Sen Fang commited on
Commit
ce105c4
·
1 Parent(s): c0bfefc
README.md ADDED
@@ -0,0 +1,21 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ ---
4
+ ## Training procedure
5
+
6
+
7
+ The following `bitsandbytes` quantization config was used during training:
8
+ - quant_method: bitsandbytes
9
+ - load_in_8bit: True
10
+ - load_in_4bit: False
11
+ - llm_int8_threshold: 6.0
12
+ - llm_int8_skip_modules: None
13
+ - llm_int8_enable_fp32_cpu_offload: False
14
+ - llm_int8_has_fp16_weight: False
15
+ - bnb_4bit_quant_type: fp4
16
+ - bnb_4bit_use_double_quant: False
17
+ - bnb_4bit_compute_dtype: float32
18
+ ### Framework versions
19
+
20
+
21
+ - PEFT 0.5.0
adapter_config.json ADDED
@@ -0,0 +1,21 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "auto_mapping": null,
3
+ "base_model_name_or_path": "/proj/berzelius-2023-175/users/x_senfa/apr_ft/llms/CodeLlama-7B-fp16",
4
+ "bias": "none",
5
+ "fan_in_fan_out": false,
6
+ "inference_mode": true,
7
+ "init_lora_weights": true,
8
+ "layers_pattern": null,
9
+ "layers_to_transform": null,
10
+ "lora_alpha": 16,
11
+ "lora_dropout": 0.05,
12
+ "modules_to_save": null,
13
+ "peft_type": "LORA",
14
+ "r": 8,
15
+ "revision": null,
16
+ "target_modules": [
17
+ "q_proj",
18
+ "v_proj"
19
+ ],
20
+ "task_type": "CAUSAL_LM"
21
+ }
adapter_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fcd3acb860e9ad211de93e279a3805dc69242cd76d8ad1688aba02ff78619b6d
3
+ size 16822989
special_tokens_map.json ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "<s>",
3
+ "eos_token": "</s>",
4
+ "pad_token": "<unk>",
5
+ "unk_token": "<unk>"
6
+ }
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1,41 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "<unk>",
5
+ "lstrip": false,
6
+ "normalized": true,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "1": {
12
+ "content": "<s>",
13
+ "lstrip": false,
14
+ "normalized": true,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "2": {
20
+ "content": "</s>",
21
+ "lstrip": false,
22
+ "normalized": true,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ }
27
+ },
28
+ "additional_special_tokens": [],
29
+ "bos_token": "<s>",
30
+ "clean_up_tokenization_spaces": false,
31
+ "eos_token": "</s>",
32
+ "legacy": null,
33
+ "model_max_length": 1024,
34
+ "pad_token": "<unk>",
35
+ "padding_side": "left",
36
+ "sp_model_kwargs": {},
37
+ "spaces_between_special_tokens": false,
38
+ "tokenizer_class": "LlamaTokenizer",
39
+ "unk_token": "<unk>",
40
+ "use_default_system_prompt": true
41
+ }
trainer_state.json ADDED
@@ -0,0 +1,904 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 2.0,
5
+ "eval_steps": 10.0,
6
+ "global_step": 1462,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.01,
13
+ "learning_rate": 0.0004999422837239569,
14
+ "loss": 0.841,
15
+ "step": 10
16
+ },
17
+ {
18
+ "epoch": 0.03,
19
+ "learning_rate": 0.000499769161545176,
20
+ "loss": 0.5851,
21
+ "step": 20
22
+ },
23
+ {
24
+ "epoch": 0.04,
25
+ "learning_rate": 0.0004994807133993966,
26
+ "loss": 0.582,
27
+ "step": 30
28
+ },
29
+ {
30
+ "epoch": 0.05,
31
+ "learning_rate": 0.0004990770724718415,
32
+ "loss": 0.5389,
33
+ "step": 40
34
+ },
35
+ {
36
+ "epoch": 0.07,
37
+ "learning_rate": 0.0004985584251357201,
38
+ "loss": 0.5349,
39
+ "step": 50
40
+ },
41
+ {
42
+ "epoch": 0.08,
43
+ "learning_rate": 0.000497925010866175,
44
+ "loss": 0.5181,
45
+ "step": 60
46
+ },
47
+ {
48
+ "epoch": 0.1,
49
+ "learning_rate": 0.0004971771221297088,
50
+ "loss": 0.5041,
51
+ "step": 70
52
+ },
53
+ {
54
+ "epoch": 0.11,
55
+ "learning_rate": 0.0004963151042491437,
56
+ "loss": 0.5311,
57
+ "step": 80
58
+ },
59
+ {
60
+ "epoch": 0.12,
61
+ "learning_rate": 0.0004953393552441752,
62
+ "loss": 0.4979,
63
+ "step": 90
64
+ },
65
+ {
66
+ "epoch": 0.14,
67
+ "learning_rate": 0.0004942503256475948,
68
+ "loss": 0.5101,
69
+ "step": 100
70
+ },
71
+ {
72
+ "epoch": 0.15,
73
+ "learning_rate": 0.000493048518297265,
74
+ "loss": 0.5458,
75
+ "step": 110
76
+ },
77
+ {
78
+ "epoch": 0.16,
79
+ "learning_rate": 0.0004917344881039438,
80
+ "loss": 0.5137,
81
+ "step": 120
82
+ },
83
+ {
84
+ "epoch": 0.18,
85
+ "learning_rate": 0.0004903088417950664,
86
+ "loss": 0.4865,
87
+ "step": 130
88
+ },
89
+ {
90
+ "epoch": 0.19,
91
+ "learning_rate": 0.0004887722376345999,
92
+ "loss": 0.5004,
93
+ "step": 140
94
+ },
95
+ {
96
+ "epoch": 0.21,
97
+ "learning_rate": 0.00048712538511910353,
98
+ "loss": 0.4931,
99
+ "step": 150
100
+ },
101
+ {
102
+ "epoch": 0.22,
103
+ "learning_rate": 0.0004853690446501323,
104
+ "loss": 0.4855,
105
+ "step": 160
106
+ },
107
+ {
108
+ "epoch": 0.23,
109
+ "learning_rate": 0.00048350402718313703,
110
+ "loss": 0.5288,
111
+ "step": 170
112
+ },
113
+ {
114
+ "epoch": 0.25,
115
+ "learning_rate": 0.00048153119385302114,
116
+ "loss": 0.4931,
117
+ "step": 180
118
+ },
119
+ {
120
+ "epoch": 0.26,
121
+ "learning_rate": 0.00047945145557652923,
122
+ "loss": 0.5064,
123
+ "step": 190
124
+ },
125
+ {
126
+ "epoch": 0.27,
127
+ "learning_rate": 0.000477265772631649,
128
+ "loss": 0.5217,
129
+ "step": 200
130
+ },
131
+ {
132
+ "epoch": 0.29,
133
+ "learning_rate": 0.00047497515421422205,
134
+ "loss": 0.4952,
135
+ "step": 210
136
+ },
137
+ {
138
+ "epoch": 0.3,
139
+ "learning_rate": 0.00047258065797196744,
140
+ "loss": 0.5501,
141
+ "step": 220
142
+ },
143
+ {
144
+ "epoch": 0.31,
145
+ "learning_rate": 0.00047008338951613397,
146
+ "loss": 0.5033,
147
+ "step": 230
148
+ },
149
+ {
150
+ "epoch": 0.33,
151
+ "learning_rate": 0.000467484501911006,
152
+ "loss": 0.5217,
153
+ "step": 240
154
+ },
155
+ {
156
+ "epoch": 0.34,
157
+ "learning_rate": 0.000464785195141499,
158
+ "loss": 0.5051,
159
+ "step": 250
160
+ },
161
+ {
162
+ "epoch": 0.36,
163
+ "learning_rate": 0.00046198671555909,
164
+ "loss": 0.5066,
165
+ "step": 260
166
+ },
167
+ {
168
+ "epoch": 0.37,
169
+ "learning_rate": 0.0004590903553063396,
170
+ "loss": 0.4972,
171
+ "step": 270
172
+ },
173
+ {
174
+ "epoch": 0.38,
175
+ "learning_rate": 0.0004560974517202708,
176
+ "loss": 0.5008,
177
+ "step": 280
178
+ },
179
+ {
180
+ "epoch": 0.4,
181
+ "learning_rate": 0.00045300938671487994,
182
+ "loss": 0.5091,
183
+ "step": 290
184
+ },
185
+ {
186
+ "epoch": 0.41,
187
+ "learning_rate": 0.0004498275861430654,
188
+ "loss": 0.4764,
189
+ "step": 300
190
+ },
191
+ {
192
+ "epoch": 0.42,
193
+ "learning_rate": 0.00044655351913826794,
194
+ "loss": 0.506,
195
+ "step": 310
196
+ },
197
+ {
198
+ "epoch": 0.44,
199
+ "learning_rate": 0.00044318869743612796,
200
+ "loss": 0.521,
201
+ "step": 320
202
+ },
203
+ {
204
+ "epoch": 0.45,
205
+ "learning_rate": 0.000439734674676471,
206
+ "loss": 0.4923,
207
+ "step": 330
208
+ },
209
+ {
210
+ "epoch": 0.47,
211
+ "learning_rate": 0.00043619304568594545,
212
+ "loss": 0.5319,
213
+ "step": 340
214
+ },
215
+ {
216
+ "epoch": 0.48,
217
+ "learning_rate": 0.0004325654457416431,
218
+ "loss": 0.4804,
219
+ "step": 350
220
+ },
221
+ {
222
+ "epoch": 0.49,
223
+ "learning_rate": 0.00042922848418325,
224
+ "loss": 0.5062,
225
+ "step": 360
226
+ },
227
+ {
228
+ "epoch": 0.51,
229
+ "learning_rate": 0.0004254421861612533,
230
+ "loss": 0.4685,
231
+ "step": 370
232
+ },
233
+ {
234
+ "epoch": 0.52,
235
+ "learning_rate": 0.000421574881182088,
236
+ "loss": 0.502,
237
+ "step": 380
238
+ },
239
+ {
240
+ "epoch": 0.53,
241
+ "learning_rate": 0.0004176283548972877,
242
+ "loss": 0.5225,
243
+ "step": 390
244
+ },
245
+ {
246
+ "epoch": 0.55,
247
+ "learning_rate": 0.0004136044295372563,
248
+ "loss": 0.5375,
249
+ "step": 400
250
+ },
251
+ {
252
+ "epoch": 0.56,
253
+ "learning_rate": 0.00040950496306988853,
254
+ "loss": 0.5128,
255
+ "step": 410
256
+ },
257
+ {
258
+ "epoch": 0.57,
259
+ "learning_rate": 0.00040533184834269057,
260
+ "loss": 0.5079,
261
+ "step": 420
262
+ },
263
+ {
264
+ "epoch": 0.59,
265
+ "learning_rate": 0.0004010870122087948,
266
+ "loss": 0.5297,
267
+ "step": 430
268
+ },
269
+ {
270
+ "epoch": 0.6,
271
+ "learning_rate": 0.0003967724146372738,
272
+ "loss": 0.5069,
273
+ "step": 440
274
+ },
275
+ {
276
+ "epoch": 0.62,
277
+ "learning_rate": 0.0003923900478081629,
278
+ "loss": 0.5238,
279
+ "step": 450
280
+ },
281
+ {
282
+ "epoch": 0.63,
283
+ "learning_rate": 0.0003879419351926115,
284
+ "loss": 0.5052,
285
+ "step": 460
286
+ },
287
+ {
288
+ "epoch": 0.64,
289
+ "learning_rate": 0.0003834301306185841,
290
+ "loss": 0.5068,
291
+ "step": 470
292
+ },
293
+ {
294
+ "epoch": 0.66,
295
+ "learning_rate": 0.0003788567173225469,
296
+ "loss": 0.497,
297
+ "step": 480
298
+ },
299
+ {
300
+ "epoch": 0.67,
301
+ "learning_rate": 0.0003742238069875736,
302
+ "loss": 0.5203,
303
+ "step": 490
304
+ },
305
+ {
306
+ "epoch": 0.68,
307
+ "learning_rate": 0.00036953353876831866,
308
+ "loss": 0.4843,
309
+ "step": 500
310
+ },
311
+ {
312
+ "epoch": 0.7,
313
+ "learning_rate": 0.000364788078303304,
314
+ "loss": 0.4976,
315
+ "step": 510
316
+ },
317
+ {
318
+ "epoch": 0.71,
319
+ "learning_rate": 0.000359989616714979,
320
+ "loss": 0.5101,
321
+ "step": 520
322
+ },
323
+ {
324
+ "epoch": 0.73,
325
+ "learning_rate": 0.0003551403695980124,
326
+ "loss": 0.485,
327
+ "step": 530
328
+ },
329
+ {
330
+ "epoch": 0.74,
331
+ "learning_rate": 0.0003502425759962859,
332
+ "loss": 0.5139,
333
+ "step": 540
334
+ },
335
+ {
336
+ "epoch": 0.75,
337
+ "learning_rate": 0.0003452984973690597,
338
+ "loss": 0.507,
339
+ "step": 550
340
+ },
341
+ {
342
+ "epoch": 0.77,
343
+ "learning_rate": 0.0003403104165467883,
344
+ "loss": 0.4868,
345
+ "step": 560
346
+ },
347
+ {
348
+ "epoch": 0.78,
349
+ "learning_rate": 0.000335280636677069,
350
+ "loss": 0.5278,
351
+ "step": 570
352
+ },
353
+ {
354
+ "epoch": 0.79,
355
+ "learning_rate": 0.00033021148016120915,
356
+ "loss": 0.4709,
357
+ "step": 580
358
+ },
359
+ {
360
+ "epoch": 0.81,
361
+ "learning_rate": 0.0003251052875819029,
362
+ "loss": 0.5023,
363
+ "step": 590
364
+ },
365
+ {
366
+ "epoch": 0.82,
367
+ "learning_rate": 0.0003199644166225136,
368
+ "loss": 0.499,
369
+ "step": 600
370
+ },
371
+ {
372
+ "epoch": 0.83,
373
+ "learning_rate": 0.00031479124097846047,
374
+ "loss": 0.5014,
375
+ "step": 610
376
+ },
377
+ {
378
+ "epoch": 0.85,
379
+ "learning_rate": 0.0003095881492612115,
380
+ "loss": 0.5224,
381
+ "step": 620
382
+ },
383
+ {
384
+ "epoch": 0.86,
385
+ "learning_rate": 0.0003043575438953893,
386
+ "loss": 0.5063,
387
+ "step": 630
388
+ },
389
+ {
390
+ "epoch": 0.88,
391
+ "learning_rate": 0.00029910184000949916,
392
+ "loss": 0.5092,
393
+ "step": 640
394
+ },
395
+ {
396
+ "epoch": 0.89,
397
+ "learning_rate": 0.0002938234643207913,
398
+ "loss": 0.498,
399
+ "step": 650
400
+ },
401
+ {
402
+ "epoch": 0.9,
403
+ "learning_rate": 0.0002885248540147722,
404
+ "loss": 0.4861,
405
+ "step": 660
406
+ },
407
+ {
408
+ "epoch": 0.92,
409
+ "learning_rate": 0.00028320845561988255,
410
+ "loss": 0.5027,
411
+ "step": 670
412
+ },
413
+ {
414
+ "epoch": 0.93,
415
+ "learning_rate": 0.0002778767238778607,
416
+ "loss": 0.5162,
417
+ "step": 680
418
+ },
419
+ {
420
+ "epoch": 0.94,
421
+ "learning_rate": 0.0002725321206103147,
422
+ "loss": 0.495,
423
+ "step": 690
424
+ },
425
+ {
426
+ "epoch": 0.96,
427
+ "learning_rate": 0.0002671771135820249,
428
+ "loss": 0.4804,
429
+ "step": 700
430
+ },
431
+ {
432
+ "epoch": 0.97,
433
+ "learning_rate": 0.0002618141753615021,
434
+ "loss": 0.4684,
435
+ "step": 710
436
+ },
437
+ {
438
+ "epoch": 0.98,
439
+ "learning_rate": 0.0002564457821793282,
440
+ "loss": 0.4736,
441
+ "step": 720
442
+ },
443
+ {
444
+ "epoch": 1.0,
445
+ "learning_rate": 0.00025107441278480563,
446
+ "loss": 0.4928,
447
+ "step": 730
448
+ },
449
+ {
450
+ "epoch": 1.01,
451
+ "learning_rate": 0.00024570254730144417,
452
+ "loss": 0.5076,
453
+ "step": 740
454
+ },
455
+ {
456
+ "epoch": 1.03,
457
+ "learning_rate": 0.00024033266608181257,
458
+ "loss": 0.4936,
459
+ "step": 750
460
+ },
461
+ {
462
+ "epoch": 1.04,
463
+ "learning_rate": 0.00023496724856228514,
464
+ "loss": 0.5076,
465
+ "step": 760
466
+ },
467
+ {
468
+ "epoch": 1.05,
469
+ "learning_rate": 0.00022960877211821102,
470
+ "loss": 0.4881,
471
+ "step": 770
472
+ },
473
+ {
474
+ "epoch": 1.07,
475
+ "learning_rate": 0.0002242597109200352,
476
+ "loss": 0.5276,
477
+ "step": 780
478
+ },
479
+ {
480
+ "epoch": 1.08,
481
+ "learning_rate": 0.00021892253479089907,
482
+ "loss": 0.4963,
483
+ "step": 790
484
+ },
485
+ {
486
+ "epoch": 1.09,
487
+ "learning_rate": 0.00021359970806624883,
488
+ "loss": 0.4923,
489
+ "step": 800
490
+ },
491
+ {
492
+ "epoch": 1.11,
493
+ "learning_rate": 0.00020829368845597693,
494
+ "loss": 0.5217,
495
+ "step": 810
496
+ },
497
+ {
498
+ "epoch": 1.12,
499
+ "learning_rate": 0.0002030069259096234,
500
+ "loss": 0.4848,
501
+ "step": 820
502
+ },
503
+ {
504
+ "epoch": 1.14,
505
+ "learning_rate": 0.0001977418614851605,
506
+ "loss": 0.4759,
507
+ "step": 830
508
+ },
509
+ {
510
+ "epoch": 1.15,
511
+ "learning_rate": 0.0001925009262218818,
512
+ "loss": 0.5065,
513
+ "step": 840
514
+ },
515
+ {
516
+ "epoch": 1.16,
517
+ "learning_rate": 0.0001872865400179182,
518
+ "loss": 0.4642,
519
+ "step": 850
520
+ },
521
+ {
522
+ "epoch": 1.18,
523
+ "learning_rate": 0.00018210111051289815,
524
+ "loss": 0.4869,
525
+ "step": 860
526
+ },
527
+ {
528
+ "epoch": 1.19,
529
+ "learning_rate": 0.00017694703197626734,
530
+ "loss": 0.4859,
531
+ "step": 870
532
+ },
533
+ {
534
+ "epoch": 1.2,
535
+ "learning_rate": 0.0001718266842017823,
536
+ "loss": 0.4529,
537
+ "step": 880
538
+ },
539
+ {
540
+ "epoch": 1.22,
541
+ "learning_rate": 0.00016674243140868777,
542
+ "loss": 0.5037,
543
+ "step": 890
544
+ },
545
+ {
546
+ "epoch": 1.23,
547
+ "learning_rate": 0.0001616966211500851,
548
+ "loss": 0.475,
549
+ "step": 900
550
+ },
551
+ {
552
+ "epoch": 1.24,
553
+ "learning_rate": 0.0001566915832289964,
554
+ "loss": 0.4951,
555
+ "step": 910
556
+ },
557
+ {
558
+ "epoch": 1.26,
559
+ "learning_rate": 0.00015172962862262362,
560
+ "loss": 0.4845,
561
+ "step": 920
562
+ },
563
+ {
564
+ "epoch": 1.27,
565
+ "learning_rate": 0.00014681304841530107,
566
+ "loss": 0.4528,
567
+ "step": 930
568
+ },
569
+ {
570
+ "epoch": 1.29,
571
+ "learning_rate": 0.00014194411274063212,
572
+ "loss": 0.4856,
573
+ "step": 940
574
+ },
575
+ {
576
+ "epoch": 1.3,
577
+ "learning_rate": 0.00013712506973330036,
578
+ "loss": 0.4764,
579
+ "step": 950
580
+ },
581
+ {
582
+ "epoch": 1.31,
583
+ "learning_rate": 0.00013235814449103736,
584
+ "loss": 0.4775,
585
+ "step": 960
586
+ },
587
+ {
588
+ "epoch": 1.33,
589
+ "learning_rate": 0.00012764553804722867,
590
+ "loss": 0.4961,
591
+ "step": 970
592
+ },
593
+ {
594
+ "epoch": 1.34,
595
+ "learning_rate": 0.00012298942635462923,
596
+ "loss": 0.4862,
597
+ "step": 980
598
+ },
599
+ {
600
+ "epoch": 1.35,
601
+ "learning_rate": 0.00011839195928066102,
602
+ "loss": 0.485,
603
+ "step": 990
604
+ },
605
+ {
606
+ "epoch": 1.37,
607
+ "learning_rate": 0.00011385525961475404,
608
+ "loss": 0.4862,
609
+ "step": 1000
610
+ },
611
+ {
612
+ "epoch": 1.38,
613
+ "learning_rate": 0.00010938142208819007,
614
+ "loss": 0.4768,
615
+ "step": 1010
616
+ },
617
+ {
618
+ "epoch": 1.4,
619
+ "learning_rate": 0.00010497251240690254,
620
+ "loss": 0.4885,
621
+ "step": 1020
622
+ },
623
+ {
624
+ "epoch": 1.41,
625
+ "learning_rate": 0.00010063056629767691,
626
+ "loss": 0.4758,
627
+ "step": 1030
628
+ },
629
+ {
630
+ "epoch": 1.42,
631
+ "learning_rate": 9.635758856819508e-05,
632
+ "loss": 0.4676,
633
+ "step": 1040
634
+ },
635
+ {
636
+ "epoch": 1.44,
637
+ "learning_rate": 9.215555218135416e-05,
638
+ "loss": 0.4858,
639
+ "step": 1050
640
+ },
641
+ {
642
+ "epoch": 1.45,
643
+ "learning_rate": 8.802639734429043e-05,
644
+ "loss": 0.456,
645
+ "step": 1060
646
+ },
647
+ {
648
+ "epoch": 1.46,
649
+ "learning_rate": 8.39720306125274e-05,
650
+ "loss": 0.4776,
651
+ "step": 1070
652
+ },
653
+ {
654
+ "epoch": 1.48,
655
+ "learning_rate": 7.999432400966063e-05,
656
+ "loss": 0.461,
657
+ "step": 1080
658
+ },
659
+ {
660
+ "epoch": 1.49,
661
+ "learning_rate": 7.609511416298859e-05,
662
+ "loss": 0.4893,
663
+ "step": 1090
664
+ },
665
+ {
666
+ "epoch": 1.5,
667
+ "learning_rate": 7.227620145548624e-05,
668
+ "loss": 0.4917,
669
+ "step": 1100
670
+ },
671
+ {
672
+ "epoch": 1.52,
673
+ "learning_rate": 6.853934919451357e-05,
674
+ "loss": 0.5223,
675
+ "step": 1110
676
+ },
677
+ {
678
+ "epoch": 1.53,
679
+ "learning_rate": 6.488628279764369e-05,
680
+ "loss": 0.4687,
681
+ "step": 1120
682
+ },
683
+ {
684
+ "epoch": 1.55,
685
+ "learning_rate": 6.131868899598503e-05,
686
+ "loss": 0.4788,
687
+ "step": 1130
688
+ },
689
+ {
690
+ "epoch": 1.56,
691
+ "learning_rate": 5.783821505536696e-05,
692
+ "loss": 0.472,
693
+ "step": 1140
694
+ },
695
+ {
696
+ "epoch": 1.57,
697
+ "learning_rate": 5.444646801574729e-05,
698
+ "loss": 0.4664,
699
+ "step": 1150
700
+ },
701
+ {
702
+ "epoch": 1.59,
703
+ "learning_rate": 5.114501394919316e-05,
704
+ "loss": 0.4547,
705
+ "step": 1160
706
+ },
707
+ {
708
+ "epoch": 1.6,
709
+ "learning_rate": 4.7935377236778654e-05,
710
+ "loss": 0.4927,
711
+ "step": 1170
712
+ },
713
+ {
714
+ "epoch": 1.61,
715
+ "learning_rate": 4.4819039864731625e-05,
716
+ "loss": 0.4814,
717
+ "step": 1180
718
+ },
719
+ {
720
+ "epoch": 1.63,
721
+ "learning_rate": 4.1797440740156204e-05,
722
+ "loss": 0.4681,
723
+ "step": 1190
724
+ },
725
+ {
726
+ "epoch": 1.64,
727
+ "learning_rate": 3.887197502664572e-05,
728
+ "loss": 0.4979,
729
+ "step": 1200
730
+ },
731
+ {
732
+ "epoch": 1.66,
733
+ "learning_rate": 3.6043993500093466e-05,
734
+ "loss": 0.4743,
735
+ "step": 1210
736
+ },
737
+ {
738
+ "epoch": 1.67,
739
+ "learning_rate": 3.3314801924998973e-05,
740
+ "loss": 0.4792,
741
+ "step": 1220
742
+ },
743
+ {
744
+ "epoch": 1.68,
745
+ "learning_rate": 3.068566045155682e-05,
746
+ "loss": 0.4791,
747
+ "step": 1230
748
+ },
749
+ {
750
+ "epoch": 1.7,
751
+ "learning_rate": 2.8157783033807365e-05,
752
+ "loss": 0.4846,
753
+ "step": 1240
754
+ },
755
+ {
756
+ "epoch": 1.71,
757
+ "learning_rate": 2.57323368691173e-05,
758
+ "loss": 0.4684,
759
+ "step": 1250
760
+ },
761
+ {
762
+ "epoch": 1.72,
763
+ "learning_rate": 2.3410441859249538e-05,
764
+ "loss": 0.4645,
765
+ "step": 1260
766
+ },
767
+ {
768
+ "epoch": 1.74,
769
+ "learning_rate": 2.1193170093270847e-05,
770
+ "loss": 0.51,
771
+ "step": 1270
772
+ },
773
+ {
774
+ "epoch": 1.75,
775
+ "learning_rate": 1.908154535253559e-05,
776
+ "loss": 0.4606,
777
+ "step": 1280
778
+ },
779
+ {
780
+ "epoch": 1.76,
781
+ "learning_rate": 1.707654263797534e-05,
782
+ "loss": 0.4711,
783
+ "step": 1290
784
+ },
785
+ {
786
+ "epoch": 1.78,
787
+ "learning_rate": 1.5179087719911138e-05,
788
+ "loss": 0.4523,
789
+ "step": 1300
790
+ },
791
+ {
792
+ "epoch": 1.79,
793
+ "learning_rate": 1.3390056710597648e-05,
794
+ "loss": 0.4997,
795
+ "step": 1310
796
+ },
797
+ {
798
+ "epoch": 1.81,
799
+ "learning_rate": 1.1710275659695596e-05,
800
+ "loss": 0.4838,
801
+ "step": 1320
802
+ },
803
+ {
804
+ "epoch": 1.82,
805
+ "learning_rate": 1.0140520172859518e-05,
806
+ "loss": 0.4573,
807
+ "step": 1330
808
+ },
809
+ {
810
+ "epoch": 1.83,
811
+ "learning_rate": 8.681515053617473e-06,
812
+ "loss": 0.4858,
813
+ "step": 1340
814
+ },
815
+ {
816
+ "epoch": 1.85,
817
+ "learning_rate": 7.333933968707046e-06,
818
+ "loss": 0.4855,
819
+ "step": 1350
820
+ },
821
+ {
822
+ "epoch": 1.86,
823
+ "learning_rate": 6.098399137023425e-06,
824
+ "loss": 0.462,
825
+ "step": 1360
826
+ },
827
+ {
828
+ "epoch": 1.87,
829
+ "learning_rate": 4.975481042321816e-06,
830
+ "loss": 0.4766,
831
+ "step": 1370
832
+ },
833
+ {
834
+ "epoch": 1.89,
835
+ "learning_rate": 3.965698169808018e-06,
836
+ "loss": 0.5164,
837
+ "step": 1380
838
+ },
839
+ {
840
+ "epoch": 1.9,
841
+ "learning_rate": 3.069516766738195e-06,
842
+ "loss": 0.472,
843
+ "step": 1390
844
+ },
845
+ {
846
+ "epoch": 1.92,
847
+ "learning_rate": 2.2873506271382515e-06,
848
+ "loss": 0.4915,
849
+ "step": 1400
850
+ },
851
+ {
852
+ "epoch": 1.93,
853
+ "learning_rate": 1.6195609007428057e-06,
854
+ "loss": 0.4754,
855
+ "step": 1410
856
+ },
857
+ {
858
+ "epoch": 1.94,
859
+ "learning_rate": 1.066455926241383e-06,
860
+ "loss": 0.4655,
861
+ "step": 1420
862
+ },
863
+ {
864
+ "epoch": 1.96,
865
+ "learning_rate": 6.282910889090499e-07,
866
+ "loss": 0.4948,
867
+ "step": 1430
868
+ },
869
+ {
870
+ "epoch": 1.97,
871
+ "learning_rate": 3.052687026874612e-07,
872
+ "loss": 0.501,
873
+ "step": 1440
874
+ },
875
+ {
876
+ "epoch": 1.98,
877
+ "learning_rate": 9.753791677027945e-08,
878
+ "loss": 0.473,
879
+ "step": 1450
880
+ },
881
+ {
882
+ "epoch": 2.0,
883
+ "learning_rate": 5.194646736622621e-09,
884
+ "loss": 0.4731,
885
+ "step": 1460
886
+ },
887
+ {
888
+ "epoch": 2.0,
889
+ "step": 1462,
890
+ "total_flos": 3.404194232389337e+18,
891
+ "train_loss": 0.49735359360353076,
892
+ "train_runtime": 25276.0399,
893
+ "train_samples_per_second": 3.7,
894
+ "train_steps_per_second": 0.058
895
+ }
896
+ ],
897
+ "logging_steps": 10,
898
+ "max_steps": 1462,
899
+ "num_train_epochs": 2,
900
+ "save_steps": 150,
901
+ "total_flos": 3.404194232389337e+18,
902
+ "trial_name": null,
903
+ "trial_params": null
904
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9fcd955228bbc27a040d82eb946b796e1089c67ed98f185a361c324f2008c49a
3
+ size 4219