sedrickkeh commited on
Commit
b3c2038
·
verified ·
1 Parent(s): acc1a7c

End of training

Browse files
README.md CHANGED
@@ -4,6 +4,7 @@ license: llama3.1
4
  base_model: meta-llama/Llama-3.1-8B
5
  tags:
6
  - llama-factory
 
7
  - generated_from_trainer
8
  model-index:
9
  - name: oh_v1_w_v3_opengpt
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
15
 
16
  # oh_v1_w_v3_opengpt
17
 
18
- This model is a fine-tuned version of [meta-llama/Llama-3.1-8B](https://huggingface.co/meta-llama/Llama-3.1-8B) on an unknown dataset.
19
  It achieves the following results on the evaluation set:
20
  - Loss: 0.5331
21
 
 
4
  base_model: meta-llama/Llama-3.1-8B
5
  tags:
6
  - llama-factory
7
+ - full
8
  - generated_from_trainer
9
  model-index:
10
  - name: oh_v1_w_v3_opengpt
 
16
 
17
  # oh_v1_w_v3_opengpt
18
 
19
+ This model is a fine-tuned version of [meta-llama/Llama-3.1-8B](https://huggingface.co/meta-llama/Llama-3.1-8B) on the mlfoundations-dev/oh_v1_w_v3_opengpt dataset.
20
  It achieves the following results on the evaluation set:
21
  - Loss: 0.5331
22
 
all_results.json ADDED
@@ -0,0 +1,12 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.99626400996264,
3
+ "eval_loss": 0.5331025123596191,
4
+ "eval_runtime": 215.0816,
5
+ "eval_samples_per_second": 50.302,
6
+ "eval_steps_per_second": 0.395,
7
+ "total_flos": 2014651046952960.0,
8
+ "train_loss": 0.5350556444943397,
9
+ "train_runtime": 35909.5127,
10
+ "train_samples_per_second": 17.172,
11
+ "train_steps_per_second": 0.034
12
+ }
eval_results.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.99626400996264,
3
+ "eval_loss": 0.5331025123596191,
4
+ "eval_runtime": 215.0816,
5
+ "eval_samples_per_second": 50.302,
6
+ "eval_steps_per_second": 0.395
7
+ }
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.99626400996264,
3
+ "total_flos": 2014651046952960.0,
4
+ "train_loss": 0.5350556444943397,
5
+ "train_runtime": 35909.5127,
6
+ "train_samples_per_second": 17.172,
7
+ "train_steps_per_second": 0.034
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,906 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 2.99626400996264,
5
+ "eval_steps": 500,
6
+ "global_step": 1203,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.024906600249066,
13
+ "grad_norm": 10.55736720812015,
14
+ "learning_rate": 5e-06,
15
+ "loss": 0.899,
16
+ "step": 10
17
+ },
18
+ {
19
+ "epoch": 0.049813200498132,
20
+ "grad_norm": 2.4056192000192786,
21
+ "learning_rate": 5e-06,
22
+ "loss": 0.7632,
23
+ "step": 20
24
+ },
25
+ {
26
+ "epoch": 0.074719800747198,
27
+ "grad_norm": 1.0776649113170846,
28
+ "learning_rate": 5e-06,
29
+ "loss": 0.7213,
30
+ "step": 30
31
+ },
32
+ {
33
+ "epoch": 0.099626400996264,
34
+ "grad_norm": 1.2974590316911792,
35
+ "learning_rate": 5e-06,
36
+ "loss": 0.6892,
37
+ "step": 40
38
+ },
39
+ {
40
+ "epoch": 0.12453300124533001,
41
+ "grad_norm": 0.8664841939523225,
42
+ "learning_rate": 5e-06,
43
+ "loss": 0.6708,
44
+ "step": 50
45
+ },
46
+ {
47
+ "epoch": 0.149439601494396,
48
+ "grad_norm": 0.7916914326975599,
49
+ "learning_rate": 5e-06,
50
+ "loss": 0.6504,
51
+ "step": 60
52
+ },
53
+ {
54
+ "epoch": 0.17434620174346202,
55
+ "grad_norm": 0.8355113234133088,
56
+ "learning_rate": 5e-06,
57
+ "loss": 0.6407,
58
+ "step": 70
59
+ },
60
+ {
61
+ "epoch": 0.199252801992528,
62
+ "grad_norm": 0.6185509662301869,
63
+ "learning_rate": 5e-06,
64
+ "loss": 0.6377,
65
+ "step": 80
66
+ },
67
+ {
68
+ "epoch": 0.22415940224159403,
69
+ "grad_norm": 0.5793150300164642,
70
+ "learning_rate": 5e-06,
71
+ "loss": 0.6276,
72
+ "step": 90
73
+ },
74
+ {
75
+ "epoch": 0.24906600249066002,
76
+ "grad_norm": 0.5522899454819212,
77
+ "learning_rate": 5e-06,
78
+ "loss": 0.6229,
79
+ "step": 100
80
+ },
81
+ {
82
+ "epoch": 0.273972602739726,
83
+ "grad_norm": 0.6471095275689863,
84
+ "learning_rate": 5e-06,
85
+ "loss": 0.6154,
86
+ "step": 110
87
+ },
88
+ {
89
+ "epoch": 0.298879202988792,
90
+ "grad_norm": 0.6264851589702339,
91
+ "learning_rate": 5e-06,
92
+ "loss": 0.6104,
93
+ "step": 120
94
+ },
95
+ {
96
+ "epoch": 0.32378580323785805,
97
+ "grad_norm": 0.705297908293922,
98
+ "learning_rate": 5e-06,
99
+ "loss": 0.6094,
100
+ "step": 130
101
+ },
102
+ {
103
+ "epoch": 0.34869240348692404,
104
+ "grad_norm": 0.620653193497072,
105
+ "learning_rate": 5e-06,
106
+ "loss": 0.6079,
107
+ "step": 140
108
+ },
109
+ {
110
+ "epoch": 0.37359900373599003,
111
+ "grad_norm": 0.9444217688486999,
112
+ "learning_rate": 5e-06,
113
+ "loss": 0.5983,
114
+ "step": 150
115
+ },
116
+ {
117
+ "epoch": 0.398505603985056,
118
+ "grad_norm": 0.6034795779576418,
119
+ "learning_rate": 5e-06,
120
+ "loss": 0.6044,
121
+ "step": 160
122
+ },
123
+ {
124
+ "epoch": 0.42341220423412207,
125
+ "grad_norm": 0.4974203412589217,
126
+ "learning_rate": 5e-06,
127
+ "loss": 0.5962,
128
+ "step": 170
129
+ },
130
+ {
131
+ "epoch": 0.44831880448318806,
132
+ "grad_norm": 0.728230577535382,
133
+ "learning_rate": 5e-06,
134
+ "loss": 0.5975,
135
+ "step": 180
136
+ },
137
+ {
138
+ "epoch": 0.47322540473225405,
139
+ "grad_norm": 0.49214457212025137,
140
+ "learning_rate": 5e-06,
141
+ "loss": 0.5986,
142
+ "step": 190
143
+ },
144
+ {
145
+ "epoch": 0.49813200498132004,
146
+ "grad_norm": 0.5432395369586009,
147
+ "learning_rate": 5e-06,
148
+ "loss": 0.5885,
149
+ "step": 200
150
+ },
151
+ {
152
+ "epoch": 0.523038605230386,
153
+ "grad_norm": 0.5473973456491285,
154
+ "learning_rate": 5e-06,
155
+ "loss": 0.5901,
156
+ "step": 210
157
+ },
158
+ {
159
+ "epoch": 0.547945205479452,
160
+ "grad_norm": 0.5759522711461309,
161
+ "learning_rate": 5e-06,
162
+ "loss": 0.586,
163
+ "step": 220
164
+ },
165
+ {
166
+ "epoch": 0.572851805728518,
167
+ "grad_norm": 0.5583334053126371,
168
+ "learning_rate": 5e-06,
169
+ "loss": 0.5878,
170
+ "step": 230
171
+ },
172
+ {
173
+ "epoch": 0.597758405977584,
174
+ "grad_norm": 0.7058646192825392,
175
+ "learning_rate": 5e-06,
176
+ "loss": 0.5818,
177
+ "step": 240
178
+ },
179
+ {
180
+ "epoch": 0.6226650062266501,
181
+ "grad_norm": 0.5788660860127395,
182
+ "learning_rate": 5e-06,
183
+ "loss": 0.5853,
184
+ "step": 250
185
+ },
186
+ {
187
+ "epoch": 0.6475716064757161,
188
+ "grad_norm": 0.6402936090502463,
189
+ "learning_rate": 5e-06,
190
+ "loss": 0.5769,
191
+ "step": 260
192
+ },
193
+ {
194
+ "epoch": 0.6724782067247821,
195
+ "grad_norm": 0.5496743019012175,
196
+ "learning_rate": 5e-06,
197
+ "loss": 0.5762,
198
+ "step": 270
199
+ },
200
+ {
201
+ "epoch": 0.6973848069738481,
202
+ "grad_norm": 0.5455503536049673,
203
+ "learning_rate": 5e-06,
204
+ "loss": 0.5808,
205
+ "step": 280
206
+ },
207
+ {
208
+ "epoch": 0.7222914072229141,
209
+ "grad_norm": 0.5883422768851626,
210
+ "learning_rate": 5e-06,
211
+ "loss": 0.5747,
212
+ "step": 290
213
+ },
214
+ {
215
+ "epoch": 0.7471980074719801,
216
+ "grad_norm": 0.7956726883043881,
217
+ "learning_rate": 5e-06,
218
+ "loss": 0.5717,
219
+ "step": 300
220
+ },
221
+ {
222
+ "epoch": 0.772104607721046,
223
+ "grad_norm": 0.6813019491351078,
224
+ "learning_rate": 5e-06,
225
+ "loss": 0.5722,
226
+ "step": 310
227
+ },
228
+ {
229
+ "epoch": 0.797011207970112,
230
+ "grad_norm": 0.5185381604929972,
231
+ "learning_rate": 5e-06,
232
+ "loss": 0.5724,
233
+ "step": 320
234
+ },
235
+ {
236
+ "epoch": 0.821917808219178,
237
+ "grad_norm": 0.6909227667005327,
238
+ "learning_rate": 5e-06,
239
+ "loss": 0.5691,
240
+ "step": 330
241
+ },
242
+ {
243
+ "epoch": 0.8468244084682441,
244
+ "grad_norm": 0.4796775398588659,
245
+ "learning_rate": 5e-06,
246
+ "loss": 0.5652,
247
+ "step": 340
248
+ },
249
+ {
250
+ "epoch": 0.8717310087173101,
251
+ "grad_norm": 0.5758165163544798,
252
+ "learning_rate": 5e-06,
253
+ "loss": 0.5736,
254
+ "step": 350
255
+ },
256
+ {
257
+ "epoch": 0.8966376089663761,
258
+ "grad_norm": 0.5892321233613834,
259
+ "learning_rate": 5e-06,
260
+ "loss": 0.5663,
261
+ "step": 360
262
+ },
263
+ {
264
+ "epoch": 0.9215442092154421,
265
+ "grad_norm": 0.5088131774112012,
266
+ "learning_rate": 5e-06,
267
+ "loss": 0.5674,
268
+ "step": 370
269
+ },
270
+ {
271
+ "epoch": 0.9464508094645081,
272
+ "grad_norm": 0.5017508892526636,
273
+ "learning_rate": 5e-06,
274
+ "loss": 0.563,
275
+ "step": 380
276
+ },
277
+ {
278
+ "epoch": 0.9713574097135741,
279
+ "grad_norm": 0.6797455724948757,
280
+ "learning_rate": 5e-06,
281
+ "loss": 0.5674,
282
+ "step": 390
283
+ },
284
+ {
285
+ "epoch": 0.9962640099626401,
286
+ "grad_norm": 0.5319255477795009,
287
+ "learning_rate": 5e-06,
288
+ "loss": 0.562,
289
+ "step": 400
290
+ },
291
+ {
292
+ "epoch": 0.9987546699875467,
293
+ "eval_loss": 0.5648990869522095,
294
+ "eval_runtime": 215.2105,
295
+ "eval_samples_per_second": 50.272,
296
+ "eval_steps_per_second": 0.395,
297
+ "step": 401
298
+ },
299
+ {
300
+ "epoch": 1.0211706102117062,
301
+ "grad_norm": 0.6332423226665538,
302
+ "learning_rate": 5e-06,
303
+ "loss": 0.5373,
304
+ "step": 410
305
+ },
306
+ {
307
+ "epoch": 1.046077210460772,
308
+ "grad_norm": 0.5250206863412192,
309
+ "learning_rate": 5e-06,
310
+ "loss": 0.5262,
311
+ "step": 420
312
+ },
313
+ {
314
+ "epoch": 1.0709838107098382,
315
+ "grad_norm": 0.6270808658082102,
316
+ "learning_rate": 5e-06,
317
+ "loss": 0.5275,
318
+ "step": 430
319
+ },
320
+ {
321
+ "epoch": 1.095890410958904,
322
+ "grad_norm": 0.5516867253896128,
323
+ "learning_rate": 5e-06,
324
+ "loss": 0.5262,
325
+ "step": 440
326
+ },
327
+ {
328
+ "epoch": 1.1207970112079702,
329
+ "grad_norm": 0.5358792491430266,
330
+ "learning_rate": 5e-06,
331
+ "loss": 0.5319,
332
+ "step": 450
333
+ },
334
+ {
335
+ "epoch": 1.145703611457036,
336
+ "grad_norm": 0.7694988300392477,
337
+ "learning_rate": 5e-06,
338
+ "loss": 0.5251,
339
+ "step": 460
340
+ },
341
+ {
342
+ "epoch": 1.1706102117061021,
343
+ "grad_norm": 0.5425248792129497,
344
+ "learning_rate": 5e-06,
345
+ "loss": 0.5255,
346
+ "step": 470
347
+ },
348
+ {
349
+ "epoch": 1.195516811955168,
350
+ "grad_norm": 0.6608067800821592,
351
+ "learning_rate": 5e-06,
352
+ "loss": 0.5312,
353
+ "step": 480
354
+ },
355
+ {
356
+ "epoch": 1.2204234122042341,
357
+ "grad_norm": 0.5892306715949471,
358
+ "learning_rate": 5e-06,
359
+ "loss": 0.5292,
360
+ "step": 490
361
+ },
362
+ {
363
+ "epoch": 1.2453300124533002,
364
+ "grad_norm": 0.6342257885292061,
365
+ "learning_rate": 5e-06,
366
+ "loss": 0.5282,
367
+ "step": 500
368
+ },
369
+ {
370
+ "epoch": 1.270236612702366,
371
+ "grad_norm": 0.6358865929230534,
372
+ "learning_rate": 5e-06,
373
+ "loss": 0.5256,
374
+ "step": 510
375
+ },
376
+ {
377
+ "epoch": 1.2951432129514322,
378
+ "grad_norm": 0.5977464268642516,
379
+ "learning_rate": 5e-06,
380
+ "loss": 0.5216,
381
+ "step": 520
382
+ },
383
+ {
384
+ "epoch": 1.320049813200498,
385
+ "grad_norm": 0.5713712938047752,
386
+ "learning_rate": 5e-06,
387
+ "loss": 0.5246,
388
+ "step": 530
389
+ },
390
+ {
391
+ "epoch": 1.3449564134495642,
392
+ "grad_norm": 0.5806626902728351,
393
+ "learning_rate": 5e-06,
394
+ "loss": 0.522,
395
+ "step": 540
396
+ },
397
+ {
398
+ "epoch": 1.36986301369863,
399
+ "grad_norm": 0.6614502597905593,
400
+ "learning_rate": 5e-06,
401
+ "loss": 0.5232,
402
+ "step": 550
403
+ },
404
+ {
405
+ "epoch": 1.3947696139476962,
406
+ "grad_norm": 0.5826714843440824,
407
+ "learning_rate": 5e-06,
408
+ "loss": 0.5243,
409
+ "step": 560
410
+ },
411
+ {
412
+ "epoch": 1.419676214196762,
413
+ "grad_norm": 0.5870354154595686,
414
+ "learning_rate": 5e-06,
415
+ "loss": 0.5277,
416
+ "step": 570
417
+ },
418
+ {
419
+ "epoch": 1.4445828144458281,
420
+ "grad_norm": 0.5517404762107168,
421
+ "learning_rate": 5e-06,
422
+ "loss": 0.5185,
423
+ "step": 580
424
+ },
425
+ {
426
+ "epoch": 1.4694894146948942,
427
+ "grad_norm": 0.6548505841520894,
428
+ "learning_rate": 5e-06,
429
+ "loss": 0.5254,
430
+ "step": 590
431
+ },
432
+ {
433
+ "epoch": 1.4943960149439601,
434
+ "grad_norm": 0.4798373315690756,
435
+ "learning_rate": 5e-06,
436
+ "loss": 0.5223,
437
+ "step": 600
438
+ },
439
+ {
440
+ "epoch": 1.519302615193026,
441
+ "grad_norm": 0.561260368162772,
442
+ "learning_rate": 5e-06,
443
+ "loss": 0.5208,
444
+ "step": 610
445
+ },
446
+ {
447
+ "epoch": 1.544209215442092,
448
+ "grad_norm": 0.4776435289710049,
449
+ "learning_rate": 5e-06,
450
+ "loss": 0.5169,
451
+ "step": 620
452
+ },
453
+ {
454
+ "epoch": 1.5691158156911582,
455
+ "grad_norm": 0.5736062837326033,
456
+ "learning_rate": 5e-06,
457
+ "loss": 0.518,
458
+ "step": 630
459
+ },
460
+ {
461
+ "epoch": 1.5940224159402243,
462
+ "grad_norm": 0.5605397398468376,
463
+ "learning_rate": 5e-06,
464
+ "loss": 0.519,
465
+ "step": 640
466
+ },
467
+ {
468
+ "epoch": 1.6189290161892902,
469
+ "grad_norm": 0.5414003173485504,
470
+ "learning_rate": 5e-06,
471
+ "loss": 0.5141,
472
+ "step": 650
473
+ },
474
+ {
475
+ "epoch": 1.643835616438356,
476
+ "grad_norm": 0.5029725162688677,
477
+ "learning_rate": 5e-06,
478
+ "loss": 0.5123,
479
+ "step": 660
480
+ },
481
+ {
482
+ "epoch": 1.6687422166874222,
483
+ "grad_norm": 0.49094415690364374,
484
+ "learning_rate": 5e-06,
485
+ "loss": 0.5103,
486
+ "step": 670
487
+ },
488
+ {
489
+ "epoch": 1.6936488169364883,
490
+ "grad_norm": 0.6443717801298894,
491
+ "learning_rate": 5e-06,
492
+ "loss": 0.5173,
493
+ "step": 680
494
+ },
495
+ {
496
+ "epoch": 1.7185554171855542,
497
+ "grad_norm": 0.6131661065579275,
498
+ "learning_rate": 5e-06,
499
+ "loss": 0.5123,
500
+ "step": 690
501
+ },
502
+ {
503
+ "epoch": 1.74346201743462,
504
+ "grad_norm": 0.5102402228309177,
505
+ "learning_rate": 5e-06,
506
+ "loss": 0.5117,
507
+ "step": 700
508
+ },
509
+ {
510
+ "epoch": 1.7683686176836861,
511
+ "grad_norm": 0.6532768204087063,
512
+ "learning_rate": 5e-06,
513
+ "loss": 0.5116,
514
+ "step": 710
515
+ },
516
+ {
517
+ "epoch": 1.7932752179327522,
518
+ "grad_norm": 0.5359103919656766,
519
+ "learning_rate": 5e-06,
520
+ "loss": 0.5112,
521
+ "step": 720
522
+ },
523
+ {
524
+ "epoch": 1.8181818181818183,
525
+ "grad_norm": 0.4654117626073739,
526
+ "learning_rate": 5e-06,
527
+ "loss": 0.5108,
528
+ "step": 730
529
+ },
530
+ {
531
+ "epoch": 1.8430884184308842,
532
+ "grad_norm": 0.4816605018327504,
533
+ "learning_rate": 5e-06,
534
+ "loss": 0.5125,
535
+ "step": 740
536
+ },
537
+ {
538
+ "epoch": 1.86799501867995,
539
+ "grad_norm": 0.5078445355540058,
540
+ "learning_rate": 5e-06,
541
+ "loss": 0.5174,
542
+ "step": 750
543
+ },
544
+ {
545
+ "epoch": 1.8929016189290162,
546
+ "grad_norm": 0.5123586401271214,
547
+ "learning_rate": 5e-06,
548
+ "loss": 0.5103,
549
+ "step": 760
550
+ },
551
+ {
552
+ "epoch": 1.9178082191780823,
553
+ "grad_norm": 0.512594100070009,
554
+ "learning_rate": 5e-06,
555
+ "loss": 0.5125,
556
+ "step": 770
557
+ },
558
+ {
559
+ "epoch": 1.9427148194271482,
560
+ "grad_norm": 0.5251535719075618,
561
+ "learning_rate": 5e-06,
562
+ "loss": 0.5097,
563
+ "step": 780
564
+ },
565
+ {
566
+ "epoch": 1.967621419676214,
567
+ "grad_norm": 0.5598844434696433,
568
+ "learning_rate": 5e-06,
569
+ "loss": 0.5114,
570
+ "step": 790
571
+ },
572
+ {
573
+ "epoch": 1.9925280199252802,
574
+ "grad_norm": 0.5580903372050761,
575
+ "learning_rate": 5e-06,
576
+ "loss": 0.5112,
577
+ "step": 800
578
+ },
579
+ {
580
+ "epoch": 2.0,
581
+ "eval_loss": 0.5391710996627808,
582
+ "eval_runtime": 215.2899,
583
+ "eval_samples_per_second": 50.253,
584
+ "eval_steps_per_second": 0.395,
585
+ "step": 803
586
+ },
587
+ {
588
+ "epoch": 2.0174346201743463,
589
+ "grad_norm": 0.7489739683512228,
590
+ "learning_rate": 5e-06,
591
+ "loss": 0.4772,
592
+ "step": 810
593
+ },
594
+ {
595
+ "epoch": 2.0423412204234124,
596
+ "grad_norm": 0.6206410773051714,
597
+ "learning_rate": 5e-06,
598
+ "loss": 0.4773,
599
+ "step": 820
600
+ },
601
+ {
602
+ "epoch": 2.067247820672478,
603
+ "grad_norm": 0.5964416221117852,
604
+ "learning_rate": 5e-06,
605
+ "loss": 0.4774,
606
+ "step": 830
607
+ },
608
+ {
609
+ "epoch": 2.092154420921544,
610
+ "grad_norm": 0.7072977668899793,
611
+ "learning_rate": 5e-06,
612
+ "loss": 0.4734,
613
+ "step": 840
614
+ },
615
+ {
616
+ "epoch": 2.1170610211706102,
617
+ "grad_norm": 0.4869213806678473,
618
+ "learning_rate": 5e-06,
619
+ "loss": 0.4738,
620
+ "step": 850
621
+ },
622
+ {
623
+ "epoch": 2.1419676214196763,
624
+ "grad_norm": 0.5559708671866962,
625
+ "learning_rate": 5e-06,
626
+ "loss": 0.4739,
627
+ "step": 860
628
+ },
629
+ {
630
+ "epoch": 2.166874221668742,
631
+ "grad_norm": 0.590838180422237,
632
+ "learning_rate": 5e-06,
633
+ "loss": 0.4781,
634
+ "step": 870
635
+ },
636
+ {
637
+ "epoch": 2.191780821917808,
638
+ "grad_norm": 0.5449563311605545,
639
+ "learning_rate": 5e-06,
640
+ "loss": 0.4768,
641
+ "step": 880
642
+ },
643
+ {
644
+ "epoch": 2.216687422166874,
645
+ "grad_norm": 0.49509703126867355,
646
+ "learning_rate": 5e-06,
647
+ "loss": 0.4702,
648
+ "step": 890
649
+ },
650
+ {
651
+ "epoch": 2.2415940224159403,
652
+ "grad_norm": 0.583235606431074,
653
+ "learning_rate": 5e-06,
654
+ "loss": 0.4785,
655
+ "step": 900
656
+ },
657
+ {
658
+ "epoch": 2.2665006226650064,
659
+ "grad_norm": 0.5224340118110492,
660
+ "learning_rate": 5e-06,
661
+ "loss": 0.4688,
662
+ "step": 910
663
+ },
664
+ {
665
+ "epoch": 2.291407222914072,
666
+ "grad_norm": 0.5751594164855693,
667
+ "learning_rate": 5e-06,
668
+ "loss": 0.4685,
669
+ "step": 920
670
+ },
671
+ {
672
+ "epoch": 2.316313823163138,
673
+ "grad_norm": 0.5440703036591379,
674
+ "learning_rate": 5e-06,
675
+ "loss": 0.4773,
676
+ "step": 930
677
+ },
678
+ {
679
+ "epoch": 2.3412204234122043,
680
+ "grad_norm": 0.7610931862664628,
681
+ "learning_rate": 5e-06,
682
+ "loss": 0.472,
683
+ "step": 940
684
+ },
685
+ {
686
+ "epoch": 2.3661270236612704,
687
+ "grad_norm": 0.5531311055353737,
688
+ "learning_rate": 5e-06,
689
+ "loss": 0.4708,
690
+ "step": 950
691
+ },
692
+ {
693
+ "epoch": 2.391033623910336,
694
+ "grad_norm": 0.511600827539314,
695
+ "learning_rate": 5e-06,
696
+ "loss": 0.4764,
697
+ "step": 960
698
+ },
699
+ {
700
+ "epoch": 2.415940224159402,
701
+ "grad_norm": 0.5621388473604839,
702
+ "learning_rate": 5e-06,
703
+ "loss": 0.4792,
704
+ "step": 970
705
+ },
706
+ {
707
+ "epoch": 2.4408468244084682,
708
+ "grad_norm": 0.574647600528262,
709
+ "learning_rate": 5e-06,
710
+ "loss": 0.4722,
711
+ "step": 980
712
+ },
713
+ {
714
+ "epoch": 2.4657534246575343,
715
+ "grad_norm": 0.5174894983673209,
716
+ "learning_rate": 5e-06,
717
+ "loss": 0.4752,
718
+ "step": 990
719
+ },
720
+ {
721
+ "epoch": 2.4906600249066004,
722
+ "grad_norm": 0.49985821335778424,
723
+ "learning_rate": 5e-06,
724
+ "loss": 0.4824,
725
+ "step": 1000
726
+ },
727
+ {
728
+ "epoch": 2.515566625155666,
729
+ "grad_norm": 0.6533346544154589,
730
+ "learning_rate": 5e-06,
731
+ "loss": 0.468,
732
+ "step": 1010
733
+ },
734
+ {
735
+ "epoch": 2.540473225404732,
736
+ "grad_norm": 0.5759385792339666,
737
+ "learning_rate": 5e-06,
738
+ "loss": 0.4759,
739
+ "step": 1020
740
+ },
741
+ {
742
+ "epoch": 2.5653798256537983,
743
+ "grad_norm": 0.5706835517348989,
744
+ "learning_rate": 5e-06,
745
+ "loss": 0.473,
746
+ "step": 1030
747
+ },
748
+ {
749
+ "epoch": 2.5902864259028644,
750
+ "grad_norm": 0.48023447741714353,
751
+ "learning_rate": 5e-06,
752
+ "loss": 0.4758,
753
+ "step": 1040
754
+ },
755
+ {
756
+ "epoch": 2.61519302615193,
757
+ "grad_norm": 0.48468151801419096,
758
+ "learning_rate": 5e-06,
759
+ "loss": 0.4721,
760
+ "step": 1050
761
+ },
762
+ {
763
+ "epoch": 2.640099626400996,
764
+ "grad_norm": 0.5091065263641408,
765
+ "learning_rate": 5e-06,
766
+ "loss": 0.479,
767
+ "step": 1060
768
+ },
769
+ {
770
+ "epoch": 2.6650062266500623,
771
+ "grad_norm": 0.5876312291375192,
772
+ "learning_rate": 5e-06,
773
+ "loss": 0.471,
774
+ "step": 1070
775
+ },
776
+ {
777
+ "epoch": 2.6899128268991284,
778
+ "grad_norm": 0.5815028633704035,
779
+ "learning_rate": 5e-06,
780
+ "loss": 0.4762,
781
+ "step": 1080
782
+ },
783
+ {
784
+ "epoch": 2.7148194271481945,
785
+ "grad_norm": 0.5605250992780186,
786
+ "learning_rate": 5e-06,
787
+ "loss": 0.4735,
788
+ "step": 1090
789
+ },
790
+ {
791
+ "epoch": 2.73972602739726,
792
+ "grad_norm": 0.5741942655339728,
793
+ "learning_rate": 5e-06,
794
+ "loss": 0.4747,
795
+ "step": 1100
796
+ },
797
+ {
798
+ "epoch": 2.7646326276463262,
799
+ "grad_norm": 0.4775069856578055,
800
+ "learning_rate": 5e-06,
801
+ "loss": 0.4757,
802
+ "step": 1110
803
+ },
804
+ {
805
+ "epoch": 2.7895392278953923,
806
+ "grad_norm": 0.498224788090082,
807
+ "learning_rate": 5e-06,
808
+ "loss": 0.4741,
809
+ "step": 1120
810
+ },
811
+ {
812
+ "epoch": 2.8144458281444584,
813
+ "grad_norm": 0.48637748558305866,
814
+ "learning_rate": 5e-06,
815
+ "loss": 0.4767,
816
+ "step": 1130
817
+ },
818
+ {
819
+ "epoch": 2.839352428393524,
820
+ "grad_norm": 0.5090237580915907,
821
+ "learning_rate": 5e-06,
822
+ "loss": 0.4734,
823
+ "step": 1140
824
+ },
825
+ {
826
+ "epoch": 2.86425902864259,
827
+ "grad_norm": 0.4947938642895003,
828
+ "learning_rate": 5e-06,
829
+ "loss": 0.4782,
830
+ "step": 1150
831
+ },
832
+ {
833
+ "epoch": 2.8891656288916563,
834
+ "grad_norm": 0.5195966815447782,
835
+ "learning_rate": 5e-06,
836
+ "loss": 0.4803,
837
+ "step": 1160
838
+ },
839
+ {
840
+ "epoch": 2.9140722291407224,
841
+ "grad_norm": 0.5212585367090384,
842
+ "learning_rate": 5e-06,
843
+ "loss": 0.4722,
844
+ "step": 1170
845
+ },
846
+ {
847
+ "epoch": 2.9389788293897885,
848
+ "grad_norm": 0.514479905781623,
849
+ "learning_rate": 5e-06,
850
+ "loss": 0.4752,
851
+ "step": 1180
852
+ },
853
+ {
854
+ "epoch": 2.963885429638854,
855
+ "grad_norm": 0.5039971892926054,
856
+ "learning_rate": 5e-06,
857
+ "loss": 0.475,
858
+ "step": 1190
859
+ },
860
+ {
861
+ "epoch": 2.9887920298879203,
862
+ "grad_norm": 0.4924909134578761,
863
+ "learning_rate": 5e-06,
864
+ "loss": 0.4697,
865
+ "step": 1200
866
+ },
867
+ {
868
+ "epoch": 2.99626400996264,
869
+ "eval_loss": 0.5331025123596191,
870
+ "eval_runtime": 217.1359,
871
+ "eval_samples_per_second": 49.826,
872
+ "eval_steps_per_second": 0.391,
873
+ "step": 1203
874
+ },
875
+ {
876
+ "epoch": 2.99626400996264,
877
+ "step": 1203,
878
+ "total_flos": 2014651046952960.0,
879
+ "train_loss": 0.5350556444943397,
880
+ "train_runtime": 35909.5127,
881
+ "train_samples_per_second": 17.172,
882
+ "train_steps_per_second": 0.034
883
+ }
884
+ ],
885
+ "logging_steps": 10,
886
+ "max_steps": 1203,
887
+ "num_input_tokens_seen": 0,
888
+ "num_train_epochs": 3,
889
+ "save_steps": 500,
890
+ "stateful_callbacks": {
891
+ "TrainerControl": {
892
+ "args": {
893
+ "should_epoch_stop": false,
894
+ "should_evaluate": false,
895
+ "should_log": false,
896
+ "should_save": true,
897
+ "should_training_stop": true
898
+ },
899
+ "attributes": {}
900
+ }
901
+ },
902
+ "total_flos": 2014651046952960.0,
903
+ "train_batch_size": 8,
904
+ "trial_name": null,
905
+ "trial_params": null
906
+ }
training_eval_loss.png ADDED
training_loss.png ADDED