gsmyrnis commited on
Commit
3f107dc
1 Parent(s): ca075ab

End of training

Browse files
README.md CHANGED
@@ -4,6 +4,7 @@ license: llama3.1
4
  base_model: meta-llama/Meta-Llama-3.1-8B
5
  tags:
6
  - llama-factory
 
7
  - generated_from_trainer
8
  model-index:
9
  - name: llama3-1_8b_baseline_infinity_instruct
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
15
 
16
  # llama3-1_8b_baseline_infinity_instruct
17
 
18
- This model is a fine-tuned version of [meta-llama/Meta-Llama-3.1-8B](https://huggingface.co/meta-llama/Meta-Llama-3.1-8B) on an unknown dataset.
19
  It achieves the following results on the evaluation set:
20
  - Loss: 0.9010
21
 
 
4
  base_model: meta-llama/Meta-Llama-3.1-8B
5
  tags:
6
  - llama-factory
7
+ - full
8
  - generated_from_trainer
9
  model-index:
10
  - name: llama3-1_8b_baseline_infinity_instruct
 
16
 
17
  # llama3-1_8b_baseline_infinity_instruct
18
 
19
+ This model is a fine-tuned version of [meta-llama/Meta-Llama-3.1-8B](https://huggingface.co/meta-llama/Meta-Llama-3.1-8B) on the BAAI/Infinity-Instruct dataset.
20
  It achieves the following results on the evaluation set:
21
  - Loss: 0.9010
22
 
all_results.json ADDED
@@ -0,0 +1,12 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 3.0,
3
+ "eval_loss": 0.9010276198387146,
4
+ "eval_runtime": 39.042,
5
+ "eval_samples_per_second": 291.788,
6
+ "eval_steps_per_second": 1.153,
7
+ "total_flos": 5.9918252490593665e+19,
8
+ "train_loss": 0.8716646811452975,
9
+ "train_runtime": 8996.6792,
10
+ "train_samples_per_second": 72.17,
11
+ "train_steps_per_second": 0.141
12
+ }
eval_results.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 3.0,
3
+ "eval_loss": 0.9010276198387146,
4
+ "eval_runtime": 39.042,
5
+ "eval_samples_per_second": 291.788,
6
+ "eval_steps_per_second": 1.153
7
+ }
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 3.0,
3
+ "total_flos": 5.9918252490593665e+19,
4
+ "train_loss": 0.8716646811452975,
5
+ "train_runtime": 8996.6792,
6
+ "train_samples_per_second": 72.17,
7
+ "train_steps_per_second": 0.141
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,948 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 3.0,
5
+ "eval_steps": 500,
6
+ "global_step": 1269,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.02364066193853428,
13
+ "grad_norm": 1.7614086866378784,
14
+ "learning_rate": 5e-06,
15
+ "loss": 1.0709,
16
+ "step": 10
17
+ },
18
+ {
19
+ "epoch": 0.04728132387706856,
20
+ "grad_norm": 3.922980546951294,
21
+ "learning_rate": 5e-06,
22
+ "loss": 1.0131,
23
+ "step": 20
24
+ },
25
+ {
26
+ "epoch": 0.07092198581560284,
27
+ "grad_norm": 1.108715295791626,
28
+ "learning_rate": 5e-06,
29
+ "loss": 0.9972,
30
+ "step": 30
31
+ },
32
+ {
33
+ "epoch": 0.09456264775413711,
34
+ "grad_norm": 0.6008648872375488,
35
+ "learning_rate": 5e-06,
36
+ "loss": 0.983,
37
+ "step": 40
38
+ },
39
+ {
40
+ "epoch": 0.1182033096926714,
41
+ "grad_norm": 0.5573455095291138,
42
+ "learning_rate": 5e-06,
43
+ "loss": 0.9748,
44
+ "step": 50
45
+ },
46
+ {
47
+ "epoch": 0.14184397163120568,
48
+ "grad_norm": 0.6367310881614685,
49
+ "learning_rate": 5e-06,
50
+ "loss": 0.9644,
51
+ "step": 60
52
+ },
53
+ {
54
+ "epoch": 0.16548463356973994,
55
+ "grad_norm": 0.5621835589408875,
56
+ "learning_rate": 5e-06,
57
+ "loss": 0.9609,
58
+ "step": 70
59
+ },
60
+ {
61
+ "epoch": 0.18912529550827423,
62
+ "grad_norm": 0.6173996329307556,
63
+ "learning_rate": 5e-06,
64
+ "loss": 0.9607,
65
+ "step": 80
66
+ },
67
+ {
68
+ "epoch": 0.2127659574468085,
69
+ "grad_norm": 0.5042706727981567,
70
+ "learning_rate": 5e-06,
71
+ "loss": 0.9524,
72
+ "step": 90
73
+ },
74
+ {
75
+ "epoch": 0.2364066193853428,
76
+ "grad_norm": 0.7688830494880676,
77
+ "learning_rate": 5e-06,
78
+ "loss": 0.9438,
79
+ "step": 100
80
+ },
81
+ {
82
+ "epoch": 0.26004728132387706,
83
+ "grad_norm": 0.5911379456520081,
84
+ "learning_rate": 5e-06,
85
+ "loss": 0.9401,
86
+ "step": 110
87
+ },
88
+ {
89
+ "epoch": 0.28368794326241137,
90
+ "grad_norm": 0.5532689690589905,
91
+ "learning_rate": 5e-06,
92
+ "loss": 0.9468,
93
+ "step": 120
94
+ },
95
+ {
96
+ "epoch": 0.3073286052009456,
97
+ "grad_norm": 0.8925480246543884,
98
+ "learning_rate": 5e-06,
99
+ "loss": 0.9462,
100
+ "step": 130
101
+ },
102
+ {
103
+ "epoch": 0.3309692671394799,
104
+ "grad_norm": 0.5422332882881165,
105
+ "learning_rate": 5e-06,
106
+ "loss": 0.94,
107
+ "step": 140
108
+ },
109
+ {
110
+ "epoch": 0.3546099290780142,
111
+ "grad_norm": 0.5486394166946411,
112
+ "learning_rate": 5e-06,
113
+ "loss": 0.9344,
114
+ "step": 150
115
+ },
116
+ {
117
+ "epoch": 0.37825059101654845,
118
+ "grad_norm": 0.5163227915763855,
119
+ "learning_rate": 5e-06,
120
+ "loss": 0.9281,
121
+ "step": 160
122
+ },
123
+ {
124
+ "epoch": 0.40189125295508277,
125
+ "grad_norm": 0.53909832239151,
126
+ "learning_rate": 5e-06,
127
+ "loss": 0.9377,
128
+ "step": 170
129
+ },
130
+ {
131
+ "epoch": 0.425531914893617,
132
+ "grad_norm": 0.500614583492279,
133
+ "learning_rate": 5e-06,
134
+ "loss": 0.9309,
135
+ "step": 180
136
+ },
137
+ {
138
+ "epoch": 0.4491725768321513,
139
+ "grad_norm": 0.5055537819862366,
140
+ "learning_rate": 5e-06,
141
+ "loss": 0.9326,
142
+ "step": 190
143
+ },
144
+ {
145
+ "epoch": 0.4728132387706856,
146
+ "grad_norm": 0.5739602446556091,
147
+ "learning_rate": 5e-06,
148
+ "loss": 0.9334,
149
+ "step": 200
150
+ },
151
+ {
152
+ "epoch": 0.49645390070921985,
153
+ "grad_norm": 0.4969063103199005,
154
+ "learning_rate": 5e-06,
155
+ "loss": 0.9306,
156
+ "step": 210
157
+ },
158
+ {
159
+ "epoch": 0.5200945626477541,
160
+ "grad_norm": 0.4895598888397217,
161
+ "learning_rate": 5e-06,
162
+ "loss": 0.9326,
163
+ "step": 220
164
+ },
165
+ {
166
+ "epoch": 0.5437352245862884,
167
+ "grad_norm": 0.488599956035614,
168
+ "learning_rate": 5e-06,
169
+ "loss": 0.9251,
170
+ "step": 230
171
+ },
172
+ {
173
+ "epoch": 0.5673758865248227,
174
+ "grad_norm": 0.5502417087554932,
175
+ "learning_rate": 5e-06,
176
+ "loss": 0.9233,
177
+ "step": 240
178
+ },
179
+ {
180
+ "epoch": 0.5910165484633569,
181
+ "grad_norm": 0.5484869480133057,
182
+ "learning_rate": 5e-06,
183
+ "loss": 0.9177,
184
+ "step": 250
185
+ },
186
+ {
187
+ "epoch": 0.6146572104018913,
188
+ "grad_norm": 0.70972740650177,
189
+ "learning_rate": 5e-06,
190
+ "loss": 0.9204,
191
+ "step": 260
192
+ },
193
+ {
194
+ "epoch": 0.6382978723404256,
195
+ "grad_norm": 0.5443243980407715,
196
+ "learning_rate": 5e-06,
197
+ "loss": 0.9203,
198
+ "step": 270
199
+ },
200
+ {
201
+ "epoch": 0.6619385342789598,
202
+ "grad_norm": 0.4656381607055664,
203
+ "learning_rate": 5e-06,
204
+ "loss": 0.9251,
205
+ "step": 280
206
+ },
207
+ {
208
+ "epoch": 0.6855791962174941,
209
+ "grad_norm": 0.49396905303001404,
210
+ "learning_rate": 5e-06,
211
+ "loss": 0.9242,
212
+ "step": 290
213
+ },
214
+ {
215
+ "epoch": 0.7092198581560284,
216
+ "grad_norm": 0.43112456798553467,
217
+ "learning_rate": 5e-06,
218
+ "loss": 0.9238,
219
+ "step": 300
220
+ },
221
+ {
222
+ "epoch": 0.7328605200945626,
223
+ "grad_norm": 0.4588732123374939,
224
+ "learning_rate": 5e-06,
225
+ "loss": 0.9206,
226
+ "step": 310
227
+ },
228
+ {
229
+ "epoch": 0.7565011820330969,
230
+ "grad_norm": 0.5064387321472168,
231
+ "learning_rate": 5e-06,
232
+ "loss": 0.9177,
233
+ "step": 320
234
+ },
235
+ {
236
+ "epoch": 0.7801418439716312,
237
+ "grad_norm": 0.51941978931427,
238
+ "learning_rate": 5e-06,
239
+ "loss": 0.9161,
240
+ "step": 330
241
+ },
242
+ {
243
+ "epoch": 0.8037825059101655,
244
+ "grad_norm": 0.45160844922065735,
245
+ "learning_rate": 5e-06,
246
+ "loss": 0.9197,
247
+ "step": 340
248
+ },
249
+ {
250
+ "epoch": 0.8274231678486997,
251
+ "grad_norm": 0.4652918875217438,
252
+ "learning_rate": 5e-06,
253
+ "loss": 0.9182,
254
+ "step": 350
255
+ },
256
+ {
257
+ "epoch": 0.851063829787234,
258
+ "grad_norm": 0.49702754616737366,
259
+ "learning_rate": 5e-06,
260
+ "loss": 0.9115,
261
+ "step": 360
262
+ },
263
+ {
264
+ "epoch": 0.8747044917257684,
265
+ "grad_norm": 0.6036341786384583,
266
+ "learning_rate": 5e-06,
267
+ "loss": 0.9195,
268
+ "step": 370
269
+ },
270
+ {
271
+ "epoch": 0.8983451536643026,
272
+ "grad_norm": 0.4541695713996887,
273
+ "learning_rate": 5e-06,
274
+ "loss": 0.9144,
275
+ "step": 380
276
+ },
277
+ {
278
+ "epoch": 0.9219858156028369,
279
+ "grad_norm": 0.5585927963256836,
280
+ "learning_rate": 5e-06,
281
+ "loss": 0.9139,
282
+ "step": 390
283
+ },
284
+ {
285
+ "epoch": 0.9456264775413712,
286
+ "grad_norm": 0.4704740047454834,
287
+ "learning_rate": 5e-06,
288
+ "loss": 0.9107,
289
+ "step": 400
290
+ },
291
+ {
292
+ "epoch": 0.9692671394799054,
293
+ "grad_norm": 0.43893522024154663,
294
+ "learning_rate": 5e-06,
295
+ "loss": 0.9093,
296
+ "step": 410
297
+ },
298
+ {
299
+ "epoch": 0.9929078014184397,
300
+ "grad_norm": 0.4892677366733551,
301
+ "learning_rate": 5e-06,
302
+ "loss": 0.9018,
303
+ "step": 420
304
+ },
305
+ {
306
+ "epoch": 1.0,
307
+ "eval_loss": 0.9103471636772156,
308
+ "eval_runtime": 38.5799,
309
+ "eval_samples_per_second": 295.284,
310
+ "eval_steps_per_second": 1.166,
311
+ "step": 423
312
+ },
313
+ {
314
+ "epoch": 1.016548463356974,
315
+ "grad_norm": 0.5619480609893799,
316
+ "learning_rate": 5e-06,
317
+ "loss": 0.8787,
318
+ "step": 430
319
+ },
320
+ {
321
+ "epoch": 1.0401891252955082,
322
+ "grad_norm": 0.5610897541046143,
323
+ "learning_rate": 5e-06,
324
+ "loss": 0.8696,
325
+ "step": 440
326
+ },
327
+ {
328
+ "epoch": 1.0638297872340425,
329
+ "grad_norm": 0.4962644577026367,
330
+ "learning_rate": 5e-06,
331
+ "loss": 0.8629,
332
+ "step": 450
333
+ },
334
+ {
335
+ "epoch": 1.0874704491725768,
336
+ "grad_norm": 0.8102297782897949,
337
+ "learning_rate": 5e-06,
338
+ "loss": 0.8631,
339
+ "step": 460
340
+ },
341
+ {
342
+ "epoch": 1.1111111111111112,
343
+ "grad_norm": 0.4762827754020691,
344
+ "learning_rate": 5e-06,
345
+ "loss": 0.865,
346
+ "step": 470
347
+ },
348
+ {
349
+ "epoch": 1.1347517730496455,
350
+ "grad_norm": 0.5226008296012878,
351
+ "learning_rate": 5e-06,
352
+ "loss": 0.8588,
353
+ "step": 480
354
+ },
355
+ {
356
+ "epoch": 1.1583924349881798,
357
+ "grad_norm": 0.4529044032096863,
358
+ "learning_rate": 5e-06,
359
+ "loss": 0.8605,
360
+ "step": 490
361
+ },
362
+ {
363
+ "epoch": 1.1820330969267139,
364
+ "grad_norm": 0.48834407329559326,
365
+ "learning_rate": 5e-06,
366
+ "loss": 0.865,
367
+ "step": 500
368
+ },
369
+ {
370
+ "epoch": 1.2056737588652482,
371
+ "grad_norm": 0.5005617737770081,
372
+ "learning_rate": 5e-06,
373
+ "loss": 0.866,
374
+ "step": 510
375
+ },
376
+ {
377
+ "epoch": 1.2293144208037825,
378
+ "grad_norm": 0.4393448829650879,
379
+ "learning_rate": 5e-06,
380
+ "loss": 0.8652,
381
+ "step": 520
382
+ },
383
+ {
384
+ "epoch": 1.2529550827423168,
385
+ "grad_norm": 0.5248346328735352,
386
+ "learning_rate": 5e-06,
387
+ "loss": 0.8653,
388
+ "step": 530
389
+ },
390
+ {
391
+ "epoch": 1.2765957446808511,
392
+ "grad_norm": 0.5386662483215332,
393
+ "learning_rate": 5e-06,
394
+ "loss": 0.859,
395
+ "step": 540
396
+ },
397
+ {
398
+ "epoch": 1.3002364066193852,
399
+ "grad_norm": 0.48023349046707153,
400
+ "learning_rate": 5e-06,
401
+ "loss": 0.8639,
402
+ "step": 550
403
+ },
404
+ {
405
+ "epoch": 1.3238770685579198,
406
+ "grad_norm": 0.46678417921066284,
407
+ "learning_rate": 5e-06,
408
+ "loss": 0.8678,
409
+ "step": 560
410
+ },
411
+ {
412
+ "epoch": 1.3475177304964538,
413
+ "grad_norm": 0.5156953930854797,
414
+ "learning_rate": 5e-06,
415
+ "loss": 0.863,
416
+ "step": 570
417
+ },
418
+ {
419
+ "epoch": 1.3711583924349882,
420
+ "grad_norm": 0.4521232545375824,
421
+ "learning_rate": 5e-06,
422
+ "loss": 0.8661,
423
+ "step": 580
424
+ },
425
+ {
426
+ "epoch": 1.3947990543735225,
427
+ "grad_norm": 0.4961282014846802,
428
+ "learning_rate": 5e-06,
429
+ "loss": 0.8591,
430
+ "step": 590
431
+ },
432
+ {
433
+ "epoch": 1.4184397163120568,
434
+ "grad_norm": 0.4934581220149994,
435
+ "learning_rate": 5e-06,
436
+ "loss": 0.864,
437
+ "step": 600
438
+ },
439
+ {
440
+ "epoch": 1.442080378250591,
441
+ "grad_norm": 0.5448637008666992,
442
+ "learning_rate": 5e-06,
443
+ "loss": 0.8635,
444
+ "step": 610
445
+ },
446
+ {
447
+ "epoch": 1.4657210401891252,
448
+ "grad_norm": 0.4642185568809509,
449
+ "learning_rate": 5e-06,
450
+ "loss": 0.8631,
451
+ "step": 620
452
+ },
453
+ {
454
+ "epoch": 1.4893617021276595,
455
+ "grad_norm": 0.4816928803920746,
456
+ "learning_rate": 5e-06,
457
+ "loss": 0.8618,
458
+ "step": 630
459
+ },
460
+ {
461
+ "epoch": 1.5130023640661938,
462
+ "grad_norm": 0.4565974175930023,
463
+ "learning_rate": 5e-06,
464
+ "loss": 0.864,
465
+ "step": 640
466
+ },
467
+ {
468
+ "epoch": 1.5366430260047281,
469
+ "grad_norm": 0.580853283405304,
470
+ "learning_rate": 5e-06,
471
+ "loss": 0.8613,
472
+ "step": 650
473
+ },
474
+ {
475
+ "epoch": 1.5602836879432624,
476
+ "grad_norm": 0.5403729677200317,
477
+ "learning_rate": 5e-06,
478
+ "loss": 0.8636,
479
+ "step": 660
480
+ },
481
+ {
482
+ "epoch": 1.5839243498817965,
483
+ "grad_norm": 0.4570687711238861,
484
+ "learning_rate": 5e-06,
485
+ "loss": 0.8636,
486
+ "step": 670
487
+ },
488
+ {
489
+ "epoch": 1.607565011820331,
490
+ "grad_norm": 0.557025134563446,
491
+ "learning_rate": 5e-06,
492
+ "loss": 0.8634,
493
+ "step": 680
494
+ },
495
+ {
496
+ "epoch": 1.6312056737588652,
497
+ "grad_norm": 0.4970671534538269,
498
+ "learning_rate": 5e-06,
499
+ "loss": 0.8582,
500
+ "step": 690
501
+ },
502
+ {
503
+ "epoch": 1.6548463356973995,
504
+ "grad_norm": 0.5011308193206787,
505
+ "learning_rate": 5e-06,
506
+ "loss": 0.8615,
507
+ "step": 700
508
+ },
509
+ {
510
+ "epoch": 1.6784869976359338,
511
+ "grad_norm": 0.5732368230819702,
512
+ "learning_rate": 5e-06,
513
+ "loss": 0.8612,
514
+ "step": 710
515
+ },
516
+ {
517
+ "epoch": 1.702127659574468,
518
+ "grad_norm": 0.44940251111984253,
519
+ "learning_rate": 5e-06,
520
+ "loss": 0.8654,
521
+ "step": 720
522
+ },
523
+ {
524
+ "epoch": 1.7257683215130024,
525
+ "grad_norm": 0.4695453643798828,
526
+ "learning_rate": 5e-06,
527
+ "loss": 0.8655,
528
+ "step": 730
529
+ },
530
+ {
531
+ "epoch": 1.7494089834515365,
532
+ "grad_norm": 0.6582662463188171,
533
+ "learning_rate": 5e-06,
534
+ "loss": 0.863,
535
+ "step": 740
536
+ },
537
+ {
538
+ "epoch": 1.773049645390071,
539
+ "grad_norm": 0.5361992716789246,
540
+ "learning_rate": 5e-06,
541
+ "loss": 0.857,
542
+ "step": 750
543
+ },
544
+ {
545
+ "epoch": 1.7966903073286051,
546
+ "grad_norm": 0.4670429527759552,
547
+ "learning_rate": 5e-06,
548
+ "loss": 0.8614,
549
+ "step": 760
550
+ },
551
+ {
552
+ "epoch": 1.8203309692671394,
553
+ "grad_norm": 0.44174039363861084,
554
+ "learning_rate": 5e-06,
555
+ "loss": 0.8606,
556
+ "step": 770
557
+ },
558
+ {
559
+ "epoch": 1.8439716312056738,
560
+ "grad_norm": 0.4454022943973541,
561
+ "learning_rate": 5e-06,
562
+ "loss": 0.8603,
563
+ "step": 780
564
+ },
565
+ {
566
+ "epoch": 1.867612293144208,
567
+ "grad_norm": 0.46973222494125366,
568
+ "learning_rate": 5e-06,
569
+ "loss": 0.8624,
570
+ "step": 790
571
+ },
572
+ {
573
+ "epoch": 1.8912529550827424,
574
+ "grad_norm": 0.4946763515472412,
575
+ "learning_rate": 5e-06,
576
+ "loss": 0.8648,
577
+ "step": 800
578
+ },
579
+ {
580
+ "epoch": 1.9148936170212765,
581
+ "grad_norm": 0.4165429472923279,
582
+ "learning_rate": 5e-06,
583
+ "loss": 0.8584,
584
+ "step": 810
585
+ },
586
+ {
587
+ "epoch": 1.938534278959811,
588
+ "grad_norm": 0.44204238057136536,
589
+ "learning_rate": 5e-06,
590
+ "loss": 0.8615,
591
+ "step": 820
592
+ },
593
+ {
594
+ "epoch": 1.962174940898345,
595
+ "grad_norm": 0.4277856647968292,
596
+ "learning_rate": 5e-06,
597
+ "loss": 0.8603,
598
+ "step": 830
599
+ },
600
+ {
601
+ "epoch": 1.9858156028368794,
602
+ "grad_norm": 0.4897569715976715,
603
+ "learning_rate": 5e-06,
604
+ "loss": 0.8586,
605
+ "step": 840
606
+ },
607
+ {
608
+ "epoch": 2.0,
609
+ "eval_loss": 0.9001395106315613,
610
+ "eval_runtime": 38.3885,
611
+ "eval_samples_per_second": 296.756,
612
+ "eval_steps_per_second": 1.172,
613
+ "step": 846
614
+ },
615
+ {
616
+ "epoch": 2.0094562647754137,
617
+ "grad_norm": 0.673308789730072,
618
+ "learning_rate": 5e-06,
619
+ "loss": 0.8393,
620
+ "step": 850
621
+ },
622
+ {
623
+ "epoch": 2.033096926713948,
624
+ "grad_norm": 0.5594461560249329,
625
+ "learning_rate": 5e-06,
626
+ "loss": 0.8071,
627
+ "step": 860
628
+ },
629
+ {
630
+ "epoch": 2.0567375886524824,
631
+ "grad_norm": 0.4584490656852722,
632
+ "learning_rate": 5e-06,
633
+ "loss": 0.8083,
634
+ "step": 870
635
+ },
636
+ {
637
+ "epoch": 2.0803782505910164,
638
+ "grad_norm": 0.5544384121894836,
639
+ "learning_rate": 5e-06,
640
+ "loss": 0.8135,
641
+ "step": 880
642
+ },
643
+ {
644
+ "epoch": 2.104018912529551,
645
+ "grad_norm": 0.5173571705818176,
646
+ "learning_rate": 5e-06,
647
+ "loss": 0.8104,
648
+ "step": 890
649
+ },
650
+ {
651
+ "epoch": 2.127659574468085,
652
+ "grad_norm": 0.488302618265152,
653
+ "learning_rate": 5e-06,
654
+ "loss": 0.8155,
655
+ "step": 900
656
+ },
657
+ {
658
+ "epoch": 2.1513002364066196,
659
+ "grad_norm": 0.5209611058235168,
660
+ "learning_rate": 5e-06,
661
+ "loss": 0.8108,
662
+ "step": 910
663
+ },
664
+ {
665
+ "epoch": 2.1749408983451537,
666
+ "grad_norm": 0.51943439245224,
667
+ "learning_rate": 5e-06,
668
+ "loss": 0.808,
669
+ "step": 920
670
+ },
671
+ {
672
+ "epoch": 2.198581560283688,
673
+ "grad_norm": 0.5136656761169434,
674
+ "learning_rate": 5e-06,
675
+ "loss": 0.8152,
676
+ "step": 930
677
+ },
678
+ {
679
+ "epoch": 2.2222222222222223,
680
+ "grad_norm": 0.48146888613700867,
681
+ "learning_rate": 5e-06,
682
+ "loss": 0.8114,
683
+ "step": 940
684
+ },
685
+ {
686
+ "epoch": 2.2458628841607564,
687
+ "grad_norm": 0.5011332035064697,
688
+ "learning_rate": 5e-06,
689
+ "loss": 0.8133,
690
+ "step": 950
691
+ },
692
+ {
693
+ "epoch": 2.269503546099291,
694
+ "grad_norm": 0.5616235733032227,
695
+ "learning_rate": 5e-06,
696
+ "loss": 0.815,
697
+ "step": 960
698
+ },
699
+ {
700
+ "epoch": 2.293144208037825,
701
+ "grad_norm": 0.4666649401187897,
702
+ "learning_rate": 5e-06,
703
+ "loss": 0.8096,
704
+ "step": 970
705
+ },
706
+ {
707
+ "epoch": 2.3167848699763596,
708
+ "grad_norm": 0.5732521414756775,
709
+ "learning_rate": 5e-06,
710
+ "loss": 0.8098,
711
+ "step": 980
712
+ },
713
+ {
714
+ "epoch": 2.3404255319148937,
715
+ "grad_norm": 0.4544953405857086,
716
+ "learning_rate": 5e-06,
717
+ "loss": 0.8124,
718
+ "step": 990
719
+ },
720
+ {
721
+ "epoch": 2.3640661938534278,
722
+ "grad_norm": 0.44587448239326477,
723
+ "learning_rate": 5e-06,
724
+ "loss": 0.8095,
725
+ "step": 1000
726
+ },
727
+ {
728
+ "epoch": 2.3877068557919623,
729
+ "grad_norm": 0.484457403421402,
730
+ "learning_rate": 5e-06,
731
+ "loss": 0.8118,
732
+ "step": 1010
733
+ },
734
+ {
735
+ "epoch": 2.4113475177304964,
736
+ "grad_norm": 0.4641134440898895,
737
+ "learning_rate": 5e-06,
738
+ "loss": 0.8124,
739
+ "step": 1020
740
+ },
741
+ {
742
+ "epoch": 2.434988179669031,
743
+ "grad_norm": 0.49560511112213135,
744
+ "learning_rate": 5e-06,
745
+ "loss": 0.8129,
746
+ "step": 1030
747
+ },
748
+ {
749
+ "epoch": 2.458628841607565,
750
+ "grad_norm": 0.4686046540737152,
751
+ "learning_rate": 5e-06,
752
+ "loss": 0.814,
753
+ "step": 1040
754
+ },
755
+ {
756
+ "epoch": 2.482269503546099,
757
+ "grad_norm": 0.5736932158470154,
758
+ "learning_rate": 5e-06,
759
+ "loss": 0.809,
760
+ "step": 1050
761
+ },
762
+ {
763
+ "epoch": 2.5059101654846336,
764
+ "grad_norm": 0.5565879940986633,
765
+ "learning_rate": 5e-06,
766
+ "loss": 0.8192,
767
+ "step": 1060
768
+ },
769
+ {
770
+ "epoch": 2.5295508274231677,
771
+ "grad_norm": 0.5535584092140198,
772
+ "learning_rate": 5e-06,
773
+ "loss": 0.8177,
774
+ "step": 1070
775
+ },
776
+ {
777
+ "epoch": 2.5531914893617023,
778
+ "grad_norm": 0.49917906522750854,
779
+ "learning_rate": 5e-06,
780
+ "loss": 0.817,
781
+ "step": 1080
782
+ },
783
+ {
784
+ "epoch": 2.5768321513002364,
785
+ "grad_norm": 0.48814043402671814,
786
+ "learning_rate": 5e-06,
787
+ "loss": 0.8118,
788
+ "step": 1090
789
+ },
790
+ {
791
+ "epoch": 2.6004728132387704,
792
+ "grad_norm": 0.4942103326320648,
793
+ "learning_rate": 5e-06,
794
+ "loss": 0.8218,
795
+ "step": 1100
796
+ },
797
+ {
798
+ "epoch": 2.624113475177305,
799
+ "grad_norm": 0.5041037797927856,
800
+ "learning_rate": 5e-06,
801
+ "loss": 0.8142,
802
+ "step": 1110
803
+ },
804
+ {
805
+ "epoch": 2.6477541371158395,
806
+ "grad_norm": 0.4969462454319,
807
+ "learning_rate": 5e-06,
808
+ "loss": 0.8194,
809
+ "step": 1120
810
+ },
811
+ {
812
+ "epoch": 2.6713947990543736,
813
+ "grad_norm": 0.5635184049606323,
814
+ "learning_rate": 5e-06,
815
+ "loss": 0.8206,
816
+ "step": 1130
817
+ },
818
+ {
819
+ "epoch": 2.6950354609929077,
820
+ "grad_norm": 0.4972004294395447,
821
+ "learning_rate": 5e-06,
822
+ "loss": 0.8192,
823
+ "step": 1140
824
+ },
825
+ {
826
+ "epoch": 2.7186761229314422,
827
+ "grad_norm": 0.49688395857810974,
828
+ "learning_rate": 5e-06,
829
+ "loss": 0.8165,
830
+ "step": 1150
831
+ },
832
+ {
833
+ "epoch": 2.7423167848699763,
834
+ "grad_norm": 0.4726709723472595,
835
+ "learning_rate": 5e-06,
836
+ "loss": 0.8205,
837
+ "step": 1160
838
+ },
839
+ {
840
+ "epoch": 2.7659574468085104,
841
+ "grad_norm": 0.532228946685791,
842
+ "learning_rate": 5e-06,
843
+ "loss": 0.8139,
844
+ "step": 1170
845
+ },
846
+ {
847
+ "epoch": 2.789598108747045,
848
+ "grad_norm": 0.4629664421081543,
849
+ "learning_rate": 5e-06,
850
+ "loss": 0.8119,
851
+ "step": 1180
852
+ },
853
+ {
854
+ "epoch": 2.813238770685579,
855
+ "grad_norm": 0.4985387921333313,
856
+ "learning_rate": 5e-06,
857
+ "loss": 0.8206,
858
+ "step": 1190
859
+ },
860
+ {
861
+ "epoch": 2.8368794326241136,
862
+ "grad_norm": 0.48208263516426086,
863
+ "learning_rate": 5e-06,
864
+ "loss": 0.8144,
865
+ "step": 1200
866
+ },
867
+ {
868
+ "epoch": 2.8605200945626477,
869
+ "grad_norm": 0.5160050988197327,
870
+ "learning_rate": 5e-06,
871
+ "loss": 0.8179,
872
+ "step": 1210
873
+ },
874
+ {
875
+ "epoch": 2.884160756501182,
876
+ "grad_norm": 0.47810521721839905,
877
+ "learning_rate": 5e-06,
878
+ "loss": 0.8157,
879
+ "step": 1220
880
+ },
881
+ {
882
+ "epoch": 2.9078014184397163,
883
+ "grad_norm": 0.4839594066143036,
884
+ "learning_rate": 5e-06,
885
+ "loss": 0.8162,
886
+ "step": 1230
887
+ },
888
+ {
889
+ "epoch": 2.9314420803782504,
890
+ "grad_norm": 0.49858078360557556,
891
+ "learning_rate": 5e-06,
892
+ "loss": 0.8133,
893
+ "step": 1240
894
+ },
895
+ {
896
+ "epoch": 2.955082742316785,
897
+ "grad_norm": 0.5557030439376831,
898
+ "learning_rate": 5e-06,
899
+ "loss": 0.8167,
900
+ "step": 1250
901
+ },
902
+ {
903
+ "epoch": 2.978723404255319,
904
+ "grad_norm": 0.5793541073799133,
905
+ "learning_rate": 5e-06,
906
+ "loss": 0.8143,
907
+ "step": 1260
908
+ },
909
+ {
910
+ "epoch": 3.0,
911
+ "eval_loss": 0.9010276198387146,
912
+ "eval_runtime": 39.1845,
913
+ "eval_samples_per_second": 290.728,
914
+ "eval_steps_per_second": 1.148,
915
+ "step": 1269
916
+ },
917
+ {
918
+ "epoch": 3.0,
919
+ "step": 1269,
920
+ "total_flos": 5.9918252490593665e+19,
921
+ "train_loss": 0.8716646811452975,
922
+ "train_runtime": 8996.6792,
923
+ "train_samples_per_second": 72.17,
924
+ "train_steps_per_second": 0.141
925
+ }
926
+ ],
927
+ "logging_steps": 10,
928
+ "max_steps": 1269,
929
+ "num_input_tokens_seen": 0,
930
+ "num_train_epochs": 3,
931
+ "save_steps": 500,
932
+ "stateful_callbacks": {
933
+ "TrainerControl": {
934
+ "args": {
935
+ "should_epoch_stop": false,
936
+ "should_evaluate": false,
937
+ "should_log": false,
938
+ "should_save": true,
939
+ "should_training_stop": true
940
+ },
941
+ "attributes": {}
942
+ }
943
+ },
944
+ "total_flos": 5.9918252490593665e+19,
945
+ "train_batch_size": 16,
946
+ "trial_name": null,
947
+ "trial_params": null
948
+ }
training_eval_loss.png ADDED
training_loss.png ADDED