sedrickkeh commited on
Commit
af60a5c
·
verified ·
1 Parent(s): 7ecdb5d

End of training

Browse files
README.md CHANGED
@@ -4,6 +4,7 @@ license: llama3.1
4
  base_model: meta-llama/Meta-Llama-3.1-8B
5
  tags:
6
  - llama-factory
 
7
  - generated_from_trainer
8
  model-index:
9
  - name: oh-dcft-v3.1-gemini-1.5-pro
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
15
 
16
  # oh-dcft-v3.1-gemini-1.5-pro
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: 0.6309
21
 
 
4
  base_model: meta-llama/Meta-Llama-3.1-8B
5
  tags:
6
  - llama-factory
7
+ - full
8
  - generated_from_trainer
9
  model-index:
10
  - name: oh-dcft-v3.1-gemini-1.5-pro
 
16
 
17
  # oh-dcft-v3.1-gemini-1.5-pro
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-gemini-1.5-pro dataset.
20
  It achieves the following results on the evaluation set:
21
  - Loss: 0.6309
22
 
all_results.json ADDED
@@ -0,0 +1,12 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.9957446808510637,
3
+ "eval_loss": 0.6309003829956055,
4
+ "eval_runtime": 299.4868,
5
+ "eval_samples_per_second": 39.648,
6
+ "eval_steps_per_second": 0.621,
7
+ "total_flos": 2210839784325120.0,
8
+ "train_loss": 0.6033726125052481,
9
+ "train_runtime": 44159.109,
10
+ "train_samples_per_second": 15.326,
11
+ "train_steps_per_second": 0.03
12
+ }
eval_results.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.9957446808510637,
3
+ "eval_loss": 0.6309003829956055,
4
+ "eval_runtime": 299.4868,
5
+ "eval_samples_per_second": 39.648,
6
+ "eval_steps_per_second": 0.621
7
+ }
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.9957446808510637,
3
+ "total_flos": 2210839784325120.0,
4
+ "train_loss": 0.6033726125052481,
5
+ "train_runtime": 44159.109,
6
+ "train_samples_per_second": 15.326,
7
+ "train_steps_per_second": 0.03
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,990 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 2.9957446808510637,
5
+ "eval_steps": 500,
6
+ "global_step": 1320,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.02269503546099291,
13
+ "grad_norm": 2.4400690431145335,
14
+ "learning_rate": 5e-06,
15
+ "loss": 0.9537,
16
+ "step": 10
17
+ },
18
+ {
19
+ "epoch": 0.04539007092198582,
20
+ "grad_norm": 1.0371553805334768,
21
+ "learning_rate": 5e-06,
22
+ "loss": 0.8304,
23
+ "step": 20
24
+ },
25
+ {
26
+ "epoch": 0.06808510638297872,
27
+ "grad_norm": 0.8207371359291633,
28
+ "learning_rate": 5e-06,
29
+ "loss": 0.7801,
30
+ "step": 30
31
+ },
32
+ {
33
+ "epoch": 0.09078014184397164,
34
+ "grad_norm": 0.7210586495004421,
35
+ "learning_rate": 5e-06,
36
+ "loss": 0.7522,
37
+ "step": 40
38
+ },
39
+ {
40
+ "epoch": 0.11347517730496454,
41
+ "grad_norm": 0.815177982428463,
42
+ "learning_rate": 5e-06,
43
+ "loss": 0.739,
44
+ "step": 50
45
+ },
46
+ {
47
+ "epoch": 0.13617021276595745,
48
+ "grad_norm": 2.1607006669022053,
49
+ "learning_rate": 5e-06,
50
+ "loss": 0.731,
51
+ "step": 60
52
+ },
53
+ {
54
+ "epoch": 0.15886524822695036,
55
+ "grad_norm": 0.8260032479277283,
56
+ "learning_rate": 5e-06,
57
+ "loss": 0.7245,
58
+ "step": 70
59
+ },
60
+ {
61
+ "epoch": 0.18156028368794327,
62
+ "grad_norm": 0.6043557996158753,
63
+ "learning_rate": 5e-06,
64
+ "loss": 0.709,
65
+ "step": 80
66
+ },
67
+ {
68
+ "epoch": 0.20425531914893616,
69
+ "grad_norm": 0.6817812588474443,
70
+ "learning_rate": 5e-06,
71
+ "loss": 0.7065,
72
+ "step": 90
73
+ },
74
+ {
75
+ "epoch": 0.22695035460992907,
76
+ "grad_norm": 0.6633270827705589,
77
+ "learning_rate": 5e-06,
78
+ "loss": 0.7042,
79
+ "step": 100
80
+ },
81
+ {
82
+ "epoch": 0.24964539007092199,
83
+ "grad_norm": 0.8788640026205816,
84
+ "learning_rate": 5e-06,
85
+ "loss": 0.6937,
86
+ "step": 110
87
+ },
88
+ {
89
+ "epoch": 0.2723404255319149,
90
+ "grad_norm": 0.6318189161156693,
91
+ "learning_rate": 5e-06,
92
+ "loss": 0.6875,
93
+ "step": 120
94
+ },
95
+ {
96
+ "epoch": 0.2950354609929078,
97
+ "grad_norm": 0.6116775844736505,
98
+ "learning_rate": 5e-06,
99
+ "loss": 0.6882,
100
+ "step": 130
101
+ },
102
+ {
103
+ "epoch": 0.3177304964539007,
104
+ "grad_norm": 0.5442399734240484,
105
+ "learning_rate": 5e-06,
106
+ "loss": 0.6829,
107
+ "step": 140
108
+ },
109
+ {
110
+ "epoch": 0.3404255319148936,
111
+ "grad_norm": 0.772917712294573,
112
+ "learning_rate": 5e-06,
113
+ "loss": 0.6778,
114
+ "step": 150
115
+ },
116
+ {
117
+ "epoch": 0.36312056737588655,
118
+ "grad_norm": 0.5042750571740913,
119
+ "learning_rate": 5e-06,
120
+ "loss": 0.6826,
121
+ "step": 160
122
+ },
123
+ {
124
+ "epoch": 0.38581560283687943,
125
+ "grad_norm": 0.5272476451832978,
126
+ "learning_rate": 5e-06,
127
+ "loss": 0.6791,
128
+ "step": 170
129
+ },
130
+ {
131
+ "epoch": 0.4085106382978723,
132
+ "grad_norm": 0.6477667523961554,
133
+ "learning_rate": 5e-06,
134
+ "loss": 0.6778,
135
+ "step": 180
136
+ },
137
+ {
138
+ "epoch": 0.43120567375886526,
139
+ "grad_norm": 0.6825001542564298,
140
+ "learning_rate": 5e-06,
141
+ "loss": 0.6727,
142
+ "step": 190
143
+ },
144
+ {
145
+ "epoch": 0.45390070921985815,
146
+ "grad_norm": 0.7223077265284692,
147
+ "learning_rate": 5e-06,
148
+ "loss": 0.6766,
149
+ "step": 200
150
+ },
151
+ {
152
+ "epoch": 0.4765957446808511,
153
+ "grad_norm": 0.7610671419515379,
154
+ "learning_rate": 5e-06,
155
+ "loss": 0.6754,
156
+ "step": 210
157
+ },
158
+ {
159
+ "epoch": 0.49929078014184397,
160
+ "grad_norm": 0.5918397478370985,
161
+ "learning_rate": 5e-06,
162
+ "loss": 0.6707,
163
+ "step": 220
164
+ },
165
+ {
166
+ "epoch": 0.5219858156028369,
167
+ "grad_norm": 0.5937553589475805,
168
+ "learning_rate": 5e-06,
169
+ "loss": 0.6695,
170
+ "step": 230
171
+ },
172
+ {
173
+ "epoch": 0.5446808510638298,
174
+ "grad_norm": 0.6308694971546095,
175
+ "learning_rate": 5e-06,
176
+ "loss": 0.6706,
177
+ "step": 240
178
+ },
179
+ {
180
+ "epoch": 0.5673758865248227,
181
+ "grad_norm": 0.5145635067713508,
182
+ "learning_rate": 5e-06,
183
+ "loss": 0.6638,
184
+ "step": 250
185
+ },
186
+ {
187
+ "epoch": 0.5900709219858156,
188
+ "grad_norm": 0.8960582295099536,
189
+ "learning_rate": 5e-06,
190
+ "loss": 0.659,
191
+ "step": 260
192
+ },
193
+ {
194
+ "epoch": 0.6127659574468085,
195
+ "grad_norm": 0.5714201206969344,
196
+ "learning_rate": 5e-06,
197
+ "loss": 0.6648,
198
+ "step": 270
199
+ },
200
+ {
201
+ "epoch": 0.6354609929078014,
202
+ "grad_norm": 0.5218865972600176,
203
+ "learning_rate": 5e-06,
204
+ "loss": 0.6692,
205
+ "step": 280
206
+ },
207
+ {
208
+ "epoch": 0.6581560283687943,
209
+ "grad_norm": 0.908425551953448,
210
+ "learning_rate": 5e-06,
211
+ "loss": 0.6603,
212
+ "step": 290
213
+ },
214
+ {
215
+ "epoch": 0.6808510638297872,
216
+ "grad_norm": 0.5937599038511239,
217
+ "learning_rate": 5e-06,
218
+ "loss": 0.6616,
219
+ "step": 300
220
+ },
221
+ {
222
+ "epoch": 0.7035460992907802,
223
+ "grad_norm": 0.5791492049959488,
224
+ "learning_rate": 5e-06,
225
+ "loss": 0.6573,
226
+ "step": 310
227
+ },
228
+ {
229
+ "epoch": 0.7262411347517731,
230
+ "grad_norm": 0.5544912073084471,
231
+ "learning_rate": 5e-06,
232
+ "loss": 0.6561,
233
+ "step": 320
234
+ },
235
+ {
236
+ "epoch": 0.7489361702127659,
237
+ "grad_norm": 0.6161595885325214,
238
+ "learning_rate": 5e-06,
239
+ "loss": 0.6535,
240
+ "step": 330
241
+ },
242
+ {
243
+ "epoch": 0.7716312056737589,
244
+ "grad_norm": 0.5292955751782062,
245
+ "learning_rate": 5e-06,
246
+ "loss": 0.6575,
247
+ "step": 340
248
+ },
249
+ {
250
+ "epoch": 0.7943262411347518,
251
+ "grad_norm": 0.5125989549019483,
252
+ "learning_rate": 5e-06,
253
+ "loss": 0.6508,
254
+ "step": 350
255
+ },
256
+ {
257
+ "epoch": 0.8170212765957446,
258
+ "grad_norm": 0.4999207740358057,
259
+ "learning_rate": 5e-06,
260
+ "loss": 0.652,
261
+ "step": 360
262
+ },
263
+ {
264
+ "epoch": 0.8397163120567376,
265
+ "grad_norm": 0.5737129637592092,
266
+ "learning_rate": 5e-06,
267
+ "loss": 0.6487,
268
+ "step": 370
269
+ },
270
+ {
271
+ "epoch": 0.8624113475177305,
272
+ "grad_norm": 0.697556121614893,
273
+ "learning_rate": 5e-06,
274
+ "loss": 0.6537,
275
+ "step": 380
276
+ },
277
+ {
278
+ "epoch": 0.8851063829787233,
279
+ "grad_norm": 0.8803924578376542,
280
+ "learning_rate": 5e-06,
281
+ "loss": 0.6528,
282
+ "step": 390
283
+ },
284
+ {
285
+ "epoch": 0.9078014184397163,
286
+ "grad_norm": 0.7209729745886211,
287
+ "learning_rate": 5e-06,
288
+ "loss": 0.6499,
289
+ "step": 400
290
+ },
291
+ {
292
+ "epoch": 0.9304964539007092,
293
+ "grad_norm": 0.631225301000223,
294
+ "learning_rate": 5e-06,
295
+ "loss": 0.6493,
296
+ "step": 410
297
+ },
298
+ {
299
+ "epoch": 0.9531914893617022,
300
+ "grad_norm": 0.5302593707285189,
301
+ "learning_rate": 5e-06,
302
+ "loss": 0.651,
303
+ "step": 420
304
+ },
305
+ {
306
+ "epoch": 0.975886524822695,
307
+ "grad_norm": 0.5216458807768983,
308
+ "learning_rate": 5e-06,
309
+ "loss": 0.6438,
310
+ "step": 430
311
+ },
312
+ {
313
+ "epoch": 0.9985815602836879,
314
+ "grad_norm": 0.5024039028278129,
315
+ "learning_rate": 5e-06,
316
+ "loss": 0.6478,
317
+ "step": 440
318
+ },
319
+ {
320
+ "epoch": 0.9985815602836879,
321
+ "eval_loss": 0.64774489402771,
322
+ "eval_runtime": 312.1779,
323
+ "eval_samples_per_second": 38.036,
324
+ "eval_steps_per_second": 0.596,
325
+ "step": 440
326
+ },
327
+ {
328
+ "epoch": 1.0212765957446808,
329
+ "grad_norm": 0.6519926768853314,
330
+ "learning_rate": 5e-06,
331
+ "loss": 0.6161,
332
+ "step": 450
333
+ },
334
+ {
335
+ "epoch": 1.0439716312056737,
336
+ "grad_norm": 0.575055735064131,
337
+ "learning_rate": 5e-06,
338
+ "loss": 0.589,
339
+ "step": 460
340
+ },
341
+ {
342
+ "epoch": 1.0666666666666667,
343
+ "grad_norm": 0.9082080666134971,
344
+ "learning_rate": 5e-06,
345
+ "loss": 0.5912,
346
+ "step": 470
347
+ },
348
+ {
349
+ "epoch": 1.0893617021276596,
350
+ "grad_norm": 0.583715919536177,
351
+ "learning_rate": 5e-06,
352
+ "loss": 0.5913,
353
+ "step": 480
354
+ },
355
+ {
356
+ "epoch": 1.1120567375886525,
357
+ "grad_norm": 0.5975048919359871,
358
+ "learning_rate": 5e-06,
359
+ "loss": 0.5976,
360
+ "step": 490
361
+ },
362
+ {
363
+ "epoch": 1.1347517730496455,
364
+ "grad_norm": 0.5418419185217853,
365
+ "learning_rate": 5e-06,
366
+ "loss": 0.5931,
367
+ "step": 500
368
+ },
369
+ {
370
+ "epoch": 1.1574468085106382,
371
+ "grad_norm": 0.638492552707852,
372
+ "learning_rate": 5e-06,
373
+ "loss": 0.5913,
374
+ "step": 510
375
+ },
376
+ {
377
+ "epoch": 1.1801418439716311,
378
+ "grad_norm": 0.5450441529585099,
379
+ "learning_rate": 5e-06,
380
+ "loss": 0.5888,
381
+ "step": 520
382
+ },
383
+ {
384
+ "epoch": 1.202836879432624,
385
+ "grad_norm": 0.667596474347774,
386
+ "learning_rate": 5e-06,
387
+ "loss": 0.5926,
388
+ "step": 530
389
+ },
390
+ {
391
+ "epoch": 1.225531914893617,
392
+ "grad_norm": 0.5901671600943978,
393
+ "learning_rate": 5e-06,
394
+ "loss": 0.5905,
395
+ "step": 540
396
+ },
397
+ {
398
+ "epoch": 1.24822695035461,
399
+ "grad_norm": 0.5680819366472465,
400
+ "learning_rate": 5e-06,
401
+ "loss": 0.5915,
402
+ "step": 550
403
+ },
404
+ {
405
+ "epoch": 1.270921985815603,
406
+ "grad_norm": 0.7905201905390314,
407
+ "learning_rate": 5e-06,
408
+ "loss": 0.5948,
409
+ "step": 560
410
+ },
411
+ {
412
+ "epoch": 1.2936170212765958,
413
+ "grad_norm": 0.5581341122325792,
414
+ "learning_rate": 5e-06,
415
+ "loss": 0.5904,
416
+ "step": 570
417
+ },
418
+ {
419
+ "epoch": 1.3163120567375888,
420
+ "grad_norm": 0.5968971489169257,
421
+ "learning_rate": 5e-06,
422
+ "loss": 0.5927,
423
+ "step": 580
424
+ },
425
+ {
426
+ "epoch": 1.3390070921985815,
427
+ "grad_norm": 0.6127986193776547,
428
+ "learning_rate": 5e-06,
429
+ "loss": 0.5889,
430
+ "step": 590
431
+ },
432
+ {
433
+ "epoch": 1.3617021276595744,
434
+ "grad_norm": 0.5540746342979398,
435
+ "learning_rate": 5e-06,
436
+ "loss": 0.5998,
437
+ "step": 600
438
+ },
439
+ {
440
+ "epoch": 1.3843971631205674,
441
+ "grad_norm": 0.6578848756357453,
442
+ "learning_rate": 5e-06,
443
+ "loss": 0.5933,
444
+ "step": 610
445
+ },
446
+ {
447
+ "epoch": 1.4070921985815603,
448
+ "grad_norm": 0.752983431690844,
449
+ "learning_rate": 5e-06,
450
+ "loss": 0.5967,
451
+ "step": 620
452
+ },
453
+ {
454
+ "epoch": 1.4297872340425533,
455
+ "grad_norm": 0.6009296481522326,
456
+ "learning_rate": 5e-06,
457
+ "loss": 0.5886,
458
+ "step": 630
459
+ },
460
+ {
461
+ "epoch": 1.452482269503546,
462
+ "grad_norm": 0.6855926736224828,
463
+ "learning_rate": 5e-06,
464
+ "loss": 0.5931,
465
+ "step": 640
466
+ },
467
+ {
468
+ "epoch": 1.475177304964539,
469
+ "grad_norm": 0.5390133364015494,
470
+ "learning_rate": 5e-06,
471
+ "loss": 0.5926,
472
+ "step": 650
473
+ },
474
+ {
475
+ "epoch": 1.4978723404255319,
476
+ "grad_norm": 0.6338310115530009,
477
+ "learning_rate": 5e-06,
478
+ "loss": 0.5969,
479
+ "step": 660
480
+ },
481
+ {
482
+ "epoch": 1.5205673758865248,
483
+ "grad_norm": 0.5298069642997515,
484
+ "learning_rate": 5e-06,
485
+ "loss": 0.5868,
486
+ "step": 670
487
+ },
488
+ {
489
+ "epoch": 1.5432624113475177,
490
+ "grad_norm": 0.5254529387411777,
491
+ "learning_rate": 5e-06,
492
+ "loss": 0.5893,
493
+ "step": 680
494
+ },
495
+ {
496
+ "epoch": 1.5659574468085107,
497
+ "grad_norm": 0.5090891733108535,
498
+ "learning_rate": 5e-06,
499
+ "loss": 0.5937,
500
+ "step": 690
501
+ },
502
+ {
503
+ "epoch": 1.5886524822695036,
504
+ "grad_norm": 0.6276632546884366,
505
+ "learning_rate": 5e-06,
506
+ "loss": 0.5856,
507
+ "step": 700
508
+ },
509
+ {
510
+ "epoch": 1.6113475177304966,
511
+ "grad_norm": 0.5618088874975952,
512
+ "learning_rate": 5e-06,
513
+ "loss": 0.5796,
514
+ "step": 710
515
+ },
516
+ {
517
+ "epoch": 1.6340425531914895,
518
+ "grad_norm": 0.5451154576203426,
519
+ "learning_rate": 5e-06,
520
+ "loss": 0.584,
521
+ "step": 720
522
+ },
523
+ {
524
+ "epoch": 1.6567375886524822,
525
+ "grad_norm": 0.5751145914968212,
526
+ "learning_rate": 5e-06,
527
+ "loss": 0.5882,
528
+ "step": 730
529
+ },
530
+ {
531
+ "epoch": 1.6794326241134752,
532
+ "grad_norm": 0.5457326197852193,
533
+ "learning_rate": 5e-06,
534
+ "loss": 0.5893,
535
+ "step": 740
536
+ },
537
+ {
538
+ "epoch": 1.702127659574468,
539
+ "grad_norm": 0.5538870750639828,
540
+ "learning_rate": 5e-06,
541
+ "loss": 0.5895,
542
+ "step": 750
543
+ },
544
+ {
545
+ "epoch": 1.724822695035461,
546
+ "grad_norm": 0.6482629759445236,
547
+ "learning_rate": 5e-06,
548
+ "loss": 0.5883,
549
+ "step": 760
550
+ },
551
+ {
552
+ "epoch": 1.7475177304964538,
553
+ "grad_norm": 0.604796917686035,
554
+ "learning_rate": 5e-06,
555
+ "loss": 0.5908,
556
+ "step": 770
557
+ },
558
+ {
559
+ "epoch": 1.7702127659574467,
560
+ "grad_norm": 0.5570179578262068,
561
+ "learning_rate": 5e-06,
562
+ "loss": 0.5935,
563
+ "step": 780
564
+ },
565
+ {
566
+ "epoch": 1.7929078014184396,
567
+ "grad_norm": 0.5359469482507023,
568
+ "learning_rate": 5e-06,
569
+ "loss": 0.5833,
570
+ "step": 790
571
+ },
572
+ {
573
+ "epoch": 1.8156028368794326,
574
+ "grad_norm": 0.6340496302965002,
575
+ "learning_rate": 5e-06,
576
+ "loss": 0.5853,
577
+ "step": 800
578
+ },
579
+ {
580
+ "epoch": 1.8382978723404255,
581
+ "grad_norm": 0.5742298835017674,
582
+ "learning_rate": 5e-06,
583
+ "loss": 0.5823,
584
+ "step": 810
585
+ },
586
+ {
587
+ "epoch": 1.8609929078014185,
588
+ "grad_norm": 0.5827541219871901,
589
+ "learning_rate": 5e-06,
590
+ "loss": 0.5903,
591
+ "step": 820
592
+ },
593
+ {
594
+ "epoch": 1.8836879432624114,
595
+ "grad_norm": 0.528321132004614,
596
+ "learning_rate": 5e-06,
597
+ "loss": 0.5832,
598
+ "step": 830
599
+ },
600
+ {
601
+ "epoch": 1.9063829787234043,
602
+ "grad_norm": 0.565101788703942,
603
+ "learning_rate": 5e-06,
604
+ "loss": 0.5855,
605
+ "step": 840
606
+ },
607
+ {
608
+ "epoch": 1.9290780141843973,
609
+ "grad_norm": 0.601792497070637,
610
+ "learning_rate": 5e-06,
611
+ "loss": 0.5842,
612
+ "step": 850
613
+ },
614
+ {
615
+ "epoch": 1.9517730496453902,
616
+ "grad_norm": 0.5088677439673144,
617
+ "learning_rate": 5e-06,
618
+ "loss": 0.5821,
619
+ "step": 860
620
+ },
621
+ {
622
+ "epoch": 1.974468085106383,
623
+ "grad_norm": 0.6353626361966195,
624
+ "learning_rate": 5e-06,
625
+ "loss": 0.5865,
626
+ "step": 870
627
+ },
628
+ {
629
+ "epoch": 1.9971631205673759,
630
+ "grad_norm": 0.627955882058217,
631
+ "learning_rate": 5e-06,
632
+ "loss": 0.5869,
633
+ "step": 880
634
+ },
635
+ {
636
+ "epoch": 1.9994326241134752,
637
+ "eval_loss": 0.63065505027771,
638
+ "eval_runtime": 299.005,
639
+ "eval_samples_per_second": 39.712,
640
+ "eval_steps_per_second": 0.622,
641
+ "step": 881
642
+ },
643
+ {
644
+ "epoch": 2.0198581560283686,
645
+ "grad_norm": 0.7421753177971604,
646
+ "learning_rate": 5e-06,
647
+ "loss": 0.5575,
648
+ "step": 890
649
+ },
650
+ {
651
+ "epoch": 2.0425531914893615,
652
+ "grad_norm": 0.7475116659953,
653
+ "learning_rate": 5e-06,
654
+ "loss": 0.5333,
655
+ "step": 900
656
+ },
657
+ {
658
+ "epoch": 2.0652482269503545,
659
+ "grad_norm": 0.6094678843754806,
660
+ "learning_rate": 5e-06,
661
+ "loss": 0.5281,
662
+ "step": 910
663
+ },
664
+ {
665
+ "epoch": 2.0879432624113474,
666
+ "grad_norm": 0.6416679206326653,
667
+ "learning_rate": 5e-06,
668
+ "loss": 0.5307,
669
+ "step": 920
670
+ },
671
+ {
672
+ "epoch": 2.1106382978723404,
673
+ "grad_norm": 0.6149270041451872,
674
+ "learning_rate": 5e-06,
675
+ "loss": 0.5253,
676
+ "step": 930
677
+ },
678
+ {
679
+ "epoch": 2.1333333333333333,
680
+ "grad_norm": 0.5944382793137648,
681
+ "learning_rate": 5e-06,
682
+ "loss": 0.5321,
683
+ "step": 940
684
+ },
685
+ {
686
+ "epoch": 2.1560283687943262,
687
+ "grad_norm": 0.5206404660542666,
688
+ "learning_rate": 5e-06,
689
+ "loss": 0.5327,
690
+ "step": 950
691
+ },
692
+ {
693
+ "epoch": 2.178723404255319,
694
+ "grad_norm": 0.5421603421972422,
695
+ "learning_rate": 5e-06,
696
+ "loss": 0.5283,
697
+ "step": 960
698
+ },
699
+ {
700
+ "epoch": 2.201418439716312,
701
+ "grad_norm": 0.630367827768556,
702
+ "learning_rate": 5e-06,
703
+ "loss": 0.5312,
704
+ "step": 970
705
+ },
706
+ {
707
+ "epoch": 2.224113475177305,
708
+ "grad_norm": 0.5289786942278032,
709
+ "learning_rate": 5e-06,
710
+ "loss": 0.5281,
711
+ "step": 980
712
+ },
713
+ {
714
+ "epoch": 2.246808510638298,
715
+ "grad_norm": 0.6186809984064454,
716
+ "learning_rate": 5e-06,
717
+ "loss": 0.5287,
718
+ "step": 990
719
+ },
720
+ {
721
+ "epoch": 2.269503546099291,
722
+ "grad_norm": 0.5941474628916863,
723
+ "learning_rate": 5e-06,
724
+ "loss": 0.5349,
725
+ "step": 1000
726
+ },
727
+ {
728
+ "epoch": 2.2921985815602834,
729
+ "grad_norm": 0.570443876715086,
730
+ "learning_rate": 5e-06,
731
+ "loss": 0.5295,
732
+ "step": 1010
733
+ },
734
+ {
735
+ "epoch": 2.3148936170212764,
736
+ "grad_norm": 0.6792218051762158,
737
+ "learning_rate": 5e-06,
738
+ "loss": 0.5323,
739
+ "step": 1020
740
+ },
741
+ {
742
+ "epoch": 2.3375886524822693,
743
+ "grad_norm": 0.5535124105821935,
744
+ "learning_rate": 5e-06,
745
+ "loss": 0.5277,
746
+ "step": 1030
747
+ },
748
+ {
749
+ "epoch": 2.3602836879432623,
750
+ "grad_norm": 0.667112587037914,
751
+ "learning_rate": 5e-06,
752
+ "loss": 0.5276,
753
+ "step": 1040
754
+ },
755
+ {
756
+ "epoch": 2.382978723404255,
757
+ "grad_norm": 0.5987387148760719,
758
+ "learning_rate": 5e-06,
759
+ "loss": 0.5295,
760
+ "step": 1050
761
+ },
762
+ {
763
+ "epoch": 2.405673758865248,
764
+ "grad_norm": 0.5484898675236806,
765
+ "learning_rate": 5e-06,
766
+ "loss": 0.5328,
767
+ "step": 1060
768
+ },
769
+ {
770
+ "epoch": 2.428368794326241,
771
+ "grad_norm": 0.5736373550799053,
772
+ "learning_rate": 5e-06,
773
+ "loss": 0.534,
774
+ "step": 1070
775
+ },
776
+ {
777
+ "epoch": 2.451063829787234,
778
+ "grad_norm": 0.5626598264859632,
779
+ "learning_rate": 5e-06,
780
+ "loss": 0.5356,
781
+ "step": 1080
782
+ },
783
+ {
784
+ "epoch": 2.473758865248227,
785
+ "grad_norm": 0.6153434121318484,
786
+ "learning_rate": 5e-06,
787
+ "loss": 0.5306,
788
+ "step": 1090
789
+ },
790
+ {
791
+ "epoch": 2.49645390070922,
792
+ "grad_norm": 0.7252891365142108,
793
+ "learning_rate": 5e-06,
794
+ "loss": 0.5307,
795
+ "step": 1100
796
+ },
797
+ {
798
+ "epoch": 2.519148936170213,
799
+ "grad_norm": 0.6153968835692192,
800
+ "learning_rate": 5e-06,
801
+ "loss": 0.5331,
802
+ "step": 1110
803
+ },
804
+ {
805
+ "epoch": 2.541843971631206,
806
+ "grad_norm": 0.5969808676825302,
807
+ "learning_rate": 5e-06,
808
+ "loss": 0.5363,
809
+ "step": 1120
810
+ },
811
+ {
812
+ "epoch": 2.5645390070921987,
813
+ "grad_norm": 0.5692435999805617,
814
+ "learning_rate": 5e-06,
815
+ "loss": 0.5391,
816
+ "step": 1130
817
+ },
818
+ {
819
+ "epoch": 2.5872340425531917,
820
+ "grad_norm": 0.6180618030016519,
821
+ "learning_rate": 5e-06,
822
+ "loss": 0.5299,
823
+ "step": 1140
824
+ },
825
+ {
826
+ "epoch": 2.6099290780141846,
827
+ "grad_norm": 0.543137746234749,
828
+ "learning_rate": 5e-06,
829
+ "loss": 0.5374,
830
+ "step": 1150
831
+ },
832
+ {
833
+ "epoch": 2.6326241134751776,
834
+ "grad_norm": 0.5200265379748215,
835
+ "learning_rate": 5e-06,
836
+ "loss": 0.5309,
837
+ "step": 1160
838
+ },
839
+ {
840
+ "epoch": 2.65531914893617,
841
+ "grad_norm": 0.5194882503023576,
842
+ "learning_rate": 5e-06,
843
+ "loss": 0.5293,
844
+ "step": 1170
845
+ },
846
+ {
847
+ "epoch": 2.678014184397163,
848
+ "grad_norm": 0.5570786417431203,
849
+ "learning_rate": 5e-06,
850
+ "loss": 0.5316,
851
+ "step": 1180
852
+ },
853
+ {
854
+ "epoch": 2.700709219858156,
855
+ "grad_norm": 0.6007520350434941,
856
+ "learning_rate": 5e-06,
857
+ "loss": 0.5347,
858
+ "step": 1190
859
+ },
860
+ {
861
+ "epoch": 2.723404255319149,
862
+ "grad_norm": 0.5662193506846984,
863
+ "learning_rate": 5e-06,
864
+ "loss": 0.5339,
865
+ "step": 1200
866
+ },
867
+ {
868
+ "epoch": 2.746099290780142,
869
+ "grad_norm": 0.6675198025626778,
870
+ "learning_rate": 5e-06,
871
+ "loss": 0.5346,
872
+ "step": 1210
873
+ },
874
+ {
875
+ "epoch": 2.7687943262411348,
876
+ "grad_norm": 0.6689385730256848,
877
+ "learning_rate": 5e-06,
878
+ "loss": 0.5303,
879
+ "step": 1220
880
+ },
881
+ {
882
+ "epoch": 2.7914893617021277,
883
+ "grad_norm": 0.6400833035990827,
884
+ "learning_rate": 5e-06,
885
+ "loss": 0.5315,
886
+ "step": 1230
887
+ },
888
+ {
889
+ "epoch": 2.8141843971631206,
890
+ "grad_norm": 0.6835204230746162,
891
+ "learning_rate": 5e-06,
892
+ "loss": 0.5367,
893
+ "step": 1240
894
+ },
895
+ {
896
+ "epoch": 2.8368794326241136,
897
+ "grad_norm": 0.5862756798841194,
898
+ "learning_rate": 5e-06,
899
+ "loss": 0.5346,
900
+ "step": 1250
901
+ },
902
+ {
903
+ "epoch": 2.8595744680851065,
904
+ "grad_norm": 0.5727844470422598,
905
+ "learning_rate": 5e-06,
906
+ "loss": 0.5376,
907
+ "step": 1260
908
+ },
909
+ {
910
+ "epoch": 2.8822695035460995,
911
+ "grad_norm": 0.5445857583169009,
912
+ "learning_rate": 5e-06,
913
+ "loss": 0.5294,
914
+ "step": 1270
915
+ },
916
+ {
917
+ "epoch": 2.904964539007092,
918
+ "grad_norm": 0.6161117247407584,
919
+ "learning_rate": 5e-06,
920
+ "loss": 0.5309,
921
+ "step": 1280
922
+ },
923
+ {
924
+ "epoch": 2.927659574468085,
925
+ "grad_norm": 0.6027552334532725,
926
+ "learning_rate": 5e-06,
927
+ "loss": 0.5359,
928
+ "step": 1290
929
+ },
930
+ {
931
+ "epoch": 2.950354609929078,
932
+ "grad_norm": 0.6005574689486347,
933
+ "learning_rate": 5e-06,
934
+ "loss": 0.531,
935
+ "step": 1300
936
+ },
937
+ {
938
+ "epoch": 2.9730496453900708,
939
+ "grad_norm": 0.5590382510647179,
940
+ "learning_rate": 5e-06,
941
+ "loss": 0.5348,
942
+ "step": 1310
943
+ },
944
+ {
945
+ "epoch": 2.9957446808510637,
946
+ "grad_norm": 0.5507647179314145,
947
+ "learning_rate": 5e-06,
948
+ "loss": 0.5325,
949
+ "step": 1320
950
+ },
951
+ {
952
+ "epoch": 2.9957446808510637,
953
+ "eval_loss": 0.6309003829956055,
954
+ "eval_runtime": 301.6902,
955
+ "eval_samples_per_second": 39.358,
956
+ "eval_steps_per_second": 0.617,
957
+ "step": 1320
958
+ },
959
+ {
960
+ "epoch": 2.9957446808510637,
961
+ "step": 1320,
962
+ "total_flos": 2210839784325120.0,
963
+ "train_loss": 0.6033726125052481,
964
+ "train_runtime": 44159.109,
965
+ "train_samples_per_second": 15.326,
966
+ "train_steps_per_second": 0.03
967
+ }
968
+ ],
969
+ "logging_steps": 10,
970
+ "max_steps": 1320,
971
+ "num_input_tokens_seen": 0,
972
+ "num_train_epochs": 3,
973
+ "save_steps": 500,
974
+ "stateful_callbacks": {
975
+ "TrainerControl": {
976
+ "args": {
977
+ "should_epoch_stop": false,
978
+ "should_evaluate": false,
979
+ "should_log": false,
980
+ "should_save": true,
981
+ "should_training_stop": true
982
+ },
983
+ "attributes": {}
984
+ }
985
+ },
986
+ "total_flos": 2210839784325120.0,
987
+ "train_batch_size": 8,
988
+ "trial_name": null,
989
+ "trial_params": null
990
+ }
training_eval_loss.png ADDED
training_loss.png ADDED