sedrickkeh commited on
Commit
a12f1d0
1 Parent(s): 2f13b04

End of training

Browse files
README.md CHANGED
@@ -4,6 +4,7 @@ license: apache-2.0
4
  base_model: Qwen/Qwen2.5-7B
5
  tags:
6
  - llama-factory
 
7
  - generated_from_trainer
8
  model-index:
9
  - name: hp_ablations_qwen_adambeta1_0.9
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
15
 
16
  # hp_ablations_qwen_adambeta1_0.9
17
 
18
- This model is a fine-tuned version of [Qwen/Qwen2.5-7B](https://huggingface.co/Qwen/Qwen2.5-7B) on an unknown dataset.
19
  It achieves the following results on the evaluation set:
20
  - Loss: 0.6187
21
 
 
4
  base_model: Qwen/Qwen2.5-7B
5
  tags:
6
  - llama-factory
7
+ - full
8
  - generated_from_trainer
9
  model-index:
10
  - name: hp_ablations_qwen_adambeta1_0.9
 
16
 
17
  # hp_ablations_qwen_adambeta1_0.9
18
 
19
+ This model is a fine-tuned version of [Qwen/Qwen2.5-7B](https://huggingface.co/Qwen/Qwen2.5-7B) on the mlfoundations-dev/oh-dcft-v3.1-gpt-4o-mini dataset.
20
  It achieves the following results on the evaluation set:
21
  - Loss: 0.6187
22
 
all_results.json ADDED
@@ -0,0 +1,12 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.996011396011396,
3
+ "eval_loss": 0.6186709403991699,
4
+ "eval_runtime": 442.7471,
5
+ "eval_samples_per_second": 26.704,
6
+ "eval_steps_per_second": 0.418,
7
+ "total_flos": 2755219238682624.0,
8
+ "train_loss": 0.6038080361335789,
9
+ "train_runtime": 70053.326,
10
+ "train_samples_per_second": 9.619,
11
+ "train_steps_per_second": 0.019
12
+ }
eval_results.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.996011396011396,
3
+ "eval_loss": 0.6186709403991699,
4
+ "eval_runtime": 442.7471,
5
+ "eval_samples_per_second": 26.704,
6
+ "eval_steps_per_second": 0.418
7
+ }
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.996011396011396,
3
+ "total_flos": 2755219238682624.0,
4
+ "train_loss": 0.6038080361335789,
5
+ "train_runtime": 70053.326,
6
+ "train_samples_per_second": 9.619,
7
+ "train_steps_per_second": 0.019
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,983 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 2.996011396011396,
5
+ "eval_steps": 500,
6
+ "global_step": 1314,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.022792022792022793,
13
+ "grad_norm": 1.1971258233647637,
14
+ "learning_rate": 5e-06,
15
+ "loss": 0.7556,
16
+ "step": 10
17
+ },
18
+ {
19
+ "epoch": 0.045584045584045586,
20
+ "grad_norm": 0.7980484827875407,
21
+ "learning_rate": 5e-06,
22
+ "loss": 0.7003,
23
+ "step": 20
24
+ },
25
+ {
26
+ "epoch": 0.06837606837606838,
27
+ "grad_norm": 0.9071642149447434,
28
+ "learning_rate": 5e-06,
29
+ "loss": 0.6796,
30
+ "step": 30
31
+ },
32
+ {
33
+ "epoch": 0.09116809116809117,
34
+ "grad_norm": 0.8612137999379124,
35
+ "learning_rate": 5e-06,
36
+ "loss": 0.6794,
37
+ "step": 40
38
+ },
39
+ {
40
+ "epoch": 0.11396011396011396,
41
+ "grad_norm": 0.8035666098155594,
42
+ "learning_rate": 5e-06,
43
+ "loss": 0.6755,
44
+ "step": 50
45
+ },
46
+ {
47
+ "epoch": 0.13675213675213677,
48
+ "grad_norm": 0.6770011867016273,
49
+ "learning_rate": 5e-06,
50
+ "loss": 0.6551,
51
+ "step": 60
52
+ },
53
+ {
54
+ "epoch": 0.15954415954415954,
55
+ "grad_norm": 0.44801829754361133,
56
+ "learning_rate": 5e-06,
57
+ "loss": 0.652,
58
+ "step": 70
59
+ },
60
+ {
61
+ "epoch": 0.18233618233618235,
62
+ "grad_norm": 0.37755075969161783,
63
+ "learning_rate": 5e-06,
64
+ "loss": 0.655,
65
+ "step": 80
66
+ },
67
+ {
68
+ "epoch": 0.20512820512820512,
69
+ "grad_norm": 0.31540683060359476,
70
+ "learning_rate": 5e-06,
71
+ "loss": 0.6397,
72
+ "step": 90
73
+ },
74
+ {
75
+ "epoch": 0.22792022792022792,
76
+ "grad_norm": 0.3007494360593848,
77
+ "learning_rate": 5e-06,
78
+ "loss": 0.641,
79
+ "step": 100
80
+ },
81
+ {
82
+ "epoch": 0.25071225071225073,
83
+ "grad_norm": 0.33895214190090917,
84
+ "learning_rate": 5e-06,
85
+ "loss": 0.6323,
86
+ "step": 110
87
+ },
88
+ {
89
+ "epoch": 0.27350427350427353,
90
+ "grad_norm": 0.3070144915055194,
91
+ "learning_rate": 5e-06,
92
+ "loss": 0.6406,
93
+ "step": 120
94
+ },
95
+ {
96
+ "epoch": 0.2962962962962963,
97
+ "grad_norm": 0.2895750901265279,
98
+ "learning_rate": 5e-06,
99
+ "loss": 0.6416,
100
+ "step": 130
101
+ },
102
+ {
103
+ "epoch": 0.3190883190883191,
104
+ "grad_norm": 0.3078158388563675,
105
+ "learning_rate": 5e-06,
106
+ "loss": 0.6437,
107
+ "step": 140
108
+ },
109
+ {
110
+ "epoch": 0.3418803418803419,
111
+ "grad_norm": 0.3110436970966206,
112
+ "learning_rate": 5e-06,
113
+ "loss": 0.6362,
114
+ "step": 150
115
+ },
116
+ {
117
+ "epoch": 0.3646723646723647,
118
+ "grad_norm": 0.31245569584499555,
119
+ "learning_rate": 5e-06,
120
+ "loss": 0.6414,
121
+ "step": 160
122
+ },
123
+ {
124
+ "epoch": 0.38746438746438744,
125
+ "grad_norm": 0.3596651585190044,
126
+ "learning_rate": 5e-06,
127
+ "loss": 0.6332,
128
+ "step": 170
129
+ },
130
+ {
131
+ "epoch": 0.41025641025641024,
132
+ "grad_norm": 0.2963606015052807,
133
+ "learning_rate": 5e-06,
134
+ "loss": 0.6387,
135
+ "step": 180
136
+ },
137
+ {
138
+ "epoch": 0.43304843304843305,
139
+ "grad_norm": 0.30315877922859497,
140
+ "learning_rate": 5e-06,
141
+ "loss": 0.6313,
142
+ "step": 190
143
+ },
144
+ {
145
+ "epoch": 0.45584045584045585,
146
+ "grad_norm": 0.30426971666648045,
147
+ "learning_rate": 5e-06,
148
+ "loss": 0.6394,
149
+ "step": 200
150
+ },
151
+ {
152
+ "epoch": 0.47863247863247865,
153
+ "grad_norm": 0.3090940011420846,
154
+ "learning_rate": 5e-06,
155
+ "loss": 0.6357,
156
+ "step": 210
157
+ },
158
+ {
159
+ "epoch": 0.5014245014245015,
160
+ "grad_norm": 0.3030681417284509,
161
+ "learning_rate": 5e-06,
162
+ "loss": 0.6394,
163
+ "step": 220
164
+ },
165
+ {
166
+ "epoch": 0.5242165242165242,
167
+ "grad_norm": 0.3237381931795037,
168
+ "learning_rate": 5e-06,
169
+ "loss": 0.635,
170
+ "step": 230
171
+ },
172
+ {
173
+ "epoch": 0.5470085470085471,
174
+ "grad_norm": 0.3005504395534347,
175
+ "learning_rate": 5e-06,
176
+ "loss": 0.6309,
177
+ "step": 240
178
+ },
179
+ {
180
+ "epoch": 0.5698005698005698,
181
+ "grad_norm": 0.33169918765087686,
182
+ "learning_rate": 5e-06,
183
+ "loss": 0.6358,
184
+ "step": 250
185
+ },
186
+ {
187
+ "epoch": 0.5925925925925926,
188
+ "grad_norm": 0.3009298815938261,
189
+ "learning_rate": 5e-06,
190
+ "loss": 0.634,
191
+ "step": 260
192
+ },
193
+ {
194
+ "epoch": 0.6153846153846154,
195
+ "grad_norm": 0.30744307688401507,
196
+ "learning_rate": 5e-06,
197
+ "loss": 0.6311,
198
+ "step": 270
199
+ },
200
+ {
201
+ "epoch": 0.6381766381766382,
202
+ "grad_norm": 0.3279175755970601,
203
+ "learning_rate": 5e-06,
204
+ "loss": 0.6419,
205
+ "step": 280
206
+ },
207
+ {
208
+ "epoch": 0.6609686609686609,
209
+ "grad_norm": 0.3289090789927867,
210
+ "learning_rate": 5e-06,
211
+ "loss": 0.6306,
212
+ "step": 290
213
+ },
214
+ {
215
+ "epoch": 0.6837606837606838,
216
+ "grad_norm": 0.2854748875179496,
217
+ "learning_rate": 5e-06,
218
+ "loss": 0.6287,
219
+ "step": 300
220
+ },
221
+ {
222
+ "epoch": 0.7065527065527065,
223
+ "grad_norm": 0.2991563777299476,
224
+ "learning_rate": 5e-06,
225
+ "loss": 0.6286,
226
+ "step": 310
227
+ },
228
+ {
229
+ "epoch": 0.7293447293447294,
230
+ "grad_norm": 0.33511506767288557,
231
+ "learning_rate": 5e-06,
232
+ "loss": 0.6291,
233
+ "step": 320
234
+ },
235
+ {
236
+ "epoch": 0.7521367521367521,
237
+ "grad_norm": 0.3037033252170676,
238
+ "learning_rate": 5e-06,
239
+ "loss": 0.6258,
240
+ "step": 330
241
+ },
242
+ {
243
+ "epoch": 0.7749287749287749,
244
+ "grad_norm": 0.33097093854534626,
245
+ "learning_rate": 5e-06,
246
+ "loss": 0.6316,
247
+ "step": 340
248
+ },
249
+ {
250
+ "epoch": 0.7977207977207977,
251
+ "grad_norm": 0.3307444115460614,
252
+ "learning_rate": 5e-06,
253
+ "loss": 0.6387,
254
+ "step": 350
255
+ },
256
+ {
257
+ "epoch": 0.8205128205128205,
258
+ "grad_norm": 0.3151765558306375,
259
+ "learning_rate": 5e-06,
260
+ "loss": 0.6304,
261
+ "step": 360
262
+ },
263
+ {
264
+ "epoch": 0.8433048433048433,
265
+ "grad_norm": 0.3379019384905337,
266
+ "learning_rate": 5e-06,
267
+ "loss": 0.6242,
268
+ "step": 370
269
+ },
270
+ {
271
+ "epoch": 0.8660968660968661,
272
+ "grad_norm": 0.3126887513754788,
273
+ "learning_rate": 5e-06,
274
+ "loss": 0.6365,
275
+ "step": 380
276
+ },
277
+ {
278
+ "epoch": 0.8888888888888888,
279
+ "grad_norm": 0.32925730498547584,
280
+ "learning_rate": 5e-06,
281
+ "loss": 0.635,
282
+ "step": 390
283
+ },
284
+ {
285
+ "epoch": 0.9116809116809117,
286
+ "grad_norm": 0.2978036556412762,
287
+ "learning_rate": 5e-06,
288
+ "loss": 0.6382,
289
+ "step": 400
290
+ },
291
+ {
292
+ "epoch": 0.9344729344729344,
293
+ "grad_norm": 0.289640703225658,
294
+ "learning_rate": 5e-06,
295
+ "loss": 0.6345,
296
+ "step": 410
297
+ },
298
+ {
299
+ "epoch": 0.9572649572649573,
300
+ "grad_norm": 0.3026561525544134,
301
+ "learning_rate": 5e-06,
302
+ "loss": 0.6214,
303
+ "step": 420
304
+ },
305
+ {
306
+ "epoch": 0.98005698005698,
307
+ "grad_norm": 0.3397854219636717,
308
+ "learning_rate": 5e-06,
309
+ "loss": 0.6344,
310
+ "step": 430
311
+ },
312
+ {
313
+ "epoch": 0.9982905982905983,
314
+ "eval_loss": 0.6250635981559753,
315
+ "eval_runtime": 439.0667,
316
+ "eval_samples_per_second": 26.928,
317
+ "eval_steps_per_second": 0.421,
318
+ "step": 438
319
+ },
320
+ {
321
+ "epoch": 1.0034188034188034,
322
+ "grad_norm": 0.3459608098032768,
323
+ "learning_rate": 5e-06,
324
+ "loss": 0.6471,
325
+ "step": 440
326
+ },
327
+ {
328
+ "epoch": 1.0262108262108263,
329
+ "grad_norm": 0.3262562439094346,
330
+ "learning_rate": 5e-06,
331
+ "loss": 0.6014,
332
+ "step": 450
333
+ },
334
+ {
335
+ "epoch": 1.049002849002849,
336
+ "grad_norm": 0.31143402213282795,
337
+ "learning_rate": 5e-06,
338
+ "loss": 0.5952,
339
+ "step": 460
340
+ },
341
+ {
342
+ "epoch": 1.0717948717948718,
343
+ "grad_norm": 0.2910747569488903,
344
+ "learning_rate": 5e-06,
345
+ "loss": 0.596,
346
+ "step": 470
347
+ },
348
+ {
349
+ "epoch": 1.0945868945868946,
350
+ "grad_norm": 0.30400774756928156,
351
+ "learning_rate": 5e-06,
352
+ "loss": 0.6035,
353
+ "step": 480
354
+ },
355
+ {
356
+ "epoch": 1.1173789173789175,
357
+ "grad_norm": 0.33115180700526176,
358
+ "learning_rate": 5e-06,
359
+ "loss": 0.6086,
360
+ "step": 490
361
+ },
362
+ {
363
+ "epoch": 1.1401709401709401,
364
+ "grad_norm": 0.29149552599551304,
365
+ "learning_rate": 5e-06,
366
+ "loss": 0.5984,
367
+ "step": 500
368
+ },
369
+ {
370
+ "epoch": 1.162962962962963,
371
+ "grad_norm": 0.3109915110332694,
372
+ "learning_rate": 5e-06,
373
+ "loss": 0.5963,
374
+ "step": 510
375
+ },
376
+ {
377
+ "epoch": 1.1857549857549858,
378
+ "grad_norm": 0.2961129445555981,
379
+ "learning_rate": 5e-06,
380
+ "loss": 0.6021,
381
+ "step": 520
382
+ },
383
+ {
384
+ "epoch": 1.2085470085470085,
385
+ "grad_norm": 0.32307405333599015,
386
+ "learning_rate": 5e-06,
387
+ "loss": 0.5951,
388
+ "step": 530
389
+ },
390
+ {
391
+ "epoch": 1.2313390313390313,
392
+ "grad_norm": 0.288036504729705,
393
+ "learning_rate": 5e-06,
394
+ "loss": 0.6053,
395
+ "step": 540
396
+ },
397
+ {
398
+ "epoch": 1.2541310541310542,
399
+ "grad_norm": 0.33288538265730133,
400
+ "learning_rate": 5e-06,
401
+ "loss": 0.6019,
402
+ "step": 550
403
+ },
404
+ {
405
+ "epoch": 1.2769230769230768,
406
+ "grad_norm": 0.2879818949036318,
407
+ "learning_rate": 5e-06,
408
+ "loss": 0.6034,
409
+ "step": 560
410
+ },
411
+ {
412
+ "epoch": 1.2997150997150997,
413
+ "grad_norm": 0.3062524679342497,
414
+ "learning_rate": 5e-06,
415
+ "loss": 0.5975,
416
+ "step": 570
417
+ },
418
+ {
419
+ "epoch": 1.3225071225071225,
420
+ "grad_norm": 0.2720920616055293,
421
+ "learning_rate": 5e-06,
422
+ "loss": 0.5959,
423
+ "step": 580
424
+ },
425
+ {
426
+ "epoch": 1.3452991452991454,
427
+ "grad_norm": 0.2939303230115104,
428
+ "learning_rate": 5e-06,
429
+ "loss": 0.5957,
430
+ "step": 590
431
+ },
432
+ {
433
+ "epoch": 1.368091168091168,
434
+ "grad_norm": 0.3473254630486921,
435
+ "learning_rate": 5e-06,
436
+ "loss": 0.6013,
437
+ "step": 600
438
+ },
439
+ {
440
+ "epoch": 1.390883190883191,
441
+ "grad_norm": 0.3225692085792273,
442
+ "learning_rate": 5e-06,
443
+ "loss": 0.6013,
444
+ "step": 610
445
+ },
446
+ {
447
+ "epoch": 1.4136752136752135,
448
+ "grad_norm": 0.2932018066888041,
449
+ "learning_rate": 5e-06,
450
+ "loss": 0.6039,
451
+ "step": 620
452
+ },
453
+ {
454
+ "epoch": 1.4364672364672364,
455
+ "grad_norm": 0.3139198151802545,
456
+ "learning_rate": 5e-06,
457
+ "loss": 0.6019,
458
+ "step": 630
459
+ },
460
+ {
461
+ "epoch": 1.4592592592592593,
462
+ "grad_norm": 0.30804637385823225,
463
+ "learning_rate": 5e-06,
464
+ "loss": 0.5968,
465
+ "step": 640
466
+ },
467
+ {
468
+ "epoch": 1.4820512820512821,
469
+ "grad_norm": 0.29934506174738645,
470
+ "learning_rate": 5e-06,
471
+ "loss": 0.5947,
472
+ "step": 650
473
+ },
474
+ {
475
+ "epoch": 1.504843304843305,
476
+ "grad_norm": 0.2908995234453594,
477
+ "learning_rate": 5e-06,
478
+ "loss": 0.5944,
479
+ "step": 660
480
+ },
481
+ {
482
+ "epoch": 1.5276353276353276,
483
+ "grad_norm": 0.31573752727748167,
484
+ "learning_rate": 5e-06,
485
+ "loss": 0.602,
486
+ "step": 670
487
+ },
488
+ {
489
+ "epoch": 1.5504273504273505,
490
+ "grad_norm": 0.27582728820967456,
491
+ "learning_rate": 5e-06,
492
+ "loss": 0.5906,
493
+ "step": 680
494
+ },
495
+ {
496
+ "epoch": 1.573219373219373,
497
+ "grad_norm": 0.32900911461957716,
498
+ "learning_rate": 5e-06,
499
+ "loss": 0.5945,
500
+ "step": 690
501
+ },
502
+ {
503
+ "epoch": 1.596011396011396,
504
+ "grad_norm": 0.2864850746661905,
505
+ "learning_rate": 5e-06,
506
+ "loss": 0.6101,
507
+ "step": 700
508
+ },
509
+ {
510
+ "epoch": 1.6188034188034188,
511
+ "grad_norm": 0.3076750266232437,
512
+ "learning_rate": 5e-06,
513
+ "loss": 0.5971,
514
+ "step": 710
515
+ },
516
+ {
517
+ "epoch": 1.6415954415954417,
518
+ "grad_norm": 0.29013113513057376,
519
+ "learning_rate": 5e-06,
520
+ "loss": 0.6017,
521
+ "step": 720
522
+ },
523
+ {
524
+ "epoch": 1.6643874643874645,
525
+ "grad_norm": 0.32853478390443247,
526
+ "learning_rate": 5e-06,
527
+ "loss": 0.5948,
528
+ "step": 730
529
+ },
530
+ {
531
+ "epoch": 1.6871794871794872,
532
+ "grad_norm": 0.3212067064463354,
533
+ "learning_rate": 5e-06,
534
+ "loss": 0.5959,
535
+ "step": 740
536
+ },
537
+ {
538
+ "epoch": 1.7099715099715098,
539
+ "grad_norm": 0.2972760971170388,
540
+ "learning_rate": 5e-06,
541
+ "loss": 0.5933,
542
+ "step": 750
543
+ },
544
+ {
545
+ "epoch": 1.7327635327635327,
546
+ "grad_norm": 0.3094250761102705,
547
+ "learning_rate": 5e-06,
548
+ "loss": 0.5972,
549
+ "step": 760
550
+ },
551
+ {
552
+ "epoch": 1.7555555555555555,
553
+ "grad_norm": 0.2982342821177231,
554
+ "learning_rate": 5e-06,
555
+ "loss": 0.5936,
556
+ "step": 770
557
+ },
558
+ {
559
+ "epoch": 1.7783475783475784,
560
+ "grad_norm": 0.30920000199984904,
561
+ "learning_rate": 5e-06,
562
+ "loss": 0.6008,
563
+ "step": 780
564
+ },
565
+ {
566
+ "epoch": 1.8011396011396013,
567
+ "grad_norm": 0.2949027515395575,
568
+ "learning_rate": 5e-06,
569
+ "loss": 0.5984,
570
+ "step": 790
571
+ },
572
+ {
573
+ "epoch": 1.823931623931624,
574
+ "grad_norm": 0.31043498815582843,
575
+ "learning_rate": 5e-06,
576
+ "loss": 0.5851,
577
+ "step": 800
578
+ },
579
+ {
580
+ "epoch": 1.8467236467236468,
581
+ "grad_norm": 0.30810688172811324,
582
+ "learning_rate": 5e-06,
583
+ "loss": 0.593,
584
+ "step": 810
585
+ },
586
+ {
587
+ "epoch": 1.8695156695156694,
588
+ "grad_norm": 0.30623595909755325,
589
+ "learning_rate": 5e-06,
590
+ "loss": 0.5908,
591
+ "step": 820
592
+ },
593
+ {
594
+ "epoch": 1.8923076923076922,
595
+ "grad_norm": 0.327188525053139,
596
+ "learning_rate": 5e-06,
597
+ "loss": 0.5946,
598
+ "step": 830
599
+ },
600
+ {
601
+ "epoch": 1.915099715099715,
602
+ "grad_norm": 0.310903558460967,
603
+ "learning_rate": 5e-06,
604
+ "loss": 0.599,
605
+ "step": 840
606
+ },
607
+ {
608
+ "epoch": 1.937891737891738,
609
+ "grad_norm": 0.3021187702467701,
610
+ "learning_rate": 5e-06,
611
+ "loss": 0.5972,
612
+ "step": 850
613
+ },
614
+ {
615
+ "epoch": 1.9606837606837608,
616
+ "grad_norm": 0.3108930383190649,
617
+ "learning_rate": 5e-06,
618
+ "loss": 0.5985,
619
+ "step": 860
620
+ },
621
+ {
622
+ "epoch": 1.9834757834757835,
623
+ "grad_norm": 0.3507405640216123,
624
+ "learning_rate": 5e-06,
625
+ "loss": 0.5959,
626
+ "step": 870
627
+ },
628
+ {
629
+ "epoch": 1.9994301994301993,
630
+ "eval_loss": 0.6187016367912292,
631
+ "eval_runtime": 440.3195,
632
+ "eval_samples_per_second": 26.851,
633
+ "eval_steps_per_second": 0.42,
634
+ "step": 877
635
+ },
636
+ {
637
+ "epoch": 2.006837606837607,
638
+ "grad_norm": 0.33247529291148487,
639
+ "learning_rate": 5e-06,
640
+ "loss": 0.6153,
641
+ "step": 880
642
+ },
643
+ {
644
+ "epoch": 2.0296296296296297,
645
+ "grad_norm": 0.35035702544183833,
646
+ "learning_rate": 5e-06,
647
+ "loss": 0.5544,
648
+ "step": 890
649
+ },
650
+ {
651
+ "epoch": 2.0524216524216525,
652
+ "grad_norm": 0.3079737428410421,
653
+ "learning_rate": 5e-06,
654
+ "loss": 0.5695,
655
+ "step": 900
656
+ },
657
+ {
658
+ "epoch": 2.0752136752136754,
659
+ "grad_norm": 0.3546826206246178,
660
+ "learning_rate": 5e-06,
661
+ "loss": 0.5676,
662
+ "step": 910
663
+ },
664
+ {
665
+ "epoch": 2.098005698005698,
666
+ "grad_norm": 0.33546462432559726,
667
+ "learning_rate": 5e-06,
668
+ "loss": 0.5708,
669
+ "step": 920
670
+ },
671
+ {
672
+ "epoch": 2.1207977207977207,
673
+ "grad_norm": 0.28495486013565147,
674
+ "learning_rate": 5e-06,
675
+ "loss": 0.5688,
676
+ "step": 930
677
+ },
678
+ {
679
+ "epoch": 2.1435897435897435,
680
+ "grad_norm": 0.2994932879877907,
681
+ "learning_rate": 5e-06,
682
+ "loss": 0.5635,
683
+ "step": 940
684
+ },
685
+ {
686
+ "epoch": 2.1663817663817664,
687
+ "grad_norm": 0.304625525109529,
688
+ "learning_rate": 5e-06,
689
+ "loss": 0.5643,
690
+ "step": 950
691
+ },
692
+ {
693
+ "epoch": 2.1891737891737892,
694
+ "grad_norm": 0.31522776479984477,
695
+ "learning_rate": 5e-06,
696
+ "loss": 0.5694,
697
+ "step": 960
698
+ },
699
+ {
700
+ "epoch": 2.211965811965812,
701
+ "grad_norm": 0.2935531068982711,
702
+ "learning_rate": 5e-06,
703
+ "loss": 0.5689,
704
+ "step": 970
705
+ },
706
+ {
707
+ "epoch": 2.234757834757835,
708
+ "grad_norm": 0.28633664929312297,
709
+ "learning_rate": 5e-06,
710
+ "loss": 0.5585,
711
+ "step": 980
712
+ },
713
+ {
714
+ "epoch": 2.2575498575498574,
715
+ "grad_norm": 0.323299232881518,
716
+ "learning_rate": 5e-06,
717
+ "loss": 0.5676,
718
+ "step": 990
719
+ },
720
+ {
721
+ "epoch": 2.2803418803418802,
722
+ "grad_norm": 0.3023262779587687,
723
+ "learning_rate": 5e-06,
724
+ "loss": 0.5679,
725
+ "step": 1000
726
+ },
727
+ {
728
+ "epoch": 2.303133903133903,
729
+ "grad_norm": 0.32932247022115296,
730
+ "learning_rate": 5e-06,
731
+ "loss": 0.5654,
732
+ "step": 1010
733
+ },
734
+ {
735
+ "epoch": 2.325925925925926,
736
+ "grad_norm": 0.3096111275725531,
737
+ "learning_rate": 5e-06,
738
+ "loss": 0.5684,
739
+ "step": 1020
740
+ },
741
+ {
742
+ "epoch": 2.348717948717949,
743
+ "grad_norm": 0.35230664675988677,
744
+ "learning_rate": 5e-06,
745
+ "loss": 0.5697,
746
+ "step": 1030
747
+ },
748
+ {
749
+ "epoch": 2.3715099715099717,
750
+ "grad_norm": 0.2969558547430912,
751
+ "learning_rate": 5e-06,
752
+ "loss": 0.5677,
753
+ "step": 1040
754
+ },
755
+ {
756
+ "epoch": 2.394301994301994,
757
+ "grad_norm": 0.30518464062660466,
758
+ "learning_rate": 5e-06,
759
+ "loss": 0.5671,
760
+ "step": 1050
761
+ },
762
+ {
763
+ "epoch": 2.417094017094017,
764
+ "grad_norm": 0.2978848500767363,
765
+ "learning_rate": 5e-06,
766
+ "loss": 0.5754,
767
+ "step": 1060
768
+ },
769
+ {
770
+ "epoch": 2.43988603988604,
771
+ "grad_norm": 0.3003352203902967,
772
+ "learning_rate": 5e-06,
773
+ "loss": 0.5782,
774
+ "step": 1070
775
+ },
776
+ {
777
+ "epoch": 2.4626780626780627,
778
+ "grad_norm": 0.306630876746365,
779
+ "learning_rate": 5e-06,
780
+ "loss": 0.5735,
781
+ "step": 1080
782
+ },
783
+ {
784
+ "epoch": 2.4854700854700855,
785
+ "grad_norm": 0.30337205010612617,
786
+ "learning_rate": 5e-06,
787
+ "loss": 0.5711,
788
+ "step": 1090
789
+ },
790
+ {
791
+ "epoch": 2.5082621082621084,
792
+ "grad_norm": 0.3036174952543044,
793
+ "learning_rate": 5e-06,
794
+ "loss": 0.5776,
795
+ "step": 1100
796
+ },
797
+ {
798
+ "epoch": 2.5310541310541312,
799
+ "grad_norm": 0.3097038070254852,
800
+ "learning_rate": 5e-06,
801
+ "loss": 0.5711,
802
+ "step": 1110
803
+ },
804
+ {
805
+ "epoch": 2.5538461538461537,
806
+ "grad_norm": 0.3055259061973633,
807
+ "learning_rate": 5e-06,
808
+ "loss": 0.5692,
809
+ "step": 1120
810
+ },
811
+ {
812
+ "epoch": 2.5766381766381765,
813
+ "grad_norm": 0.2978967073586066,
814
+ "learning_rate": 5e-06,
815
+ "loss": 0.5687,
816
+ "step": 1130
817
+ },
818
+ {
819
+ "epoch": 2.5994301994301994,
820
+ "grad_norm": 0.3134626755759451,
821
+ "learning_rate": 5e-06,
822
+ "loss": 0.5683,
823
+ "step": 1140
824
+ },
825
+ {
826
+ "epoch": 2.6222222222222222,
827
+ "grad_norm": 0.30526868243301847,
828
+ "learning_rate": 5e-06,
829
+ "loss": 0.5686,
830
+ "step": 1150
831
+ },
832
+ {
833
+ "epoch": 2.645014245014245,
834
+ "grad_norm": 0.3094315211429326,
835
+ "learning_rate": 5e-06,
836
+ "loss": 0.5698,
837
+ "step": 1160
838
+ },
839
+ {
840
+ "epoch": 2.667806267806268,
841
+ "grad_norm": 0.2979337120230037,
842
+ "learning_rate": 5e-06,
843
+ "loss": 0.5726,
844
+ "step": 1170
845
+ },
846
+ {
847
+ "epoch": 2.690598290598291,
848
+ "grad_norm": 0.29617078222655224,
849
+ "learning_rate": 5e-06,
850
+ "loss": 0.5677,
851
+ "step": 1180
852
+ },
853
+ {
854
+ "epoch": 2.7133903133903132,
855
+ "grad_norm": 0.2874153849084469,
856
+ "learning_rate": 5e-06,
857
+ "loss": 0.5687,
858
+ "step": 1190
859
+ },
860
+ {
861
+ "epoch": 2.736182336182336,
862
+ "grad_norm": 0.2882166610791341,
863
+ "learning_rate": 5e-06,
864
+ "loss": 0.5719,
865
+ "step": 1200
866
+ },
867
+ {
868
+ "epoch": 2.758974358974359,
869
+ "grad_norm": 0.29078138854804286,
870
+ "learning_rate": 5e-06,
871
+ "loss": 0.5753,
872
+ "step": 1210
873
+ },
874
+ {
875
+ "epoch": 2.781766381766382,
876
+ "grad_norm": 0.2866124714656832,
877
+ "learning_rate": 5e-06,
878
+ "loss": 0.5652,
879
+ "step": 1220
880
+ },
881
+ {
882
+ "epoch": 2.8045584045584047,
883
+ "grad_norm": 0.2924235716239258,
884
+ "learning_rate": 5e-06,
885
+ "loss": 0.5637,
886
+ "step": 1230
887
+ },
888
+ {
889
+ "epoch": 2.827350427350427,
890
+ "grad_norm": 0.3070438197451834,
891
+ "learning_rate": 5e-06,
892
+ "loss": 0.5784,
893
+ "step": 1240
894
+ },
895
+ {
896
+ "epoch": 2.8501424501424504,
897
+ "grad_norm": 0.3055107717692077,
898
+ "learning_rate": 5e-06,
899
+ "loss": 0.5751,
900
+ "step": 1250
901
+ },
902
+ {
903
+ "epoch": 2.872934472934473,
904
+ "grad_norm": 0.3021077678301008,
905
+ "learning_rate": 5e-06,
906
+ "loss": 0.5723,
907
+ "step": 1260
908
+ },
909
+ {
910
+ "epoch": 2.8957264957264957,
911
+ "grad_norm": 0.30622160394298603,
912
+ "learning_rate": 5e-06,
913
+ "loss": 0.5732,
914
+ "step": 1270
915
+ },
916
+ {
917
+ "epoch": 2.9185185185185185,
918
+ "grad_norm": 0.28670420364394134,
919
+ "learning_rate": 5e-06,
920
+ "loss": 0.5634,
921
+ "step": 1280
922
+ },
923
+ {
924
+ "epoch": 2.9413105413105414,
925
+ "grad_norm": 0.2968532310336153,
926
+ "learning_rate": 5e-06,
927
+ "loss": 0.5636,
928
+ "step": 1290
929
+ },
930
+ {
931
+ "epoch": 2.9641025641025642,
932
+ "grad_norm": 0.35184113878950307,
933
+ "learning_rate": 5e-06,
934
+ "loss": 0.5648,
935
+ "step": 1300
936
+ },
937
+ {
938
+ "epoch": 2.9868945868945866,
939
+ "grad_norm": 0.3103099246416176,
940
+ "learning_rate": 5e-06,
941
+ "loss": 0.5746,
942
+ "step": 1310
943
+ },
944
+ {
945
+ "epoch": 2.996011396011396,
946
+ "eval_loss": 0.6186709403991699,
947
+ "eval_runtime": 440.9792,
948
+ "eval_samples_per_second": 26.811,
949
+ "eval_steps_per_second": 0.42,
950
+ "step": 1314
951
+ },
952
+ {
953
+ "epoch": 2.996011396011396,
954
+ "step": 1314,
955
+ "total_flos": 2755219238682624.0,
956
+ "train_loss": 0.6038080361335789,
957
+ "train_runtime": 70053.326,
958
+ "train_samples_per_second": 9.619,
959
+ "train_steps_per_second": 0.019
960
+ }
961
+ ],
962
+ "logging_steps": 10,
963
+ "max_steps": 1314,
964
+ "num_input_tokens_seen": 0,
965
+ "num_train_epochs": 3,
966
+ "save_steps": 500,
967
+ "stateful_callbacks": {
968
+ "TrainerControl": {
969
+ "args": {
970
+ "should_epoch_stop": false,
971
+ "should_evaluate": false,
972
+ "should_log": false,
973
+ "should_save": true,
974
+ "should_training_stop": true
975
+ },
976
+ "attributes": {}
977
+ }
978
+ },
979
+ "total_flos": 2755219238682624.0,
980
+ "train_batch_size": 8,
981
+ "trial_name": null,
982
+ "trial_params": null
983
+ }
training_eval_loss.png ADDED
training_loss.png ADDED