ryanmarten commited on
Commit
33e435d
·
verified ·
1 Parent(s): 984456b

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_all_large_0.3k
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
15
 
16
  # d1_code_all_large_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_all_large_0.3k
 
16
 
17
  # d1_code_all_large_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_all_large_0.3k dataset.
20
 
21
  ## Model description
22
 
all_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 12.30379746835443,
3
+ "total_flos": 1.4789494914731213e+17,
4
+ "train_loss": 0.5038245582682455,
5
+ "train_runtime": 5710.2834,
6
+ "train_samples_per_second": 0.719,
7
+ "train_steps_per_second": 0.02
8
+ }
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 12.30379746835443,
3
+ "total_flos": 1.4789494914731213e+17,
4
+ "train_loss": 0.5038245582682455,
5
+ "train_runtime": 5710.2834,
6
+ "train_samples_per_second": 0.719,
7
+ "train_steps_per_second": 0.02
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": 6.454556094390299,
14
+ "learning_rate": 8.333333333333333e-07,
15
+ "loss": 1.0545,
16
+ "step": 1
17
+ },
18
+ {
19
+ "epoch": 0.20253164556962025,
20
+ "grad_norm": 6.417737137955762,
21
+ "learning_rate": 1.6666666666666667e-06,
22
+ "loss": 1.0147,
23
+ "step": 2
24
+ },
25
+ {
26
+ "epoch": 0.3037974683544304,
27
+ "grad_norm": 6.397055772415459,
28
+ "learning_rate": 2.5e-06,
29
+ "loss": 1.0388,
30
+ "step": 3
31
+ },
32
+ {
33
+ "epoch": 0.4050632911392405,
34
+ "grad_norm": 5.970478154455193,
35
+ "learning_rate": 3.3333333333333333e-06,
36
+ "loss": 1.0566,
37
+ "step": 4
38
+ },
39
+ {
40
+ "epoch": 0.5063291139240507,
41
+ "grad_norm": 4.549973865423855,
42
+ "learning_rate": 4.166666666666667e-06,
43
+ "loss": 0.9862,
44
+ "step": 5
45
+ },
46
+ {
47
+ "epoch": 0.6075949367088608,
48
+ "grad_norm": 2.884485930963668,
49
+ "learning_rate": 5e-06,
50
+ "loss": 0.976,
51
+ "step": 6
52
+ },
53
+ {
54
+ "epoch": 0.7088607594936709,
55
+ "grad_norm": 2.624623338241297,
56
+ "learning_rate": 5.833333333333334e-06,
57
+ "loss": 0.9699,
58
+ "step": 7
59
+ },
60
+ {
61
+ "epoch": 0.810126582278481,
62
+ "grad_norm": 3.5832441548111302,
63
+ "learning_rate": 6.666666666666667e-06,
64
+ "loss": 0.9578,
65
+ "step": 8
66
+ },
67
+ {
68
+ "epoch": 0.9113924050632911,
69
+ "grad_norm": 4.110170880390489,
70
+ "learning_rate": 7.500000000000001e-06,
71
+ "loss": 0.9382,
72
+ "step": 9
73
+ },
74
+ {
75
+ "epoch": 1.0506329113924051,
76
+ "grad_norm": 3.886591437000566,
77
+ "learning_rate": 8.333333333333334e-06,
78
+ "loss": 0.9338,
79
+ "step": 10
80
+ },
81
+ {
82
+ "epoch": 1.1518987341772151,
83
+ "grad_norm": 3.444362366123677,
84
+ "learning_rate": 9.166666666666666e-06,
85
+ "loss": 0.9125,
86
+ "step": 11
87
+ },
88
+ {
89
+ "epoch": 1.2531645569620253,
90
+ "grad_norm": 3.183785788213602,
91
+ "learning_rate": 1e-05,
92
+ "loss": 0.9065,
93
+ "step": 12
94
+ },
95
+ {
96
+ "epoch": 1.3544303797468356,
97
+ "grad_norm": 2.2368237099623776,
98
+ "learning_rate": 9.997762161417517e-06,
99
+ "loss": 0.8338,
100
+ "step": 13
101
+ },
102
+ {
103
+ "epoch": 1.4556962025316456,
104
+ "grad_norm": 1.5273683177235289,
105
+ "learning_rate": 9.991050648838676e-06,
106
+ "loss": 0.7991,
107
+ "step": 14
108
+ },
109
+ {
110
+ "epoch": 1.5569620253164556,
111
+ "grad_norm": 1.3227943760161387,
112
+ "learning_rate": 9.979871469976197e-06,
113
+ "loss": 0.819,
114
+ "step": 15
115
+ },
116
+ {
117
+ "epoch": 1.6582278481012658,
118
+ "grad_norm": 1.4149831364858252,
119
+ "learning_rate": 9.964234631709188e-06,
120
+ "loss": 0.8206,
121
+ "step": 16
122
+ },
123
+ {
124
+ "epoch": 1.759493670886076,
125
+ "grad_norm": 1.1745154520980696,
126
+ "learning_rate": 9.944154131125643e-06,
127
+ "loss": 0.8095,
128
+ "step": 17
129
+ },
130
+ {
131
+ "epoch": 1.8607594936708862,
132
+ "grad_norm": 1.1965850692590576,
133
+ "learning_rate": 9.91964794299315e-06,
134
+ "loss": 0.8021,
135
+ "step": 18
136
+ },
137
+ {
138
+ "epoch": 1.9620253164556962,
139
+ "grad_norm": 1.1087740638507724,
140
+ "learning_rate": 9.890738003669029e-06,
141
+ "loss": 0.7678,
142
+ "step": 19
143
+ },
144
+ {
145
+ "epoch": 2.1012658227848102,
146
+ "grad_norm": 0.8896991449953787,
147
+ "learning_rate": 9.857450191464337e-06,
148
+ "loss": 0.7597,
149
+ "step": 20
150
+ },
151
+ {
152
+ "epoch": 2.2025316455696204,
153
+ "grad_norm": 0.9532045822053368,
154
+ "learning_rate": 9.819814303479268e-06,
155
+ "loss": 0.7567,
156
+ "step": 21
157
+ },
158
+ {
159
+ "epoch": 2.3037974683544302,
160
+ "grad_norm": 0.9726543976613604,
161
+ "learning_rate": 9.777864028930705e-06,
162
+ "loss": 0.7245,
163
+ "step": 22
164
+ },
165
+ {
166
+ "epoch": 2.4050632911392404,
167
+ "grad_norm": 0.8945259026314462,
168
+ "learning_rate": 9.731636918995821e-06,
169
+ "loss": 0.7267,
170
+ "step": 23
171
+ },
172
+ {
173
+ "epoch": 2.5063291139240507,
174
+ "grad_norm": 0.8696330849922621,
175
+ "learning_rate": 9.681174353198687e-06,
176
+ "loss": 0.6757,
177
+ "step": 24
178
+ },
179
+ {
180
+ "epoch": 2.607594936708861,
181
+ "grad_norm": 0.8945043909600005,
182
+ "learning_rate": 9.626521502369984e-06,
183
+ "loss": 0.7068,
184
+ "step": 25
185
+ },
186
+ {
187
+ "epoch": 2.708860759493671,
188
+ "grad_norm": 0.7517601035844069,
189
+ "learning_rate": 9.567727288213005e-06,
190
+ "loss": 0.7087,
191
+ "step": 26
192
+ },
193
+ {
194
+ "epoch": 2.810126582278481,
195
+ "grad_norm": 0.7583761414292917,
196
+ "learning_rate": 9.504844339512096e-06,
197
+ "loss": 0.6867,
198
+ "step": 27
199
+ },
200
+ {
201
+ "epoch": 2.911392405063291,
202
+ "grad_norm": 0.8085046373388372,
203
+ "learning_rate": 9.437928945022772e-06,
204
+ "loss": 0.7023,
205
+ "step": 28
206
+ },
207
+ {
208
+ "epoch": 3.050632911392405,
209
+ "grad_norm": 0.7716504406586172,
210
+ "learning_rate": 9.36704100308565e-06,
211
+ "loss": 0.6412,
212
+ "step": 29
213
+ },
214
+ {
215
+ "epoch": 3.151898734177215,
216
+ "grad_norm": 0.8078413603181949,
217
+ "learning_rate": 9.292243968009332e-06,
218
+ "loss": 0.64,
219
+ "step": 30
220
+ },
221
+ {
222
+ "epoch": 3.2531645569620253,
223
+ "grad_norm": 0.730865860825069,
224
+ "learning_rate": 9.213604793270196e-06,
225
+ "loss": 0.6487,
226
+ "step": 31
227
+ },
228
+ {
229
+ "epoch": 3.3544303797468356,
230
+ "grad_norm": 0.6456456255392042,
231
+ "learning_rate": 9.131193871579975e-06,
232
+ "loss": 0.6279,
233
+ "step": 32
234
+ },
235
+ {
236
+ "epoch": 3.4556962025316453,
237
+ "grad_norm": 0.7069707689429814,
238
+ "learning_rate": 9.045084971874738e-06,
239
+ "loss": 0.6612,
240
+ "step": 33
241
+ },
242
+ {
243
+ "epoch": 3.5569620253164556,
244
+ "grad_norm": 0.6854078477032485,
245
+ "learning_rate": 8.955355173281709e-06,
246
+ "loss": 0.6314,
247
+ "step": 34
248
+ },
249
+ {
250
+ "epoch": 3.6582278481012658,
251
+ "grad_norm": 0.6916883772345461,
252
+ "learning_rate": 8.862084796122998e-06,
253
+ "loss": 0.634,
254
+ "step": 35
255
+ },
256
+ {
257
+ "epoch": 3.759493670886076,
258
+ "grad_norm": 0.6348149724624935,
259
+ "learning_rate": 8.765357330018056e-06,
260
+ "loss": 0.5864,
261
+ "step": 36
262
+ },
263
+ {
264
+ "epoch": 3.8607594936708862,
265
+ "grad_norm": 0.6768513298499309,
266
+ "learning_rate": 8.665259359149132e-06,
267
+ "loss": 0.5963,
268
+ "step": 37
269
+ },
270
+ {
271
+ "epoch": 3.962025316455696,
272
+ "grad_norm": 0.7044734514285598,
273
+ "learning_rate": 8.561880484756726e-06,
274
+ "loss": 0.6281,
275
+ "step": 38
276
+ },
277
+ {
278
+ "epoch": 4.10126582278481,
279
+ "grad_norm": 0.6357590042657805,
280
+ "learning_rate": 8.455313244934324e-06,
281
+ "loss": 0.5561,
282
+ "step": 39
283
+ },
284
+ {
285
+ "epoch": 4.2025316455696204,
286
+ "grad_norm": 0.6061488200858313,
287
+ "learning_rate": 8.345653031794292e-06,
288
+ "loss": 0.5678,
289
+ "step": 40
290
+ },
291
+ {
292
+ "epoch": 4.30379746835443,
293
+ "grad_norm": 0.5543856848671779,
294
+ "learning_rate": 8.232998006078998e-06,
295
+ "loss": 0.5662,
296
+ "step": 41
297
+ },
298
+ {
299
+ "epoch": 4.405063291139241,
300
+ "grad_norm": 0.616391389615191,
301
+ "learning_rate": 8.117449009293668e-06,
302
+ "loss": 0.5315,
303
+ "step": 42
304
+ },
305
+ {
306
+ "epoch": 4.506329113924051,
307
+ "grad_norm": 0.7372836903174991,
308
+ "learning_rate": 7.99910947343957e-06,
309
+ "loss": 0.5546,
310
+ "step": 43
311
+ },
312
+ {
313
+ "epoch": 4.6075949367088604,
314
+ "grad_norm": 0.6191135413381138,
315
+ "learning_rate": 7.87808532842837e-06,
316
+ "loss": 0.5316,
317
+ "step": 44
318
+ },
319
+ {
320
+ "epoch": 4.708860759493671,
321
+ "grad_norm": 0.6742260760353104,
322
+ "learning_rate": 7.754484907260513e-06,
323
+ "loss": 0.5741,
324
+ "step": 45
325
+ },
326
+ {
327
+ "epoch": 4.810126582278481,
328
+ "grad_norm": 0.6419298888465984,
329
+ "learning_rate": 7.628418849052523e-06,
330
+ "loss": 0.5193,
331
+ "step": 46
332
+ },
333
+ {
334
+ "epoch": 4.911392405063291,
335
+ "grad_norm": 0.6794663123446099,
336
+ "learning_rate": 7.500000000000001e-06,
337
+ "loss": 0.5354,
338
+ "step": 47
339
+ },
340
+ {
341
+ "epoch": 5.050632911392405,
342
+ "grad_norm": 0.5622606680501858,
343
+ "learning_rate": 7.369343312364994e-06,
344
+ "loss": 0.5143,
345
+ "step": 48
346
+ },
347
+ {
348
+ "epoch": 5.151898734177215,
349
+ "grad_norm": 0.557968749514869,
350
+ "learning_rate": 7.236565741578163e-06,
351
+ "loss": 0.4758,
352
+ "step": 49
353
+ },
354
+ {
355
+ "epoch": 5.253164556962025,
356
+ "grad_norm": 0.6285164975043734,
357
+ "learning_rate": 7.101786141547829e-06,
358
+ "loss": 0.498,
359
+ "step": 50
360
+ },
361
+ {
362
+ "epoch": 5.3544303797468356,
363
+ "grad_norm": 0.5508254900573698,
364
+ "learning_rate": 6.965125158269619e-06,
365
+ "loss": 0.4918,
366
+ "step": 51
367
+ },
368
+ {
369
+ "epoch": 5.455696202531645,
370
+ "grad_norm": 0.6573706547008377,
371
+ "learning_rate": 6.8267051218319766e-06,
372
+ "loss": 0.4897,
373
+ "step": 52
374
+ },
375
+ {
376
+ "epoch": 5.556962025316456,
377
+ "grad_norm": 0.5972950225279897,
378
+ "learning_rate": 6.686649936914151e-06,
379
+ "loss": 0.4818,
380
+ "step": 53
381
+ },
382
+ {
383
+ "epoch": 5.658227848101266,
384
+ "grad_norm": 0.5801150362152928,
385
+ "learning_rate": 6.545084971874738e-06,
386
+ "loss": 0.5278,
387
+ "step": 54
388
+ },
389
+ {
390
+ "epoch": 5.759493670886076,
391
+ "grad_norm": 0.5714074206166933,
392
+ "learning_rate": 6.402136946530014e-06,
393
+ "loss": 0.4475,
394
+ "step": 55
395
+ },
396
+ {
397
+ "epoch": 5.860759493670886,
398
+ "grad_norm": 0.6990262068725134,
399
+ "learning_rate": 6.257933818722544e-06,
400
+ "loss": 0.45,
401
+ "step": 56
402
+ },
403
+ {
404
+ "epoch": 5.962025316455696,
405
+ "grad_norm": 0.612195471912547,
406
+ "learning_rate": 6.112604669781572e-06,
407
+ "loss": 0.4678,
408
+ "step": 57
409
+ },
410
+ {
411
+ "epoch": 6.10126582278481,
412
+ "grad_norm": 0.6372612171416383,
413
+ "learning_rate": 5.9662795889777666e-06,
414
+ "loss": 0.4615,
415
+ "step": 58
416
+ },
417
+ {
418
+ "epoch": 6.2025316455696204,
419
+ "grad_norm": 0.5561127158427576,
420
+ "learning_rate": 5.819089557075689e-06,
421
+ "loss": 0.4388,
422
+ "step": 59
423
+ },
424
+ {
425
+ "epoch": 6.30379746835443,
426
+ "grad_norm": 0.6764720947393291,
427
+ "learning_rate": 5.671166329088278e-06,
428
+ "loss": 0.4694,
429
+ "step": 60
430
+ },
431
+ {
432
+ "epoch": 6.405063291139241,
433
+ "grad_norm": 0.5435832040476095,
434
+ "learning_rate": 5.522642316338268e-06,
435
+ "loss": 0.4345,
436
+ "step": 61
437
+ },
438
+ {
439
+ "epoch": 6.506329113924051,
440
+ "grad_norm": 0.634318429580686,
441
+ "learning_rate": 5.373650467932122e-06,
442
+ "loss": 0.3914,
443
+ "step": 62
444
+ },
445
+ {
446
+ "epoch": 6.6075949367088604,
447
+ "grad_norm": 0.5378270846095503,
448
+ "learning_rate": 5.224324151752575e-06,
449
+ "loss": 0.4026,
450
+ "step": 63
451
+ },
452
+ {
453
+ "epoch": 6.708860759493671,
454
+ "grad_norm": 0.4812034904225687,
455
+ "learning_rate": 5.074797035076319e-06,
456
+ "loss": 0.3814,
457
+ "step": 64
458
+ },
459
+ {
460
+ "epoch": 6.810126582278481,
461
+ "grad_norm": 0.6986551150169362,
462
+ "learning_rate": 4.9252029649236835e-06,
463
+ "loss": 0.3906,
464
+ "step": 65
465
+ },
466
+ {
467
+ "epoch": 6.911392405063291,
468
+ "grad_norm": 0.5080316842246274,
469
+ "learning_rate": 4.775675848247427e-06,
470
+ "loss": 0.4156,
471
+ "step": 66
472
+ },
473
+ {
474
+ "epoch": 7.050632911392405,
475
+ "grad_norm": 0.6522549284181087,
476
+ "learning_rate": 4.626349532067879e-06,
477
+ "loss": 0.4038,
478
+ "step": 67
479
+ },
480
+ {
481
+ "epoch": 7.151898734177215,
482
+ "grad_norm": 0.5871592991384735,
483
+ "learning_rate": 4.477357683661734e-06,
484
+ "loss": 0.3242,
485
+ "step": 68
486
+ },
487
+ {
488
+ "epoch": 7.253164556962025,
489
+ "grad_norm": 0.7491792949733449,
490
+ "learning_rate": 4.3288336709117246e-06,
491
+ "loss": 0.4044,
492
+ "step": 69
493
+ },
494
+ {
495
+ "epoch": 7.3544303797468356,
496
+ "grad_norm": 0.6192149001344914,
497
+ "learning_rate": 4.180910442924312e-06,
498
+ "loss": 0.3757,
499
+ "step": 70
500
+ },
501
+ {
502
+ "epoch": 7.455696202531645,
503
+ "grad_norm": 0.5923337987187833,
504
+ "learning_rate": 4.033720411022235e-06,
505
+ "loss": 0.375,
506
+ "step": 71
507
+ },
508
+ {
509
+ "epoch": 7.556962025316456,
510
+ "grad_norm": 0.6244472713338677,
511
+ "learning_rate": 3.887395330218429e-06,
512
+ "loss": 0.342,
513
+ "step": 72
514
+ },
515
+ {
516
+ "epoch": 7.658227848101266,
517
+ "grad_norm": 0.5877970431530837,
518
+ "learning_rate": 3.7420661812774577e-06,
519
+ "loss": 0.3663,
520
+ "step": 73
521
+ },
522
+ {
523
+ "epoch": 7.759493670886076,
524
+ "grad_norm": 0.5139611168189747,
525
+ "learning_rate": 3.5978630534699873e-06,
526
+ "loss": 0.3469,
527
+ "step": 74
528
+ },
529
+ {
530
+ "epoch": 7.860759493670886,
531
+ "grad_norm": 0.6375466905321525,
532
+ "learning_rate": 3.4549150281252635e-06,
533
+ "loss": 0.3507,
534
+ "step": 75
535
+ },
536
+ {
537
+ "epoch": 7.962025316455696,
538
+ "grad_norm": 0.623233174164931,
539
+ "learning_rate": 3.3133500630858507e-06,
540
+ "loss": 0.3416,
541
+ "step": 76
542
+ },
543
+ {
544
+ "epoch": 8.10126582278481,
545
+ "grad_norm": 0.7043410018742996,
546
+ "learning_rate": 3.173294878168025e-06,
547
+ "loss": 0.3062,
548
+ "step": 77
549
+ },
550
+ {
551
+ "epoch": 8.20253164556962,
552
+ "grad_norm": 0.6265939100728176,
553
+ "learning_rate": 3.0348748417303826e-06,
554
+ "loss": 0.3424,
555
+ "step": 78
556
+ },
557
+ {
558
+ "epoch": 8.30379746835443,
559
+ "grad_norm": 0.5476956354154895,
560
+ "learning_rate": 2.8982138584521734e-06,
561
+ "loss": 0.3477,
562
+ "step": 79
563
+ },
564
+ {
565
+ "epoch": 8.405063291139241,
566
+ "grad_norm": 0.8387684314122273,
567
+ "learning_rate": 2.7634342584218364e-06,
568
+ "loss": 0.2966,
569
+ "step": 80
570
+ },
571
+ {
572
+ "epoch": 8.50632911392405,
573
+ "grad_norm": 0.6219439618490264,
574
+ "learning_rate": 2.6306566876350072e-06,
575
+ "loss": 0.3116,
576
+ "step": 81
577
+ },
578
+ {
579
+ "epoch": 8.60759493670886,
580
+ "grad_norm": 0.5752960287884662,
581
+ "learning_rate": 2.5000000000000015e-06,
582
+ "loss": 0.3159,
583
+ "step": 82
584
+ },
585
+ {
586
+ "epoch": 8.708860759493671,
587
+ "grad_norm": 0.5768589622756186,
588
+ "learning_rate": 2.371581150947476e-06,
589
+ "loss": 0.3184,
590
+ "step": 83
591
+ },
592
+ {
593
+ "epoch": 8.810126582278482,
594
+ "grad_norm": 0.6410650599630946,
595
+ "learning_rate": 2.245515092739488e-06,
596
+ "loss": 0.3208,
597
+ "step": 84
598
+ },
599
+ {
600
+ "epoch": 8.91139240506329,
601
+ "grad_norm": 0.5407301259788028,
602
+ "learning_rate": 2.1219146715716332e-06,
603
+ "loss": 0.3268,
604
+ "step": 85
605
+ },
606
+ {
607
+ "epoch": 9.050632911392405,
608
+ "grad_norm": 0.5566954462690553,
609
+ "learning_rate": 2.0008905265604316e-06,
610
+ "loss": 0.3139,
611
+ "step": 86
612
+ },
613
+ {
614
+ "epoch": 9.151898734177216,
615
+ "grad_norm": 0.5302867488116099,
616
+ "learning_rate": 1.8825509907063328e-06,
617
+ "loss": 0.2942,
618
+ "step": 87
619
+ },
620
+ {
621
+ "epoch": 9.253164556962025,
622
+ "grad_norm": 0.550994856130443,
623
+ "learning_rate": 1.7670019939210025e-06,
624
+ "loss": 0.2742,
625
+ "step": 88
626
+ },
627
+ {
628
+ "epoch": 9.354430379746836,
629
+ "grad_norm": 0.5831740601335943,
630
+ "learning_rate": 1.6543469682057105e-06,
631
+ "loss": 0.2998,
632
+ "step": 89
633
+ },
634
+ {
635
+ "epoch": 9.455696202531646,
636
+ "grad_norm": 0.5443597602523821,
637
+ "learning_rate": 1.544686755065677e-06,
638
+ "loss": 0.2789,
639
+ "step": 90
640
+ },
641
+ {
642
+ "epoch": 9.556962025316455,
643
+ "grad_norm": 0.5529724450594875,
644
+ "learning_rate": 1.438119515243277e-06,
645
+ "loss": 0.2716,
646
+ "step": 91
647
+ },
648
+ {
649
+ "epoch": 9.658227848101266,
650
+ "grad_norm": 0.5542774061602812,
651
+ "learning_rate": 1.3347406408508695e-06,
652
+ "loss": 0.3038,
653
+ "step": 92
654
+ },
655
+ {
656
+ "epoch": 9.759493670886076,
657
+ "grad_norm": 0.5124770637416871,
658
+ "learning_rate": 1.234642669981946e-06,
659
+ "loss": 0.2947,
660
+ "step": 93
661
+ },
662
+ {
663
+ "epoch": 9.860759493670885,
664
+ "grad_norm": 0.5321332704363179,
665
+ "learning_rate": 1.137915203877003e-06,
666
+ "loss": 0.2753,
667
+ "step": 94
668
+ },
669
+ {
670
+ "epoch": 9.962025316455696,
671
+ "grad_norm": 0.5071141597255457,
672
+ "learning_rate": 1.044644826718295e-06,
673
+ "loss": 0.2931,
674
+ "step": 95
675
+ },
676
+ {
677
+ "epoch": 10.10126582278481,
678
+ "grad_norm": 0.5075278407092699,
679
+ "learning_rate": 9.549150281252633e-07,
680
+ "loss": 0.2742,
681
+ "step": 96
682
+ },
683
+ {
684
+ "epoch": 10.20253164556962,
685
+ "grad_norm": 0.4983778336916034,
686
+ "learning_rate": 8.688061284200266e-07,
687
+ "loss": 0.2446,
688
+ "step": 97
689
+ },
690
+ {
691
+ "epoch": 10.30379746835443,
692
+ "grad_norm": 0.48331118277540047,
693
+ "learning_rate": 7.863952067298042e-07,
694
+ "loss": 0.2609,
695
+ "step": 98
696
+ },
697
+ {
698
+ "epoch": 10.405063291139241,
699
+ "grad_norm": 0.4814152403038732,
700
+ "learning_rate": 7.077560319906696e-07,
701
+ "loss": 0.2403,
702
+ "step": 99
703
+ },
704
+ {
705
+ "epoch": 10.50632911392405,
706
+ "grad_norm": 0.5660703504839063,
707
+ "learning_rate": 6.329589969143518e-07,
708
+ "loss": 0.2876,
709
+ "step": 100
710
+ },
711
+ {
712
+ "epoch": 10.60759493670886,
713
+ "grad_norm": 0.49598484519578945,
714
+ "learning_rate": 5.620710549772295e-07,
715
+ "loss": 0.277,
716
+ "step": 101
717
+ },
718
+ {
719
+ "epoch": 10.708860759493671,
720
+ "grad_norm": 0.47886642738337454,
721
+ "learning_rate": 4.951556604879049e-07,
722
+ "loss": 0.2838,
723
+ "step": 102
724
+ },
725
+ {
726
+ "epoch": 10.810126582278482,
727
+ "grad_norm": 0.5173907661727798,
728
+ "learning_rate": 4.322727117869951e-07,
729
+ "loss": 0.256,
730
+ "step": 103
731
+ },
732
+ {
733
+ "epoch": 10.91139240506329,
734
+ "grad_norm": 0.48104938905225675,
735
+ "learning_rate": 3.734784976300165e-07,
736
+ "loss": 0.2958,
737
+ "step": 104
738
+ },
739
+ {
740
+ "epoch": 11.050632911392405,
741
+ "grad_norm": 0.4497186932206841,
742
+ "learning_rate": 3.18825646801314e-07,
743
+ "loss": 0.2405,
744
+ "step": 105
745
+ },
746
+ {
747
+ "epoch": 11.151898734177216,
748
+ "grad_norm": 0.5459247872594127,
749
+ "learning_rate": 2.6836308100417874e-07,
750
+ "loss": 0.2295,
751
+ "step": 106
752
+ },
753
+ {
754
+ "epoch": 11.253164556962025,
755
+ "grad_norm": 0.4937165422111695,
756
+ "learning_rate": 2.2213597106929608e-07,
757
+ "loss": 0.2759,
758
+ "step": 107
759
+ },
760
+ {
761
+ "epoch": 11.354430379746836,
762
+ "grad_norm": 0.48808210203806407,
763
+ "learning_rate": 1.801856965207338e-07,
764
+ "loss": 0.2662,
765
+ "step": 108
766
+ },
767
+ {
768
+ "epoch": 11.455696202531646,
769
+ "grad_norm": 0.4709659384878408,
770
+ "learning_rate": 1.4254980853566248e-07,
771
+ "loss": 0.2473,
772
+ "step": 109
773
+ },
774
+ {
775
+ "epoch": 11.556962025316455,
776
+ "grad_norm": 0.4546457843533063,
777
+ "learning_rate": 1.0926199633097156e-07,
778
+ "loss": 0.3028,
779
+ "step": 110
780
+ },
781
+ {
782
+ "epoch": 11.658227848101266,
783
+ "grad_norm": 0.435537410815735,
784
+ "learning_rate": 8.035205700685167e-08,
785
+ "loss": 0.2382,
786
+ "step": 111
787
+ },
788
+ {
789
+ "epoch": 11.759493670886076,
790
+ "grad_norm": 0.4563587492866212,
791
+ "learning_rate": 5.584586887435739e-08,
792
+ "loss": 0.2674,
793
+ "step": 112
794
+ },
795
+ {
796
+ "epoch": 11.860759493670885,
797
+ "grad_norm": 0.41954247071137973,
798
+ "learning_rate": 3.576536829081323e-08,
799
+ "loss": 0.2338,
800
+ "step": 113
801
+ },
802
+ {
803
+ "epoch": 11.962025316455696,
804
+ "grad_norm": 0.4387686702433981,
805
+ "learning_rate": 2.012853002380466e-08,
806
+ "loss": 0.2923,
807
+ "step": 114
808
+ },
809
+ {
810
+ "epoch": 12.10126582278481,
811
+ "grad_norm": 0.43045896415758966,
812
+ "learning_rate": 8.949351161324227e-09,
813
+ "loss": 0.2907,
814
+ "step": 115
815
+ },
816
+ {
817
+ "epoch": 12.20253164556962,
818
+ "grad_norm": 0.41266970718616575,
819
+ "learning_rate": 2.237838582483387e-09,
820
+ "loss": 0.2661,
821
+ "step": 116
822
+ },
823
+ {
824
+ "epoch": 12.30379746835443,
825
+ "grad_norm": 0.5074238547486939,
826
+ "learning_rate": 0.0,
827
+ "loss": 0.2524,
828
+ "step": 117
829
+ },
830
+ {
831
+ "epoch": 12.30379746835443,
832
+ "step": 117,
833
+ "total_flos": 1.4789494914731213e+17,
834
+ "train_loss": 0.5038245582682455,
835
+ "train_runtime": 5710.2834,
836
+ "train_samples_per_second": 0.719,
837
+ "train_steps_per_second": 0.02
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": 1.4789494914731213e+17,
858
+ "train_batch_size": 1,
859
+ "trial_name": null,
860
+ "trial_params": null
861
+ }
training_loss.png ADDED