sedrickkeh commited on
Commit
eb2c3c9
·
verified ·
1 Parent(s): 6877bc9

End of training

Browse files
README.md CHANGED
@@ -4,6 +4,7 @@ license: llama3.1
4
  base_model: meta-llama/Llama-3.1-8B
5
  tags:
6
  - llama-factory
 
7
  - generated_from_trainer
8
  model-index:
9
  - name: oh-dcft-v1-no-curation
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
15
 
16
  # oh-dcft-v1-no-curation
17
 
18
- This model is a fine-tuned version of [meta-llama/Llama-3.1-8B](https://huggingface.co/meta-llama/Llama-3.1-8B) on an unknown dataset.
19
  It achieves the following results on the evaluation set:
20
  - Loss: 0.5306
21
 
 
4
  base_model: meta-llama/Llama-3.1-8B
5
  tags:
6
  - llama-factory
7
+ - full
8
  - generated_from_trainer
9
  model-index:
10
  - name: oh-dcft-v1-no-curation
 
16
 
17
  # oh-dcft-v1-no-curation
18
 
19
+ This model is a fine-tuned version of [meta-llama/Llama-3.1-8B](https://huggingface.co/meta-llama/Llama-3.1-8B) on the mlfoundations-dev/oh-dcft-v1-no-curation-sharegpt-format dataset.
20
  It achieves the following results on the evaluation set:
21
  - Loss: 0.5306
22
 
all_results.json ADDED
@@ -0,0 +1,12 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.994385527136619,
3
+ "eval_loss": 0.5305783748626709,
4
+ "eval_runtime": 135.5906,
5
+ "eval_samples_per_second": 79.607,
6
+ "eval_steps_per_second": 0.627,
7
+ "total_flos": 2009625935216640.0,
8
+ "train_loss": 0.5358249799410502,
9
+ "train_runtime": 20182.0753,
10
+ "train_samples_per_second": 30.484,
11
+ "train_steps_per_second": 0.059
12
+ }
eval_results.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.994385527136619,
3
+ "eval_loss": 0.5305783748626709,
4
+ "eval_runtime": 135.5906,
5
+ "eval_samples_per_second": 79.607,
6
+ "eval_steps_per_second": 0.627
7
+ }
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.994385527136619,
3
+ "total_flos": 2009625935216640.0,
4
+ "train_loss": 0.5358249799410502,
5
+ "train_runtime": 20182.0753,
6
+ "train_samples_per_second": 30.484,
7
+ "train_steps_per_second": 0.059
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,906 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 2.994385527136619,
5
+ "eval_steps": 500,
6
+ "global_step": 1200,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.024953212726138492,
13
+ "grad_norm": 8.501412658362785,
14
+ "learning_rate": 5e-06,
15
+ "loss": 0.9043,
16
+ "step": 10
17
+ },
18
+ {
19
+ "epoch": 0.049906425452276984,
20
+ "grad_norm": 4.686127489377169,
21
+ "learning_rate": 5e-06,
22
+ "loss": 0.7713,
23
+ "step": 20
24
+ },
25
+ {
26
+ "epoch": 0.07485963817841547,
27
+ "grad_norm": 1.065750712888582,
28
+ "learning_rate": 5e-06,
29
+ "loss": 0.7261,
30
+ "step": 30
31
+ },
32
+ {
33
+ "epoch": 0.09981285090455397,
34
+ "grad_norm": 1.1573725910946995,
35
+ "learning_rate": 5e-06,
36
+ "loss": 0.6939,
37
+ "step": 40
38
+ },
39
+ {
40
+ "epoch": 0.12476606363069245,
41
+ "grad_norm": 1.0825665225776888,
42
+ "learning_rate": 5e-06,
43
+ "loss": 0.6773,
44
+ "step": 50
45
+ },
46
+ {
47
+ "epoch": 0.14971927635683094,
48
+ "grad_norm": 0.8282989762012689,
49
+ "learning_rate": 5e-06,
50
+ "loss": 0.6551,
51
+ "step": 60
52
+ },
53
+ {
54
+ "epoch": 0.17467248908296942,
55
+ "grad_norm": 0.8431312859314775,
56
+ "learning_rate": 5e-06,
57
+ "loss": 0.6458,
58
+ "step": 70
59
+ },
60
+ {
61
+ "epoch": 0.19962570180910794,
62
+ "grad_norm": 0.7090142603822943,
63
+ "learning_rate": 5e-06,
64
+ "loss": 0.6379,
65
+ "step": 80
66
+ },
67
+ {
68
+ "epoch": 0.22457891453524642,
69
+ "grad_norm": 0.6113443247263295,
70
+ "learning_rate": 5e-06,
71
+ "loss": 0.6258,
72
+ "step": 90
73
+ },
74
+ {
75
+ "epoch": 0.2495321272613849,
76
+ "grad_norm": 0.5723628494735549,
77
+ "learning_rate": 5e-06,
78
+ "loss": 0.6294,
79
+ "step": 100
80
+ },
81
+ {
82
+ "epoch": 0.2744853399875234,
83
+ "grad_norm": 0.5986798385585294,
84
+ "learning_rate": 5e-06,
85
+ "loss": 0.616,
86
+ "step": 110
87
+ },
88
+ {
89
+ "epoch": 0.2994385527136619,
90
+ "grad_norm": 0.6703694452116971,
91
+ "learning_rate": 5e-06,
92
+ "loss": 0.61,
93
+ "step": 120
94
+ },
95
+ {
96
+ "epoch": 0.32439176543980036,
97
+ "grad_norm": 0.48278079529549894,
98
+ "learning_rate": 5e-06,
99
+ "loss": 0.6088,
100
+ "step": 130
101
+ },
102
+ {
103
+ "epoch": 0.34934497816593885,
104
+ "grad_norm": 0.5383252054302604,
105
+ "learning_rate": 5e-06,
106
+ "loss": 0.6097,
107
+ "step": 140
108
+ },
109
+ {
110
+ "epoch": 0.37429819089207733,
111
+ "grad_norm": 0.5904470071991698,
112
+ "learning_rate": 5e-06,
113
+ "loss": 0.6053,
114
+ "step": 150
115
+ },
116
+ {
117
+ "epoch": 0.39925140361821587,
118
+ "grad_norm": 0.5818363085286041,
119
+ "learning_rate": 5e-06,
120
+ "loss": 0.6014,
121
+ "step": 160
122
+ },
123
+ {
124
+ "epoch": 0.42420461634435436,
125
+ "grad_norm": 0.6764973062341755,
126
+ "learning_rate": 5e-06,
127
+ "loss": 0.5993,
128
+ "step": 170
129
+ },
130
+ {
131
+ "epoch": 0.44915782907049284,
132
+ "grad_norm": 0.6999555821582358,
133
+ "learning_rate": 5e-06,
134
+ "loss": 0.5991,
135
+ "step": 180
136
+ },
137
+ {
138
+ "epoch": 0.4741110417966313,
139
+ "grad_norm": 0.7880847284498282,
140
+ "learning_rate": 5e-06,
141
+ "loss": 0.5961,
142
+ "step": 190
143
+ },
144
+ {
145
+ "epoch": 0.4990642545227698,
146
+ "grad_norm": 0.7979122334833642,
147
+ "learning_rate": 5e-06,
148
+ "loss": 0.5916,
149
+ "step": 200
150
+ },
151
+ {
152
+ "epoch": 0.5240174672489083,
153
+ "grad_norm": 0.5793813415501323,
154
+ "learning_rate": 5e-06,
155
+ "loss": 0.5921,
156
+ "step": 210
157
+ },
158
+ {
159
+ "epoch": 0.5489706799750468,
160
+ "grad_norm": 0.5540868891147879,
161
+ "learning_rate": 5e-06,
162
+ "loss": 0.5806,
163
+ "step": 220
164
+ },
165
+ {
166
+ "epoch": 0.5739238927011853,
167
+ "grad_norm": 0.5338207990715906,
168
+ "learning_rate": 5e-06,
169
+ "loss": 0.5851,
170
+ "step": 230
171
+ },
172
+ {
173
+ "epoch": 0.5988771054273238,
174
+ "grad_norm": 0.5622641527094587,
175
+ "learning_rate": 5e-06,
176
+ "loss": 0.5848,
177
+ "step": 240
178
+ },
179
+ {
180
+ "epoch": 0.6238303181534622,
181
+ "grad_norm": 0.5384076976816572,
182
+ "learning_rate": 5e-06,
183
+ "loss": 0.5791,
184
+ "step": 250
185
+ },
186
+ {
187
+ "epoch": 0.6487835308796007,
188
+ "grad_norm": 0.6033419963307188,
189
+ "learning_rate": 5e-06,
190
+ "loss": 0.5842,
191
+ "step": 260
192
+ },
193
+ {
194
+ "epoch": 0.6737367436057392,
195
+ "grad_norm": 0.5342190421176904,
196
+ "learning_rate": 5e-06,
197
+ "loss": 0.5802,
198
+ "step": 270
199
+ },
200
+ {
201
+ "epoch": 0.6986899563318777,
202
+ "grad_norm": 0.5282395994026974,
203
+ "learning_rate": 5e-06,
204
+ "loss": 0.5769,
205
+ "step": 280
206
+ },
207
+ {
208
+ "epoch": 0.7236431690580162,
209
+ "grad_norm": 0.6039215573643518,
210
+ "learning_rate": 5e-06,
211
+ "loss": 0.5811,
212
+ "step": 290
213
+ },
214
+ {
215
+ "epoch": 0.7485963817841547,
216
+ "grad_norm": 0.5625470869148316,
217
+ "learning_rate": 5e-06,
218
+ "loss": 0.5748,
219
+ "step": 300
220
+ },
221
+ {
222
+ "epoch": 0.7735495945102931,
223
+ "grad_norm": 0.6938197396793522,
224
+ "learning_rate": 5e-06,
225
+ "loss": 0.576,
226
+ "step": 310
227
+ },
228
+ {
229
+ "epoch": 0.7985028072364317,
230
+ "grad_norm": 0.5199860087817598,
231
+ "learning_rate": 5e-06,
232
+ "loss": 0.5733,
233
+ "step": 320
234
+ },
235
+ {
236
+ "epoch": 0.8234560199625702,
237
+ "grad_norm": 0.5347102211468582,
238
+ "learning_rate": 5e-06,
239
+ "loss": 0.5713,
240
+ "step": 330
241
+ },
242
+ {
243
+ "epoch": 0.8484092326887087,
244
+ "grad_norm": 0.49920752644836863,
245
+ "learning_rate": 5e-06,
246
+ "loss": 0.5747,
247
+ "step": 340
248
+ },
249
+ {
250
+ "epoch": 0.8733624454148472,
251
+ "grad_norm": 0.5742360043218735,
252
+ "learning_rate": 5e-06,
253
+ "loss": 0.5672,
254
+ "step": 350
255
+ },
256
+ {
257
+ "epoch": 0.8983156581409857,
258
+ "grad_norm": 0.6082105330996729,
259
+ "learning_rate": 5e-06,
260
+ "loss": 0.5648,
261
+ "step": 360
262
+ },
263
+ {
264
+ "epoch": 0.9232688708671242,
265
+ "grad_norm": 0.5296253264006645,
266
+ "learning_rate": 5e-06,
267
+ "loss": 0.5701,
268
+ "step": 370
269
+ },
270
+ {
271
+ "epoch": 0.9482220835932627,
272
+ "grad_norm": 0.5862019120409135,
273
+ "learning_rate": 5e-06,
274
+ "loss": 0.5693,
275
+ "step": 380
276
+ },
277
+ {
278
+ "epoch": 0.9731752963194011,
279
+ "grad_norm": 0.5004204006192343,
280
+ "learning_rate": 5e-06,
281
+ "loss": 0.5667,
282
+ "step": 390
283
+ },
284
+ {
285
+ "epoch": 0.9981285090455396,
286
+ "grad_norm": 0.5111639493787322,
287
+ "learning_rate": 5e-06,
288
+ "loss": 0.5574,
289
+ "step": 400
290
+ },
291
+ {
292
+ "epoch": 0.9981285090455396,
293
+ "eval_loss": 0.5630170702934265,
294
+ "eval_runtime": 136.6299,
295
+ "eval_samples_per_second": 79.002,
296
+ "eval_steps_per_second": 0.622,
297
+ "step": 400
298
+ },
299
+ {
300
+ "epoch": 1.023081721771678,
301
+ "grad_norm": 0.5280851311810567,
302
+ "learning_rate": 5e-06,
303
+ "loss": 0.5296,
304
+ "step": 410
305
+ },
306
+ {
307
+ "epoch": 1.0480349344978166,
308
+ "grad_norm": 0.8751970650854215,
309
+ "learning_rate": 5e-06,
310
+ "loss": 0.5324,
311
+ "step": 420
312
+ },
313
+ {
314
+ "epoch": 1.072988147223955,
315
+ "grad_norm": 0.5756582314387694,
316
+ "learning_rate": 5e-06,
317
+ "loss": 0.5245,
318
+ "step": 430
319
+ },
320
+ {
321
+ "epoch": 1.0979413599500936,
322
+ "grad_norm": 0.5619488808417233,
323
+ "learning_rate": 5e-06,
324
+ "loss": 0.5277,
325
+ "step": 440
326
+ },
327
+ {
328
+ "epoch": 1.122894572676232,
329
+ "grad_norm": 0.8057161437263283,
330
+ "learning_rate": 5e-06,
331
+ "loss": 0.5255,
332
+ "step": 450
333
+ },
334
+ {
335
+ "epoch": 1.1478477854023705,
336
+ "grad_norm": 0.581447349352516,
337
+ "learning_rate": 5e-06,
338
+ "loss": 0.5246,
339
+ "step": 460
340
+ },
341
+ {
342
+ "epoch": 1.172800998128509,
343
+ "grad_norm": 0.5907089920281934,
344
+ "learning_rate": 5e-06,
345
+ "loss": 0.5308,
346
+ "step": 470
347
+ },
348
+ {
349
+ "epoch": 1.1977542108546475,
350
+ "grad_norm": 0.527607335995899,
351
+ "learning_rate": 5e-06,
352
+ "loss": 0.5314,
353
+ "step": 480
354
+ },
355
+ {
356
+ "epoch": 1.222707423580786,
357
+ "grad_norm": 0.6625045440230309,
358
+ "learning_rate": 5e-06,
359
+ "loss": 0.5271,
360
+ "step": 490
361
+ },
362
+ {
363
+ "epoch": 1.2476606363069245,
364
+ "grad_norm": 0.49276396345156864,
365
+ "learning_rate": 5e-06,
366
+ "loss": 0.5307,
367
+ "step": 500
368
+ },
369
+ {
370
+ "epoch": 1.272613849033063,
371
+ "grad_norm": 0.47964009848881595,
372
+ "learning_rate": 5e-06,
373
+ "loss": 0.525,
374
+ "step": 510
375
+ },
376
+ {
377
+ "epoch": 1.2975670617592014,
378
+ "grad_norm": 0.6903279013904727,
379
+ "learning_rate": 5e-06,
380
+ "loss": 0.5173,
381
+ "step": 520
382
+ },
383
+ {
384
+ "epoch": 1.32252027448534,
385
+ "grad_norm": 0.6235250926674746,
386
+ "learning_rate": 5e-06,
387
+ "loss": 0.5241,
388
+ "step": 530
389
+ },
390
+ {
391
+ "epoch": 1.3474734872114784,
392
+ "grad_norm": 0.7115742016246728,
393
+ "learning_rate": 5e-06,
394
+ "loss": 0.5209,
395
+ "step": 540
396
+ },
397
+ {
398
+ "epoch": 1.372426699937617,
399
+ "grad_norm": 0.5305573388431144,
400
+ "learning_rate": 5e-06,
401
+ "loss": 0.527,
402
+ "step": 550
403
+ },
404
+ {
405
+ "epoch": 1.3973799126637554,
406
+ "grad_norm": 0.5804195692188839,
407
+ "learning_rate": 5e-06,
408
+ "loss": 0.5253,
409
+ "step": 560
410
+ },
411
+ {
412
+ "epoch": 1.4223331253898939,
413
+ "grad_norm": 0.49733338644989056,
414
+ "learning_rate": 5e-06,
415
+ "loss": 0.523,
416
+ "step": 570
417
+ },
418
+ {
419
+ "epoch": 1.4472863381160324,
420
+ "grad_norm": 0.5635813062598766,
421
+ "learning_rate": 5e-06,
422
+ "loss": 0.518,
423
+ "step": 580
424
+ },
425
+ {
426
+ "epoch": 1.472239550842171,
427
+ "grad_norm": 0.5233795326173047,
428
+ "learning_rate": 5e-06,
429
+ "loss": 0.5184,
430
+ "step": 590
431
+ },
432
+ {
433
+ "epoch": 1.4971927635683095,
434
+ "grad_norm": 0.5340519491855418,
435
+ "learning_rate": 5e-06,
436
+ "loss": 0.5199,
437
+ "step": 600
438
+ },
439
+ {
440
+ "epoch": 1.522145976294448,
441
+ "grad_norm": 0.6382502305796666,
442
+ "learning_rate": 5e-06,
443
+ "loss": 0.5152,
444
+ "step": 610
445
+ },
446
+ {
447
+ "epoch": 1.5470991890205865,
448
+ "grad_norm": 0.6114232456070381,
449
+ "learning_rate": 5e-06,
450
+ "loss": 0.5202,
451
+ "step": 620
452
+ },
453
+ {
454
+ "epoch": 1.572052401746725,
455
+ "grad_norm": 0.5267522341441786,
456
+ "learning_rate": 5e-06,
457
+ "loss": 0.5194,
458
+ "step": 630
459
+ },
460
+ {
461
+ "epoch": 1.5970056144728635,
462
+ "grad_norm": 0.49682168460425896,
463
+ "learning_rate": 5e-06,
464
+ "loss": 0.5235,
465
+ "step": 640
466
+ },
467
+ {
468
+ "epoch": 1.621958827199002,
469
+ "grad_norm": 0.5492682660691606,
470
+ "learning_rate": 5e-06,
471
+ "loss": 0.5224,
472
+ "step": 650
473
+ },
474
+ {
475
+ "epoch": 1.6469120399251405,
476
+ "grad_norm": 0.5160022677697064,
477
+ "learning_rate": 5e-06,
478
+ "loss": 0.5136,
479
+ "step": 660
480
+ },
481
+ {
482
+ "epoch": 1.671865252651279,
483
+ "grad_norm": 0.5108996597287156,
484
+ "learning_rate": 5e-06,
485
+ "loss": 0.5143,
486
+ "step": 670
487
+ },
488
+ {
489
+ "epoch": 1.6968184653774174,
490
+ "grad_norm": 0.5398052585347178,
491
+ "learning_rate": 5e-06,
492
+ "loss": 0.5142,
493
+ "step": 680
494
+ },
495
+ {
496
+ "epoch": 1.721771678103556,
497
+ "grad_norm": 0.5061450862678053,
498
+ "learning_rate": 5e-06,
499
+ "loss": 0.5131,
500
+ "step": 690
501
+ },
502
+ {
503
+ "epoch": 1.7467248908296944,
504
+ "grad_norm": 0.5157055149054185,
505
+ "learning_rate": 5e-06,
506
+ "loss": 0.508,
507
+ "step": 700
508
+ },
509
+ {
510
+ "epoch": 1.7716781035558329,
511
+ "grad_norm": 0.5204430984353897,
512
+ "learning_rate": 5e-06,
513
+ "loss": 0.5175,
514
+ "step": 710
515
+ },
516
+ {
517
+ "epoch": 1.7966313162819714,
518
+ "grad_norm": 0.5357849253293278,
519
+ "learning_rate": 5e-06,
520
+ "loss": 0.5168,
521
+ "step": 720
522
+ },
523
+ {
524
+ "epoch": 1.8215845290081099,
525
+ "grad_norm": 0.7132216372119276,
526
+ "learning_rate": 5e-06,
527
+ "loss": 0.5192,
528
+ "step": 730
529
+ },
530
+ {
531
+ "epoch": 1.8465377417342483,
532
+ "grad_norm": 0.6769653937286553,
533
+ "learning_rate": 5e-06,
534
+ "loss": 0.515,
535
+ "step": 740
536
+ },
537
+ {
538
+ "epoch": 1.8714909544603868,
539
+ "grad_norm": 0.47758034322098186,
540
+ "learning_rate": 5e-06,
541
+ "loss": 0.5113,
542
+ "step": 750
543
+ },
544
+ {
545
+ "epoch": 1.8964441671865253,
546
+ "grad_norm": 0.45410954217292865,
547
+ "learning_rate": 5e-06,
548
+ "loss": 0.508,
549
+ "step": 760
550
+ },
551
+ {
552
+ "epoch": 1.9213973799126638,
553
+ "grad_norm": 0.5514628147384041,
554
+ "learning_rate": 5e-06,
555
+ "loss": 0.5111,
556
+ "step": 770
557
+ },
558
+ {
559
+ "epoch": 1.9463505926388023,
560
+ "grad_norm": 0.5437873076115038,
561
+ "learning_rate": 5e-06,
562
+ "loss": 0.515,
563
+ "step": 780
564
+ },
565
+ {
566
+ "epoch": 1.9713038053649408,
567
+ "grad_norm": 0.5047758942757364,
568
+ "learning_rate": 5e-06,
569
+ "loss": 0.509,
570
+ "step": 790
571
+ },
572
+ {
573
+ "epoch": 1.9962570180910792,
574
+ "grad_norm": 0.4712219343056804,
575
+ "learning_rate": 5e-06,
576
+ "loss": 0.5138,
577
+ "step": 800
578
+ },
579
+ {
580
+ "epoch": 1.9987523393636932,
581
+ "eval_loss": 0.5372687578201294,
582
+ "eval_runtime": 138.155,
583
+ "eval_samples_per_second": 78.13,
584
+ "eval_steps_per_second": 0.615,
585
+ "step": 801
586
+ },
587
+ {
588
+ "epoch": 2.0212102308172177,
589
+ "grad_norm": 0.5647461852845252,
590
+ "learning_rate": 5e-06,
591
+ "loss": 0.478,
592
+ "step": 810
593
+ },
594
+ {
595
+ "epoch": 2.046163443543356,
596
+ "grad_norm": 0.6651732696313954,
597
+ "learning_rate": 5e-06,
598
+ "loss": 0.4739,
599
+ "step": 820
600
+ },
601
+ {
602
+ "epoch": 2.0711166562694947,
603
+ "grad_norm": 0.6344890617063265,
604
+ "learning_rate": 5e-06,
605
+ "loss": 0.4742,
606
+ "step": 830
607
+ },
608
+ {
609
+ "epoch": 2.096069868995633,
610
+ "grad_norm": 0.5551443463065495,
611
+ "learning_rate": 5e-06,
612
+ "loss": 0.4757,
613
+ "step": 840
614
+ },
615
+ {
616
+ "epoch": 2.1210230817217717,
617
+ "grad_norm": 0.5821542029735364,
618
+ "learning_rate": 5e-06,
619
+ "loss": 0.4745,
620
+ "step": 850
621
+ },
622
+ {
623
+ "epoch": 2.14597629444791,
624
+ "grad_norm": 0.5036890967140374,
625
+ "learning_rate": 5e-06,
626
+ "loss": 0.4741,
627
+ "step": 860
628
+ },
629
+ {
630
+ "epoch": 2.1709295071740486,
631
+ "grad_norm": 0.526252717008255,
632
+ "learning_rate": 5e-06,
633
+ "loss": 0.4752,
634
+ "step": 870
635
+ },
636
+ {
637
+ "epoch": 2.195882719900187,
638
+ "grad_norm": 0.5161295092624307,
639
+ "learning_rate": 5e-06,
640
+ "loss": 0.4708,
641
+ "step": 880
642
+ },
643
+ {
644
+ "epoch": 2.2208359326263256,
645
+ "grad_norm": 0.5979999195074699,
646
+ "learning_rate": 5e-06,
647
+ "loss": 0.474,
648
+ "step": 890
649
+ },
650
+ {
651
+ "epoch": 2.245789145352464,
652
+ "grad_norm": 0.548385025393527,
653
+ "learning_rate": 5e-06,
654
+ "loss": 0.4725,
655
+ "step": 900
656
+ },
657
+ {
658
+ "epoch": 2.2707423580786026,
659
+ "grad_norm": 0.5786454120324753,
660
+ "learning_rate": 5e-06,
661
+ "loss": 0.4721,
662
+ "step": 910
663
+ },
664
+ {
665
+ "epoch": 2.295695570804741,
666
+ "grad_norm": 0.49033248942630286,
667
+ "learning_rate": 5e-06,
668
+ "loss": 0.4754,
669
+ "step": 920
670
+ },
671
+ {
672
+ "epoch": 2.3206487835308796,
673
+ "grad_norm": 0.578944877860593,
674
+ "learning_rate": 5e-06,
675
+ "loss": 0.4771,
676
+ "step": 930
677
+ },
678
+ {
679
+ "epoch": 2.345601996257018,
680
+ "grad_norm": 0.5444827802066028,
681
+ "learning_rate": 5e-06,
682
+ "loss": 0.478,
683
+ "step": 940
684
+ },
685
+ {
686
+ "epoch": 2.3705552089831565,
687
+ "grad_norm": 0.5738362287944658,
688
+ "learning_rate": 5e-06,
689
+ "loss": 0.4706,
690
+ "step": 950
691
+ },
692
+ {
693
+ "epoch": 2.395508421709295,
694
+ "grad_norm": 0.5497412052181665,
695
+ "learning_rate": 5e-06,
696
+ "loss": 0.4742,
697
+ "step": 960
698
+ },
699
+ {
700
+ "epoch": 2.4204616344354335,
701
+ "grad_norm": 0.6321664132907767,
702
+ "learning_rate": 5e-06,
703
+ "loss": 0.4745,
704
+ "step": 970
705
+ },
706
+ {
707
+ "epoch": 2.445414847161572,
708
+ "grad_norm": 0.5155397917374155,
709
+ "learning_rate": 5e-06,
710
+ "loss": 0.4714,
711
+ "step": 980
712
+ },
713
+ {
714
+ "epoch": 2.4703680598877105,
715
+ "grad_norm": 0.48000015580082134,
716
+ "learning_rate": 5e-06,
717
+ "loss": 0.4737,
718
+ "step": 990
719
+ },
720
+ {
721
+ "epoch": 2.495321272613849,
722
+ "grad_norm": 0.5646370042066121,
723
+ "learning_rate": 5e-06,
724
+ "loss": 0.4737,
725
+ "step": 1000
726
+ },
727
+ {
728
+ "epoch": 2.5202744853399874,
729
+ "grad_norm": 0.5488929805385689,
730
+ "learning_rate": 5e-06,
731
+ "loss": 0.479,
732
+ "step": 1010
733
+ },
734
+ {
735
+ "epoch": 2.545227698066126,
736
+ "grad_norm": 0.5690240798527287,
737
+ "learning_rate": 5e-06,
738
+ "loss": 0.4735,
739
+ "step": 1020
740
+ },
741
+ {
742
+ "epoch": 2.5701809107922644,
743
+ "grad_norm": 0.5241263467703002,
744
+ "learning_rate": 5e-06,
745
+ "loss": 0.4759,
746
+ "step": 1030
747
+ },
748
+ {
749
+ "epoch": 2.595134123518403,
750
+ "grad_norm": 0.4966621491972494,
751
+ "learning_rate": 5e-06,
752
+ "loss": 0.4741,
753
+ "step": 1040
754
+ },
755
+ {
756
+ "epoch": 2.6200873362445414,
757
+ "grad_norm": 0.6751968467864414,
758
+ "learning_rate": 5e-06,
759
+ "loss": 0.4798,
760
+ "step": 1050
761
+ },
762
+ {
763
+ "epoch": 2.64504054897068,
764
+ "grad_norm": 0.6383918524471734,
765
+ "learning_rate": 5e-06,
766
+ "loss": 0.4754,
767
+ "step": 1060
768
+ },
769
+ {
770
+ "epoch": 2.6699937616968183,
771
+ "grad_norm": 0.5256722819070684,
772
+ "learning_rate": 5e-06,
773
+ "loss": 0.4726,
774
+ "step": 1070
775
+ },
776
+ {
777
+ "epoch": 2.694946974422957,
778
+ "grad_norm": 0.5129557694221921,
779
+ "learning_rate": 5e-06,
780
+ "loss": 0.4738,
781
+ "step": 1080
782
+ },
783
+ {
784
+ "epoch": 2.7199001871490953,
785
+ "grad_norm": 0.49845285895513675,
786
+ "learning_rate": 5e-06,
787
+ "loss": 0.4788,
788
+ "step": 1090
789
+ },
790
+ {
791
+ "epoch": 2.744853399875234,
792
+ "grad_norm": 0.6463893459607853,
793
+ "learning_rate": 5e-06,
794
+ "loss": 0.4772,
795
+ "step": 1100
796
+ },
797
+ {
798
+ "epoch": 2.7698066126013723,
799
+ "grad_norm": 0.5326072861407999,
800
+ "learning_rate": 5e-06,
801
+ "loss": 0.4756,
802
+ "step": 1110
803
+ },
804
+ {
805
+ "epoch": 2.7947598253275108,
806
+ "grad_norm": 0.5085270847721544,
807
+ "learning_rate": 5e-06,
808
+ "loss": 0.4741,
809
+ "step": 1120
810
+ },
811
+ {
812
+ "epoch": 2.8197130380536493,
813
+ "grad_norm": 0.5495621538149744,
814
+ "learning_rate": 5e-06,
815
+ "loss": 0.4745,
816
+ "step": 1130
817
+ },
818
+ {
819
+ "epoch": 2.8446662507797877,
820
+ "grad_norm": 0.48958337112517436,
821
+ "learning_rate": 5e-06,
822
+ "loss": 0.4715,
823
+ "step": 1140
824
+ },
825
+ {
826
+ "epoch": 2.8696194635059262,
827
+ "grad_norm": 0.5857338163975798,
828
+ "learning_rate": 5e-06,
829
+ "loss": 0.4747,
830
+ "step": 1150
831
+ },
832
+ {
833
+ "epoch": 2.8945726762320647,
834
+ "grad_norm": 0.48260180409447406,
835
+ "learning_rate": 5e-06,
836
+ "loss": 0.4715,
837
+ "step": 1160
838
+ },
839
+ {
840
+ "epoch": 2.919525888958203,
841
+ "grad_norm": 0.5058713259660398,
842
+ "learning_rate": 5e-06,
843
+ "loss": 0.4722,
844
+ "step": 1170
845
+ },
846
+ {
847
+ "epoch": 2.944479101684342,
848
+ "grad_norm": 0.544953115514363,
849
+ "learning_rate": 5e-06,
850
+ "loss": 0.4765,
851
+ "step": 1180
852
+ },
853
+ {
854
+ "epoch": 2.96943231441048,
855
+ "grad_norm": 0.4680283080397261,
856
+ "learning_rate": 5e-06,
857
+ "loss": 0.4756,
858
+ "step": 1190
859
+ },
860
+ {
861
+ "epoch": 2.994385527136619,
862
+ "grad_norm": 0.52273100015014,
863
+ "learning_rate": 5e-06,
864
+ "loss": 0.4712,
865
+ "step": 1200
866
+ },
867
+ {
868
+ "epoch": 2.994385527136619,
869
+ "eval_loss": 0.5305783748626709,
870
+ "eval_runtime": 137.7646,
871
+ "eval_samples_per_second": 78.351,
872
+ "eval_steps_per_second": 0.617,
873
+ "step": 1200
874
+ },
875
+ {
876
+ "epoch": 2.994385527136619,
877
+ "step": 1200,
878
+ "total_flos": 2009625935216640.0,
879
+ "train_loss": 0.5358249799410502,
880
+ "train_runtime": 20182.0753,
881
+ "train_samples_per_second": 30.484,
882
+ "train_steps_per_second": 0.059
883
+ }
884
+ ],
885
+ "logging_steps": 10,
886
+ "max_steps": 1200,
887
+ "num_input_tokens_seen": 0,
888
+ "num_train_epochs": 3,
889
+ "save_steps": 500,
890
+ "stateful_callbacks": {
891
+ "TrainerControl": {
892
+ "args": {
893
+ "should_epoch_stop": false,
894
+ "should_evaluate": false,
895
+ "should_log": false,
896
+ "should_save": true,
897
+ "should_training_stop": true
898
+ },
899
+ "attributes": {}
900
+ }
901
+ },
902
+ "total_flos": 2009625935216640.0,
903
+ "train_batch_size": 8,
904
+ "trial_name": null,
905
+ "trial_params": null
906
+ }
training_eval_loss.png ADDED
training_loss.png ADDED