sedrickkeh commited on
Commit
2dae4fa
·
verified ·
1 Parent(s): 7cee295

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_camel_biology_gpt-4o-mini
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
15
 
16
  # oh_v1_w_v3_camel_biology_gpt-4o-mini
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: 0.5550
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_camel_biology_gpt-4o-mini
 
16
 
17
  # oh_v1_w_v3_camel_biology_gpt-4o-mini
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_camel_biology_gpt-4o-mini dataset.
20
  It achieves the following results on the evaluation set:
21
  - Loss: 0.5550
22
 
all_results.json ADDED
@@ -0,0 +1,12 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.9962358845671266,
3
+ "eval_loss": 0.555037796497345,
4
+ "eval_runtime": 213.0252,
5
+ "eval_samples_per_second": 50.388,
6
+ "eval_steps_per_second": 0.394,
7
+ "total_flos": 1999575711744000.0,
8
+ "train_loss": 0.5572832309620664,
9
+ "train_runtime": 35786.5658,
10
+ "train_samples_per_second": 17.096,
11
+ "train_steps_per_second": 0.033
12
+ }
eval_results.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.9962358845671266,
3
+ "eval_loss": 0.555037796497345,
4
+ "eval_runtime": 213.0252,
5
+ "eval_samples_per_second": 50.388,
6
+ "eval_steps_per_second": 0.394
7
+ }
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.9962358845671266,
3
+ "total_flos": 1999575711744000.0,
4
+ "train_loss": 0.5572832309620664,
5
+ "train_runtime": 35786.5658,
6
+ "train_samples_per_second": 17.096,
7
+ "train_steps_per_second": 0.033
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,899 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 2.9962358845671266,
5
+ "eval_steps": 500,
6
+ "global_step": 1194,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.025094102885821833,
13
+ "grad_norm": 11.718188698364322,
14
+ "learning_rate": 5e-06,
15
+ "loss": 0.9141,
16
+ "step": 10
17
+ },
18
+ {
19
+ "epoch": 0.050188205771643665,
20
+ "grad_norm": 2.128194193417522,
21
+ "learning_rate": 5e-06,
22
+ "loss": 0.7836,
23
+ "step": 20
24
+ },
25
+ {
26
+ "epoch": 0.07528230865746549,
27
+ "grad_norm": 2.3757769329973657,
28
+ "learning_rate": 5e-06,
29
+ "loss": 0.7498,
30
+ "step": 30
31
+ },
32
+ {
33
+ "epoch": 0.10037641154328733,
34
+ "grad_norm": 1.103206879933866,
35
+ "learning_rate": 5e-06,
36
+ "loss": 0.7188,
37
+ "step": 40
38
+ },
39
+ {
40
+ "epoch": 0.12547051442910917,
41
+ "grad_norm": 0.9274722897954563,
42
+ "learning_rate": 5e-06,
43
+ "loss": 0.7002,
44
+ "step": 50
45
+ },
46
+ {
47
+ "epoch": 0.15056461731493098,
48
+ "grad_norm": 0.84713375375135,
49
+ "learning_rate": 5e-06,
50
+ "loss": 0.6765,
51
+ "step": 60
52
+ },
53
+ {
54
+ "epoch": 0.17565872020075282,
55
+ "grad_norm": 0.7850634011267035,
56
+ "learning_rate": 5e-06,
57
+ "loss": 0.6679,
58
+ "step": 70
59
+ },
60
+ {
61
+ "epoch": 0.20075282308657466,
62
+ "grad_norm": 0.6763495874297807,
63
+ "learning_rate": 5e-06,
64
+ "loss": 0.6554,
65
+ "step": 80
66
+ },
67
+ {
68
+ "epoch": 0.2258469259723965,
69
+ "grad_norm": 0.6460889441423061,
70
+ "learning_rate": 5e-06,
71
+ "loss": 0.6537,
72
+ "step": 90
73
+ },
74
+ {
75
+ "epoch": 0.25094102885821834,
76
+ "grad_norm": 0.6381839585959109,
77
+ "learning_rate": 5e-06,
78
+ "loss": 0.6469,
79
+ "step": 100
80
+ },
81
+ {
82
+ "epoch": 0.27603513174404015,
83
+ "grad_norm": 0.5934854411636431,
84
+ "learning_rate": 5e-06,
85
+ "loss": 0.6408,
86
+ "step": 110
87
+ },
88
+ {
89
+ "epoch": 0.30112923462986196,
90
+ "grad_norm": 0.5908335131781053,
91
+ "learning_rate": 5e-06,
92
+ "loss": 0.6315,
93
+ "step": 120
94
+ },
95
+ {
96
+ "epoch": 0.32622333751568383,
97
+ "grad_norm": 0.6207583357857969,
98
+ "learning_rate": 5e-06,
99
+ "loss": 0.6295,
100
+ "step": 130
101
+ },
102
+ {
103
+ "epoch": 0.35131744040150564,
104
+ "grad_norm": 0.5777218463460926,
105
+ "learning_rate": 5e-06,
106
+ "loss": 0.6353,
107
+ "step": 140
108
+ },
109
+ {
110
+ "epoch": 0.37641154328732745,
111
+ "grad_norm": 0.8031543297119933,
112
+ "learning_rate": 5e-06,
113
+ "loss": 0.6257,
114
+ "step": 150
115
+ },
116
+ {
117
+ "epoch": 0.4015056461731493,
118
+ "grad_norm": 0.6565972172028122,
119
+ "learning_rate": 5e-06,
120
+ "loss": 0.6223,
121
+ "step": 160
122
+ },
123
+ {
124
+ "epoch": 0.42659974905897113,
125
+ "grad_norm": 0.5758668737369184,
126
+ "learning_rate": 5e-06,
127
+ "loss": 0.6238,
128
+ "step": 170
129
+ },
130
+ {
131
+ "epoch": 0.451693851944793,
132
+ "grad_norm": 0.5079092787071162,
133
+ "learning_rate": 5e-06,
134
+ "loss": 0.6143,
135
+ "step": 180
136
+ },
137
+ {
138
+ "epoch": 0.4767879548306148,
139
+ "grad_norm": 0.6379157115232267,
140
+ "learning_rate": 5e-06,
141
+ "loss": 0.6199,
142
+ "step": 190
143
+ },
144
+ {
145
+ "epoch": 0.5018820577164367,
146
+ "grad_norm": 0.5743755502713278,
147
+ "learning_rate": 5e-06,
148
+ "loss": 0.6129,
149
+ "step": 200
150
+ },
151
+ {
152
+ "epoch": 0.5269761606022585,
153
+ "grad_norm": 0.7914784193340709,
154
+ "learning_rate": 5e-06,
155
+ "loss": 0.609,
156
+ "step": 210
157
+ },
158
+ {
159
+ "epoch": 0.5520702634880803,
160
+ "grad_norm": 0.6052863906299388,
161
+ "learning_rate": 5e-06,
162
+ "loss": 0.6191,
163
+ "step": 220
164
+ },
165
+ {
166
+ "epoch": 0.5771643663739021,
167
+ "grad_norm": 0.6079363974587813,
168
+ "learning_rate": 5e-06,
169
+ "loss": 0.6036,
170
+ "step": 230
171
+ },
172
+ {
173
+ "epoch": 0.6022584692597239,
174
+ "grad_norm": 0.6317013898545292,
175
+ "learning_rate": 5e-06,
176
+ "loss": 0.6101,
177
+ "step": 240
178
+ },
179
+ {
180
+ "epoch": 0.6273525721455459,
181
+ "grad_norm": 0.6772767650378945,
182
+ "learning_rate": 5e-06,
183
+ "loss": 0.6066,
184
+ "step": 250
185
+ },
186
+ {
187
+ "epoch": 0.6524466750313677,
188
+ "grad_norm": 0.7659397239810531,
189
+ "learning_rate": 5e-06,
190
+ "loss": 0.6085,
191
+ "step": 260
192
+ },
193
+ {
194
+ "epoch": 0.6775407779171895,
195
+ "grad_norm": 0.8301922378020979,
196
+ "learning_rate": 5e-06,
197
+ "loss": 0.6054,
198
+ "step": 270
199
+ },
200
+ {
201
+ "epoch": 0.7026348808030113,
202
+ "grad_norm": 0.4548488940244719,
203
+ "learning_rate": 5e-06,
204
+ "loss": 0.6039,
205
+ "step": 280
206
+ },
207
+ {
208
+ "epoch": 0.7277289836888331,
209
+ "grad_norm": 0.6991563441452833,
210
+ "learning_rate": 5e-06,
211
+ "loss": 0.6041,
212
+ "step": 290
213
+ },
214
+ {
215
+ "epoch": 0.7528230865746549,
216
+ "grad_norm": 0.7252866874449959,
217
+ "learning_rate": 5e-06,
218
+ "loss": 0.5982,
219
+ "step": 300
220
+ },
221
+ {
222
+ "epoch": 0.7779171894604768,
223
+ "grad_norm": 0.5621384726961683,
224
+ "learning_rate": 5e-06,
225
+ "loss": 0.5984,
226
+ "step": 310
227
+ },
228
+ {
229
+ "epoch": 0.8030112923462986,
230
+ "grad_norm": 0.5544205898879206,
231
+ "learning_rate": 5e-06,
232
+ "loss": 0.5938,
233
+ "step": 320
234
+ },
235
+ {
236
+ "epoch": 0.8281053952321205,
237
+ "grad_norm": 0.6591982327209397,
238
+ "learning_rate": 5e-06,
239
+ "loss": 0.5949,
240
+ "step": 330
241
+ },
242
+ {
243
+ "epoch": 0.8531994981179423,
244
+ "grad_norm": 0.4966951651690375,
245
+ "learning_rate": 5e-06,
246
+ "loss": 0.5939,
247
+ "step": 340
248
+ },
249
+ {
250
+ "epoch": 0.8782936010037641,
251
+ "grad_norm": 0.7936946361545921,
252
+ "learning_rate": 5e-06,
253
+ "loss": 0.5952,
254
+ "step": 350
255
+ },
256
+ {
257
+ "epoch": 0.903387703889586,
258
+ "grad_norm": 0.6049775996861211,
259
+ "learning_rate": 5e-06,
260
+ "loss": 0.5828,
261
+ "step": 360
262
+ },
263
+ {
264
+ "epoch": 0.9284818067754078,
265
+ "grad_norm": 0.5343638273778424,
266
+ "learning_rate": 5e-06,
267
+ "loss": 0.5851,
268
+ "step": 370
269
+ },
270
+ {
271
+ "epoch": 0.9535759096612296,
272
+ "grad_norm": 0.5723040103769105,
273
+ "learning_rate": 5e-06,
274
+ "loss": 0.5912,
275
+ "step": 380
276
+ },
277
+ {
278
+ "epoch": 0.9786700125470514,
279
+ "grad_norm": 0.5959442881234533,
280
+ "learning_rate": 5e-06,
281
+ "loss": 0.5915,
282
+ "step": 390
283
+ },
284
+ {
285
+ "epoch": 0.998745294855709,
286
+ "eval_loss": 0.5871068239212036,
287
+ "eval_runtime": 212.6533,
288
+ "eval_samples_per_second": 50.477,
289
+ "eval_steps_per_second": 0.395,
290
+ "step": 398
291
+ },
292
+ {
293
+ "epoch": 1.0037641154328734,
294
+ "grad_norm": 0.5888812035860913,
295
+ "learning_rate": 5e-06,
296
+ "loss": 0.5826,
297
+ "step": 400
298
+ },
299
+ {
300
+ "epoch": 1.0288582183186952,
301
+ "grad_norm": 0.496571742956618,
302
+ "learning_rate": 5e-06,
303
+ "loss": 0.5559,
304
+ "step": 410
305
+ },
306
+ {
307
+ "epoch": 1.053952321204517,
308
+ "grad_norm": 0.6292730066506935,
309
+ "learning_rate": 5e-06,
310
+ "loss": 0.5513,
311
+ "step": 420
312
+ },
313
+ {
314
+ "epoch": 1.0790464240903388,
315
+ "grad_norm": 0.588641857548959,
316
+ "learning_rate": 5e-06,
317
+ "loss": 0.5507,
318
+ "step": 430
319
+ },
320
+ {
321
+ "epoch": 1.1041405269761606,
322
+ "grad_norm": 0.6769239280680559,
323
+ "learning_rate": 5e-06,
324
+ "loss": 0.5503,
325
+ "step": 440
326
+ },
327
+ {
328
+ "epoch": 1.1292346298619824,
329
+ "grad_norm": 0.5417691042097619,
330
+ "learning_rate": 5e-06,
331
+ "loss": 0.548,
332
+ "step": 450
333
+ },
334
+ {
335
+ "epoch": 1.1543287327478042,
336
+ "grad_norm": 0.5991902330546697,
337
+ "learning_rate": 5e-06,
338
+ "loss": 0.5384,
339
+ "step": 460
340
+ },
341
+ {
342
+ "epoch": 1.179422835633626,
343
+ "grad_norm": 0.5820599790063293,
344
+ "learning_rate": 5e-06,
345
+ "loss": 0.5487,
346
+ "step": 470
347
+ },
348
+ {
349
+ "epoch": 1.2045169385194479,
350
+ "grad_norm": 0.5519292876234776,
351
+ "learning_rate": 5e-06,
352
+ "loss": 0.5427,
353
+ "step": 480
354
+ },
355
+ {
356
+ "epoch": 1.2296110414052697,
357
+ "grad_norm": 0.49516603784025776,
358
+ "learning_rate": 5e-06,
359
+ "loss": 0.546,
360
+ "step": 490
361
+ },
362
+ {
363
+ "epoch": 1.2547051442910915,
364
+ "grad_norm": 0.5743423113593866,
365
+ "learning_rate": 5e-06,
366
+ "loss": 0.5393,
367
+ "step": 500
368
+ },
369
+ {
370
+ "epoch": 1.2797992471769133,
371
+ "grad_norm": 0.5535831554138544,
372
+ "learning_rate": 5e-06,
373
+ "loss": 0.547,
374
+ "step": 510
375
+ },
376
+ {
377
+ "epoch": 1.3048933500627353,
378
+ "grad_norm": 0.5141692394527988,
379
+ "learning_rate": 5e-06,
380
+ "loss": 0.5411,
381
+ "step": 520
382
+ },
383
+ {
384
+ "epoch": 1.3299874529485571,
385
+ "grad_norm": 0.4854378839815677,
386
+ "learning_rate": 5e-06,
387
+ "loss": 0.5442,
388
+ "step": 530
389
+ },
390
+ {
391
+ "epoch": 1.355081555834379,
392
+ "grad_norm": 0.46518640648830495,
393
+ "learning_rate": 5e-06,
394
+ "loss": 0.5427,
395
+ "step": 540
396
+ },
397
+ {
398
+ "epoch": 1.3801756587202008,
399
+ "grad_norm": 0.534031920116045,
400
+ "learning_rate": 5e-06,
401
+ "loss": 0.5417,
402
+ "step": 550
403
+ },
404
+ {
405
+ "epoch": 1.4052697616060226,
406
+ "grad_norm": 0.5911980981423581,
407
+ "learning_rate": 5e-06,
408
+ "loss": 0.5435,
409
+ "step": 560
410
+ },
411
+ {
412
+ "epoch": 1.4303638644918444,
413
+ "grad_norm": 0.5763366884781533,
414
+ "learning_rate": 5e-06,
415
+ "loss": 0.5435,
416
+ "step": 570
417
+ },
418
+ {
419
+ "epoch": 1.4554579673776662,
420
+ "grad_norm": 0.6044164192343925,
421
+ "learning_rate": 5e-06,
422
+ "loss": 0.5458,
423
+ "step": 580
424
+ },
425
+ {
426
+ "epoch": 1.480552070263488,
427
+ "grad_norm": 0.5136929530377343,
428
+ "learning_rate": 5e-06,
429
+ "loss": 0.5433,
430
+ "step": 590
431
+ },
432
+ {
433
+ "epoch": 1.50564617314931,
434
+ "grad_norm": 0.6480890499633274,
435
+ "learning_rate": 5e-06,
436
+ "loss": 0.5412,
437
+ "step": 600
438
+ },
439
+ {
440
+ "epoch": 1.5307402760351319,
441
+ "grad_norm": 0.5778909116986595,
442
+ "learning_rate": 5e-06,
443
+ "loss": 0.5366,
444
+ "step": 610
445
+ },
446
+ {
447
+ "epoch": 1.5558343789209537,
448
+ "grad_norm": 0.554397297340124,
449
+ "learning_rate": 5e-06,
450
+ "loss": 0.5424,
451
+ "step": 620
452
+ },
453
+ {
454
+ "epoch": 1.5809284818067755,
455
+ "grad_norm": 0.512796007100603,
456
+ "learning_rate": 5e-06,
457
+ "loss": 0.5408,
458
+ "step": 630
459
+ },
460
+ {
461
+ "epoch": 1.6060225846925973,
462
+ "grad_norm": 0.4758665935260214,
463
+ "learning_rate": 5e-06,
464
+ "loss": 0.5418,
465
+ "step": 640
466
+ },
467
+ {
468
+ "epoch": 1.631116687578419,
469
+ "grad_norm": 0.4968644389707275,
470
+ "learning_rate": 5e-06,
471
+ "loss": 0.535,
472
+ "step": 650
473
+ },
474
+ {
475
+ "epoch": 1.656210790464241,
476
+ "grad_norm": 0.5920011846841533,
477
+ "learning_rate": 5e-06,
478
+ "loss": 0.54,
479
+ "step": 660
480
+ },
481
+ {
482
+ "epoch": 1.6813048933500627,
483
+ "grad_norm": 0.5905169032972927,
484
+ "learning_rate": 5e-06,
485
+ "loss": 0.5381,
486
+ "step": 670
487
+ },
488
+ {
489
+ "epoch": 1.7063989962358845,
490
+ "grad_norm": 0.6296689735308911,
491
+ "learning_rate": 5e-06,
492
+ "loss": 0.5358,
493
+ "step": 680
494
+ },
495
+ {
496
+ "epoch": 1.7314930991217063,
497
+ "grad_norm": 0.47536620723191336,
498
+ "learning_rate": 5e-06,
499
+ "loss": 0.5379,
500
+ "step": 690
501
+ },
502
+ {
503
+ "epoch": 1.7565872020075282,
504
+ "grad_norm": 0.5683042454275448,
505
+ "learning_rate": 5e-06,
506
+ "loss": 0.5353,
507
+ "step": 700
508
+ },
509
+ {
510
+ "epoch": 1.78168130489335,
511
+ "grad_norm": 0.5865132456497922,
512
+ "learning_rate": 5e-06,
513
+ "loss": 0.5341,
514
+ "step": 710
515
+ },
516
+ {
517
+ "epoch": 1.8067754077791718,
518
+ "grad_norm": 0.5863809886365764,
519
+ "learning_rate": 5e-06,
520
+ "loss": 0.5385,
521
+ "step": 720
522
+ },
523
+ {
524
+ "epoch": 1.8318695106649936,
525
+ "grad_norm": 0.47137949702370147,
526
+ "learning_rate": 5e-06,
527
+ "loss": 0.5349,
528
+ "step": 730
529
+ },
530
+ {
531
+ "epoch": 1.8569636135508154,
532
+ "grad_norm": 0.5139487832974206,
533
+ "learning_rate": 5e-06,
534
+ "loss": 0.5351,
535
+ "step": 740
536
+ },
537
+ {
538
+ "epoch": 1.8820577164366374,
539
+ "grad_norm": 0.5728562358927324,
540
+ "learning_rate": 5e-06,
541
+ "loss": 0.5387,
542
+ "step": 750
543
+ },
544
+ {
545
+ "epoch": 1.9071518193224593,
546
+ "grad_norm": 0.6768594215912966,
547
+ "learning_rate": 5e-06,
548
+ "loss": 0.5364,
549
+ "step": 760
550
+ },
551
+ {
552
+ "epoch": 1.932245922208281,
553
+ "grad_norm": 0.5485345915469325,
554
+ "learning_rate": 5e-06,
555
+ "loss": 0.5394,
556
+ "step": 770
557
+ },
558
+ {
559
+ "epoch": 1.9573400250941029,
560
+ "grad_norm": 0.5148704568756698,
561
+ "learning_rate": 5e-06,
562
+ "loss": 0.5358,
563
+ "step": 780
564
+ },
565
+ {
566
+ "epoch": 1.9824341279799247,
567
+ "grad_norm": 0.48466535956929385,
568
+ "learning_rate": 5e-06,
569
+ "loss": 0.5304,
570
+ "step": 790
571
+ },
572
+ {
573
+ "epoch": 2.0,
574
+ "eval_loss": 0.5621475577354431,
575
+ "eval_runtime": 213.3338,
576
+ "eval_samples_per_second": 50.316,
577
+ "eval_steps_per_second": 0.394,
578
+ "step": 797
579
+ },
580
+ {
581
+ "epoch": 2.0075282308657467,
582
+ "grad_norm": 0.6573006701729232,
583
+ "learning_rate": 5e-06,
584
+ "loss": 0.5167,
585
+ "step": 800
586
+ },
587
+ {
588
+ "epoch": 2.0326223337515685,
589
+ "grad_norm": 0.7871500030029548,
590
+ "learning_rate": 5e-06,
591
+ "loss": 0.4942,
592
+ "step": 810
593
+ },
594
+ {
595
+ "epoch": 2.0577164366373903,
596
+ "grad_norm": 0.5640563491273413,
597
+ "learning_rate": 5e-06,
598
+ "loss": 0.4956,
599
+ "step": 820
600
+ },
601
+ {
602
+ "epoch": 2.082810539523212,
603
+ "grad_norm": 0.7511306820222987,
604
+ "learning_rate": 5e-06,
605
+ "loss": 0.4938,
606
+ "step": 830
607
+ },
608
+ {
609
+ "epoch": 2.107904642409034,
610
+ "grad_norm": 0.5817278040608698,
611
+ "learning_rate": 5e-06,
612
+ "loss": 0.4925,
613
+ "step": 840
614
+ },
615
+ {
616
+ "epoch": 2.132998745294856,
617
+ "grad_norm": 0.5443042434696377,
618
+ "learning_rate": 5e-06,
619
+ "loss": 0.4924,
620
+ "step": 850
621
+ },
622
+ {
623
+ "epoch": 2.1580928481806776,
624
+ "grad_norm": 0.5307201281775036,
625
+ "learning_rate": 5e-06,
626
+ "loss": 0.5,
627
+ "step": 860
628
+ },
629
+ {
630
+ "epoch": 2.1831869510664994,
631
+ "grad_norm": 0.5134207088872962,
632
+ "learning_rate": 5e-06,
633
+ "loss": 0.4948,
634
+ "step": 870
635
+ },
636
+ {
637
+ "epoch": 2.208281053952321,
638
+ "grad_norm": 0.6926705508270693,
639
+ "learning_rate": 5e-06,
640
+ "loss": 0.497,
641
+ "step": 880
642
+ },
643
+ {
644
+ "epoch": 2.233375156838143,
645
+ "grad_norm": 0.6946700857047731,
646
+ "learning_rate": 5e-06,
647
+ "loss": 0.4922,
648
+ "step": 890
649
+ },
650
+ {
651
+ "epoch": 2.258469259723965,
652
+ "grad_norm": 0.5473927966647341,
653
+ "learning_rate": 5e-06,
654
+ "loss": 0.4992,
655
+ "step": 900
656
+ },
657
+ {
658
+ "epoch": 2.2835633626097867,
659
+ "grad_norm": 0.6500276609885082,
660
+ "learning_rate": 5e-06,
661
+ "loss": 0.4932,
662
+ "step": 910
663
+ },
664
+ {
665
+ "epoch": 2.3086574654956085,
666
+ "grad_norm": 0.5514076764002837,
667
+ "learning_rate": 5e-06,
668
+ "loss": 0.4915,
669
+ "step": 920
670
+ },
671
+ {
672
+ "epoch": 2.3337515683814303,
673
+ "grad_norm": 0.49943208945627254,
674
+ "learning_rate": 5e-06,
675
+ "loss": 0.4999,
676
+ "step": 930
677
+ },
678
+ {
679
+ "epoch": 2.358845671267252,
680
+ "grad_norm": 0.6060583960970398,
681
+ "learning_rate": 5e-06,
682
+ "loss": 0.4953,
683
+ "step": 940
684
+ },
685
+ {
686
+ "epoch": 2.383939774153074,
687
+ "grad_norm": 0.5320264242785274,
688
+ "learning_rate": 5e-06,
689
+ "loss": 0.4964,
690
+ "step": 950
691
+ },
692
+ {
693
+ "epoch": 2.4090338770388957,
694
+ "grad_norm": 0.5858837169303328,
695
+ "learning_rate": 5e-06,
696
+ "loss": 0.5006,
697
+ "step": 960
698
+ },
699
+ {
700
+ "epoch": 2.4341279799247175,
701
+ "grad_norm": 0.5441932891457427,
702
+ "learning_rate": 5e-06,
703
+ "loss": 0.4927,
704
+ "step": 970
705
+ },
706
+ {
707
+ "epoch": 2.4592220828105393,
708
+ "grad_norm": 0.49825625085166125,
709
+ "learning_rate": 5e-06,
710
+ "loss": 0.4986,
711
+ "step": 980
712
+ },
713
+ {
714
+ "epoch": 2.484316185696361,
715
+ "grad_norm": 0.5693348465052496,
716
+ "learning_rate": 5e-06,
717
+ "loss": 0.4923,
718
+ "step": 990
719
+ },
720
+ {
721
+ "epoch": 2.509410288582183,
722
+ "grad_norm": 0.5198895218070472,
723
+ "learning_rate": 5e-06,
724
+ "loss": 0.4961,
725
+ "step": 1000
726
+ },
727
+ {
728
+ "epoch": 2.5345043914680048,
729
+ "grad_norm": 0.5822359089459672,
730
+ "learning_rate": 5e-06,
731
+ "loss": 0.4948,
732
+ "step": 1010
733
+ },
734
+ {
735
+ "epoch": 2.5595984943538266,
736
+ "grad_norm": 0.7250677792140846,
737
+ "learning_rate": 5e-06,
738
+ "loss": 0.4988,
739
+ "step": 1020
740
+ },
741
+ {
742
+ "epoch": 2.584692597239649,
743
+ "grad_norm": 0.49974126038635164,
744
+ "learning_rate": 5e-06,
745
+ "loss": 0.4915,
746
+ "step": 1030
747
+ },
748
+ {
749
+ "epoch": 2.6097867001254706,
750
+ "grad_norm": 0.4964480117175315,
751
+ "learning_rate": 5e-06,
752
+ "loss": 0.4902,
753
+ "step": 1040
754
+ },
755
+ {
756
+ "epoch": 2.6348808030112925,
757
+ "grad_norm": 0.5288705472818386,
758
+ "learning_rate": 5e-06,
759
+ "loss": 0.497,
760
+ "step": 1050
761
+ },
762
+ {
763
+ "epoch": 2.6599749058971143,
764
+ "grad_norm": 0.4795064924818655,
765
+ "learning_rate": 5e-06,
766
+ "loss": 0.5,
767
+ "step": 1060
768
+ },
769
+ {
770
+ "epoch": 2.685069008782936,
771
+ "grad_norm": 0.5639285671354385,
772
+ "learning_rate": 5e-06,
773
+ "loss": 0.496,
774
+ "step": 1070
775
+ },
776
+ {
777
+ "epoch": 2.710163111668758,
778
+ "grad_norm": 0.4890610843848083,
779
+ "learning_rate": 5e-06,
780
+ "loss": 0.497,
781
+ "step": 1080
782
+ },
783
+ {
784
+ "epoch": 2.7352572145545797,
785
+ "grad_norm": 0.5718819316899499,
786
+ "learning_rate": 5e-06,
787
+ "loss": 0.4948,
788
+ "step": 1090
789
+ },
790
+ {
791
+ "epoch": 2.7603513174404015,
792
+ "grad_norm": 0.564322354041535,
793
+ "learning_rate": 5e-06,
794
+ "loss": 0.492,
795
+ "step": 1100
796
+ },
797
+ {
798
+ "epoch": 2.7854454203262233,
799
+ "grad_norm": 0.5775080175263282,
800
+ "learning_rate": 5e-06,
801
+ "loss": 0.4957,
802
+ "step": 1110
803
+ },
804
+ {
805
+ "epoch": 2.810539523212045,
806
+ "grad_norm": 0.553905770960013,
807
+ "learning_rate": 5e-06,
808
+ "loss": 0.4993,
809
+ "step": 1120
810
+ },
811
+ {
812
+ "epoch": 2.835633626097867,
813
+ "grad_norm": 0.5725357035362282,
814
+ "learning_rate": 5e-06,
815
+ "loss": 0.4972,
816
+ "step": 1130
817
+ },
818
+ {
819
+ "epoch": 2.8607277289836888,
820
+ "grad_norm": 0.5476057683573488,
821
+ "learning_rate": 5e-06,
822
+ "loss": 0.4928,
823
+ "step": 1140
824
+ },
825
+ {
826
+ "epoch": 2.8858218318695106,
827
+ "grad_norm": 0.49125993261554296,
828
+ "learning_rate": 5e-06,
829
+ "loss": 0.4908,
830
+ "step": 1150
831
+ },
832
+ {
833
+ "epoch": 2.9109159347553324,
834
+ "grad_norm": 0.530302463940866,
835
+ "learning_rate": 5e-06,
836
+ "loss": 0.4962,
837
+ "step": 1160
838
+ },
839
+ {
840
+ "epoch": 2.936010037641154,
841
+ "grad_norm": 0.47444202016080034,
842
+ "learning_rate": 5e-06,
843
+ "loss": 0.498,
844
+ "step": 1170
845
+ },
846
+ {
847
+ "epoch": 2.961104140526976,
848
+ "grad_norm": 0.49422829602754365,
849
+ "learning_rate": 5e-06,
850
+ "loss": 0.4885,
851
+ "step": 1180
852
+ },
853
+ {
854
+ "epoch": 2.9861982434127983,
855
+ "grad_norm": 0.47369939752959883,
856
+ "learning_rate": 5e-06,
857
+ "loss": 0.4907,
858
+ "step": 1190
859
+ },
860
+ {
861
+ "epoch": 2.9962358845671266,
862
+ "eval_loss": 0.555037796497345,
863
+ "eval_runtime": 214.1596,
864
+ "eval_samples_per_second": 50.121,
865
+ "eval_steps_per_second": 0.392,
866
+ "step": 1194
867
+ },
868
+ {
869
+ "epoch": 2.9962358845671266,
870
+ "step": 1194,
871
+ "total_flos": 1999575711744000.0,
872
+ "train_loss": 0.5572832309620664,
873
+ "train_runtime": 35786.5658,
874
+ "train_samples_per_second": 17.096,
875
+ "train_steps_per_second": 0.033
876
+ }
877
+ ],
878
+ "logging_steps": 10,
879
+ "max_steps": 1194,
880
+ "num_input_tokens_seen": 0,
881
+ "num_train_epochs": 3,
882
+ "save_steps": 500,
883
+ "stateful_callbacks": {
884
+ "TrainerControl": {
885
+ "args": {
886
+ "should_epoch_stop": false,
887
+ "should_evaluate": false,
888
+ "should_log": false,
889
+ "should_save": true,
890
+ "should_training_stop": true
891
+ },
892
+ "attributes": {}
893
+ }
894
+ },
895
+ "total_flos": 1999575711744000.0,
896
+ "train_batch_size": 8,
897
+ "trial_name": null,
898
+ "trial_params": null
899
+ }
training_eval_loss.png ADDED
training_loss.png ADDED