gulaschnascher4000 commited on
Commit
dc4635f
·
verified ·
1 Parent(s): d9dfc9b

End of training

Browse files
README.md CHANGED
@@ -4,6 +4,7 @@ license: llama3.2
4
  base_model: meta-llama/Llama-3.2-3B
5
  tags:
6
  - llama-factory
 
7
  - generated_from_trainer
8
  model-index:
9
  - name: lora_0-1_3B
@@ -15,9 +16,9 @@ should probably proofread and complete it, then remove this comment. -->
15
 
16
  # lora_0-1_3B
17
 
18
- This model is a fine-tuned version of [meta-llama/Llama-3.2-3B](https://huggingface.co/meta-llama/Llama-3.2-3B) on the None dataset.
19
  It achieves the following results on the evaluation set:
20
- - Loss: 1.6384
21
 
22
  ## Model description
23
 
 
4
  base_model: meta-llama/Llama-3.2-3B
5
  tags:
6
  - llama-factory
7
+ - lora
8
  - generated_from_trainer
9
  model-index:
10
  - name: lora_0-1_3B
 
16
 
17
  # lora_0-1_3B
18
 
19
+ This model is a fine-tuned version of [meta-llama/Llama-3.2-3B](https://huggingface.co/meta-llama/Llama-3.2-3B) on the gulaschnascher4000/stream-dataset-0-2 and the identity-chatgulaschpt datasets.
20
  It achieves the following results on the evaluation set:
21
+ - Loss: 1.6369
22
 
23
  ## Model description
24
 
all_results.json ADDED
@@ -0,0 +1,12 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 1.0,
3
+ "eval_loss": 1.6368554830551147,
4
+ "eval_runtime": 51.0365,
5
+ "eval_samples_per_second": 77.298,
6
+ "eval_steps_per_second": 19.339,
7
+ "total_flos": 2.3250303254423142e+17,
8
+ "train_loss": 1.7593931747986389,
9
+ "train_runtime": 934.3961,
10
+ "train_samples_per_second": 37.998,
11
+ "train_steps_per_second": 1.188
12
+ }
eval_results.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 1.0,
3
+ "eval_loss": 1.6368554830551147,
4
+ "eval_runtime": 51.0365,
5
+ "eval_samples_per_second": 77.298,
6
+ "eval_steps_per_second": 19.339
7
+ }
runs/Jan14_00-59-17_pss5xry2f0aq/events.out.tfevents.1736817359.pss5xry2f0aq.30019.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:62c7850f34057d31f5d4a7bdbc7901fdd42143f8f24e10ce5210237a9ca152e2
3
+ size 359
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 1.0,
3
+ "total_flos": 2.3250303254423142e+17,
4
+ "train_loss": 1.7593931747986389,
5
+ "train_runtime": 934.3961,
6
+ "train_samples_per_second": 37.998,
7
+ "train_steps_per_second": 1.188
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,835 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 1.0,
5
+ "eval_steps": 500,
6
+ "global_step": 1110,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.009009009009009009,
13
+ "grad_norm": 1.3664734363555908,
14
+ "learning_rate": 9.00900900900901e-06,
15
+ "loss": 3.0355,
16
+ "step": 10
17
+ },
18
+ {
19
+ "epoch": 0.018018018018018018,
20
+ "grad_norm": 1.426148533821106,
21
+ "learning_rate": 1.801801801801802e-05,
22
+ "loss": 3.0333,
23
+ "step": 20
24
+ },
25
+ {
26
+ "epoch": 0.02702702702702703,
27
+ "grad_norm": 1.4102978706359863,
28
+ "learning_rate": 2.702702702702703e-05,
29
+ "loss": 2.8445,
30
+ "step": 30
31
+ },
32
+ {
33
+ "epoch": 0.036036036036036036,
34
+ "grad_norm": 1.448880910873413,
35
+ "learning_rate": 3.603603603603604e-05,
36
+ "loss": 2.5113,
37
+ "step": 40
38
+ },
39
+ {
40
+ "epoch": 0.04504504504504504,
41
+ "grad_norm": 1.8392173051834106,
42
+ "learning_rate": 4.5045045045045046e-05,
43
+ "loss": 2.2516,
44
+ "step": 50
45
+ },
46
+ {
47
+ "epoch": 0.05405405405405406,
48
+ "grad_norm": 1.6239687204360962,
49
+ "learning_rate": 5.405405405405406e-05,
50
+ "loss": 2.1521,
51
+ "step": 60
52
+ },
53
+ {
54
+ "epoch": 0.06306306306306306,
55
+ "grad_norm": 1.783531904220581,
56
+ "learning_rate": 6.306306306306306e-05,
57
+ "loss": 2.0766,
58
+ "step": 70
59
+ },
60
+ {
61
+ "epoch": 0.07207207207207207,
62
+ "grad_norm": 1.878064513206482,
63
+ "learning_rate": 7.207207207207208e-05,
64
+ "loss": 2.0584,
65
+ "step": 80
66
+ },
67
+ {
68
+ "epoch": 0.08108108108108109,
69
+ "grad_norm": 1.793707013130188,
70
+ "learning_rate": 8.108108108108109e-05,
71
+ "loss": 2.0741,
72
+ "step": 90
73
+ },
74
+ {
75
+ "epoch": 0.09009009009009009,
76
+ "grad_norm": 1.812543511390686,
77
+ "learning_rate": 9.009009009009009e-05,
78
+ "loss": 1.9833,
79
+ "step": 100
80
+ },
81
+ {
82
+ "epoch": 0.0990990990990991,
83
+ "grad_norm": 2.002195119857788,
84
+ "learning_rate": 9.90990990990991e-05,
85
+ "loss": 1.9653,
86
+ "step": 110
87
+ },
88
+ {
89
+ "epoch": 0.10810810810810811,
90
+ "grad_norm": 2.367985963821411,
91
+ "learning_rate": 9.99799753559161e-05,
92
+ "loss": 2.0073,
93
+ "step": 120
94
+ },
95
+ {
96
+ "epoch": 0.11711711711711711,
97
+ "grad_norm": 1.732643485069275,
98
+ "learning_rate": 9.991077495600804e-05,
99
+ "loss": 1.9519,
100
+ "step": 130
101
+ },
102
+ {
103
+ "epoch": 0.12612612612612611,
104
+ "grad_norm": 1.9506319761276245,
105
+ "learning_rate": 9.979221999536699e-05,
106
+ "loss": 1.9233,
107
+ "step": 140
108
+ },
109
+ {
110
+ "epoch": 0.13513513513513514,
111
+ "grad_norm": 1.7935253381729126,
112
+ "learning_rate": 9.962442770775675e-05,
113
+ "loss": 1.9716,
114
+ "step": 150
115
+ },
116
+ {
117
+ "epoch": 0.14414414414414414,
118
+ "grad_norm": 2.0075876712799072,
119
+ "learning_rate": 9.940756401555898e-05,
120
+ "loss": 1.8546,
121
+ "step": 160
122
+ },
123
+ {
124
+ "epoch": 0.15315315315315314,
125
+ "grad_norm": 1.8393422365188599,
126
+ "learning_rate": 9.914184336569974e-05,
127
+ "loss": 1.9295,
128
+ "step": 170
129
+ },
130
+ {
131
+ "epoch": 0.16216216216216217,
132
+ "grad_norm": 2.006378650665283,
133
+ "learning_rate": 9.882752851759248e-05,
134
+ "loss": 1.872,
135
+ "step": 180
136
+ },
137
+ {
138
+ "epoch": 0.17117117117117117,
139
+ "grad_norm": 1.8561218976974487,
140
+ "learning_rate": 9.846493028330678e-05,
141
+ "loss": 1.8979,
142
+ "step": 190
143
+ },
144
+ {
145
+ "epoch": 0.18018018018018017,
146
+ "grad_norm": 1.9337742328643799,
147
+ "learning_rate": 9.805440722022014e-05,
148
+ "loss": 1.8461,
149
+ "step": 200
150
+ },
151
+ {
152
+ "epoch": 0.1891891891891892,
153
+ "grad_norm": 1.990472435951233,
154
+ "learning_rate": 9.759636527645633e-05,
155
+ "loss": 1.885,
156
+ "step": 210
157
+ },
158
+ {
159
+ "epoch": 0.1981981981981982,
160
+ "grad_norm": 2.0939085483551025,
161
+ "learning_rate": 9.709125738946125e-05,
162
+ "loss": 1.8196,
163
+ "step": 220
164
+ },
165
+ {
166
+ "epoch": 0.2072072072072072,
167
+ "grad_norm": 2.165142297744751,
168
+ "learning_rate": 9.65395830381131e-05,
169
+ "loss": 1.8212,
170
+ "step": 230
171
+ },
172
+ {
173
+ "epoch": 0.21621621621621623,
174
+ "grad_norm": 1.9272476434707642,
175
+ "learning_rate": 9.594188774880982e-05,
176
+ "loss": 1.7787,
177
+ "step": 240
178
+ },
179
+ {
180
+ "epoch": 0.22522522522522523,
181
+ "grad_norm": 1.9800828695297241,
182
+ "learning_rate": 9.5298762556022e-05,
183
+ "loss": 1.8144,
184
+ "step": 250
185
+ },
186
+ {
187
+ "epoch": 0.23423423423423423,
188
+ "grad_norm": 2.138906240463257,
189
+ "learning_rate": 9.46108434178452e-05,
190
+ "loss": 1.8595,
191
+ "step": 260
192
+ },
193
+ {
194
+ "epoch": 0.24324324324324326,
195
+ "grad_norm": 2.0844318866729736,
196
+ "learning_rate": 9.387881058712888e-05,
197
+ "loss": 1.8435,
198
+ "step": 270
199
+ },
200
+ {
201
+ "epoch": 0.25225225225225223,
202
+ "grad_norm": 2.002121925354004,
203
+ "learning_rate": 9.310338793880458e-05,
204
+ "loss": 1.7743,
205
+ "step": 280
206
+ },
207
+ {
208
+ "epoch": 0.26126126126126126,
209
+ "grad_norm": 2.004910469055176,
210
+ "learning_rate": 9.228534225407781e-05,
211
+ "loss": 1.7949,
212
+ "step": 290
213
+ },
214
+ {
215
+ "epoch": 0.2702702702702703,
216
+ "grad_norm": 2.2141101360321045,
217
+ "learning_rate": 9.142548246219212e-05,
218
+ "loss": 1.7682,
219
+ "step": 300
220
+ },
221
+ {
222
+ "epoch": 0.27927927927927926,
223
+ "grad_norm": 2.036374807357788,
224
+ "learning_rate": 9.05246588405146e-05,
225
+ "loss": 1.761,
226
+ "step": 310
227
+ },
228
+ {
229
+ "epoch": 0.2882882882882883,
230
+ "grad_norm": 2.0280163288116455,
231
+ "learning_rate": 8.958376217373428e-05,
232
+ "loss": 1.8201,
233
+ "step": 320
234
+ },
235
+ {
236
+ "epoch": 0.2972972972972973,
237
+ "grad_norm": 2.244485378265381,
238
+ "learning_rate": 8.860372287300431e-05,
239
+ "loss": 1.762,
240
+ "step": 330
241
+ },
242
+ {
243
+ "epoch": 0.3063063063063063,
244
+ "grad_norm": 2.02524471282959,
245
+ "learning_rate": 8.758551005589967e-05,
246
+ "loss": 1.7456,
247
+ "step": 340
248
+ },
249
+ {
250
+ "epoch": 0.3153153153153153,
251
+ "grad_norm": 2.269183397293091,
252
+ "learning_rate": 8.653013058809944e-05,
253
+ "loss": 1.7447,
254
+ "step": 350
255
+ },
256
+ {
257
+ "epoch": 0.32432432432432434,
258
+ "grad_norm": 2.273630142211914,
259
+ "learning_rate": 8.543862808774192e-05,
260
+ "loss": 1.7693,
261
+ "step": 360
262
+ },
263
+ {
264
+ "epoch": 0.3333333333333333,
265
+ "grad_norm": 1.9659849405288696,
266
+ "learning_rate": 8.43120818934367e-05,
267
+ "loss": 1.7175,
268
+ "step": 370
269
+ },
270
+ {
271
+ "epoch": 0.34234234234234234,
272
+ "grad_norm": 2.624666690826416,
273
+ "learning_rate": 8.315160599695434e-05,
274
+ "loss": 1.7573,
275
+ "step": 380
276
+ },
277
+ {
278
+ "epoch": 0.35135135135135137,
279
+ "grad_norm": 2.3735556602478027,
280
+ "learning_rate": 8.195834794164925e-05,
281
+ "loss": 1.7034,
282
+ "step": 390
283
+ },
284
+ {
285
+ "epoch": 0.36036036036036034,
286
+ "grad_norm": 2.209813117980957,
287
+ "learning_rate": 8.073348768770462e-05,
288
+ "loss": 1.7617,
289
+ "step": 400
290
+ },
291
+ {
292
+ "epoch": 0.36936936936936937,
293
+ "grad_norm": 2.607027530670166,
294
+ "learning_rate": 7.947823644532198e-05,
295
+ "loss": 1.7826,
296
+ "step": 410
297
+ },
298
+ {
299
+ "epoch": 0.3783783783783784,
300
+ "grad_norm": 2.2740437984466553,
301
+ "learning_rate": 7.81938354770089e-05,
302
+ "loss": 1.7294,
303
+ "step": 420
304
+ },
305
+ {
306
+ "epoch": 0.38738738738738737,
307
+ "grad_norm": 2.4003989696502686,
308
+ "learning_rate": 7.688155487014936e-05,
309
+ "loss": 1.7352,
310
+ "step": 430
311
+ },
312
+ {
313
+ "epoch": 0.3963963963963964,
314
+ "grad_norm": 2.295224905014038,
315
+ "learning_rate": 7.554269228107044e-05,
316
+ "loss": 1.7472,
317
+ "step": 440
318
+ },
319
+ {
320
+ "epoch": 0.40540540540540543,
321
+ "grad_norm": 2.1371731758117676,
322
+ "learning_rate": 7.417857165184723e-05,
323
+ "loss": 1.66,
324
+ "step": 450
325
+ },
326
+ {
327
+ "epoch": 0.4144144144144144,
328
+ "grad_norm": 2.535508155822754,
329
+ "learning_rate": 7.279054190111506e-05,
330
+ "loss": 1.6889,
331
+ "step": 460
332
+ },
333
+ {
334
+ "epoch": 0.42342342342342343,
335
+ "grad_norm": 2.1276588439941406,
336
+ "learning_rate": 7.137997559018346e-05,
337
+ "loss": 1.7011,
338
+ "step": 470
339
+ },
340
+ {
341
+ "epoch": 0.43243243243243246,
342
+ "grad_norm": 2.2915713787078857,
343
+ "learning_rate": 6.994826756577082e-05,
344
+ "loss": 1.6353,
345
+ "step": 480
346
+ },
347
+ {
348
+ "epoch": 0.44144144144144143,
349
+ "grad_norm": 2.1963353157043457,
350
+ "learning_rate": 6.849683358070217e-05,
351
+ "loss": 1.6553,
352
+ "step": 490
353
+ },
354
+ {
355
+ "epoch": 0.45045045045045046,
356
+ "grad_norm": 2.3190746307373047,
357
+ "learning_rate": 6.702710889393369e-05,
358
+ "loss": 1.6471,
359
+ "step": 500
360
+ },
361
+ {
362
+ "epoch": 0.45045045045045046,
363
+ "eval_loss": 1.7359974384307861,
364
+ "eval_runtime": 51.5503,
365
+ "eval_samples_per_second": 76.527,
366
+ "eval_steps_per_second": 19.146,
367
+ "step": 500
368
+ },
369
+ {
370
+ "epoch": 0.4594594594594595,
371
+ "grad_norm": 2.2826740741729736,
372
+ "learning_rate": 6.554054685128856e-05,
373
+ "loss": 1.7566,
374
+ "step": 510
375
+ },
376
+ {
377
+ "epoch": 0.46846846846846846,
378
+ "grad_norm": 2.4409995079040527,
379
+ "learning_rate": 6.403861744830747e-05,
380
+ "loss": 1.7117,
381
+ "step": 520
382
+ },
383
+ {
384
+ "epoch": 0.4774774774774775,
385
+ "grad_norm": 2.375598907470703,
386
+ "learning_rate": 6.252280587663492e-05,
387
+ "loss": 1.6775,
388
+ "step": 530
389
+ },
390
+ {
391
+ "epoch": 0.4864864864864865,
392
+ "grad_norm": 2.470062255859375,
393
+ "learning_rate": 6.099461105537889e-05,
394
+ "loss": 1.6699,
395
+ "step": 540
396
+ },
397
+ {
398
+ "epoch": 0.4954954954954955,
399
+ "grad_norm": 2.439343214035034,
400
+ "learning_rate": 5.945554414889582e-05,
401
+ "loss": 1.6072,
402
+ "step": 550
403
+ },
404
+ {
405
+ "epoch": 0.5045045045045045,
406
+ "grad_norm": 2.3654677867889404,
407
+ "learning_rate": 5.790712707246705e-05,
408
+ "loss": 1.6006,
409
+ "step": 560
410
+ },
411
+ {
412
+ "epoch": 0.5135135135135135,
413
+ "grad_norm": 2.3721349239349365,
414
+ "learning_rate": 5.6350890987343944e-05,
415
+ "loss": 1.6737,
416
+ "step": 570
417
+ },
418
+ {
419
+ "epoch": 0.5225225225225225,
420
+ "grad_norm": 2.794997453689575,
421
+ "learning_rate": 5.478837478665021e-05,
422
+ "loss": 1.695,
423
+ "step": 580
424
+ },
425
+ {
426
+ "epoch": 0.5315315315315315,
427
+ "grad_norm": 2.6293647289276123,
428
+ "learning_rate": 5.322112357363841e-05,
429
+ "loss": 1.6577,
430
+ "step": 590
431
+ },
432
+ {
433
+ "epoch": 0.5405405405405406,
434
+ "grad_norm": 2.333765745162964,
435
+ "learning_rate": 5.1650687133805674e-05,
436
+ "loss": 1.6792,
437
+ "step": 600
438
+ },
439
+ {
440
+ "epoch": 0.5495495495495496,
441
+ "grad_norm": 2.4596219062805176,
442
+ "learning_rate": 5.007861840237923e-05,
443
+ "loss": 1.6937,
444
+ "step": 610
445
+ },
446
+ {
447
+ "epoch": 0.5585585585585585,
448
+ "grad_norm": 2.6094493865966797,
449
+ "learning_rate": 4.850647192868735e-05,
450
+ "loss": 1.6714,
451
+ "step": 620
452
+ },
453
+ {
454
+ "epoch": 0.5675675675675675,
455
+ "grad_norm": 2.353469133377075,
456
+ "learning_rate": 4.69358023389342e-05,
457
+ "loss": 1.6349,
458
+ "step": 630
459
+ },
460
+ {
461
+ "epoch": 0.5765765765765766,
462
+ "grad_norm": 2.481355667114258,
463
+ "learning_rate": 4.536816279889865e-05,
464
+ "loss": 1.5774,
465
+ "step": 640
466
+ },
467
+ {
468
+ "epoch": 0.5855855855855856,
469
+ "grad_norm": 2.60036563873291,
470
+ "learning_rate": 4.380510347807725e-05,
471
+ "loss": 1.6312,
472
+ "step": 650
473
+ },
474
+ {
475
+ "epoch": 0.5945945945945946,
476
+ "grad_norm": 2.5595216751098633,
477
+ "learning_rate": 4.224817001679011e-05,
478
+ "loss": 1.6359,
479
+ "step": 660
480
+ },
481
+ {
482
+ "epoch": 0.6036036036036037,
483
+ "grad_norm": 2.7325408458709717,
484
+ "learning_rate": 4.06989019977654e-05,
485
+ "loss": 1.7161,
486
+ "step": 670
487
+ },
488
+ {
489
+ "epoch": 0.6126126126126126,
490
+ "grad_norm": 2.384521007537842,
491
+ "learning_rate": 3.9158831423714036e-05,
492
+ "loss": 1.6495,
493
+ "step": 680
494
+ },
495
+ {
496
+ "epoch": 0.6216216216216216,
497
+ "grad_norm": 2.7080259323120117,
498
+ "learning_rate": 3.762948120239988e-05,
499
+ "loss": 1.6727,
500
+ "step": 690
501
+ },
502
+ {
503
+ "epoch": 0.6306306306306306,
504
+ "grad_norm": 2.678630828857422,
505
+ "learning_rate": 3.6112363640703476e-05,
506
+ "loss": 1.6925,
507
+ "step": 700
508
+ },
509
+ {
510
+ "epoch": 0.6396396396396397,
511
+ "grad_norm": 2.457289457321167,
512
+ "learning_rate": 3.4608978949168614e-05,
513
+ "loss": 1.6202,
514
+ "step": 710
515
+ },
516
+ {
517
+ "epoch": 0.6486486486486487,
518
+ "grad_norm": 2.3443291187286377,
519
+ "learning_rate": 3.312081375851038e-05,
520
+ "loss": 1.5941,
521
+ "step": 720
522
+ },
523
+ {
524
+ "epoch": 0.6576576576576577,
525
+ "grad_norm": 2.434515953063965,
526
+ "learning_rate": 3.164933964955174e-05,
527
+ "loss": 1.622,
528
+ "step": 730
529
+ },
530
+ {
531
+ "epoch": 0.6666666666666666,
532
+ "grad_norm": 2.5307750701904297,
533
+ "learning_rate": 3.019601169804216e-05,
534
+ "loss": 1.6565,
535
+ "step": 740
536
+ },
537
+ {
538
+ "epoch": 0.6756756756756757,
539
+ "grad_norm": 2.6859347820281982,
540
+ "learning_rate": 2.876226703579761e-05,
541
+ "loss": 1.6155,
542
+ "step": 750
543
+ },
544
+ {
545
+ "epoch": 0.6846846846846847,
546
+ "grad_norm": 2.6468803882598877,
547
+ "learning_rate": 2.7349523429584305e-05,
548
+ "loss": 1.6692,
549
+ "step": 760
550
+ },
551
+ {
552
+ "epoch": 0.6936936936936937,
553
+ "grad_norm": 2.4167847633361816,
554
+ "learning_rate": 2.595917787915176e-05,
555
+ "loss": 1.5915,
556
+ "step": 770
557
+ },
558
+ {
559
+ "epoch": 0.7027027027027027,
560
+ "grad_norm": 2.6717164516448975,
561
+ "learning_rate": 2.459260523580154e-05,
562
+ "loss": 1.6659,
563
+ "step": 780
564
+ },
565
+ {
566
+ "epoch": 0.7117117117117117,
567
+ "grad_norm": 2.5674118995666504,
568
+ "learning_rate": 2.325115684285743e-05,
569
+ "loss": 1.5425,
570
+ "step": 790
571
+ },
572
+ {
573
+ "epoch": 0.7207207207207207,
574
+ "grad_norm": 2.6461567878723145,
575
+ "learning_rate": 2.1936159199381743e-05,
576
+ "loss": 1.6597,
577
+ "step": 800
578
+ },
579
+ {
580
+ "epoch": 0.7297297297297297,
581
+ "grad_norm": 2.5922305583953857,
582
+ "learning_rate": 2.0648912648459074e-05,
583
+ "loss": 1.6627,
584
+ "step": 810
585
+ },
586
+ {
587
+ "epoch": 0.7387387387387387,
588
+ "grad_norm": 2.553170680999756,
589
+ "learning_rate": 1.939069009134433e-05,
590
+ "loss": 1.5385,
591
+ "step": 820
592
+ },
593
+ {
594
+ "epoch": 0.7477477477477478,
595
+ "grad_norm": 2.7178986072540283,
596
+ "learning_rate": 1.8162735728746975e-05,
597
+ "loss": 1.5861,
598
+ "step": 830
599
+ },
600
+ {
601
+ "epoch": 0.7567567567567568,
602
+ "grad_norm": 2.497939348220825,
603
+ "learning_rate": 1.6966263830495936e-05,
604
+ "loss": 1.694,
605
+ "step": 840
606
+ },
607
+ {
608
+ "epoch": 0.7657657657657657,
609
+ "grad_norm": 2.7136847972869873,
610
+ "learning_rate": 1.5802457534801718e-05,
611
+ "loss": 1.6814,
612
+ "step": 850
613
+ },
614
+ {
615
+ "epoch": 0.7747747747747747,
616
+ "grad_norm": 2.701145887374878,
617
+ "learning_rate": 1.4672467678303386e-05,
618
+ "loss": 1.5897,
619
+ "step": 860
620
+ },
621
+ {
622
+ "epoch": 0.7837837837837838,
623
+ "grad_norm": 2.8379249572753906,
624
+ "learning_rate": 1.3577411658056966e-05,
625
+ "loss": 1.6266,
626
+ "step": 870
627
+ },
628
+ {
629
+ "epoch": 0.7927927927927928,
630
+ "grad_norm": 2.6730990409851074,
631
+ "learning_rate": 1.2518372326590877e-05,
632
+ "loss": 1.5935,
633
+ "step": 880
634
+ },
635
+ {
636
+ "epoch": 0.8018018018018018,
637
+ "grad_norm": 2.6736879348754883,
638
+ "learning_rate": 1.149639692112095e-05,
639
+ "loss": 1.5966,
640
+ "step": 890
641
+ },
642
+ {
643
+ "epoch": 0.8108108108108109,
644
+ "grad_norm": 2.9796481132507324,
645
+ "learning_rate": 1.0512496027983714e-05,
646
+ "loss": 1.6568,
647
+ "step": 900
648
+ },
649
+ {
650
+ "epoch": 0.8198198198198198,
651
+ "grad_norm": 2.821319103240967,
652
+ "learning_rate": 9.56764258331226e-06,
653
+ "loss": 1.6065,
654
+ "step": 910
655
+ },
656
+ {
657
+ "epoch": 0.8288288288288288,
658
+ "grad_norm": 2.5861129760742188,
659
+ "learning_rate": 8.662770910942692e-06,
660
+ "loss": 1.5545,
661
+ "step": 920
662
+ },
663
+ {
664
+ "epoch": 0.8378378378378378,
665
+ "grad_norm": 2.6395463943481445,
666
+ "learning_rate": 7.798775798502483e-06,
667
+ "loss": 1.5567,
668
+ "step": 930
669
+ },
670
+ {
671
+ "epoch": 0.8468468468468469,
672
+ "grad_norm": 2.8468408584594727,
673
+ "learning_rate": 6.976511612594622e-06,
674
+ "loss": 1.5972,
675
+ "step": 940
676
+ },
677
+ {
678
+ "epoch": 0.8558558558558559,
679
+ "grad_norm": 3.0037944316864014,
680
+ "learning_rate": 6.1967914539521e-06,
681
+ "loss": 1.6084,
682
+ "step": 950
683
+ },
684
+ {
685
+ "epoch": 0.8648648648648649,
686
+ "grad_norm": 2.7935662269592285,
687
+ "learning_rate": 5.460386353398583e-06,
688
+ "loss": 1.571,
689
+ "step": 960
690
+ },
691
+ {
692
+ "epoch": 0.8738738738738738,
693
+ "grad_norm": 2.817194938659668,
694
+ "learning_rate": 4.768024509410096e-06,
695
+ "loss": 1.5944,
696
+ "step": 970
697
+ },
698
+ {
699
+ "epoch": 0.8828828828828829,
700
+ "grad_norm": 2.5486297607421875,
701
+ "learning_rate": 4.120390568031673e-06,
702
+ "loss": 1.5694,
703
+ "step": 980
704
+ },
705
+ {
706
+ "epoch": 0.8918918918918919,
707
+ "grad_norm": 2.557070255279541,
708
+ "learning_rate": 3.51812494586114e-06,
709
+ "loss": 1.6251,
710
+ "step": 990
711
+ },
712
+ {
713
+ "epoch": 0.9009009009009009,
714
+ "grad_norm": 2.4766974449157715,
715
+ "learning_rate": 2.9618231967694532e-06,
716
+ "loss": 1.6062,
717
+ "step": 1000
718
+ },
719
+ {
720
+ "epoch": 0.9009009009009009,
721
+ "eval_loss": 1.6383798122406006,
722
+ "eval_runtime": 51.1987,
723
+ "eval_samples_per_second": 77.053,
724
+ "eval_steps_per_second": 19.278,
725
+ "step": 1000
726
+ },
727
+ {
728
+ "epoch": 0.9099099099099099,
729
+ "grad_norm": 2.7714474201202393,
730
+ "learning_rate": 2.452035422983734e-06,
731
+ "loss": 1.5405,
732
+ "step": 1010
733
+ },
734
+ {
735
+ "epoch": 0.918918918918919,
736
+ "grad_norm": 2.9670562744140625,
737
+ "learning_rate": 1.9892657311155248e-06,
738
+ "loss": 1.6195,
739
+ "step": 1020
740
+ },
741
+ {
742
+ "epoch": 0.9279279279279279,
743
+ "grad_norm": 2.3906571865081787,
744
+ "learning_rate": 1.5739717336720084e-06,
745
+ "loss": 1.5761,
746
+ "step": 1030
747
+ },
748
+ {
749
+ "epoch": 0.9369369369369369,
750
+ "grad_norm": 2.843130588531494,
751
+ "learning_rate": 1.2065640965432001e-06,
752
+ "loss": 1.5647,
753
+ "step": 1040
754
+ },
755
+ {
756
+ "epoch": 0.9459459459459459,
757
+ "grad_norm": 2.8360941410064697,
758
+ "learning_rate": 8.874061329125938e-07,
759
+ "loss": 1.6573,
760
+ "step": 1050
761
+ },
762
+ {
763
+ "epoch": 0.954954954954955,
764
+ "grad_norm": 2.885618209838867,
765
+ "learning_rate": 6.168134439928364e-07,
766
+ "loss": 1.5961,
767
+ "step": 1060
768
+ },
769
+ {
770
+ "epoch": 0.963963963963964,
771
+ "grad_norm": 2.8238613605499268,
772
+ "learning_rate": 3.950536069415756e-07,
773
+ "loss": 1.6878,
774
+ "step": 1070
775
+ },
776
+ {
777
+ "epoch": 0.972972972972973,
778
+ "grad_norm": 2.663311719894409,
779
+ "learning_rate": 2.223459102662695e-07,
780
+ "loss": 1.6297,
781
+ "step": 1080
782
+ },
783
+ {
784
+ "epoch": 0.9819819819819819,
785
+ "grad_norm": 2.4804253578186035,
786
+ "learning_rate": 9.886113697944454e-08,
787
+ "loss": 1.6147,
788
+ "step": 1090
789
+ },
790
+ {
791
+ "epoch": 0.990990990990991,
792
+ "grad_norm": 2.756866216659546,
793
+ "learning_rate": 2.4721395718924068e-08,
794
+ "loss": 1.6267,
795
+ "step": 1100
796
+ },
797
+ {
798
+ "epoch": 1.0,
799
+ "grad_norm": 2.6657907962799072,
800
+ "learning_rate": 0.0,
801
+ "loss": 1.5991,
802
+ "step": 1110
803
+ },
804
+ {
805
+ "epoch": 1.0,
806
+ "step": 1110,
807
+ "total_flos": 2.3250303254423142e+17,
808
+ "train_loss": 1.7593931747986389,
809
+ "train_runtime": 934.3961,
810
+ "train_samples_per_second": 37.998,
811
+ "train_steps_per_second": 1.188
812
+ }
813
+ ],
814
+ "logging_steps": 10,
815
+ "max_steps": 1110,
816
+ "num_input_tokens_seen": 0,
817
+ "num_train_epochs": 1,
818
+ "save_steps": 50000,
819
+ "stateful_callbacks": {
820
+ "TrainerControl": {
821
+ "args": {
822
+ "should_epoch_stop": false,
823
+ "should_evaluate": false,
824
+ "should_log": false,
825
+ "should_save": true,
826
+ "should_training_stop": true
827
+ },
828
+ "attributes": {}
829
+ }
830
+ },
831
+ "total_flos": 2.3250303254423142e+17,
832
+ "train_batch_size": 8,
833
+ "trial_name": null,
834
+ "trial_params": null
835
+ }
training_eval_loss.png ADDED
training_loss.png ADDED