gsmyrnis commited on
Commit
4efbb76
·
verified ·
1 Parent(s): 530a020

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 +567 -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: qwen2-5_sky_t1_2-5k_rewrite_r1_distill_llama70b
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
15
 
16
  # qwen2-5_sky_t1_2-5k_rewrite_r1_distill_llama70b
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: qwen2-5_sky_t1_2-5k_rewrite_r1_distill_llama70b
 
16
 
17
  # qwen2-5_sky_t1_2-5k_rewrite_r1_distill_llama70b
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/sky_t1_2-5k_rewrite_r1_distill_llama70b dataset.
20
 
21
  ## Model description
22
 
all_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 3.0,
3
+ "total_flos": 20372648230912.0,
4
+ "train_loss": 0.4803738268216451,
5
+ "train_runtime": 543.6715,
6
+ "train_samples_per_second": 13.243,
7
+ "train_steps_per_second": 0.138
8
+ }
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 3.0,
3
+ "total_flos": 20372648230912.0,
4
+ "train_loss": 0.4803738268216451,
5
+ "train_runtime": 543.6715,
6
+ "train_samples_per_second": 13.243,
7
+ "train_steps_per_second": 0.138
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,567 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 3.0,
5
+ "eval_steps": 500,
6
+ "global_step": 75,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.04,
13
+ "grad_norm": 4.877281767256086,
14
+ "learning_rate": 1.25e-06,
15
+ "loss": 0.8206,
16
+ "step": 1
17
+ },
18
+ {
19
+ "epoch": 0.08,
20
+ "grad_norm": 5.369627920053383,
21
+ "learning_rate": 2.5e-06,
22
+ "loss": 0.8398,
23
+ "step": 2
24
+ },
25
+ {
26
+ "epoch": 0.12,
27
+ "grad_norm": 4.819203480978856,
28
+ "learning_rate": 3.7500000000000005e-06,
29
+ "loss": 0.8349,
30
+ "step": 3
31
+ },
32
+ {
33
+ "epoch": 0.16,
34
+ "grad_norm": 4.708855694010613,
35
+ "learning_rate": 5e-06,
36
+ "loss": 0.794,
37
+ "step": 4
38
+ },
39
+ {
40
+ "epoch": 0.2,
41
+ "grad_norm": 3.6373310096703357,
42
+ "learning_rate": 6.25e-06,
43
+ "loss": 0.7663,
44
+ "step": 5
45
+ },
46
+ {
47
+ "epoch": 0.24,
48
+ "grad_norm": 2.135279898196607,
49
+ "learning_rate": 7.500000000000001e-06,
50
+ "loss": 0.6478,
51
+ "step": 6
52
+ },
53
+ {
54
+ "epoch": 0.28,
55
+ "grad_norm": 2.5655576510980604,
56
+ "learning_rate": 8.750000000000001e-06,
57
+ "loss": 0.6477,
58
+ "step": 7
59
+ },
60
+ {
61
+ "epoch": 0.32,
62
+ "grad_norm": 2.8068625830587983,
63
+ "learning_rate": 1e-05,
64
+ "loss": 0.6968,
65
+ "step": 8
66
+ },
67
+ {
68
+ "epoch": 0.36,
69
+ "grad_norm": 2.0405627443941867,
70
+ "learning_rate": 9.994504457428557e-06,
71
+ "loss": 0.6165,
72
+ "step": 9
73
+ },
74
+ {
75
+ "epoch": 0.4,
76
+ "grad_norm": 2.088955022175797,
77
+ "learning_rate": 9.978029910109491e-06,
78
+ "loss": 0.6045,
79
+ "step": 10
80
+ },
81
+ {
82
+ "epoch": 0.44,
83
+ "grad_norm": 1.6551751609174368,
84
+ "learning_rate": 9.950612572673255e-06,
85
+ "loss": 0.579,
86
+ "step": 11
87
+ },
88
+ {
89
+ "epoch": 0.48,
90
+ "grad_norm": 1.5329473345986582,
91
+ "learning_rate": 9.91231271437788e-06,
92
+ "loss": 0.5634,
93
+ "step": 12
94
+ },
95
+ {
96
+ "epoch": 0.52,
97
+ "grad_norm": 1.4220679781894365,
98
+ "learning_rate": 9.863214526624065e-06,
99
+ "loss": 0.5766,
100
+ "step": 13
101
+ },
102
+ {
103
+ "epoch": 0.56,
104
+ "grad_norm": 1.49388738084358,
105
+ "learning_rate": 9.803425937884202e-06,
106
+ "loss": 0.5681,
107
+ "step": 14
108
+ },
109
+ {
110
+ "epoch": 0.6,
111
+ "grad_norm": 1.2077926761043856,
112
+ "learning_rate": 9.733078376452172e-06,
113
+ "loss": 0.5909,
114
+ "step": 15
115
+ },
116
+ {
117
+ "epoch": 0.64,
118
+ "grad_norm": 1.1712939279273036,
119
+ "learning_rate": 9.652326481535434e-06,
120
+ "loss": 0.555,
121
+ "step": 16
122
+ },
123
+ {
124
+ "epoch": 0.68,
125
+ "grad_norm": 1.1371718201572385,
126
+ "learning_rate": 9.561347763324484e-06,
127
+ "loss": 0.5554,
128
+ "step": 17
129
+ },
130
+ {
131
+ "epoch": 0.72,
132
+ "grad_norm": 1.0446533550508397,
133
+ "learning_rate": 9.460342212786933e-06,
134
+ "loss": 0.5547,
135
+ "step": 18
136
+ },
137
+ {
138
+ "epoch": 0.76,
139
+ "grad_norm": 0.7760261762286178,
140
+ "learning_rate": 9.349531862043952e-06,
141
+ "loss": 0.5153,
142
+ "step": 19
143
+ },
144
+ {
145
+ "epoch": 0.8,
146
+ "grad_norm": 0.9073177851696704,
147
+ "learning_rate": 9.229160296295488e-06,
148
+ "loss": 0.5672,
149
+ "step": 20
150
+ },
151
+ {
152
+ "epoch": 0.84,
153
+ "grad_norm": 0.8474699152217943,
154
+ "learning_rate": 9.099492118367123e-06,
155
+ "loss": 0.5615,
156
+ "step": 21
157
+ },
158
+ {
159
+ "epoch": 0.88,
160
+ "grad_norm": 0.8094336108303068,
161
+ "learning_rate": 8.960812367055646e-06,
162
+ "loss": 0.532,
163
+ "step": 22
164
+ },
165
+ {
166
+ "epoch": 0.92,
167
+ "grad_norm": 0.8195192520724021,
168
+ "learning_rate": 8.81342589055191e-06,
169
+ "loss": 0.5685,
170
+ "step": 23
171
+ },
172
+ {
173
+ "epoch": 0.96,
174
+ "grad_norm": 0.8032789127659623,
175
+ "learning_rate": 8.657656676318346e-06,
176
+ "loss": 0.5798,
177
+ "step": 24
178
+ },
179
+ {
180
+ "epoch": 1.0,
181
+ "grad_norm": 0.7820268577079116,
182
+ "learning_rate": 8.49384713889421e-06,
183
+ "loss": 0.5492,
184
+ "step": 25
185
+ },
186
+ {
187
+ "epoch": 1.04,
188
+ "grad_norm": 0.7692042488256751,
189
+ "learning_rate": 8.32235736719411e-06,
190
+ "loss": 0.4943,
191
+ "step": 26
192
+ },
193
+ {
194
+ "epoch": 1.08,
195
+ "grad_norm": 0.7090181776748236,
196
+ "learning_rate": 8.143564332954426e-06,
197
+ "loss": 0.4435,
198
+ "step": 27
199
+ },
200
+ {
201
+ "epoch": 1.12,
202
+ "grad_norm": 0.65702841278586,
203
+ "learning_rate": 7.957861062067614e-06,
204
+ "loss": 0.4319,
205
+ "step": 28
206
+ },
207
+ {
208
+ "epoch": 1.16,
209
+ "grad_norm": 0.675448734499891,
210
+ "learning_rate": 7.765655770625997e-06,
211
+ "loss": 0.4383,
212
+ "step": 29
213
+ },
214
+ {
215
+ "epoch": 1.2,
216
+ "grad_norm": 0.7807304911969501,
217
+ "learning_rate": 7.56737096757421e-06,
218
+ "loss": 0.4815,
219
+ "step": 30
220
+ },
221
+ {
222
+ "epoch": 1.24,
223
+ "grad_norm": 0.697811006200286,
224
+ "learning_rate": 7.363442525942827e-06,
225
+ "loss": 0.4587,
226
+ "step": 31
227
+ },
228
+ {
229
+ "epoch": 1.28,
230
+ "grad_norm": 0.7882685643397841,
231
+ "learning_rate": 7.1543187247048525e-06,
232
+ "loss": 0.4446,
233
+ "step": 32
234
+ },
235
+ {
236
+ "epoch": 1.32,
237
+ "grad_norm": 0.7379298166757972,
238
+ "learning_rate": 6.9404592633612486e-06,
239
+ "loss": 0.4305,
240
+ "step": 33
241
+ },
242
+ {
243
+ "epoch": 1.3599999999999999,
244
+ "grad_norm": 0.761037521053593,
245
+ "learning_rate": 6.722334251421665e-06,
246
+ "loss": 0.4743,
247
+ "step": 34
248
+ },
249
+ {
250
+ "epoch": 1.4,
251
+ "grad_norm": 0.7824996053631937,
252
+ "learning_rate": 6.500423175001705e-06,
253
+ "loss": 0.4381,
254
+ "step": 35
255
+ },
256
+ {
257
+ "epoch": 1.44,
258
+ "grad_norm": 0.7386455759838737,
259
+ "learning_rate": 6.275213842808383e-06,
260
+ "loss": 0.4806,
261
+ "step": 36
262
+ },
263
+ {
264
+ "epoch": 1.48,
265
+ "grad_norm": 0.7526911520339383,
266
+ "learning_rate": 6.047201313830724e-06,
267
+ "loss": 0.4603,
268
+ "step": 37
269
+ },
270
+ {
271
+ "epoch": 1.52,
272
+ "grad_norm": 0.6600966457635797,
273
+ "learning_rate": 5.816886809092651e-06,
274
+ "loss": 0.4155,
275
+ "step": 38
276
+ },
277
+ {
278
+ "epoch": 1.56,
279
+ "grad_norm": 0.7714058248821409,
280
+ "learning_rate": 5.584776609860414e-06,
281
+ "loss": 0.4588,
282
+ "step": 39
283
+ },
284
+ {
285
+ "epoch": 1.6,
286
+ "grad_norm": 0.7134529075059547,
287
+ "learning_rate": 5.351380944726465e-06,
288
+ "loss": 0.443,
289
+ "step": 40
290
+ },
291
+ {
292
+ "epoch": 1.6400000000000001,
293
+ "grad_norm": 0.6530128164819851,
294
+ "learning_rate": 5.117212868016303e-06,
295
+ "loss": 0.4313,
296
+ "step": 41
297
+ },
298
+ {
299
+ "epoch": 1.6800000000000002,
300
+ "grad_norm": 0.6968246237025671,
301
+ "learning_rate": 4.882787131983698e-06,
302
+ "loss": 0.4539,
303
+ "step": 42
304
+ },
305
+ {
306
+ "epoch": 1.72,
307
+ "grad_norm": 0.7211717885656892,
308
+ "learning_rate": 4.6486190552735375e-06,
309
+ "loss": 0.4248,
310
+ "step": 43
311
+ },
312
+ {
313
+ "epoch": 1.76,
314
+ "grad_norm": 0.725165174893242,
315
+ "learning_rate": 4.415223390139588e-06,
316
+ "loss": 0.4299,
317
+ "step": 44
318
+ },
319
+ {
320
+ "epoch": 1.8,
321
+ "grad_norm": 0.6282730028813539,
322
+ "learning_rate": 4.183113190907349e-06,
323
+ "loss": 0.4198,
324
+ "step": 45
325
+ },
326
+ {
327
+ "epoch": 1.8399999999999999,
328
+ "grad_norm": 0.6824715174152934,
329
+ "learning_rate": 3.952798686169279e-06,
330
+ "loss": 0.4391,
331
+ "step": 46
332
+ },
333
+ {
334
+ "epoch": 1.88,
335
+ "grad_norm": 0.6679676027831118,
336
+ "learning_rate": 3.7247861571916183e-06,
337
+ "loss": 0.469,
338
+ "step": 47
339
+ },
340
+ {
341
+ "epoch": 1.92,
342
+ "grad_norm": 0.674340396045629,
343
+ "learning_rate": 3.4995768249982975e-06,
344
+ "loss": 0.4518,
345
+ "step": 48
346
+ },
347
+ {
348
+ "epoch": 1.96,
349
+ "grad_norm": 0.7333007620608074,
350
+ "learning_rate": 3.2776657485783357e-06,
351
+ "loss": 0.4824,
352
+ "step": 49
353
+ },
354
+ {
355
+ "epoch": 2.0,
356
+ "grad_norm": 0.7309087708096496,
357
+ "learning_rate": 3.059540736638751e-06,
358
+ "loss": 0.4069,
359
+ "step": 50
360
+ },
361
+ {
362
+ "epoch": 2.04,
363
+ "grad_norm": 0.7051822551292913,
364
+ "learning_rate": 2.8456812752951483e-06,
365
+ "loss": 0.3536,
366
+ "step": 51
367
+ },
368
+ {
369
+ "epoch": 2.08,
370
+ "grad_norm": 0.689099788100928,
371
+ "learning_rate": 2.636557474057173e-06,
372
+ "loss": 0.387,
373
+ "step": 52
374
+ },
375
+ {
376
+ "epoch": 2.12,
377
+ "grad_norm": 0.6646060871398509,
378
+ "learning_rate": 2.4326290324257896e-06,
379
+ "loss": 0.3654,
380
+ "step": 53
381
+ },
382
+ {
383
+ "epoch": 2.16,
384
+ "grad_norm": 0.6581663922816163,
385
+ "learning_rate": 2.234344229374003e-06,
386
+ "loss": 0.3556,
387
+ "step": 54
388
+ },
389
+ {
390
+ "epoch": 2.2,
391
+ "grad_norm": 0.6153881116079951,
392
+ "learning_rate": 2.042138937932388e-06,
393
+ "loss": 0.3776,
394
+ "step": 55
395
+ },
396
+ {
397
+ "epoch": 2.24,
398
+ "grad_norm": 0.6071374038319348,
399
+ "learning_rate": 1.856435667045577e-06,
400
+ "loss": 0.3705,
401
+ "step": 56
402
+ },
403
+ {
404
+ "epoch": 2.2800000000000002,
405
+ "grad_norm": 0.6993574061286455,
406
+ "learning_rate": 1.677642632805892e-06,
407
+ "loss": 0.3954,
408
+ "step": 57
409
+ },
410
+ {
411
+ "epoch": 2.32,
412
+ "grad_norm": 0.6193275767542719,
413
+ "learning_rate": 1.5061528611057917e-06,
414
+ "loss": 0.3456,
415
+ "step": 58
416
+ },
417
+ {
418
+ "epoch": 2.36,
419
+ "grad_norm": 0.6355788571188572,
420
+ "learning_rate": 1.3423433236816563e-06,
421
+ "loss": 0.3473,
422
+ "step": 59
423
+ },
424
+ {
425
+ "epoch": 2.4,
426
+ "grad_norm": 0.5948954936656425,
427
+ "learning_rate": 1.186574109448091e-06,
428
+ "loss": 0.3602,
429
+ "step": 60
430
+ },
431
+ {
432
+ "epoch": 2.44,
433
+ "grad_norm": 0.64634470157691,
434
+ "learning_rate": 1.0391876329443534e-06,
435
+ "loss": 0.3689,
436
+ "step": 61
437
+ },
438
+ {
439
+ "epoch": 2.48,
440
+ "grad_norm": 0.6754148904921049,
441
+ "learning_rate": 9.005078816328772e-07,
442
+ "loss": 0.3757,
443
+ "step": 62
444
+ },
445
+ {
446
+ "epoch": 2.52,
447
+ "grad_norm": 0.6274024265467717,
448
+ "learning_rate": 7.708397037045129e-07,
449
+ "loss": 0.3699,
450
+ "step": 63
451
+ },
452
+ {
453
+ "epoch": 2.56,
454
+ "grad_norm": 0.6201142033583393,
455
+ "learning_rate": 6.50468137956049e-07,
456
+ "loss": 0.3298,
457
+ "step": 64
458
+ },
459
+ {
460
+ "epoch": 2.6,
461
+ "grad_norm": 0.6790501930611452,
462
+ "learning_rate": 5.396577872130676e-07,
463
+ "loss": 0.3964,
464
+ "step": 65
465
+ },
466
+ {
467
+ "epoch": 2.64,
468
+ "grad_norm": 0.6401906672741758,
469
+ "learning_rate": 4.386522366755169e-07,
470
+ "loss": 0.3982,
471
+ "step": 66
472
+ },
473
+ {
474
+ "epoch": 2.68,
475
+ "grad_norm": 0.6511102281434116,
476
+ "learning_rate": 3.4767351846456744e-07,
477
+ "loss": 0.3749,
478
+ "step": 67
479
+ },
480
+ {
481
+ "epoch": 2.7199999999999998,
482
+ "grad_norm": 0.664712761433258,
483
+ "learning_rate": 2.669216235478295e-07,
484
+ "loss": 0.3711,
485
+ "step": 68
486
+ },
487
+ {
488
+ "epoch": 2.76,
489
+ "grad_norm": 0.5664202074794812,
490
+ "learning_rate": 1.9657406211579966e-07,
491
+ "loss": 0.3356,
492
+ "step": 69
493
+ },
494
+ {
495
+ "epoch": 2.8,
496
+ "grad_norm": 0.6530867966383043,
497
+ "learning_rate": 1.3678547337593494e-07,
498
+ "loss": 0.3626,
499
+ "step": 70
500
+ },
501
+ {
502
+ "epoch": 2.84,
503
+ "grad_norm": 0.6322961093826559,
504
+ "learning_rate": 8.768728562211948e-08,
505
+ "loss": 0.3477,
506
+ "step": 71
507
+ },
508
+ {
509
+ "epoch": 2.88,
510
+ "grad_norm": 0.5931867745636884,
511
+ "learning_rate": 4.9387427326745287e-08,
512
+ "loss": 0.3456,
513
+ "step": 72
514
+ },
515
+ {
516
+ "epoch": 2.92,
517
+ "grad_norm": 0.6598198778563147,
518
+ "learning_rate": 2.1970089890509527e-08,
519
+ "loss": 0.3885,
520
+ "step": 73
521
+ },
522
+ {
523
+ "epoch": 2.96,
524
+ "grad_norm": 0.6197517869539098,
525
+ "learning_rate": 5.495542571443135e-09,
526
+ "loss": 0.3517,
527
+ "step": 74
528
+ },
529
+ {
530
+ "epoch": 3.0,
531
+ "grad_norm": 0.612110955289339,
532
+ "learning_rate": 0.0,
533
+ "loss": 0.3648,
534
+ "step": 75
535
+ },
536
+ {
537
+ "epoch": 3.0,
538
+ "step": 75,
539
+ "total_flos": 20372648230912.0,
540
+ "train_loss": 0.4803738268216451,
541
+ "train_runtime": 543.6715,
542
+ "train_samples_per_second": 13.243,
543
+ "train_steps_per_second": 0.138
544
+ }
545
+ ],
546
+ "logging_steps": 1,
547
+ "max_steps": 75,
548
+ "num_input_tokens_seen": 0,
549
+ "num_train_epochs": 3,
550
+ "save_steps": 500,
551
+ "stateful_callbacks": {
552
+ "TrainerControl": {
553
+ "args": {
554
+ "should_epoch_stop": false,
555
+ "should_evaluate": false,
556
+ "should_log": false,
557
+ "should_save": true,
558
+ "should_training_stop": true
559
+ },
560
+ "attributes": {}
561
+ }
562
+ },
563
+ "total_flos": 20372648230912.0,
564
+ "train_batch_size": 1,
565
+ "trial_name": null,
566
+ "trial_params": null
567
+ }
training_loss.png ADDED