sedrickkeh commited on
Commit
8ff744b
1 Parent(s): 5c9cc7f

End of training

Browse files
README.md CHANGED
@@ -4,6 +4,7 @@ license: gemma
4
  base_model: google/gemma-2-9b
5
  tags:
6
  - llama-factory
 
7
  - generated_from_trainer
8
  model-index:
9
  - name: hp_ablations_gemma_adambeta2_0.9995
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
15
 
16
  # hp_ablations_gemma_adambeta2_0.9995
17
 
18
- This model is a fine-tuned version of [google/gemma-2-9b](https://huggingface.co/google/gemma-2-9b) on an unknown dataset.
19
  It achieves the following results on the evaluation set:
20
  - Loss: 0.6002
21
 
 
4
  base_model: google/gemma-2-9b
5
  tags:
6
  - llama-factory
7
+ - full
8
  - generated_from_trainer
9
  model-index:
10
  - name: hp_ablations_gemma_adambeta2_0.9995
 
16
 
17
  # hp_ablations_gemma_adambeta2_0.9995
18
 
19
+ This model is a fine-tuned version of [google/gemma-2-9b](https://huggingface.co/google/gemma-2-9b) on the mlfoundations-dev/oh-dcft-v3.1-gpt-4o-mini dataset.
20
  It achieves the following results on the evaluation set:
21
  - Loss: 0.6002
22
 
all_results.json ADDED
@@ -0,0 +1,12 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.9991537376586743,
3
+ "eval_loss": 0.6002166867256165,
4
+ "eval_runtime": 696.9645,
5
+ "eval_samples_per_second": 17.133,
6
+ "eval_steps_per_second": 0.537,
7
+ "total_flos": 5064195066298368.0,
8
+ "train_loss": 0.5481024000140579,
9
+ "train_runtime": 121980.0389,
10
+ "train_samples_per_second": 5.58,
11
+ "train_steps_per_second": 0.011
12
+ }
eval_results.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.9991537376586743,
3
+ "eval_loss": 0.6002166867256165,
4
+ "eval_runtime": 696.9645,
5
+ "eval_samples_per_second": 17.133,
6
+ "eval_steps_per_second": 0.537
7
+ }
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.9991537376586743,
3
+ "total_flos": 5064195066298368.0,
4
+ "train_loss": 0.5481024000140579,
5
+ "train_runtime": 121980.0389,
6
+ "train_samples_per_second": 5.58,
7
+ "train_steps_per_second": 0.011
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,990 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 2.9991537376586743,
5
+ "eval_steps": 500,
6
+ "global_step": 1329,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.022566995768688293,
13
+ "grad_norm": 1.1089675466467734,
14
+ "learning_rate": 5e-06,
15
+ "loss": 0.7442,
16
+ "step": 10
17
+ },
18
+ {
19
+ "epoch": 0.045133991537376586,
20
+ "grad_norm": 0.9032651687709552,
21
+ "learning_rate": 5e-06,
22
+ "loss": 0.6806,
23
+ "step": 20
24
+ },
25
+ {
26
+ "epoch": 0.06770098730606489,
27
+ "grad_norm": 0.661414123301935,
28
+ "learning_rate": 5e-06,
29
+ "loss": 0.663,
30
+ "step": 30
31
+ },
32
+ {
33
+ "epoch": 0.09026798307475317,
34
+ "grad_norm": 0.684191997191131,
35
+ "learning_rate": 5e-06,
36
+ "loss": 0.6404,
37
+ "step": 40
38
+ },
39
+ {
40
+ "epoch": 0.11283497884344147,
41
+ "grad_norm": 0.7968850136147834,
42
+ "learning_rate": 5e-06,
43
+ "loss": 0.6338,
44
+ "step": 50
45
+ },
46
+ {
47
+ "epoch": 0.13540197461212977,
48
+ "grad_norm": 0.6025980511091419,
49
+ "learning_rate": 5e-06,
50
+ "loss": 0.6332,
51
+ "step": 60
52
+ },
53
+ {
54
+ "epoch": 0.15796897038081806,
55
+ "grad_norm": 0.6896440149282623,
56
+ "learning_rate": 5e-06,
57
+ "loss": 0.629,
58
+ "step": 70
59
+ },
60
+ {
61
+ "epoch": 0.18053596614950634,
62
+ "grad_norm": 0.6167482460724676,
63
+ "learning_rate": 5e-06,
64
+ "loss": 0.6305,
65
+ "step": 80
66
+ },
67
+ {
68
+ "epoch": 0.20310296191819463,
69
+ "grad_norm": 0.7092392264057279,
70
+ "learning_rate": 5e-06,
71
+ "loss": 0.6283,
72
+ "step": 90
73
+ },
74
+ {
75
+ "epoch": 0.22566995768688294,
76
+ "grad_norm": 0.6757765981920996,
77
+ "learning_rate": 5e-06,
78
+ "loss": 0.6274,
79
+ "step": 100
80
+ },
81
+ {
82
+ "epoch": 0.24823695345557123,
83
+ "grad_norm": 0.602546276691872,
84
+ "learning_rate": 5e-06,
85
+ "loss": 0.6181,
86
+ "step": 110
87
+ },
88
+ {
89
+ "epoch": 0.27080394922425954,
90
+ "grad_norm": 0.6184862063130019,
91
+ "learning_rate": 5e-06,
92
+ "loss": 0.6173,
93
+ "step": 120
94
+ },
95
+ {
96
+ "epoch": 0.2933709449929478,
97
+ "grad_norm": 0.9212523299186395,
98
+ "learning_rate": 5e-06,
99
+ "loss": 0.6186,
100
+ "step": 130
101
+ },
102
+ {
103
+ "epoch": 0.3159379407616361,
104
+ "grad_norm": 0.6085522366969077,
105
+ "learning_rate": 5e-06,
106
+ "loss": 0.6134,
107
+ "step": 140
108
+ },
109
+ {
110
+ "epoch": 0.3385049365303244,
111
+ "grad_norm": 0.6256921238889772,
112
+ "learning_rate": 5e-06,
113
+ "loss": 0.6141,
114
+ "step": 150
115
+ },
116
+ {
117
+ "epoch": 0.3610719322990127,
118
+ "grad_norm": 0.606645633530053,
119
+ "learning_rate": 5e-06,
120
+ "loss": 0.6158,
121
+ "step": 160
122
+ },
123
+ {
124
+ "epoch": 0.383638928067701,
125
+ "grad_norm": 0.6401613005630999,
126
+ "learning_rate": 5e-06,
127
+ "loss": 0.6118,
128
+ "step": 170
129
+ },
130
+ {
131
+ "epoch": 0.40620592383638926,
132
+ "grad_norm": 0.6243251601570871,
133
+ "learning_rate": 5e-06,
134
+ "loss": 0.6088,
135
+ "step": 180
136
+ },
137
+ {
138
+ "epoch": 0.4287729196050776,
139
+ "grad_norm": 0.6242404065068139,
140
+ "learning_rate": 5e-06,
141
+ "loss": 0.6034,
142
+ "step": 190
143
+ },
144
+ {
145
+ "epoch": 0.4513399153737659,
146
+ "grad_norm": 0.6121729700895588,
147
+ "learning_rate": 5e-06,
148
+ "loss": 0.6099,
149
+ "step": 200
150
+ },
151
+ {
152
+ "epoch": 0.47390691114245415,
153
+ "grad_norm": 0.5969802976661266,
154
+ "learning_rate": 5e-06,
155
+ "loss": 0.6063,
156
+ "step": 210
157
+ },
158
+ {
159
+ "epoch": 0.49647390691114246,
160
+ "grad_norm": 0.6129982980567859,
161
+ "learning_rate": 5e-06,
162
+ "loss": 0.598,
163
+ "step": 220
164
+ },
165
+ {
166
+ "epoch": 0.5190409026798307,
167
+ "grad_norm": 0.5668533739328568,
168
+ "learning_rate": 5e-06,
169
+ "loss": 0.6068,
170
+ "step": 230
171
+ },
172
+ {
173
+ "epoch": 0.5416078984485191,
174
+ "grad_norm": 0.7232774678335098,
175
+ "learning_rate": 5e-06,
176
+ "loss": 0.6057,
177
+ "step": 240
178
+ },
179
+ {
180
+ "epoch": 0.5641748942172073,
181
+ "grad_norm": 0.6039422383722364,
182
+ "learning_rate": 5e-06,
183
+ "loss": 0.5998,
184
+ "step": 250
185
+ },
186
+ {
187
+ "epoch": 0.5867418899858956,
188
+ "grad_norm": 0.7087899032677945,
189
+ "learning_rate": 5e-06,
190
+ "loss": 0.6025,
191
+ "step": 260
192
+ },
193
+ {
194
+ "epoch": 0.609308885754584,
195
+ "grad_norm": 0.553100874749091,
196
+ "learning_rate": 5e-06,
197
+ "loss": 0.5979,
198
+ "step": 270
199
+ },
200
+ {
201
+ "epoch": 0.6318758815232722,
202
+ "grad_norm": 0.66241431256307,
203
+ "learning_rate": 5e-06,
204
+ "loss": 0.6046,
205
+ "step": 280
206
+ },
207
+ {
208
+ "epoch": 0.6544428772919605,
209
+ "grad_norm": 0.832301067216732,
210
+ "learning_rate": 5e-06,
211
+ "loss": 0.598,
212
+ "step": 290
213
+ },
214
+ {
215
+ "epoch": 0.6770098730606487,
216
+ "grad_norm": 0.9576819834043614,
217
+ "learning_rate": 5e-06,
218
+ "loss": 0.6042,
219
+ "step": 300
220
+ },
221
+ {
222
+ "epoch": 0.6995768688293371,
223
+ "grad_norm": 0.560752931056728,
224
+ "learning_rate": 5e-06,
225
+ "loss": 0.5973,
226
+ "step": 310
227
+ },
228
+ {
229
+ "epoch": 0.7221438645980254,
230
+ "grad_norm": 0.5478632639125564,
231
+ "learning_rate": 5e-06,
232
+ "loss": 0.6005,
233
+ "step": 320
234
+ },
235
+ {
236
+ "epoch": 0.7447108603667136,
237
+ "grad_norm": 0.5604724061444557,
238
+ "learning_rate": 5e-06,
239
+ "loss": 0.5962,
240
+ "step": 330
241
+ },
242
+ {
243
+ "epoch": 0.767277856135402,
244
+ "grad_norm": 0.5858187723986242,
245
+ "learning_rate": 5e-06,
246
+ "loss": 0.5956,
247
+ "step": 340
248
+ },
249
+ {
250
+ "epoch": 0.7898448519040903,
251
+ "grad_norm": 0.6186940151586905,
252
+ "learning_rate": 5e-06,
253
+ "loss": 0.597,
254
+ "step": 350
255
+ },
256
+ {
257
+ "epoch": 0.8124118476727785,
258
+ "grad_norm": 0.580787958623627,
259
+ "learning_rate": 5e-06,
260
+ "loss": 0.5959,
261
+ "step": 360
262
+ },
263
+ {
264
+ "epoch": 0.8349788434414669,
265
+ "grad_norm": 0.6446875040901882,
266
+ "learning_rate": 5e-06,
267
+ "loss": 0.6031,
268
+ "step": 370
269
+ },
270
+ {
271
+ "epoch": 0.8575458392101551,
272
+ "grad_norm": 0.5607098113086376,
273
+ "learning_rate": 5e-06,
274
+ "loss": 0.5997,
275
+ "step": 380
276
+ },
277
+ {
278
+ "epoch": 0.8801128349788434,
279
+ "grad_norm": 0.6331077663206254,
280
+ "learning_rate": 5e-06,
281
+ "loss": 0.5936,
282
+ "step": 390
283
+ },
284
+ {
285
+ "epoch": 0.9026798307475318,
286
+ "grad_norm": 0.5922081888201195,
287
+ "learning_rate": 5e-06,
288
+ "loss": 0.596,
289
+ "step": 400
290
+ },
291
+ {
292
+ "epoch": 0.92524682651622,
293
+ "grad_norm": 0.5930063432040521,
294
+ "learning_rate": 5e-06,
295
+ "loss": 0.5941,
296
+ "step": 410
297
+ },
298
+ {
299
+ "epoch": 0.9478138222849083,
300
+ "grad_norm": 0.5819547842258083,
301
+ "learning_rate": 5e-06,
302
+ "loss": 0.5946,
303
+ "step": 420
304
+ },
305
+ {
306
+ "epoch": 0.9703808180535967,
307
+ "grad_norm": 0.6092660320439263,
308
+ "learning_rate": 5e-06,
309
+ "loss": 0.5886,
310
+ "step": 430
311
+ },
312
+ {
313
+ "epoch": 0.9929478138222849,
314
+ "grad_norm": 0.7102470041653363,
315
+ "learning_rate": 5e-06,
316
+ "loss": 0.5934,
317
+ "step": 440
318
+ },
319
+ {
320
+ "epoch": 0.9997179125528914,
321
+ "eval_loss": 0.5912763476371765,
322
+ "eval_runtime": 697.555,
323
+ "eval_samples_per_second": 17.118,
324
+ "eval_steps_per_second": 0.536,
325
+ "step": 443
326
+ },
327
+ {
328
+ "epoch": 1.0155148095909732,
329
+ "grad_norm": 0.7217656578243299,
330
+ "learning_rate": 5e-06,
331
+ "loss": 0.6009,
332
+ "step": 450
333
+ },
334
+ {
335
+ "epoch": 1.0380818053596614,
336
+ "grad_norm": 0.5922099753160804,
337
+ "learning_rate": 5e-06,
338
+ "loss": 0.5282,
339
+ "step": 460
340
+ },
341
+ {
342
+ "epoch": 1.0606488011283497,
343
+ "grad_norm": 0.6407127507440616,
344
+ "learning_rate": 5e-06,
345
+ "loss": 0.543,
346
+ "step": 470
347
+ },
348
+ {
349
+ "epoch": 1.0832157968970382,
350
+ "grad_norm": 0.576580888488731,
351
+ "learning_rate": 5e-06,
352
+ "loss": 0.5308,
353
+ "step": 480
354
+ },
355
+ {
356
+ "epoch": 1.1057827926657264,
357
+ "grad_norm": 0.6154989306400179,
358
+ "learning_rate": 5e-06,
359
+ "loss": 0.5411,
360
+ "step": 490
361
+ },
362
+ {
363
+ "epoch": 1.1283497884344147,
364
+ "grad_norm": 0.5803110773211014,
365
+ "learning_rate": 5e-06,
366
+ "loss": 0.5399,
367
+ "step": 500
368
+ },
369
+ {
370
+ "epoch": 1.150916784203103,
371
+ "grad_norm": 0.6165899465430926,
372
+ "learning_rate": 5e-06,
373
+ "loss": 0.5411,
374
+ "step": 510
375
+ },
376
+ {
377
+ "epoch": 1.1734837799717912,
378
+ "grad_norm": 0.5838143629087249,
379
+ "learning_rate": 5e-06,
380
+ "loss": 0.5336,
381
+ "step": 520
382
+ },
383
+ {
384
+ "epoch": 1.1960507757404795,
385
+ "grad_norm": 0.6587792003103263,
386
+ "learning_rate": 5e-06,
387
+ "loss": 0.5399,
388
+ "step": 530
389
+ },
390
+ {
391
+ "epoch": 1.2186177715091677,
392
+ "grad_norm": 0.5699038771851571,
393
+ "learning_rate": 5e-06,
394
+ "loss": 0.5373,
395
+ "step": 540
396
+ },
397
+ {
398
+ "epoch": 1.2411847672778562,
399
+ "grad_norm": 0.6061015503667232,
400
+ "learning_rate": 5e-06,
401
+ "loss": 0.5421,
402
+ "step": 550
403
+ },
404
+ {
405
+ "epoch": 1.2637517630465445,
406
+ "grad_norm": 0.5832963025889784,
407
+ "learning_rate": 5e-06,
408
+ "loss": 0.544,
409
+ "step": 560
410
+ },
411
+ {
412
+ "epoch": 1.2863187588152327,
413
+ "grad_norm": 0.5816672164385919,
414
+ "learning_rate": 5e-06,
415
+ "loss": 0.5412,
416
+ "step": 570
417
+ },
418
+ {
419
+ "epoch": 1.308885754583921,
420
+ "grad_norm": 0.5590405534898907,
421
+ "learning_rate": 5e-06,
422
+ "loss": 0.5411,
423
+ "step": 580
424
+ },
425
+ {
426
+ "epoch": 1.3314527503526092,
427
+ "grad_norm": 0.5836929150351833,
428
+ "learning_rate": 5e-06,
429
+ "loss": 0.5411,
430
+ "step": 590
431
+ },
432
+ {
433
+ "epoch": 1.3540197461212977,
434
+ "grad_norm": 0.6281841450404138,
435
+ "learning_rate": 5e-06,
436
+ "loss": 0.5464,
437
+ "step": 600
438
+ },
439
+ {
440
+ "epoch": 1.376586741889986,
441
+ "grad_norm": 0.8123084972654747,
442
+ "learning_rate": 5e-06,
443
+ "loss": 0.5391,
444
+ "step": 610
445
+ },
446
+ {
447
+ "epoch": 1.3991537376586742,
448
+ "grad_norm": 0.5793145395265858,
449
+ "learning_rate": 5e-06,
450
+ "loss": 0.5434,
451
+ "step": 620
452
+ },
453
+ {
454
+ "epoch": 1.4217207334273625,
455
+ "grad_norm": 0.5812729927118028,
456
+ "learning_rate": 5e-06,
457
+ "loss": 0.5415,
458
+ "step": 630
459
+ },
460
+ {
461
+ "epoch": 1.4442877291960508,
462
+ "grad_norm": 0.6664458859786708,
463
+ "learning_rate": 5e-06,
464
+ "loss": 0.5412,
465
+ "step": 640
466
+ },
467
+ {
468
+ "epoch": 1.466854724964739,
469
+ "grad_norm": 0.5647691806534823,
470
+ "learning_rate": 5e-06,
471
+ "loss": 0.5421,
472
+ "step": 650
473
+ },
474
+ {
475
+ "epoch": 1.4894217207334273,
476
+ "grad_norm": 0.6261523442650948,
477
+ "learning_rate": 5e-06,
478
+ "loss": 0.5459,
479
+ "step": 660
480
+ },
481
+ {
482
+ "epoch": 1.5119887165021155,
483
+ "grad_norm": 0.5871251244408628,
484
+ "learning_rate": 5e-06,
485
+ "loss": 0.5405,
486
+ "step": 670
487
+ },
488
+ {
489
+ "epoch": 1.5345557122708038,
490
+ "grad_norm": 0.5429570025219707,
491
+ "learning_rate": 5e-06,
492
+ "loss": 0.5346,
493
+ "step": 680
494
+ },
495
+ {
496
+ "epoch": 1.5571227080394923,
497
+ "grad_norm": 0.6033700862868088,
498
+ "learning_rate": 5e-06,
499
+ "loss": 0.5368,
500
+ "step": 690
501
+ },
502
+ {
503
+ "epoch": 1.5796897038081805,
504
+ "grad_norm": 0.6119780310178549,
505
+ "learning_rate": 5e-06,
506
+ "loss": 0.5411,
507
+ "step": 700
508
+ },
509
+ {
510
+ "epoch": 1.6022566995768688,
511
+ "grad_norm": 0.5711515525605414,
512
+ "learning_rate": 5e-06,
513
+ "loss": 0.5438,
514
+ "step": 710
515
+ },
516
+ {
517
+ "epoch": 1.6248236953455573,
518
+ "grad_norm": 0.6045746499276777,
519
+ "learning_rate": 5e-06,
520
+ "loss": 0.5518,
521
+ "step": 720
522
+ },
523
+ {
524
+ "epoch": 1.6473906911142455,
525
+ "grad_norm": 0.5481917881896207,
526
+ "learning_rate": 5e-06,
527
+ "loss": 0.535,
528
+ "step": 730
529
+ },
530
+ {
531
+ "epoch": 1.6699576868829338,
532
+ "grad_norm": 0.6887524559132724,
533
+ "learning_rate": 5e-06,
534
+ "loss": 0.5509,
535
+ "step": 740
536
+ },
537
+ {
538
+ "epoch": 1.692524682651622,
539
+ "grad_norm": 0.6373707446598362,
540
+ "learning_rate": 5e-06,
541
+ "loss": 0.5422,
542
+ "step": 750
543
+ },
544
+ {
545
+ "epoch": 1.7150916784203103,
546
+ "grad_norm": 0.5959637343984598,
547
+ "learning_rate": 5e-06,
548
+ "loss": 0.5411,
549
+ "step": 760
550
+ },
551
+ {
552
+ "epoch": 1.7376586741889986,
553
+ "grad_norm": 0.6663181609027282,
554
+ "learning_rate": 5e-06,
555
+ "loss": 0.5418,
556
+ "step": 770
557
+ },
558
+ {
559
+ "epoch": 1.7602256699576868,
560
+ "grad_norm": 0.6225777365866713,
561
+ "learning_rate": 5e-06,
562
+ "loss": 0.5377,
563
+ "step": 780
564
+ },
565
+ {
566
+ "epoch": 1.782792665726375,
567
+ "grad_norm": 0.5772084554676634,
568
+ "learning_rate": 5e-06,
569
+ "loss": 0.5413,
570
+ "step": 790
571
+ },
572
+ {
573
+ "epoch": 1.8053596614950633,
574
+ "grad_norm": 0.5970786767490915,
575
+ "learning_rate": 5e-06,
576
+ "loss": 0.5363,
577
+ "step": 800
578
+ },
579
+ {
580
+ "epoch": 1.8279266572637518,
581
+ "grad_norm": 0.6556464568292062,
582
+ "learning_rate": 5e-06,
583
+ "loss": 0.547,
584
+ "step": 810
585
+ },
586
+ {
587
+ "epoch": 1.85049365303244,
588
+ "grad_norm": 0.6665921946620136,
589
+ "learning_rate": 5e-06,
590
+ "loss": 0.5344,
591
+ "step": 820
592
+ },
593
+ {
594
+ "epoch": 1.8730606488011283,
595
+ "grad_norm": 0.6146565230270853,
596
+ "learning_rate": 5e-06,
597
+ "loss": 0.5467,
598
+ "step": 830
599
+ },
600
+ {
601
+ "epoch": 1.8956276445698168,
602
+ "grad_norm": 0.5627387129268103,
603
+ "learning_rate": 5e-06,
604
+ "loss": 0.5465,
605
+ "step": 840
606
+ },
607
+ {
608
+ "epoch": 1.918194640338505,
609
+ "grad_norm": 0.5464136766635778,
610
+ "learning_rate": 5e-06,
611
+ "loss": 0.5343,
612
+ "step": 850
613
+ },
614
+ {
615
+ "epoch": 1.9407616361071933,
616
+ "grad_norm": 0.6141386058908015,
617
+ "learning_rate": 5e-06,
618
+ "loss": 0.5394,
619
+ "step": 860
620
+ },
621
+ {
622
+ "epoch": 1.9633286318758816,
623
+ "grad_norm": 0.6464922675747387,
624
+ "learning_rate": 5e-06,
625
+ "loss": 0.5457,
626
+ "step": 870
627
+ },
628
+ {
629
+ "epoch": 1.9858956276445698,
630
+ "grad_norm": 0.6921324613776051,
631
+ "learning_rate": 5e-06,
632
+ "loss": 0.5441,
633
+ "step": 880
634
+ },
635
+ {
636
+ "epoch": 1.9994358251057829,
637
+ "eval_loss": 0.587400496006012,
638
+ "eval_runtime": 695.1905,
639
+ "eval_samples_per_second": 17.177,
640
+ "eval_steps_per_second": 0.538,
641
+ "step": 886
642
+ },
643
+ {
644
+ "epoch": 2.008462623413258,
645
+ "grad_norm": 1.1610399960653068,
646
+ "learning_rate": 5e-06,
647
+ "loss": 0.568,
648
+ "step": 890
649
+ },
650
+ {
651
+ "epoch": 2.0310296191819464,
652
+ "grad_norm": 0.6487581234182378,
653
+ "learning_rate": 5e-06,
654
+ "loss": 0.4787,
655
+ "step": 900
656
+ },
657
+ {
658
+ "epoch": 2.0535966149506346,
659
+ "grad_norm": 0.6328774131189219,
660
+ "learning_rate": 5e-06,
661
+ "loss": 0.4804,
662
+ "step": 910
663
+ },
664
+ {
665
+ "epoch": 2.076163610719323,
666
+ "grad_norm": 0.6041122255586804,
667
+ "learning_rate": 5e-06,
668
+ "loss": 0.4728,
669
+ "step": 920
670
+ },
671
+ {
672
+ "epoch": 2.098730606488011,
673
+ "grad_norm": 0.6042930918406503,
674
+ "learning_rate": 5e-06,
675
+ "loss": 0.4803,
676
+ "step": 930
677
+ },
678
+ {
679
+ "epoch": 2.1212976022566994,
680
+ "grad_norm": 0.6007680878710548,
681
+ "learning_rate": 5e-06,
682
+ "loss": 0.4843,
683
+ "step": 940
684
+ },
685
+ {
686
+ "epoch": 2.143864598025388,
687
+ "grad_norm": 0.5977697264779503,
688
+ "learning_rate": 5e-06,
689
+ "loss": 0.4861,
690
+ "step": 950
691
+ },
692
+ {
693
+ "epoch": 2.1664315937940763,
694
+ "grad_norm": 0.6224632801497896,
695
+ "learning_rate": 5e-06,
696
+ "loss": 0.4824,
697
+ "step": 960
698
+ },
699
+ {
700
+ "epoch": 2.1889985895627646,
701
+ "grad_norm": 0.7298479275194018,
702
+ "learning_rate": 5e-06,
703
+ "loss": 0.4807,
704
+ "step": 970
705
+ },
706
+ {
707
+ "epoch": 2.211565585331453,
708
+ "grad_norm": 0.6192992961102274,
709
+ "learning_rate": 5e-06,
710
+ "loss": 0.476,
711
+ "step": 980
712
+ },
713
+ {
714
+ "epoch": 2.234132581100141,
715
+ "grad_norm": 0.580537121968891,
716
+ "learning_rate": 5e-06,
717
+ "loss": 0.4757,
718
+ "step": 990
719
+ },
720
+ {
721
+ "epoch": 2.2566995768688294,
722
+ "grad_norm": 0.59921408257906,
723
+ "learning_rate": 5e-06,
724
+ "loss": 0.4848,
725
+ "step": 1000
726
+ },
727
+ {
728
+ "epoch": 2.2792665726375176,
729
+ "grad_norm": 0.6091899429354127,
730
+ "learning_rate": 5e-06,
731
+ "loss": 0.491,
732
+ "step": 1010
733
+ },
734
+ {
735
+ "epoch": 2.301833568406206,
736
+ "grad_norm": 0.710504212436636,
737
+ "learning_rate": 5e-06,
738
+ "loss": 0.4853,
739
+ "step": 1020
740
+ },
741
+ {
742
+ "epoch": 2.324400564174894,
743
+ "grad_norm": 0.6579165047877594,
744
+ "learning_rate": 5e-06,
745
+ "loss": 0.4897,
746
+ "step": 1030
747
+ },
748
+ {
749
+ "epoch": 2.3469675599435824,
750
+ "grad_norm": 0.6639077938288619,
751
+ "learning_rate": 5e-06,
752
+ "loss": 0.4875,
753
+ "step": 1040
754
+ },
755
+ {
756
+ "epoch": 2.3695345557122707,
757
+ "grad_norm": 0.6130071384626814,
758
+ "learning_rate": 5e-06,
759
+ "loss": 0.4853,
760
+ "step": 1050
761
+ },
762
+ {
763
+ "epoch": 2.392101551480959,
764
+ "grad_norm": 0.6666570081599825,
765
+ "learning_rate": 5e-06,
766
+ "loss": 0.4848,
767
+ "step": 1060
768
+ },
769
+ {
770
+ "epoch": 2.414668547249647,
771
+ "grad_norm": 0.7611806796886672,
772
+ "learning_rate": 5e-06,
773
+ "loss": 0.4857,
774
+ "step": 1070
775
+ },
776
+ {
777
+ "epoch": 2.4372355430183354,
778
+ "grad_norm": 0.7898919198900027,
779
+ "learning_rate": 5e-06,
780
+ "loss": 0.4891,
781
+ "step": 1080
782
+ },
783
+ {
784
+ "epoch": 2.459802538787024,
785
+ "grad_norm": 0.5773849556240105,
786
+ "learning_rate": 5e-06,
787
+ "loss": 0.4871,
788
+ "step": 1090
789
+ },
790
+ {
791
+ "epoch": 2.4823695345557124,
792
+ "grad_norm": 0.6135977931263356,
793
+ "learning_rate": 5e-06,
794
+ "loss": 0.4902,
795
+ "step": 1100
796
+ },
797
+ {
798
+ "epoch": 2.5049365303244007,
799
+ "grad_norm": 0.594132854203967,
800
+ "learning_rate": 5e-06,
801
+ "loss": 0.4934,
802
+ "step": 1110
803
+ },
804
+ {
805
+ "epoch": 2.527503526093089,
806
+ "grad_norm": 0.6537675061104099,
807
+ "learning_rate": 5e-06,
808
+ "loss": 0.4841,
809
+ "step": 1120
810
+ },
811
+ {
812
+ "epoch": 2.550070521861777,
813
+ "grad_norm": 0.7811576129975868,
814
+ "learning_rate": 5e-06,
815
+ "loss": 0.4891,
816
+ "step": 1130
817
+ },
818
+ {
819
+ "epoch": 2.5726375176304654,
820
+ "grad_norm": 0.5924950779737967,
821
+ "learning_rate": 5e-06,
822
+ "loss": 0.4874,
823
+ "step": 1140
824
+ },
825
+ {
826
+ "epoch": 2.5952045133991537,
827
+ "grad_norm": 0.6437126383257294,
828
+ "learning_rate": 5e-06,
829
+ "loss": 0.4884,
830
+ "step": 1150
831
+ },
832
+ {
833
+ "epoch": 2.617771509167842,
834
+ "grad_norm": 0.618147281886192,
835
+ "learning_rate": 5e-06,
836
+ "loss": 0.4921,
837
+ "step": 1160
838
+ },
839
+ {
840
+ "epoch": 2.64033850493653,
841
+ "grad_norm": 0.6453257647203345,
842
+ "learning_rate": 5e-06,
843
+ "loss": 0.481,
844
+ "step": 1170
845
+ },
846
+ {
847
+ "epoch": 2.6629055007052185,
848
+ "grad_norm": 0.632332824068504,
849
+ "learning_rate": 5e-06,
850
+ "loss": 0.4918,
851
+ "step": 1180
852
+ },
853
+ {
854
+ "epoch": 2.685472496473907,
855
+ "grad_norm": 0.5721042513789194,
856
+ "learning_rate": 5e-06,
857
+ "loss": 0.4925,
858
+ "step": 1190
859
+ },
860
+ {
861
+ "epoch": 2.7080394922425954,
862
+ "grad_norm": 0.6132070533315946,
863
+ "learning_rate": 5e-06,
864
+ "loss": 0.4911,
865
+ "step": 1200
866
+ },
867
+ {
868
+ "epoch": 2.7306064880112837,
869
+ "grad_norm": 0.5673966352972261,
870
+ "learning_rate": 5e-06,
871
+ "loss": 0.4924,
872
+ "step": 1210
873
+ },
874
+ {
875
+ "epoch": 2.753173483779972,
876
+ "grad_norm": 0.5773699546800514,
877
+ "learning_rate": 5e-06,
878
+ "loss": 0.489,
879
+ "step": 1220
880
+ },
881
+ {
882
+ "epoch": 2.77574047954866,
883
+ "grad_norm": 0.5857505263001362,
884
+ "learning_rate": 5e-06,
885
+ "loss": 0.4955,
886
+ "step": 1230
887
+ },
888
+ {
889
+ "epoch": 2.7983074753173485,
890
+ "grad_norm": 0.6473498773975065,
891
+ "learning_rate": 5e-06,
892
+ "loss": 0.4938,
893
+ "step": 1240
894
+ },
895
+ {
896
+ "epoch": 2.8208744710860367,
897
+ "grad_norm": 0.5740719012369875,
898
+ "learning_rate": 5e-06,
899
+ "loss": 0.4941,
900
+ "step": 1250
901
+ },
902
+ {
903
+ "epoch": 2.843441466854725,
904
+ "grad_norm": 0.561559067194783,
905
+ "learning_rate": 5e-06,
906
+ "loss": 0.4958,
907
+ "step": 1260
908
+ },
909
+ {
910
+ "epoch": 2.8660084626234132,
911
+ "grad_norm": 0.5993497581530252,
912
+ "learning_rate": 5e-06,
913
+ "loss": 0.4928,
914
+ "step": 1270
915
+ },
916
+ {
917
+ "epoch": 2.8885754583921015,
918
+ "grad_norm": 0.5741956451302742,
919
+ "learning_rate": 5e-06,
920
+ "loss": 0.4909,
921
+ "step": 1280
922
+ },
923
+ {
924
+ "epoch": 2.9111424541607898,
925
+ "grad_norm": 0.5947983449561617,
926
+ "learning_rate": 5e-06,
927
+ "loss": 0.4946,
928
+ "step": 1290
929
+ },
930
+ {
931
+ "epoch": 2.933709449929478,
932
+ "grad_norm": 0.5946148545340252,
933
+ "learning_rate": 5e-06,
934
+ "loss": 0.4902,
935
+ "step": 1300
936
+ },
937
+ {
938
+ "epoch": 2.9562764456981663,
939
+ "grad_norm": 0.6578162427861818,
940
+ "learning_rate": 5e-06,
941
+ "loss": 0.4939,
942
+ "step": 1310
943
+ },
944
+ {
945
+ "epoch": 2.9788434414668545,
946
+ "grad_norm": 0.575594449194069,
947
+ "learning_rate": 5e-06,
948
+ "loss": 0.4904,
949
+ "step": 1320
950
+ },
951
+ {
952
+ "epoch": 2.9991537376586743,
953
+ "eval_loss": 0.6002166867256165,
954
+ "eval_runtime": 700.2243,
955
+ "eval_samples_per_second": 17.053,
956
+ "eval_steps_per_second": 0.534,
957
+ "step": 1329
958
+ },
959
+ {
960
+ "epoch": 2.9991537376586743,
961
+ "step": 1329,
962
+ "total_flos": 5064195066298368.0,
963
+ "train_loss": 0.5481024000140579,
964
+ "train_runtime": 121980.0389,
965
+ "train_samples_per_second": 5.58,
966
+ "train_steps_per_second": 0.011
967
+ }
968
+ ],
969
+ "logging_steps": 10,
970
+ "max_steps": 1329,
971
+ "num_input_tokens_seen": 0,
972
+ "num_train_epochs": 3,
973
+ "save_steps": 500,
974
+ "stateful_callbacks": {
975
+ "TrainerControl": {
976
+ "args": {
977
+ "should_epoch_stop": false,
978
+ "should_evaluate": false,
979
+ "should_log": false,
980
+ "should_save": true,
981
+ "should_training_stop": true
982
+ },
983
+ "attributes": {}
984
+ }
985
+ },
986
+ "total_flos": 5064195066298368.0,
987
+ "train_batch_size": 4,
988
+ "trial_name": null,
989
+ "trial_params": null
990
+ }
training_eval_loss.png ADDED
training_loss.png ADDED