "step": 38
124
  },
125
  {
126
- "epoch": 0.6,
127
  "learning_rate": 2e-05,
128
- "loss": 1.5115,
129
  "step": 40
130
  },
131
  {
132
- "epoch": 0.63,
133
  "learning_rate": 2e-05,
134
- "loss": 1.2874,
135
  "step": 42
136
  },
137
  {
138
- "epoch": 0.66,
139
  "learning_rate": 2e-05,
140
- "loss": 1.4912,
141
  "step": 44
142
  },
143
  {
144
- "epoch": 0.69,
145
  "learning_rate": 2e-05,
146
- "loss": 1.7937,
147
  "step": 46
148
  },
149
- {
150
- "epoch": 0.72,
151
- "learning_rate": 2e-05,
152
- "loss": 2.2967,
153
- "step": 48
154
- },
155
- {
156
- "epoch": 0.75,
157
- "learning_rate": 2e-05,
158
- "loss": 2.3339,
159
- "step": 50
160
- },
161
- {
162
- "epoch": 0.78,
163
- "learning_rate": 2e-05,
164
- "loss": 1.3627,
165
- "step": 52
166
- },
167
- {
168
- "epoch": 0.81,
169
- "learning_rate": 2e-05,
170
- "loss": 1.3231,
171
- "step": 54
172
- },
173
- {
174
- "epoch": 0.84,
175
- "learning_rate": 2e-05,
176
- "loss": 1.3223,
177
- "step": 56
178
- },
179
- {
180
- "epoch": 0.86,
181
- "learning_rate": 2e-05,
182
- "loss": 1.5476,
183
- "step": 58
184
- },
185
- {
186
- "epoch": 0.89,
187
- "learning_rate": 2e-05,
188
- "loss": 1.6677,
189
- "step": 60
190
- },
191
- {
192
- "epoch": 0.92,
193
- "learning_rate": 2e-05,
194
- "loss": 1.845,
195
- "step": 62
196
- },
197
- {
198
- "epoch": 0.95,
199
- "learning_rate": 2e-05,
200
- "loss": 2.2604,
201
- "step": 64
202
- },
203
  {
204
  "epoch": 0.98,
205
  "learning_rate": 2e-05,
206
- "loss": 1.5046,
207
- "step": 66
208
- },
209
- {
210
- "epoch": 1.01,
211
- "learning_rate": 2e-05,
212
- "loss": 2.3164,
213
- "step": 68
214
- },
215
- {
216
- "epoch": 1.04,
217
- "learning_rate": 2e-05,
218
- "loss": 1.8736,
219
- "step": 70
220
- },
221
- {
222
- "epoch": 1.07,
223
- "learning_rate": 2e-05,
224
- "loss": 1.3723,
225
- "step": 72
226
- },
227
- {
228
- "epoch": 1.1,
229
- "learning_rate": 2e-05,
230
- "loss": 1.2697,
231
- "step": 74
232
- },
233
- {
234
- "epoch": 1.13,
235
- "learning_rate": 2e-05,
236
- "loss": 1.3229,
237
- "step": 76
238
- },
239
- {
240
- "epoch": 1.16,
241
- "learning_rate": 2e-05,
242
- "loss": 1.3046,
243
- "step": 78
244
- },
245
- {
246
- "epoch": 1.19,
247
- "learning_rate": 2e-05,
248
- "loss": 1.6217,
249
- "step": 80
250
- },
251
- {
252
- "epoch": 1.22,
253
- "learning_rate": 2e-05,
254
- "loss": 1.7905,
255
- "step": 82
256
- },
257
- {
258
- "epoch": 1.25,
259
- "learning_rate": 2e-05,
260
- "loss": 1.9723,
261
- "step": 84
262
- },
263
- {
264
- "epoch": 1.28,
265
- "learning_rate": 2e-05,
266
- "loss": 1.5563,
267
- "step": 86
268
- },
269
- {
270
- "epoch": 1.31,
271
- "learning_rate": 2e-05,
272
- "loss": 1.2439,
273
- "step": 88
274
- },
275
- {
276
- "epoch": 1.34,
277
- "learning_rate": 2e-05,
278
- "loss": 1.2835,
279
- "step": 90
280
- },
281
- {
282
- "epoch": 1.37,
283
- "learning_rate": 2e-05,
284
- "loss": 1.2859,
285
- "step": 92
286
- },
287
- {
288
- "epoch": 1.4,
289
- "learning_rate": 2e-05,
290
- "loss": 1.1242,
291
- "step": 94
292
- },
293
- {
294
- "epoch": 1.43,
295
- "learning_rate": 2e-05,
296
- "loss": 1.5188,
297
- "step": 96
298
- },
299
- {
300
- "epoch": 1.46,
301
- "learning_rate": 2e-05,
302
- "loss": 1.7652,
303
- "step": 98
304
- },
305
- {
306
- "epoch": 1.49,
307
- "learning_rate": 2e-05,
308
- "loss": 2.2685,
309
- "step": 100
310
- },
311
- {
312
- "epoch": 1.52,
313
- "learning_rate": 2e-05,
314
- "loss": 1.774,
315
- "step": 102
316
- },
317
- {
318
- "epoch": 1.55,
319
- "learning_rate": 2e-05,
320
- "loss": 1.2512,
321
- "step": 104
322
- },
323
- {
324
- "epoch": 1.58,
325
- "learning_rate": 2e-05,
326
- "loss": 1.3014,
327
- "step": 106
328
- },
329
- {
330
- "epoch": 1.61,
331
- "learning_rate": 2e-05,
332
- "loss": 1.282,
333
- "step": 108
334
- },
335
- {
336
- "epoch": 1.64,
337
- "learning_rate": 2e-05,
338
- "loss": 1.2282,
339
- "step": 110
340
- },
341
- {
342
- "epoch": 1.67,
343
- "learning_rate": 2e-05,
344
- "loss": 1.4717,
345
- "step": 112
346
- },
347
- {
348
- "epoch": 1.7,
349
- "learning_rate": 2e-05,
350
- "loss": 1.7285,
351
- "step": 114
352
- },
353
- {
354
- "epoch": 1.73,
355
- "learning_rate": 2e-05,
356
- "loss": 1.9227,
357
- "step": 116
358
- },
359
- {
360
- "epoch": 1.76,
361
- "learning_rate": 2e-05,
362
- "loss": 1.9831,
363
- "step": 118
364
- },
365
- {
366
- "epoch": 1.79,
367
- "learning_rate": 2e-05,
368
- "loss": 1.1965,
369
- "step": 120
370
- },
371
- {
372
- "epoch": 1.82,
373
- "learning_rate": 2e-05,
374
- "loss": 1.1885,
375
- "step": 122
376
- },
377
- {
378
- "epoch": 1.85,
379
- "learning_rate": 2e-05,
380
- "loss": 1.1176,
381
- "step": 124
382
- },
383
- {
384
- "epoch": 1.88,
385
- "learning_rate": 2e-05,
386
- "loss": 1.3141,
387
- "step": 126
388
- },
389
- {
390
- "epoch": 1.91,
391
- "learning_rate": 2e-05,
392
- "loss": 1.5024,
393
- "step": 128
394
- },
395
- {
396
- "epoch": 1.94,
397
- "learning_rate": 2e-05,
398
- "loss": 1.6938,
399
- "step": 130
400
- },
401
- {
402
- "epoch": 1.97,
403
- "learning_rate": 2e-05,
404
- "loss": 1.7779,
405
- "step": 132
406
- },
407
- {
408
- "epoch": 2.0,
409
- "learning_rate": 2e-05,
410
- "loss": 1.2812,
411
- "step": 134
412
- },
413
- {
414
- "epoch": 2.03,
415
- "learning_rate": 2e-05,
416
- "loss": 1.6223,
417
- "step": 136
418
- },
419
- {
420
- "epoch": 2.06,
421
- "learning_rate": 2e-05,
422
- "loss": 1.3561,
423
- "step": 138
424
- },
425
- {
426
- "epoch": 2.09,
427
- "learning_rate": 2e-05,
428
- "loss": 1.1687,
429
- "step": 140
430
- },
431
- {
432
- "epoch": 2.12,
433
- "learning_rate": 2e-05,
434
- "loss": 1.1408,
435
- "step": 142
436
- },
437
- {
438
- "epoch": 2.15,
439
- "learning_rate": 2e-05,
440
- "loss": 1.1346,
441
- "step": 144
442
- },
443
- {
444
- "epoch": 2.18,
445
- "learning_rate": 2e-05,
446
- "loss": 1.2752,
447
- "step": 146
448
- },
449
- {
450
- "epoch": 2.21,
451
- "learning_rate": 2e-05,
452
- "loss": 1.4722,
453
- "step": 148
454
- },
455
- {
456
- "epoch": 2.24,
457
- "learning_rate": 2e-05,
458
- "loss": 1.7086,
459
- "step": 150
460
- },
461
- {
462
- "epoch": 2.27,
463
- "learning_rate": 2e-05,
464
- "loss": 2.5066,
465
- "step": 152
466
- },
467
- {
468
- "epoch": 2.3,
469
- "learning_rate": 2e-05,
470
- "loss": 1.3293,
471
- "step": 154
472
- },
473
- {
474
- "epoch": 2.33,
475
- "learning_rate": 2e-05,
476
- "loss": 1.3019,
477
- "step": 156
478
- },
479
- {
480
- "epoch": 2.36,
481
- "learning_rate": 2e-05,
482
- "loss": 1.2667,
483
- "step": 158
484
- },
485
- {
486
- "epoch": 2.39,
487
- "learning_rate": 2e-05,
488
- "loss": 1.0921,
489
- "step": 160
490
- },
491
- {
492
- "epoch": 2.42,
493
- "learning_rate": 2e-05,
494
- "loss": 1.3771,
495
- "step": 162
496
- },
497
- {
498
- "epoch": 2.45,
499
- "learning_rate": 2e-05,
500
- "loss": 1.507,
501
- "step": 164
502
- },
503
- {
504
- "epoch": 2.48,
505
- "learning_rate": 2e-05,
506
- "loss": 1.7428,
507
- "step": 166
508
- },
509
- {
510
- "epoch": 2.51,
511
- "learning_rate": 2e-05,
512
- "loss": 2.16,
513
- "step": 168
514
- },
515
- {
516
- "epoch": 2.53,
517
- "learning_rate": 2e-05,
518
- "loss": 1.7357,
519
- "step": 170
520
- },
521
- {
522
- "epoch": 2.56,
523
- "learning_rate": 2e-05,
524
- "loss": 1.1363,
525
- "step": 172
526
- },
527
- {
528
- "epoch": 2.59,
529
- "learning_rate": 2e-05,
530
- "loss": 1.1947,
531
- "step": 174
532
- },
533
- {
534
- "epoch": 2.62,
535
- "learning_rate": 2e-05,
536
- "loss": 1.0145,
537
- "step": 176
538
- },
539
- {
540
- "epoch": 2.65,
541
- "learning_rate": 2e-05,
542
- "loss": 1.1838,
543
- "step": 178
544
- },
545
- {
546
- "epoch": 2.68,
547
- "learning_rate": 2e-05,
548
- "loss": 1.4005,
549
- "step": 180
550
- },
551
- {
552
- "epoch": 2.71,
553
- "learning_rate": 2e-05,
554
- "loss": 1.5415,
555
- "step": 182
556
- },
557
- {
558
- "epoch": 2.74,
559
- "learning_rate": 2e-05,
560
- "loss": 1.6624,
561
- "step": 184
562
- },
563
- {
564
- "epoch": 2.77,
565
- "learning_rate": 2e-05,
566
- "loss": 1.2388,
567
- "step": 186
568
- },
569
- {
570
- "epoch": 2.8,
571
- "learning_rate": 2e-05,
572
- "loss": 1.1796,
573
- "step": 188
574
- },
575
- {
576
- "epoch": 2.83,
577
- "learning_rate": 2e-05,
578
- "loss": 1.1539,
579
- "step": 190
580
- },
581
- {
582
- "epoch": 2.86,
583
- "learning_rate": 2e-05,
584
- "loss": 1.1312,
585
- "step": 192
586
- },
587
- {
588
- "epoch": 2.89,
589
- "learning_rate": 2e-05,
590
- "loss": 1.1102,
591
- "step": 194
592
- },
593
- {
594
- "epoch": 2.92,
595
- "learning_rate": 2e-05,
596
- "loss": 1.3282,
597
- "step": 196
598
- },
599
- {
600
- "epoch": 2.95,
601
- "learning_rate": 2e-05,
602
- "loss": 1.5463,
603
- "step": 198
604
- },
605
- {
606
- "epoch": 2.98,
607
- "learning_rate": 2e-05,
608
- "loss": 1.8497,
609
- "step": 200
610
  }
611
  ],
612
  "logging_steps": 2,
613
- "max_steps": 201,
614
  "num_input_tokens_seen": 0,
615
- "num_train_epochs": 3,
616
  "save_steps": 500,
617
- "total_flos": 2.64245700432384e+16,
618
  "train_batch_size": 1,
619
  "trial_name": null,
620
  "trial_params": null
 
1
  {
2
  "best_metric": null,
3
  "best_model_checkpoint": null,
4
+ "epoch": 0.9846153846153847,
5
  "eval_steps": 500,
6
+ "global_step": 48,
7
  "is_hyper_param_search": false,
8
  "is_local_process_zero": true,
9
  "is_world_process_zero": true,
10
  "log_history": [
11
  {
12
+ "epoch": 0.04,
13
  "learning_rate": 2e-05,
14
+ "loss": 1.8538,
15
  "step": 2
16
  },
17
  {
18
+ "epoch": 0.08,
19
  "learning_rate": 2e-05,
20
+ "loss": 2.2228,
21
  "step": 4
22
  },
23
  {
24
+ "epoch": 0.12,
25
  "learning_rate": 2e-05,
26
+ "loss": 2.246,
27
  "step": 6
28
  },
29
  {
30
+ "epoch": 0.16,
31
  "learning_rate": 2e-05,
32
+ "loss": 2.3182,
33
  "step": 8
34
  },
35
  {
36
+ "epoch": 0.21,
37
  "learning_rate": 2e-05,
38
+ "loss": 2.4862,
39
  "step": 10
40
  },
41
  {
42
+ "epoch": 0.25,
43
  "learning_rate": 2e-05,
44
+ "loss": 2.8649,
45
  "step": 12
46
  },
47
  {
48
+ "epoch": 0.29,
49
  "learning_rate": 2e-05,
50
+ "loss": 1.9152,
51
  "step": 14
52
  },
53
  {
54
+ "epoch": 0.33,
55
  "learning_rate": 2e-05,
56
+ "loss": 1.8919,
57
  "step": 16
58
  },
59
  {
60
+ "epoch": 0.37,
61
  "learning_rate": 2e-05,
62
+ "loss": 2.2002,
63
  "step": 18
64
  },
65
  {
66
+ "epoch": 0.41,
67
  "learning_rate": 2e-05,
68
+ "loss": 2.1343,
69
  "step": 20
70
  },
71
  {
72
+ "epoch": 0.45,
73
  "learning_rate": 2e-05,
74
+ "loss": 2.3809,
75
  "step": 22
76
  },
77
  {
78
+ "epoch": 0.49,
79
  "learning_rate": 2e-05,
80
+ "loss": 2.65,
81
  "step": 24
82
  },
83
  {
84
+ "epoch": 0.53,
85
  "learning_rate": 2e-05,
86
+ "loss": 1.8329,
87
  "step": 26
88
  },
89
  {
90
+ "epoch": 0.57,
91
  "learning_rate": 2e-05,
92
+ "loss": 2.0931,
93
  "step": 28
94
  },
95
  {
96
+ "epoch": 0.62,
97
  "learning_rate": 2e-05,
98
+ "loss": 2.3318,
99
  "step": 30
100
  },
101
  {
102
+ "epoch": 0.66,
103
  "learning_rate": 2e-05,
104
+ "loss": 2.0929,
105
  "step": 32
106
  },
107
  {
108
+ "epoch": 0.7,
109
  "learning_rate": 2e-05,
110
+ "loss": 2.1388,
111
  "step": 34
112
  },
113
  {
114
+ "epoch": 0.74,
115
  "learning_rate": 2e-05,
116
+ "loss": 2.5487,
117
  "step": 36
118
  },
119
  {
120
+ "epoch": 0.78,
121
  "learning_rate": 2e-05,
122
+ "loss": 1.8645,
123
  "step": 38
124
  },
125
  {
126
+ "epoch": 0.82,
127
  "learning_rate": 2e-05,
128
+ "loss": 2.1712,
129
  "step": 40
130
  },
131
  {
132
+ "epoch": 0.86,
133
  "learning_rate": 2e-05,
134
+ "loss": 2.2737,
135
  "step": 42
136
  },
137
  {
138
+ "epoch": 0.9,
139
  "learning_rate": 2e-05,
140
+ "loss": 2.3037,
141
  "step": 44
142
  },
143
  {
144
+ "epoch": 0.94,
145
  "learning_rate": 2e-05,
146
+ "loss": 2.1125,
147
  "step": 46
148
  },
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
149
  {
150
  "epoch": 0.98,
151
  "learning_rate": 2e-05,
152
+ "loss": 2.4756,
153
+ "step": 48
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
154
  }
155
  ],
156
  "logging_steps": 2,
157
+ "max_steps": 48,
158
  "num_input_tokens_seen": 0,
159
+ "num_train_epochs": 1,
160
  "save_steps": 500,
161
+ "total_flos": 6721315396724736.0,
162
  "train_batch_size": 1,
163
  "trial_name": null,
164
  "trial_params": null
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:2fd3c9cc4cd9d87c77700a7c9c613470c0e75a919ff24c748c9aad4601ff9c04
3
  size 4728
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d74e83a165fa225d16722a027928462e982b5e88fc5d6526c61bb1e20e6decc8
3
  size 4728