aadityap commited on
Commit
3086564
·
verified ·
1 Parent(s): 0f2df80

Model save

Browse files
Files changed (4) hide show
  1. README.md +61 -0
  2. all_results.json +9 -0
  3. train_results.json +9 -0
  4. trainer_state.json +490 -0
README.md ADDED
@@ -0,0 +1,61 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ license: mit
4
+ base_model: deepseek-ai/DeepSeek-R1-Distill-Qwen-32B
5
+ tags:
6
+ - trl
7
+ - sft
8
+ - generated_from_trainer
9
+ model-index:
10
+ - name: models-ttt-problem-10-replicate-2-step1
11
+ results: []
12
+ ---
13
+
14
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
15
+ should probably proofread and complete it, then remove this comment. -->
16
+
17
+ # models-ttt-problem-10-replicate-2-step1
18
+
19
+ This model is a fine-tuned version of [deepseek-ai/DeepSeek-R1-Distill-Qwen-32B](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-32B) on an unknown dataset.
20
+
21
+ ## Model description
22
+
23
+ More information needed
24
+
25
+ ## Intended uses & limitations
26
+
27
+ More information needed
28
+
29
+ ## Training and evaluation data
30
+
31
+ More information needed
32
+
33
+ ## Training procedure
34
+
35
+ ### Training hyperparameters
36
+
37
+ The following hyperparameters were used during training:
38
+ - learning_rate: 5e-05
39
+ - train_batch_size: 1
40
+ - eval_batch_size: 1
41
+ - seed: 42
42
+ - distributed_type: multi-GPU
43
+ - num_devices: 8
44
+ - total_train_batch_size: 8
45
+ - total_eval_batch_size: 8
46
+ - optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
47
+ - lr_scheduler_type: cosine
48
+ - lr_scheduler_warmup_ratio: 0.1
49
+ - num_epochs: 2
50
+
51
+ ### Training results
52
+
53
+
54
+
55
+ ### Framework versions
56
+
57
+ - PEFT 0.13.2
58
+ - Transformers 4.47.0.dev0
59
+ - Pytorch 2.4.0+cu121
60
+ - Datasets 3.1.0
61
+ - Tokenizers 0.20.3
all_results.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.0,
3
+ "total_flos": 570726028935168.0,
4
+ "train_loss": 0.11890232225414366,
5
+ "train_runtime": 1057.719,
6
+ "train_samples": 250,
7
+ "train_samples_per_second": 0.473,
8
+ "train_steps_per_second": 0.061
9
+ }
train_results.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.0,
3
+ "total_flos": 570726028935168.0,
4
+ "train_loss": 0.11890232225414366,
5
+ "train_runtime": 1057.719,
6
+ "train_samples": 250,
7
+ "train_samples_per_second": 0.473,
8
+ "train_steps_per_second": 0.061
9
+ }
trainer_state.json ADDED
@@ -0,0 +1,490 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 2.0,
5
+ "eval_steps": 500,
6
+ "global_step": 64,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.03125,
13
+ "grad_norm": 0.042211709723423994,
14
+ "learning_rate": 7.142857142857143e-06,
15
+ "loss": 0.1826,
16
+ "step": 1
17
+ },
18
+ {
19
+ "epoch": 0.0625,
20
+ "grad_norm": 0.040523357558985305,
21
+ "learning_rate": 1.4285714285714285e-05,
22
+ "loss": 0.1697,
23
+ "step": 2
24
+ },
25
+ {
26
+ "epoch": 0.09375,
27
+ "grad_norm": 0.04167011745239068,
28
+ "learning_rate": 2.1428571428571428e-05,
29
+ "loss": 0.1922,
30
+ "step": 3
31
+ },
32
+ {
33
+ "epoch": 0.125,
34
+ "grad_norm": 0.04332694019851011,
35
+ "learning_rate": 2.857142857142857e-05,
36
+ "loss": 0.1826,
37
+ "step": 4
38
+ },
39
+ {
40
+ "epoch": 0.15625,
41
+ "grad_norm": 0.04289781626308719,
42
+ "learning_rate": 3.571428571428572e-05,
43
+ "loss": 0.1933,
44
+ "step": 5
45
+ },
46
+ {
47
+ "epoch": 0.1875,
48
+ "grad_norm": 0.048927401557601384,
49
+ "learning_rate": 4.2857142857142856e-05,
50
+ "loss": 0.1792,
51
+ "step": 6
52
+ },
53
+ {
54
+ "epoch": 0.21875,
55
+ "grad_norm": 0.04917933368077925,
56
+ "learning_rate": 5e-05,
57
+ "loss": 0.1785,
58
+ "step": 7
59
+ },
60
+ {
61
+ "epoch": 0.25,
62
+ "grad_norm": 0.042531202393428634,
63
+ "learning_rate": 4.996203791083291e-05,
64
+ "loss": 0.1782,
65
+ "step": 8
66
+ },
67
+ {
68
+ "epoch": 0.28125,
69
+ "grad_norm": 0.03423472324986091,
70
+ "learning_rate": 4.984826693294874e-05,
71
+ "loss": 0.1553,
72
+ "step": 9
73
+ },
74
+ {
75
+ "epoch": 0.3125,
76
+ "grad_norm": 0.02536269873026519,
77
+ "learning_rate": 4.965903258506806e-05,
78
+ "loss": 0.162,
79
+ "step": 10
80
+ },
81
+ {
82
+ "epoch": 0.34375,
83
+ "grad_norm": 0.03414579032479303,
84
+ "learning_rate": 4.9394909565685894e-05,
85
+ "loss": 0.1805,
86
+ "step": 11
87
+ },
88
+ {
89
+ "epoch": 0.375,
90
+ "grad_norm": 0.04499054740361273,
91
+ "learning_rate": 4.905670000773126e-05,
92
+ "loss": 0.1633,
93
+ "step": 12
94
+ },
95
+ {
96
+ "epoch": 0.40625,
97
+ "grad_norm": 0.054239841431854424,
98
+ "learning_rate": 4.864543104251587e-05,
99
+ "loss": 0.1632,
100
+ "step": 13
101
+ },
102
+ {
103
+ "epoch": 0.4375,
104
+ "grad_norm": 0.04808977247007115,
105
+ "learning_rate": 4.8162351680370044e-05,
106
+ "loss": 0.1491,
107
+ "step": 14
108
+ },
109
+ {
110
+ "epoch": 0.46875,
111
+ "grad_norm": 0.04220777198225847,
112
+ "learning_rate": 4.760892901743944e-05,
113
+ "loss": 0.1504,
114
+ "step": 15
115
+ },
116
+ {
117
+ "epoch": 0.5,
118
+ "grad_norm": 0.02944080547232296,
119
+ "learning_rate": 4.698684378016222e-05,
120
+ "loss": 0.1476,
121
+ "step": 16
122
+ },
123
+ {
124
+ "epoch": 0.53125,
125
+ "grad_norm": 0.022946753850242432,
126
+ "learning_rate": 4.629798522095818e-05,
127
+ "loss": 0.1398,
128
+ "step": 17
129
+ },
130
+ {
131
+ "epoch": 0.5625,
132
+ "grad_norm": 0.018713023972775928,
133
+ "learning_rate": 4.554444538063113e-05,
134
+ "loss": 0.1586,
135
+ "step": 18
136
+ },
137
+ {
138
+ "epoch": 0.59375,
139
+ "grad_norm": 0.01989174806605767,
140
+ "learning_rate": 4.4728512734909844e-05,
141
+ "loss": 0.1378,
142
+ "step": 19
143
+ },
144
+ {
145
+ "epoch": 0.625,
146
+ "grad_norm": 0.023054897121146254,
147
+ "learning_rate": 4.385266524442241e-05,
148
+ "loss": 0.1334,
149
+ "step": 20
150
+ },
151
+ {
152
+ "epoch": 0.65625,
153
+ "grad_norm": 0.02399880426486958,
154
+ "learning_rate": 4.2919562829211283e-05,
155
+ "loss": 0.1246,
156
+ "step": 21
157
+ },
158
+ {
159
+ "epoch": 0.6875,
160
+ "grad_norm": 0.027211398286697646,
161
+ "learning_rate": 4.193203929064353e-05,
162
+ "loss": 0.142,
163
+ "step": 22
164
+ },
165
+ {
166
+ "epoch": 0.71875,
167
+ "grad_norm": 0.02336043662637302,
168
+ "learning_rate": 4.089309370524921e-05,
169
+ "loss": 0.1173,
170
+ "step": 23
171
+ },
172
+ {
173
+ "epoch": 0.75,
174
+ "grad_norm": 0.025261125777024232,
175
+ "learning_rate": 3.9805881316624506e-05,
176
+ "loss": 0.1259,
177
+ "step": 24
178
+ },
179
+ {
180
+ "epoch": 0.78125,
181
+ "grad_norm": 0.02143183000772084,
182
+ "learning_rate": 3.867370395306068e-05,
183
+ "loss": 0.1205,
184
+ "step": 25
185
+ },
186
+ {
187
+ "epoch": 0.8125,
188
+ "grad_norm": 0.022292087028719917,
189
+ "learning_rate": 3.7500000000000003e-05,
190
+ "loss": 0.1155,
191
+ "step": 26
192
+ },
193
+ {
194
+ "epoch": 0.84375,
195
+ "grad_norm": 0.019679819265901136,
196
+ "learning_rate": 3.628833395777224e-05,
197
+ "loss": 0.1232,
198
+ "step": 27
199
+ },
200
+ {
201
+ "epoch": 0.875,
202
+ "grad_norm": 0.020714109057838017,
203
+ "learning_rate": 3.504238561632424e-05,
204
+ "loss": 0.1168,
205
+ "step": 28
206
+ },
207
+ {
208
+ "epoch": 0.90625,
209
+ "grad_norm": 0.019403140215305675,
210
+ "learning_rate": 3.376593887981887e-05,
211
+ "loss": 0.1025,
212
+ "step": 29
213
+ },
214
+ {
215
+ "epoch": 0.9375,
216
+ "grad_norm": 0.018873000340986552,
217
+ "learning_rate": 3.246287027504237e-05,
218
+ "loss": 0.1075,
219
+ "step": 30
220
+ },
221
+ {
222
+ "epoch": 0.96875,
223
+ "grad_norm": 0.019583048283476286,
224
+ "learning_rate": 3.1137137178519985e-05,
225
+ "loss": 0.0966,
226
+ "step": 31
227
+ },
228
+ {
229
+ "epoch": 1.0,
230
+ "grad_norm": 0.020023991728142915,
231
+ "learning_rate": 2.9792765798093465e-05,
232
+ "loss": 0.1077,
233
+ "step": 32
234
+ },
235
+ {
236
+ "epoch": 1.03125,
237
+ "grad_norm": 0.02014336988909588,
238
+ "learning_rate": 2.8433838945460205e-05,
239
+ "loss": 0.114,
240
+ "step": 33
241
+ },
242
+ {
243
+ "epoch": 1.0625,
244
+ "grad_norm": 0.018645793178802845,
245
+ "learning_rate": 2.7064483636808313e-05,
246
+ "loss": 0.1129,
247
+ "step": 34
248
+ },
249
+ {
250
+ "epoch": 1.09375,
251
+ "grad_norm": 0.017769270453618723,
252
+ "learning_rate": 2.5688858559204053e-05,
253
+ "loss": 0.0885,
254
+ "step": 35
255
+ },
256
+ {
257
+ "epoch": 1.125,
258
+ "grad_norm": 0.018237269189223717,
259
+ "learning_rate": 2.4311141440795953e-05,
260
+ "loss": 0.1085,
261
+ "step": 36
262
+ },
263
+ {
264
+ "epoch": 1.15625,
265
+ "grad_norm": 0.019420564407489134,
266
+ "learning_rate": 2.2935516363191693e-05,
267
+ "loss": 0.1009,
268
+ "step": 37
269
+ },
270
+ {
271
+ "epoch": 1.1875,
272
+ "grad_norm": 0.01818503869300628,
273
+ "learning_rate": 2.1566161054539798e-05,
274
+ "loss": 0.0909,
275
+ "step": 38
276
+ },
277
+ {
278
+ "epoch": 1.21875,
279
+ "grad_norm": 0.01977177642557231,
280
+ "learning_rate": 2.0207234201906547e-05,
281
+ "loss": 0.0993,
282
+ "step": 39
283
+ },
284
+ {
285
+ "epoch": 1.25,
286
+ "grad_norm": 0.018860196742925606,
287
+ "learning_rate": 1.8862862821480025e-05,
288
+ "loss": 0.0881,
289
+ "step": 40
290
+ },
291
+ {
292
+ "epoch": 1.28125,
293
+ "grad_norm": 0.01789538986871461,
294
+ "learning_rate": 1.7537129724957642e-05,
295
+ "loss": 0.1031,
296
+ "step": 41
297
+ },
298
+ {
299
+ "epoch": 1.3125,
300
+ "grad_norm": 0.019440146647284384,
301
+ "learning_rate": 1.6234061120181142e-05,
302
+ "loss": 0.1011,
303
+ "step": 42
304
+ },
305
+ {
306
+ "epoch": 1.34375,
307
+ "grad_norm": 0.01773027173909912,
308
+ "learning_rate": 1.495761438367577e-05,
309
+ "loss": 0.0873,
310
+ "step": 43
311
+ },
312
+ {
313
+ "epoch": 1.375,
314
+ "grad_norm": 0.018053120975854698,
315
+ "learning_rate": 1.3711666042227772e-05,
316
+ "loss": 0.0942,
317
+ "step": 44
318
+ },
319
+ {
320
+ "epoch": 1.40625,
321
+ "grad_norm": 0.01599530366821666,
322
+ "learning_rate": 1.2500000000000006e-05,
323
+ "loss": 0.0864,
324
+ "step": 45
325
+ },
326
+ {
327
+ "epoch": 1.4375,
328
+ "grad_norm": 0.018506143523795423,
329
+ "learning_rate": 1.1326296046939333e-05,
330
+ "loss": 0.0897,
331
+ "step": 46
332
+ },
333
+ {
334
+ "epoch": 1.46875,
335
+ "grad_norm": 0.019285636800618775,
336
+ "learning_rate": 1.0194118683375503e-05,
337
+ "loss": 0.0945,
338
+ "step": 47
339
+ },
340
+ {
341
+ "epoch": 1.5,
342
+ "grad_norm": 0.01851733435975437,
343
+ "learning_rate": 9.106906294750805e-06,
344
+ "loss": 0.089,
345
+ "step": 48
346
+ },
347
+ {
348
+ "epoch": 1.53125,
349
+ "grad_norm": 0.01615116629140626,
350
+ "learning_rate": 8.067960709356478e-06,
351
+ "loss": 0.0829,
352
+ "step": 49
353
+ },
354
+ {
355
+ "epoch": 1.5625,
356
+ "grad_norm": 0.01999530228835859,
357
+ "learning_rate": 7.080437170788723e-06,
358
+ "loss": 0.0806,
359
+ "step": 50
360
+ },
361
+ {
362
+ "epoch": 1.59375,
363
+ "grad_norm": 0.018698060874040544,
364
+ "learning_rate": 6.147334755577596e-06,
365
+ "loss": 0.0745,
366
+ "step": 51
367
+ },
368
+ {
369
+ "epoch": 1.625,
370
+ "grad_norm": 0.020647083613217352,
371
+ "learning_rate": 5.271487265090163e-06,
372
+ "loss": 0.0762,
373
+ "step": 52
374
+ },
375
+ {
376
+ "epoch": 1.65625,
377
+ "grad_norm": 0.018963763952432634,
378
+ "learning_rate": 4.4555546193688735e-06,
379
+ "loss": 0.0767,
380
+ "step": 53
381
+ },
382
+ {
383
+ "epoch": 1.6875,
384
+ "grad_norm": 0.016352645860581026,
385
+ "learning_rate": 3.7020147790418263e-06,
386
+ "loss": 0.0886,
387
+ "step": 54
388
+ },
389
+ {
390
+ "epoch": 1.71875,
391
+ "grad_norm": 0.018189380607389477,
392
+ "learning_rate": 3.013156219837776e-06,
393
+ "loss": 0.0872,
394
+ "step": 55
395
+ },
396
+ {
397
+ "epoch": 1.75,
398
+ "grad_norm": 0.016586579136002214,
399
+ "learning_rate": 2.391070982560564e-06,
400
+ "loss": 0.0748,
401
+ "step": 56
402
+ },
403
+ {
404
+ "epoch": 1.78125,
405
+ "grad_norm": 0.017951491429409636,
406
+ "learning_rate": 1.837648319629956e-06,
407
+ "loss": 0.111,
408
+ "step": 57
409
+ },
410
+ {
411
+ "epoch": 1.8125,
412
+ "grad_norm": 0.016524711157425227,
413
+ "learning_rate": 1.3545689574841342e-06,
414
+ "loss": 0.0906,
415
+ "step": 58
416
+ },
417
+ {
418
+ "epoch": 1.84375,
419
+ "grad_norm": 0.016444127025366004,
420
+ "learning_rate": 9.432999922687396e-07,
421
+ "loss": 0.0817,
422
+ "step": 59
423
+ },
424
+ {
425
+ "epoch": 1.875,
426
+ "grad_norm": 0.01799234707191234,
427
+ "learning_rate": 6.050904343141095e-07,
428
+ "loss": 0.0969,
429
+ "step": 60
430
+ },
431
+ {
432
+ "epoch": 1.90625,
433
+ "grad_norm": 0.020873493821065264,
434
+ "learning_rate": 3.4096741493194197e-07,
435
+ "loss": 0.0921,
436
+ "step": 61
437
+ },
438
+ {
439
+ "epoch": 1.9375,
440
+ "grad_norm": 0.016975329426932043,
441
+ "learning_rate": 1.517330670512629e-07,
442
+ "loss": 0.0818,
443
+ "step": 62
444
+ },
445
+ {
446
+ "epoch": 1.96875,
447
+ "grad_norm": 0.016262557073136383,
448
+ "learning_rate": 3.796208916709565e-08,
449
+ "loss": 0.0846,
450
+ "step": 63
451
+ },
452
+ {
453
+ "epoch": 2.0,
454
+ "grad_norm": 0.016105491523942778,
455
+ "learning_rate": 0.0,
456
+ "loss": 0.0837,
457
+ "step": 64
458
+ },
459
+ {
460
+ "epoch": 2.0,
461
+ "step": 64,
462
+ "total_flos": 570726028935168.0,
463
+ "train_loss": 0.11890232225414366,
464
+ "train_runtime": 1057.719,
465
+ "train_samples_per_second": 0.473,
466
+ "train_steps_per_second": 0.061
467
+ }
468
+ ],
469
+ "logging_steps": 1,
470
+ "max_steps": 64,
471
+ "num_input_tokens_seen": 0,
472
+ "num_train_epochs": 2,
473
+ "save_steps": 500,
474
+ "stateful_callbacks": {
475
+ "TrainerControl": {
476
+ "args": {
477
+ "should_epoch_stop": false,
478
+ "should_evaluate": false,
479
+ "should_log": false,
480
+ "should_save": true,
481
+ "should_training_stop": true
482
+ },
483
+ "attributes": {}
484
+ }
485
+ },
486
+ "total_flos": 570726028935168.0,
487
+ "train_batch_size": 1,
488
+ "trial_name": null,
489
+ "trial_params": null
490
+ }