Sen Fang commited on
Commit
8b75f58
·
1 Parent(s): c0c975e
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 sha256:54a40ea5390b7e70b58f71020391dd1fb3913f5ed08f95da6ddf408a69e1c074
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 +1,850 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 2.0,
5
+ "eval_steps": 10.0,
6
+ "global_step": 1374,
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": 0.0004999346542461473,
14
+ "loss": 0.4698,
15
+ "step": 10
16
+ },
17
+ {
18
+ "epoch": 0.03,
19
+ "learning_rate": 0.0004997386511451298,
20
+ "loss": 0.2312,
21
+ "step": 20
22
+ },
23
+ {
24
+ "epoch": 0.04,
25
+ "learning_rate": 0.0004994120931607106,
26
+ "loss": 0.1855,
27
+ "step": 30
28
+ },
29
+ {
30
+ "epoch": 0.06,
31
+ "learning_rate": 0.0004989551510063109,
32
+ "loss": 0.1778,
33
+ "step": 40
34
+ },
35
+ {
36
+ "epoch": 0.07,
37
+ "learning_rate": 0.0004983680635557671,
38
+ "loss": 0.1728,
39
+ "step": 50
40
+ },
41
+ {
42
+ "epoch": 0.09,
43
+ "learning_rate": 0.0004976511377184557,
44
+ "loss": 0.182,
45
+ "step": 60
46
+ },
47
+ {
48
+ "epoch": 0.1,
49
+ "learning_rate": 0.0004968047482788508,
50
+ "loss": 0.1752,
51
+ "step": 70
52
+ },
53
+ {
54
+ "epoch": 0.12,
55
+ "learning_rate": 0.0004958293377006004,
56
+ "loss": 0.1742,
57
+ "step": 80
58
+ },
59
+ {
60
+ "epoch": 0.13,
61
+ "learning_rate": 0.0004947254158952209,
62
+ "loss": 0.1686,
63
+ "step": 90
64
+ },
65
+ {
66
+ "epoch": 0.15,
67
+ "learning_rate": 0.0004934935599555328,
68
+ "loss": 0.1674,
69
+ "step": 100
70
+ },
71
+ {
72
+ "epoch": 0.16,
73
+ "learning_rate": 0.0004921344138539761,
74
+ "loss": 0.1665,
75
+ "step": 110
76
+ },
77
+ {
78
+ "epoch": 0.17,
79
+ "learning_rate": 0.000490648688105964,
80
+ "loss": 0.1673,
81
+ "step": 120
82
+ },
83
+ {
84
+ "epoch": 0.19,
85
+ "learning_rate": 0.0004890371593984484,
86
+ "loss": 0.1706,
87
+ "step": 130
88
+ },
89
+ {
90
+ "epoch": 0.2,
91
+ "learning_rate": 0.0004873006701838952,
92
+ "loss": 0.1674,
93
+ "step": 140
94
+ },
95
+ {
96
+ "epoch": 0.22,
97
+ "learning_rate": 0.0004854401282398789,
98
+ "loss": 0.1662,
99
+ "step": 150
100
+ },
101
+ {
102
+ "epoch": 0.23,
103
+ "learning_rate": 0.0004834565061945266,
104
+ "loss": 0.1615,
105
+ "step": 160
106
+ },
107
+ {
108
+ "epoch": 0.25,
109
+ "learning_rate": 0.0004813508410180617,
110
+ "loss": 0.1643,
111
+ "step": 170
112
+ },
113
+ {
114
+ "epoch": 0.26,
115
+ "learning_rate": 0.00047912423348071056,
116
+ "loss": 0.1634,
117
+ "step": 180
118
+ },
119
+ {
120
+ "epoch": 0.28,
121
+ "learning_rate": 0.00047677784757725784,
122
+ "loss": 0.1552,
123
+ "step": 190
124
+ },
125
+ {
126
+ "epoch": 0.29,
127
+ "learning_rate": 0.00047431290991854904,
128
+ "loss": 0.1605,
129
+ "step": 200
130
+ },
131
+ {
132
+ "epoch": 0.31,
133
+ "learning_rate": 0.0004717307090902601,
134
+ "loss": 0.1523,
135
+ "step": 210
136
+ },
137
+ {
138
+ "epoch": 0.32,
139
+ "learning_rate": 0.00046903259497926896,
140
+ "loss": 0.1603,
141
+ "step": 220
142
+ },
143
+ {
144
+ "epoch": 0.33,
145
+ "learning_rate": 0.00046621997806798004,
146
+ "loss": 0.1599,
147
+ "step": 230
148
+ },
149
+ {
150
+ "epoch": 0.35,
151
+ "learning_rate": 0.00046329432869697236,
152
+ "loss": 0.1624,
153
+ "step": 240
154
+ },
155
+ {
156
+ "epoch": 0.36,
157
+ "learning_rate": 0.0004602571762963551,
158
+ "loss": 0.1669,
159
+ "step": 250
160
+ },
161
+ {
162
+ "epoch": 0.38,
163
+ "learning_rate": 0.0004571101085862337,
164
+ "loss": 0.1579,
165
+ "step": 260
166
+ },
167
+ {
168
+ "epoch": 0.39,
169
+ "learning_rate": 0.0004538547707467038,
170
+ "loss": 0.1665,
171
+ "step": 270
172
+ },
173
+ {
174
+ "epoch": 0.41,
175
+ "learning_rate": 0.0004504928645578067,
176
+ "loss": 0.1632,
177
+ "step": 280
178
+ },
179
+ {
180
+ "epoch": 0.42,
181
+ "learning_rate": 0.0004470261475098967,
182
+ "loss": 0.1571,
183
+ "step": 290
184
+ },
185
+ {
186
+ "epoch": 0.44,
187
+ "learning_rate": 0.00044345643188488506,
188
+ "loss": 0.154,
189
+ "step": 300
190
+ },
191
+ {
192
+ "epoch": 0.45,
193
+ "learning_rate": 0.00043978558380884013,
194
+ "loss": 0.1504,
195
+ "step": 310
196
+ },
197
+ {
198
+ "epoch": 0.47,
199
+ "learning_rate": 0.00043601552227644035,
200
+ "loss": 0.1634,
201
+ "step": 320
202
+ },
203
+ {
204
+ "epoch": 0.48,
205
+ "learning_rate": 0.00043214821814778904,
206
+ "loss": 0.147,
207
+ "step": 330
208
+ },
209
+ {
210
+ "epoch": 0.49,
211
+ "learning_rate": 0.0004281856931181155,
212
+ "loss": 0.1493,
213
+ "step": 340
214
+ },
215
+ {
216
+ "epoch": 0.51,
217
+ "learning_rate": 0.0004241300186609015,
218
+ "loss": 0.1543,
219
+ "step": 350
220
+ },
221
+ {
222
+ "epoch": 0.52,
223
+ "learning_rate": 0.0004199833149449853,
224
+ "loss": 0.1519,
225
+ "step": 360
226
+ },
227
+ {
228
+ "epoch": 0.54,
229
+ "learning_rate": 0.00041574774972620956,
230
+ "loss": 0.152,
231
+ "step": 370
232
+ },
233
+ {
234
+ "epoch": 0.55,
235
+ "learning_rate": 0.00041142553721419184,
236
+ "loss": 0.155,
237
+ "step": 380
238
+ },
239
+ {
240
+ "epoch": 0.57,
241
+ "learning_rate": 0.0004070189369148116,
242
+ "loss": 0.1585,
243
+ "step": 390
244
+ },
245
+ {
246
+ "epoch": 0.58,
247
+ "learning_rate": 0.0004025302524490166,
248
+ "loss": 0.1518,
249
+ "step": 400
250
+ },
251
+ {
252
+ "epoch": 0.6,
253
+ "learning_rate": 0.0003979618303485687,
254
+ "loss": 0.1536,
255
+ "step": 410
256
+ },
257
+ {
258
+ "epoch": 0.61,
259
+ "learning_rate": 0.0003933160588293564,
260
+ "loss": 0.1477,
261
+ "step": 420
262
+ },
263
+ {
264
+ "epoch": 0.63,
265
+ "learning_rate": 0.00038859536654291705,
266
+ "loss": 0.1494,
267
+ "step": 430
268
+ },
269
+ {
270
+ "epoch": 0.64,
271
+ "learning_rate": 0.00038380222130681986,
272
+ "loss": 0.1578,
273
+ "step": 440
274
+ },
275
+ {
276
+ "epoch": 0.66,
277
+ "learning_rate": 0.00037893912881457507,
278
+ "loss": 0.1533,
279
+ "step": 450
280
+ },
281
+ {
282
+ "epoch": 0.67,
283
+ "learning_rate": 0.00037400863132574223,
284
+ "loss": 0.1511,
285
+ "step": 460
286
+ },
287
+ {
288
+ "epoch": 0.68,
289
+ "learning_rate": 0.0003690133063369238,
290
+ "loss": 0.1548,
291
+ "step": 470
292
+ },
293
+ {
294
+ "epoch": 0.7,
295
+ "learning_rate": 0.00036395576523433673,
296
+ "loss": 0.1459,
297
+ "step": 480
298
+ },
299
+ {
300
+ "epoch": 0.71,
301
+ "learning_rate": 0.0003588386519286688,
302
+ "loss": 0.1518,
303
+ "step": 490
304
+ },
305
+ {
306
+ "epoch": 0.73,
307
+ "learning_rate": 0.00035366464147293223,
308
+ "loss": 0.1529,
309
+ "step": 500
310
+ },
311
+ {
312
+ "epoch": 0.74,
313
+ "learning_rate": 0.0003484364386640365,
314
+ "loss": 0.1608,
315
+ "step": 510
316
+ },
317
+ {
318
+ "epoch": 0.76,
319
+ "learning_rate": 0.00034315677662881205,
320
+ "loss": 0.1616,
321
+ "step": 520
322
+ },
323
+ {
324
+ "epoch": 0.77,
325
+ "learning_rate": 0.0003378284153952254,
326
+ "loss": 0.1549,
327
+ "step": 530
328
+ },
329
+ {
330
+ "epoch": 0.79,
331
+ "learning_rate": 0.00033245414044952927,
332
+ "loss": 0.155,
333
+ "step": 540
334
+ },
335
+ {
336
+ "epoch": 0.8,
337
+ "learning_rate": 0.00032703676128010553,
338
+ "loss": 0.1514,
339
+ "step": 550
340
+ },
341
+ {
342
+ "epoch": 0.82,
343
+ "learning_rate": 0.0003215791099087601,
344
+ "loss": 0.1462,
345
+ "step": 560
346
+ },
347
+ {
348
+ "epoch": 0.83,
349
+ "learning_rate": 0.0003160840394102379,
350
+ "loss": 0.1528,
351
+ "step": 570
352
+ },
353
+ {
354
+ "epoch": 0.84,
355
+ "learning_rate": 0.0003105544224207326,
356
+ "loss": 0.1517,
357
+ "step": 580
358
+ },
359
+ {
360
+ "epoch": 0.86,
361
+ "learning_rate": 0.00030499314963616987,
362
+ "loss": 0.1532,
363
+ "step": 590
364
+ },
365
+ {
366
+ "epoch": 0.87,
367
+ "learning_rate": 0.0002994031283010493,
368
+ "loss": 0.155,
369
+ "step": 600
370
+ },
371
+ {
372
+ "epoch": 0.89,
373
+ "learning_rate": 0.00029378728068863683,
374
+ "loss": 0.1499,
375
+ "step": 610
376
+ },
377
+ {
378
+ "epoch": 0.9,
379
+ "learning_rate": 0.00028814854257329825,
380
+ "loss": 0.1591,
381
+ "step": 620
382
+ },
383
+ {
384
+ "epoch": 0.92,
385
+ "learning_rate": 0.00028248986169577693,
386
+ "loss": 0.1501,
387
+ "step": 630
388
+ },
389
+ {
390
+ "epoch": 0.93,
391
+ "learning_rate": 0.00027681419622221514,
392
+ "loss": 0.1449,
393
+ "step": 640
394
+ },
395
+ {
396
+ "epoch": 0.95,
397
+ "learning_rate": 0.00027112451319772447,
398
+ "loss": 0.1423,
399
+ "step": 650
400
+ },
401
+ {
402
+ "epoch": 0.96,
403
+ "learning_rate": 0.00026542378699531645,
404
+ "loss": 0.1462,
405
+ "step": 660
406
+ },
407
+ {
408
+ "epoch": 0.98,
409
+ "learning_rate": 0.00025971499776100064,
410
+ "loss": 0.1469,
411
+ "step": 670
412
+ },
413
+ {
414
+ "epoch": 0.99,
415
+ "learning_rate": 0.00025400112985586587,
416
+ "loss": 0.1418,
417
+ "step": 680
418
+ },
419
+ {
420
+ "epoch": 1.0,
421
+ "learning_rate": 0.00024828517029595767,
422
+ "loss": 0.1397,
423
+ "step": 690
424
+ },
425
+ {
426
+ "epoch": 1.02,
427
+ "learning_rate": 0.00024257010719076723,
428
+ "loss": 0.142,
429
+ "step": 700
430
+ },
431
+ {
432
+ "epoch": 1.03,
433
+ "learning_rate": 0.00023685892818115019,
434
+ "loss": 0.1395,
435
+ "step": 710
436
+ },
437
+ {
438
+ "epoch": 1.05,
439
+ "learning_rate": 0.00023115461887748858,
440
+ "loss": 0.1392,
441
+ "step": 720
442
+ },
443
+ {
444
+ "epoch": 1.06,
445
+ "learning_rate": 0.00022546016129891568,
446
+ "loss": 0.1458,
447
+ "step": 730
448
+ },
449
+ {
450
+ "epoch": 1.08,
451
+ "learning_rate": 0.0002197785323144176,
452
+ "loss": 0.1341,
453
+ "step": 740
454
+ },
455
+ {
456
+ "epoch": 1.09,
457
+ "learning_rate": 0.00021411270208662698,
458
+ "loss": 0.1475,
459
+ "step": 750
460
+ },
461
+ {
462
+ "epoch": 1.11,
463
+ "learning_rate": 0.00020846563251912355,
464
+ "loss": 0.149,
465
+ "step": 760
466
+ },
467
+ {
468
+ "epoch": 1.12,
469
+ "learning_rate": 0.00020284027570805074,
470
+ "loss": 0.1399,
471
+ "step": 770
472
+ },
473
+ {
474
+ "epoch": 1.14,
475
+ "learning_rate": 0.00019723957239886066,
476
+ "loss": 0.1455,
477
+ "step": 780
478
+ },
479
+ {
480
+ "epoch": 1.15,
481
+ "learning_rate": 0.00019166645044899205,
482
+ "loss": 0.147,
483
+ "step": 790
484
+ },
485
+ {
486
+ "epoch": 1.16,
487
+ "learning_rate": 0.00018612382329728588,
488
+ "loss": 0.1499,
489
+ "step": 800
490
+ },
491
+ {
492
+ "epoch": 1.18,
493
+ "learning_rate": 0.00018061458844093858,
494
+ "loss": 0.1353,
495
+ "step": 810
496
+ },
497
+ {
498
+ "epoch": 1.19,
499
+ "learning_rate": 0.0001751416259207889,
500
+ "loss": 0.1433,
501
+ "step": 820
502
+ },
503
+ {
504
+ "epoch": 1.21,
505
+ "learning_rate": 0.00016970779681573046,
506
+ "loss": 0.1487,
507
+ "step": 830
508
+ },
509
+ {
510
+ "epoch": 1.22,
511
+ "learning_rate": 0.00016431594174703645,
512
+ "loss": 0.1459,
513
+ "step": 840
514
+ },
515
+ {
516
+ "epoch": 1.24,
517
+ "learning_rate": 0.00015896887939338,
518
+ "loss": 0.153,
519
+ "step": 850
520
+ },
521
+ {
522
+ "epoch": 1.25,
523
+ "learning_rate": 0.0001536694050173242,
524
+ "loss": 0.1411,
525
+ "step": 860
526
+ },
527
+ {
528
+ "epoch": 1.27,
529
+ "learning_rate": 0.00014842028900405421,
530
+ "loss": 0.1426,
531
+ "step": 870
532
+ },
533
+ {
534
+ "epoch": 1.28,
535
+ "learning_rate": 0.00014322427541311347,
536
+ "loss": 0.1455,
537
+ "step": 880
538
+ },
539
+ {
540
+ "epoch": 1.3,
541
+ "learning_rate": 0.0001380840805439031,
542
+ "loss": 0.1485,
543
+ "step": 890
544
+ },
545
+ {
546
+ "epoch": 1.31,
547
+ "learning_rate": 0.00013300239151569252,
548
+ "loss": 0.152,
549
+ "step": 900
550
+ },
551
+ {
552
+ "epoch": 1.32,
553
+ "learning_rate": 0.00012798186486288483,
554
+ "loss": 0.145,
555
+ "step": 910
556
+ },
557
+ {
558
+ "epoch": 1.34,
559
+ "learning_rate": 0.0001230251251462708,
560
+ "loss": 0.1311,
561
+ "step": 920
562
+ },
563
+ {
564
+ "epoch": 1.35,
565
+ "learning_rate": 0.00011813476358099823,
566
+ "loss": 0.1411,
567
+ "step": 930
568
+ },
569
+ {
570
+ "epoch": 1.37,
571
+ "learning_rate": 0.0001133133366819717,
572
+ "loss": 0.1384,
573
+ "step": 940
574
+ },
575
+ {
576
+ "epoch": 1.38,
577
+ "learning_rate": 0.00010856336492739407,
578
+ "loss": 0.1455,
579
+ "step": 950
580
+ },
581
+ {
582
+ "epoch": 1.4,
583
+ "learning_rate": 0.00010388733144114604,
584
+ "loss": 0.1307,
585
+ "step": 960
586
+ },
587
+ {
588
+ "epoch": 1.41,
589
+ "learning_rate": 9.928768069469313e-05,
590
+ "loss": 0.1404,
591
+ "step": 970
592
+ },
593
+ {
594
+ "epoch": 1.43,
595
+ "learning_rate": 9.476681722919939e-05,
596
+ "loss": 0.1333,
597
+ "step": 980
598
+ },
599
+ {
600
+ "epoch": 1.44,
601
+ "learning_rate": 9.032710439851444e-05,
602
+ "loss": 0.144,
603
+ "step": 990
604
+ },
605
+ {
606
+ "epoch": 1.46,
607
+ "learning_rate": 8.597086313369271e-05,
608
+ "loss": 0.1499,
609
+ "step": 1000
610
+ },
611
+ {
612
+ "epoch": 1.47,
613
+ "learning_rate": 8.170037072968967e-05,
614
+ "loss": 0.1471,
615
+ "step": 1010
616
+ },
617
+ {
618
+ "epoch": 1.48,
619
+ "learning_rate": 7.751785965486893e-05,
620
+ "loss": 0.1312,
621
+ "step": 1020
622
+ },
623
+ {
624
+ "epoch": 1.5,
625
+ "learning_rate": 7.342551638394385e-05,
626
+ "loss": 0.1466,
627
+ "step": 1030
628
+ },
629
+ {
630
+ "epoch": 1.51,
631
+ "learning_rate": 6.942548025496312e-05,
632
+ "loss": 0.147,
633
+ "step": 1040
634
+ },
635
+ {
636
+ "epoch": 1.53,
637
+ "learning_rate": 6.551984235093692e-05,
638
+ "loss": 0.1375,
639
+ "step": 1050
640
+ },
641
+ {
642
+ "epoch": 1.54,
643
+ "learning_rate": 6.171064440669014e-05,
644
+ "loss": 0.1398,
645
+ "step": 1060
646
+ },
647
+ {
648
+ "epoch": 1.56,
649
+ "learning_rate": 5.7999877741512744e-05,
650
+ "loss": 0.1337,
651
+ "step": 1070
652
+ },
653
+ {
654
+ "epoch": 1.57,
655
+ "learning_rate": 5.438948221816559e-05,
656
+ "loss": 0.1351,
657
+ "step": 1080
658
+ },
659
+ {
660
+ "epoch": 1.59,
661
+ "learning_rate": 5.088134522878601e-05,
662
+ "loss": 0.1417,
663
+ "step": 1090
664
+ },
665
+ {
666
+ "epoch": 1.6,
667
+ "learning_rate": 4.74773007082237e-05,
668
+ "loss": 0.14,
669
+ "step": 1100
670
+ },
671
+ {
672
+ "epoch": 1.62,
673
+ "learning_rate": 4.417912817532133e-05,
674
+ "loss": 0.1423,
675
+ "step": 1110
676
+ },
677
+ {
678
+ "epoch": 1.63,
679
+ "learning_rate": 4.098855180264285e-05,
680
+ "loss": 0.1327,
681
+ "step": 1120
682
+ },
683
+ {
684
+ "epoch": 1.64,
685
+ "learning_rate": 3.7907239515134697e-05,
686
+ "loss": 0.1405,
687
+ "step": 1130
688
+ },
689
+ {
690
+ "epoch": 1.66,
691
+ "learning_rate": 3.493680211819103e-05,
692
+ "loss": 0.1444,
693
+ "step": 1140
694
+ },
695
+ {
696
+ "epoch": 1.67,
697
+ "learning_rate": 3.207879245557976e-05,
698
+ "loss": 0.1421,
699
+ "step": 1150
700
+ },
701
+ {
702
+ "epoch": 1.69,
703
+ "learning_rate": 2.933470459766821e-05,
704
+ "loss": 0.1389,
705
+ "step": 1160
706
+ },
707
+ {
708
+ "epoch": 1.7,
709
+ "learning_rate": 2.6705973060374117e-05,
710
+ "loss": 0.1457,
711
+ "step": 1170
712
+ },
713
+ {
714
+ "epoch": 1.72,
715
+ "learning_rate": 2.4193972055249343e-05,
716
+ "loss": 0.1451,
717
+ "step": 1180
718
+ },
719
+ {
720
+ "epoch": 1.73,
721
+ "learning_rate": 2.180001477108867e-05,
722
+ "loss": 0.1343,
723
+ "step": 1190
724
+ },
725
+ {
726
+ "epoch": 1.75,
727
+ "learning_rate": 1.9525352687439546e-05,
728
+ "loss": 0.1267,
729
+ "step": 1200
730
+ },
731
+ {
732
+ "epoch": 1.76,
733
+ "learning_rate": 1.737117492037091e-05,
734
+ "loss": 0.1364,
735
+ "step": 1210
736
+ },
737
+ {
738
+ "epoch": 1.78,
739
+ "learning_rate": 1.533860760084374e-05,
740
+ "loss": 0.1511,
741
+ "step": 1220
742
+ },
743
+ {
744
+ "epoch": 1.79,
745
+ "learning_rate": 1.3428713286008004e-05,
746
+ "loss": 0.1418,
747
+ "step": 1230
748
+ },
749
+ {
750
+ "epoch": 1.8,
751
+ "learning_rate": 1.1642490403733995e-05,
752
+ "loss": 0.1403,
753
+ "step": 1240
754
+ },
755
+ {
756
+ "epoch": 1.82,
757
+ "learning_rate": 9.980872730667972e-06,
758
+ "loss": 0.1433,
759
+ "step": 1250
760
+ },
761
+ {
762
+ "epoch": 1.83,
763
+ "learning_rate": 8.444728904085736e-06,
764
+ "loss": 0.1491,
765
+ "step": 1260
766
+ },
767
+ {
768
+ "epoch": 1.85,
769
+ "learning_rate": 7.0348619677982066e-06,
770
+ "loss": 0.1332,
771
+ "step": 1270
772
+ },
773
+ {
774
+ "epoch": 1.86,
775
+ "learning_rate": 5.752008952347648e-06,
776
+ "loss": 0.1351,
777
+ "step": 1280
778
+ },
779
+ {
780
+ "epoch": 1.88,
781
+ "learning_rate": 4.596840489713095e-06,
782
+ "loss": 0.1471,
783
+ "step": 1290
784
+ },
785
+ {
786
+ "epoch": 1.89,
787
+ "learning_rate": 3.5699604627266734e-06,
788
+ "loss": 0.1572,
789
+ "step": 1300
790
+ },
791
+ {
792
+ "epoch": 1.91,
793
+ "learning_rate": 2.6719056893842463e-06,
794
+ "loss": 0.1436,
795
+ "step": 1310
796
+ },
797
+ {
798
+ "epoch": 1.92,
799
+ "learning_rate": 1.9031456422151373e-06,
800
+ "loss": 0.1482,
801
+ "step": 1320
802
+ },
803
+ {
804
+ "epoch": 1.94,
805
+ "learning_rate": 1.2640822028578414e-06,
806
+ "loss": 0.1386,
807
+ "step": 1330
808
+ },
809
+ {
810
+ "epoch": 1.95,
811
+ "learning_rate": 7.550494519699968e-07,
812
+ "loss": 0.1294,
813
+ "step": 1340
814
+ },
815
+ {
816
+ "epoch": 1.97,
817
+ "learning_rate": 3.7631349458230877e-07,
818
+ "loss": 0.1439,
819
+ "step": 1350
820
+ },
821
+ {
822
+ "epoch": 1.98,
823
+ "learning_rate": 1.280723209880208e-07,
824
+ "loss": 0.135,
825
+ "step": 1360
826
+ },
827
+ {
828
+ "epoch": 1.99,
829
+ "learning_rate": 1.0455703240069836e-08,
830
+ "loss": 0.1345,
831
+ "step": 1370
832
+ },
833
+ {
834
+ "epoch": 2.0,
835
+ "step": 1374,
836
+ "total_flos": 3.2608758262451405e+18,
837
+ "train_loss": 0.1525881653095784,
838
+ "train_runtime": 23876.5124,
839
+ "train_samples_per_second": 3.681,
840
+ "train_steps_per_second": 0.058
841
+ }
842
+ ],
843
+ "logging_steps": 10,
844
+ "max_steps": 1374,
845
+ "num_train_epochs": 2,
846
+ "save_steps": 150,
847
+ "total_flos": 3.2608758262451405e+18,
848
+ "trial_name": null,
849
+ "trial_params": null
850
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:14996c320a649dbed236c1e4101b14b3dcb2d54952bfbad1a1f79b2939c77455
3
+ size 4219