double7 commited on
Commit
72c63c8
1 Parent(s): 976ed98

Upload 7 files

Browse files
config.json ADDED
@@ -0,0 +1,29 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "/data0/yangs/llama-160m",
3
+ "architectures": [
4
+ "LlamaForCausalLM"
5
+ ],
6
+ "attention_bias": false,
7
+ "attention_dropout": 0.0,
8
+ "bos_token_id": 0,
9
+ "eos_token_id": 2,
10
+ "hidden_act": "silu",
11
+ "hidden_size": 768,
12
+ "initializer_range": 0.02,
13
+ "intermediate_size": 3072,
14
+ "max_position_embeddings": 2048,
15
+ "model_type": "llama",
16
+ "num_attention_heads": 12,
17
+ "num_hidden_layers": 12,
18
+ "num_key_value_heads": 12,
19
+ "pad_token_id": 1,
20
+ "pretraining_tp": 1,
21
+ "rms_norm_eps": 1e-06,
22
+ "rope_scaling": null,
23
+ "rope_theta": 10000.0,
24
+ "tie_word_embeddings": false,
25
+ "torch_dtype": "float32",
26
+ "transformers_version": "4.37.0.dev0",
27
+ "use_cache": true,
28
+ "vocab_size": 32000
29
+ }
generation_config.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 0,
4
+ "eos_token_id": 2,
5
+ "pad_token_id": 1,
6
+ "transformers_version": "4.37.0.dev0"
7
+ }
special_tokens_map.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "<s>",
4
+ "lstrip": false,
5
+ "normalized": true,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "eos_token": {
10
+ "content": "</s>",
11
+ "lstrip": false,
12
+ "normalized": true,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": "<unk>",
17
+ "unk_token": {
18
+ "content": "<unk>",
19
+ "lstrip": false,
20
+ "normalized": true,
21
+ "rstrip": false,
22
+ "single_word": false
23
+ }
24
+ }
tokenizer.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347
3
+ size 499723
tokenizer_config.json ADDED
@@ -0,0 +1,43 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_bos_token": true,
3
+ "add_eos_token": false,
4
+ "added_tokens_decoder": {
5
+ "0": {
6
+ "content": "<unk>",
7
+ "lstrip": false,
8
+ "normalized": true,
9
+ "rstrip": false,
10
+ "single_word": false,
11
+ "special": true
12
+ },
13
+ "1": {
14
+ "content": "<s>",
15
+ "lstrip": false,
16
+ "normalized": true,
17
+ "rstrip": false,
18
+ "single_word": false,
19
+ "special": true
20
+ },
21
+ "2": {
22
+ "content": "</s>",
23
+ "lstrip": false,
24
+ "normalized": true,
25
+ "rstrip": false,
26
+ "single_word": false,
27
+ "special": true
28
+ }
29
+ },
30
+ "bos_token": "<s>",
31
+ "clean_up_tokenization_spaces": false,
32
+ "eos_token": "</s>",
33
+ "legacy": true,
34
+ "model_max_length": 2048,
35
+ "pad_token": "<unk>",
36
+ "padding_side": "right",
37
+ "sp_model_kwargs": {},
38
+ "spaces_between_special_tokens": false,
39
+ "tokenizer_class": "LlamaTokenizer",
40
+ "unk_token": "<unk>",
41
+ "use_default_system_prompt": false,
42
+ "use_fast": true
43
+ }
trainer_state.json ADDED
@@ -0,0 +1,546 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 2.9980481457384514,
5
+ "eval_steps": 500,
6
+ "global_step": 4320,
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": 7.692307692307694e-06,
14
+ "loss": 3.0678,
15
+ "step": 50
16
+ },
17
+ {
18
+ "epoch": 0.07,
19
+ "learning_rate": 1.5384615384615387e-05,
20
+ "loss": 2.3204,
21
+ "step": 100
22
+ },
23
+ {
24
+ "epoch": 0.1,
25
+ "learning_rate": 1.999887567062682e-05,
26
+ "loss": 1.9946,
27
+ "step": 150
28
+ },
29
+ {
30
+ "epoch": 0.14,
31
+ "learning_rate": 1.998622986851668e-05,
32
+ "loss": 1.8239,
33
+ "step": 200
34
+ },
35
+ {
36
+ "epoch": 0.17,
37
+ "learning_rate": 1.9959550682646195e-05,
38
+ "loss": 1.7682,
39
+ "step": 250
40
+ },
41
+ {
42
+ "epoch": 0.21,
43
+ "learning_rate": 1.9918875604567105e-05,
44
+ "loss": 1.6983,
45
+ "step": 300
46
+ },
47
+ {
48
+ "epoch": 0.24,
49
+ "learning_rate": 1.9864261793886955e-05,
50
+ "loss": 1.6675,
51
+ "step": 350
52
+ },
53
+ {
54
+ "epoch": 0.28,
55
+ "learning_rate": 1.9795785997944227e-05,
56
+ "loss": 1.6711,
57
+ "step": 400
58
+ },
59
+ {
60
+ "epoch": 0.31,
61
+ "learning_rate": 1.971354444395734e-05,
62
+ "loss": 1.6544,
63
+ "step": 450
64
+ },
65
+ {
66
+ "epoch": 0.35,
67
+ "learning_rate": 1.9617652703799102e-05,
68
+ "loss": 1.6135,
69
+ "step": 500
70
+ },
71
+ {
72
+ "epoch": 0.38,
73
+ "learning_rate": 1.9508245531586613e-05,
74
+ "loss": 1.6339,
75
+ "step": 550
76
+ },
77
+ {
78
+ "epoch": 0.42,
79
+ "learning_rate": 1.9385476674314883e-05,
80
+ "loss": 1.5796,
81
+ "step": 600
82
+ },
83
+ {
84
+ "epoch": 0.45,
85
+ "learning_rate": 1.9249518655800262e-05,
86
+ "loss": 1.5763,
87
+ "step": 650
88
+ },
89
+ {
90
+ "epoch": 0.49,
91
+ "learning_rate": 1.9100562534237318e-05,
92
+ "loss": 1.5615,
93
+ "step": 700
94
+ },
95
+ {
96
+ "epoch": 0.52,
97
+ "learning_rate": 1.8938817633709796e-05,
98
+ "loss": 1.5427,
99
+ "step": 750
100
+ },
101
+ {
102
+ "epoch": 0.56,
103
+ "learning_rate": 1.8764511250033053e-05,
104
+ "loss": 1.5316,
105
+ "step": 800
106
+ },
107
+ {
108
+ "epoch": 0.59,
109
+ "learning_rate": 1.8577888331341302e-05,
110
+ "loss": 1.5422,
111
+ "step": 850
112
+ },
113
+ {
114
+ "epoch": 0.62,
115
+ "learning_rate": 1.837921113386847e-05,
116
+ "loss": 1.5417,
117
+ "step": 900
118
+ },
119
+ {
120
+ "epoch": 0.66,
121
+ "learning_rate": 1.816875885340649e-05,
122
+ "loss": 1.5356,
123
+ "step": 950
124
+ },
125
+ {
126
+ "epoch": 0.69,
127
+ "learning_rate": 1.7946827232958858e-05,
128
+ "loss": 1.5343,
129
+ "step": 1000
130
+ },
131
+ {
132
+ "epoch": 0.73,
133
+ "learning_rate": 1.7713728147140857e-05,
134
+ "loss": 1.5125,
135
+ "step": 1050
136
+ },
137
+ {
138
+ "epoch": 0.76,
139
+ "learning_rate": 1.7469789163910416e-05,
140
+ "loss": 1.5159,
141
+ "step": 1100
142
+ },
143
+ {
144
+ "epoch": 0.8,
145
+ "learning_rate": 1.7215353084245564e-05,
146
+ "loss": 1.5107,
147
+ "step": 1150
148
+ },
149
+ {
150
+ "epoch": 0.83,
151
+ "learning_rate": 1.6950777460415324e-05,
152
+ "loss": 1.5229,
153
+ "step": 1200
154
+ },
155
+ {
156
+ "epoch": 0.87,
157
+ "learning_rate": 1.667643409352098e-05,
158
+ "loss": 1.4952,
159
+ "step": 1250
160
+ },
161
+ {
162
+ "epoch": 0.9,
163
+ "learning_rate": 1.6392708511013874e-05,
164
+ "loss": 1.5025,
165
+ "step": 1300
166
+ },
167
+ {
168
+ "epoch": 0.94,
169
+ "learning_rate": 1.6099999424923913e-05,
170
+ "loss": 1.488,
171
+ "step": 1350
172
+ },
173
+ {
174
+ "epoch": 0.97,
175
+ "learning_rate": 1.5798718171560124e-05,
176
+ "loss": 1.5125,
177
+ "step": 1400
178
+ },
179
+ {
180
+ "epoch": 1.01,
181
+ "learning_rate": 1.548928813347067e-05,
182
+ "loss": 1.506,
183
+ "step": 1450
184
+ },
185
+ {
186
+ "epoch": 1.04,
187
+ "learning_rate": 1.5172144144474582e-05,
188
+ "loss": 1.4895,
189
+ "step": 1500
190
+ },
191
+ {
192
+ "epoch": 1.08,
193
+ "learning_rate": 1.4847731878601348e-05,
194
+ "loss": 1.4411,
195
+ "step": 1550
196
+ },
197
+ {
198
+ "epoch": 1.11,
199
+ "learning_rate": 1.4516507223797018e-05,
200
+ "loss": 1.4307,
201
+ "step": 1600
202
+ },
203
+ {
204
+ "epoch": 1.15,
205
+ "learning_rate": 1.4178935641276982e-05,
206
+ "loss": 1.4289,
207
+ "step": 1650
208
+ },
209
+ {
210
+ "epoch": 1.18,
211
+ "learning_rate": 1.3835491511425678e-05,
212
+ "loss": 1.4333,
213
+ "step": 1700
214
+ },
215
+ {
216
+ "epoch": 1.21,
217
+ "learning_rate": 1.3486657467162423e-05,
218
+ "loss": 1.4518,
219
+ "step": 1750
220
+ },
221
+ {
222
+ "epoch": 1.25,
223
+ "learning_rate": 1.3132923715710191e-05,
224
+ "loss": 1.4258,
225
+ "step": 1800
226
+ },
227
+ {
228
+ "epoch": 1.28,
229
+ "learning_rate": 1.277478734972041e-05,
230
+ "loss": 1.4344,
231
+ "step": 1850
232
+ },
233
+ {
234
+ "epoch": 1.32,
235
+ "learning_rate": 1.2412751648721855e-05,
236
+ "loss": 1.434,
237
+ "step": 1900
238
+ },
239
+ {
240
+ "epoch": 1.35,
241
+ "learning_rate": 1.2047325371875286e-05,
242
+ "loss": 1.4473,
243
+ "step": 1950
244
+ },
245
+ {
246
+ "epoch": 1.39,
247
+ "learning_rate": 1.1679022043027706e-05,
248
+ "loss": 1.4234,
249
+ "step": 2000
250
+ },
251
+ {
252
+ "epoch": 1.42,
253
+ "learning_rate": 1.1308359229070914e-05,
254
+ "loss": 1.4298,
255
+ "step": 2050
256
+ },
257
+ {
258
+ "epoch": 1.46,
259
+ "learning_rate": 1.093585781261852e-05,
260
+ "loss": 1.4147,
261
+ "step": 2100
262
+ },
263
+ {
264
+ "epoch": 1.49,
265
+ "learning_rate": 1.0562041260023406e-05,
266
+ "loss": 1.4194,
267
+ "step": 2150
268
+ },
269
+ {
270
+ "epoch": 1.53,
271
+ "learning_rate": 1.018743488576437e-05,
272
+ "loss": 1.4264,
273
+ "step": 2200
274
+ },
275
+ {
276
+ "epoch": 1.56,
277
+ "learning_rate": 9.812565114235635e-06,
278
+ "loss": 1.4327,
279
+ "step": 2250
280
+ },
281
+ {
282
+ "epoch": 1.6,
283
+ "learning_rate": 9.437958739976597e-06,
284
+ "loss": 1.4231,
285
+ "step": 2300
286
+ },
287
+ {
288
+ "epoch": 1.63,
289
+ "learning_rate": 9.064142187381482e-06,
290
+ "loss": 1.397,
291
+ "step": 2350
292
+ },
293
+ {
294
+ "epoch": 1.67,
295
+ "learning_rate": 8.691640770929086e-06,
296
+ "loss": 1.4213,
297
+ "step": 2400
298
+ },
299
+ {
300
+ "epoch": 1.7,
301
+ "learning_rate": 8.3209779569723e-06,
302
+ "loss": 1.3827,
303
+ "step": 2450
304
+ },
305
+ {
306
+ "epoch": 1.73,
307
+ "learning_rate": 7.952674628124715e-06,
308
+ "loss": 1.4029,
309
+ "step": 2500
310
+ },
311
+ {
312
+ "epoch": 1.77,
313
+ "learning_rate": 7.587248351278149e-06,
314
+ "loss": 1.4026,
315
+ "step": 2550
316
+ },
317
+ {
318
+ "epoch": 1.8,
319
+ "learning_rate": 7.2252126502795935e-06,
320
+ "loss": 1.4032,
321
+ "step": 2600
322
+ },
323
+ {
324
+ "epoch": 1.84,
325
+ "learning_rate": 6.867076284289811e-06,
326
+ "loss": 1.4065,
327
+ "step": 2650
328
+ },
329
+ {
330
+ "epoch": 1.87,
331
+ "learning_rate": 6.513342532837581e-06,
332
+ "loss": 1.4033,
333
+ "step": 2700
334
+ },
335
+ {
336
+ "epoch": 1.91,
337
+ "learning_rate": 6.164508488574326e-06,
338
+ "loss": 1.4236,
339
+ "step": 2750
340
+ },
341
+ {
342
+ "epoch": 1.94,
343
+ "learning_rate": 5.821064358723021e-06,
344
+ "loss": 1.4105,
345
+ "step": 2800
346
+ },
347
+ {
348
+ "epoch": 1.98,
349
+ "learning_rate": 5.483492776202986e-06,
350
+ "loss": 1.4443,
351
+ "step": 2850
352
+ },
353
+ {
354
+ "epoch": 2.01,
355
+ "learning_rate": 5.152268121398656e-06,
356
+ "loss": 1.3914,
357
+ "step": 2900
358
+ },
359
+ {
360
+ "epoch": 2.05,
361
+ "learning_rate": 4.827855855525423e-06,
362
+ "loss": 1.3694,
363
+ "step": 2950
364
+ },
365
+ {
366
+ "epoch": 2.08,
367
+ "learning_rate": 4.510711866529335e-06,
368
+ "loss": 1.3685,
369
+ "step": 3000
370
+ },
371
+ {
372
+ "epoch": 2.12,
373
+ "learning_rate": 4.201281828439878e-06,
374
+ "loss": 1.38,
375
+ "step": 3050
376
+ },
377
+ {
378
+ "epoch": 2.15,
379
+ "learning_rate": 3.9000005750760915e-06,
380
+ "loss": 1.3944,
381
+ "step": 3100
382
+ },
383
+ {
384
+ "epoch": 2.19,
385
+ "learning_rate": 3.6072914889861287e-06,
386
+ "loss": 1.3632,
387
+ "step": 3150
388
+ },
389
+ {
390
+ "epoch": 2.22,
391
+ "learning_rate": 3.323565906479024e-06,
392
+ "loss": 1.3923,
393
+ "step": 3200
394
+ },
395
+ {
396
+ "epoch": 2.26,
397
+ "learning_rate": 3.0492225395846777e-06,
398
+ "loss": 1.3598,
399
+ "step": 3250
400
+ },
401
+ {
402
+ "epoch": 2.29,
403
+ "learning_rate": 2.7846469157544353e-06,
404
+ "loss": 1.3582,
405
+ "step": 3300
406
+ },
407
+ {
408
+ "epoch": 2.32,
409
+ "learning_rate": 2.530210836089586e-06,
410
+ "loss": 1.4019,
411
+ "step": 3350
412
+ },
413
+ {
414
+ "epoch": 2.36,
415
+ "learning_rate": 2.2862718528591434e-06,
416
+ "loss": 1.3822,
417
+ "step": 3400
418
+ },
419
+ {
420
+ "epoch": 2.39,
421
+ "learning_rate": 2.0531727670411418e-06,
422
+ "loss": 1.4255,
423
+ "step": 3450
424
+ },
425
+ {
426
+ "epoch": 2.43,
427
+ "learning_rate": 1.831241146593512e-06,
428
+ "loss": 1.3501,
429
+ "step": 3500
430
+ },
431
+ {
432
+ "epoch": 2.46,
433
+ "learning_rate": 1.620788866131533e-06,
434
+ "loss": 1.3792,
435
+ "step": 3550
436
+ },
437
+ {
438
+ "epoch": 2.5,
439
+ "learning_rate": 1.4221116686587e-06,
440
+ "loss": 1.3632,
441
+ "step": 3600
442
+ },
443
+ {
444
+ "epoch": 2.53,
445
+ "learning_rate": 1.2354887499669487e-06,
446
+ "loss": 1.389,
447
+ "step": 3650
448
+ },
449
+ {
450
+ "epoch": 2.57,
451
+ "learning_rate": 1.0611823662902088e-06,
452
+ "loss": 1.3764,
453
+ "step": 3700
454
+ },
455
+ {
456
+ "epoch": 2.6,
457
+ "learning_rate": 8.994374657626847e-07,
458
+ "loss": 1.3873,
459
+ "step": 3750
460
+ },
461
+ {
462
+ "epoch": 2.64,
463
+ "learning_rate": 7.504813441997382e-07,
464
+ "loss": 1.3481,
465
+ "step": 3800
466
+ },
467
+ {
468
+ "epoch": 2.67,
469
+ "learning_rate": 6.145233256851202e-07,
470
+ "loss": 1.3685,
471
+ "step": 3850
472
+ },
473
+ {
474
+ "epoch": 2.71,
475
+ "learning_rate": 4.91754468413388e-07,
476
+ "loss": 1.3712,
477
+ "step": 3900
478
+ },
479
+ {
480
+ "epoch": 2.74,
481
+ "learning_rate": 3.823472962008978e-07,
482
+ "loss": 1.3702,
483
+ "step": 3950
484
+ },
485
+ {
486
+ "epoch": 2.78,
487
+ "learning_rate": 2.864555560426596e-07,
488
+ "loss": 1.3731,
489
+ "step": 4000
490
+ },
491
+ {
492
+ "epoch": 2.81,
493
+ "learning_rate": 2.0421400205577591e-07,
494
+ "loss": 1.392,
495
+ "step": 4050
496
+ },
497
+ {
498
+ "epoch": 2.85,
499
+ "learning_rate": 1.3573820611304745e-07,
500
+ "loss": 1.3628,
501
+ "step": 4100
502
+ },
503
+ {
504
+ "epoch": 2.88,
505
+ "learning_rate": 8.112439543289663e-08,
506
+ "loss": 1.367,
507
+ "step": 4150
508
+ },
509
+ {
510
+ "epoch": 2.91,
511
+ "learning_rate": 4.044931735380564e-08,
512
+ "loss": 1.371,
513
+ "step": 4200
514
+ },
515
+ {
516
+ "epoch": 2.95,
517
+ "learning_rate": 1.3770131483322734e-08,
518
+ "loss": 1.3728,
519
+ "step": 4250
520
+ },
521
+ {
522
+ "epoch": 2.98,
523
+ "learning_rate": 1.1243293731799665e-09,
524
+ "loss": 1.3754,
525
+ "step": 4300
526
+ },
527
+ {
528
+ "epoch": 3.0,
529
+ "step": 4320,
530
+ "total_flos": 4.683004912987013e+17,
531
+ "train_loss": 1.4905582423563357,
532
+ "train_runtime": 19585.3727,
533
+ "train_samples_per_second": 14.126,
534
+ "train_steps_per_second": 0.221
535
+ }
536
+ ],
537
+ "logging_steps": 50,
538
+ "max_steps": 4320,
539
+ "num_input_tokens_seen": 0,
540
+ "num_train_epochs": 3,
541
+ "save_steps": 1200,
542
+ "total_flos": 4.683004912987013e+17,
543
+ "train_batch_size": 2,
544
+ "trial_name": null,
545
+ "trial_params": null
546
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f89e6d423314f571519b58181726e976526e530a845a65d710f27578736c0222
3
+ size 4283