ryanmarten commited on
Commit
4a9cad7
·
verified ·
1 Parent(s): 71288b8

End of training

Browse files
Files changed (5) hide show
  1. README.md +2 -1
  2. all_results.json +8 -0
  3. train_results.json +8 -0
  4. trainer_state.json +861 -0
  5. training_loss.png +0 -0
README.md CHANGED
@@ -4,6 +4,7 @@ license: apache-2.0
4
  base_model: Qwen/Qwen2.5-7B-Instruct
5
  tags:
6
  - llama-factory
 
7
  - generated_from_trainer
8
  model-index:
9
  - name: d1_code_longest_0.3k
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
15
 
16
  # d1_code_longest_0.3k
17
 
18
- This model is a fine-tuned version of [Qwen/Qwen2.5-7B-Instruct](https://huggingface.co/Qwen/Qwen2.5-7B-Instruct) on an unknown dataset.
19
 
20
  ## Model description
21
 
 
4
  base_model: Qwen/Qwen2.5-7B-Instruct
5
  tags:
6
  - llama-factory
7
+ - full
8
  - generated_from_trainer
9
  model-index:
10
  - name: d1_code_longest_0.3k
 
16
 
17
  # d1_code_longest_0.3k
18
 
19
+ This model is a fine-tuned version of [Qwen/Qwen2.5-7B-Instruct](https://huggingface.co/Qwen/Qwen2.5-7B-Instruct) on the mlfoundations-dev/d1_code_longest_0.3k dataset.
20
 
21
  ## Model description
22
 
all_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 12.30379746835443,
3
+ "total_flos": 1.5758524630748365e+17,
4
+ "train_loss": 0.5164603565493201,
5
+ "train_runtime": 5508.0362,
6
+ "train_samples_per_second": 0.746,
7
+ "train_steps_per_second": 0.021
8
+ }
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 12.30379746835443,
3
+ "total_flos": 1.5758524630748365e+17,
4
+ "train_loss": 0.5164603565493201,
5
+ "train_runtime": 5508.0362,
6
+ "train_samples_per_second": 0.746,
7
+ "train_steps_per_second": 0.021
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,861 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 12.30379746835443,
5
+ "eval_steps": 500,
6
+ "global_step": 117,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.10126582278481013,
13
+ "grad_norm": 6.166710783670563,
14
+ "learning_rate": 8.333333333333333e-07,
15
+ "loss": 0.9829,
16
+ "step": 1
17
+ },
18
+ {
19
+ "epoch": 0.20253164556962025,
20
+ "grad_norm": 6.361623319742254,
21
+ "learning_rate": 1.6666666666666667e-06,
22
+ "loss": 1.0447,
23
+ "step": 2
24
+ },
25
+ {
26
+ "epoch": 0.3037974683544304,
27
+ "grad_norm": 6.616695944493182,
28
+ "learning_rate": 2.5e-06,
29
+ "loss": 1.0673,
30
+ "step": 3
31
+ },
32
+ {
33
+ "epoch": 0.4050632911392405,
34
+ "grad_norm": 5.884116222556751,
35
+ "learning_rate": 3.3333333333333333e-06,
36
+ "loss": 1.0159,
37
+ "step": 4
38
+ },
39
+ {
40
+ "epoch": 0.5063291139240507,
41
+ "grad_norm": 4.530788873812888,
42
+ "learning_rate": 4.166666666666667e-06,
43
+ "loss": 0.9789,
44
+ "step": 5
45
+ },
46
+ {
47
+ "epoch": 0.6075949367088608,
48
+ "grad_norm": 2.7528904756405916,
49
+ "learning_rate": 5e-06,
50
+ "loss": 0.946,
51
+ "step": 6
52
+ },
53
+ {
54
+ "epoch": 0.7088607594936709,
55
+ "grad_norm": 2.4860173529738514,
56
+ "learning_rate": 5.833333333333334e-06,
57
+ "loss": 0.9604,
58
+ "step": 7
59
+ },
60
+ {
61
+ "epoch": 0.810126582278481,
62
+ "grad_norm": 3.7622258998085596,
63
+ "learning_rate": 6.666666666666667e-06,
64
+ "loss": 0.9377,
65
+ "step": 8
66
+ },
67
+ {
68
+ "epoch": 0.9113924050632911,
69
+ "grad_norm": 4.307665757929639,
70
+ "learning_rate": 7.500000000000001e-06,
71
+ "loss": 0.9678,
72
+ "step": 9
73
+ },
74
+ {
75
+ "epoch": 1.0506329113924051,
76
+ "grad_norm": 4.0395954346652045,
77
+ "learning_rate": 8.333333333333334e-06,
78
+ "loss": 0.9771,
79
+ "step": 10
80
+ },
81
+ {
82
+ "epoch": 1.1518987341772151,
83
+ "grad_norm": 3.3957279512382406,
84
+ "learning_rate": 9.166666666666666e-06,
85
+ "loss": 0.8963,
86
+ "step": 11
87
+ },
88
+ {
89
+ "epoch": 1.2531645569620253,
90
+ "grad_norm": 3.1111142570953945,
91
+ "learning_rate": 1e-05,
92
+ "loss": 0.9014,
93
+ "step": 12
94
+ },
95
+ {
96
+ "epoch": 1.3544303797468356,
97
+ "grad_norm": 2.364341875085718,
98
+ "learning_rate": 9.997762161417517e-06,
99
+ "loss": 0.8638,
100
+ "step": 13
101
+ },
102
+ {
103
+ "epoch": 1.4556962025316456,
104
+ "grad_norm": 1.5851033336379674,
105
+ "learning_rate": 9.991050648838676e-06,
106
+ "loss": 0.8325,
107
+ "step": 14
108
+ },
109
+ {
110
+ "epoch": 1.5569620253164556,
111
+ "grad_norm": 1.4980198864988055,
112
+ "learning_rate": 9.979871469976197e-06,
113
+ "loss": 0.7846,
114
+ "step": 15
115
+ },
116
+ {
117
+ "epoch": 1.6582278481012658,
118
+ "grad_norm": 1.4584988306087168,
119
+ "learning_rate": 9.964234631709188e-06,
120
+ "loss": 0.7994,
121
+ "step": 16
122
+ },
123
+ {
124
+ "epoch": 1.759493670886076,
125
+ "grad_norm": 1.3260698118659737,
126
+ "learning_rate": 9.944154131125643e-06,
127
+ "loss": 0.8356,
128
+ "step": 17
129
+ },
130
+ {
131
+ "epoch": 1.8607594936708862,
132
+ "grad_norm": 1.196582916094783,
133
+ "learning_rate": 9.91964794299315e-06,
134
+ "loss": 0.7772,
135
+ "step": 18
136
+ },
137
+ {
138
+ "epoch": 1.9620253164556962,
139
+ "grad_norm": 1.043300550372986,
140
+ "learning_rate": 9.890738003669029e-06,
141
+ "loss": 0.7712,
142
+ "step": 19
143
+ },
144
+ {
145
+ "epoch": 2.1012658227848102,
146
+ "grad_norm": 0.9881112255272267,
147
+ "learning_rate": 9.857450191464337e-06,
148
+ "loss": 0.7684,
149
+ "step": 20
150
+ },
151
+ {
152
+ "epoch": 2.2025316455696204,
153
+ "grad_norm": 0.9769809317612732,
154
+ "learning_rate": 9.819814303479268e-06,
155
+ "loss": 0.7519,
156
+ "step": 21
157
+ },
158
+ {
159
+ "epoch": 2.3037974683544302,
160
+ "grad_norm": 0.8897204736798004,
161
+ "learning_rate": 9.777864028930705e-06,
162
+ "loss": 0.7214,
163
+ "step": 22
164
+ },
165
+ {
166
+ "epoch": 2.4050632911392404,
167
+ "grad_norm": 1.1156867024494346,
168
+ "learning_rate": 9.731636918995821e-06,
169
+ "loss": 0.7218,
170
+ "step": 23
171
+ },
172
+ {
173
+ "epoch": 2.5063291139240507,
174
+ "grad_norm": 0.8249350363259335,
175
+ "learning_rate": 9.681174353198687e-06,
176
+ "loss": 0.7137,
177
+ "step": 24
178
+ },
179
+ {
180
+ "epoch": 2.607594936708861,
181
+ "grad_norm": 0.8123049973633097,
182
+ "learning_rate": 9.626521502369984e-06,
183
+ "loss": 0.698,
184
+ "step": 25
185
+ },
186
+ {
187
+ "epoch": 2.708860759493671,
188
+ "grad_norm": 0.7491557925370659,
189
+ "learning_rate": 9.567727288213005e-06,
190
+ "loss": 0.7091,
191
+ "step": 26
192
+ },
193
+ {
194
+ "epoch": 2.810126582278481,
195
+ "grad_norm": 0.8735150128201048,
196
+ "learning_rate": 9.504844339512096e-06,
197
+ "loss": 0.692,
198
+ "step": 27
199
+ },
200
+ {
201
+ "epoch": 2.911392405063291,
202
+ "grad_norm": 0.7925464600601956,
203
+ "learning_rate": 9.437928945022772e-06,
204
+ "loss": 0.7143,
205
+ "step": 28
206
+ },
207
+ {
208
+ "epoch": 3.050632911392405,
209
+ "grad_norm": 0.7165551367920177,
210
+ "learning_rate": 9.36704100308565e-06,
211
+ "loss": 0.6815,
212
+ "step": 29
213
+ },
214
+ {
215
+ "epoch": 3.151898734177215,
216
+ "grad_norm": 0.7216095752807774,
217
+ "learning_rate": 9.292243968009332e-06,
218
+ "loss": 0.6725,
219
+ "step": 30
220
+ },
221
+ {
222
+ "epoch": 3.2531645569620253,
223
+ "grad_norm": 0.7319165600762544,
224
+ "learning_rate": 9.213604793270196e-06,
225
+ "loss": 0.6368,
226
+ "step": 31
227
+ },
228
+ {
229
+ "epoch": 3.3544303797468356,
230
+ "grad_norm": 0.8042531889838589,
231
+ "learning_rate": 9.131193871579975e-06,
232
+ "loss": 0.6493,
233
+ "step": 32
234
+ },
235
+ {
236
+ "epoch": 3.4556962025316453,
237
+ "grad_norm": 0.7673927084015323,
238
+ "learning_rate": 9.045084971874738e-06,
239
+ "loss": 0.6493,
240
+ "step": 33
241
+ },
242
+ {
243
+ "epoch": 3.5569620253164556,
244
+ "grad_norm": 0.7199824300876746,
245
+ "learning_rate": 8.955355173281709e-06,
246
+ "loss": 0.6229,
247
+ "step": 34
248
+ },
249
+ {
250
+ "epoch": 3.6582278481012658,
251
+ "grad_norm": 0.7681905580545264,
252
+ "learning_rate": 8.862084796122998e-06,
253
+ "loss": 0.638,
254
+ "step": 35
255
+ },
256
+ {
257
+ "epoch": 3.759493670886076,
258
+ "grad_norm": 0.7031656323708602,
259
+ "learning_rate": 8.765357330018056e-06,
260
+ "loss": 0.6239,
261
+ "step": 36
262
+ },
263
+ {
264
+ "epoch": 3.8607594936708862,
265
+ "grad_norm": 0.5994707954353558,
266
+ "learning_rate": 8.665259359149132e-06,
267
+ "loss": 0.6095,
268
+ "step": 37
269
+ },
270
+ {
271
+ "epoch": 3.962025316455696,
272
+ "grad_norm": 0.5940391386425359,
273
+ "learning_rate": 8.561880484756726e-06,
274
+ "loss": 0.6043,
275
+ "step": 38
276
+ },
277
+ {
278
+ "epoch": 4.10126582278481,
279
+ "grad_norm": 0.8009562004971592,
280
+ "learning_rate": 8.455313244934324e-06,
281
+ "loss": 0.6053,
282
+ "step": 39
283
+ },
284
+ {
285
+ "epoch": 4.2025316455696204,
286
+ "grad_norm": 0.6899995087542556,
287
+ "learning_rate": 8.345653031794292e-06,
288
+ "loss": 0.5589,
289
+ "step": 40
290
+ },
291
+ {
292
+ "epoch": 4.30379746835443,
293
+ "grad_norm": 0.590919631317703,
294
+ "learning_rate": 8.232998006078998e-06,
295
+ "loss": 0.5538,
296
+ "step": 41
297
+ },
298
+ {
299
+ "epoch": 4.405063291139241,
300
+ "grad_norm": 0.6456293792676087,
301
+ "learning_rate": 8.117449009293668e-06,
302
+ "loss": 0.572,
303
+ "step": 42
304
+ },
305
+ {
306
+ "epoch": 4.506329113924051,
307
+ "grad_norm": 0.6469736779657014,
308
+ "learning_rate": 7.99910947343957e-06,
309
+ "loss": 0.5576,
310
+ "step": 43
311
+ },
312
+ {
313
+ "epoch": 4.6075949367088604,
314
+ "grad_norm": 0.6542464704518391,
315
+ "learning_rate": 7.87808532842837e-06,
316
+ "loss": 0.558,
317
+ "step": 44
318
+ },
319
+ {
320
+ "epoch": 4.708860759493671,
321
+ "grad_norm": 0.5743410125239509,
322
+ "learning_rate": 7.754484907260513e-06,
323
+ "loss": 0.5497,
324
+ "step": 45
325
+ },
326
+ {
327
+ "epoch": 4.810126582278481,
328
+ "grad_norm": 0.7132685729885268,
329
+ "learning_rate": 7.628418849052523e-06,
330
+ "loss": 0.5584,
331
+ "step": 46
332
+ },
333
+ {
334
+ "epoch": 4.911392405063291,
335
+ "grad_norm": 0.5429667720524944,
336
+ "learning_rate": 7.500000000000001e-06,
337
+ "loss": 0.545,
338
+ "step": 47
339
+ },
340
+ {
341
+ "epoch": 5.050632911392405,
342
+ "grad_norm": 0.7077161332627818,
343
+ "learning_rate": 7.369343312364994e-06,
344
+ "loss": 0.5441,
345
+ "step": 48
346
+ },
347
+ {
348
+ "epoch": 5.151898734177215,
349
+ "grad_norm": 0.5871123451372706,
350
+ "learning_rate": 7.236565741578163e-06,
351
+ "loss": 0.5123,
352
+ "step": 49
353
+ },
354
+ {
355
+ "epoch": 5.253164556962025,
356
+ "grad_norm": 0.49115314171982294,
357
+ "learning_rate": 7.101786141547829e-06,
358
+ "loss": 0.4975,
359
+ "step": 50
360
+ },
361
+ {
362
+ "epoch": 5.3544303797468356,
363
+ "grad_norm": 0.6792073834840421,
364
+ "learning_rate": 6.965125158269619e-06,
365
+ "loss": 0.5027,
366
+ "step": 51
367
+ },
368
+ {
369
+ "epoch": 5.455696202531645,
370
+ "grad_norm": 0.5427481034018683,
371
+ "learning_rate": 6.8267051218319766e-06,
372
+ "loss": 0.5105,
373
+ "step": 52
374
+ },
375
+ {
376
+ "epoch": 5.556962025316456,
377
+ "grad_norm": 0.6081404237694144,
378
+ "learning_rate": 6.686649936914151e-06,
379
+ "loss": 0.5094,
380
+ "step": 53
381
+ },
382
+ {
383
+ "epoch": 5.658227848101266,
384
+ "grad_norm": 0.6058953141713973,
385
+ "learning_rate": 6.545084971874738e-06,
386
+ "loss": 0.4464,
387
+ "step": 54
388
+ },
389
+ {
390
+ "epoch": 5.759493670886076,
391
+ "grad_norm": 0.5429116331543787,
392
+ "learning_rate": 6.402136946530014e-06,
393
+ "loss": 0.4788,
394
+ "step": 55
395
+ },
396
+ {
397
+ "epoch": 5.860759493670886,
398
+ "grad_norm": 0.5625127147066366,
399
+ "learning_rate": 6.257933818722544e-06,
400
+ "loss": 0.4802,
401
+ "step": 56
402
+ },
403
+ {
404
+ "epoch": 5.962025316455696,
405
+ "grad_norm": 0.6018989851826122,
406
+ "learning_rate": 6.112604669781572e-06,
407
+ "loss": 0.4843,
408
+ "step": 57
409
+ },
410
+ {
411
+ "epoch": 6.10126582278481,
412
+ "grad_norm": 0.5388943441225789,
413
+ "learning_rate": 5.9662795889777666e-06,
414
+ "loss": 0.4573,
415
+ "step": 58
416
+ },
417
+ {
418
+ "epoch": 6.2025316455696204,
419
+ "grad_norm": 0.5683110661372949,
420
+ "learning_rate": 5.819089557075689e-06,
421
+ "loss": 0.4315,
422
+ "step": 59
423
+ },
424
+ {
425
+ "epoch": 6.30379746835443,
426
+ "grad_norm": 0.5187201700440711,
427
+ "learning_rate": 5.671166329088278e-06,
428
+ "loss": 0.4385,
429
+ "step": 60
430
+ },
431
+ {
432
+ "epoch": 6.405063291139241,
433
+ "grad_norm": 0.5669675733829938,
434
+ "learning_rate": 5.522642316338268e-06,
435
+ "loss": 0.4216,
436
+ "step": 61
437
+ },
438
+ {
439
+ "epoch": 6.506329113924051,
440
+ "grad_norm": 0.5129178319513059,
441
+ "learning_rate": 5.373650467932122e-06,
442
+ "loss": 0.427,
443
+ "step": 62
444
+ },
445
+ {
446
+ "epoch": 6.6075949367088604,
447
+ "grad_norm": 0.5450550510951897,
448
+ "learning_rate": 5.224324151752575e-06,
449
+ "loss": 0.4269,
450
+ "step": 63
451
+ },
452
+ {
453
+ "epoch": 6.708860759493671,
454
+ "grad_norm": 0.5754420283713282,
455
+ "learning_rate": 5.074797035076319e-06,
456
+ "loss": 0.442,
457
+ "step": 64
458
+ },
459
+ {
460
+ "epoch": 6.810126582278481,
461
+ "grad_norm": 0.49191142476709515,
462
+ "learning_rate": 4.9252029649236835e-06,
463
+ "loss": 0.4342,
464
+ "step": 65
465
+ },
466
+ {
467
+ "epoch": 6.911392405063291,
468
+ "grad_norm": 0.5234434258205786,
469
+ "learning_rate": 4.775675848247427e-06,
470
+ "loss": 0.4292,
471
+ "step": 66
472
+ },
473
+ {
474
+ "epoch": 7.050632911392405,
475
+ "grad_norm": 0.5109662641475552,
476
+ "learning_rate": 4.626349532067879e-06,
477
+ "loss": 0.4168,
478
+ "step": 67
479
+ },
480
+ {
481
+ "epoch": 7.151898734177215,
482
+ "grad_norm": 0.5002978625419157,
483
+ "learning_rate": 4.477357683661734e-06,
484
+ "loss": 0.3871,
485
+ "step": 68
486
+ },
487
+ {
488
+ "epoch": 7.253164556962025,
489
+ "grad_norm": 0.5061314744739948,
490
+ "learning_rate": 4.3288336709117246e-06,
491
+ "loss": 0.3875,
492
+ "step": 69
493
+ },
494
+ {
495
+ "epoch": 7.3544303797468356,
496
+ "grad_norm": 0.6006225790016789,
497
+ "learning_rate": 4.180910442924312e-06,
498
+ "loss": 0.3825,
499
+ "step": 70
500
+ },
501
+ {
502
+ "epoch": 7.455696202531645,
503
+ "grad_norm": 0.5079368956298324,
504
+ "learning_rate": 4.033720411022235e-06,
505
+ "loss": 0.4008,
506
+ "step": 71
507
+ },
508
+ {
509
+ "epoch": 7.556962025316456,
510
+ "grad_norm": 0.548047739058599,
511
+ "learning_rate": 3.887395330218429e-06,
512
+ "loss": 0.3932,
513
+ "step": 72
514
+ },
515
+ {
516
+ "epoch": 7.658227848101266,
517
+ "grad_norm": 0.49930557552790156,
518
+ "learning_rate": 3.7420661812774577e-06,
519
+ "loss": 0.3446,
520
+ "step": 73
521
+ },
522
+ {
523
+ "epoch": 7.759493670886076,
524
+ "grad_norm": 0.5013495045541969,
525
+ "learning_rate": 3.5978630534699873e-06,
526
+ "loss": 0.3759,
527
+ "step": 74
528
+ },
529
+ {
530
+ "epoch": 7.860759493670886,
531
+ "grad_norm": 0.4947628036069991,
532
+ "learning_rate": 3.4549150281252635e-06,
533
+ "loss": 0.3771,
534
+ "step": 75
535
+ },
536
+ {
537
+ "epoch": 7.962025316455696,
538
+ "grad_norm": 0.5159919253735209,
539
+ "learning_rate": 3.3133500630858507e-06,
540
+ "loss": 0.3681,
541
+ "step": 76
542
+ },
543
+ {
544
+ "epoch": 8.10126582278481,
545
+ "grad_norm": 0.5071977268333723,
546
+ "learning_rate": 3.173294878168025e-06,
547
+ "loss": 0.3479,
548
+ "step": 77
549
+ },
550
+ {
551
+ "epoch": 8.20253164556962,
552
+ "grad_norm": 0.5211792049459344,
553
+ "learning_rate": 3.0348748417303826e-06,
554
+ "loss": 0.3468,
555
+ "step": 78
556
+ },
557
+ {
558
+ "epoch": 8.30379746835443,
559
+ "grad_norm": 0.5210688446476981,
560
+ "learning_rate": 2.8982138584521734e-06,
561
+ "loss": 0.341,
562
+ "step": 79
563
+ },
564
+ {
565
+ "epoch": 8.405063291139241,
566
+ "grad_norm": 0.5620152430317302,
567
+ "learning_rate": 2.7634342584218364e-06,
568
+ "loss": 0.3379,
569
+ "step": 80
570
+ },
571
+ {
572
+ "epoch": 8.50632911392405,
573
+ "grad_norm": 0.4934024735643696,
574
+ "learning_rate": 2.6306566876350072e-06,
575
+ "loss": 0.3553,
576
+ "step": 81
577
+ },
578
+ {
579
+ "epoch": 8.60759493670886,
580
+ "grad_norm": 0.5620316090128035,
581
+ "learning_rate": 2.5000000000000015e-06,
582
+ "loss": 0.3436,
583
+ "step": 82
584
+ },
585
+ {
586
+ "epoch": 8.708860759493671,
587
+ "grad_norm": 0.4900110812468828,
588
+ "learning_rate": 2.371581150947476e-06,
589
+ "loss": 0.3586,
590
+ "step": 83
591
+ },
592
+ {
593
+ "epoch": 8.810126582278482,
594
+ "grad_norm": 0.5242688539902147,
595
+ "learning_rate": 2.245515092739488e-06,
596
+ "loss": 0.3185,
597
+ "step": 84
598
+ },
599
+ {
600
+ "epoch": 8.91139240506329,
601
+ "grad_norm": 0.49075412772195337,
602
+ "learning_rate": 2.1219146715716332e-06,
603
+ "loss": 0.3007,
604
+ "step": 85
605
+ },
606
+ {
607
+ "epoch": 9.050632911392405,
608
+ "grad_norm": 0.5465801208579395,
609
+ "learning_rate": 2.0008905265604316e-06,
610
+ "loss": 0.3055,
611
+ "step": 86
612
+ },
613
+ {
614
+ "epoch": 9.151898734177216,
615
+ "grad_norm": 0.5468450266115664,
616
+ "learning_rate": 1.8825509907063328e-06,
617
+ "loss": 0.3092,
618
+ "step": 87
619
+ },
620
+ {
621
+ "epoch": 9.253164556962025,
622
+ "grad_norm": 0.5088152889789159,
623
+ "learning_rate": 1.7670019939210025e-06,
624
+ "loss": 0.3248,
625
+ "step": 88
626
+ },
627
+ {
628
+ "epoch": 9.354430379746836,
629
+ "grad_norm": 0.5176473893651191,
630
+ "learning_rate": 1.6543469682057105e-06,
631
+ "loss": 0.3006,
632
+ "step": 89
633
+ },
634
+ {
635
+ "epoch": 9.455696202531646,
636
+ "grad_norm": 0.49784877091018886,
637
+ "learning_rate": 1.544686755065677e-06,
638
+ "loss": 0.3181,
639
+ "step": 90
640
+ },
641
+ {
642
+ "epoch": 9.556962025316455,
643
+ "grad_norm": 0.5310739290997998,
644
+ "learning_rate": 1.438119515243277e-06,
645
+ "loss": 0.308,
646
+ "step": 91
647
+ },
648
+ {
649
+ "epoch": 9.658227848101266,
650
+ "grad_norm": 0.5179413997897845,
651
+ "learning_rate": 1.3347406408508695e-06,
652
+ "loss": 0.295,
653
+ "step": 92
654
+ },
655
+ {
656
+ "epoch": 9.759493670886076,
657
+ "grad_norm": 0.553058850478278,
658
+ "learning_rate": 1.234642669981946e-06,
659
+ "loss": 0.3063,
660
+ "step": 93
661
+ },
662
+ {
663
+ "epoch": 9.860759493670885,
664
+ "grad_norm": 0.5043907566802052,
665
+ "learning_rate": 1.137915203877003e-06,
666
+ "loss": 0.332,
667
+ "step": 94
668
+ },
669
+ {
670
+ "epoch": 9.962025316455696,
671
+ "grad_norm": 0.5055188027879971,
672
+ "learning_rate": 1.044644826718295e-06,
673
+ "loss": 0.3089,
674
+ "step": 95
675
+ },
676
+ {
677
+ "epoch": 10.10126582278481,
678
+ "grad_norm": 0.4881762000345773,
679
+ "learning_rate": 9.549150281252633e-07,
680
+ "loss": 0.307,
681
+ "step": 96
682
+ },
683
+ {
684
+ "epoch": 10.20253164556962,
685
+ "grad_norm": 0.4574073085718341,
686
+ "learning_rate": 8.688061284200266e-07,
687
+ "loss": 0.284,
688
+ "step": 97
689
+ },
690
+ {
691
+ "epoch": 10.30379746835443,
692
+ "grad_norm": 0.4779443110518862,
693
+ "learning_rate": 7.863952067298042e-07,
694
+ "loss": 0.2952,
695
+ "step": 98
696
+ },
697
+ {
698
+ "epoch": 10.405063291139241,
699
+ "grad_norm": 0.47835483708174475,
700
+ "learning_rate": 7.077560319906696e-07,
701
+ "loss": 0.252,
702
+ "step": 99
703
+ },
704
+ {
705
+ "epoch": 10.50632911392405,
706
+ "grad_norm": 0.547245288212164,
707
+ "learning_rate": 6.329589969143518e-07,
708
+ "loss": 0.2975,
709
+ "step": 100
710
+ },
711
+ {
712
+ "epoch": 10.60759493670886,
713
+ "grad_norm": 0.4995321835139682,
714
+ "learning_rate": 5.620710549772295e-07,
715
+ "loss": 0.2693,
716
+ "step": 101
717
+ },
718
+ {
719
+ "epoch": 10.708860759493671,
720
+ "grad_norm": 0.4735989189101935,
721
+ "learning_rate": 4.951556604879049e-07,
722
+ "loss": 0.3174,
723
+ "step": 102
724
+ },
725
+ {
726
+ "epoch": 10.810126582278482,
727
+ "grad_norm": 0.4639177277753312,
728
+ "learning_rate": 4.322727117869951e-07,
729
+ "loss": 0.2988,
730
+ "step": 103
731
+ },
732
+ {
733
+ "epoch": 10.91139240506329,
734
+ "grad_norm": 0.47629194630723604,
735
+ "learning_rate": 3.734784976300165e-07,
736
+ "loss": 0.2878,
737
+ "step": 104
738
+ },
739
+ {
740
+ "epoch": 11.050632911392405,
741
+ "grad_norm": 0.4735370844852891,
742
+ "learning_rate": 3.18825646801314e-07,
743
+ "loss": 0.2954,
744
+ "step": 105
745
+ },
746
+ {
747
+ "epoch": 11.151898734177216,
748
+ "grad_norm": 0.5268056430033645,
749
+ "learning_rate": 2.6836308100417874e-07,
750
+ "loss": 0.2896,
751
+ "step": 106
752
+ },
753
+ {
754
+ "epoch": 11.253164556962025,
755
+ "grad_norm": 0.5369135402096533,
756
+ "learning_rate": 2.2213597106929608e-07,
757
+ "loss": 0.2926,
758
+ "step": 107
759
+ },
760
+ {
761
+ "epoch": 11.354430379746836,
762
+ "grad_norm": 0.4547745160184931,
763
+ "learning_rate": 1.801856965207338e-07,
764
+ "loss": 0.2856,
765
+ "step": 108
766
+ },
767
+ {
768
+ "epoch": 11.455696202531646,
769
+ "grad_norm": 0.5115777299478187,
770
+ "learning_rate": 1.4254980853566248e-07,
771
+ "loss": 0.2548,
772
+ "step": 109
773
+ },
774
+ {
775
+ "epoch": 11.556962025316455,
776
+ "grad_norm": 0.4805530419903368,
777
+ "learning_rate": 1.0926199633097156e-07,
778
+ "loss": 0.2548,
779
+ "step": 110
780
+ },
781
+ {
782
+ "epoch": 11.658227848101266,
783
+ "grad_norm": 0.45391787027753194,
784
+ "learning_rate": 8.035205700685167e-08,
785
+ "loss": 0.2937,
786
+ "step": 111
787
+ },
788
+ {
789
+ "epoch": 11.759493670886076,
790
+ "grad_norm": 0.4575324517290467,
791
+ "learning_rate": 5.584586887435739e-08,
792
+ "loss": 0.2625,
793
+ "step": 112
794
+ },
795
+ {
796
+ "epoch": 11.860759493670885,
797
+ "grad_norm": 0.43014512418830014,
798
+ "learning_rate": 3.576536829081323e-08,
799
+ "loss": 0.2661,
800
+ "step": 113
801
+ },
802
+ {
803
+ "epoch": 11.962025316455696,
804
+ "grad_norm": 0.48309302437169416,
805
+ "learning_rate": 2.012853002380466e-08,
806
+ "loss": 0.3091,
807
+ "step": 114
808
+ },
809
+ {
810
+ "epoch": 12.10126582278481,
811
+ "grad_norm": 0.4454898577756698,
812
+ "learning_rate": 8.949351161324227e-09,
813
+ "loss": 0.2936,
814
+ "step": 115
815
+ },
816
+ {
817
+ "epoch": 12.20253164556962,
818
+ "grad_norm": 0.4323222828785268,
819
+ "learning_rate": 2.237838582483387e-09,
820
+ "loss": 0.2743,
821
+ "step": 116
822
+ },
823
+ {
824
+ "epoch": 12.30379746835443,
825
+ "grad_norm": 0.4487114468883012,
826
+ "learning_rate": 0.0,
827
+ "loss": 0.2999,
828
+ "step": 117
829
+ },
830
+ {
831
+ "epoch": 12.30379746835443,
832
+ "step": 117,
833
+ "total_flos": 1.5758524630748365e+17,
834
+ "train_loss": 0.5164603565493201,
835
+ "train_runtime": 5508.0362,
836
+ "train_samples_per_second": 0.746,
837
+ "train_steps_per_second": 0.021
838
+ }
839
+ ],
840
+ "logging_steps": 1,
841
+ "max_steps": 117,
842
+ "num_input_tokens_seen": 0,
843
+ "num_train_epochs": 13,
844
+ "save_steps": 500,
845
+ "stateful_callbacks": {
846
+ "TrainerControl": {
847
+ "args": {
848
+ "should_epoch_stop": false,
849
+ "should_evaluate": false,
850
+ "should_log": false,
851
+ "should_save": true,
852
+ "should_training_stop": true
853
+ },
854
+ "attributes": {}
855
+ }
856
+ },
857
+ "total_flos": 1.5758524630748365e+17,
858
+ "train_batch_size": 1,
859
+ "trial_name": null,
860
+ "trial_params": null
861
+ }
training_loss.png ADDED