sedrickkeh commited on
Commit
f0d949d
1 Parent(s): 4c8a6c1

End of training

Browse files
README.md CHANGED
@@ -4,6 +4,7 @@ license: llama3.1
4
  base_model: meta-llama/Llama-3.1-8B
5
  tags:
6
  - llama-factory
 
7
  - generated_from_trainer
8
  model-index:
9
  - name: oh_v1_w_v3_alpaca_threshold90_it
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
15
 
16
  # oh_v1_w_v3_alpaca_threshold90_it
17
 
18
- This model is a fine-tuned version of [meta-llama/Llama-3.1-8B](https://huggingface.co/meta-llama/Llama-3.1-8B) on an unknown dataset.
19
  It achieves the following results on the evaluation set:
20
  - Loss: 0.5371
21
 
 
4
  base_model: meta-llama/Llama-3.1-8B
5
  tags:
6
  - llama-factory
7
+ - full
8
  - generated_from_trainer
9
  model-index:
10
  - name: oh_v1_w_v3_alpaca_threshold90_it
 
16
 
17
  # oh_v1_w_v3_alpaca_threshold90_it
18
 
19
+ This model is a fine-tuned version of [meta-llama/Llama-3.1-8B](https://huggingface.co/meta-llama/Llama-3.1-8B) on the mlfoundations-dev/oh_v1_w_v3_alpaca_threshold90_it dataset.
20
  It achieves the following results on the evaluation set:
21
  - Loss: 0.5371
22
 
all_results.json ADDED
@@ -0,0 +1,12 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.9981447124304266,
3
+ "eval_loss": 0.537095308303833,
4
+ "eval_runtime": 218.4519,
5
+ "eval_samples_per_second": 49.86,
6
+ "eval_steps_per_second": 0.394,
7
+ "total_flos": 2029726382161920.0,
8
+ "train_loss": 0.5400018749063952,
9
+ "train_runtime": 36478.2162,
10
+ "train_samples_per_second": 17.019,
11
+ "train_steps_per_second": 0.033
12
+ }
eval_results.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.9981447124304266,
3
+ "eval_loss": 0.537095308303833,
4
+ "eval_runtime": 218.4519,
5
+ "eval_samples_per_second": 49.86,
6
+ "eval_steps_per_second": 0.394
7
+ }
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.9981447124304266,
3
+ "total_flos": 2029726382161920.0,
4
+ "train_loss": 0.5400018749063952,
5
+ "train_runtime": 36478.2162,
6
+ "train_samples_per_second": 17.019,
7
+ "train_steps_per_second": 0.033
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,913 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 2.9981447124304266,
5
+ "eval_steps": 500,
6
+ "global_step": 1212,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.024737167594310452,
13
+ "grad_norm": 3.4830790839772257,
14
+ "learning_rate": 5e-06,
15
+ "loss": 0.8916,
16
+ "step": 10
17
+ },
18
+ {
19
+ "epoch": 0.049474335188620905,
20
+ "grad_norm": 6.942725060810088,
21
+ "learning_rate": 5e-06,
22
+ "loss": 0.7677,
23
+ "step": 20
24
+ },
25
+ {
26
+ "epoch": 0.07421150278293136,
27
+ "grad_norm": 1.8951062871081399,
28
+ "learning_rate": 5e-06,
29
+ "loss": 0.7288,
30
+ "step": 30
31
+ },
32
+ {
33
+ "epoch": 0.09894867037724181,
34
+ "grad_norm": 1.9357008010716068,
35
+ "learning_rate": 5e-06,
36
+ "loss": 0.7031,
37
+ "step": 40
38
+ },
39
+ {
40
+ "epoch": 0.12368583797155226,
41
+ "grad_norm": 0.9499312221395673,
42
+ "learning_rate": 5e-06,
43
+ "loss": 0.6844,
44
+ "step": 50
45
+ },
46
+ {
47
+ "epoch": 0.14842300556586271,
48
+ "grad_norm": 0.7455987403828673,
49
+ "learning_rate": 5e-06,
50
+ "loss": 0.6598,
51
+ "step": 60
52
+ },
53
+ {
54
+ "epoch": 0.17316017316017315,
55
+ "grad_norm": 0.6470160824911549,
56
+ "learning_rate": 5e-06,
57
+ "loss": 0.6485,
58
+ "step": 70
59
+ },
60
+ {
61
+ "epoch": 0.19789734075448362,
62
+ "grad_norm": 0.6635217517468123,
63
+ "learning_rate": 5e-06,
64
+ "loss": 0.6398,
65
+ "step": 80
66
+ },
67
+ {
68
+ "epoch": 0.22263450834879406,
69
+ "grad_norm": 0.5563048334006576,
70
+ "learning_rate": 5e-06,
71
+ "loss": 0.6378,
72
+ "step": 90
73
+ },
74
+ {
75
+ "epoch": 0.24737167594310452,
76
+ "grad_norm": 0.685513800699662,
77
+ "learning_rate": 5e-06,
78
+ "loss": 0.6307,
79
+ "step": 100
80
+ },
81
+ {
82
+ "epoch": 0.272108843537415,
83
+ "grad_norm": 0.8408848874695213,
84
+ "learning_rate": 5e-06,
85
+ "loss": 0.6176,
86
+ "step": 110
87
+ },
88
+ {
89
+ "epoch": 0.29684601113172543,
90
+ "grad_norm": 0.6773751764933612,
91
+ "learning_rate": 5e-06,
92
+ "loss": 0.6178,
93
+ "step": 120
94
+ },
95
+ {
96
+ "epoch": 0.32158317872603587,
97
+ "grad_norm": 0.7027587036277353,
98
+ "learning_rate": 5e-06,
99
+ "loss": 0.6133,
100
+ "step": 130
101
+ },
102
+ {
103
+ "epoch": 0.3463203463203463,
104
+ "grad_norm": 0.6157245730205899,
105
+ "learning_rate": 5e-06,
106
+ "loss": 0.6092,
107
+ "step": 140
108
+ },
109
+ {
110
+ "epoch": 0.37105751391465674,
111
+ "grad_norm": 0.5802260294344969,
112
+ "learning_rate": 5e-06,
113
+ "loss": 0.6046,
114
+ "step": 150
115
+ },
116
+ {
117
+ "epoch": 0.39579468150896724,
118
+ "grad_norm": 0.6344807037013861,
119
+ "learning_rate": 5e-06,
120
+ "loss": 0.6009,
121
+ "step": 160
122
+ },
123
+ {
124
+ "epoch": 0.4205318491032777,
125
+ "grad_norm": 0.48480545471576164,
126
+ "learning_rate": 5e-06,
127
+ "loss": 0.6082,
128
+ "step": 170
129
+ },
130
+ {
131
+ "epoch": 0.4452690166975881,
132
+ "grad_norm": 0.6380631405566793,
133
+ "learning_rate": 5e-06,
134
+ "loss": 0.6004,
135
+ "step": 180
136
+ },
137
+ {
138
+ "epoch": 0.47000618429189855,
139
+ "grad_norm": 0.5505169198792789,
140
+ "learning_rate": 5e-06,
141
+ "loss": 0.5994,
142
+ "step": 190
143
+ },
144
+ {
145
+ "epoch": 0.49474335188620905,
146
+ "grad_norm": 0.6242896844382176,
147
+ "learning_rate": 5e-06,
148
+ "loss": 0.5985,
149
+ "step": 200
150
+ },
151
+ {
152
+ "epoch": 0.5194805194805194,
153
+ "grad_norm": 0.5255950166858494,
154
+ "learning_rate": 5e-06,
155
+ "loss": 0.6022,
156
+ "step": 210
157
+ },
158
+ {
159
+ "epoch": 0.54421768707483,
160
+ "grad_norm": 0.5516723643293726,
161
+ "learning_rate": 5e-06,
162
+ "loss": 0.5927,
163
+ "step": 220
164
+ },
165
+ {
166
+ "epoch": 0.5689548546691404,
167
+ "grad_norm": 0.584042711614131,
168
+ "learning_rate": 5e-06,
169
+ "loss": 0.5925,
170
+ "step": 230
171
+ },
172
+ {
173
+ "epoch": 0.5936920222634509,
174
+ "grad_norm": 0.6260676537261808,
175
+ "learning_rate": 5e-06,
176
+ "loss": 0.5928,
177
+ "step": 240
178
+ },
179
+ {
180
+ "epoch": 0.6184291898577613,
181
+ "grad_norm": 0.5176619963131881,
182
+ "learning_rate": 5e-06,
183
+ "loss": 0.5858,
184
+ "step": 250
185
+ },
186
+ {
187
+ "epoch": 0.6431663574520717,
188
+ "grad_norm": 0.5024236252235582,
189
+ "learning_rate": 5e-06,
190
+ "loss": 0.5879,
191
+ "step": 260
192
+ },
193
+ {
194
+ "epoch": 0.6679035250463822,
195
+ "grad_norm": 0.6803247050243845,
196
+ "learning_rate": 5e-06,
197
+ "loss": 0.5844,
198
+ "step": 270
199
+ },
200
+ {
201
+ "epoch": 0.6926406926406926,
202
+ "grad_norm": 0.6015791834486387,
203
+ "learning_rate": 5e-06,
204
+ "loss": 0.5755,
205
+ "step": 280
206
+ },
207
+ {
208
+ "epoch": 0.717377860235003,
209
+ "grad_norm": 0.5299909658526826,
210
+ "learning_rate": 5e-06,
211
+ "loss": 0.5843,
212
+ "step": 290
213
+ },
214
+ {
215
+ "epoch": 0.7421150278293135,
216
+ "grad_norm": 0.5088932556423367,
217
+ "learning_rate": 5e-06,
218
+ "loss": 0.5864,
219
+ "step": 300
220
+ },
221
+ {
222
+ "epoch": 0.766852195423624,
223
+ "grad_norm": 0.5964832923195739,
224
+ "learning_rate": 5e-06,
225
+ "loss": 0.5822,
226
+ "step": 310
227
+ },
228
+ {
229
+ "epoch": 0.7915893630179345,
230
+ "grad_norm": 0.5860489284099033,
231
+ "learning_rate": 5e-06,
232
+ "loss": 0.5727,
233
+ "step": 320
234
+ },
235
+ {
236
+ "epoch": 0.8163265306122449,
237
+ "grad_norm": 0.5219580208519626,
238
+ "learning_rate": 5e-06,
239
+ "loss": 0.5794,
240
+ "step": 330
241
+ },
242
+ {
243
+ "epoch": 0.8410636982065554,
244
+ "grad_norm": 0.5269259293045121,
245
+ "learning_rate": 5e-06,
246
+ "loss": 0.5793,
247
+ "step": 340
248
+ },
249
+ {
250
+ "epoch": 0.8658008658008658,
251
+ "grad_norm": 0.5291953882291388,
252
+ "learning_rate": 5e-06,
253
+ "loss": 0.5744,
254
+ "step": 350
255
+ },
256
+ {
257
+ "epoch": 0.8905380333951762,
258
+ "grad_norm": 0.521358302515632,
259
+ "learning_rate": 5e-06,
260
+ "loss": 0.5707,
261
+ "step": 360
262
+ },
263
+ {
264
+ "epoch": 0.9152752009894867,
265
+ "grad_norm": 0.5774675957700983,
266
+ "learning_rate": 5e-06,
267
+ "loss": 0.5765,
268
+ "step": 370
269
+ },
270
+ {
271
+ "epoch": 0.9400123685837971,
272
+ "grad_norm": 0.5172232609124243,
273
+ "learning_rate": 5e-06,
274
+ "loss": 0.5714,
275
+ "step": 380
276
+ },
277
+ {
278
+ "epoch": 0.9647495361781077,
279
+ "grad_norm": 0.5240273423167071,
280
+ "learning_rate": 5e-06,
281
+ "loss": 0.5682,
282
+ "step": 390
283
+ },
284
+ {
285
+ "epoch": 0.9894867037724181,
286
+ "grad_norm": 0.6149272559928857,
287
+ "learning_rate": 5e-06,
288
+ "loss": 0.5719,
289
+ "step": 400
290
+ },
291
+ {
292
+ "epoch": 0.9993815708101422,
293
+ "eval_loss": 0.5683358907699585,
294
+ "eval_runtime": 219.0047,
295
+ "eval_samples_per_second": 49.734,
296
+ "eval_steps_per_second": 0.393,
297
+ "step": 404
298
+ },
299
+ {
300
+ "epoch": 1.0142238713667284,
301
+ "grad_norm": 0.5073078170608074,
302
+ "learning_rate": 5e-06,
303
+ "loss": 0.547,
304
+ "step": 410
305
+ },
306
+ {
307
+ "epoch": 1.0389610389610389,
308
+ "grad_norm": 0.6437533127258526,
309
+ "learning_rate": 5e-06,
310
+ "loss": 0.5316,
311
+ "step": 420
312
+ },
313
+ {
314
+ "epoch": 1.0636982065553493,
315
+ "grad_norm": 0.6100879209869073,
316
+ "learning_rate": 5e-06,
317
+ "loss": 0.5301,
318
+ "step": 430
319
+ },
320
+ {
321
+ "epoch": 1.08843537414966,
322
+ "grad_norm": 0.48108127450705374,
323
+ "learning_rate": 5e-06,
324
+ "loss": 0.5294,
325
+ "step": 440
326
+ },
327
+ {
328
+ "epoch": 1.1131725417439704,
329
+ "grad_norm": 0.5868928526612365,
330
+ "learning_rate": 5e-06,
331
+ "loss": 0.5332,
332
+ "step": 450
333
+ },
334
+ {
335
+ "epoch": 1.1379097093382808,
336
+ "grad_norm": 0.6202895625151273,
337
+ "learning_rate": 5e-06,
338
+ "loss": 0.535,
339
+ "step": 460
340
+ },
341
+ {
342
+ "epoch": 1.1626468769325913,
343
+ "grad_norm": 0.5075954994852542,
344
+ "learning_rate": 5e-06,
345
+ "loss": 0.5326,
346
+ "step": 470
347
+ },
348
+ {
349
+ "epoch": 1.1873840445269017,
350
+ "grad_norm": 0.4728741075680983,
351
+ "learning_rate": 5e-06,
352
+ "loss": 0.532,
353
+ "step": 480
354
+ },
355
+ {
356
+ "epoch": 1.2121212121212122,
357
+ "grad_norm": 0.5081500756875807,
358
+ "learning_rate": 5e-06,
359
+ "loss": 0.5348,
360
+ "step": 490
361
+ },
362
+ {
363
+ "epoch": 1.2368583797155226,
364
+ "grad_norm": 0.6910919006687529,
365
+ "learning_rate": 5e-06,
366
+ "loss": 0.5197,
367
+ "step": 500
368
+ },
369
+ {
370
+ "epoch": 1.261595547309833,
371
+ "grad_norm": 0.5501274010525514,
372
+ "learning_rate": 5e-06,
373
+ "loss": 0.5222,
374
+ "step": 510
375
+ },
376
+ {
377
+ "epoch": 1.2863327149041435,
378
+ "grad_norm": 0.7316007600357546,
379
+ "learning_rate": 5e-06,
380
+ "loss": 0.5335,
381
+ "step": 520
382
+ },
383
+ {
384
+ "epoch": 1.311069882498454,
385
+ "grad_norm": 0.5057200127850379,
386
+ "learning_rate": 5e-06,
387
+ "loss": 0.5288,
388
+ "step": 530
389
+ },
390
+ {
391
+ "epoch": 1.3358070500927643,
392
+ "grad_norm": 0.5719065403986006,
393
+ "learning_rate": 5e-06,
394
+ "loss": 0.529,
395
+ "step": 540
396
+ },
397
+ {
398
+ "epoch": 1.3605442176870748,
399
+ "grad_norm": 0.5660341689340015,
400
+ "learning_rate": 5e-06,
401
+ "loss": 0.5282,
402
+ "step": 550
403
+ },
404
+ {
405
+ "epoch": 1.3852813852813852,
406
+ "grad_norm": 0.7015838028229988,
407
+ "learning_rate": 5e-06,
408
+ "loss": 0.5296,
409
+ "step": 560
410
+ },
411
+ {
412
+ "epoch": 1.4100185528756957,
413
+ "grad_norm": 0.5640160639184895,
414
+ "learning_rate": 5e-06,
415
+ "loss": 0.531,
416
+ "step": 570
417
+ },
418
+ {
419
+ "epoch": 1.434755720470006,
420
+ "grad_norm": 0.5241187267275058,
421
+ "learning_rate": 5e-06,
422
+ "loss": 0.5277,
423
+ "step": 580
424
+ },
425
+ {
426
+ "epoch": 1.4594928880643168,
427
+ "grad_norm": 0.5322661450321486,
428
+ "learning_rate": 5e-06,
429
+ "loss": 0.5223,
430
+ "step": 590
431
+ },
432
+ {
433
+ "epoch": 1.4842300556586272,
434
+ "grad_norm": 0.5414175662054105,
435
+ "learning_rate": 5e-06,
436
+ "loss": 0.5253,
437
+ "step": 600
438
+ },
439
+ {
440
+ "epoch": 1.5089672232529376,
441
+ "grad_norm": 0.5120018621643256,
442
+ "learning_rate": 5e-06,
443
+ "loss": 0.5171,
444
+ "step": 610
445
+ },
446
+ {
447
+ "epoch": 1.533704390847248,
448
+ "grad_norm": 0.5872164754716582,
449
+ "learning_rate": 5e-06,
450
+ "loss": 0.5263,
451
+ "step": 620
452
+ },
453
+ {
454
+ "epoch": 1.5584415584415585,
455
+ "grad_norm": 0.4661784659324605,
456
+ "learning_rate": 5e-06,
457
+ "loss": 0.5217,
458
+ "step": 630
459
+ },
460
+ {
461
+ "epoch": 1.583178726035869,
462
+ "grad_norm": 0.464230352567236,
463
+ "learning_rate": 5e-06,
464
+ "loss": 0.5204,
465
+ "step": 640
466
+ },
467
+ {
468
+ "epoch": 1.6079158936301794,
469
+ "grad_norm": 0.5032827937357744,
470
+ "learning_rate": 5e-06,
471
+ "loss": 0.5216,
472
+ "step": 650
473
+ },
474
+ {
475
+ "epoch": 1.6326530612244898,
476
+ "grad_norm": 0.4996792160145028,
477
+ "learning_rate": 5e-06,
478
+ "loss": 0.5218,
479
+ "step": 660
480
+ },
481
+ {
482
+ "epoch": 1.6573902288188003,
483
+ "grad_norm": 0.5271632514608294,
484
+ "learning_rate": 5e-06,
485
+ "loss": 0.5226,
486
+ "step": 670
487
+ },
488
+ {
489
+ "epoch": 1.6821273964131107,
490
+ "grad_norm": 0.49347623420693476,
491
+ "learning_rate": 5e-06,
492
+ "loss": 0.5193,
493
+ "step": 680
494
+ },
495
+ {
496
+ "epoch": 1.7068645640074211,
497
+ "grad_norm": 0.6049309413421875,
498
+ "learning_rate": 5e-06,
499
+ "loss": 0.5247,
500
+ "step": 690
501
+ },
502
+ {
503
+ "epoch": 1.7316017316017316,
504
+ "grad_norm": 0.5973360853484692,
505
+ "learning_rate": 5e-06,
506
+ "loss": 0.516,
507
+ "step": 700
508
+ },
509
+ {
510
+ "epoch": 1.756338899196042,
511
+ "grad_norm": 0.49072766464621603,
512
+ "learning_rate": 5e-06,
513
+ "loss": 0.5179,
514
+ "step": 710
515
+ },
516
+ {
517
+ "epoch": 1.7810760667903525,
518
+ "grad_norm": 0.5091642966772914,
519
+ "learning_rate": 5e-06,
520
+ "loss": 0.5181,
521
+ "step": 720
522
+ },
523
+ {
524
+ "epoch": 1.805813234384663,
525
+ "grad_norm": 0.5801588473122157,
526
+ "learning_rate": 5e-06,
527
+ "loss": 0.5177,
528
+ "step": 730
529
+ },
530
+ {
531
+ "epoch": 1.8305504019789733,
532
+ "grad_norm": 0.48516680968832,
533
+ "learning_rate": 5e-06,
534
+ "loss": 0.5231,
535
+ "step": 740
536
+ },
537
+ {
538
+ "epoch": 1.8552875695732838,
539
+ "grad_norm": 0.46039086459013556,
540
+ "learning_rate": 5e-06,
541
+ "loss": 0.5249,
542
+ "step": 750
543
+ },
544
+ {
545
+ "epoch": 1.8800247371675942,
546
+ "grad_norm": 0.5083902833654592,
547
+ "learning_rate": 5e-06,
548
+ "loss": 0.518,
549
+ "step": 760
550
+ },
551
+ {
552
+ "epoch": 1.9047619047619047,
553
+ "grad_norm": 0.5109759169478353,
554
+ "learning_rate": 5e-06,
555
+ "loss": 0.5177,
556
+ "step": 770
557
+ },
558
+ {
559
+ "epoch": 1.929499072356215,
560
+ "grad_norm": 0.4767006976535441,
561
+ "learning_rate": 5e-06,
562
+ "loss": 0.5132,
563
+ "step": 780
564
+ },
565
+ {
566
+ "epoch": 1.9542362399505255,
567
+ "grad_norm": 0.5035389849175771,
568
+ "learning_rate": 5e-06,
569
+ "loss": 0.5181,
570
+ "step": 790
571
+ },
572
+ {
573
+ "epoch": 1.978973407544836,
574
+ "grad_norm": 0.5072097192727141,
575
+ "learning_rate": 5e-06,
576
+ "loss": 0.5166,
577
+ "step": 800
578
+ },
579
+ {
580
+ "epoch": 1.9987631416202845,
581
+ "eval_loss": 0.5430302023887634,
582
+ "eval_runtime": 219.3095,
583
+ "eval_samples_per_second": 49.665,
584
+ "eval_steps_per_second": 0.392,
585
+ "step": 808
586
+ },
587
+ {
588
+ "epoch": 2.0037105751391464,
589
+ "grad_norm": 0.5742594976510665,
590
+ "learning_rate": 5e-06,
591
+ "loss": 0.5105,
592
+ "step": 810
593
+ },
594
+ {
595
+ "epoch": 2.028447742733457,
596
+ "grad_norm": 0.5679263759068595,
597
+ "learning_rate": 5e-06,
598
+ "loss": 0.4801,
599
+ "step": 820
600
+ },
601
+ {
602
+ "epoch": 2.0531849103277673,
603
+ "grad_norm": 0.5672377634558515,
604
+ "learning_rate": 5e-06,
605
+ "loss": 0.4818,
606
+ "step": 830
607
+ },
608
+ {
609
+ "epoch": 2.0779220779220777,
610
+ "grad_norm": 0.6793815683511014,
611
+ "learning_rate": 5e-06,
612
+ "loss": 0.4756,
613
+ "step": 840
614
+ },
615
+ {
616
+ "epoch": 2.102659245516388,
617
+ "grad_norm": 0.5158458141043201,
618
+ "learning_rate": 5e-06,
619
+ "loss": 0.4768,
620
+ "step": 850
621
+ },
622
+ {
623
+ "epoch": 2.1273964131106986,
624
+ "grad_norm": 0.6722479592255648,
625
+ "learning_rate": 5e-06,
626
+ "loss": 0.481,
627
+ "step": 860
628
+ },
629
+ {
630
+ "epoch": 2.1521335807050095,
631
+ "grad_norm": 0.622090680752648,
632
+ "learning_rate": 5e-06,
633
+ "loss": 0.4806,
634
+ "step": 870
635
+ },
636
+ {
637
+ "epoch": 2.17687074829932,
638
+ "grad_norm": 0.5446538279562969,
639
+ "learning_rate": 5e-06,
640
+ "loss": 0.4792,
641
+ "step": 880
642
+ },
643
+ {
644
+ "epoch": 2.2016079158936304,
645
+ "grad_norm": 0.9897157820333273,
646
+ "learning_rate": 5e-06,
647
+ "loss": 0.4743,
648
+ "step": 890
649
+ },
650
+ {
651
+ "epoch": 2.226345083487941,
652
+ "grad_norm": 0.5709797478235871,
653
+ "learning_rate": 5e-06,
654
+ "loss": 0.4745,
655
+ "step": 900
656
+ },
657
+ {
658
+ "epoch": 2.2510822510822512,
659
+ "grad_norm": 0.5508207898306552,
660
+ "learning_rate": 5e-06,
661
+ "loss": 0.4825,
662
+ "step": 910
663
+ },
664
+ {
665
+ "epoch": 2.2758194186765617,
666
+ "grad_norm": 0.4968712039591529,
667
+ "learning_rate": 5e-06,
668
+ "loss": 0.4805,
669
+ "step": 920
670
+ },
671
+ {
672
+ "epoch": 2.300556586270872,
673
+ "grad_norm": 0.5588550182736718,
674
+ "learning_rate": 5e-06,
675
+ "loss": 0.4816,
676
+ "step": 930
677
+ },
678
+ {
679
+ "epoch": 2.3252937538651826,
680
+ "grad_norm": 0.6560301930501052,
681
+ "learning_rate": 5e-06,
682
+ "loss": 0.4834,
683
+ "step": 940
684
+ },
685
+ {
686
+ "epoch": 2.350030921459493,
687
+ "grad_norm": 0.47583877726868273,
688
+ "learning_rate": 5e-06,
689
+ "loss": 0.4785,
690
+ "step": 950
691
+ },
692
+ {
693
+ "epoch": 2.3747680890538034,
694
+ "grad_norm": 0.5212062104260379,
695
+ "learning_rate": 5e-06,
696
+ "loss": 0.4787,
697
+ "step": 960
698
+ },
699
+ {
700
+ "epoch": 2.399505256648114,
701
+ "grad_norm": 0.47451320868129626,
702
+ "learning_rate": 5e-06,
703
+ "loss": 0.4829,
704
+ "step": 970
705
+ },
706
+ {
707
+ "epoch": 2.4242424242424243,
708
+ "grad_norm": 0.4830738101162483,
709
+ "learning_rate": 5e-06,
710
+ "loss": 0.4843,
711
+ "step": 980
712
+ },
713
+ {
714
+ "epoch": 2.4489795918367347,
715
+ "grad_norm": 0.5182106761363315,
716
+ "learning_rate": 5e-06,
717
+ "loss": 0.4757,
718
+ "step": 990
719
+ },
720
+ {
721
+ "epoch": 2.473716759431045,
722
+ "grad_norm": 0.5153536365130167,
723
+ "learning_rate": 5e-06,
724
+ "loss": 0.4768,
725
+ "step": 1000
726
+ },
727
+ {
728
+ "epoch": 2.4984539270253556,
729
+ "grad_norm": 0.5417873882437457,
730
+ "learning_rate": 5e-06,
731
+ "loss": 0.4782,
732
+ "step": 1010
733
+ },
734
+ {
735
+ "epoch": 2.523191094619666,
736
+ "grad_norm": 0.5527273801359924,
737
+ "learning_rate": 5e-06,
738
+ "loss": 0.475,
739
+ "step": 1020
740
+ },
741
+ {
742
+ "epoch": 2.5479282622139765,
743
+ "grad_norm": 0.5335187564602286,
744
+ "learning_rate": 5e-06,
745
+ "loss": 0.4792,
746
+ "step": 1030
747
+ },
748
+ {
749
+ "epoch": 2.572665429808287,
750
+ "grad_norm": 0.5000885945895702,
751
+ "learning_rate": 5e-06,
752
+ "loss": 0.4769,
753
+ "step": 1040
754
+ },
755
+ {
756
+ "epoch": 2.5974025974025974,
757
+ "grad_norm": 0.518335000088379,
758
+ "learning_rate": 5e-06,
759
+ "loss": 0.4771,
760
+ "step": 1050
761
+ },
762
+ {
763
+ "epoch": 2.622139764996908,
764
+ "grad_norm": 0.5506780400189991,
765
+ "learning_rate": 5e-06,
766
+ "loss": 0.4813,
767
+ "step": 1060
768
+ },
769
+ {
770
+ "epoch": 2.6468769325912183,
771
+ "grad_norm": 0.5009935325893752,
772
+ "learning_rate": 5e-06,
773
+ "loss": 0.479,
774
+ "step": 1070
775
+ },
776
+ {
777
+ "epoch": 2.6716141001855287,
778
+ "grad_norm": 0.5265944850039301,
779
+ "learning_rate": 5e-06,
780
+ "loss": 0.4808,
781
+ "step": 1080
782
+ },
783
+ {
784
+ "epoch": 2.696351267779839,
785
+ "grad_norm": 0.5426869881704958,
786
+ "learning_rate": 5e-06,
787
+ "loss": 0.4798,
788
+ "step": 1090
789
+ },
790
+ {
791
+ "epoch": 2.7210884353741496,
792
+ "grad_norm": 0.6372802273548948,
793
+ "learning_rate": 5e-06,
794
+ "loss": 0.4824,
795
+ "step": 1100
796
+ },
797
+ {
798
+ "epoch": 2.74582560296846,
799
+ "grad_norm": 0.5506322753225094,
800
+ "learning_rate": 5e-06,
801
+ "loss": 0.4748,
802
+ "step": 1110
803
+ },
804
+ {
805
+ "epoch": 2.7705627705627704,
806
+ "grad_norm": 0.6432775550069307,
807
+ "learning_rate": 5e-06,
808
+ "loss": 0.4785,
809
+ "step": 1120
810
+ },
811
+ {
812
+ "epoch": 2.795299938157081,
813
+ "grad_norm": 0.5422596013158468,
814
+ "learning_rate": 5e-06,
815
+ "loss": 0.4756,
816
+ "step": 1130
817
+ },
818
+ {
819
+ "epoch": 2.8200371057513913,
820
+ "grad_norm": 0.5766402488022242,
821
+ "learning_rate": 5e-06,
822
+ "loss": 0.4813,
823
+ "step": 1140
824
+ },
825
+ {
826
+ "epoch": 2.8447742733457018,
827
+ "grad_norm": 0.6098697051701109,
828
+ "learning_rate": 5e-06,
829
+ "loss": 0.4813,
830
+ "step": 1150
831
+ },
832
+ {
833
+ "epoch": 2.869511440940012,
834
+ "grad_norm": 0.6756833035066511,
835
+ "learning_rate": 5e-06,
836
+ "loss": 0.4774,
837
+ "step": 1160
838
+ },
839
+ {
840
+ "epoch": 2.8942486085343226,
841
+ "grad_norm": 0.6408060819703033,
842
+ "learning_rate": 5e-06,
843
+ "loss": 0.4762,
844
+ "step": 1170
845
+ },
846
+ {
847
+ "epoch": 2.9189857761286335,
848
+ "grad_norm": 0.5130121997574635,
849
+ "learning_rate": 5e-06,
850
+ "loss": 0.4764,
851
+ "step": 1180
852
+ },
853
+ {
854
+ "epoch": 2.9437229437229435,
855
+ "grad_norm": 0.5162312463500585,
856
+ "learning_rate": 5e-06,
857
+ "loss": 0.4777,
858
+ "step": 1190
859
+ },
860
+ {
861
+ "epoch": 2.9684601113172544,
862
+ "grad_norm": 0.5924512448077986,
863
+ "learning_rate": 5e-06,
864
+ "loss": 0.4791,
865
+ "step": 1200
866
+ },
867
+ {
868
+ "epoch": 2.9931972789115644,
869
+ "grad_norm": 0.502845968511808,
870
+ "learning_rate": 5e-06,
871
+ "loss": 0.4761,
872
+ "step": 1210
873
+ },
874
+ {
875
+ "epoch": 2.9981447124304266,
876
+ "eval_loss": 0.537095308303833,
877
+ "eval_runtime": 219.5171,
878
+ "eval_samples_per_second": 49.618,
879
+ "eval_steps_per_second": 0.392,
880
+ "step": 1212
881
+ },
882
+ {
883
+ "epoch": 2.9981447124304266,
884
+ "step": 1212,
885
+ "total_flos": 2029726382161920.0,
886
+ "train_loss": 0.5400018749063952,
887
+ "train_runtime": 36478.2162,
888
+ "train_samples_per_second": 17.019,
889
+ "train_steps_per_second": 0.033
890
+ }
891
+ ],
892
+ "logging_steps": 10,
893
+ "max_steps": 1212,
894
+ "num_input_tokens_seen": 0,
895
+ "num_train_epochs": 3,
896
+ "save_steps": 500,
897
+ "stateful_callbacks": {
898
+ "TrainerControl": {
899
+ "args": {
900
+ "should_epoch_stop": false,
901
+ "should_evaluate": false,
902
+ "should_log": false,
903
+ "should_save": true,
904
+ "should_training_stop": true
905
+ },
906
+ "attributes": {}
907
+ }
908
+ },
909
+ "total_flos": 2029726382161920.0,
910
+ "train_batch_size": 8,
911
+ "trial_name": null,
912
+ "trial_params": null
913
+ }
training_eval_loss.png ADDED
training_loss.png ADDED