sedrickkeh commited on
Commit
be30218
·
verified ·
1 Parent(s): 6e7c715

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 +693 -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: 3k_globalbatchsize96_lr4e5_epochs3
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
15
 
16
  # 3k_globalbatchsize96_lr4e5_epochs3
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: 3k_globalbatchsize96_lr4e5_epochs3
 
16
 
17
  # 3k_globalbatchsize96_lr4e5_epochs3
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/openthoughts_3000 dataset.
20
 
21
  ## Model description
22
 
all_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.976,
3
+ "total_flos": 1.818538711009198e+17,
4
+ "train_loss": 0.5000655266546434,
5
+ "train_runtime": 9080.5183,
6
+ "train_samples_per_second": 0.99,
7
+ "train_steps_per_second": 0.01
8
+ }
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.976,
3
+ "total_flos": 1.818538711009198e+17,
4
+ "train_loss": 0.5000655266546434,
5
+ "train_runtime": 9080.5183,
6
+ "train_samples_per_second": 0.99,
7
+ "train_steps_per_second": 0.01
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,693 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 2.976,
5
+ "eval_steps": 500,
6
+ "global_step": 93,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.032,
13
+ "grad_norm": 6.08677457937853,
14
+ "learning_rate": 4.000000000000001e-06,
15
+ "loss": 0.8709,
16
+ "step": 1
17
+ },
18
+ {
19
+ "epoch": 0.064,
20
+ "grad_norm": 5.862503603501722,
21
+ "learning_rate": 8.000000000000001e-06,
22
+ "loss": 0.8625,
23
+ "step": 2
24
+ },
25
+ {
26
+ "epoch": 0.096,
27
+ "grad_norm": 4.5723036455815205,
28
+ "learning_rate": 1.2e-05,
29
+ "loss": 0.839,
30
+ "step": 3
31
+ },
32
+ {
33
+ "epoch": 0.128,
34
+ "grad_norm": 2.0695238404961547,
35
+ "learning_rate": 1.6000000000000003e-05,
36
+ "loss": 0.7376,
37
+ "step": 4
38
+ },
39
+ {
40
+ "epoch": 0.16,
41
+ "grad_norm": 5.589012282844458,
42
+ "learning_rate": 2e-05,
43
+ "loss": 0.825,
44
+ "step": 5
45
+ },
46
+ {
47
+ "epoch": 0.192,
48
+ "grad_norm": 8.490554966738163,
49
+ "learning_rate": 2.4e-05,
50
+ "loss": 0.8478,
51
+ "step": 6
52
+ },
53
+ {
54
+ "epoch": 0.224,
55
+ "grad_norm": 6.29894861103046,
56
+ "learning_rate": 2.8e-05,
57
+ "loss": 0.7658,
58
+ "step": 7
59
+ },
60
+ {
61
+ "epoch": 0.256,
62
+ "grad_norm": 3.558435531359445,
63
+ "learning_rate": 3.2000000000000005e-05,
64
+ "loss": 0.7555,
65
+ "step": 8
66
+ },
67
+ {
68
+ "epoch": 0.288,
69
+ "grad_norm": 2.722863714076688,
70
+ "learning_rate": 3.6e-05,
71
+ "loss": 0.7274,
72
+ "step": 9
73
+ },
74
+ {
75
+ "epoch": 0.32,
76
+ "grad_norm": 2.088640250697761,
77
+ "learning_rate": 4e-05,
78
+ "loss": 0.6627,
79
+ "step": 10
80
+ },
81
+ {
82
+ "epoch": 0.352,
83
+ "grad_norm": 1.5096876602344986,
84
+ "learning_rate": 3.998567509632663e-05,
85
+ "loss": 0.6817,
86
+ "step": 11
87
+ },
88
+ {
89
+ "epoch": 0.384,
90
+ "grad_norm": 1.6896637753034105,
91
+ "learning_rate": 3.9942720905593045e-05,
92
+ "loss": 0.6761,
93
+ "step": 12
94
+ },
95
+ {
96
+ "epoch": 0.416,
97
+ "grad_norm": 1.5060644054718806,
98
+ "learning_rate": 3.98711989592637e-05,
99
+ "loss": 0.6519,
100
+ "step": 13
101
+ },
102
+ {
103
+ "epoch": 0.448,
104
+ "grad_norm": 1.4049366508424377,
105
+ "learning_rate": 3.9771211711837774e-05,
106
+ "loss": 0.6333,
107
+ "step": 14
108
+ },
109
+ {
110
+ "epoch": 0.48,
111
+ "grad_norm": 1.411091031460123,
112
+ "learning_rate": 3.9642902394084056e-05,
113
+ "loss": 0.5874,
114
+ "step": 15
115
+ },
116
+ {
117
+ "epoch": 0.512,
118
+ "grad_norm": 0.8883192699998052,
119
+ "learning_rate": 3.948645480786427e-05,
120
+ "loss": 0.6116,
121
+ "step": 16
122
+ },
123
+ {
124
+ "epoch": 0.544,
125
+ "grad_norm": 1.129257283929009,
126
+ "learning_rate": 3.930209306283867e-05,
127
+ "loss": 0.5852,
128
+ "step": 17
129
+ },
130
+ {
131
+ "epoch": 0.576,
132
+ "grad_norm": 0.901306643045694,
133
+ "learning_rate": 3.909008125543111e-05,
134
+ "loss": 0.5821,
135
+ "step": 18
136
+ },
137
+ {
138
+ "epoch": 0.608,
139
+ "grad_norm": 0.9465933700636773,
140
+ "learning_rate": 3.885072309051346e-05,
141
+ "loss": 0.5833,
142
+ "step": 19
143
+ },
144
+ {
145
+ "epoch": 0.64,
146
+ "grad_norm": 0.8046120689103757,
147
+ "learning_rate": 3.858436144635131e-05,
148
+ "loss": 0.5597,
149
+ "step": 20
150
+ },
151
+ {
152
+ "epoch": 0.672,
153
+ "grad_norm": 0.9228859026973337,
154
+ "learning_rate": 3.829137788343415e-05,
155
+ "loss": 0.5571,
156
+ "step": 21
157
+ },
158
+ {
159
+ "epoch": 0.704,
160
+ "grad_norm": 0.9202261747479373,
161
+ "learning_rate": 3.797219209789365e-05,
162
+ "loss": 0.6007,
163
+ "step": 22
164
+ },
165
+ {
166
+ "epoch": 0.736,
167
+ "grad_norm": 0.9336252436543613,
168
+ "learning_rate": 3.762726132029298e-05,
169
+ "loss": 0.5618,
170
+ "step": 23
171
+ },
172
+ {
173
+ "epoch": 0.768,
174
+ "grad_norm": 1.2080051557421407,
175
+ "learning_rate": 3.725707966064846e-05,
176
+ "loss": 0.6057,
177
+ "step": 24
178
+ },
179
+ {
180
+ "epoch": 0.8,
181
+ "grad_norm": 1.1231601773572062,
182
+ "learning_rate": 3.686217740062169e-05,
183
+ "loss": 0.5603,
184
+ "step": 25
185
+ },
186
+ {
187
+ "epoch": 0.832,
188
+ "grad_norm": 0.8382981788712457,
189
+ "learning_rate": 3.644312023389621e-05,
190
+ "loss": 0.5491,
191
+ "step": 26
192
+ },
193
+ {
194
+ "epoch": 0.864,
195
+ "grad_norm": 1.1862939971134692,
196
+ "learning_rate": 3.600050845582669e-05,
197
+ "loss": 0.5887,
198
+ "step": 27
199
+ },
200
+ {
201
+ "epoch": 0.896,
202
+ "grad_norm": 1.0957634618366314,
203
+ "learning_rate": 3.5534976103521716e-05,
204
+ "loss": 0.5958,
205
+ "step": 28
206
+ },
207
+ {
208
+ "epoch": 0.928,
209
+ "grad_norm": 0.6590817382859444,
210
+ "learning_rate": 3.504719004759163e-05,
211
+ "loss": 0.5528,
212
+ "step": 29
213
+ },
214
+ {
215
+ "epoch": 0.96,
216
+ "grad_norm": 0.875749033008583,
217
+ "learning_rate": 3.4537849036862874e-05,
218
+ "loss": 0.565,
219
+ "step": 30
220
+ },
221
+ {
222
+ "epoch": 0.992,
223
+ "grad_norm": 0.783563683071678,
224
+ "learning_rate": 3.400768269742702e-05,
225
+ "loss": 0.5595,
226
+ "step": 31
227
+ },
228
+ {
229
+ "epoch": 1.024,
230
+ "grad_norm": 1.4075572002103194,
231
+ "learning_rate": 3.345745048745838e-05,
232
+ "loss": 0.919,
233
+ "step": 32
234
+ },
235
+ {
236
+ "epoch": 1.056,
237
+ "grad_norm": 0.7062499705952857,
238
+ "learning_rate": 3.288794060929754e-05,
239
+ "loss": 0.4404,
240
+ "step": 33
241
+ },
242
+ {
243
+ "epoch": 1.088,
244
+ "grad_norm": 0.9967541412280615,
245
+ "learning_rate": 3.229996888035908e-05,
246
+ "loss": 0.4984,
247
+ "step": 34
248
+ },
249
+ {
250
+ "epoch": 1.12,
251
+ "grad_norm": 0.8310704240859569,
252
+ "learning_rate": 3.169437756448095e-05,
253
+ "loss": 0.4807,
254
+ "step": 35
255
+ },
256
+ {
257
+ "epoch": 1.152,
258
+ "grad_norm": 0.7016314595292313,
259
+ "learning_rate": 3.107203416538969e-05,
260
+ "loss": 0.4703,
261
+ "step": 36
262
+ },
263
+ {
264
+ "epoch": 1.184,
265
+ "grad_norm": 1.4372935885534768,
266
+ "learning_rate": 3.0433830184009694e-05,
267
+ "loss": 0.4739,
268
+ "step": 37
269
+ },
270
+ {
271
+ "epoch": 1.216,
272
+ "grad_norm": 0.8650508524606009,
273
+ "learning_rate": 2.9780679841396668e-05,
274
+ "loss": 0.4525,
275
+ "step": 38
276
+ },
277
+ {
278
+ "epoch": 1.248,
279
+ "grad_norm": 1.0766618304625992,
280
+ "learning_rate": 2.9113518769124836e-05,
281
+ "loss": 0.4987,
282
+ "step": 39
283
+ },
284
+ {
285
+ "epoch": 1.28,
286
+ "grad_norm": 0.6767070574855524,
287
+ "learning_rate": 2.843330266900368e-05,
288
+ "loss": 0.4475,
289
+ "step": 40
290
+ },
291
+ {
292
+ "epoch": 1.312,
293
+ "grad_norm": 1.026763400096595,
294
+ "learning_rate": 2.774100594404435e-05,
295
+ "loss": 0.4667,
296
+ "step": 41
297
+ },
298
+ {
299
+ "epoch": 1.3439999999999999,
300
+ "grad_norm": 0.9002290625499892,
301
+ "learning_rate": 2.703762030263666e-05,
302
+ "loss": 0.4916,
303
+ "step": 42
304
+ },
305
+ {
306
+ "epoch": 1.376,
307
+ "grad_norm": 0.9903797867735974,
308
+ "learning_rate": 2.632415333793648e-05,
309
+ "loss": 0.4771,
310
+ "step": 43
311
+ },
312
+ {
313
+ "epoch": 1.408,
314
+ "grad_norm": 0.7218070561744779,
315
+ "learning_rate": 2.5601627084498146e-05,
316
+ "loss": 0.407,
317
+ "step": 44
318
+ },
319
+ {
320
+ "epoch": 1.44,
321
+ "grad_norm": 0.8313194407823631,
322
+ "learning_rate": 2.4871076554219838e-05,
323
+ "loss": 0.4442,
324
+ "step": 45
325
+ },
326
+ {
327
+ "epoch": 1.472,
328
+ "grad_norm": 0.8236730874850681,
329
+ "learning_rate": 2.413354825369906e-05,
330
+ "loss": 0.5223,
331
+ "step": 46
332
+ },
333
+ {
334
+ "epoch": 1.504,
335
+ "grad_norm": 0.6125019115754542,
336
+ "learning_rate": 2.3390098685121938e-05,
337
+ "loss": 0.42,
338
+ "step": 47
339
+ },
340
+ {
341
+ "epoch": 1.536,
342
+ "grad_norm": 0.6737099841054438,
343
+ "learning_rate": 2.264179283283405e-05,
344
+ "loss": 0.4665,
345
+ "step": 48
346
+ },
347
+ {
348
+ "epoch": 1.568,
349
+ "grad_norm": 0.5907056602966384,
350
+ "learning_rate": 2.1889702637760627e-05,
351
+ "loss": 0.4445,
352
+ "step": 49
353
+ },
354
+ {
355
+ "epoch": 1.6,
356
+ "grad_norm": 0.6965345367451425,
357
+ "learning_rate": 2.1134905461861486e-05,
358
+ "loss": 0.5221,
359
+ "step": 50
360
+ },
361
+ {
362
+ "epoch": 1.6320000000000001,
363
+ "grad_norm": 0.5006105897500711,
364
+ "learning_rate": 2.0378482544820383e-05,
365
+ "loss": 0.4218,
366
+ "step": 51
367
+ },
368
+ {
369
+ "epoch": 1.6640000000000001,
370
+ "grad_norm": 0.6174830888739168,
371
+ "learning_rate": 1.9621517455179627e-05,
372
+ "loss": 0.476,
373
+ "step": 52
374
+ },
375
+ {
376
+ "epoch": 1.696,
377
+ "grad_norm": 0.503990909520708,
378
+ "learning_rate": 1.886509453813852e-05,
379
+ "loss": 0.4275,
380
+ "step": 53
381
+ },
382
+ {
383
+ "epoch": 1.728,
384
+ "grad_norm": 0.5974502021657286,
385
+ "learning_rate": 1.8110297362239376e-05,
386
+ "loss": 0.4757,
387
+ "step": 54
388
+ },
389
+ {
390
+ "epoch": 1.76,
391
+ "grad_norm": 0.551329913445271,
392
+ "learning_rate": 1.735820716716596e-05,
393
+ "loss": 0.4757,
394
+ "step": 55
395
+ },
396
+ {
397
+ "epoch": 1.792,
398
+ "grad_norm": 0.579064058765967,
399
+ "learning_rate": 1.660990131487807e-05,
400
+ "loss": 0.4182,
401
+ "step": 56
402
+ },
403
+ {
404
+ "epoch": 1.8239999999999998,
405
+ "grad_norm": 0.548961385728647,
406
+ "learning_rate": 1.586645174630094e-05,
407
+ "loss": 0.4731,
408
+ "step": 57
409
+ },
410
+ {
411
+ "epoch": 1.8559999999999999,
412
+ "grad_norm": 0.5764590922933827,
413
+ "learning_rate": 1.5128923445780163e-05,
414
+ "loss": 0.4271,
415
+ "step": 58
416
+ },
417
+ {
418
+ "epoch": 1.888,
419
+ "grad_norm": 0.5781209646736115,
420
+ "learning_rate": 1.4398372915501862e-05,
421
+ "loss": 0.4644,
422
+ "step": 59
423
+ },
424
+ {
425
+ "epoch": 1.92,
426
+ "grad_norm": 0.42862823912049036,
427
+ "learning_rate": 1.3675846662063521e-05,
428
+ "loss": 0.4071,
429
+ "step": 60
430
+ },
431
+ {
432
+ "epoch": 1.952,
433
+ "grad_norm": 0.569159061133135,
434
+ "learning_rate": 1.296237969736334e-05,
435
+ "loss": 0.4561,
436
+ "step": 61
437
+ },
438
+ {
439
+ "epoch": 1.984,
440
+ "grad_norm": 0.47530501617214926,
441
+ "learning_rate": 1.2258994055955658e-05,
442
+ "loss": 0.3817,
443
+ "step": 62
444
+ },
445
+ {
446
+ "epoch": 2.016,
447
+ "grad_norm": 0.9251054706846084,
448
+ "learning_rate": 1.156669733099632e-05,
449
+ "loss": 0.7898,
450
+ "step": 63
451
+ },
452
+ {
453
+ "epoch": 2.048,
454
+ "grad_norm": 0.5716376901076641,
455
+ "learning_rate": 1.0886481230875172e-05,
456
+ "loss": 0.3525,
457
+ "step": 64
458
+ },
459
+ {
460
+ "epoch": 2.08,
461
+ "grad_norm": 0.5177936896448316,
462
+ "learning_rate": 1.0219320158603337e-05,
463
+ "loss": 0.3394,
464
+ "step": 65
465
+ },
466
+ {
467
+ "epoch": 2.112,
468
+ "grad_norm": 0.5247806458447061,
469
+ "learning_rate": 9.566169815990311e-06,
470
+ "loss": 0.3834,
471
+ "step": 66
472
+ },
473
+ {
474
+ "epoch": 2.144,
475
+ "grad_norm": 0.5455234154497576,
476
+ "learning_rate": 8.92796583461031e-06,
477
+ "loss": 0.3577,
478
+ "step": 67
479
+ },
480
+ {
481
+ "epoch": 2.176,
482
+ "grad_norm": 0.5925096096878631,
483
+ "learning_rate": 8.305622435519058e-06,
484
+ "loss": 0.3831,
485
+ "step": 68
486
+ },
487
+ {
488
+ "epoch": 2.208,
489
+ "grad_norm": 0.6820182428585542,
490
+ "learning_rate": 7.70003111964093e-06,
491
+ "loss": 0.376,
492
+ "step": 69
493
+ },
494
+ {
495
+ "epoch": 2.24,
496
+ "grad_norm": 0.6320620904769954,
497
+ "learning_rate": 7.112059390702459e-06,
498
+ "loss": 0.3715,
499
+ "step": 70
500
+ },
501
+ {
502
+ "epoch": 2.2720000000000002,
503
+ "grad_norm": 0.4928197929862798,
504
+ "learning_rate": 6.542549512541623e-06,
505
+ "loss": 0.3713,
506
+ "step": 71
507
+ },
508
+ {
509
+ "epoch": 2.304,
510
+ "grad_norm": 0.4767133735569691,
511
+ "learning_rate": 5.9923173025729895e-06,
512
+ "loss": 0.3303,
513
+ "step": 72
514
+ },
515
+ {
516
+ "epoch": 2.336,
517
+ "grad_norm": 0.7059312169326228,
518
+ "learning_rate": 5.462150963137125e-06,
519
+ "loss": 0.4568,
520
+ "step": 73
521
+ },
522
+ {
523
+ "epoch": 2.368,
524
+ "grad_norm": 0.5003954153261982,
525
+ "learning_rate": 4.952809952408375e-06,
526
+ "loss": 0.3514,
527
+ "step": 74
528
+ },
529
+ {
530
+ "epoch": 2.4,
531
+ "grad_norm": 0.5944579046423205,
532
+ "learning_rate": 4.465023896478293e-06,
533
+ "loss": 0.3627,
534
+ "step": 75
535
+ },
536
+ {
537
+ "epoch": 2.432,
538
+ "grad_norm": 0.4180864471254852,
539
+ "learning_rate": 3.999491544173311e-06,
540
+ "loss": 0.3054,
541
+ "step": 76
542
+ },
543
+ {
544
+ "epoch": 2.464,
545
+ "grad_norm": 0.44019542396333683,
546
+ "learning_rate": 3.5568797661038004e-06,
547
+ "loss": 0.375,
548
+ "step": 77
549
+ },
550
+ {
551
+ "epoch": 2.496,
552
+ "grad_norm": 0.4294310051147678,
553
+ "learning_rate": 3.137822599378315e-06,
554
+ "loss": 0.3537,
555
+ "step": 78
556
+ },
557
+ {
558
+ "epoch": 2.528,
559
+ "grad_norm": 0.3888176996168452,
560
+ "learning_rate": 2.7429203393515426e-06,
561
+ "loss": 0.378,
562
+ "step": 79
563
+ },
564
+ {
565
+ "epoch": 2.56,
566
+ "grad_norm": 0.3576643131385393,
567
+ "learning_rate": 2.372738679707023e-06,
568
+ "loss": 0.3232,
569
+ "step": 80
570
+ },
571
+ {
572
+ "epoch": 2.592,
573
+ "grad_norm": 0.3696160994931973,
574
+ "learning_rate": 2.02780790210636e-06,
575
+ "loss": 0.3542,
576
+ "step": 81
577
+ },
578
+ {
579
+ "epoch": 2.624,
580
+ "grad_norm": 0.36882809964571234,
581
+ "learning_rate": 1.7086221165658544e-06,
582
+ "loss": 0.351,
583
+ "step": 82
584
+ },
585
+ {
586
+ "epoch": 2.656,
587
+ "grad_norm": 0.34603236345776744,
588
+ "learning_rate": 1.4156385536486973e-06,
589
+ "loss": 0.3212,
590
+ "step": 83
591
+ },
592
+ {
593
+ "epoch": 2.6879999999999997,
594
+ "grad_norm": 0.3805023899734686,
595
+ "learning_rate": 1.1492769094865475e-06,
596
+ "loss": 0.3744,
597
+ "step": 84
598
+ },
599
+ {
600
+ "epoch": 2.7199999999999998,
601
+ "grad_norm": 0.3005431184449355,
602
+ "learning_rate": 9.099187445688984e-07,
603
+ "loss": 0.3071,
604
+ "step": 85
605
+ },
606
+ {
607
+ "epoch": 2.752,
608
+ "grad_norm": 0.3625697661026582,
609
+ "learning_rate": 6.979069371613345e-07,
610
+ "loss": 0.3755,
611
+ "step": 86
612
+ },
613
+ {
614
+ "epoch": 2.784,
615
+ "grad_norm": 0.30922554074419895,
616
+ "learning_rate": 5.135451921357337e-07,
617
+ "loss": 0.2993,
618
+ "step": 87
619
+ },
620
+ {
621
+ "epoch": 2.816,
622
+ "grad_norm": 0.3209396414531254,
623
+ "learning_rate": 3.570976059159481e-07,
624
+ "loss": 0.3725,
625
+ "step": 88
626
+ },
627
+ {
628
+ "epoch": 2.848,
629
+ "grad_norm": 0.3188890164441534,
630
+ "learning_rate": 2.2878828816222942e-07,
631
+ "loss": 0.3691,
632
+ "step": 89
633
+ },
634
+ {
635
+ "epoch": 2.88,
636
+ "grad_norm": 0.30765746035254077,
637
+ "learning_rate": 1.2880104073630163e-07,
638
+ "loss": 0.3218,
639
+ "step": 90
640
+ },
641
+ {
642
+ "epoch": 2.912,
643
+ "grad_norm": 0.3085708159717203,
644
+ "learning_rate": 5.7279094406959e-08,
645
+ "loss": 0.3625,
646
+ "step": 91
647
+ },
648
+ {
649
+ "epoch": 2.944,
650
+ "grad_norm": 0.3092205424287526,
651
+ "learning_rate": 1.4324903673370583e-08,
652
+ "loss": 0.3543,
653
+ "step": 92
654
+ },
655
+ {
656
+ "epoch": 2.976,
657
+ "grad_norm": 0.3121521435612877,
658
+ "learning_rate": 0.0,
659
+ "loss": 0.3917,
660
+ "step": 93
661
+ },
662
+ {
663
+ "epoch": 2.976,
664
+ "step": 93,
665
+ "total_flos": 1.818538711009198e+17,
666
+ "train_loss": 0.5000655266546434,
667
+ "train_runtime": 9080.5183,
668
+ "train_samples_per_second": 0.99,
669
+ "train_steps_per_second": 0.01
670
+ }
671
+ ],
672
+ "logging_steps": 1.0,
673
+ "max_steps": 93,
674
+ "num_input_tokens_seen": 0,
675
+ "num_train_epochs": 3,
676
+ "save_steps": 500,
677
+ "stateful_callbacks": {
678
+ "TrainerControl": {
679
+ "args": {
680
+ "should_epoch_stop": false,
681
+ "should_evaluate": false,
682
+ "should_log": false,
683
+ "should_save": true,
684
+ "should_training_stop": true
685
+ },
686
+ "attributes": {}
687
+ }
688
+ },
689
+ "total_flos": 1.818538711009198e+17,
690
+ "train_batch_size": 1,
691
+ "trial_name": null,
692
+ "trial_params": null
693
+ }
training_loss.png ADDED