sedrickkeh commited on
Commit
584e879
·
verified ·
1 Parent(s): 982c22d

End of training

Browse files
README.md CHANGED
@@ -4,6 +4,7 @@ license: apache-2.0
4
  base_model: mistralai/Mistral-7B-v0.1
5
  tags:
6
  - llama-factory
 
7
  - generated_from_trainer
8
  model-index:
9
  - name: hp_ablations_mistral_bsz512_dcftv1.2
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
15
 
16
  # hp_ablations_mistral_bsz512_dcftv1.2
17
 
18
- This model is a fine-tuned version of [mistralai/Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1) on an unknown dataset.
19
  It achieves the following results on the evaluation set:
20
  - Loss: 0.0765
21
 
 
4
  base_model: mistralai/Mistral-7B-v0.1
5
  tags:
6
  - llama-factory
7
+ - full
8
  - generated_from_trainer
9
  model-index:
10
  - name: hp_ablations_mistral_bsz512_dcftv1.2
 
16
 
17
  # hp_ablations_mistral_bsz512_dcftv1.2
18
 
19
+ This model is a fine-tuned version of [mistralai/Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1) on the mlfoundations-dev/oh-dcft-v1.2_no-curation_gpt-4o-mini dataset.
20
  It achieves the following results on the evaluation set:
21
  - Loss: 0.0765
22
 
all_results.json ADDED
@@ -0,0 +1,12 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.99695843190267,
3
+ "eval_loss": 0.07649385929107666,
4
+ "eval_runtime": 382.2734,
5
+ "eval_samples_per_second": 26.068,
6
+ "eval_steps_per_second": 0.408,
7
+ "total_flos": 1854056851046400.0,
8
+ "train_loss": 0.49424083467636865,
9
+ "train_runtime": 64065.8676,
10
+ "train_samples_per_second": 8.866,
11
+ "train_steps_per_second": 0.017
12
+ }
eval_results.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.99695843190267,
3
+ "eval_loss": 0.07649385929107666,
4
+ "eval_runtime": 382.2734,
5
+ "eval_samples_per_second": 26.068,
6
+ "eval_steps_per_second": 0.408
7
+ }
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.99695843190267,
3
+ "total_flos": 1854056851046400.0,
4
+ "train_loss": 0.49424083467636865,
5
+ "train_runtime": 64065.8676,
6
+ "train_samples_per_second": 8.866,
7
+ "train_steps_per_second": 0.017
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,836 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 2.99695843190267,
5
+ "eval_steps": 500,
6
+ "global_step": 1107,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.027036160865157147,
13
+ "grad_norm": 3.523985384534832,
14
+ "learning_rate": 5e-06,
15
+ "loss": 0.8602,
16
+ "step": 10
17
+ },
18
+ {
19
+ "epoch": 0.054072321730314295,
20
+ "grad_norm": 2.330230157876199,
21
+ "learning_rate": 5e-06,
22
+ "loss": 0.6873,
23
+ "step": 20
24
+ },
25
+ {
26
+ "epoch": 0.08110848259547145,
27
+ "grad_norm": 2.692866742411895,
28
+ "learning_rate": 5e-06,
29
+ "loss": 0.6489,
30
+ "step": 30
31
+ },
32
+ {
33
+ "epoch": 0.10814464346062859,
34
+ "grad_norm": 1.742013451785388,
35
+ "learning_rate": 5e-06,
36
+ "loss": 0.6363,
37
+ "step": 40
38
+ },
39
+ {
40
+ "epoch": 0.13518080432578575,
41
+ "grad_norm": 1.9825861616522564,
42
+ "learning_rate": 5e-06,
43
+ "loss": 0.6212,
44
+ "step": 50
45
+ },
46
+ {
47
+ "epoch": 0.1622169651909429,
48
+ "grad_norm": 1.963028689906959,
49
+ "learning_rate": 5e-06,
50
+ "loss": 0.6158,
51
+ "step": 60
52
+ },
53
+ {
54
+ "epoch": 0.18925312605610004,
55
+ "grad_norm": 1.5400134410581494,
56
+ "learning_rate": 5e-06,
57
+ "loss": 0.6126,
58
+ "step": 70
59
+ },
60
+ {
61
+ "epoch": 0.21628928692125718,
62
+ "grad_norm": 1.7957880103549582,
63
+ "learning_rate": 5e-06,
64
+ "loss": 0.6046,
65
+ "step": 80
66
+ },
67
+ {
68
+ "epoch": 0.24332544778641432,
69
+ "grad_norm": 2.0825972162611626,
70
+ "learning_rate": 5e-06,
71
+ "loss": 0.6015,
72
+ "step": 90
73
+ },
74
+ {
75
+ "epoch": 0.2703616086515715,
76
+ "grad_norm": 1.5684743614874819,
77
+ "learning_rate": 5e-06,
78
+ "loss": 0.6044,
79
+ "step": 100
80
+ },
81
+ {
82
+ "epoch": 0.29739776951672864,
83
+ "grad_norm": 1.5536576895608207,
84
+ "learning_rate": 5e-06,
85
+ "loss": 0.6009,
86
+ "step": 110
87
+ },
88
+ {
89
+ "epoch": 0.3244339303818858,
90
+ "grad_norm": 1.9211273900718062,
91
+ "learning_rate": 5e-06,
92
+ "loss": 0.5959,
93
+ "step": 120
94
+ },
95
+ {
96
+ "epoch": 0.3514700912470429,
97
+ "grad_norm": 1.857918426411918,
98
+ "learning_rate": 5e-06,
99
+ "loss": 0.5981,
100
+ "step": 130
101
+ },
102
+ {
103
+ "epoch": 0.37850625211220007,
104
+ "grad_norm": 2.480111887309551,
105
+ "learning_rate": 5e-06,
106
+ "loss": 0.5972,
107
+ "step": 140
108
+ },
109
+ {
110
+ "epoch": 0.4055424129773572,
111
+ "grad_norm": 1.9435865546624187,
112
+ "learning_rate": 5e-06,
113
+ "loss": 0.589,
114
+ "step": 150
115
+ },
116
+ {
117
+ "epoch": 0.43257857384251436,
118
+ "grad_norm": 2.256806266648931,
119
+ "learning_rate": 5e-06,
120
+ "loss": 0.5905,
121
+ "step": 160
122
+ },
123
+ {
124
+ "epoch": 0.4596147347076715,
125
+ "grad_norm": 1.9395141957859183,
126
+ "learning_rate": 5e-06,
127
+ "loss": 0.5907,
128
+ "step": 170
129
+ },
130
+ {
131
+ "epoch": 0.48665089557282865,
132
+ "grad_norm": 1.70470436825217,
133
+ "learning_rate": 5e-06,
134
+ "loss": 0.5891,
135
+ "step": 180
136
+ },
137
+ {
138
+ "epoch": 0.5136870564379858,
139
+ "grad_norm": 1.2803454947922437,
140
+ "learning_rate": 5e-06,
141
+ "loss": 0.5864,
142
+ "step": 190
143
+ },
144
+ {
145
+ "epoch": 0.540723217303143,
146
+ "grad_norm": 1.6583430834053514,
147
+ "learning_rate": 5e-06,
148
+ "loss": 0.5845,
149
+ "step": 200
150
+ },
151
+ {
152
+ "epoch": 0.5677593781683001,
153
+ "grad_norm": 1.6236160787848246,
154
+ "learning_rate": 5e-06,
155
+ "loss": 0.5873,
156
+ "step": 210
157
+ },
158
+ {
159
+ "epoch": 0.5947955390334573,
160
+ "grad_norm": 1.4448675180275712,
161
+ "learning_rate": 5e-06,
162
+ "loss": 0.5829,
163
+ "step": 220
164
+ },
165
+ {
166
+ "epoch": 0.6218316998986144,
167
+ "grad_norm": 1.5221126449990794,
168
+ "learning_rate": 5e-06,
169
+ "loss": 0.5844,
170
+ "step": 230
171
+ },
172
+ {
173
+ "epoch": 0.6488678607637716,
174
+ "grad_norm": 1.3401517373172736,
175
+ "learning_rate": 5e-06,
176
+ "loss": 0.581,
177
+ "step": 240
178
+ },
179
+ {
180
+ "epoch": 0.6759040216289287,
181
+ "grad_norm": 3.0974221570456697,
182
+ "learning_rate": 5e-06,
183
+ "loss": 0.5751,
184
+ "step": 250
185
+ },
186
+ {
187
+ "epoch": 0.7029401824940859,
188
+ "grad_norm": 1.7078786998431879,
189
+ "learning_rate": 5e-06,
190
+ "loss": 0.5786,
191
+ "step": 260
192
+ },
193
+ {
194
+ "epoch": 0.729976343359243,
195
+ "grad_norm": 1.3452565169314334,
196
+ "learning_rate": 5e-06,
197
+ "loss": 0.5802,
198
+ "step": 270
199
+ },
200
+ {
201
+ "epoch": 0.7570125042244001,
202
+ "grad_norm": 1.2754169661691006,
203
+ "learning_rate": 5e-06,
204
+ "loss": 0.5753,
205
+ "step": 280
206
+ },
207
+ {
208
+ "epoch": 0.7840486650895573,
209
+ "grad_norm": 1.561813747345879,
210
+ "learning_rate": 5e-06,
211
+ "loss": 0.5756,
212
+ "step": 290
213
+ },
214
+ {
215
+ "epoch": 0.8110848259547144,
216
+ "grad_norm": 1.4562454979417123,
217
+ "learning_rate": 5e-06,
218
+ "loss": 0.5754,
219
+ "step": 300
220
+ },
221
+ {
222
+ "epoch": 0.8381209868198716,
223
+ "grad_norm": 1.3655414659589415,
224
+ "learning_rate": 5e-06,
225
+ "loss": 0.5731,
226
+ "step": 310
227
+ },
228
+ {
229
+ "epoch": 0.8651571476850287,
230
+ "grad_norm": 1.57353294016275,
231
+ "learning_rate": 5e-06,
232
+ "loss": 0.5721,
233
+ "step": 320
234
+ },
235
+ {
236
+ "epoch": 0.8921933085501859,
237
+ "grad_norm": 1.9418137004471465,
238
+ "learning_rate": 5e-06,
239
+ "loss": 0.5736,
240
+ "step": 330
241
+ },
242
+ {
243
+ "epoch": 0.919229469415343,
244
+ "grad_norm": 1.6960343893725316,
245
+ "learning_rate": 5e-06,
246
+ "loss": 0.5806,
247
+ "step": 340
248
+ },
249
+ {
250
+ "epoch": 0.9462656302805001,
251
+ "grad_norm": 2.406507142621058,
252
+ "learning_rate": 5e-06,
253
+ "loss": 0.5714,
254
+ "step": 350
255
+ },
256
+ {
257
+ "epoch": 0.9733017911456573,
258
+ "grad_norm": 1.9230224466359063,
259
+ "learning_rate": 5e-06,
260
+ "loss": 0.5704,
261
+ "step": 360
262
+ },
263
+ {
264
+ "epoch": 0.9976343359242987,
265
+ "eval_loss": 0.07099956274032593,
266
+ "eval_runtime": 383.3884,
267
+ "eval_samples_per_second": 25.992,
268
+ "eval_steps_per_second": 0.407,
269
+ "step": 369
270
+ },
271
+ {
272
+ "epoch": 1.0023656640757013,
273
+ "grad_norm": 3.4559216993428326,
274
+ "learning_rate": 5e-06,
275
+ "loss": 0.5655,
276
+ "step": 370
277
+ },
278
+ {
279
+ "epoch": 1.0294018249408583,
280
+ "grad_norm": 2.6945339527227783,
281
+ "learning_rate": 5e-06,
282
+ "loss": 0.4803,
283
+ "step": 380
284
+ },
285
+ {
286
+ "epoch": 1.0564379858060156,
287
+ "grad_norm": 2.274860352040799,
288
+ "learning_rate": 5e-06,
289
+ "loss": 0.4751,
290
+ "step": 390
291
+ },
292
+ {
293
+ "epoch": 1.0834741466711728,
294
+ "grad_norm": 1.9418080331159586,
295
+ "learning_rate": 5e-06,
296
+ "loss": 0.4775,
297
+ "step": 400
298
+ },
299
+ {
300
+ "epoch": 1.1105103075363298,
301
+ "grad_norm": 1.3221563752390588,
302
+ "learning_rate": 5e-06,
303
+ "loss": 0.4751,
304
+ "step": 410
305
+ },
306
+ {
307
+ "epoch": 1.1375464684014869,
308
+ "grad_norm": 1.6270959849174909,
309
+ "learning_rate": 5e-06,
310
+ "loss": 0.4773,
311
+ "step": 420
312
+ },
313
+ {
314
+ "epoch": 1.1645826292666441,
315
+ "grad_norm": 1.723481598695817,
316
+ "learning_rate": 5e-06,
317
+ "loss": 0.4861,
318
+ "step": 430
319
+ },
320
+ {
321
+ "epoch": 1.1916187901318014,
322
+ "grad_norm": 1.6689122667194243,
323
+ "learning_rate": 5e-06,
324
+ "loss": 0.4771,
325
+ "step": 440
326
+ },
327
+ {
328
+ "epoch": 1.2186549509969584,
329
+ "grad_norm": 1.8852129756960698,
330
+ "learning_rate": 5e-06,
331
+ "loss": 0.4817,
332
+ "step": 450
333
+ },
334
+ {
335
+ "epoch": 1.2456911118621157,
336
+ "grad_norm": 1.5781803973122046,
337
+ "learning_rate": 5e-06,
338
+ "loss": 0.4832,
339
+ "step": 460
340
+ },
341
+ {
342
+ "epoch": 1.2727272727272727,
343
+ "grad_norm": 1.8932565449503365,
344
+ "learning_rate": 5e-06,
345
+ "loss": 0.4855,
346
+ "step": 470
347
+ },
348
+ {
349
+ "epoch": 1.29976343359243,
350
+ "grad_norm": 1.5040934591134398,
351
+ "learning_rate": 5e-06,
352
+ "loss": 0.4816,
353
+ "step": 480
354
+ },
355
+ {
356
+ "epoch": 1.326799594457587,
357
+ "grad_norm": 1.415624345433887,
358
+ "learning_rate": 5e-06,
359
+ "loss": 0.4817,
360
+ "step": 490
361
+ },
362
+ {
363
+ "epoch": 1.3538357553227442,
364
+ "grad_norm": 1.4726186128545236,
365
+ "learning_rate": 5e-06,
366
+ "loss": 0.4859,
367
+ "step": 500
368
+ },
369
+ {
370
+ "epoch": 1.3808719161879013,
371
+ "grad_norm": 1.371837855586058,
372
+ "learning_rate": 5e-06,
373
+ "loss": 0.4862,
374
+ "step": 510
375
+ },
376
+ {
377
+ "epoch": 1.4079080770530585,
378
+ "grad_norm": 1.625255953470612,
379
+ "learning_rate": 5e-06,
380
+ "loss": 0.4899,
381
+ "step": 520
382
+ },
383
+ {
384
+ "epoch": 1.4349442379182156,
385
+ "grad_norm": 1.4470657655644708,
386
+ "learning_rate": 5e-06,
387
+ "loss": 0.484,
388
+ "step": 530
389
+ },
390
+ {
391
+ "epoch": 1.4619803987833728,
392
+ "grad_norm": 1.4168708504506906,
393
+ "learning_rate": 5e-06,
394
+ "loss": 0.489,
395
+ "step": 540
396
+ },
397
+ {
398
+ "epoch": 1.4890165596485299,
399
+ "grad_norm": 1.2953354115079219,
400
+ "learning_rate": 5e-06,
401
+ "loss": 0.4876,
402
+ "step": 550
403
+ },
404
+ {
405
+ "epoch": 1.5160527205136871,
406
+ "grad_norm": 1.2905587316106748,
407
+ "learning_rate": 5e-06,
408
+ "loss": 0.4898,
409
+ "step": 560
410
+ },
411
+ {
412
+ "epoch": 1.5430888813788441,
413
+ "grad_norm": 1.874461693755812,
414
+ "learning_rate": 5e-06,
415
+ "loss": 0.4852,
416
+ "step": 570
417
+ },
418
+ {
419
+ "epoch": 1.5701250422440014,
420
+ "grad_norm": 1.537393515627057,
421
+ "learning_rate": 5e-06,
422
+ "loss": 0.4874,
423
+ "step": 580
424
+ },
425
+ {
426
+ "epoch": 1.5971612031091587,
427
+ "grad_norm": 1.7234212392856714,
428
+ "learning_rate": 5e-06,
429
+ "loss": 0.4911,
430
+ "step": 590
431
+ },
432
+ {
433
+ "epoch": 1.6241973639743157,
434
+ "grad_norm": 1.4569000028167551,
435
+ "learning_rate": 5e-06,
436
+ "loss": 0.487,
437
+ "step": 600
438
+ },
439
+ {
440
+ "epoch": 1.6512335248394727,
441
+ "grad_norm": 1.4876997193606485,
442
+ "learning_rate": 5e-06,
443
+ "loss": 0.4854,
444
+ "step": 610
445
+ },
446
+ {
447
+ "epoch": 1.67826968570463,
448
+ "grad_norm": 1.4853131850089583,
449
+ "learning_rate": 5e-06,
450
+ "loss": 0.4902,
451
+ "step": 620
452
+ },
453
+ {
454
+ "epoch": 1.7053058465697872,
455
+ "grad_norm": 1.4854501350323384,
456
+ "learning_rate": 5e-06,
457
+ "loss": 0.4952,
458
+ "step": 630
459
+ },
460
+ {
461
+ "epoch": 1.7323420074349443,
462
+ "grad_norm": 1.613201730070182,
463
+ "learning_rate": 5e-06,
464
+ "loss": 0.493,
465
+ "step": 640
466
+ },
467
+ {
468
+ "epoch": 1.7593781683001013,
469
+ "grad_norm": 1.3411867074544503,
470
+ "learning_rate": 5e-06,
471
+ "loss": 0.4904,
472
+ "step": 650
473
+ },
474
+ {
475
+ "epoch": 1.7864143291652586,
476
+ "grad_norm": 1.3453881021060534,
477
+ "learning_rate": 5e-06,
478
+ "loss": 0.4879,
479
+ "step": 660
480
+ },
481
+ {
482
+ "epoch": 1.8134504900304158,
483
+ "grad_norm": 1.4275860747925428,
484
+ "learning_rate": 5e-06,
485
+ "loss": 0.4904,
486
+ "step": 670
487
+ },
488
+ {
489
+ "epoch": 1.8404866508955728,
490
+ "grad_norm": 1.3712075307477265,
491
+ "learning_rate": 5e-06,
492
+ "loss": 0.4935,
493
+ "step": 680
494
+ },
495
+ {
496
+ "epoch": 1.8675228117607299,
497
+ "grad_norm": 1.2986482215538881,
498
+ "learning_rate": 5e-06,
499
+ "loss": 0.4917,
500
+ "step": 690
501
+ },
502
+ {
503
+ "epoch": 1.8945589726258871,
504
+ "grad_norm": 1.2770662158812232,
505
+ "learning_rate": 5e-06,
506
+ "loss": 0.4943,
507
+ "step": 700
508
+ },
509
+ {
510
+ "epoch": 1.9215951334910444,
511
+ "grad_norm": 1.2971456378708284,
512
+ "learning_rate": 5e-06,
513
+ "loss": 0.4994,
514
+ "step": 710
515
+ },
516
+ {
517
+ "epoch": 1.9486312943562014,
518
+ "grad_norm": 1.2953045522832038,
519
+ "learning_rate": 5e-06,
520
+ "loss": 0.4927,
521
+ "step": 720
522
+ },
523
+ {
524
+ "epoch": 1.9756674552213584,
525
+ "grad_norm": 1.2187798944947157,
526
+ "learning_rate": 5e-06,
527
+ "loss": 0.4886,
528
+ "step": 730
529
+ },
530
+ {
531
+ "epoch": 1.9972963839134843,
532
+ "eval_loss": 0.07174264639616013,
533
+ "eval_runtime": 383.5066,
534
+ "eval_samples_per_second": 25.984,
535
+ "eval_steps_per_second": 0.407,
536
+ "step": 738
537
+ },
538
+ {
539
+ "epoch": 2.0047313281514025,
540
+ "grad_norm": 3.2606644157882583,
541
+ "learning_rate": 5e-06,
542
+ "loss": 0.477,
543
+ "step": 740
544
+ },
545
+ {
546
+ "epoch": 2.0317674890165596,
547
+ "grad_norm": 1.997812835463083,
548
+ "learning_rate": 5e-06,
549
+ "loss": 0.3854,
550
+ "step": 750
551
+ },
552
+ {
553
+ "epoch": 2.0588036498817166,
554
+ "grad_norm": 1.9996540417166961,
555
+ "learning_rate": 5e-06,
556
+ "loss": 0.3864,
557
+ "step": 760
558
+ },
559
+ {
560
+ "epoch": 2.085839810746874,
561
+ "grad_norm": 1.8618681796717789,
562
+ "learning_rate": 5e-06,
563
+ "loss": 0.3812,
564
+ "step": 770
565
+ },
566
+ {
567
+ "epoch": 2.112875971612031,
568
+ "grad_norm": 1.4915524529602713,
569
+ "learning_rate": 5e-06,
570
+ "loss": 0.3811,
571
+ "step": 780
572
+ },
573
+ {
574
+ "epoch": 2.139912132477188,
575
+ "grad_norm": 1.5723106961760522,
576
+ "learning_rate": 5e-06,
577
+ "loss": 0.3808,
578
+ "step": 790
579
+ },
580
+ {
581
+ "epoch": 2.1669482933423456,
582
+ "grad_norm": 1.714466030885258,
583
+ "learning_rate": 5e-06,
584
+ "loss": 0.3876,
585
+ "step": 800
586
+ },
587
+ {
588
+ "epoch": 2.1939844542075027,
589
+ "grad_norm": 1.7511244660613634,
590
+ "learning_rate": 5e-06,
591
+ "loss": 0.3862,
592
+ "step": 810
593
+ },
594
+ {
595
+ "epoch": 2.2210206150726597,
596
+ "grad_norm": 1.5027273200313567,
597
+ "learning_rate": 5e-06,
598
+ "loss": 0.3873,
599
+ "step": 820
600
+ },
601
+ {
602
+ "epoch": 2.2480567759378167,
603
+ "grad_norm": 1.6128980496405356,
604
+ "learning_rate": 5e-06,
605
+ "loss": 0.3896,
606
+ "step": 830
607
+ },
608
+ {
609
+ "epoch": 2.2750929368029738,
610
+ "grad_norm": 1.561276115932866,
611
+ "learning_rate": 5e-06,
612
+ "loss": 0.3909,
613
+ "step": 840
614
+ },
615
+ {
616
+ "epoch": 2.3021290976681312,
617
+ "grad_norm": 1.7787495204510098,
618
+ "learning_rate": 5e-06,
619
+ "loss": 0.3866,
620
+ "step": 850
621
+ },
622
+ {
623
+ "epoch": 2.3291652585332883,
624
+ "grad_norm": 1.5802735443144562,
625
+ "learning_rate": 5e-06,
626
+ "loss": 0.3896,
627
+ "step": 860
628
+ },
629
+ {
630
+ "epoch": 2.3562014193984453,
631
+ "grad_norm": 1.5469387511116455,
632
+ "learning_rate": 5e-06,
633
+ "loss": 0.3948,
634
+ "step": 870
635
+ },
636
+ {
637
+ "epoch": 2.3832375802636028,
638
+ "grad_norm": 1.6780934080456225,
639
+ "learning_rate": 5e-06,
640
+ "loss": 0.3937,
641
+ "step": 880
642
+ },
643
+ {
644
+ "epoch": 2.41027374112876,
645
+ "grad_norm": 1.5538724349535749,
646
+ "learning_rate": 5e-06,
647
+ "loss": 0.3933,
648
+ "step": 890
649
+ },
650
+ {
651
+ "epoch": 2.437309901993917,
652
+ "grad_norm": 1.6919213854745063,
653
+ "learning_rate": 5e-06,
654
+ "loss": 0.3927,
655
+ "step": 900
656
+ },
657
+ {
658
+ "epoch": 2.464346062859074,
659
+ "grad_norm": 1.6467399942324181,
660
+ "learning_rate": 5e-06,
661
+ "loss": 0.3916,
662
+ "step": 910
663
+ },
664
+ {
665
+ "epoch": 2.4913822237242313,
666
+ "grad_norm": 1.5494538660407549,
667
+ "learning_rate": 5e-06,
668
+ "loss": 0.393,
669
+ "step": 920
670
+ },
671
+ {
672
+ "epoch": 2.5184183845893884,
673
+ "grad_norm": 1.6061582119048823,
674
+ "learning_rate": 5e-06,
675
+ "loss": 0.392,
676
+ "step": 930
677
+ },
678
+ {
679
+ "epoch": 2.5454545454545454,
680
+ "grad_norm": 1.5497867717979459,
681
+ "learning_rate": 5e-06,
682
+ "loss": 0.3996,
683
+ "step": 940
684
+ },
685
+ {
686
+ "epoch": 2.5724907063197024,
687
+ "grad_norm": 1.8696325727031804,
688
+ "learning_rate": 5e-06,
689
+ "loss": 0.3949,
690
+ "step": 950
691
+ },
692
+ {
693
+ "epoch": 2.59952686718486,
694
+ "grad_norm": 1.6545052949984496,
695
+ "learning_rate": 5e-06,
696
+ "loss": 0.4002,
697
+ "step": 960
698
+ },
699
+ {
700
+ "epoch": 2.626563028050017,
701
+ "grad_norm": 1.5124642558655546,
702
+ "learning_rate": 5e-06,
703
+ "loss": 0.3988,
704
+ "step": 970
705
+ },
706
+ {
707
+ "epoch": 2.653599188915174,
708
+ "grad_norm": 1.559510310440385,
709
+ "learning_rate": 5e-06,
710
+ "loss": 0.4024,
711
+ "step": 980
712
+ },
713
+ {
714
+ "epoch": 2.6806353497803315,
715
+ "grad_norm": 1.676052966396514,
716
+ "learning_rate": 5e-06,
717
+ "loss": 0.4007,
718
+ "step": 990
719
+ },
720
+ {
721
+ "epoch": 2.7076715106454885,
722
+ "grad_norm": 2.1046446839691,
723
+ "learning_rate": 5e-06,
724
+ "loss": 0.3952,
725
+ "step": 1000
726
+ },
727
+ {
728
+ "epoch": 2.7347076715106455,
729
+ "grad_norm": 2.0451177849286464,
730
+ "learning_rate": 5e-06,
731
+ "loss": 0.4021,
732
+ "step": 1010
733
+ },
734
+ {
735
+ "epoch": 2.7617438323758026,
736
+ "grad_norm": 1.6691707308120913,
737
+ "learning_rate": 5e-06,
738
+ "loss": 0.4046,
739
+ "step": 1020
740
+ },
741
+ {
742
+ "epoch": 2.7887799932409596,
743
+ "grad_norm": 1.4465197945931527,
744
+ "learning_rate": 5e-06,
745
+ "loss": 0.4022,
746
+ "step": 1030
747
+ },
748
+ {
749
+ "epoch": 2.815816154106117,
750
+ "grad_norm": 1.4351190888202614,
751
+ "learning_rate": 5e-06,
752
+ "loss": 0.4014,
753
+ "step": 1040
754
+ },
755
+ {
756
+ "epoch": 2.842852314971274,
757
+ "grad_norm": 1.485706611946695,
758
+ "learning_rate": 5e-06,
759
+ "loss": 0.4034,
760
+ "step": 1050
761
+ },
762
+ {
763
+ "epoch": 2.869888475836431,
764
+ "grad_norm": 1.4860832182563435,
765
+ "learning_rate": 5e-06,
766
+ "loss": 0.3996,
767
+ "step": 1060
768
+ },
769
+ {
770
+ "epoch": 2.8969246367015886,
771
+ "grad_norm": 1.5630718102725172,
772
+ "learning_rate": 5e-06,
773
+ "loss": 0.4077,
774
+ "step": 1070
775
+ },
776
+ {
777
+ "epoch": 2.9239607975667457,
778
+ "grad_norm": 1.4737818975824717,
779
+ "learning_rate": 5e-06,
780
+ "loss": 0.4027,
781
+ "step": 1080
782
+ },
783
+ {
784
+ "epoch": 2.9509969584319027,
785
+ "grad_norm": 1.5487795543993597,
786
+ "learning_rate": 5e-06,
787
+ "loss": 0.406,
788
+ "step": 1090
789
+ },
790
+ {
791
+ "epoch": 2.9780331192970597,
792
+ "grad_norm": 1.3628828414709748,
793
+ "learning_rate": 5e-06,
794
+ "loss": 0.4047,
795
+ "step": 1100
796
+ },
797
+ {
798
+ "epoch": 2.99695843190267,
799
+ "eval_loss": 0.07649385929107666,
800
+ "eval_runtime": 385.8472,
801
+ "eval_samples_per_second": 25.826,
802
+ "eval_steps_per_second": 0.404,
803
+ "step": 1107
804
+ },
805
+ {
806
+ "epoch": 2.99695843190267,
807
+ "step": 1107,
808
+ "total_flos": 1854056851046400.0,
809
+ "train_loss": 0.49424083467636865,
810
+ "train_runtime": 64065.8676,
811
+ "train_samples_per_second": 8.866,
812
+ "train_steps_per_second": 0.017
813
+ }
814
+ ],
815
+ "logging_steps": 10,
816
+ "max_steps": 1107,
817
+ "num_input_tokens_seen": 0,
818
+ "num_train_epochs": 3,
819
+ "save_steps": 500,
820
+ "stateful_callbacks": {
821
+ "TrainerControl": {
822
+ "args": {
823
+ "should_epoch_stop": false,
824
+ "should_evaluate": false,
825
+ "should_log": false,
826
+ "should_save": true,
827
+ "should_training_stop": true
828
+ },
829
+ "attributes": {}
830
+ }
831
+ },
832
+ "total_flos": 1854056851046400.0,
833
+ "train_batch_size": 8,
834
+ "trial_name": null,
835
+ "trial_params": null
836
+ }
training_eval_loss.png ADDED
training_loss.png ADDED