sedrickkeh commited on
Commit
87866fc
·
verified ·
1 Parent(s): 6b3adea

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-llama-3.1-nemotron-70b
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
15
 
16
  # oh-dcft-v3.1-llama-3.1-nemotron-70b
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: 0.5541
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-llama-3.1-nemotron-70b
 
16
 
17
  # oh-dcft-v3.1-llama-3.1-nemotron-70b
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-llama-3.1-nemotron-70b dataset.
20
  It achieves the following results on the evaluation set:
21
  - Loss: 0.5541
22
 
all_results.json ADDED
@@ -0,0 +1,12 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.9993044284720614,
3
+ "eval_loss": 0.5541288256645203,
4
+ "eval_runtime": 570.3234,
5
+ "eval_samples_per_second": 25.47,
6
+ "eval_steps_per_second": 0.398,
7
+ "total_flos": 2708325846220800.0,
8
+ "train_loss": 0.542901256162588,
9
+ "train_runtime": 94751.0453,
10
+ "train_samples_per_second": 8.738,
11
+ "train_steps_per_second": 0.017
12
+ }
eval_results.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.9993044284720614,
3
+ "eval_loss": 0.5541288256645203,
4
+ "eval_runtime": 570.3234,
5
+ "eval_samples_per_second": 25.47,
6
+ "eval_steps_per_second": 0.398
7
+ }
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.9993044284720614,
3
+ "total_flos": 2708325846220800.0,
4
+ "train_loss": 0.542901256162588,
5
+ "train_runtime": 94751.0453,
6
+ "train_samples_per_second": 8.738,
7
+ "train_steps_per_second": 0.017
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,1193 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 2.9993044284720614,
5
+ "eval_steps": 500,
6
+ "global_step": 1617,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.018548574078367727,
13
+ "grad_norm": 2.947240343866928,
14
+ "learning_rate": 5e-06,
15
+ "loss": 0.9092,
16
+ "step": 10
17
+ },
18
+ {
19
+ "epoch": 0.037097148156735454,
20
+ "grad_norm": 4.807938505196804,
21
+ "learning_rate": 5e-06,
22
+ "loss": 0.8033,
23
+ "step": 20
24
+ },
25
+ {
26
+ "epoch": 0.055645722235103175,
27
+ "grad_norm": 3.003116346754329,
28
+ "learning_rate": 5e-06,
29
+ "loss": 0.7597,
30
+ "step": 30
31
+ },
32
+ {
33
+ "epoch": 0.07419429631347091,
34
+ "grad_norm": 1.4718449165697798,
35
+ "learning_rate": 5e-06,
36
+ "loss": 0.7382,
37
+ "step": 40
38
+ },
39
+ {
40
+ "epoch": 0.09274287039183862,
41
+ "grad_norm": 1.5160653638400317,
42
+ "learning_rate": 5e-06,
43
+ "loss": 0.7179,
44
+ "step": 50
45
+ },
46
+ {
47
+ "epoch": 0.11129144447020635,
48
+ "grad_norm": 1.2376263065355404,
49
+ "learning_rate": 5e-06,
50
+ "loss": 0.6967,
51
+ "step": 60
52
+ },
53
+ {
54
+ "epoch": 0.12984001854857408,
55
+ "grad_norm": 1.2615134334932208,
56
+ "learning_rate": 5e-06,
57
+ "loss": 0.6892,
58
+ "step": 70
59
+ },
60
+ {
61
+ "epoch": 0.14838859262694182,
62
+ "grad_norm": 0.6297064773701695,
63
+ "learning_rate": 5e-06,
64
+ "loss": 0.6645,
65
+ "step": 80
66
+ },
67
+ {
68
+ "epoch": 0.16693716670530953,
69
+ "grad_norm": 0.8430227369470438,
70
+ "learning_rate": 5e-06,
71
+ "loss": 0.6739,
72
+ "step": 90
73
+ },
74
+ {
75
+ "epoch": 0.18548574078367724,
76
+ "grad_norm": 0.5602106795501006,
77
+ "learning_rate": 5e-06,
78
+ "loss": 0.6596,
79
+ "step": 100
80
+ },
81
+ {
82
+ "epoch": 0.20403431486204499,
83
+ "grad_norm": 0.5414728553841807,
84
+ "learning_rate": 5e-06,
85
+ "loss": 0.6419,
86
+ "step": 110
87
+ },
88
+ {
89
+ "epoch": 0.2225828889404127,
90
+ "grad_norm": 0.581237350923138,
91
+ "learning_rate": 5e-06,
92
+ "loss": 0.6475,
93
+ "step": 120
94
+ },
95
+ {
96
+ "epoch": 0.24113146301878044,
97
+ "grad_norm": 0.9110329106881332,
98
+ "learning_rate": 5e-06,
99
+ "loss": 0.6394,
100
+ "step": 130
101
+ },
102
+ {
103
+ "epoch": 0.25968003709714815,
104
+ "grad_norm": 0.7117787694171418,
105
+ "learning_rate": 5e-06,
106
+ "loss": 0.6349,
107
+ "step": 140
108
+ },
109
+ {
110
+ "epoch": 0.2782286111755159,
111
+ "grad_norm": 0.9097641559079228,
112
+ "learning_rate": 5e-06,
113
+ "loss": 0.6307,
114
+ "step": 150
115
+ },
116
+ {
117
+ "epoch": 0.29677718525388364,
118
+ "grad_norm": 0.925373877659069,
119
+ "learning_rate": 5e-06,
120
+ "loss": 0.6221,
121
+ "step": 160
122
+ },
123
+ {
124
+ "epoch": 0.3153257593322513,
125
+ "grad_norm": 0.56437914746603,
126
+ "learning_rate": 5e-06,
127
+ "loss": 0.6321,
128
+ "step": 170
129
+ },
130
+ {
131
+ "epoch": 0.33387433341061906,
132
+ "grad_norm": 0.8197823974891302,
133
+ "learning_rate": 5e-06,
134
+ "loss": 0.6306,
135
+ "step": 180
136
+ },
137
+ {
138
+ "epoch": 0.3524229074889868,
139
+ "grad_norm": 0.7402190785036176,
140
+ "learning_rate": 5e-06,
141
+ "loss": 0.632,
142
+ "step": 190
143
+ },
144
+ {
145
+ "epoch": 0.3709714815673545,
146
+ "grad_norm": 0.4858062774844838,
147
+ "learning_rate": 5e-06,
148
+ "loss": 0.633,
149
+ "step": 200
150
+ },
151
+ {
152
+ "epoch": 0.38952005564572223,
153
+ "grad_norm": 0.5126564143958117,
154
+ "learning_rate": 5e-06,
155
+ "loss": 0.6406,
156
+ "step": 210
157
+ },
158
+ {
159
+ "epoch": 0.40806862972408997,
160
+ "grad_norm": 0.48094170267698244,
161
+ "learning_rate": 5e-06,
162
+ "loss": 0.6246,
163
+ "step": 220
164
+ },
165
+ {
166
+ "epoch": 0.4266172038024577,
167
+ "grad_norm": 0.6524447128350448,
168
+ "learning_rate": 5e-06,
169
+ "loss": 0.607,
170
+ "step": 230
171
+ },
172
+ {
173
+ "epoch": 0.4451657778808254,
174
+ "grad_norm": 0.5518493275013766,
175
+ "learning_rate": 5e-06,
176
+ "loss": 0.6243,
177
+ "step": 240
178
+ },
179
+ {
180
+ "epoch": 0.46371435195919314,
181
+ "grad_norm": 0.6558097677670804,
182
+ "learning_rate": 5e-06,
183
+ "loss": 0.6171,
184
+ "step": 250
185
+ },
186
+ {
187
+ "epoch": 0.4822629260375609,
188
+ "grad_norm": 0.7137291594764235,
189
+ "learning_rate": 5e-06,
190
+ "loss": 0.606,
191
+ "step": 260
192
+ },
193
+ {
194
+ "epoch": 0.5008115001159286,
195
+ "grad_norm": 0.5275248289780116,
196
+ "learning_rate": 5e-06,
197
+ "loss": 0.6102,
198
+ "step": 270
199
+ },
200
+ {
201
+ "epoch": 0.5193600741942963,
202
+ "grad_norm": 0.8348067083552699,
203
+ "learning_rate": 5e-06,
204
+ "loss": 0.616,
205
+ "step": 280
206
+ },
207
+ {
208
+ "epoch": 0.537908648272664,
209
+ "grad_norm": 0.6106750334627583,
210
+ "learning_rate": 5e-06,
211
+ "loss": 0.6142,
212
+ "step": 290
213
+ },
214
+ {
215
+ "epoch": 0.5564572223510318,
216
+ "grad_norm": 0.6769490656578356,
217
+ "learning_rate": 5e-06,
218
+ "loss": 0.6079,
219
+ "step": 300
220
+ },
221
+ {
222
+ "epoch": 0.5750057964293995,
223
+ "grad_norm": 0.9615740472605508,
224
+ "learning_rate": 5e-06,
225
+ "loss": 0.6055,
226
+ "step": 310
227
+ },
228
+ {
229
+ "epoch": 0.5935543705077673,
230
+ "grad_norm": 0.5778708733971128,
231
+ "learning_rate": 5e-06,
232
+ "loss": 0.6028,
233
+ "step": 320
234
+ },
235
+ {
236
+ "epoch": 0.6121029445861349,
237
+ "grad_norm": 0.4960676395424829,
238
+ "learning_rate": 5e-06,
239
+ "loss": 0.6086,
240
+ "step": 330
241
+ },
242
+ {
243
+ "epoch": 0.6306515186645026,
244
+ "grad_norm": 0.5758649696479989,
245
+ "learning_rate": 5e-06,
246
+ "loss": 0.603,
247
+ "step": 340
248
+ },
249
+ {
250
+ "epoch": 0.6492000927428704,
251
+ "grad_norm": 0.6291485309740135,
252
+ "learning_rate": 5e-06,
253
+ "loss": 0.6058,
254
+ "step": 350
255
+ },
256
+ {
257
+ "epoch": 0.6677486668212381,
258
+ "grad_norm": 0.6179917359269285,
259
+ "learning_rate": 5e-06,
260
+ "loss": 0.5953,
261
+ "step": 360
262
+ },
263
+ {
264
+ "epoch": 0.6862972408996059,
265
+ "grad_norm": 0.552294533130827,
266
+ "learning_rate": 5e-06,
267
+ "loss": 0.6024,
268
+ "step": 370
269
+ },
270
+ {
271
+ "epoch": 0.7048458149779736,
272
+ "grad_norm": 0.5774022034655474,
273
+ "learning_rate": 5e-06,
274
+ "loss": 0.607,
275
+ "step": 380
276
+ },
277
+ {
278
+ "epoch": 0.7233943890563413,
279
+ "grad_norm": 0.7214292693401807,
280
+ "learning_rate": 5e-06,
281
+ "loss": 0.5974,
282
+ "step": 390
283
+ },
284
+ {
285
+ "epoch": 0.741942963134709,
286
+ "grad_norm": 0.6477577613968241,
287
+ "learning_rate": 5e-06,
288
+ "loss": 0.5972,
289
+ "step": 400
290
+ },
291
+ {
292
+ "epoch": 0.7604915372130767,
293
+ "grad_norm": 0.6340600580365108,
294
+ "learning_rate": 5e-06,
295
+ "loss": 0.5816,
296
+ "step": 410
297
+ },
298
+ {
299
+ "epoch": 0.7790401112914445,
300
+ "grad_norm": 0.5680706241521782,
301
+ "learning_rate": 5e-06,
302
+ "loss": 0.5757,
303
+ "step": 420
304
+ },
305
+ {
306
+ "epoch": 0.7975886853698122,
307
+ "grad_norm": 0.6389171777087581,
308
+ "learning_rate": 5e-06,
309
+ "loss": 0.595,
310
+ "step": 430
311
+ },
312
+ {
313
+ "epoch": 0.8161372594481799,
314
+ "grad_norm": 0.6107276360180472,
315
+ "learning_rate": 5e-06,
316
+ "loss": 0.5936,
317
+ "step": 440
318
+ },
319
+ {
320
+ "epoch": 0.8346858335265477,
321
+ "grad_norm": 0.5997875839762729,
322
+ "learning_rate": 5e-06,
323
+ "loss": 0.5856,
324
+ "step": 450
325
+ },
326
+ {
327
+ "epoch": 0.8532344076049154,
328
+ "grad_norm": 0.5683771733599212,
329
+ "learning_rate": 5e-06,
330
+ "loss": 0.5943,
331
+ "step": 460
332
+ },
333
+ {
334
+ "epoch": 0.871782981683283,
335
+ "grad_norm": 0.7010536508894063,
336
+ "learning_rate": 5e-06,
337
+ "loss": 0.5833,
338
+ "step": 470
339
+ },
340
+ {
341
+ "epoch": 0.8903315557616508,
342
+ "grad_norm": 0.5908990476203293,
343
+ "learning_rate": 5e-06,
344
+ "loss": 0.599,
345
+ "step": 480
346
+ },
347
+ {
348
+ "epoch": 0.9088801298400185,
349
+ "grad_norm": 0.6071694127371104,
350
+ "learning_rate": 5e-06,
351
+ "loss": 0.5768,
352
+ "step": 490
353
+ },
354
+ {
355
+ "epoch": 0.9274287039183863,
356
+ "grad_norm": 0.592858345014562,
357
+ "learning_rate": 5e-06,
358
+ "loss": 0.5839,
359
+ "step": 500
360
+ },
361
+ {
362
+ "epoch": 0.945977277996754,
363
+ "grad_norm": 0.800429521567568,
364
+ "learning_rate": 5e-06,
365
+ "loss": 0.5797,
366
+ "step": 510
367
+ },
368
+ {
369
+ "epoch": 0.9645258520751218,
370
+ "grad_norm": 0.7189609643326995,
371
+ "learning_rate": 5e-06,
372
+ "loss": 0.5833,
373
+ "step": 520
374
+ },
375
+ {
376
+ "epoch": 0.9830744261534895,
377
+ "grad_norm": 0.6891202812176819,
378
+ "learning_rate": 5e-06,
379
+ "loss": 0.5897,
380
+ "step": 530
381
+ },
382
+ {
383
+ "epoch": 0.9997681428240204,
384
+ "eval_loss": 0.5783212780952454,
385
+ "eval_runtime": 568.0594,
386
+ "eval_samples_per_second": 25.571,
387
+ "eval_steps_per_second": 0.4,
388
+ "step": 539
389
+ },
390
+ {
391
+ "epoch": 1.0016230002318571,
392
+ "grad_norm": 1.508531876731451,
393
+ "learning_rate": 5e-06,
394
+ "loss": 0.6241,
395
+ "step": 540
396
+ },
397
+ {
398
+ "epoch": 1.0201715743102249,
399
+ "grad_norm": 0.6966255000443538,
400
+ "learning_rate": 5e-06,
401
+ "loss": 0.5375,
402
+ "step": 550
403
+ },
404
+ {
405
+ "epoch": 1.0387201483885926,
406
+ "grad_norm": 0.890924162169634,
407
+ "learning_rate": 5e-06,
408
+ "loss": 0.5428,
409
+ "step": 560
410
+ },
411
+ {
412
+ "epoch": 1.0572687224669604,
413
+ "grad_norm": 0.8758158076398361,
414
+ "learning_rate": 5e-06,
415
+ "loss": 0.5265,
416
+ "step": 570
417
+ },
418
+ {
419
+ "epoch": 1.075817296545328,
420
+ "grad_norm": 0.5756897797693913,
421
+ "learning_rate": 5e-06,
422
+ "loss": 0.5341,
423
+ "step": 580
424
+ },
425
+ {
426
+ "epoch": 1.0943658706236958,
427
+ "grad_norm": 0.6406486451101234,
428
+ "learning_rate": 5e-06,
429
+ "loss": 0.525,
430
+ "step": 590
431
+ },
432
+ {
433
+ "epoch": 1.1129144447020636,
434
+ "grad_norm": 0.6180252143784474,
435
+ "learning_rate": 5e-06,
436
+ "loss": 0.538,
437
+ "step": 600
438
+ },
439
+ {
440
+ "epoch": 1.1314630187804313,
441
+ "grad_norm": 0.5607170765041661,
442
+ "learning_rate": 5e-06,
443
+ "loss": 0.5298,
444
+ "step": 610
445
+ },
446
+ {
447
+ "epoch": 1.150011592858799,
448
+ "grad_norm": 0.5660990160046208,
449
+ "learning_rate": 5e-06,
450
+ "loss": 0.5234,
451
+ "step": 620
452
+ },
453
+ {
454
+ "epoch": 1.1685601669371668,
455
+ "grad_norm": 0.7049943454448735,
456
+ "learning_rate": 5e-06,
457
+ "loss": 0.5314,
458
+ "step": 630
459
+ },
460
+ {
461
+ "epoch": 1.1871087410155345,
462
+ "grad_norm": 0.6322757671168315,
463
+ "learning_rate": 5e-06,
464
+ "loss": 0.5314,
465
+ "step": 640
466
+ },
467
+ {
468
+ "epoch": 1.205657315093902,
469
+ "grad_norm": 0.6315492733495492,
470
+ "learning_rate": 5e-06,
471
+ "loss": 0.5359,
472
+ "step": 650
473
+ },
474
+ {
475
+ "epoch": 1.2242058891722698,
476
+ "grad_norm": 0.5824676673790568,
477
+ "learning_rate": 5e-06,
478
+ "loss": 0.5314,
479
+ "step": 660
480
+ },
481
+ {
482
+ "epoch": 1.2427544632506375,
483
+ "grad_norm": 0.6054188119886005,
484
+ "learning_rate": 5e-06,
485
+ "loss": 0.5305,
486
+ "step": 670
487
+ },
488
+ {
489
+ "epoch": 1.2613030373290053,
490
+ "grad_norm": 0.7111463099630733,
491
+ "learning_rate": 5e-06,
492
+ "loss": 0.5427,
493
+ "step": 680
494
+ },
495
+ {
496
+ "epoch": 1.279851611407373,
497
+ "grad_norm": 0.5473676666520529,
498
+ "learning_rate": 5e-06,
499
+ "loss": 0.5304,
500
+ "step": 690
501
+ },
502
+ {
503
+ "epoch": 1.2984001854857408,
504
+ "grad_norm": 0.5677624386493364,
505
+ "learning_rate": 5e-06,
506
+ "loss": 0.5301,
507
+ "step": 700
508
+ },
509
+ {
510
+ "epoch": 1.3169487595641085,
511
+ "grad_norm": 0.5554522651946812,
512
+ "learning_rate": 5e-06,
513
+ "loss": 0.5216,
514
+ "step": 710
515
+ },
516
+ {
517
+ "epoch": 1.3354973336424762,
518
+ "grad_norm": 0.5796555010460428,
519
+ "learning_rate": 5e-06,
520
+ "loss": 0.5465,
521
+ "step": 720
522
+ },
523
+ {
524
+ "epoch": 1.354045907720844,
525
+ "grad_norm": 0.7811681513039581,
526
+ "learning_rate": 5e-06,
527
+ "loss": 0.5432,
528
+ "step": 730
529
+ },
530
+ {
531
+ "epoch": 1.3725944817992117,
532
+ "grad_norm": 0.5405617867913124,
533
+ "learning_rate": 5e-06,
534
+ "loss": 0.5318,
535
+ "step": 740
536
+ },
537
+ {
538
+ "epoch": 1.3911430558775795,
539
+ "grad_norm": 0.6158407644675764,
540
+ "learning_rate": 5e-06,
541
+ "loss": 0.5282,
542
+ "step": 750
543
+ },
544
+ {
545
+ "epoch": 1.4096916299559472,
546
+ "grad_norm": 0.5535023339764817,
547
+ "learning_rate": 5e-06,
548
+ "loss": 0.5204,
549
+ "step": 760
550
+ },
551
+ {
552
+ "epoch": 1.428240204034315,
553
+ "grad_norm": 0.5811059240915003,
554
+ "learning_rate": 5e-06,
555
+ "loss": 0.5262,
556
+ "step": 770
557
+ },
558
+ {
559
+ "epoch": 1.4467887781126825,
560
+ "grad_norm": 0.5321304586809559,
561
+ "learning_rate": 5e-06,
562
+ "loss": 0.5264,
563
+ "step": 780
564
+ },
565
+ {
566
+ "epoch": 1.4653373521910504,
567
+ "grad_norm": 0.5552151667844475,
568
+ "learning_rate": 5e-06,
569
+ "loss": 0.5264,
570
+ "step": 790
571
+ },
572
+ {
573
+ "epoch": 1.483885926269418,
574
+ "grad_norm": 0.5361161307842528,
575
+ "learning_rate": 5e-06,
576
+ "loss": 0.5268,
577
+ "step": 800
578
+ },
579
+ {
580
+ "epoch": 1.502434500347786,
581
+ "grad_norm": 0.547926650755314,
582
+ "learning_rate": 5e-06,
583
+ "loss": 0.5301,
584
+ "step": 810
585
+ },
586
+ {
587
+ "epoch": 1.5209830744261534,
588
+ "grad_norm": 0.542860060621558,
589
+ "learning_rate": 5e-06,
590
+ "loss": 0.5193,
591
+ "step": 820
592
+ },
593
+ {
594
+ "epoch": 1.5395316485045212,
595
+ "grad_norm": 0.5952263875330506,
596
+ "learning_rate": 5e-06,
597
+ "loss": 0.5216,
598
+ "step": 830
599
+ },
600
+ {
601
+ "epoch": 1.558080222582889,
602
+ "grad_norm": 0.6060219873288911,
603
+ "learning_rate": 5e-06,
604
+ "loss": 0.5259,
605
+ "step": 840
606
+ },
607
+ {
608
+ "epoch": 1.5766287966612567,
609
+ "grad_norm": 0.5526009181452085,
610
+ "learning_rate": 5e-06,
611
+ "loss": 0.5241,
612
+ "step": 850
613
+ },
614
+ {
615
+ "epoch": 1.5951773707396244,
616
+ "grad_norm": 0.7655982897227043,
617
+ "learning_rate": 5e-06,
618
+ "loss": 0.5267,
619
+ "step": 860
620
+ },
621
+ {
622
+ "epoch": 1.6137259448179921,
623
+ "grad_norm": 0.584300099468923,
624
+ "learning_rate": 5e-06,
625
+ "loss": 0.5317,
626
+ "step": 870
627
+ },
628
+ {
629
+ "epoch": 1.6322745188963599,
630
+ "grad_norm": 0.643161799376352,
631
+ "learning_rate": 5e-06,
632
+ "loss": 0.527,
633
+ "step": 880
634
+ },
635
+ {
636
+ "epoch": 1.6508230929747274,
637
+ "grad_norm": 0.5972896345510548,
638
+ "learning_rate": 5e-06,
639
+ "loss": 0.5194,
640
+ "step": 890
641
+ },
642
+ {
643
+ "epoch": 1.6693716670530954,
644
+ "grad_norm": 0.5653434512662245,
645
+ "learning_rate": 5e-06,
646
+ "loss": 0.5206,
647
+ "step": 900
648
+ },
649
+ {
650
+ "epoch": 1.6879202411314629,
651
+ "grad_norm": 0.6357358361838954,
652
+ "learning_rate": 5e-06,
653
+ "loss": 0.515,
654
+ "step": 910
655
+ },
656
+ {
657
+ "epoch": 1.7064688152098308,
658
+ "grad_norm": 0.4604990510970151,
659
+ "learning_rate": 5e-06,
660
+ "loss": 0.5304,
661
+ "step": 920
662
+ },
663
+ {
664
+ "epoch": 1.7250173892881984,
665
+ "grad_norm": 0.546451119916009,
666
+ "learning_rate": 5e-06,
667
+ "loss": 0.5228,
668
+ "step": 930
669
+ },
670
+ {
671
+ "epoch": 1.7435659633665663,
672
+ "grad_norm": 0.6039459271117709,
673
+ "learning_rate": 5e-06,
674
+ "loss": 0.5066,
675
+ "step": 940
676
+ },
677
+ {
678
+ "epoch": 1.7621145374449338,
679
+ "grad_norm": 0.5296235201454201,
680
+ "learning_rate": 5e-06,
681
+ "loss": 0.5219,
682
+ "step": 950
683
+ },
684
+ {
685
+ "epoch": 1.7806631115233018,
686
+ "grad_norm": 0.7121437475875894,
687
+ "learning_rate": 5e-06,
688
+ "loss": 0.519,
689
+ "step": 960
690
+ },
691
+ {
692
+ "epoch": 1.7992116856016693,
693
+ "grad_norm": 0.6767258868529336,
694
+ "learning_rate": 5e-06,
695
+ "loss": 0.5155,
696
+ "step": 970
697
+ },
698
+ {
699
+ "epoch": 1.817760259680037,
700
+ "grad_norm": 0.6347972264963303,
701
+ "learning_rate": 5e-06,
702
+ "loss": 0.5056,
703
+ "step": 980
704
+ },
705
+ {
706
+ "epoch": 1.8363088337584048,
707
+ "grad_norm": 0.5014537431107586,
708
+ "learning_rate": 5e-06,
709
+ "loss": 0.5178,
710
+ "step": 990
711
+ },
712
+ {
713
+ "epoch": 1.8548574078367726,
714
+ "grad_norm": 0.5154954740926758,
715
+ "learning_rate": 5e-06,
716
+ "loss": 0.5235,
717
+ "step": 1000
718
+ },
719
+ {
720
+ "epoch": 1.8734059819151403,
721
+ "grad_norm": 0.7714296144853684,
722
+ "learning_rate": 5e-06,
723
+ "loss": 0.5291,
724
+ "step": 1010
725
+ },
726
+ {
727
+ "epoch": 1.891954555993508,
728
+ "grad_norm": 0.5251823522650411,
729
+ "learning_rate": 5e-06,
730
+ "loss": 0.5136,
731
+ "step": 1020
732
+ },
733
+ {
734
+ "epoch": 1.9105031300718758,
735
+ "grad_norm": 0.5661699899829179,
736
+ "learning_rate": 5e-06,
737
+ "loss": 0.5174,
738
+ "step": 1030
739
+ },
740
+ {
741
+ "epoch": 1.9290517041502433,
742
+ "grad_norm": 0.7742483564691807,
743
+ "learning_rate": 5e-06,
744
+ "loss": 0.524,
745
+ "step": 1040
746
+ },
747
+ {
748
+ "epoch": 1.9476002782286113,
749
+ "grad_norm": 0.6651979093977842,
750
+ "learning_rate": 5e-06,
751
+ "loss": 0.5117,
752
+ "step": 1050
753
+ },
754
+ {
755
+ "epoch": 1.9661488523069788,
756
+ "grad_norm": 0.4966107646307623,
757
+ "learning_rate": 5e-06,
758
+ "loss": 0.5101,
759
+ "step": 1060
760
+ },
761
+ {
762
+ "epoch": 1.9846974263853467,
763
+ "grad_norm": 0.5477414666916758,
764
+ "learning_rate": 5e-06,
765
+ "loss": 0.519,
766
+ "step": 1070
767
+ },
768
+ {
769
+ "epoch": 1.9995362856480408,
770
+ "eval_loss": 0.5572099685668945,
771
+ "eval_runtime": 571.2353,
772
+ "eval_samples_per_second": 25.429,
773
+ "eval_steps_per_second": 0.397,
774
+ "step": 1078
775
+ },
776
+ {
777
+ "epoch": 2.0032460004637143,
778
+ "grad_norm": 0.9550000425676828,
779
+ "learning_rate": 5e-06,
780
+ "loss": 0.5612,
781
+ "step": 1080
782
+ },
783
+ {
784
+ "epoch": 2.0217945745420822,
785
+ "grad_norm": 0.7749538423166364,
786
+ "learning_rate": 5e-06,
787
+ "loss": 0.4796,
788
+ "step": 1090
789
+ },
790
+ {
791
+ "epoch": 2.0403431486204497,
792
+ "grad_norm": 0.6196120958317889,
793
+ "learning_rate": 5e-06,
794
+ "loss": 0.4717,
795
+ "step": 1100
796
+ },
797
+ {
798
+ "epoch": 2.0588917226988177,
799
+ "grad_norm": 0.7503113770111682,
800
+ "learning_rate": 5e-06,
801
+ "loss": 0.4705,
802
+ "step": 1110
803
+ },
804
+ {
805
+ "epoch": 2.0774402967771852,
806
+ "grad_norm": 0.8195010086999319,
807
+ "learning_rate": 5e-06,
808
+ "loss": 0.4661,
809
+ "step": 1120
810
+ },
811
+ {
812
+ "epoch": 2.095988870855553,
813
+ "grad_norm": 0.7634806874449389,
814
+ "learning_rate": 5e-06,
815
+ "loss": 0.4674,
816
+ "step": 1130
817
+ },
818
+ {
819
+ "epoch": 2.1145374449339207,
820
+ "grad_norm": 0.5282535148779405,
821
+ "learning_rate": 5e-06,
822
+ "loss": 0.4645,
823
+ "step": 1140
824
+ },
825
+ {
826
+ "epoch": 2.1330860190122882,
827
+ "grad_norm": 0.8193534272594567,
828
+ "learning_rate": 5e-06,
829
+ "loss": 0.4676,
830
+ "step": 1150
831
+ },
832
+ {
833
+ "epoch": 2.151634593090656,
834
+ "grad_norm": 0.6337110972085372,
835
+ "learning_rate": 5e-06,
836
+ "loss": 0.4657,
837
+ "step": 1160
838
+ },
839
+ {
840
+ "epoch": 2.1701831671690237,
841
+ "grad_norm": 0.5914383150324656,
842
+ "learning_rate": 5e-06,
843
+ "loss": 0.4656,
844
+ "step": 1170
845
+ },
846
+ {
847
+ "epoch": 2.1887317412473917,
848
+ "grad_norm": 0.5943506604962782,
849
+ "learning_rate": 5e-06,
850
+ "loss": 0.4583,
851
+ "step": 1180
852
+ },
853
+ {
854
+ "epoch": 2.207280315325759,
855
+ "grad_norm": 0.5899210327294049,
856
+ "learning_rate": 5e-06,
857
+ "loss": 0.469,
858
+ "step": 1190
859
+ },
860
+ {
861
+ "epoch": 2.225828889404127,
862
+ "grad_norm": 0.611916980187844,
863
+ "learning_rate": 5e-06,
864
+ "loss": 0.4715,
865
+ "step": 1200
866
+ },
867
+ {
868
+ "epoch": 2.2443774634824947,
869
+ "grad_norm": 0.5355848691260124,
870
+ "learning_rate": 5e-06,
871
+ "loss": 0.4584,
872
+ "step": 1210
873
+ },
874
+ {
875
+ "epoch": 2.2629260375608626,
876
+ "grad_norm": 0.9219025948142408,
877
+ "learning_rate": 5e-06,
878
+ "loss": 0.4772,
879
+ "step": 1220
880
+ },
881
+ {
882
+ "epoch": 2.28147461163923,
883
+ "grad_norm": 0.6356945379257924,
884
+ "learning_rate": 5e-06,
885
+ "loss": 0.4701,
886
+ "step": 1230
887
+ },
888
+ {
889
+ "epoch": 2.300023185717598,
890
+ "grad_norm": 0.5513716882387442,
891
+ "learning_rate": 5e-06,
892
+ "loss": 0.4619,
893
+ "step": 1240
894
+ },
895
+ {
896
+ "epoch": 2.3185717597959656,
897
+ "grad_norm": 0.5749300964975611,
898
+ "learning_rate": 5e-06,
899
+ "loss": 0.4734,
900
+ "step": 1250
901
+ },
902
+ {
903
+ "epoch": 2.3371203338743336,
904
+ "grad_norm": 0.536668517924369,
905
+ "learning_rate": 5e-06,
906
+ "loss": 0.4746,
907
+ "step": 1260
908
+ },
909
+ {
910
+ "epoch": 2.355668907952701,
911
+ "grad_norm": 0.6278038505879705,
912
+ "learning_rate": 5e-06,
913
+ "loss": 0.4647,
914
+ "step": 1270
915
+ },
916
+ {
917
+ "epoch": 2.374217482031069,
918
+ "grad_norm": 0.7026457321857001,
919
+ "learning_rate": 5e-06,
920
+ "loss": 0.459,
921
+ "step": 1280
922
+ },
923
+ {
924
+ "epoch": 2.3927660561094366,
925
+ "grad_norm": 0.4965239092186108,
926
+ "learning_rate": 5e-06,
927
+ "loss": 0.4687,
928
+ "step": 1290
929
+ },
930
+ {
931
+ "epoch": 2.411314630187804,
932
+ "grad_norm": 0.6026630222667435,
933
+ "learning_rate": 5e-06,
934
+ "loss": 0.4636,
935
+ "step": 1300
936
+ },
937
+ {
938
+ "epoch": 2.429863204266172,
939
+ "grad_norm": 0.5323286663903507,
940
+ "learning_rate": 5e-06,
941
+ "loss": 0.4798,
942
+ "step": 1310
943
+ },
944
+ {
945
+ "epoch": 2.4484117783445396,
946
+ "grad_norm": 0.5602695467114859,
947
+ "learning_rate": 5e-06,
948
+ "loss": 0.4743,
949
+ "step": 1320
950
+ },
951
+ {
952
+ "epoch": 2.4669603524229076,
953
+ "grad_norm": 0.5102675731281984,
954
+ "learning_rate": 5e-06,
955
+ "loss": 0.479,
956
+ "step": 1330
957
+ },
958
+ {
959
+ "epoch": 2.485508926501275,
960
+ "grad_norm": 0.6117163824572253,
961
+ "learning_rate": 5e-06,
962
+ "loss": 0.4667,
963
+ "step": 1340
964
+ },
965
+ {
966
+ "epoch": 2.504057500579643,
967
+ "grad_norm": 0.5650168400314336,
968
+ "learning_rate": 5e-06,
969
+ "loss": 0.4756,
970
+ "step": 1350
971
+ },
972
+ {
973
+ "epoch": 2.5226060746580106,
974
+ "grad_norm": 0.5983202563411003,
975
+ "learning_rate": 5e-06,
976
+ "loss": 0.4656,
977
+ "step": 1360
978
+ },
979
+ {
980
+ "epoch": 2.5411546487363785,
981
+ "grad_norm": 0.5753074854215314,
982
+ "learning_rate": 5e-06,
983
+ "loss": 0.4769,
984
+ "step": 1370
985
+ },
986
+ {
987
+ "epoch": 2.559703222814746,
988
+ "grad_norm": 0.5728210847879419,
989
+ "learning_rate": 5e-06,
990
+ "loss": 0.4715,
991
+ "step": 1380
992
+ },
993
+ {
994
+ "epoch": 2.578251796893114,
995
+ "grad_norm": 0.6211225492010692,
996
+ "learning_rate": 5e-06,
997
+ "loss": 0.466,
998
+ "step": 1390
999
+ },
1000
+ {
1001
+ "epoch": 2.5968003709714815,
1002
+ "grad_norm": 0.6343921659926295,
1003
+ "learning_rate": 5e-06,
1004
+ "loss": 0.4719,
1005
+ "step": 1400
1006
+ },
1007
+ {
1008
+ "epoch": 2.615348945049849,
1009
+ "grad_norm": 0.5369714001226162,
1010
+ "learning_rate": 5e-06,
1011
+ "loss": 0.4657,
1012
+ "step": 1410
1013
+ },
1014
+ {
1015
+ "epoch": 2.633897519128217,
1016
+ "grad_norm": 0.6567365723948313,
1017
+ "learning_rate": 5e-06,
1018
+ "loss": 0.4649,
1019
+ "step": 1420
1020
+ },
1021
+ {
1022
+ "epoch": 2.652446093206585,
1023
+ "grad_norm": 0.540386802693628,
1024
+ "learning_rate": 5e-06,
1025
+ "loss": 0.4734,
1026
+ "step": 1430
1027
+ },
1028
+ {
1029
+ "epoch": 2.6709946672849525,
1030
+ "grad_norm": 0.5879368969729212,
1031
+ "learning_rate": 5e-06,
1032
+ "loss": 0.4696,
1033
+ "step": 1440
1034
+ },
1035
+ {
1036
+ "epoch": 2.68954324136332,
1037
+ "grad_norm": 0.6738162707613969,
1038
+ "learning_rate": 5e-06,
1039
+ "loss": 0.4626,
1040
+ "step": 1450
1041
+ },
1042
+ {
1043
+ "epoch": 2.708091815441688,
1044
+ "grad_norm": 0.536400620430425,
1045
+ "learning_rate": 5e-06,
1046
+ "loss": 0.4683,
1047
+ "step": 1460
1048
+ },
1049
+ {
1050
+ "epoch": 2.7266403895200555,
1051
+ "grad_norm": 0.6155879189493075,
1052
+ "learning_rate": 5e-06,
1053
+ "loss": 0.4724,
1054
+ "step": 1470
1055
+ },
1056
+ {
1057
+ "epoch": 2.7451889635984235,
1058
+ "grad_norm": 0.549109259484142,
1059
+ "learning_rate": 5e-06,
1060
+ "loss": 0.4756,
1061
+ "step": 1480
1062
+ },
1063
+ {
1064
+ "epoch": 2.763737537676791,
1065
+ "grad_norm": 0.6863931463768982,
1066
+ "learning_rate": 5e-06,
1067
+ "loss": 0.4765,
1068
+ "step": 1490
1069
+ },
1070
+ {
1071
+ "epoch": 2.782286111755159,
1072
+ "grad_norm": 0.7188901458634751,
1073
+ "learning_rate": 5e-06,
1074
+ "loss": 0.4798,
1075
+ "step": 1500
1076
+ },
1077
+ {
1078
+ "epoch": 2.8008346858335265,
1079
+ "grad_norm": 0.5414397122063939,
1080
+ "learning_rate": 5e-06,
1081
+ "loss": 0.4694,
1082
+ "step": 1510
1083
+ },
1084
+ {
1085
+ "epoch": 2.8193832599118944,
1086
+ "grad_norm": 0.5257691036404979,
1087
+ "learning_rate": 5e-06,
1088
+ "loss": 0.4672,
1089
+ "step": 1520
1090
+ },
1091
+ {
1092
+ "epoch": 2.837931833990262,
1093
+ "grad_norm": 0.5415229151538448,
1094
+ "learning_rate": 5e-06,
1095
+ "loss": 0.4704,
1096
+ "step": 1530
1097
+ },
1098
+ {
1099
+ "epoch": 2.85648040806863,
1100
+ "grad_norm": 0.5702613917862038,
1101
+ "learning_rate": 5e-06,
1102
+ "loss": 0.4795,
1103
+ "step": 1540
1104
+ },
1105
+ {
1106
+ "epoch": 2.8750289821469974,
1107
+ "grad_norm": 0.6269031796704044,
1108
+ "learning_rate": 5e-06,
1109
+ "loss": 0.4626,
1110
+ "step": 1550
1111
+ },
1112
+ {
1113
+ "epoch": 2.893577556225365,
1114
+ "grad_norm": 0.6589343281148584,
1115
+ "learning_rate": 5e-06,
1116
+ "loss": 0.4767,
1117
+ "step": 1560
1118
+ },
1119
+ {
1120
+ "epoch": 2.912126130303733,
1121
+ "grad_norm": 0.7261065025479309,
1122
+ "learning_rate": 5e-06,
1123
+ "loss": 0.4681,
1124
+ "step": 1570
1125
+ },
1126
+ {
1127
+ "epoch": 2.930674704382101,
1128
+ "grad_norm": 0.5636416585164459,
1129
+ "learning_rate": 5e-06,
1130
+ "loss": 0.4732,
1131
+ "step": 1580
1132
+ },
1133
+ {
1134
+ "epoch": 2.9492232784604684,
1135
+ "grad_norm": 0.5677677401790827,
1136
+ "learning_rate": 5e-06,
1137
+ "loss": 0.4732,
1138
+ "step": 1590
1139
+ },
1140
+ {
1141
+ "epoch": 2.967771852538836,
1142
+ "grad_norm": 0.6434371938896902,
1143
+ "learning_rate": 5e-06,
1144
+ "loss": 0.4713,
1145
+ "step": 1600
1146
+ },
1147
+ {
1148
+ "epoch": 2.986320426617204,
1149
+ "grad_norm": 0.6181729556001256,
1150
+ "learning_rate": 5e-06,
1151
+ "loss": 0.469,
1152
+ "step": 1610
1153
+ },
1154
+ {
1155
+ "epoch": 2.9993044284720614,
1156
+ "eval_loss": 0.5541288256645203,
1157
+ "eval_runtime": 569.4978,
1158
+ "eval_samples_per_second": 25.507,
1159
+ "eval_steps_per_second": 0.399,
1160
+ "step": 1617
1161
+ },
1162
+ {
1163
+ "epoch": 2.9993044284720614,
1164
+ "step": 1617,
1165
+ "total_flos": 2708325846220800.0,
1166
+ "train_loss": 0.542901256162588,
1167
+ "train_runtime": 94751.0453,
1168
+ "train_samples_per_second": 8.738,
1169
+ "train_steps_per_second": 0.017
1170
+ }
1171
+ ],
1172
+ "logging_steps": 10,
1173
+ "max_steps": 1617,
1174
+ "num_input_tokens_seen": 0,
1175
+ "num_train_epochs": 3,
1176
+ "save_steps": 500,
1177
+ "stateful_callbacks": {
1178
+ "TrainerControl": {
1179
+ "args": {
1180
+ "should_epoch_stop": false,
1181
+ "should_evaluate": false,
1182
+ "should_log": false,
1183
+ "should_save": true,
1184
+ "should_training_stop": true
1185
+ },
1186
+ "attributes": {}
1187
+ }
1188
+ },
1189
+ "total_flos": 2708325846220800.0,
1190
+ "train_batch_size": 8,
1191
+ "trial_name": null,
1192
+ "trial_params": null
1193
+ }
training_eval_loss.png ADDED
training_loss.png ADDED