sedrickkeh commited on
Commit
cfc1cd2
·
verified ·
1 Parent(s): ce5fe85

End of training

Browse files
Files changed (5) hide show
  1. README.md +2 -1
  2. all_results.json +8 -0
  3. train_results.json +8 -0
  4. trainer_state.json +693 -0
  5. training_loss.png +0 -0
README.md CHANGED
@@ -4,6 +4,7 @@ license: apache-2.0
4
  base_model: Qwen/Qwen2.5-7B-Instruct
5
  tags:
6
  - llama-factory
 
7
  - generated_from_trainer
8
  model-index:
9
  - name: 3k_globalbatchsize96_lr2e5_epochs3
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
15
 
16
  # 3k_globalbatchsize96_lr2e5_epochs3
17
 
18
- This model is a fine-tuned version of [Qwen/Qwen2.5-7B-Instruct](https://huggingface.co/Qwen/Qwen2.5-7B-Instruct) on an unknown dataset.
19
 
20
  ## Model description
21
 
 
4
  base_model: Qwen/Qwen2.5-7B-Instruct
5
  tags:
6
  - llama-factory
7
+ - full
8
  - generated_from_trainer
9
  model-index:
10
  - name: 3k_globalbatchsize96_lr2e5_epochs3
 
16
 
17
  # 3k_globalbatchsize96_lr2e5_epochs3
18
 
19
+ This model is a fine-tuned version of [Qwen/Qwen2.5-7B-Instruct](https://huggingface.co/Qwen/Qwen2.5-7B-Instruct) on the mlfoundations-dev/openthoughts_3000 dataset.
20
 
21
  ## Model description
22
 
all_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.976,
3
+ "total_flos": 1.818538711009198e+17,
4
+ "train_loss": 0.53916282679445,
5
+ "train_runtime": 9093.0931,
6
+ "train_samples_per_second": 0.989,
7
+ "train_steps_per_second": 0.01
8
+ }
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.976,
3
+ "total_flos": 1.818538711009198e+17,
4
+ "train_loss": 0.53916282679445,
5
+ "train_runtime": 9093.0931,
6
+ "train_samples_per_second": 0.989,
7
+ "train_steps_per_second": 0.01
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,693 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 2.976,
5
+ "eval_steps": 500,
6
+ "global_step": 93,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.032,
13
+ "grad_norm": 6.086521555575525,
14
+ "learning_rate": 2.0000000000000003e-06,
15
+ "loss": 0.8709,
16
+ "step": 1
17
+ },
18
+ {
19
+ "epoch": 0.064,
20
+ "grad_norm": 5.862669428136825,
21
+ "learning_rate": 4.000000000000001e-06,
22
+ "loss": 0.8625,
23
+ "step": 2
24
+ },
25
+ {
26
+ "epoch": 0.096,
27
+ "grad_norm": 5.665430905084772,
28
+ "learning_rate": 6e-06,
29
+ "loss": 0.8709,
30
+ "step": 3
31
+ },
32
+ {
33
+ "epoch": 0.128,
34
+ "grad_norm": 3.9717562012437484,
35
+ "learning_rate": 8.000000000000001e-06,
36
+ "loss": 0.7764,
37
+ "step": 4
38
+ },
39
+ {
40
+ "epoch": 0.16,
41
+ "grad_norm": 2.2462420454815217,
42
+ "learning_rate": 1e-05,
43
+ "loss": 0.7977,
44
+ "step": 5
45
+ },
46
+ {
47
+ "epoch": 0.192,
48
+ "grad_norm": 4.275087411533909,
49
+ "learning_rate": 1.2e-05,
50
+ "loss": 0.8235,
51
+ "step": 6
52
+ },
53
+ {
54
+ "epoch": 0.224,
55
+ "grad_norm": 4.540127453722676,
56
+ "learning_rate": 1.4e-05,
57
+ "loss": 0.7536,
58
+ "step": 7
59
+ },
60
+ {
61
+ "epoch": 0.256,
62
+ "grad_norm": 5.958818552449376,
63
+ "learning_rate": 1.6000000000000003e-05,
64
+ "loss": 0.7817,
65
+ "step": 8
66
+ },
67
+ {
68
+ "epoch": 0.288,
69
+ "grad_norm": 4.615179916049989,
70
+ "learning_rate": 1.8e-05,
71
+ "loss": 0.7716,
72
+ "step": 9
73
+ },
74
+ {
75
+ "epoch": 0.32,
76
+ "grad_norm": 2.3655854780686663,
77
+ "learning_rate": 2e-05,
78
+ "loss": 0.6932,
79
+ "step": 10
80
+ },
81
+ {
82
+ "epoch": 0.352,
83
+ "grad_norm": 2.478544642789441,
84
+ "learning_rate": 1.9992837548163315e-05,
85
+ "loss": 0.7029,
86
+ "step": 11
87
+ },
88
+ {
89
+ "epoch": 0.384,
90
+ "grad_norm": 1.9787874769183103,
91
+ "learning_rate": 1.9971360452796523e-05,
92
+ "loss": 0.7003,
93
+ "step": 12
94
+ },
95
+ {
96
+ "epoch": 0.416,
97
+ "grad_norm": 1.3386567233728712,
98
+ "learning_rate": 1.993559947963185e-05,
99
+ "loss": 0.6677,
100
+ "step": 13
101
+ },
102
+ {
103
+ "epoch": 0.448,
104
+ "grad_norm": 1.177972539602327,
105
+ "learning_rate": 1.9885605855918887e-05,
106
+ "loss": 0.6495,
107
+ "step": 14
108
+ },
109
+ {
110
+ "epoch": 0.48,
111
+ "grad_norm": 1.1019717904815112,
112
+ "learning_rate": 1.9821451197042028e-05,
113
+ "loss": 0.5967,
114
+ "step": 15
115
+ },
116
+ {
117
+ "epoch": 0.512,
118
+ "grad_norm": 0.972479825971204,
119
+ "learning_rate": 1.9743227403932135e-05,
120
+ "loss": 0.6233,
121
+ "step": 16
122
+ },
123
+ {
124
+ "epoch": 0.544,
125
+ "grad_norm": 0.9169380714246801,
126
+ "learning_rate": 1.9651046531419335e-05,
127
+ "loss": 0.5939,
128
+ "step": 17
129
+ },
130
+ {
131
+ "epoch": 0.576,
132
+ "grad_norm": 0.7484870741589357,
133
+ "learning_rate": 1.9545040627715554e-05,
134
+ "loss": 0.5913,
135
+ "step": 18
136
+ },
137
+ {
138
+ "epoch": 0.608,
139
+ "grad_norm": 0.8433187788934118,
140
+ "learning_rate": 1.942536154525673e-05,
141
+ "loss": 0.5926,
142
+ "step": 19
143
+ },
144
+ {
145
+ "epoch": 0.64,
146
+ "grad_norm": 0.7004246074241133,
147
+ "learning_rate": 1.9292180723175656e-05,
148
+ "loss": 0.5679,
149
+ "step": 20
150
+ },
151
+ {
152
+ "epoch": 0.672,
153
+ "grad_norm": 0.9310125911747377,
154
+ "learning_rate": 1.9145688941717074e-05,
155
+ "loss": 0.5644,
156
+ "step": 21
157
+ },
158
+ {
159
+ "epoch": 0.704,
160
+ "grad_norm": 0.7970495324653322,
161
+ "learning_rate": 1.8986096048946826e-05,
162
+ "loss": 0.608,
163
+ "step": 22
164
+ },
165
+ {
166
+ "epoch": 0.736,
167
+ "grad_norm": 0.8089072723260873,
168
+ "learning_rate": 1.881363066014649e-05,
169
+ "loss": 0.5677,
170
+ "step": 23
171
+ },
172
+ {
173
+ "epoch": 0.768,
174
+ "grad_norm": 1.0367436501959517,
175
+ "learning_rate": 1.862853983032423e-05,
176
+ "loss": 0.6124,
177
+ "step": 24
178
+ },
179
+ {
180
+ "epoch": 0.8,
181
+ "grad_norm": 0.6549565017520195,
182
+ "learning_rate": 1.8431088700310846e-05,
183
+ "loss": 0.565,
184
+ "step": 25
185
+ },
186
+ {
187
+ "epoch": 0.832,
188
+ "grad_norm": 0.8100479455474157,
189
+ "learning_rate": 1.8221560116948103e-05,
190
+ "loss": 0.5536,
191
+ "step": 26
192
+ },
193
+ {
194
+ "epoch": 0.864,
195
+ "grad_norm": 0.6655173946648272,
196
+ "learning_rate": 1.8000254227913346e-05,
197
+ "loss": 0.5927,
198
+ "step": 27
199
+ },
200
+ {
201
+ "epoch": 0.896,
202
+ "grad_norm": 0.6489458832967611,
203
+ "learning_rate": 1.7767488051760858e-05,
204
+ "loss": 0.5996,
205
+ "step": 28
206
+ },
207
+ {
208
+ "epoch": 0.928,
209
+ "grad_norm": 0.6556791371183601,
210
+ "learning_rate": 1.7523595023795814e-05,
211
+ "loss": 0.5565,
212
+ "step": 29
213
+ },
214
+ {
215
+ "epoch": 0.96,
216
+ "grad_norm": 0.6656680446819412,
217
+ "learning_rate": 1.7268924518431437e-05,
218
+ "loss": 0.5691,
219
+ "step": 30
220
+ },
221
+ {
222
+ "epoch": 0.992,
223
+ "grad_norm": 0.6975728008241743,
224
+ "learning_rate": 1.700384134871351e-05,
225
+ "loss": 0.5634,
226
+ "step": 31
227
+ },
228
+ {
229
+ "epoch": 1.024,
230
+ "grad_norm": 1.0956954168489788,
231
+ "learning_rate": 1.672872524372919e-05,
232
+ "loss": 0.9416,
233
+ "step": 32
234
+ },
235
+ {
236
+ "epoch": 1.056,
237
+ "grad_norm": 0.48720084397077673,
238
+ "learning_rate": 1.644397030464877e-05,
239
+ "loss": 0.4704,
240
+ "step": 33
241
+ },
242
+ {
243
+ "epoch": 1.088,
244
+ "grad_norm": 0.7718389138248645,
245
+ "learning_rate": 1.614998444017954e-05,
246
+ "loss": 0.5326,
247
+ "step": 34
248
+ },
249
+ {
250
+ "epoch": 1.12,
251
+ "grad_norm": 0.6152935642153501,
252
+ "learning_rate": 1.5847188782240473e-05,
253
+ "loss": 0.5174,
254
+ "step": 35
255
+ },
256
+ {
257
+ "epoch": 1.152,
258
+ "grad_norm": 0.5860900719025863,
259
+ "learning_rate": 1.5536017082694846e-05,
260
+ "loss": 0.5031,
261
+ "step": 36
262
+ },
263
+ {
264
+ "epoch": 1.184,
265
+ "grad_norm": 0.9098930960008285,
266
+ "learning_rate": 1.5216915092004847e-05,
267
+ "loss": 0.4993,
268
+ "step": 37
269
+ },
270
+ {
271
+ "epoch": 1.216,
272
+ "grad_norm": 0.6433773311840126,
273
+ "learning_rate": 1.4890339920698334e-05,
274
+ "loss": 0.4832,
275
+ "step": 38
276
+ },
277
+ {
278
+ "epoch": 1.248,
279
+ "grad_norm": 0.574508872316764,
280
+ "learning_rate": 1.4556759384562418e-05,
281
+ "loss": 0.5301,
282
+ "step": 39
283
+ },
284
+ {
285
+ "epoch": 1.28,
286
+ "grad_norm": 0.6177654141668454,
287
+ "learning_rate": 1.421665133450184e-05,
288
+ "loss": 0.4809,
289
+ "step": 40
290
+ },
291
+ {
292
+ "epoch": 1.312,
293
+ "grad_norm": 0.5915067861936215,
294
+ "learning_rate": 1.3870502972022175e-05,
295
+ "loss": 0.4947,
296
+ "step": 41
297
+ },
298
+ {
299
+ "epoch": 1.3439999999999999,
300
+ "grad_norm": 0.6003229031077386,
301
+ "learning_rate": 1.351881015131833e-05,
302
+ "loss": 0.5285,
303
+ "step": 42
304
+ },
305
+ {
306
+ "epoch": 1.376,
307
+ "grad_norm": 0.5286636651815204,
308
+ "learning_rate": 1.316207666896824e-05,
309
+ "loss": 0.5134,
310
+ "step": 43
311
+ },
312
+ {
313
+ "epoch": 1.408,
314
+ "grad_norm": 0.5729916541561343,
315
+ "learning_rate": 1.2800813542249073e-05,
316
+ "loss": 0.4392,
317
+ "step": 44
318
+ },
319
+ {
320
+ "epoch": 1.44,
321
+ "grad_norm": 0.5389286220120857,
322
+ "learning_rate": 1.2435538277109919e-05,
323
+ "loss": 0.4811,
324
+ "step": 45
325
+ },
326
+ {
327
+ "epoch": 1.472,
328
+ "grad_norm": 0.5162482888115615,
329
+ "learning_rate": 1.206677412684953e-05,
330
+ "loss": 0.5603,
331
+ "step": 46
332
+ },
333
+ {
334
+ "epoch": 1.504,
335
+ "grad_norm": 0.5836229105696747,
336
+ "learning_rate": 1.1695049342560969e-05,
337
+ "loss": 0.452,
338
+ "step": 47
339
+ },
340
+ {
341
+ "epoch": 1.536,
342
+ "grad_norm": 0.5251402934459473,
343
+ "learning_rate": 1.1320896416417026e-05,
344
+ "loss": 0.5007,
345
+ "step": 48
346
+ },
347
+ {
348
+ "epoch": 1.568,
349
+ "grad_norm": 0.49581981068641073,
350
+ "learning_rate": 1.0944851318880314e-05,
351
+ "loss": 0.4815,
352
+ "step": 49
353
+ },
354
+ {
355
+ "epoch": 1.6,
356
+ "grad_norm": 0.6328540516916794,
357
+ "learning_rate": 1.0567452730930743e-05,
358
+ "loss": 0.559,
359
+ "step": 50
360
+ },
361
+ {
362
+ "epoch": 1.6320000000000001,
363
+ "grad_norm": 0.48288716820093464,
364
+ "learning_rate": 1.0189241272410191e-05,
365
+ "loss": 0.4568,
366
+ "step": 51
367
+ },
368
+ {
369
+ "epoch": 1.6640000000000001,
370
+ "grad_norm": 0.43453357593747727,
371
+ "learning_rate": 9.810758727589814e-06,
372
+ "loss": 0.5136,
373
+ "step": 52
374
+ },
375
+ {
376
+ "epoch": 1.696,
377
+ "grad_norm": 0.43509974292321796,
378
+ "learning_rate": 9.43254726906926e-06,
379
+ "loss": 0.4613,
380
+ "step": 53
381
+ },
382
+ {
383
+ "epoch": 1.728,
384
+ "grad_norm": 0.5105120033270906,
385
+ "learning_rate": 9.055148681119688e-06,
386
+ "loss": 0.5104,
387
+ "step": 54
388
+ },
389
+ {
390
+ "epoch": 1.76,
391
+ "grad_norm": 0.4655178250091279,
392
+ "learning_rate": 8.67910358358298e-06,
393
+ "loss": 0.515,
394
+ "step": 55
395
+ },
396
+ {
397
+ "epoch": 1.792,
398
+ "grad_norm": 0.4271675109217912,
399
+ "learning_rate": 8.304950657439034e-06,
400
+ "loss": 0.4594,
401
+ "step": 56
402
+ },
403
+ {
404
+ "epoch": 1.8239999999999998,
405
+ "grad_norm": 0.49279836130724164,
406
+ "learning_rate": 7.93322587315047e-06,
407
+ "loss": 0.5102,
408
+ "step": 57
409
+ },
410
+ {
411
+ "epoch": 1.8559999999999999,
412
+ "grad_norm": 0.4312602763649386,
413
+ "learning_rate": 7.564461722890082e-06,
414
+ "loss": 0.461,
415
+ "step": 58
416
+ },
417
+ {
418
+ "epoch": 1.888,
419
+ "grad_norm": 0.39722581396114753,
420
+ "learning_rate": 7.199186457750931e-06,
421
+ "loss": 0.5036,
422
+ "step": 59
423
+ },
424
+ {
425
+ "epoch": 1.92,
426
+ "grad_norm": 0.3707840503108745,
427
+ "learning_rate": 6.837923331031761e-06,
428
+ "loss": 0.4449,
429
+ "step": 60
430
+ },
431
+ {
432
+ "epoch": 1.952,
433
+ "grad_norm": 0.5482704117501644,
434
+ "learning_rate": 6.48118984868167e-06,
435
+ "loss": 0.492,
436
+ "step": 61
437
+ },
438
+ {
439
+ "epoch": 1.984,
440
+ "grad_norm": 0.37776373909746913,
441
+ "learning_rate": 6.129497027977829e-06,
442
+ "loss": 0.4131,
443
+ "step": 62
444
+ },
445
+ {
446
+ "epoch": 2.016,
447
+ "grad_norm": 0.8547764035968157,
448
+ "learning_rate": 5.78334866549816e-06,
449
+ "loss": 0.8721,
450
+ "step": 63
451
+ },
452
+ {
453
+ "epoch": 2.048,
454
+ "grad_norm": 0.4825739846598577,
455
+ "learning_rate": 5.443240615437586e-06,
456
+ "loss": 0.4129,
457
+ "step": 64
458
+ },
459
+ {
460
+ "epoch": 2.08,
461
+ "grad_norm": 0.48613435690189827,
462
+ "learning_rate": 5.109660079301668e-06,
463
+ "loss": 0.4016,
464
+ "step": 65
465
+ },
466
+ {
467
+ "epoch": 2.112,
468
+ "grad_norm": 0.5173122404541247,
469
+ "learning_rate": 4.783084907995156e-06,
470
+ "loss": 0.4602,
471
+ "step": 66
472
+ },
473
+ {
474
+ "epoch": 2.144,
475
+ "grad_norm": 0.4177033225672536,
476
+ "learning_rate": 4.463982917305155e-06,
477
+ "loss": 0.4306,
478
+ "step": 67
479
+ },
480
+ {
481
+ "epoch": 2.176,
482
+ "grad_norm": 0.47335210102568437,
483
+ "learning_rate": 4.152811217759529e-06,
484
+ "loss": 0.4554,
485
+ "step": 68
486
+ },
487
+ {
488
+ "epoch": 2.208,
489
+ "grad_norm": 0.5212910634677221,
490
+ "learning_rate": 3.850015559820465e-06,
491
+ "loss": 0.4453,
492
+ "step": 69
493
+ },
494
+ {
495
+ "epoch": 2.24,
496
+ "grad_norm": 0.4937857543597342,
497
+ "learning_rate": 3.5560296953512296e-06,
498
+ "loss": 0.4501,
499
+ "step": 70
500
+ },
501
+ {
502
+ "epoch": 2.2720000000000002,
503
+ "grad_norm": 0.456794357715941,
504
+ "learning_rate": 3.2712747562708115e-06,
505
+ "loss": 0.4418,
506
+ "step": 71
507
+ },
508
+ {
509
+ "epoch": 2.304,
510
+ "grad_norm": 0.41067614923626466,
511
+ "learning_rate": 2.9961586512864947e-06,
512
+ "loss": 0.3938,
513
+ "step": 72
514
+ },
515
+ {
516
+ "epoch": 2.336,
517
+ "grad_norm": 0.5398460548951893,
518
+ "learning_rate": 2.7310754815685627e-06,
519
+ "loss": 0.5264,
520
+ "step": 73
521
+ },
522
+ {
523
+ "epoch": 2.368,
524
+ "grad_norm": 0.39421897779539533,
525
+ "learning_rate": 2.4764049762041874e-06,
526
+ "loss": 0.4214,
527
+ "step": 74
528
+ },
529
+ {
530
+ "epoch": 2.4,
531
+ "grad_norm": 0.45677772727826993,
532
+ "learning_rate": 2.2325119482391466e-06,
533
+ "loss": 0.4438,
534
+ "step": 75
535
+ },
536
+ {
537
+ "epoch": 2.432,
538
+ "grad_norm": 0.3614990193323558,
539
+ "learning_rate": 1.9997457720866554e-06,
540
+ "loss": 0.3733,
541
+ "step": 76
542
+ },
543
+ {
544
+ "epoch": 2.464,
545
+ "grad_norm": 0.4657826146229637,
546
+ "learning_rate": 1.7784398830519002e-06,
547
+ "loss": 0.4528,
548
+ "step": 77
549
+ },
550
+ {
551
+ "epoch": 2.496,
552
+ "grad_norm": 0.37950794177613073,
553
+ "learning_rate": 1.5689112996891576e-06,
554
+ "loss": 0.4219,
555
+ "step": 78
556
+ },
557
+ {
558
+ "epoch": 2.528,
559
+ "grad_norm": 0.40926112347112487,
560
+ "learning_rate": 1.3714601696757713e-06,
561
+ "loss": 0.4557,
562
+ "step": 79
563
+ },
564
+ {
565
+ "epoch": 2.56,
566
+ "grad_norm": 0.3444002577522251,
567
+ "learning_rate": 1.1863693398535115e-06,
568
+ "loss": 0.3943,
569
+ "step": 80
570
+ },
571
+ {
572
+ "epoch": 2.592,
573
+ "grad_norm": 0.3404781664005299,
574
+ "learning_rate": 1.01390395105318e-06,
575
+ "loss": 0.4281,
576
+ "step": 81
577
+ },
578
+ {
579
+ "epoch": 2.624,
580
+ "grad_norm": 0.3448953746542976,
581
+ "learning_rate": 8.543110582829272e-07,
582
+ "loss": 0.4293,
583
+ "step": 82
584
+ },
585
+ {
586
+ "epoch": 2.656,
587
+ "grad_norm": 0.320896402093994,
588
+ "learning_rate": 7.078192768243486e-07,
589
+ "loss": 0.3955,
590
+ "step": 83
591
+ },
592
+ {
593
+ "epoch": 2.6879999999999997,
594
+ "grad_norm": 0.3635340073918507,
595
+ "learning_rate": 5.746384547432738e-07,
596
+ "loss": 0.4496,
597
+ "step": 84
598
+ },
599
+ {
600
+ "epoch": 2.7199999999999998,
601
+ "grad_norm": 0.3153399588065185,
602
+ "learning_rate": 4.549593722844492e-07,
603
+ "loss": 0.3784,
604
+ "step": 85
605
+ },
606
+ {
607
+ "epoch": 2.752,
608
+ "grad_norm": 0.32229126417822246,
609
+ "learning_rate": 3.4895346858066723e-07,
610
+ "loss": 0.4526,
611
+ "step": 86
612
+ },
613
+ {
614
+ "epoch": 2.784,
615
+ "grad_norm": 0.33084039803529036,
616
+ "learning_rate": 2.5677259606786686e-07,
617
+ "loss": 0.3698,
618
+ "step": 87
619
+ },
620
+ {
621
+ "epoch": 2.816,
622
+ "grad_norm": 0.3521643205936658,
623
+ "learning_rate": 1.7854880295797406e-07,
624
+ "loss": 0.4528,
625
+ "step": 88
626
+ },
627
+ {
628
+ "epoch": 2.848,
629
+ "grad_norm": 0.3379615978848595,
630
+ "learning_rate": 1.1439414408111471e-07,
631
+ "loss": 0.4464,
632
+ "step": 89
633
+ },
634
+ {
635
+ "epoch": 2.88,
636
+ "grad_norm": 0.3383417229866119,
637
+ "learning_rate": 6.440052036815081e-08,
638
+ "loss": 0.3912,
639
+ "step": 90
640
+ },
641
+ {
642
+ "epoch": 2.912,
643
+ "grad_norm": 0.3088668241871979,
644
+ "learning_rate": 2.86395472034795e-08,
645
+ "loss": 0.4426,
646
+ "step": 91
647
+ },
648
+ {
649
+ "epoch": 2.944,
650
+ "grad_norm": 0.32476651809192353,
651
+ "learning_rate": 7.162451836685291e-09,
652
+ "loss": 0.4281,
653
+ "step": 92
654
+ },
655
+ {
656
+ "epoch": 2.976,
657
+ "grad_norm": 0.3358967063808013,
658
+ "learning_rate": 0.0,
659
+ "loss": 0.4736,
660
+ "step": 93
661
+ },
662
+ {
663
+ "epoch": 2.976,
664
+ "step": 93,
665
+ "total_flos": 1.818538711009198e+17,
666
+ "train_loss": 0.53916282679445,
667
+ "train_runtime": 9093.0931,
668
+ "train_samples_per_second": 0.989,
669
+ "train_steps_per_second": 0.01
670
+ }
671
+ ],
672
+ "logging_steps": 1.0,
673
+ "max_steps": 93,
674
+ "num_input_tokens_seen": 0,
675
+ "num_train_epochs": 3,
676
+ "save_steps": 500,
677
+ "stateful_callbacks": {
678
+ "TrainerControl": {
679
+ "args": {
680
+ "should_epoch_stop": false,
681
+ "should_evaluate": false,
682
+ "should_log": false,
683
+ "should_save": true,
684
+ "should_training_stop": true
685
+ },
686
+ "attributes": {}
687
+ }
688
+ },
689
+ "total_flos": 1.818538711009198e+17,
690
+ "train_batch_size": 1,
691
+ "trial_name": null,
692
+ "trial_params": null
693
+ }
training_loss.png ADDED