sedrickkeh commited on
Commit
1a42a7a
·
verified ·
1 Parent(s): 27cd8d2

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: hp_ablations_llama3_epoch3
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
15
 
16
  # hp_ablations_llama3_epoch3
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: 0.6414
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: hp_ablations_llama3_epoch3
 
16
 
17
  # hp_ablations_llama3_epoch3
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-dcft-v3.1-gpt-4o-mini dataset.
20
  It achieves the following results on the evaluation set:
21
  - Loss: 0.6414
22
 
all_results.json ADDED
@@ -0,0 +1,12 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.9946777054997042,
3
+ "eval_loss": 0.6413525938987732,
4
+ "eval_runtime": 226.2987,
5
+ "eval_samples_per_second": 50.327,
6
+ "eval_steps_per_second": 0.393,
7
+ "total_flos": 2120178393415680.0,
8
+ "train_loss": 0.6193434227887676,
9
+ "train_runtime": 37811.5197,
10
+ "train_samples_per_second": 17.167,
11
+ "train_steps_per_second": 0.033
12
+ }
eval_results.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.9946777054997042,
3
+ "eval_loss": 0.6413525938987732,
4
+ "eval_runtime": 226.2987,
5
+ "eval_samples_per_second": 50.327,
6
+ "eval_steps_per_second": 0.393
7
+ }
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.9946777054997042,
3
+ "total_flos": 2120178393415680.0,
4
+ "train_loss": 0.6193434227887676,
5
+ "train_runtime": 37811.5197,
6
+ "train_samples_per_second": 17.167,
7
+ "train_steps_per_second": 0.033
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,948 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 2.9946777054997042,
5
+ "eval_steps": 500,
6
+ "global_step": 1266,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.02365464222353637,
13
+ "grad_norm": 2.64785060100586,
14
+ "learning_rate": 5e-06,
15
+ "loss": 0.8881,
16
+ "step": 10
17
+ },
18
+ {
19
+ "epoch": 0.04730928444707274,
20
+ "grad_norm": 4.492435661685993,
21
+ "learning_rate": 5e-06,
22
+ "loss": 0.7888,
23
+ "step": 20
24
+ },
25
+ {
26
+ "epoch": 0.0709639266706091,
27
+ "grad_norm": 1.1116541886300364,
28
+ "learning_rate": 5e-06,
29
+ "loss": 0.7578,
30
+ "step": 30
31
+ },
32
+ {
33
+ "epoch": 0.09461856889414548,
34
+ "grad_norm": 1.218928037085855,
35
+ "learning_rate": 5e-06,
36
+ "loss": 0.7402,
37
+ "step": 40
38
+ },
39
+ {
40
+ "epoch": 0.11827321111768184,
41
+ "grad_norm": 1.020418802489519,
42
+ "learning_rate": 5e-06,
43
+ "loss": 0.7286,
44
+ "step": 50
45
+ },
46
+ {
47
+ "epoch": 0.1419278533412182,
48
+ "grad_norm": 0.962265802957487,
49
+ "learning_rate": 5e-06,
50
+ "loss": 0.7167,
51
+ "step": 60
52
+ },
53
+ {
54
+ "epoch": 0.16558249556475457,
55
+ "grad_norm": 0.7842406295728257,
56
+ "learning_rate": 5e-06,
57
+ "loss": 0.7122,
58
+ "step": 70
59
+ },
60
+ {
61
+ "epoch": 0.18923713778829096,
62
+ "grad_norm": 0.8009401339234955,
63
+ "learning_rate": 5e-06,
64
+ "loss": 0.7019,
65
+ "step": 80
66
+ },
67
+ {
68
+ "epoch": 0.21289178001182732,
69
+ "grad_norm": 0.7411954553666439,
70
+ "learning_rate": 5e-06,
71
+ "loss": 0.6865,
72
+ "step": 90
73
+ },
74
+ {
75
+ "epoch": 0.23654642223536368,
76
+ "grad_norm": 0.6604551640687066,
77
+ "learning_rate": 5e-06,
78
+ "loss": 0.6854,
79
+ "step": 100
80
+ },
81
+ {
82
+ "epoch": 0.26020106445890007,
83
+ "grad_norm": 0.5272790934190837,
84
+ "learning_rate": 5e-06,
85
+ "loss": 0.6904,
86
+ "step": 110
87
+ },
88
+ {
89
+ "epoch": 0.2838557066824364,
90
+ "grad_norm": 0.5812760583772804,
91
+ "learning_rate": 5e-06,
92
+ "loss": 0.6776,
93
+ "step": 120
94
+ },
95
+ {
96
+ "epoch": 0.3075103489059728,
97
+ "grad_norm": 0.5288883948662314,
98
+ "learning_rate": 5e-06,
99
+ "loss": 0.6738,
100
+ "step": 130
101
+ },
102
+ {
103
+ "epoch": 0.33116499112950915,
104
+ "grad_norm": 0.63694474679749,
105
+ "learning_rate": 5e-06,
106
+ "loss": 0.6822,
107
+ "step": 140
108
+ },
109
+ {
110
+ "epoch": 0.35481963335304556,
111
+ "grad_norm": 0.8869276352236491,
112
+ "learning_rate": 5e-06,
113
+ "loss": 0.6649,
114
+ "step": 150
115
+ },
116
+ {
117
+ "epoch": 0.3784742755765819,
118
+ "grad_norm": 0.518289407793539,
119
+ "learning_rate": 5e-06,
120
+ "loss": 0.6696,
121
+ "step": 160
122
+ },
123
+ {
124
+ "epoch": 0.4021289178001183,
125
+ "grad_norm": 0.5358648478939547,
126
+ "learning_rate": 5e-06,
127
+ "loss": 0.6671,
128
+ "step": 170
129
+ },
130
+ {
131
+ "epoch": 0.42578356002365464,
132
+ "grad_norm": 0.797540982889909,
133
+ "learning_rate": 5e-06,
134
+ "loss": 0.6743,
135
+ "step": 180
136
+ },
137
+ {
138
+ "epoch": 0.449438202247191,
139
+ "grad_norm": 0.4908769384958429,
140
+ "learning_rate": 5e-06,
141
+ "loss": 0.6704,
142
+ "step": 190
143
+ },
144
+ {
145
+ "epoch": 0.47309284447072736,
146
+ "grad_norm": 0.5763974723432601,
147
+ "learning_rate": 5e-06,
148
+ "loss": 0.6692,
149
+ "step": 200
150
+ },
151
+ {
152
+ "epoch": 0.4967474866942638,
153
+ "grad_norm": 0.9377426988511532,
154
+ "learning_rate": 5e-06,
155
+ "loss": 0.6696,
156
+ "step": 210
157
+ },
158
+ {
159
+ "epoch": 0.5204021289178001,
160
+ "grad_norm": 0.576249879822413,
161
+ "learning_rate": 5e-06,
162
+ "loss": 0.6688,
163
+ "step": 220
164
+ },
165
+ {
166
+ "epoch": 0.5440567711413364,
167
+ "grad_norm": 0.5017180552678443,
168
+ "learning_rate": 5e-06,
169
+ "loss": 0.6605,
170
+ "step": 230
171
+ },
172
+ {
173
+ "epoch": 0.5677114133648729,
174
+ "grad_norm": 0.46598378368467036,
175
+ "learning_rate": 5e-06,
176
+ "loss": 0.6649,
177
+ "step": 240
178
+ },
179
+ {
180
+ "epoch": 0.5913660555884093,
181
+ "grad_norm": 0.44951386010612127,
182
+ "learning_rate": 5e-06,
183
+ "loss": 0.6618,
184
+ "step": 250
185
+ },
186
+ {
187
+ "epoch": 0.6150206978119456,
188
+ "grad_norm": 0.6107309306047823,
189
+ "learning_rate": 5e-06,
190
+ "loss": 0.6525,
191
+ "step": 260
192
+ },
193
+ {
194
+ "epoch": 0.638675340035482,
195
+ "grad_norm": 0.6063385890447376,
196
+ "learning_rate": 5e-06,
197
+ "loss": 0.664,
198
+ "step": 270
199
+ },
200
+ {
201
+ "epoch": 0.6623299822590183,
202
+ "grad_norm": 0.62161938246522,
203
+ "learning_rate": 5e-06,
204
+ "loss": 0.6658,
205
+ "step": 280
206
+ },
207
+ {
208
+ "epoch": 0.6859846244825547,
209
+ "grad_norm": 0.48803716140240055,
210
+ "learning_rate": 5e-06,
211
+ "loss": 0.6599,
212
+ "step": 290
213
+ },
214
+ {
215
+ "epoch": 0.7096392667060911,
216
+ "grad_norm": 0.5643745069932464,
217
+ "learning_rate": 5e-06,
218
+ "loss": 0.6536,
219
+ "step": 300
220
+ },
221
+ {
222
+ "epoch": 0.7332939089296274,
223
+ "grad_norm": 0.45536574485759423,
224
+ "learning_rate": 5e-06,
225
+ "loss": 0.662,
226
+ "step": 310
227
+ },
228
+ {
229
+ "epoch": 0.7569485511531638,
230
+ "grad_norm": 0.5246164914927769,
231
+ "learning_rate": 5e-06,
232
+ "loss": 0.656,
233
+ "step": 320
234
+ },
235
+ {
236
+ "epoch": 0.7806031933767001,
237
+ "grad_norm": 0.5653234939153071,
238
+ "learning_rate": 5e-06,
239
+ "loss": 0.6541,
240
+ "step": 330
241
+ },
242
+ {
243
+ "epoch": 0.8042578356002366,
244
+ "grad_norm": 0.553750057338791,
245
+ "learning_rate": 5e-06,
246
+ "loss": 0.6539,
247
+ "step": 340
248
+ },
249
+ {
250
+ "epoch": 0.8279124778237729,
251
+ "grad_norm": 0.5823597490849275,
252
+ "learning_rate": 5e-06,
253
+ "loss": 0.6602,
254
+ "step": 350
255
+ },
256
+ {
257
+ "epoch": 0.8515671200473093,
258
+ "grad_norm": 0.432948815077171,
259
+ "learning_rate": 5e-06,
260
+ "loss": 0.6537,
261
+ "step": 360
262
+ },
263
+ {
264
+ "epoch": 0.8752217622708457,
265
+ "grad_norm": 0.44732587095055915,
266
+ "learning_rate": 5e-06,
267
+ "loss": 0.6571,
268
+ "step": 370
269
+ },
270
+ {
271
+ "epoch": 0.898876404494382,
272
+ "grad_norm": 0.45966183440739067,
273
+ "learning_rate": 5e-06,
274
+ "loss": 0.6478,
275
+ "step": 380
276
+ },
277
+ {
278
+ "epoch": 0.9225310467179184,
279
+ "grad_norm": 0.6457761210048674,
280
+ "learning_rate": 5e-06,
281
+ "loss": 0.6552,
282
+ "step": 390
283
+ },
284
+ {
285
+ "epoch": 0.9461856889414547,
286
+ "grad_norm": 0.4724156783274873,
287
+ "learning_rate": 5e-06,
288
+ "loss": 0.6467,
289
+ "step": 400
290
+ },
291
+ {
292
+ "epoch": 0.9698403311649911,
293
+ "grad_norm": 0.4559392408256043,
294
+ "learning_rate": 5e-06,
295
+ "loss": 0.6492,
296
+ "step": 410
297
+ },
298
+ {
299
+ "epoch": 0.9934949733885275,
300
+ "grad_norm": 0.5566904074437511,
301
+ "learning_rate": 5e-06,
302
+ "loss": 0.6483,
303
+ "step": 420
304
+ },
305
+ {
306
+ "epoch": 0.9982259018332348,
307
+ "eval_loss": 0.6504195332527161,
308
+ "eval_runtime": 226.3163,
309
+ "eval_samples_per_second": 50.323,
310
+ "eval_steps_per_second": 0.393,
311
+ "step": 422
312
+ },
313
+ {
314
+ "epoch": 1.0171496156120639,
315
+ "grad_norm": 0.5100430406130203,
316
+ "learning_rate": 5e-06,
317
+ "loss": 0.6374,
318
+ "step": 430
319
+ },
320
+ {
321
+ "epoch": 1.0408042578356003,
322
+ "grad_norm": 0.6182142994254941,
323
+ "learning_rate": 5e-06,
324
+ "loss": 0.6056,
325
+ "step": 440
326
+ },
327
+ {
328
+ "epoch": 1.0644589000591367,
329
+ "grad_norm": 0.612159821355247,
330
+ "learning_rate": 5e-06,
331
+ "loss": 0.6041,
332
+ "step": 450
333
+ },
334
+ {
335
+ "epoch": 1.0881135422826729,
336
+ "grad_norm": 0.620568436985301,
337
+ "learning_rate": 5e-06,
338
+ "loss": 0.6125,
339
+ "step": 460
340
+ },
341
+ {
342
+ "epoch": 1.1117681845062093,
343
+ "grad_norm": 0.48353189963783877,
344
+ "learning_rate": 5e-06,
345
+ "loss": 0.6114,
346
+ "step": 470
347
+ },
348
+ {
349
+ "epoch": 1.1354228267297457,
350
+ "grad_norm": 0.553405903828158,
351
+ "learning_rate": 5e-06,
352
+ "loss": 0.6114,
353
+ "step": 480
354
+ },
355
+ {
356
+ "epoch": 1.1590774689532821,
357
+ "grad_norm": 0.48890661916776557,
358
+ "learning_rate": 5e-06,
359
+ "loss": 0.6086,
360
+ "step": 490
361
+ },
362
+ {
363
+ "epoch": 1.1827321111768185,
364
+ "grad_norm": 0.4626708227952605,
365
+ "learning_rate": 5e-06,
366
+ "loss": 0.6048,
367
+ "step": 500
368
+ },
369
+ {
370
+ "epoch": 1.2063867534003547,
371
+ "grad_norm": 0.5562602420898135,
372
+ "learning_rate": 5e-06,
373
+ "loss": 0.6045,
374
+ "step": 510
375
+ },
376
+ {
377
+ "epoch": 1.2300413956238911,
378
+ "grad_norm": 0.5583615267258394,
379
+ "learning_rate": 5e-06,
380
+ "loss": 0.6106,
381
+ "step": 520
382
+ },
383
+ {
384
+ "epoch": 1.2536960378474276,
385
+ "grad_norm": 0.5120824375752575,
386
+ "learning_rate": 5e-06,
387
+ "loss": 0.6116,
388
+ "step": 530
389
+ },
390
+ {
391
+ "epoch": 1.277350680070964,
392
+ "grad_norm": 0.5477302994052297,
393
+ "learning_rate": 5e-06,
394
+ "loss": 0.6119,
395
+ "step": 540
396
+ },
397
+ {
398
+ "epoch": 1.3010053222945004,
399
+ "grad_norm": 0.6186813978530775,
400
+ "learning_rate": 5e-06,
401
+ "loss": 0.6102,
402
+ "step": 550
403
+ },
404
+ {
405
+ "epoch": 1.3246599645180366,
406
+ "grad_norm": 0.4482998491087788,
407
+ "learning_rate": 5e-06,
408
+ "loss": 0.6082,
409
+ "step": 560
410
+ },
411
+ {
412
+ "epoch": 1.348314606741573,
413
+ "grad_norm": 0.8072488882089637,
414
+ "learning_rate": 5e-06,
415
+ "loss": 0.6082,
416
+ "step": 570
417
+ },
418
+ {
419
+ "epoch": 1.3719692489651094,
420
+ "grad_norm": 0.48728798126789036,
421
+ "learning_rate": 5e-06,
422
+ "loss": 0.6118,
423
+ "step": 580
424
+ },
425
+ {
426
+ "epoch": 1.3956238911886458,
427
+ "grad_norm": 0.4366149357515769,
428
+ "learning_rate": 5e-06,
429
+ "loss": 0.6123,
430
+ "step": 590
431
+ },
432
+ {
433
+ "epoch": 1.4192785334121822,
434
+ "grad_norm": 0.5669347135329578,
435
+ "learning_rate": 5e-06,
436
+ "loss": 0.5995,
437
+ "step": 600
438
+ },
439
+ {
440
+ "epoch": 1.4429331756357184,
441
+ "grad_norm": 0.43980097172154653,
442
+ "learning_rate": 5e-06,
443
+ "loss": 0.6012,
444
+ "step": 610
445
+ },
446
+ {
447
+ "epoch": 1.4665878178592548,
448
+ "grad_norm": 0.4546181925945345,
449
+ "learning_rate": 5e-06,
450
+ "loss": 0.6039,
451
+ "step": 620
452
+ },
453
+ {
454
+ "epoch": 1.4902424600827913,
455
+ "grad_norm": 0.4439868785027596,
456
+ "learning_rate": 5e-06,
457
+ "loss": 0.6145,
458
+ "step": 630
459
+ },
460
+ {
461
+ "epoch": 1.5138971023063275,
462
+ "grad_norm": 0.4840991507057293,
463
+ "learning_rate": 5e-06,
464
+ "loss": 0.6128,
465
+ "step": 640
466
+ },
467
+ {
468
+ "epoch": 1.537551744529864,
469
+ "grad_norm": 0.5028857797195146,
470
+ "learning_rate": 5e-06,
471
+ "loss": 0.6037,
472
+ "step": 650
473
+ },
474
+ {
475
+ "epoch": 1.5612063867534003,
476
+ "grad_norm": 0.6295444122174948,
477
+ "learning_rate": 5e-06,
478
+ "loss": 0.6081,
479
+ "step": 660
480
+ },
481
+ {
482
+ "epoch": 1.5848610289769367,
483
+ "grad_norm": 0.5574140089154338,
484
+ "learning_rate": 5e-06,
485
+ "loss": 0.605,
486
+ "step": 670
487
+ },
488
+ {
489
+ "epoch": 1.6085156712004731,
490
+ "grad_norm": 0.5278087358672146,
491
+ "learning_rate": 5e-06,
492
+ "loss": 0.609,
493
+ "step": 680
494
+ },
495
+ {
496
+ "epoch": 1.6321703134240093,
497
+ "grad_norm": 0.5407251516890358,
498
+ "learning_rate": 5e-06,
499
+ "loss": 0.6023,
500
+ "step": 690
501
+ },
502
+ {
503
+ "epoch": 1.655824955647546,
504
+ "grad_norm": 0.47669538747864815,
505
+ "learning_rate": 5e-06,
506
+ "loss": 0.6098,
507
+ "step": 700
508
+ },
509
+ {
510
+ "epoch": 1.6794795978710821,
511
+ "grad_norm": 0.5117309364481941,
512
+ "learning_rate": 5e-06,
513
+ "loss": 0.6083,
514
+ "step": 710
515
+ },
516
+ {
517
+ "epoch": 1.7031342400946186,
518
+ "grad_norm": 0.4622792038975577,
519
+ "learning_rate": 5e-06,
520
+ "loss": 0.6145,
521
+ "step": 720
522
+ },
523
+ {
524
+ "epoch": 1.726788882318155,
525
+ "grad_norm": 0.476155429063029,
526
+ "learning_rate": 5e-06,
527
+ "loss": 0.6102,
528
+ "step": 730
529
+ },
530
+ {
531
+ "epoch": 1.7504435245416912,
532
+ "grad_norm": 0.6072162079153387,
533
+ "learning_rate": 5e-06,
534
+ "loss": 0.6047,
535
+ "step": 740
536
+ },
537
+ {
538
+ "epoch": 1.7740981667652278,
539
+ "grad_norm": 0.4580614369890106,
540
+ "learning_rate": 5e-06,
541
+ "loss": 0.6104,
542
+ "step": 750
543
+ },
544
+ {
545
+ "epoch": 1.797752808988764,
546
+ "grad_norm": 0.48163931069349186,
547
+ "learning_rate": 5e-06,
548
+ "loss": 0.6152,
549
+ "step": 760
550
+ },
551
+ {
552
+ "epoch": 1.8214074512123004,
553
+ "grad_norm": 0.44236726486886196,
554
+ "learning_rate": 5e-06,
555
+ "loss": 0.6159,
556
+ "step": 770
557
+ },
558
+ {
559
+ "epoch": 1.8450620934358368,
560
+ "grad_norm": 0.4140807715487973,
561
+ "learning_rate": 5e-06,
562
+ "loss": 0.6077,
563
+ "step": 780
564
+ },
565
+ {
566
+ "epoch": 1.868716735659373,
567
+ "grad_norm": 0.49794149670638754,
568
+ "learning_rate": 5e-06,
569
+ "loss": 0.6064,
570
+ "step": 790
571
+ },
572
+ {
573
+ "epoch": 1.8923713778829097,
574
+ "grad_norm": 0.48140566757242875,
575
+ "learning_rate": 5e-06,
576
+ "loss": 0.613,
577
+ "step": 800
578
+ },
579
+ {
580
+ "epoch": 1.9160260201064458,
581
+ "grad_norm": 0.4407179231154673,
582
+ "learning_rate": 5e-06,
583
+ "loss": 0.6126,
584
+ "step": 810
585
+ },
586
+ {
587
+ "epoch": 1.9396806623299823,
588
+ "grad_norm": 0.41696082412030117,
589
+ "learning_rate": 5e-06,
590
+ "loss": 0.5994,
591
+ "step": 820
592
+ },
593
+ {
594
+ "epoch": 1.9633353045535187,
595
+ "grad_norm": 0.47981289616269807,
596
+ "learning_rate": 5e-06,
597
+ "loss": 0.612,
598
+ "step": 830
599
+ },
600
+ {
601
+ "epoch": 1.9869899467770549,
602
+ "grad_norm": 0.4484580880097136,
603
+ "learning_rate": 5e-06,
604
+ "loss": 0.5968,
605
+ "step": 840
606
+ },
607
+ {
608
+ "epoch": 1.9988172678888232,
609
+ "eval_loss": 0.6400034427642822,
610
+ "eval_runtime": 225.8288,
611
+ "eval_samples_per_second": 50.432,
612
+ "eval_steps_per_second": 0.394,
613
+ "step": 845
614
+ },
615
+ {
616
+ "epoch": 2.0106445890005915,
617
+ "grad_norm": 0.6332180702757985,
618
+ "learning_rate": 5e-06,
619
+ "loss": 0.5995,
620
+ "step": 850
621
+ },
622
+ {
623
+ "epoch": 2.0342992312241277,
624
+ "grad_norm": 0.5223111355986098,
625
+ "learning_rate": 5e-06,
626
+ "loss": 0.5669,
627
+ "step": 860
628
+ },
629
+ {
630
+ "epoch": 2.057953873447664,
631
+ "grad_norm": 0.5560692373029835,
632
+ "learning_rate": 5e-06,
633
+ "loss": 0.5668,
634
+ "step": 870
635
+ },
636
+ {
637
+ "epoch": 2.0816085156712005,
638
+ "grad_norm": 0.5874280616479215,
639
+ "learning_rate": 5e-06,
640
+ "loss": 0.559,
641
+ "step": 880
642
+ },
643
+ {
644
+ "epoch": 2.1052631578947367,
645
+ "grad_norm": 0.5381432115880647,
646
+ "learning_rate": 5e-06,
647
+ "loss": 0.5615,
648
+ "step": 890
649
+ },
650
+ {
651
+ "epoch": 2.1289178001182734,
652
+ "grad_norm": 0.5340607102638687,
653
+ "learning_rate": 5e-06,
654
+ "loss": 0.567,
655
+ "step": 900
656
+ },
657
+ {
658
+ "epoch": 2.1525724423418096,
659
+ "grad_norm": 0.4709298987555601,
660
+ "learning_rate": 5e-06,
661
+ "loss": 0.5616,
662
+ "step": 910
663
+ },
664
+ {
665
+ "epoch": 2.1762270845653457,
666
+ "grad_norm": 0.47531624539694045,
667
+ "learning_rate": 5e-06,
668
+ "loss": 0.5634,
669
+ "step": 920
670
+ },
671
+ {
672
+ "epoch": 2.1998817267888824,
673
+ "grad_norm": 0.5690095865272987,
674
+ "learning_rate": 5e-06,
675
+ "loss": 0.5627,
676
+ "step": 930
677
+ },
678
+ {
679
+ "epoch": 2.2235363690124186,
680
+ "grad_norm": 0.6002806476085656,
681
+ "learning_rate": 5e-06,
682
+ "loss": 0.5643,
683
+ "step": 940
684
+ },
685
+ {
686
+ "epoch": 2.247191011235955,
687
+ "grad_norm": 0.5868800123289198,
688
+ "learning_rate": 5e-06,
689
+ "loss": 0.5702,
690
+ "step": 950
691
+ },
692
+ {
693
+ "epoch": 2.2708456534594914,
694
+ "grad_norm": 0.545743902564753,
695
+ "learning_rate": 5e-06,
696
+ "loss": 0.564,
697
+ "step": 960
698
+ },
699
+ {
700
+ "epoch": 2.2945002956830276,
701
+ "grad_norm": 0.5069047371924982,
702
+ "learning_rate": 5e-06,
703
+ "loss": 0.5657,
704
+ "step": 970
705
+ },
706
+ {
707
+ "epoch": 2.3181549379065642,
708
+ "grad_norm": 0.5832343780575133,
709
+ "learning_rate": 5e-06,
710
+ "loss": 0.5696,
711
+ "step": 980
712
+ },
713
+ {
714
+ "epoch": 2.3418095801301004,
715
+ "grad_norm": 0.49163574506732044,
716
+ "learning_rate": 5e-06,
717
+ "loss": 0.5673,
718
+ "step": 990
719
+ },
720
+ {
721
+ "epoch": 2.365464222353637,
722
+ "grad_norm": 0.5769534565432972,
723
+ "learning_rate": 5e-06,
724
+ "loss": 0.5674,
725
+ "step": 1000
726
+ },
727
+ {
728
+ "epoch": 2.3891188645771733,
729
+ "grad_norm": 0.4748456336186354,
730
+ "learning_rate": 5e-06,
731
+ "loss": 0.5656,
732
+ "step": 1010
733
+ },
734
+ {
735
+ "epoch": 2.4127735068007095,
736
+ "grad_norm": 0.5571397878943326,
737
+ "learning_rate": 5e-06,
738
+ "loss": 0.5632,
739
+ "step": 1020
740
+ },
741
+ {
742
+ "epoch": 2.436428149024246,
743
+ "grad_norm": 0.4977538685424081,
744
+ "learning_rate": 5e-06,
745
+ "loss": 0.5662,
746
+ "step": 1030
747
+ },
748
+ {
749
+ "epoch": 2.4600827912477823,
750
+ "grad_norm": 0.6212689514507728,
751
+ "learning_rate": 5e-06,
752
+ "loss": 0.5669,
753
+ "step": 1040
754
+ },
755
+ {
756
+ "epoch": 2.483737433471319,
757
+ "grad_norm": 0.5004435631831575,
758
+ "learning_rate": 5e-06,
759
+ "loss": 0.5694,
760
+ "step": 1050
761
+ },
762
+ {
763
+ "epoch": 2.507392075694855,
764
+ "grad_norm": 0.5290951851110267,
765
+ "learning_rate": 5e-06,
766
+ "loss": 0.5686,
767
+ "step": 1060
768
+ },
769
+ {
770
+ "epoch": 2.5310467179183913,
771
+ "grad_norm": 0.48790551419012185,
772
+ "learning_rate": 5e-06,
773
+ "loss": 0.5709,
774
+ "step": 1070
775
+ },
776
+ {
777
+ "epoch": 2.554701360141928,
778
+ "grad_norm": 0.5234534909018554,
779
+ "learning_rate": 5e-06,
780
+ "loss": 0.5718,
781
+ "step": 1080
782
+ },
783
+ {
784
+ "epoch": 2.578356002365464,
785
+ "grad_norm": 0.4907774872609342,
786
+ "learning_rate": 5e-06,
787
+ "loss": 0.5661,
788
+ "step": 1090
789
+ },
790
+ {
791
+ "epoch": 2.6020106445890008,
792
+ "grad_norm": 0.44774060132900434,
793
+ "learning_rate": 5e-06,
794
+ "loss": 0.5681,
795
+ "step": 1100
796
+ },
797
+ {
798
+ "epoch": 2.625665286812537,
799
+ "grad_norm": 0.5280954839046109,
800
+ "learning_rate": 5e-06,
801
+ "loss": 0.566,
802
+ "step": 1110
803
+ },
804
+ {
805
+ "epoch": 2.649319929036073,
806
+ "grad_norm": 0.6085042539902774,
807
+ "learning_rate": 5e-06,
808
+ "loss": 0.5773,
809
+ "step": 1120
810
+ },
811
+ {
812
+ "epoch": 2.67297457125961,
813
+ "grad_norm": 0.5793916767265476,
814
+ "learning_rate": 5e-06,
815
+ "loss": 0.5646,
816
+ "step": 1130
817
+ },
818
+ {
819
+ "epoch": 2.696629213483146,
820
+ "grad_norm": 0.452089081498202,
821
+ "learning_rate": 5e-06,
822
+ "loss": 0.5653,
823
+ "step": 1140
824
+ },
825
+ {
826
+ "epoch": 2.7202838557066826,
827
+ "grad_norm": 0.548967392597449,
828
+ "learning_rate": 5e-06,
829
+ "loss": 0.5689,
830
+ "step": 1150
831
+ },
832
+ {
833
+ "epoch": 2.743938497930219,
834
+ "grad_norm": 0.5154350449304891,
835
+ "learning_rate": 5e-06,
836
+ "loss": 0.5751,
837
+ "step": 1160
838
+ },
839
+ {
840
+ "epoch": 2.767593140153755,
841
+ "grad_norm": 0.5408130206005414,
842
+ "learning_rate": 5e-06,
843
+ "loss": 0.5659,
844
+ "step": 1170
845
+ },
846
+ {
847
+ "epoch": 2.7912477823772917,
848
+ "grad_norm": 0.448067927236881,
849
+ "learning_rate": 5e-06,
850
+ "loss": 0.5696,
851
+ "step": 1180
852
+ },
853
+ {
854
+ "epoch": 2.814902424600828,
855
+ "grad_norm": 0.46851404269934493,
856
+ "learning_rate": 5e-06,
857
+ "loss": 0.5692,
858
+ "step": 1190
859
+ },
860
+ {
861
+ "epoch": 2.8385570668243645,
862
+ "grad_norm": 0.4742832176992908,
863
+ "learning_rate": 5e-06,
864
+ "loss": 0.5676,
865
+ "step": 1200
866
+ },
867
+ {
868
+ "epoch": 2.8622117090479007,
869
+ "grad_norm": 0.5257873035609077,
870
+ "learning_rate": 5e-06,
871
+ "loss": 0.569,
872
+ "step": 1210
873
+ },
874
+ {
875
+ "epoch": 2.885866351271437,
876
+ "grad_norm": 0.4515773409971667,
877
+ "learning_rate": 5e-06,
878
+ "loss": 0.563,
879
+ "step": 1220
880
+ },
881
+ {
882
+ "epoch": 2.9095209934949735,
883
+ "grad_norm": 0.4991731794016107,
884
+ "learning_rate": 5e-06,
885
+ "loss": 0.5695,
886
+ "step": 1230
887
+ },
888
+ {
889
+ "epoch": 2.9331756357185097,
890
+ "grad_norm": 0.5065264471878053,
891
+ "learning_rate": 5e-06,
892
+ "loss": 0.5706,
893
+ "step": 1240
894
+ },
895
+ {
896
+ "epoch": 2.9568302779420463,
897
+ "grad_norm": 0.44403389629974643,
898
+ "learning_rate": 5e-06,
899
+ "loss": 0.5786,
900
+ "step": 1250
901
+ },
902
+ {
903
+ "epoch": 2.9804849201655825,
904
+ "grad_norm": 0.5042460817434247,
905
+ "learning_rate": 5e-06,
906
+ "loss": 0.572,
907
+ "step": 1260
908
+ },
909
+ {
910
+ "epoch": 2.9946777054997042,
911
+ "eval_loss": 0.6413525938987732,
912
+ "eval_runtime": 226.2058,
913
+ "eval_samples_per_second": 50.348,
914
+ "eval_steps_per_second": 0.393,
915
+ "step": 1266
916
+ },
917
+ {
918
+ "epoch": 2.9946777054997042,
919
+ "step": 1266,
920
+ "total_flos": 2120178393415680.0,
921
+ "train_loss": 0.6193434227887676,
922
+ "train_runtime": 37811.5197,
923
+ "train_samples_per_second": 17.167,
924
+ "train_steps_per_second": 0.033
925
+ }
926
+ ],
927
+ "logging_steps": 10,
928
+ "max_steps": 1266,
929
+ "num_input_tokens_seen": 0,
930
+ "num_train_epochs": 3,
931
+ "save_steps": 500,
932
+ "stateful_callbacks": {
933
+ "TrainerControl": {
934
+ "args": {
935
+ "should_epoch_stop": false,
936
+ "should_evaluate": false,
937
+ "should_log": false,
938
+ "should_save": true,
939
+ "should_training_stop": true
940
+ },
941
+ "attributes": {}
942
+ }
943
+ },
944
+ "total_flos": 2120178393415680.0,
945
+ "train_batch_size": 8,
946
+ "trial_name": null,
947
+ "trial_params": null
948
+ }
training_eval_loss.png ADDED
training_loss.png ADDED