sedrickkeh commited on
Commit
7e8c455
·
verified ·
1 Parent(s): 71749e4

End of training

Browse files
README.md CHANGED
@@ -4,6 +4,7 @@ license: llama3
4
  base_model: meta-llama/Meta-Llama-3-8B
5
  tags:
6
  - llama-factory
 
7
  - generated_from_trainer
8
  model-index:
9
  - name: OH_DCFT_V3_wo_evol_instruct_140k
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
15
 
16
  # OH_DCFT_V3_wo_evol_instruct_140k
17
 
18
- This model is a fine-tuned version of [meta-llama/Meta-Llama-3-8B](https://huggingface.co/meta-llama/Meta-Llama-3-8B) on an unknown dataset.
19
  It achieves the following results on the evaluation set:
20
  - Loss: 0.6165
21
 
 
4
  base_model: meta-llama/Meta-Llama-3-8B
5
  tags:
6
  - llama-factory
7
+ - full
8
  - generated_from_trainer
9
  model-index:
10
  - name: OH_DCFT_V3_wo_evol_instruct_140k
 
16
 
17
  # OH_DCFT_V3_wo_evol_instruct_140k
18
 
19
+ This model is a fine-tuned version of [meta-llama/Meta-Llama-3-8B](https://huggingface.co/meta-llama/Meta-Llama-3-8B) on the mlfoundations-dev/OH_DCFT_V3_wo_evol_instruct_140k dataset.
20
  It achieves the following results on the evaluation set:
21
  - Loss: 0.6165
22
 
all_results.json ADDED
@@ -0,0 +1,12 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 3.0,
3
+ "eval_loss": 0.6164913177490234,
4
+ "eval_runtime": 203.8494,
5
+ "eval_samples_per_second": 49.836,
6
+ "eval_steps_per_second": 0.392,
7
+ "total_flos": 1894048365281280.0,
8
+ "train_loss": 0.5957063028603713,
9
+ "train_runtime": 34171.7918,
10
+ "train_samples_per_second": 16.944,
11
+ "train_steps_per_second": 0.033
12
+ }
eval_results.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 3.0,
3
+ "eval_loss": 0.6164913177490234,
4
+ "eval_runtime": 203.8494,
5
+ "eval_samples_per_second": 49.836,
6
+ "eval_steps_per_second": 0.392
7
+ }
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 3.0,
3
+ "total_flos": 1894048365281280.0,
4
+ "train_loss": 0.5957063028603713,
5
+ "train_runtime": 34171.7918,
6
+ "train_samples_per_second": 16.944,
7
+ "train_steps_per_second": 0.033
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,857 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 3.0,
5
+ "eval_steps": 500,
6
+ "global_step": 1131,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.026525198938992044,
13
+ "grad_norm": 2.329897814074136,
14
+ "learning_rate": 5e-06,
15
+ "loss": 0.8884,
16
+ "step": 10
17
+ },
18
+ {
19
+ "epoch": 0.05305039787798409,
20
+ "grad_norm": 3.8595726803473838,
21
+ "learning_rate": 5e-06,
22
+ "loss": 0.7961,
23
+ "step": 20
24
+ },
25
+ {
26
+ "epoch": 0.07957559681697612,
27
+ "grad_norm": 1.096843401953888,
28
+ "learning_rate": 5e-06,
29
+ "loss": 0.7478,
30
+ "step": 30
31
+ },
32
+ {
33
+ "epoch": 0.10610079575596817,
34
+ "grad_norm": 1.1211395081081943,
35
+ "learning_rate": 5e-06,
36
+ "loss": 0.719,
37
+ "step": 40
38
+ },
39
+ {
40
+ "epoch": 0.13262599469496023,
41
+ "grad_norm": 1.4282528066175646,
42
+ "learning_rate": 5e-06,
43
+ "loss": 0.7052,
44
+ "step": 50
45
+ },
46
+ {
47
+ "epoch": 0.15915119363395225,
48
+ "grad_norm": 0.7596233206271922,
49
+ "learning_rate": 5e-06,
50
+ "loss": 0.694,
51
+ "step": 60
52
+ },
53
+ {
54
+ "epoch": 0.1856763925729443,
55
+ "grad_norm": 1.2777068139599712,
56
+ "learning_rate": 5e-06,
57
+ "loss": 0.6882,
58
+ "step": 70
59
+ },
60
+ {
61
+ "epoch": 0.21220159151193635,
62
+ "grad_norm": 1.1383877936666695,
63
+ "learning_rate": 5e-06,
64
+ "loss": 0.6771,
65
+ "step": 80
66
+ },
67
+ {
68
+ "epoch": 0.23872679045092837,
69
+ "grad_norm": 1.0060703202699885,
70
+ "learning_rate": 5e-06,
71
+ "loss": 0.6806,
72
+ "step": 90
73
+ },
74
+ {
75
+ "epoch": 0.26525198938992045,
76
+ "grad_norm": 1.0179544972489034,
77
+ "learning_rate": 5e-06,
78
+ "loss": 0.6665,
79
+ "step": 100
80
+ },
81
+ {
82
+ "epoch": 0.2917771883289125,
83
+ "grad_norm": 0.6720815026983992,
84
+ "learning_rate": 5e-06,
85
+ "loss": 0.6637,
86
+ "step": 110
87
+ },
88
+ {
89
+ "epoch": 0.3183023872679045,
90
+ "grad_norm": 0.9241587286794601,
91
+ "learning_rate": 5e-06,
92
+ "loss": 0.6546,
93
+ "step": 120
94
+ },
95
+ {
96
+ "epoch": 0.3448275862068966,
97
+ "grad_norm": 0.8824991950738043,
98
+ "learning_rate": 5e-06,
99
+ "loss": 0.6522,
100
+ "step": 130
101
+ },
102
+ {
103
+ "epoch": 0.3713527851458886,
104
+ "grad_norm": 0.8302884563721783,
105
+ "learning_rate": 5e-06,
106
+ "loss": 0.653,
107
+ "step": 140
108
+ },
109
+ {
110
+ "epoch": 0.3978779840848806,
111
+ "grad_norm": 0.8327701504024553,
112
+ "learning_rate": 5e-06,
113
+ "loss": 0.6455,
114
+ "step": 150
115
+ },
116
+ {
117
+ "epoch": 0.4244031830238727,
118
+ "grad_norm": 0.5492876289580202,
119
+ "learning_rate": 5e-06,
120
+ "loss": 0.6553,
121
+ "step": 160
122
+ },
123
+ {
124
+ "epoch": 0.4509283819628647,
125
+ "grad_norm": 0.7119323938911869,
126
+ "learning_rate": 5e-06,
127
+ "loss": 0.6448,
128
+ "step": 170
129
+ },
130
+ {
131
+ "epoch": 0.47745358090185674,
132
+ "grad_norm": 0.8042997280772343,
133
+ "learning_rate": 5e-06,
134
+ "loss": 0.6403,
135
+ "step": 180
136
+ },
137
+ {
138
+ "epoch": 0.5039787798408488,
139
+ "grad_norm": 0.7810674802506331,
140
+ "learning_rate": 5e-06,
141
+ "loss": 0.655,
142
+ "step": 190
143
+ },
144
+ {
145
+ "epoch": 0.5305039787798409,
146
+ "grad_norm": 0.5391416674289061,
147
+ "learning_rate": 5e-06,
148
+ "loss": 0.6398,
149
+ "step": 200
150
+ },
151
+ {
152
+ "epoch": 0.5570291777188329,
153
+ "grad_norm": 0.6885851666539189,
154
+ "learning_rate": 5e-06,
155
+ "loss": 0.6452,
156
+ "step": 210
157
+ },
158
+ {
159
+ "epoch": 0.583554376657825,
160
+ "grad_norm": 0.6168676023297931,
161
+ "learning_rate": 5e-06,
162
+ "loss": 0.6443,
163
+ "step": 220
164
+ },
165
+ {
166
+ "epoch": 0.610079575596817,
167
+ "grad_norm": 0.5971017602698001,
168
+ "learning_rate": 5e-06,
169
+ "loss": 0.636,
170
+ "step": 230
171
+ },
172
+ {
173
+ "epoch": 0.636604774535809,
174
+ "grad_norm": 0.6022582014461996,
175
+ "learning_rate": 5e-06,
176
+ "loss": 0.642,
177
+ "step": 240
178
+ },
179
+ {
180
+ "epoch": 0.6631299734748011,
181
+ "grad_norm": 0.500260575545034,
182
+ "learning_rate": 5e-06,
183
+ "loss": 0.6344,
184
+ "step": 250
185
+ },
186
+ {
187
+ "epoch": 0.6896551724137931,
188
+ "grad_norm": 0.537336125918856,
189
+ "learning_rate": 5e-06,
190
+ "loss": 0.639,
191
+ "step": 260
192
+ },
193
+ {
194
+ "epoch": 0.7161803713527851,
195
+ "grad_norm": 0.5365620197699547,
196
+ "learning_rate": 5e-06,
197
+ "loss": 0.6403,
198
+ "step": 270
199
+ },
200
+ {
201
+ "epoch": 0.7427055702917772,
202
+ "grad_norm": 0.7683273733374665,
203
+ "learning_rate": 5e-06,
204
+ "loss": 0.6401,
205
+ "step": 280
206
+ },
207
+ {
208
+ "epoch": 0.7692307692307693,
209
+ "grad_norm": 0.5276716709186424,
210
+ "learning_rate": 5e-06,
211
+ "loss": 0.6338,
212
+ "step": 290
213
+ },
214
+ {
215
+ "epoch": 0.7957559681697612,
216
+ "grad_norm": 0.5074059045052902,
217
+ "learning_rate": 5e-06,
218
+ "loss": 0.6371,
219
+ "step": 300
220
+ },
221
+ {
222
+ "epoch": 0.8222811671087533,
223
+ "grad_norm": 0.67204833216127,
224
+ "learning_rate": 5e-06,
225
+ "loss": 0.639,
226
+ "step": 310
227
+ },
228
+ {
229
+ "epoch": 0.8488063660477454,
230
+ "grad_norm": 0.5619464110309291,
231
+ "learning_rate": 5e-06,
232
+ "loss": 0.6427,
233
+ "step": 320
234
+ },
235
+ {
236
+ "epoch": 0.8753315649867374,
237
+ "grad_norm": 0.5724558992220249,
238
+ "learning_rate": 5e-06,
239
+ "loss": 0.6294,
240
+ "step": 330
241
+ },
242
+ {
243
+ "epoch": 0.9018567639257294,
244
+ "grad_norm": 0.619777380357849,
245
+ "learning_rate": 5e-06,
246
+ "loss": 0.6334,
247
+ "step": 340
248
+ },
249
+ {
250
+ "epoch": 0.9283819628647215,
251
+ "grad_norm": 0.7434000365286857,
252
+ "learning_rate": 5e-06,
253
+ "loss": 0.6329,
254
+ "step": 350
255
+ },
256
+ {
257
+ "epoch": 0.9549071618037135,
258
+ "grad_norm": 0.5490872236122448,
259
+ "learning_rate": 5e-06,
260
+ "loss": 0.6286,
261
+ "step": 360
262
+ },
263
+ {
264
+ "epoch": 0.9814323607427056,
265
+ "grad_norm": 0.584736306154939,
266
+ "learning_rate": 5e-06,
267
+ "loss": 0.6296,
268
+ "step": 370
269
+ },
270
+ {
271
+ "epoch": 1.0,
272
+ "eval_loss": 0.6237567663192749,
273
+ "eval_runtime": 202.8984,
274
+ "eval_samples_per_second": 50.069,
275
+ "eval_steps_per_second": 0.394,
276
+ "step": 377
277
+ },
278
+ {
279
+ "epoch": 1.0079575596816976,
280
+ "grad_norm": 1.012512560233075,
281
+ "learning_rate": 5e-06,
282
+ "loss": 0.614,
283
+ "step": 380
284
+ },
285
+ {
286
+ "epoch": 1.0344827586206897,
287
+ "grad_norm": 0.6423484790332888,
288
+ "learning_rate": 5e-06,
289
+ "loss": 0.5872,
290
+ "step": 390
291
+ },
292
+ {
293
+ "epoch": 1.0610079575596818,
294
+ "grad_norm": 0.6707809448038629,
295
+ "learning_rate": 5e-06,
296
+ "loss": 0.5889,
297
+ "step": 400
298
+ },
299
+ {
300
+ "epoch": 1.0875331564986737,
301
+ "grad_norm": 0.6650816829620442,
302
+ "learning_rate": 5e-06,
303
+ "loss": 0.5807,
304
+ "step": 410
305
+ },
306
+ {
307
+ "epoch": 1.1140583554376657,
308
+ "grad_norm": 0.6224384353734153,
309
+ "learning_rate": 5e-06,
310
+ "loss": 0.5915,
311
+ "step": 420
312
+ },
313
+ {
314
+ "epoch": 1.1405835543766578,
315
+ "grad_norm": 0.5684723509980193,
316
+ "learning_rate": 5e-06,
317
+ "loss": 0.5868,
318
+ "step": 430
319
+ },
320
+ {
321
+ "epoch": 1.16710875331565,
322
+ "grad_norm": 0.7059253953598508,
323
+ "learning_rate": 5e-06,
324
+ "loss": 0.5766,
325
+ "step": 440
326
+ },
327
+ {
328
+ "epoch": 1.193633952254642,
329
+ "grad_norm": 0.523842539770435,
330
+ "learning_rate": 5e-06,
331
+ "loss": 0.5848,
332
+ "step": 450
333
+ },
334
+ {
335
+ "epoch": 1.2201591511936338,
336
+ "grad_norm": 0.6714226112627073,
337
+ "learning_rate": 5e-06,
338
+ "loss": 0.5832,
339
+ "step": 460
340
+ },
341
+ {
342
+ "epoch": 1.246684350132626,
343
+ "grad_norm": 0.7286525903858807,
344
+ "learning_rate": 5e-06,
345
+ "loss": 0.589,
346
+ "step": 470
347
+ },
348
+ {
349
+ "epoch": 1.273209549071618,
350
+ "grad_norm": 0.52664344746201,
351
+ "learning_rate": 5e-06,
352
+ "loss": 0.5781,
353
+ "step": 480
354
+ },
355
+ {
356
+ "epoch": 1.29973474801061,
357
+ "grad_norm": 0.5394135765946012,
358
+ "learning_rate": 5e-06,
359
+ "loss": 0.5793,
360
+ "step": 490
361
+ },
362
+ {
363
+ "epoch": 1.3262599469496021,
364
+ "grad_norm": 0.5473942505095254,
365
+ "learning_rate": 5e-06,
366
+ "loss": 0.5865,
367
+ "step": 500
368
+ },
369
+ {
370
+ "epoch": 1.3527851458885942,
371
+ "grad_norm": 0.5239209443855906,
372
+ "learning_rate": 5e-06,
373
+ "loss": 0.5851,
374
+ "step": 510
375
+ },
376
+ {
377
+ "epoch": 1.3793103448275863,
378
+ "grad_norm": 0.638268567207473,
379
+ "learning_rate": 5e-06,
380
+ "loss": 0.5868,
381
+ "step": 520
382
+ },
383
+ {
384
+ "epoch": 1.4058355437665782,
385
+ "grad_norm": 0.5569448150591939,
386
+ "learning_rate": 5e-06,
387
+ "loss": 0.5851,
388
+ "step": 530
389
+ },
390
+ {
391
+ "epoch": 1.4323607427055702,
392
+ "grad_norm": 0.5739975276281734,
393
+ "learning_rate": 5e-06,
394
+ "loss": 0.5807,
395
+ "step": 540
396
+ },
397
+ {
398
+ "epoch": 1.4588859416445623,
399
+ "grad_norm": 0.5662395827613842,
400
+ "learning_rate": 5e-06,
401
+ "loss": 0.5768,
402
+ "step": 550
403
+ },
404
+ {
405
+ "epoch": 1.4854111405835544,
406
+ "grad_norm": 0.7163603774804506,
407
+ "learning_rate": 5e-06,
408
+ "loss": 0.5843,
409
+ "step": 560
410
+ },
411
+ {
412
+ "epoch": 1.5119363395225465,
413
+ "grad_norm": 0.7163733609893056,
414
+ "learning_rate": 5e-06,
415
+ "loss": 0.5917,
416
+ "step": 570
417
+ },
418
+ {
419
+ "epoch": 1.5384615384615383,
420
+ "grad_norm": 0.6180542754631709,
421
+ "learning_rate": 5e-06,
422
+ "loss": 0.5817,
423
+ "step": 580
424
+ },
425
+ {
426
+ "epoch": 1.5649867374005306,
427
+ "grad_norm": 0.5638514267037327,
428
+ "learning_rate": 5e-06,
429
+ "loss": 0.5862,
430
+ "step": 590
431
+ },
432
+ {
433
+ "epoch": 1.5915119363395225,
434
+ "grad_norm": 0.561015542053421,
435
+ "learning_rate": 5e-06,
436
+ "loss": 0.5815,
437
+ "step": 600
438
+ },
439
+ {
440
+ "epoch": 1.6180371352785146,
441
+ "grad_norm": 0.5019925924846618,
442
+ "learning_rate": 5e-06,
443
+ "loss": 0.5854,
444
+ "step": 610
445
+ },
446
+ {
447
+ "epoch": 1.6445623342175066,
448
+ "grad_norm": 0.6456145062380878,
449
+ "learning_rate": 5e-06,
450
+ "loss": 0.584,
451
+ "step": 620
452
+ },
453
+ {
454
+ "epoch": 1.6710875331564987,
455
+ "grad_norm": 0.6273758065445275,
456
+ "learning_rate": 5e-06,
457
+ "loss": 0.5794,
458
+ "step": 630
459
+ },
460
+ {
461
+ "epoch": 1.6976127320954908,
462
+ "grad_norm": 0.47537875001219887,
463
+ "learning_rate": 5e-06,
464
+ "loss": 0.5831,
465
+ "step": 640
466
+ },
467
+ {
468
+ "epoch": 1.7241379310344827,
469
+ "grad_norm": 0.5799224686148554,
470
+ "learning_rate": 5e-06,
471
+ "loss": 0.5876,
472
+ "step": 650
473
+ },
474
+ {
475
+ "epoch": 1.750663129973475,
476
+ "grad_norm": 0.6477709788042249,
477
+ "learning_rate": 5e-06,
478
+ "loss": 0.5786,
479
+ "step": 660
480
+ },
481
+ {
482
+ "epoch": 1.7771883289124668,
483
+ "grad_norm": 0.5398722777232823,
484
+ "learning_rate": 5e-06,
485
+ "loss": 0.5788,
486
+ "step": 670
487
+ },
488
+ {
489
+ "epoch": 1.8037135278514589,
490
+ "grad_norm": 0.503790352480121,
491
+ "learning_rate": 5e-06,
492
+ "loss": 0.5789,
493
+ "step": 680
494
+ },
495
+ {
496
+ "epoch": 1.830238726790451,
497
+ "grad_norm": 0.5980642131822352,
498
+ "learning_rate": 5e-06,
499
+ "loss": 0.5836,
500
+ "step": 690
501
+ },
502
+ {
503
+ "epoch": 1.8567639257294428,
504
+ "grad_norm": 0.5144731472377694,
505
+ "learning_rate": 5e-06,
506
+ "loss": 0.5818,
507
+ "step": 700
508
+ },
509
+ {
510
+ "epoch": 1.8832891246684351,
511
+ "grad_norm": 0.748547689970371,
512
+ "learning_rate": 5e-06,
513
+ "loss": 0.5882,
514
+ "step": 710
515
+ },
516
+ {
517
+ "epoch": 1.909814323607427,
518
+ "grad_norm": 0.5454719348703825,
519
+ "learning_rate": 5e-06,
520
+ "loss": 0.5857,
521
+ "step": 720
522
+ },
523
+ {
524
+ "epoch": 1.936339522546419,
525
+ "grad_norm": 0.537701148828446,
526
+ "learning_rate": 5e-06,
527
+ "loss": 0.5919,
528
+ "step": 730
529
+ },
530
+ {
531
+ "epoch": 1.9628647214854111,
532
+ "grad_norm": 0.5333812418899406,
533
+ "learning_rate": 5e-06,
534
+ "loss": 0.5804,
535
+ "step": 740
536
+ },
537
+ {
538
+ "epoch": 1.9893899204244032,
539
+ "grad_norm": 0.5945192673823688,
540
+ "learning_rate": 5e-06,
541
+ "loss": 0.5875,
542
+ "step": 750
543
+ },
544
+ {
545
+ "epoch": 2.0,
546
+ "eval_loss": 0.6138430833816528,
547
+ "eval_runtime": 204.1448,
548
+ "eval_samples_per_second": 49.764,
549
+ "eval_steps_per_second": 0.392,
550
+ "step": 754
551
+ },
552
+ {
553
+ "epoch": 2.0159151193633953,
554
+ "grad_norm": 0.70666737481995,
555
+ "learning_rate": 5e-06,
556
+ "loss": 0.5507,
557
+ "step": 760
558
+ },
559
+ {
560
+ "epoch": 2.042440318302387,
561
+ "grad_norm": 0.6653710012047342,
562
+ "learning_rate": 5e-06,
563
+ "loss": 0.5366,
564
+ "step": 770
565
+ },
566
+ {
567
+ "epoch": 2.0689655172413794,
568
+ "grad_norm": 0.6392050869043902,
569
+ "learning_rate": 5e-06,
570
+ "loss": 0.5362,
571
+ "step": 780
572
+ },
573
+ {
574
+ "epoch": 2.0954907161803713,
575
+ "grad_norm": 0.552417268591106,
576
+ "learning_rate": 5e-06,
577
+ "loss": 0.537,
578
+ "step": 790
579
+ },
580
+ {
581
+ "epoch": 2.1220159151193636,
582
+ "grad_norm": 0.5237418246816455,
583
+ "learning_rate": 5e-06,
584
+ "loss": 0.5387,
585
+ "step": 800
586
+ },
587
+ {
588
+ "epoch": 2.1485411140583555,
589
+ "grad_norm": 0.5931401358357342,
590
+ "learning_rate": 5e-06,
591
+ "loss": 0.5383,
592
+ "step": 810
593
+ },
594
+ {
595
+ "epoch": 2.1750663129973473,
596
+ "grad_norm": 0.5859178511093684,
597
+ "learning_rate": 5e-06,
598
+ "loss": 0.5385,
599
+ "step": 820
600
+ },
601
+ {
602
+ "epoch": 2.2015915119363396,
603
+ "grad_norm": 0.54174318871251,
604
+ "learning_rate": 5e-06,
605
+ "loss": 0.536,
606
+ "step": 830
607
+ },
608
+ {
609
+ "epoch": 2.2281167108753315,
610
+ "grad_norm": 0.6719837144625476,
611
+ "learning_rate": 5e-06,
612
+ "loss": 0.5443,
613
+ "step": 840
614
+ },
615
+ {
616
+ "epoch": 2.2546419098143238,
617
+ "grad_norm": 0.5913384023217347,
618
+ "learning_rate": 5e-06,
619
+ "loss": 0.5275,
620
+ "step": 850
621
+ },
622
+ {
623
+ "epoch": 2.2811671087533156,
624
+ "grad_norm": 0.5589992603235397,
625
+ "learning_rate": 5e-06,
626
+ "loss": 0.5324,
627
+ "step": 860
628
+ },
629
+ {
630
+ "epoch": 2.3076923076923075,
631
+ "grad_norm": 0.7009013738100872,
632
+ "learning_rate": 5e-06,
633
+ "loss": 0.5358,
634
+ "step": 870
635
+ },
636
+ {
637
+ "epoch": 2.3342175066313,
638
+ "grad_norm": 0.5192518963562719,
639
+ "learning_rate": 5e-06,
640
+ "loss": 0.5397,
641
+ "step": 880
642
+ },
643
+ {
644
+ "epoch": 2.3607427055702916,
645
+ "grad_norm": 0.5673317232053819,
646
+ "learning_rate": 5e-06,
647
+ "loss": 0.5441,
648
+ "step": 890
649
+ },
650
+ {
651
+ "epoch": 2.387267904509284,
652
+ "grad_norm": 0.6565705800504016,
653
+ "learning_rate": 5e-06,
654
+ "loss": 0.5376,
655
+ "step": 900
656
+ },
657
+ {
658
+ "epoch": 2.413793103448276,
659
+ "grad_norm": 0.6497220724543952,
660
+ "learning_rate": 5e-06,
661
+ "loss": 0.5309,
662
+ "step": 910
663
+ },
664
+ {
665
+ "epoch": 2.4403183023872677,
666
+ "grad_norm": 0.5907005582908587,
667
+ "learning_rate": 5e-06,
668
+ "loss": 0.535,
669
+ "step": 920
670
+ },
671
+ {
672
+ "epoch": 2.46684350132626,
673
+ "grad_norm": 0.6229830599060672,
674
+ "learning_rate": 5e-06,
675
+ "loss": 0.5497,
676
+ "step": 930
677
+ },
678
+ {
679
+ "epoch": 2.493368700265252,
680
+ "grad_norm": 0.5474753923784902,
681
+ "learning_rate": 5e-06,
682
+ "loss": 0.5409,
683
+ "step": 940
684
+ },
685
+ {
686
+ "epoch": 2.519893899204244,
687
+ "grad_norm": 0.5472523872985887,
688
+ "learning_rate": 5e-06,
689
+ "loss": 0.536,
690
+ "step": 950
691
+ },
692
+ {
693
+ "epoch": 2.546419098143236,
694
+ "grad_norm": 0.9882268568658281,
695
+ "learning_rate": 5e-06,
696
+ "loss": 0.5375,
697
+ "step": 960
698
+ },
699
+ {
700
+ "epoch": 2.5729442970822283,
701
+ "grad_norm": 0.6076358216087694,
702
+ "learning_rate": 5e-06,
703
+ "loss": 0.5417,
704
+ "step": 970
705
+ },
706
+ {
707
+ "epoch": 2.59946949602122,
708
+ "grad_norm": 0.6233857713542503,
709
+ "learning_rate": 5e-06,
710
+ "loss": 0.543,
711
+ "step": 980
712
+ },
713
+ {
714
+ "epoch": 2.6259946949602124,
715
+ "grad_norm": 0.662156931341361,
716
+ "learning_rate": 5e-06,
717
+ "loss": 0.5437,
718
+ "step": 990
719
+ },
720
+ {
721
+ "epoch": 2.6525198938992043,
722
+ "grad_norm": 0.520667113721229,
723
+ "learning_rate": 5e-06,
724
+ "loss": 0.5392,
725
+ "step": 1000
726
+ },
727
+ {
728
+ "epoch": 2.679045092838196,
729
+ "grad_norm": 0.5971607456495881,
730
+ "learning_rate": 5e-06,
731
+ "loss": 0.5357,
732
+ "step": 1010
733
+ },
734
+ {
735
+ "epoch": 2.7055702917771884,
736
+ "grad_norm": 0.5939890330376003,
737
+ "learning_rate": 5e-06,
738
+ "loss": 0.5371,
739
+ "step": 1020
740
+ },
741
+ {
742
+ "epoch": 2.7320954907161803,
743
+ "grad_norm": 0.5773060764723967,
744
+ "learning_rate": 5e-06,
745
+ "loss": 0.5359,
746
+ "step": 1030
747
+ },
748
+ {
749
+ "epoch": 2.7586206896551726,
750
+ "grad_norm": 0.5827565949490094,
751
+ "learning_rate": 5e-06,
752
+ "loss": 0.5431,
753
+ "step": 1040
754
+ },
755
+ {
756
+ "epoch": 2.7851458885941645,
757
+ "grad_norm": 0.6314529885890559,
758
+ "learning_rate": 5e-06,
759
+ "loss": 0.5444,
760
+ "step": 1050
761
+ },
762
+ {
763
+ "epoch": 2.8116710875331563,
764
+ "grad_norm": 0.5422009039427094,
765
+ "learning_rate": 5e-06,
766
+ "loss": 0.54,
767
+ "step": 1060
768
+ },
769
+ {
770
+ "epoch": 2.8381962864721486,
771
+ "grad_norm": 0.6028981522236778,
772
+ "learning_rate": 5e-06,
773
+ "loss": 0.5433,
774
+ "step": 1070
775
+ },
776
+ {
777
+ "epoch": 2.8647214854111405,
778
+ "grad_norm": 0.5970004720986751,
779
+ "learning_rate": 5e-06,
780
+ "loss": 0.5461,
781
+ "step": 1080
782
+ },
783
+ {
784
+ "epoch": 2.8912466843501328,
785
+ "grad_norm": 0.574798524733574,
786
+ "learning_rate": 5e-06,
787
+ "loss": 0.5397,
788
+ "step": 1090
789
+ },
790
+ {
791
+ "epoch": 2.9177718832891246,
792
+ "grad_norm": 0.5790776127397061,
793
+ "learning_rate": 5e-06,
794
+ "loss": 0.5404,
795
+ "step": 1100
796
+ },
797
+ {
798
+ "epoch": 2.9442970822281165,
799
+ "grad_norm": 0.5249016599382924,
800
+ "learning_rate": 5e-06,
801
+ "loss": 0.5461,
802
+ "step": 1110
803
+ },
804
+ {
805
+ "epoch": 2.970822281167109,
806
+ "grad_norm": 0.6578189789009287,
807
+ "learning_rate": 5e-06,
808
+ "loss": 0.5387,
809
+ "step": 1120
810
+ },
811
+ {
812
+ "epoch": 2.9973474801061006,
813
+ "grad_norm": 0.5226173945421619,
814
+ "learning_rate": 5e-06,
815
+ "loss": 0.5391,
816
+ "step": 1130
817
+ },
818
+ {
819
+ "epoch": 3.0,
820
+ "eval_loss": 0.6164913177490234,
821
+ "eval_runtime": 204.5208,
822
+ "eval_samples_per_second": 49.672,
823
+ "eval_steps_per_second": 0.391,
824
+ "step": 1131
825
+ },
826
+ {
827
+ "epoch": 3.0,
828
+ "step": 1131,
829
+ "total_flos": 1894048365281280.0,
830
+ "train_loss": 0.5957063028603713,
831
+ "train_runtime": 34171.7918,
832
+ "train_samples_per_second": 16.944,
833
+ "train_steps_per_second": 0.033
834
+ }
835
+ ],
836
+ "logging_steps": 10,
837
+ "max_steps": 1131,
838
+ "num_input_tokens_seen": 0,
839
+ "num_train_epochs": 3,
840
+ "save_steps": 500,
841
+ "stateful_callbacks": {
842
+ "TrainerControl": {
843
+ "args": {
844
+ "should_epoch_stop": false,
845
+ "should_evaluate": false,
846
+ "should_log": false,
847
+ "should_save": true,
848
+ "should_training_stop": true
849
+ },
850
+ "attributes": {}
851
+ }
852
+ },
853
+ "total_flos": 1894048365281280.0,
854
+ "train_batch_size": 8,
855
+ "trial_name": null,
856
+ "trial_params": null
857
+ }
training_eval_loss.png ADDED
training_loss.png ADDED