sedrickkeh commited on
Commit
1e46add
·
verified ·
1 Parent(s): ee655fe

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-3.1-8b
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
15
 
16
  # oh-dcft-v3.1-llama-3.1-8b
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: 0.3671
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-3.1-8b
 
16
 
17
  # oh-dcft-v3.1-llama-3.1-8b
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-3.1-8b dataset.
20
  It achieves the following results on the evaluation set:
21
  - Loss: 0.3671
22
 
all_results.json ADDED
@@ -0,0 +1,12 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.996777029006739,
3
+ "eval_loss": 0.3670854866504669,
4
+ "eval_runtime": 284.8697,
5
+ "eval_samples_per_second": 40.355,
6
+ "eval_steps_per_second": 0.632,
7
+ "total_flos": 2140488220016640.0,
8
+ "train_loss": 0.33715206393986613,
9
+ "train_runtime": 42576.9653,
10
+ "train_samples_per_second": 15.39,
11
+ "train_steps_per_second": 0.03
12
+ }
eval_results.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.996777029006739,
3
+ "eval_loss": 0.3670854866504669,
4
+ "eval_runtime": 284.8697,
5
+ "eval_samples_per_second": 40.355,
6
+ "eval_steps_per_second": 0.632
7
+ }
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.996777029006739,
3
+ "total_flos": 2140488220016640.0,
4
+ "train_loss": 0.33715206393986613,
5
+ "train_runtime": 42576.9653,
6
+ "train_samples_per_second": 15.39,
7
+ "train_steps_per_second": 0.03
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,955 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 2.996777029006739,
5
+ "eval_steps": 500,
6
+ "global_step": 1278,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.023439789041898623,
13
+ "grad_norm": 9.967006175549518,
14
+ "learning_rate": 5e-06,
15
+ "loss": 0.6132,
16
+ "step": 10
17
+ },
18
+ {
19
+ "epoch": 0.046879578083797245,
20
+ "grad_norm": 1.708901732081446,
21
+ "learning_rate": 5e-06,
22
+ "loss": 0.5268,
23
+ "step": 20
24
+ },
25
+ {
26
+ "epoch": 0.07031936712569586,
27
+ "grad_norm": 1.6601249126276376,
28
+ "learning_rate": 5e-06,
29
+ "loss": 0.4939,
30
+ "step": 30
31
+ },
32
+ {
33
+ "epoch": 0.09375915616759449,
34
+ "grad_norm": 0.9058674995172575,
35
+ "learning_rate": 5e-06,
36
+ "loss": 0.4729,
37
+ "step": 40
38
+ },
39
+ {
40
+ "epoch": 0.11719894520949312,
41
+ "grad_norm": 1.1831753366659354,
42
+ "learning_rate": 5e-06,
43
+ "loss": 0.4606,
44
+ "step": 50
45
+ },
46
+ {
47
+ "epoch": 0.14063873425139173,
48
+ "grad_norm": 0.6857017193079774,
49
+ "learning_rate": 5e-06,
50
+ "loss": 0.4448,
51
+ "step": 60
52
+ },
53
+ {
54
+ "epoch": 0.16407852329329037,
55
+ "grad_norm": 0.665824933523209,
56
+ "learning_rate": 5e-06,
57
+ "loss": 0.4416,
58
+ "step": 70
59
+ },
60
+ {
61
+ "epoch": 0.18751831233518898,
62
+ "grad_norm": 1.5098116393031806,
63
+ "learning_rate": 5e-06,
64
+ "loss": 0.436,
65
+ "step": 80
66
+ },
67
+ {
68
+ "epoch": 0.2109581013770876,
69
+ "grad_norm": 0.8814351421187862,
70
+ "learning_rate": 5e-06,
71
+ "loss": 0.4249,
72
+ "step": 90
73
+ },
74
+ {
75
+ "epoch": 0.23439789041898623,
76
+ "grad_norm": 0.5818056460875761,
77
+ "learning_rate": 5e-06,
78
+ "loss": 0.416,
79
+ "step": 100
80
+ },
81
+ {
82
+ "epoch": 0.25783767946088487,
83
+ "grad_norm": 0.6332178617430071,
84
+ "learning_rate": 5e-06,
85
+ "loss": 0.4143,
86
+ "step": 110
87
+ },
88
+ {
89
+ "epoch": 0.28127746850278346,
90
+ "grad_norm": 0.6385133089285979,
91
+ "learning_rate": 5e-06,
92
+ "loss": 0.4142,
93
+ "step": 120
94
+ },
95
+ {
96
+ "epoch": 0.3047172575446821,
97
+ "grad_norm": 0.5190122301041274,
98
+ "learning_rate": 5e-06,
99
+ "loss": 0.4054,
100
+ "step": 130
101
+ },
102
+ {
103
+ "epoch": 0.32815704658658074,
104
+ "grad_norm": 0.72677604074983,
105
+ "learning_rate": 5e-06,
106
+ "loss": 0.4061,
107
+ "step": 140
108
+ },
109
+ {
110
+ "epoch": 0.3515968356284793,
111
+ "grad_norm": 0.5862867262610982,
112
+ "learning_rate": 5e-06,
113
+ "loss": 0.4002,
114
+ "step": 150
115
+ },
116
+ {
117
+ "epoch": 0.37503662467037796,
118
+ "grad_norm": 0.4943669370299919,
119
+ "learning_rate": 5e-06,
120
+ "loss": 0.4078,
121
+ "step": 160
122
+ },
123
+ {
124
+ "epoch": 0.3984764137122766,
125
+ "grad_norm": 0.617733036177013,
126
+ "learning_rate": 5e-06,
127
+ "loss": 0.396,
128
+ "step": 170
129
+ },
130
+ {
131
+ "epoch": 0.4219162027541752,
132
+ "grad_norm": 0.501523217408356,
133
+ "learning_rate": 5e-06,
134
+ "loss": 0.398,
135
+ "step": 180
136
+ },
137
+ {
138
+ "epoch": 0.4453559917960738,
139
+ "grad_norm": 0.5513765393124738,
140
+ "learning_rate": 5e-06,
141
+ "loss": 0.3915,
142
+ "step": 190
143
+ },
144
+ {
145
+ "epoch": 0.46879578083797246,
146
+ "grad_norm": 0.48423414375531915,
147
+ "learning_rate": 5e-06,
148
+ "loss": 0.3928,
149
+ "step": 200
150
+ },
151
+ {
152
+ "epoch": 0.4922355698798711,
153
+ "grad_norm": 0.6668034319234166,
154
+ "learning_rate": 5e-06,
155
+ "loss": 0.3897,
156
+ "step": 210
157
+ },
158
+ {
159
+ "epoch": 0.5156753589217697,
160
+ "grad_norm": 0.6844513719684475,
161
+ "learning_rate": 5e-06,
162
+ "loss": 0.3943,
163
+ "step": 220
164
+ },
165
+ {
166
+ "epoch": 0.5391151479636683,
167
+ "grad_norm": 0.5634978239261383,
168
+ "learning_rate": 5e-06,
169
+ "loss": 0.3867,
170
+ "step": 230
171
+ },
172
+ {
173
+ "epoch": 0.5625549370055669,
174
+ "grad_norm": 0.48253201050555344,
175
+ "learning_rate": 5e-06,
176
+ "loss": 0.3951,
177
+ "step": 240
178
+ },
179
+ {
180
+ "epoch": 0.5859947260474656,
181
+ "grad_norm": 0.5583859071799139,
182
+ "learning_rate": 5e-06,
183
+ "loss": 0.3896,
184
+ "step": 250
185
+ },
186
+ {
187
+ "epoch": 0.6094345150893642,
188
+ "grad_norm": 0.5725388703107269,
189
+ "learning_rate": 5e-06,
190
+ "loss": 0.3888,
191
+ "step": 260
192
+ },
193
+ {
194
+ "epoch": 0.6328743041312628,
195
+ "grad_norm": 0.8156082811133835,
196
+ "learning_rate": 5e-06,
197
+ "loss": 0.379,
198
+ "step": 270
199
+ },
200
+ {
201
+ "epoch": 0.6563140931731615,
202
+ "grad_norm": 0.5966575531920639,
203
+ "learning_rate": 5e-06,
204
+ "loss": 0.3864,
205
+ "step": 280
206
+ },
207
+ {
208
+ "epoch": 0.6797538822150601,
209
+ "grad_norm": 0.48850438172794114,
210
+ "learning_rate": 5e-06,
211
+ "loss": 0.3873,
212
+ "step": 290
213
+ },
214
+ {
215
+ "epoch": 0.7031936712569586,
216
+ "grad_norm": 0.5962793861443083,
217
+ "learning_rate": 5e-06,
218
+ "loss": 0.3869,
219
+ "step": 300
220
+ },
221
+ {
222
+ "epoch": 0.7266334602988573,
223
+ "grad_norm": 0.5171258822649074,
224
+ "learning_rate": 5e-06,
225
+ "loss": 0.3818,
226
+ "step": 310
227
+ },
228
+ {
229
+ "epoch": 0.7500732493407559,
230
+ "grad_norm": 0.5592073208479841,
231
+ "learning_rate": 5e-06,
232
+ "loss": 0.3755,
233
+ "step": 320
234
+ },
235
+ {
236
+ "epoch": 0.7735130383826545,
237
+ "grad_norm": 0.533904690846359,
238
+ "learning_rate": 5e-06,
239
+ "loss": 0.3745,
240
+ "step": 330
241
+ },
242
+ {
243
+ "epoch": 0.7969528274245532,
244
+ "grad_norm": 0.548421936851489,
245
+ "learning_rate": 5e-06,
246
+ "loss": 0.3794,
247
+ "step": 340
248
+ },
249
+ {
250
+ "epoch": 0.8203926164664518,
251
+ "grad_norm": 0.5826483187799658,
252
+ "learning_rate": 5e-06,
253
+ "loss": 0.3729,
254
+ "step": 350
255
+ },
256
+ {
257
+ "epoch": 0.8438324055083504,
258
+ "grad_norm": 0.65088471853365,
259
+ "learning_rate": 5e-06,
260
+ "loss": 0.3728,
261
+ "step": 360
262
+ },
263
+ {
264
+ "epoch": 0.8672721945502491,
265
+ "grad_norm": 0.7534193430854838,
266
+ "learning_rate": 5e-06,
267
+ "loss": 0.3763,
268
+ "step": 370
269
+ },
270
+ {
271
+ "epoch": 0.8907119835921476,
272
+ "grad_norm": 0.6782662978614515,
273
+ "learning_rate": 5e-06,
274
+ "loss": 0.3794,
275
+ "step": 380
276
+ },
277
+ {
278
+ "epoch": 0.9141517726340463,
279
+ "grad_norm": 0.4771544648840299,
280
+ "learning_rate": 5e-06,
281
+ "loss": 0.3702,
282
+ "step": 390
283
+ },
284
+ {
285
+ "epoch": 0.9375915616759449,
286
+ "grad_norm": 0.5535379054118112,
287
+ "learning_rate": 5e-06,
288
+ "loss": 0.3768,
289
+ "step": 400
290
+ },
291
+ {
292
+ "epoch": 0.9610313507178435,
293
+ "grad_norm": 0.4608571118764034,
294
+ "learning_rate": 5e-06,
295
+ "loss": 0.3788,
296
+ "step": 410
297
+ },
298
+ {
299
+ "epoch": 0.9844711397597422,
300
+ "grad_norm": 0.5668677920657946,
301
+ "learning_rate": 5e-06,
302
+ "loss": 0.3673,
303
+ "step": 420
304
+ },
305
+ {
306
+ "epoch": 0.9985350131848814,
307
+ "eval_loss": 0.37145182490348816,
308
+ "eval_runtime": 299.8322,
309
+ "eval_samples_per_second": 38.341,
310
+ "eval_steps_per_second": 0.6,
311
+ "step": 426
312
+ },
313
+ {
314
+ "epoch": 1.0084969235276882,
315
+ "grad_norm": 0.6984558481581081,
316
+ "learning_rate": 5e-06,
317
+ "loss": 0.3801,
318
+ "step": 430
319
+ },
320
+ {
321
+ "epoch": 1.0319367125695869,
322
+ "grad_norm": 0.5239390968899379,
323
+ "learning_rate": 5e-06,
324
+ "loss": 0.3221,
325
+ "step": 440
326
+ },
327
+ {
328
+ "epoch": 1.0553765016114856,
329
+ "grad_norm": 0.5695148407791144,
330
+ "learning_rate": 5e-06,
331
+ "loss": 0.3257,
332
+ "step": 450
333
+ },
334
+ {
335
+ "epoch": 1.078816290653384,
336
+ "grad_norm": 0.5709165769711674,
337
+ "learning_rate": 5e-06,
338
+ "loss": 0.3205,
339
+ "step": 460
340
+ },
341
+ {
342
+ "epoch": 1.1022560796952827,
343
+ "grad_norm": 0.5285237377857552,
344
+ "learning_rate": 5e-06,
345
+ "loss": 0.3267,
346
+ "step": 470
347
+ },
348
+ {
349
+ "epoch": 1.1256958687371814,
350
+ "grad_norm": 0.5164839735723259,
351
+ "learning_rate": 5e-06,
352
+ "loss": 0.3269,
353
+ "step": 480
354
+ },
355
+ {
356
+ "epoch": 1.1491356577790799,
357
+ "grad_norm": 0.4651823010378352,
358
+ "learning_rate": 5e-06,
359
+ "loss": 0.3299,
360
+ "step": 490
361
+ },
362
+ {
363
+ "epoch": 1.1725754468209786,
364
+ "grad_norm": 0.4611493940935449,
365
+ "learning_rate": 5e-06,
366
+ "loss": 0.3307,
367
+ "step": 500
368
+ },
369
+ {
370
+ "epoch": 1.1960152358628773,
371
+ "grad_norm": 0.5290456767213961,
372
+ "learning_rate": 5e-06,
373
+ "loss": 0.3338,
374
+ "step": 510
375
+ },
376
+ {
377
+ "epoch": 1.219455024904776,
378
+ "grad_norm": 0.49326918332510095,
379
+ "learning_rate": 5e-06,
380
+ "loss": 0.3253,
381
+ "step": 520
382
+ },
383
+ {
384
+ "epoch": 1.2428948139466744,
385
+ "grad_norm": 0.551961171965543,
386
+ "learning_rate": 5e-06,
387
+ "loss": 0.3296,
388
+ "step": 530
389
+ },
390
+ {
391
+ "epoch": 1.2663346029885731,
392
+ "grad_norm": 0.49768749823200925,
393
+ "learning_rate": 5e-06,
394
+ "loss": 0.3256,
395
+ "step": 540
396
+ },
397
+ {
398
+ "epoch": 1.2897743920304716,
399
+ "grad_norm": 0.5626038647037341,
400
+ "learning_rate": 5e-06,
401
+ "loss": 0.3255,
402
+ "step": 550
403
+ },
404
+ {
405
+ "epoch": 1.3132141810723703,
406
+ "grad_norm": 0.5103540222464168,
407
+ "learning_rate": 5e-06,
408
+ "loss": 0.3225,
409
+ "step": 560
410
+ },
411
+ {
412
+ "epoch": 1.336653970114269,
413
+ "grad_norm": 0.540424196440726,
414
+ "learning_rate": 5e-06,
415
+ "loss": 0.3264,
416
+ "step": 570
417
+ },
418
+ {
419
+ "epoch": 1.3600937591561677,
420
+ "grad_norm": 0.5745930638703757,
421
+ "learning_rate": 5e-06,
422
+ "loss": 0.3196,
423
+ "step": 580
424
+ },
425
+ {
426
+ "epoch": 1.3835335481980662,
427
+ "grad_norm": 0.6445462088875488,
428
+ "learning_rate": 5e-06,
429
+ "loss": 0.3244,
430
+ "step": 590
431
+ },
432
+ {
433
+ "epoch": 1.4069733372399649,
434
+ "grad_norm": 0.481096562270643,
435
+ "learning_rate": 5e-06,
436
+ "loss": 0.3224,
437
+ "step": 600
438
+ },
439
+ {
440
+ "epoch": 1.4304131262818633,
441
+ "grad_norm": 0.5092132313164293,
442
+ "learning_rate": 5e-06,
443
+ "loss": 0.3251,
444
+ "step": 610
445
+ },
446
+ {
447
+ "epoch": 1.453852915323762,
448
+ "grad_norm": 0.5067387262720331,
449
+ "learning_rate": 5e-06,
450
+ "loss": 0.3265,
451
+ "step": 620
452
+ },
453
+ {
454
+ "epoch": 1.4772927043656607,
455
+ "grad_norm": 0.48943368667359805,
456
+ "learning_rate": 5e-06,
457
+ "loss": 0.3279,
458
+ "step": 630
459
+ },
460
+ {
461
+ "epoch": 1.5007324934075594,
462
+ "grad_norm": 0.5183176032348761,
463
+ "learning_rate": 5e-06,
464
+ "loss": 0.3227,
465
+ "step": 640
466
+ },
467
+ {
468
+ "epoch": 1.524172282449458,
469
+ "grad_norm": 0.5247150397228669,
470
+ "learning_rate": 5e-06,
471
+ "loss": 0.3181,
472
+ "step": 650
473
+ },
474
+ {
475
+ "epoch": 1.5476120714913566,
476
+ "grad_norm": 0.5298290612904054,
477
+ "learning_rate": 5e-06,
478
+ "loss": 0.3275,
479
+ "step": 660
480
+ },
481
+ {
482
+ "epoch": 1.571051860533255,
483
+ "grad_norm": 0.5335846082176007,
484
+ "learning_rate": 5e-06,
485
+ "loss": 0.3228,
486
+ "step": 670
487
+ },
488
+ {
489
+ "epoch": 1.5944916495751538,
490
+ "grad_norm": 0.49588298624982247,
491
+ "learning_rate": 5e-06,
492
+ "loss": 0.3199,
493
+ "step": 680
494
+ },
495
+ {
496
+ "epoch": 1.6179314386170525,
497
+ "grad_norm": 0.5142614839866478,
498
+ "learning_rate": 5e-06,
499
+ "loss": 0.3317,
500
+ "step": 690
501
+ },
502
+ {
503
+ "epoch": 1.6413712276589512,
504
+ "grad_norm": 0.4957666564649026,
505
+ "learning_rate": 5e-06,
506
+ "loss": 0.3239,
507
+ "step": 700
508
+ },
509
+ {
510
+ "epoch": 1.6648110167008499,
511
+ "grad_norm": 0.5027765035894468,
512
+ "learning_rate": 5e-06,
513
+ "loss": 0.3208,
514
+ "step": 710
515
+ },
516
+ {
517
+ "epoch": 1.6882508057427483,
518
+ "grad_norm": 0.6791514332148134,
519
+ "learning_rate": 5e-06,
520
+ "loss": 0.3184,
521
+ "step": 720
522
+ },
523
+ {
524
+ "epoch": 1.7116905947846468,
525
+ "grad_norm": 0.4925105147109606,
526
+ "learning_rate": 5e-06,
527
+ "loss": 0.3252,
528
+ "step": 730
529
+ },
530
+ {
531
+ "epoch": 1.7351303838265455,
532
+ "grad_norm": 0.5049447941188098,
533
+ "learning_rate": 5e-06,
534
+ "loss": 0.3201,
535
+ "step": 740
536
+ },
537
+ {
538
+ "epoch": 1.7585701728684442,
539
+ "grad_norm": 0.544693187707417,
540
+ "learning_rate": 5e-06,
541
+ "loss": 0.3258,
542
+ "step": 750
543
+ },
544
+ {
545
+ "epoch": 1.7820099619103429,
546
+ "grad_norm": 0.4756047714296392,
547
+ "learning_rate": 5e-06,
548
+ "loss": 0.3231,
549
+ "step": 760
550
+ },
551
+ {
552
+ "epoch": 1.8054497509522416,
553
+ "grad_norm": 0.5599548628855956,
554
+ "learning_rate": 5e-06,
555
+ "loss": 0.3246,
556
+ "step": 770
557
+ },
558
+ {
559
+ "epoch": 1.82888953999414,
560
+ "grad_norm": 0.4756090546668893,
561
+ "learning_rate": 5e-06,
562
+ "loss": 0.3219,
563
+ "step": 780
564
+ },
565
+ {
566
+ "epoch": 1.8523293290360385,
567
+ "grad_norm": 0.6124115056884903,
568
+ "learning_rate": 5e-06,
569
+ "loss": 0.3217,
570
+ "step": 790
571
+ },
572
+ {
573
+ "epoch": 1.8757691180779372,
574
+ "grad_norm": 0.46580289937046976,
575
+ "learning_rate": 5e-06,
576
+ "loss": 0.3259,
577
+ "step": 800
578
+ },
579
+ {
580
+ "epoch": 1.899208907119836,
581
+ "grad_norm": 0.5943245513187589,
582
+ "learning_rate": 5e-06,
583
+ "loss": 0.3197,
584
+ "step": 810
585
+ },
586
+ {
587
+ "epoch": 1.9226486961617346,
588
+ "grad_norm": 0.49042860120039294,
589
+ "learning_rate": 5e-06,
590
+ "loss": 0.3226,
591
+ "step": 820
592
+ },
593
+ {
594
+ "epoch": 1.9460884852036333,
595
+ "grad_norm": 0.5267257339110174,
596
+ "learning_rate": 5e-06,
597
+ "loss": 0.3221,
598
+ "step": 830
599
+ },
600
+ {
601
+ "epoch": 1.9695282742455318,
602
+ "grad_norm": 0.717669514593216,
603
+ "learning_rate": 5e-06,
604
+ "loss": 0.3282,
605
+ "step": 840
606
+ },
607
+ {
608
+ "epoch": 1.9929680632874303,
609
+ "grad_norm": 0.46905846352186026,
610
+ "learning_rate": 5e-06,
611
+ "loss": 0.319,
612
+ "step": 850
613
+ },
614
+ {
615
+ "epoch": 1.99765602109581,
616
+ "eval_loss": 0.36132651567459106,
617
+ "eval_runtime": 296.4899,
618
+ "eval_samples_per_second": 38.774,
619
+ "eval_steps_per_second": 0.607,
620
+ "step": 852
621
+ },
622
+ {
623
+ "epoch": 2.0169938470553763,
624
+ "grad_norm": 0.6240807456372731,
625
+ "learning_rate": 5e-06,
626
+ "loss": 0.3065,
627
+ "step": 860
628
+ },
629
+ {
630
+ "epoch": 2.040433636097275,
631
+ "grad_norm": 0.6288943822511728,
632
+ "learning_rate": 5e-06,
633
+ "loss": 0.2738,
634
+ "step": 870
635
+ },
636
+ {
637
+ "epoch": 2.0638734251391737,
638
+ "grad_norm": 0.5311356208759552,
639
+ "learning_rate": 5e-06,
640
+ "loss": 0.2731,
641
+ "step": 880
642
+ },
643
+ {
644
+ "epoch": 2.0873132141810724,
645
+ "grad_norm": 0.56075413432555,
646
+ "learning_rate": 5e-06,
647
+ "loss": 0.2744,
648
+ "step": 890
649
+ },
650
+ {
651
+ "epoch": 2.110753003222971,
652
+ "grad_norm": 0.5639112603035239,
653
+ "learning_rate": 5e-06,
654
+ "loss": 0.2751,
655
+ "step": 900
656
+ },
657
+ {
658
+ "epoch": 2.13419279226487,
659
+ "grad_norm": 0.5591523409924748,
660
+ "learning_rate": 5e-06,
661
+ "loss": 0.2765,
662
+ "step": 910
663
+ },
664
+ {
665
+ "epoch": 2.157632581306768,
666
+ "grad_norm": 0.5663817052316904,
667
+ "learning_rate": 5e-06,
668
+ "loss": 0.2737,
669
+ "step": 920
670
+ },
671
+ {
672
+ "epoch": 2.1810723703486667,
673
+ "grad_norm": 0.623581090691007,
674
+ "learning_rate": 5e-06,
675
+ "loss": 0.2741,
676
+ "step": 930
677
+ },
678
+ {
679
+ "epoch": 2.2045121593905654,
680
+ "grad_norm": 0.47856596919111055,
681
+ "learning_rate": 5e-06,
682
+ "loss": 0.2756,
683
+ "step": 940
684
+ },
685
+ {
686
+ "epoch": 2.227951948432464,
687
+ "grad_norm": 0.48512193272255605,
688
+ "learning_rate": 5e-06,
689
+ "loss": 0.2737,
690
+ "step": 950
691
+ },
692
+ {
693
+ "epoch": 2.251391737474363,
694
+ "grad_norm": 0.5099859245096886,
695
+ "learning_rate": 5e-06,
696
+ "loss": 0.2796,
697
+ "step": 960
698
+ },
699
+ {
700
+ "epoch": 2.2748315265162615,
701
+ "grad_norm": 0.5305125040779604,
702
+ "learning_rate": 5e-06,
703
+ "loss": 0.273,
704
+ "step": 970
705
+ },
706
+ {
707
+ "epoch": 2.2982713155581598,
708
+ "grad_norm": 0.557582967315411,
709
+ "learning_rate": 5e-06,
710
+ "loss": 0.273,
711
+ "step": 980
712
+ },
713
+ {
714
+ "epoch": 2.3217111046000585,
715
+ "grad_norm": 0.5807484141453545,
716
+ "learning_rate": 5e-06,
717
+ "loss": 0.2755,
718
+ "step": 990
719
+ },
720
+ {
721
+ "epoch": 2.345150893641957,
722
+ "grad_norm": 0.49544753860231366,
723
+ "learning_rate": 5e-06,
724
+ "loss": 0.281,
725
+ "step": 1000
726
+ },
727
+ {
728
+ "epoch": 2.368590682683856,
729
+ "grad_norm": 0.5035116546897159,
730
+ "learning_rate": 5e-06,
731
+ "loss": 0.275,
732
+ "step": 1010
733
+ },
734
+ {
735
+ "epoch": 2.3920304717257546,
736
+ "grad_norm": 0.5229341109864724,
737
+ "learning_rate": 5e-06,
738
+ "loss": 0.2778,
739
+ "step": 1020
740
+ },
741
+ {
742
+ "epoch": 2.4154702607676533,
743
+ "grad_norm": 0.5788744389040322,
744
+ "learning_rate": 5e-06,
745
+ "loss": 0.2843,
746
+ "step": 1030
747
+ },
748
+ {
749
+ "epoch": 2.438910049809552,
750
+ "grad_norm": 0.587705401996759,
751
+ "learning_rate": 5e-06,
752
+ "loss": 0.278,
753
+ "step": 1040
754
+ },
755
+ {
756
+ "epoch": 2.46234983885145,
757
+ "grad_norm": 0.526667348496211,
758
+ "learning_rate": 5e-06,
759
+ "loss": 0.2758,
760
+ "step": 1050
761
+ },
762
+ {
763
+ "epoch": 2.485789627893349,
764
+ "grad_norm": 0.5557320228470611,
765
+ "learning_rate": 5e-06,
766
+ "loss": 0.2749,
767
+ "step": 1060
768
+ },
769
+ {
770
+ "epoch": 2.5092294169352476,
771
+ "grad_norm": 0.5455334258984845,
772
+ "learning_rate": 5e-06,
773
+ "loss": 0.2829,
774
+ "step": 1070
775
+ },
776
+ {
777
+ "epoch": 2.5326692059771463,
778
+ "grad_norm": 0.6049156598816975,
779
+ "learning_rate": 5e-06,
780
+ "loss": 0.2792,
781
+ "step": 1080
782
+ },
783
+ {
784
+ "epoch": 2.556108995019045,
785
+ "grad_norm": 0.6063878997150756,
786
+ "learning_rate": 5e-06,
787
+ "loss": 0.2768,
788
+ "step": 1090
789
+ },
790
+ {
791
+ "epoch": 2.5795487840609432,
792
+ "grad_norm": 0.5807051957808216,
793
+ "learning_rate": 5e-06,
794
+ "loss": 0.2837,
795
+ "step": 1100
796
+ },
797
+ {
798
+ "epoch": 2.602988573102842,
799
+ "grad_norm": 0.48690825278368993,
800
+ "learning_rate": 5e-06,
801
+ "loss": 0.2732,
802
+ "step": 1110
803
+ },
804
+ {
805
+ "epoch": 2.6264283621447406,
806
+ "grad_norm": 0.47298208429795785,
807
+ "learning_rate": 5e-06,
808
+ "loss": 0.2838,
809
+ "step": 1120
810
+ },
811
+ {
812
+ "epoch": 2.6498681511866393,
813
+ "grad_norm": 0.5117446887050897,
814
+ "learning_rate": 5e-06,
815
+ "loss": 0.2799,
816
+ "step": 1130
817
+ },
818
+ {
819
+ "epoch": 2.673307940228538,
820
+ "grad_norm": 0.5788651437910809,
821
+ "learning_rate": 5e-06,
822
+ "loss": 0.2831,
823
+ "step": 1140
824
+ },
825
+ {
826
+ "epoch": 2.6967477292704367,
827
+ "grad_norm": 0.48115847246690635,
828
+ "learning_rate": 5e-06,
829
+ "loss": 0.2795,
830
+ "step": 1150
831
+ },
832
+ {
833
+ "epoch": 2.7201875183123354,
834
+ "grad_norm": 0.4752829940802366,
835
+ "learning_rate": 5e-06,
836
+ "loss": 0.282,
837
+ "step": 1160
838
+ },
839
+ {
840
+ "epoch": 2.7436273073542337,
841
+ "grad_norm": 0.4817048174863979,
842
+ "learning_rate": 5e-06,
843
+ "loss": 0.2798,
844
+ "step": 1170
845
+ },
846
+ {
847
+ "epoch": 2.7670670963961324,
848
+ "grad_norm": 0.5163232884368958,
849
+ "learning_rate": 5e-06,
850
+ "loss": 0.2785,
851
+ "step": 1180
852
+ },
853
+ {
854
+ "epoch": 2.790506885438031,
855
+ "grad_norm": 0.49724121376113645,
856
+ "learning_rate": 5e-06,
857
+ "loss": 0.2871,
858
+ "step": 1190
859
+ },
860
+ {
861
+ "epoch": 2.8139466744799297,
862
+ "grad_norm": 0.575214274563183,
863
+ "learning_rate": 5e-06,
864
+ "loss": 0.277,
865
+ "step": 1200
866
+ },
867
+ {
868
+ "epoch": 2.8373864635218284,
869
+ "grad_norm": 0.49524252636931043,
870
+ "learning_rate": 5e-06,
871
+ "loss": 0.284,
872
+ "step": 1210
873
+ },
874
+ {
875
+ "epoch": 2.8608262525637267,
876
+ "grad_norm": 0.5731018313488945,
877
+ "learning_rate": 5e-06,
878
+ "loss": 0.279,
879
+ "step": 1220
880
+ },
881
+ {
882
+ "epoch": 2.8842660416056254,
883
+ "grad_norm": 0.5359950242882849,
884
+ "learning_rate": 5e-06,
885
+ "loss": 0.2805,
886
+ "step": 1230
887
+ },
888
+ {
889
+ "epoch": 2.907705830647524,
890
+ "grad_norm": 0.5534733032853985,
891
+ "learning_rate": 5e-06,
892
+ "loss": 0.2819,
893
+ "step": 1240
894
+ },
895
+ {
896
+ "epoch": 2.9311456196894228,
897
+ "grad_norm": 0.5653588586425916,
898
+ "learning_rate": 5e-06,
899
+ "loss": 0.276,
900
+ "step": 1250
901
+ },
902
+ {
903
+ "epoch": 2.9545854087313215,
904
+ "grad_norm": 0.5400026090208886,
905
+ "learning_rate": 5e-06,
906
+ "loss": 0.2795,
907
+ "step": 1260
908
+ },
909
+ {
910
+ "epoch": 2.97802519777322,
911
+ "grad_norm": 0.5145229483499503,
912
+ "learning_rate": 5e-06,
913
+ "loss": 0.2817,
914
+ "step": 1270
915
+ },
916
+ {
917
+ "epoch": 2.996777029006739,
918
+ "eval_loss": 0.3670854866504669,
919
+ "eval_runtime": 285.3181,
920
+ "eval_samples_per_second": 40.292,
921
+ "eval_steps_per_second": 0.631,
922
+ "step": 1278
923
+ },
924
+ {
925
+ "epoch": 2.996777029006739,
926
+ "step": 1278,
927
+ "total_flos": 2140488220016640.0,
928
+ "train_loss": 0.33715206393986613,
929
+ "train_runtime": 42576.9653,
930
+ "train_samples_per_second": 15.39,
931
+ "train_steps_per_second": 0.03
932
+ }
933
+ ],
934
+ "logging_steps": 10,
935
+ "max_steps": 1278,
936
+ "num_input_tokens_seen": 0,
937
+ "num_train_epochs": 3,
938
+ "save_steps": 500,
939
+ "stateful_callbacks": {
940
+ "TrainerControl": {
941
+ "args": {
942
+ "should_epoch_stop": false,
943
+ "should_evaluate": false,
944
+ "should_log": false,
945
+ "should_save": true,
946
+ "should_training_stop": true
947
+ },
948
+ "attributes": {}
949
+ }
950
+ },
951
+ "total_flos": 2140488220016640.0,
952
+ "train_batch_size": 8,
953
+ "trial_name": null,
954
+ "trial_params": null
955
+ }
training_eval_loss.png ADDED
training_loss.png ADDED