sedrickkeh commited on
Commit
9ceeb0a
·
verified ·
1 Parent(s): 566f855

End of training

Browse files
README.md CHANGED
@@ -4,6 +4,7 @@ license: llama3.1
4
  base_model: meta-llama/Meta-Llama-3.1-8B
5
  tags:
6
  - llama-factory
 
7
  - generated_from_trainer
8
  model-index:
9
  - name: oh_v1.3_metamath_x2
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
15
 
16
  # oh_v1.3_metamath_x2
17
 
18
- This model is a fine-tuned version of [meta-llama/Meta-Llama-3.1-8B](https://huggingface.co/meta-llama/Meta-Llama-3.1-8B) on an unknown dataset.
19
  It achieves the following results on the evaluation set:
20
  - Loss: 0.7103
21
 
 
4
  base_model: meta-llama/Meta-Llama-3.1-8B
5
  tags:
6
  - llama-factory
7
+ - full
8
  - generated_from_trainer
9
  model-index:
10
  - name: oh_v1.3_metamath_x2
 
16
 
17
  # oh_v1.3_metamath_x2
18
 
19
+ This model is a fine-tuned version of [meta-llama/Meta-Llama-3.1-8B](https://huggingface.co/meta-llama/Meta-Llama-3.1-8B) on the mlfoundations-dev/oh_v1.3_metamath_x2 dataset.
20
  It achieves the following results on the evaluation set:
21
  - Loss: 0.7103
22
 
all_results.json ADDED
@@ -0,0 +1,12 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.9956063268892796,
3
+ "eval_loss": 0.7103263735771179,
4
+ "eval_runtime": 302.3687,
5
+ "eval_samples_per_second": 25.356,
6
+ "eval_steps_per_second": 0.397,
7
+ "total_flos": 1426922353459200.0,
8
+ "train_loss": 0.6913109551852857,
9
+ "train_runtime": 50624.2334,
10
+ "train_samples_per_second": 8.632,
11
+ "train_steps_per_second": 0.017
12
+ }
eval_results.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.9956063268892796,
3
+ "eval_loss": 0.7103263735771179,
4
+ "eval_runtime": 302.3687,
5
+ "eval_samples_per_second": 25.356,
6
+ "eval_steps_per_second": 0.397
7
+ }
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.9956063268892796,
3
+ "total_flos": 1426922353459200.0,
4
+ "train_loss": 0.6913109551852857,
5
+ "train_runtime": 50624.2334,
6
+ "train_samples_per_second": 8.632,
7
+ "train_steps_per_second": 0.017
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,661 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 2.9956063268892796,
5
+ "eval_steps": 500,
6
+ "global_step": 852,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.0351493848857645,
13
+ "grad_norm": 6.865658287926604,
14
+ "learning_rate": 5e-06,
15
+ "loss": 1.0224,
16
+ "step": 10
17
+ },
18
+ {
19
+ "epoch": 0.070298769771529,
20
+ "grad_norm": 1.300129454515723,
21
+ "learning_rate": 5e-06,
22
+ "loss": 0.8893,
23
+ "step": 20
24
+ },
25
+ {
26
+ "epoch": 0.1054481546572935,
27
+ "grad_norm": 1.662141333626877,
28
+ "learning_rate": 5e-06,
29
+ "loss": 0.8461,
30
+ "step": 30
31
+ },
32
+ {
33
+ "epoch": 0.140597539543058,
34
+ "grad_norm": 1.1867317674423972,
35
+ "learning_rate": 5e-06,
36
+ "loss": 0.8221,
37
+ "step": 40
38
+ },
39
+ {
40
+ "epoch": 0.1757469244288225,
41
+ "grad_norm": 1.3406663541941202,
42
+ "learning_rate": 5e-06,
43
+ "loss": 0.8082,
44
+ "step": 50
45
+ },
46
+ {
47
+ "epoch": 0.210896309314587,
48
+ "grad_norm": 0.9748883756928828,
49
+ "learning_rate": 5e-06,
50
+ "loss": 0.7926,
51
+ "step": 60
52
+ },
53
+ {
54
+ "epoch": 0.2460456942003515,
55
+ "grad_norm": 0.9049088495136057,
56
+ "learning_rate": 5e-06,
57
+ "loss": 0.7786,
58
+ "step": 70
59
+ },
60
+ {
61
+ "epoch": 0.281195079086116,
62
+ "grad_norm": 0.7756246074861972,
63
+ "learning_rate": 5e-06,
64
+ "loss": 0.7672,
65
+ "step": 80
66
+ },
67
+ {
68
+ "epoch": 0.3163444639718805,
69
+ "grad_norm": 0.6803113923201367,
70
+ "learning_rate": 5e-06,
71
+ "loss": 0.7666,
72
+ "step": 90
73
+ },
74
+ {
75
+ "epoch": 0.351493848857645,
76
+ "grad_norm": 0.7087390396178761,
77
+ "learning_rate": 5e-06,
78
+ "loss": 0.7626,
79
+ "step": 100
80
+ },
81
+ {
82
+ "epoch": 0.3866432337434095,
83
+ "grad_norm": 0.6904558645177432,
84
+ "learning_rate": 5e-06,
85
+ "loss": 0.7558,
86
+ "step": 110
87
+ },
88
+ {
89
+ "epoch": 0.421792618629174,
90
+ "grad_norm": 0.5406381121382178,
91
+ "learning_rate": 5e-06,
92
+ "loss": 0.751,
93
+ "step": 120
94
+ },
95
+ {
96
+ "epoch": 0.45694200351493847,
97
+ "grad_norm": 0.7263515515258443,
98
+ "learning_rate": 5e-06,
99
+ "loss": 0.7541,
100
+ "step": 130
101
+ },
102
+ {
103
+ "epoch": 0.492091388400703,
104
+ "grad_norm": 0.6381225954297634,
105
+ "learning_rate": 5e-06,
106
+ "loss": 0.7507,
107
+ "step": 140
108
+ },
109
+ {
110
+ "epoch": 0.5272407732864675,
111
+ "grad_norm": 0.9282926398227679,
112
+ "learning_rate": 5e-06,
113
+ "loss": 0.7425,
114
+ "step": 150
115
+ },
116
+ {
117
+ "epoch": 0.562390158172232,
118
+ "grad_norm": 0.703837658050583,
119
+ "learning_rate": 5e-06,
120
+ "loss": 0.7473,
121
+ "step": 160
122
+ },
123
+ {
124
+ "epoch": 0.5975395430579965,
125
+ "grad_norm": 0.7762432087380096,
126
+ "learning_rate": 5e-06,
127
+ "loss": 0.7408,
128
+ "step": 170
129
+ },
130
+ {
131
+ "epoch": 0.632688927943761,
132
+ "grad_norm": 0.5947662547404722,
133
+ "learning_rate": 5e-06,
134
+ "loss": 0.7347,
135
+ "step": 180
136
+ },
137
+ {
138
+ "epoch": 0.6678383128295254,
139
+ "grad_norm": 0.628944705791063,
140
+ "learning_rate": 5e-06,
141
+ "loss": 0.7406,
142
+ "step": 190
143
+ },
144
+ {
145
+ "epoch": 0.70298769771529,
146
+ "grad_norm": 0.5977406811055224,
147
+ "learning_rate": 5e-06,
148
+ "loss": 0.7347,
149
+ "step": 200
150
+ },
151
+ {
152
+ "epoch": 0.7381370826010545,
153
+ "grad_norm": 0.582107769314153,
154
+ "learning_rate": 5e-06,
155
+ "loss": 0.737,
156
+ "step": 210
157
+ },
158
+ {
159
+ "epoch": 0.773286467486819,
160
+ "grad_norm": 0.6326552735959291,
161
+ "learning_rate": 5e-06,
162
+ "loss": 0.7328,
163
+ "step": 220
164
+ },
165
+ {
166
+ "epoch": 0.8084358523725835,
167
+ "grad_norm": 0.6637547116847639,
168
+ "learning_rate": 5e-06,
169
+ "loss": 0.7311,
170
+ "step": 230
171
+ },
172
+ {
173
+ "epoch": 0.843585237258348,
174
+ "grad_norm": 0.6997143410926964,
175
+ "learning_rate": 5e-06,
176
+ "loss": 0.7341,
177
+ "step": 240
178
+ },
179
+ {
180
+ "epoch": 0.8787346221441125,
181
+ "grad_norm": 0.6162729226466245,
182
+ "learning_rate": 5e-06,
183
+ "loss": 0.7332,
184
+ "step": 250
185
+ },
186
+ {
187
+ "epoch": 0.9138840070298769,
188
+ "grad_norm": 0.6199166403621413,
189
+ "learning_rate": 5e-06,
190
+ "loss": 0.7262,
191
+ "step": 260
192
+ },
193
+ {
194
+ "epoch": 0.9490333919156415,
195
+ "grad_norm": 0.6034966296550427,
196
+ "learning_rate": 5e-06,
197
+ "loss": 0.729,
198
+ "step": 270
199
+ },
200
+ {
201
+ "epoch": 0.984182776801406,
202
+ "grad_norm": 0.6195682554180708,
203
+ "learning_rate": 5e-06,
204
+ "loss": 0.7264,
205
+ "step": 280
206
+ },
207
+ {
208
+ "epoch": 0.9982425307557118,
209
+ "eval_loss": 0.721891462802887,
210
+ "eval_runtime": 302.9942,
211
+ "eval_samples_per_second": 25.304,
212
+ "eval_steps_per_second": 0.396,
213
+ "step": 284
214
+ },
215
+ {
216
+ "epoch": 1.0197715289982425,
217
+ "grad_norm": 0.6617437136575776,
218
+ "learning_rate": 5e-06,
219
+ "loss": 0.7503,
220
+ "step": 290
221
+ },
222
+ {
223
+ "epoch": 1.054920913884007,
224
+ "grad_norm": 0.596469668606961,
225
+ "learning_rate": 5e-06,
226
+ "loss": 0.6752,
227
+ "step": 300
228
+ },
229
+ {
230
+ "epoch": 1.0900702987697715,
231
+ "grad_norm": 0.7286257897811691,
232
+ "learning_rate": 5e-06,
233
+ "loss": 0.6745,
234
+ "step": 310
235
+ },
236
+ {
237
+ "epoch": 1.1252196836555362,
238
+ "grad_norm": 0.6703518701287363,
239
+ "learning_rate": 5e-06,
240
+ "loss": 0.6772,
241
+ "step": 320
242
+ },
243
+ {
244
+ "epoch": 1.1603690685413006,
245
+ "grad_norm": 0.6678193952959378,
246
+ "learning_rate": 5e-06,
247
+ "loss": 0.6757,
248
+ "step": 330
249
+ },
250
+ {
251
+ "epoch": 1.195518453427065,
252
+ "grad_norm": 0.6272061731880971,
253
+ "learning_rate": 5e-06,
254
+ "loss": 0.6772,
255
+ "step": 340
256
+ },
257
+ {
258
+ "epoch": 1.2306678383128296,
259
+ "grad_norm": 0.5618776589312474,
260
+ "learning_rate": 5e-06,
261
+ "loss": 0.6767,
262
+ "step": 350
263
+ },
264
+ {
265
+ "epoch": 1.265817223198594,
266
+ "grad_norm": 0.656461597570214,
267
+ "learning_rate": 5e-06,
268
+ "loss": 0.6742,
269
+ "step": 360
270
+ },
271
+ {
272
+ "epoch": 1.3009666080843585,
273
+ "grad_norm": 0.95088298783439,
274
+ "learning_rate": 5e-06,
275
+ "loss": 0.6803,
276
+ "step": 370
277
+ },
278
+ {
279
+ "epoch": 1.336115992970123,
280
+ "grad_norm": 0.6256062888068228,
281
+ "learning_rate": 5e-06,
282
+ "loss": 0.6789,
283
+ "step": 380
284
+ },
285
+ {
286
+ "epoch": 1.3712653778558876,
287
+ "grad_norm": 0.5178529345876333,
288
+ "learning_rate": 5e-06,
289
+ "loss": 0.6757,
290
+ "step": 390
291
+ },
292
+ {
293
+ "epoch": 1.406414762741652,
294
+ "grad_norm": 0.6633111117626306,
295
+ "learning_rate": 5e-06,
296
+ "loss": 0.6786,
297
+ "step": 400
298
+ },
299
+ {
300
+ "epoch": 1.4415641476274166,
301
+ "grad_norm": 0.5753214727933854,
302
+ "learning_rate": 5e-06,
303
+ "loss": 0.6686,
304
+ "step": 410
305
+ },
306
+ {
307
+ "epoch": 1.476713532513181,
308
+ "grad_norm": 0.7023169996268164,
309
+ "learning_rate": 5e-06,
310
+ "loss": 0.674,
311
+ "step": 420
312
+ },
313
+ {
314
+ "epoch": 1.5118629173989455,
315
+ "grad_norm": 0.601050736097527,
316
+ "learning_rate": 5e-06,
317
+ "loss": 0.676,
318
+ "step": 430
319
+ },
320
+ {
321
+ "epoch": 1.54701230228471,
322
+ "grad_norm": 0.6375081303020413,
323
+ "learning_rate": 5e-06,
324
+ "loss": 0.6779,
325
+ "step": 440
326
+ },
327
+ {
328
+ "epoch": 1.5821616871704745,
329
+ "grad_norm": 0.6076189491485879,
330
+ "learning_rate": 5e-06,
331
+ "loss": 0.6811,
332
+ "step": 450
333
+ },
334
+ {
335
+ "epoch": 1.6173110720562391,
336
+ "grad_norm": 0.6123755131309624,
337
+ "learning_rate": 5e-06,
338
+ "loss": 0.6749,
339
+ "step": 460
340
+ },
341
+ {
342
+ "epoch": 1.6524604569420034,
343
+ "grad_norm": 0.5847478738087437,
344
+ "learning_rate": 5e-06,
345
+ "loss": 0.6747,
346
+ "step": 470
347
+ },
348
+ {
349
+ "epoch": 1.687609841827768,
350
+ "grad_norm": 0.6041574755100807,
351
+ "learning_rate": 5e-06,
352
+ "loss": 0.6681,
353
+ "step": 480
354
+ },
355
+ {
356
+ "epoch": 1.7227592267135325,
357
+ "grad_norm": 0.6972984159432736,
358
+ "learning_rate": 5e-06,
359
+ "loss": 0.6686,
360
+ "step": 490
361
+ },
362
+ {
363
+ "epoch": 1.757908611599297,
364
+ "grad_norm": 0.592332789109461,
365
+ "learning_rate": 5e-06,
366
+ "loss": 0.6763,
367
+ "step": 500
368
+ },
369
+ {
370
+ "epoch": 1.7930579964850615,
371
+ "grad_norm": 0.7081266254056617,
372
+ "learning_rate": 5e-06,
373
+ "loss": 0.6707,
374
+ "step": 510
375
+ },
376
+ {
377
+ "epoch": 1.828207381370826,
378
+ "grad_norm": 0.5655551892586738,
379
+ "learning_rate": 5e-06,
380
+ "loss": 0.675,
381
+ "step": 520
382
+ },
383
+ {
384
+ "epoch": 1.8633567662565906,
385
+ "grad_norm": 0.5912936045849521,
386
+ "learning_rate": 5e-06,
387
+ "loss": 0.6746,
388
+ "step": 530
389
+ },
390
+ {
391
+ "epoch": 1.8985061511423549,
392
+ "grad_norm": 0.6418543236430647,
393
+ "learning_rate": 5e-06,
394
+ "loss": 0.6743,
395
+ "step": 540
396
+ },
397
+ {
398
+ "epoch": 1.9336555360281196,
399
+ "grad_norm": 0.8406203952305934,
400
+ "learning_rate": 5e-06,
401
+ "loss": 0.6777,
402
+ "step": 550
403
+ },
404
+ {
405
+ "epoch": 1.968804920913884,
406
+ "grad_norm": 0.6740623987469322,
407
+ "learning_rate": 5e-06,
408
+ "loss": 0.6783,
409
+ "step": 560
410
+ },
411
+ {
412
+ "epoch": 1.9969244288224957,
413
+ "eval_loss": 0.7088373899459839,
414
+ "eval_runtime": 302.5633,
415
+ "eval_samples_per_second": 25.34,
416
+ "eval_steps_per_second": 0.397,
417
+ "step": 568
418
+ },
419
+ {
420
+ "epoch": 2.0043936731107204,
421
+ "grad_norm": 0.8982450142296012,
422
+ "learning_rate": 5e-06,
423
+ "loss": 0.7079,
424
+ "step": 570
425
+ },
426
+ {
427
+ "epoch": 2.039543057996485,
428
+ "grad_norm": 0.9691008221084222,
429
+ "learning_rate": 5e-06,
430
+ "loss": 0.6203,
431
+ "step": 580
432
+ },
433
+ {
434
+ "epoch": 2.0746924428822497,
435
+ "grad_norm": 0.6447824341633516,
436
+ "learning_rate": 5e-06,
437
+ "loss": 0.6244,
438
+ "step": 590
439
+ },
440
+ {
441
+ "epoch": 2.109841827768014,
442
+ "grad_norm": 0.7064323342581214,
443
+ "learning_rate": 5e-06,
444
+ "loss": 0.6189,
445
+ "step": 600
446
+ },
447
+ {
448
+ "epoch": 2.1449912126537787,
449
+ "grad_norm": 0.5819596596280016,
450
+ "learning_rate": 5e-06,
451
+ "loss": 0.6207,
452
+ "step": 610
453
+ },
454
+ {
455
+ "epoch": 2.180140597539543,
456
+ "grad_norm": 0.7981926624790863,
457
+ "learning_rate": 5e-06,
458
+ "loss": 0.6203,
459
+ "step": 620
460
+ },
461
+ {
462
+ "epoch": 2.2152899824253076,
463
+ "grad_norm": 0.6853162161955834,
464
+ "learning_rate": 5e-06,
465
+ "loss": 0.6281,
466
+ "step": 630
467
+ },
468
+ {
469
+ "epoch": 2.2504393673110723,
470
+ "grad_norm": 0.6819271490957453,
471
+ "learning_rate": 5e-06,
472
+ "loss": 0.6246,
473
+ "step": 640
474
+ },
475
+ {
476
+ "epoch": 2.2855887521968365,
477
+ "grad_norm": 0.678545369804577,
478
+ "learning_rate": 5e-06,
479
+ "loss": 0.6277,
480
+ "step": 650
481
+ },
482
+ {
483
+ "epoch": 2.3207381370826012,
484
+ "grad_norm": 0.6597702524075268,
485
+ "learning_rate": 5e-06,
486
+ "loss": 0.631,
487
+ "step": 660
488
+ },
489
+ {
490
+ "epoch": 2.3558875219683655,
491
+ "grad_norm": 0.5352899370053985,
492
+ "learning_rate": 5e-06,
493
+ "loss": 0.627,
494
+ "step": 670
495
+ },
496
+ {
497
+ "epoch": 2.39103690685413,
498
+ "grad_norm": 0.575976735916134,
499
+ "learning_rate": 5e-06,
500
+ "loss": 0.6252,
501
+ "step": 680
502
+ },
503
+ {
504
+ "epoch": 2.4261862917398944,
505
+ "grad_norm": 0.6538224434833726,
506
+ "learning_rate": 5e-06,
507
+ "loss": 0.631,
508
+ "step": 690
509
+ },
510
+ {
511
+ "epoch": 2.461335676625659,
512
+ "grad_norm": 0.8281376962806699,
513
+ "learning_rate": 5e-06,
514
+ "loss": 0.6238,
515
+ "step": 700
516
+ },
517
+ {
518
+ "epoch": 2.4964850615114234,
519
+ "grad_norm": 0.5971561231648772,
520
+ "learning_rate": 5e-06,
521
+ "loss": 0.6244,
522
+ "step": 710
523
+ },
524
+ {
525
+ "epoch": 2.531634446397188,
526
+ "grad_norm": 0.5668390272889466,
527
+ "learning_rate": 5e-06,
528
+ "loss": 0.6254,
529
+ "step": 720
530
+ },
531
+ {
532
+ "epoch": 2.5667838312829527,
533
+ "grad_norm": 0.7378544776528181,
534
+ "learning_rate": 5e-06,
535
+ "loss": 0.6248,
536
+ "step": 730
537
+ },
538
+ {
539
+ "epoch": 2.601933216168717,
540
+ "grad_norm": 0.6067368368819991,
541
+ "learning_rate": 5e-06,
542
+ "loss": 0.6256,
543
+ "step": 740
544
+ },
545
+ {
546
+ "epoch": 2.6370826010544817,
547
+ "grad_norm": 0.6816545127839443,
548
+ "learning_rate": 5e-06,
549
+ "loss": 0.6286,
550
+ "step": 750
551
+ },
552
+ {
553
+ "epoch": 2.672231985940246,
554
+ "grad_norm": 0.787032141068753,
555
+ "learning_rate": 5e-06,
556
+ "loss": 0.628,
557
+ "step": 760
558
+ },
559
+ {
560
+ "epoch": 2.7073813708260106,
561
+ "grad_norm": 0.6393338928189319,
562
+ "learning_rate": 5e-06,
563
+ "loss": 0.6267,
564
+ "step": 770
565
+ },
566
+ {
567
+ "epoch": 2.7425307557117753,
568
+ "grad_norm": 0.5562264277034894,
569
+ "learning_rate": 5e-06,
570
+ "loss": 0.6261,
571
+ "step": 780
572
+ },
573
+ {
574
+ "epoch": 2.7776801405975395,
575
+ "grad_norm": 0.5896436524802737,
576
+ "learning_rate": 5e-06,
577
+ "loss": 0.6256,
578
+ "step": 790
579
+ },
580
+ {
581
+ "epoch": 2.812829525483304,
582
+ "grad_norm": 0.5828475505687344,
583
+ "learning_rate": 5e-06,
584
+ "loss": 0.6247,
585
+ "step": 800
586
+ },
587
+ {
588
+ "epoch": 2.8479789103690685,
589
+ "grad_norm": 0.634394806473084,
590
+ "learning_rate": 5e-06,
591
+ "loss": 0.6269,
592
+ "step": 810
593
+ },
594
+ {
595
+ "epoch": 2.883128295254833,
596
+ "grad_norm": 0.6117384621451529,
597
+ "learning_rate": 5e-06,
598
+ "loss": 0.6279,
599
+ "step": 820
600
+ },
601
+ {
602
+ "epoch": 2.9182776801405974,
603
+ "grad_norm": 0.5540272640106404,
604
+ "learning_rate": 5e-06,
605
+ "loss": 0.6212,
606
+ "step": 830
607
+ },
608
+ {
609
+ "epoch": 2.953427065026362,
610
+ "grad_norm": 0.5600169828318418,
611
+ "learning_rate": 5e-06,
612
+ "loss": 0.6282,
613
+ "step": 840
614
+ },
615
+ {
616
+ "epoch": 2.9885764499121263,
617
+ "grad_norm": 0.7592332443324643,
618
+ "learning_rate": 5e-06,
619
+ "loss": 0.6277,
620
+ "step": 850
621
+ },
622
+ {
623
+ "epoch": 2.9956063268892796,
624
+ "eval_loss": 0.7103263735771179,
625
+ "eval_runtime": 302.7761,
626
+ "eval_samples_per_second": 25.322,
627
+ "eval_steps_per_second": 0.396,
628
+ "step": 852
629
+ },
630
+ {
631
+ "epoch": 2.9956063268892796,
632
+ "step": 852,
633
+ "total_flos": 1426922353459200.0,
634
+ "train_loss": 0.6913109551852857,
635
+ "train_runtime": 50624.2334,
636
+ "train_samples_per_second": 8.632,
637
+ "train_steps_per_second": 0.017
638
+ }
639
+ ],
640
+ "logging_steps": 10,
641
+ "max_steps": 852,
642
+ "num_input_tokens_seen": 0,
643
+ "num_train_epochs": 3,
644
+ "save_steps": 500,
645
+ "stateful_callbacks": {
646
+ "TrainerControl": {
647
+ "args": {
648
+ "should_epoch_stop": false,
649
+ "should_evaluate": false,
650
+ "should_log": false,
651
+ "should_save": true,
652
+ "should_training_stop": true
653
+ },
654
+ "attributes": {}
655
+ }
656
+ },
657
+ "total_flos": 1426922353459200.0,
658
+ "train_batch_size": 8,
659
+ "trial_name": null,
660
+ "trial_params": null
661
+ }
training_eval_loss.png ADDED
training_loss.png ADDED