AmberYifan commited on
Commit
593585e
·
verified ·
1 Parent(s): 80995d6

End of training

Browse files
Files changed (5) hide show
  1. README.md +2 -1
  2. all_results.json +8 -0
  3. train_results.json +8 -0
  4. trainer_state.json +694 -0
  5. training_loss.png +0 -0
README.md CHANGED
@@ -4,6 +4,7 @@ license: apache-2.0
4
  base_model: AmberYifan/qwen2.5-7b-instruct-full-pretrain-junk-tweet-1m-en-gpt
5
  tags:
6
  - llama-factory
 
7
  - generated_from_trainer
8
  model-index:
9
  - name: qwen2.5-7b-instruct-full-pretrain-junk-tweet-1m-en-gpt-sft
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
15
 
16
  # qwen2.5-7b-instruct-full-pretrain-junk-tweet-1m-en-gpt-sft
17
 
18
- This model is a fine-tuned version of [AmberYifan/qwen2.5-7b-instruct-full-pretrain-junk-tweet-1m-en-gpt](https://huggingface.co/AmberYifan/qwen2.5-7b-instruct-full-pretrain-junk-tweet-1m-en-gpt) on an unknown dataset.
19
 
20
  ## Model description
21
 
 
4
  base_model: AmberYifan/qwen2.5-7b-instruct-full-pretrain-junk-tweet-1m-en-gpt
5
  tags:
6
  - llama-factory
7
+ - full
8
  - generated_from_trainer
9
  model-index:
10
  - name: qwen2.5-7b-instruct-full-pretrain-junk-tweet-1m-en-gpt-sft
 
16
 
17
  # qwen2.5-7b-instruct-full-pretrain-junk-tweet-1m-en-gpt-sft
18
 
19
+ This model is a fine-tuned version of [AmberYifan/qwen2.5-7b-instruct-full-pretrain-junk-tweet-1m-en-gpt](https://huggingface.co/AmberYifan/qwen2.5-7b-instruct-full-pretrain-junk-tweet-1m-en-gpt) on the alpaca_en dataset.
20
 
21
  ## Model description
22
 
all_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 3.0,
3
+ "total_flos": 3418309214208.0,
4
+ "train_loss": 0.6678480519273411,
5
+ "train_runtime": 1116.3574,
6
+ "train_samples_per_second": 13.437,
7
+ "train_steps_per_second": 0.841
8
+ }
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 3.0,
3
+ "total_flos": 3418309214208.0,
4
+ "train_loss": 0.6678480519273411,
5
+ "train_runtime": 1116.3574,
6
+ "train_samples_per_second": 13.437,
7
+ "train_steps_per_second": 0.841
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,694 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 3.0,
6
+ "eval_steps": 500,
7
+ "global_step": 939,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.032,
14
+ "grad_norm": 39.1096352326515,
15
+ "learning_rate": 9.574468085106384e-07,
16
+ "loss": 2.3056,
17
+ "step": 10
18
+ },
19
+ {
20
+ "epoch": 0.064,
21
+ "grad_norm": 16.899076690390917,
22
+ "learning_rate": 2.021276595744681e-06,
23
+ "loss": 1.7717,
24
+ "step": 20
25
+ },
26
+ {
27
+ "epoch": 0.096,
28
+ "grad_norm": 11.308563003803815,
29
+ "learning_rate": 3.0851063829787237e-06,
30
+ "loss": 1.3068,
31
+ "step": 30
32
+ },
33
+ {
34
+ "epoch": 0.128,
35
+ "grad_norm": 10.274901415072474,
36
+ "learning_rate": 4.148936170212766e-06,
37
+ "loss": 1.1424,
38
+ "step": 40
39
+ },
40
+ {
41
+ "epoch": 0.16,
42
+ "grad_norm": 9.548706234065625,
43
+ "learning_rate": 5.212765957446809e-06,
44
+ "loss": 1.1309,
45
+ "step": 50
46
+ },
47
+ {
48
+ "epoch": 0.192,
49
+ "grad_norm": 15.83870685772359,
50
+ "learning_rate": 6.276595744680851e-06,
51
+ "loss": 1.1985,
52
+ "step": 60
53
+ },
54
+ {
55
+ "epoch": 0.224,
56
+ "grad_norm": 11.382728815514827,
57
+ "learning_rate": 7.340425531914894e-06,
58
+ "loss": 1.1048,
59
+ "step": 70
60
+ },
61
+ {
62
+ "epoch": 0.256,
63
+ "grad_norm": 11.650131589337152,
64
+ "learning_rate": 8.404255319148937e-06,
65
+ "loss": 1.1035,
66
+ "step": 80
67
+ },
68
+ {
69
+ "epoch": 0.288,
70
+ "grad_norm": 8.568640515520194,
71
+ "learning_rate": 9.46808510638298e-06,
72
+ "loss": 1.16,
73
+ "step": 90
74
+ },
75
+ {
76
+ "epoch": 0.32,
77
+ "grad_norm": 8.08410170361631,
78
+ "learning_rate": 9.999136119166803e-06,
79
+ "loss": 1.155,
80
+ "step": 100
81
+ },
82
+ {
83
+ "epoch": 0.352,
84
+ "grad_norm": 7.884413737662858,
85
+ "learning_rate": 9.9922268634943e-06,
86
+ "loss": 1.1018,
87
+ "step": 110
88
+ },
89
+ {
90
+ "epoch": 0.384,
91
+ "grad_norm": 8.691042220015294,
92
+ "learning_rate": 9.978417901361958e-06,
93
+ "loss": 1.2083,
94
+ "step": 120
95
+ },
96
+ {
97
+ "epoch": 0.416,
98
+ "grad_norm": 7.225665647795149,
99
+ "learning_rate": 9.95772831799724e-06,
100
+ "loss": 1.1581,
101
+ "step": 130
102
+ },
103
+ {
104
+ "epoch": 0.448,
105
+ "grad_norm": 8.24708045856726,
106
+ "learning_rate": 9.930186708264902e-06,
107
+ "loss": 1.1921,
108
+ "step": 140
109
+ },
110
+ {
111
+ "epoch": 0.48,
112
+ "grad_norm": 7.869724265398933,
113
+ "learning_rate": 9.895831137146319e-06,
114
+ "loss": 1.139,
115
+ "step": 150
116
+ },
117
+ {
118
+ "epoch": 0.512,
119
+ "grad_norm": 11.419142424683425,
120
+ "learning_rate": 9.854709087130261e-06,
121
+ "loss": 1.1154,
122
+ "step": 160
123
+ },
124
+ {
125
+ "epoch": 0.544,
126
+ "grad_norm": 7.9813979209772,
127
+ "learning_rate": 9.80687739258782e-06,
128
+ "loss": 1.1782,
129
+ "step": 170
130
+ },
131
+ {
132
+ "epoch": 0.576,
133
+ "grad_norm": 7.70519368380932,
134
+ "learning_rate": 9.7524021612222e-06,
135
+ "loss": 1.2527,
136
+ "step": 180
137
+ },
138
+ {
139
+ "epoch": 0.608,
140
+ "grad_norm": 9.08198271166994,
141
+ "learning_rate": 9.691358682701927e-06,
142
+ "loss": 1.2212,
143
+ "step": 190
144
+ },
145
+ {
146
+ "epoch": 0.64,
147
+ "grad_norm": 7.813420466215621,
148
+ "learning_rate": 9.623831324603755e-06,
149
+ "loss": 1.1228,
150
+ "step": 200
151
+ },
152
+ {
153
+ "epoch": 0.672,
154
+ "grad_norm": 10.205549379727113,
155
+ "learning_rate": 9.549913415809084e-06,
156
+ "loss": 1.1546,
157
+ "step": 210
158
+ },
159
+ {
160
+ "epoch": 0.704,
161
+ "grad_norm": 6.887996653451635,
162
+ "learning_rate": 9.469707117515068e-06,
163
+ "loss": 1.0508,
164
+ "step": 220
165
+ },
166
+ {
167
+ "epoch": 0.736,
168
+ "grad_norm": 7.340913854172283,
169
+ "learning_rate": 9.383323282038632e-06,
170
+ "loss": 1.1068,
171
+ "step": 230
172
+ },
173
+ {
174
+ "epoch": 0.768,
175
+ "grad_norm": 6.499910309613238,
176
+ "learning_rate": 9.29088129960862e-06,
177
+ "loss": 1.0816,
178
+ "step": 240
179
+ },
180
+ {
181
+ "epoch": 0.8,
182
+ "grad_norm": 6.572574784980115,
183
+ "learning_rate": 9.192508933357753e-06,
184
+ "loss": 1.0927,
185
+ "step": 250
186
+ },
187
+ {
188
+ "epoch": 0.832,
189
+ "grad_norm": 9.46358573340417,
190
+ "learning_rate": 9.088342142742493e-06,
191
+ "loss": 1.158,
192
+ "step": 260
193
+ },
194
+ {
195
+ "epoch": 0.864,
196
+ "grad_norm": 7.105176072989092,
197
+ "learning_rate": 8.978524895634842e-06,
198
+ "loss": 1.2067,
199
+ "step": 270
200
+ },
201
+ {
202
+ "epoch": 0.896,
203
+ "grad_norm": 6.4823504227364666,
204
+ "learning_rate": 8.86320896934581e-06,
205
+ "loss": 1.1178,
206
+ "step": 280
207
+ },
208
+ {
209
+ "epoch": 0.928,
210
+ "grad_norm": 7.93249487558393,
211
+ "learning_rate": 8.742553740855507e-06,
212
+ "loss": 1.1742,
213
+ "step": 290
214
+ },
215
+ {
216
+ "epoch": 0.96,
217
+ "grad_norm": 7.227291656812382,
218
+ "learning_rate": 8.616725966539831e-06,
219
+ "loss": 1.1389,
220
+ "step": 300
221
+ },
222
+ {
223
+ "epoch": 0.992,
224
+ "grad_norm": 13.901826169729663,
225
+ "learning_rate": 8.485899551698166e-06,
226
+ "loss": 1.161,
227
+ "step": 310
228
+ },
229
+ {
230
+ "epoch": 1.0224,
231
+ "grad_norm": 5.1219616235543555,
232
+ "learning_rate": 8.350255310200611e-06,
233
+ "loss": 0.7863,
234
+ "step": 320
235
+ },
236
+ {
237
+ "epoch": 1.0544,
238
+ "grad_norm": 8.548368757001553,
239
+ "learning_rate": 8.209980714586955e-06,
240
+ "loss": 0.5962,
241
+ "step": 330
242
+ },
243
+ {
244
+ "epoch": 1.0864,
245
+ "grad_norm": 92.54495701445232,
246
+ "learning_rate": 8.065269636962765e-06,
247
+ "loss": 0.5823,
248
+ "step": 340
249
+ },
250
+ {
251
+ "epoch": 1.1184,
252
+ "grad_norm": 4.6475189267999,
253
+ "learning_rate": 7.916322081050708e-06,
254
+ "loss": 0.5123,
255
+ "step": 350
256
+ },
257
+ {
258
+ "epoch": 1.1504,
259
+ "grad_norm": 5.5267222999889265,
260
+ "learning_rate": 7.76334390576742e-06,
261
+ "loss": 0.5728,
262
+ "step": 360
263
+ },
264
+ {
265
+ "epoch": 1.1824,
266
+ "grad_norm": 8.723856258103218,
267
+ "learning_rate": 7.60654654070796e-06,
268
+ "loss": 0.5965,
269
+ "step": 370
270
+ },
271
+ {
272
+ "epoch": 1.2144,
273
+ "grad_norm": 10.083042960722508,
274
+ "learning_rate": 7.446146693931111e-06,
275
+ "loss": 0.6133,
276
+ "step": 380
277
+ },
278
+ {
279
+ "epoch": 1.2464,
280
+ "grad_norm": 6.236613141647098,
281
+ "learning_rate": 7.282366052449351e-06,
282
+ "loss": 0.5545,
283
+ "step": 390
284
+ },
285
+ {
286
+ "epoch": 1.2784,
287
+ "grad_norm": 7.900489916591024,
288
+ "learning_rate": 7.115430975837457e-06,
289
+ "loss": 0.5395,
290
+ "step": 400
291
+ },
292
+ {
293
+ "epoch": 1.3104,
294
+ "grad_norm": 6.751119383827855,
295
+ "learning_rate": 6.945572183383229e-06,
296
+ "loss": 0.5185,
297
+ "step": 410
298
+ },
299
+ {
300
+ "epoch": 1.3424,
301
+ "grad_norm": 6.591602664759515,
302
+ "learning_rate": 6.773024435212678e-06,
303
+ "loss": 0.58,
304
+ "step": 420
305
+ },
306
+ {
307
+ "epoch": 1.3744,
308
+ "grad_norm": 5.020883690781568,
309
+ "learning_rate": 6.598026207830428e-06,
310
+ "loss": 0.582,
311
+ "step": 430
312
+ },
313
+ {
314
+ "epoch": 1.4064,
315
+ "grad_norm": 4.358741025969657,
316
+ "learning_rate": 6.4208193645237314e-06,
317
+ "loss": 0.6483,
318
+ "step": 440
319
+ },
320
+ {
321
+ "epoch": 1.4384000000000001,
322
+ "grad_norm": 6.125196557266611,
323
+ "learning_rate": 6.241648821085666e-06,
324
+ "loss": 0.5695,
325
+ "step": 450
326
+ },
327
+ {
328
+ "epoch": 1.4704,
329
+ "grad_norm": 6.1820432941203185,
330
+ "learning_rate": 6.060762207319479e-06,
331
+ "loss": 0.5888,
332
+ "step": 460
333
+ },
334
+ {
335
+ "epoch": 1.5024,
336
+ "grad_norm": 6.987894758967655,
337
+ "learning_rate": 5.878409524791931e-06,
338
+ "loss": 0.5993,
339
+ "step": 470
340
+ },
341
+ {
342
+ "epoch": 1.5344,
343
+ "grad_norm": 7.385598211189569,
344
+ "learning_rate": 5.694842801308651e-06,
345
+ "loss": 0.5826,
346
+ "step": 480
347
+ },
348
+ {
349
+ "epoch": 1.5664,
350
+ "grad_norm": 6.995583217988908,
351
+ "learning_rate": 5.510315742589042e-06,
352
+ "loss": 0.6501,
353
+ "step": 490
354
+ },
355
+ {
356
+ "epoch": 1.5984,
357
+ "grad_norm": 6.721928240858902,
358
+ "learning_rate": 5.325083381622165e-06,
359
+ "loss": 0.6005,
360
+ "step": 500
361
+ },
362
+ {
363
+ "epoch": 1.6303999999999998,
364
+ "grad_norm": 7.196864925847178,
365
+ "learning_rate": 5.139401726188208e-06,
366
+ "loss": 0.6475,
367
+ "step": 510
368
+ },
369
+ {
370
+ "epoch": 1.6623999999999999,
371
+ "grad_norm": 4.449569168084183,
372
+ "learning_rate": 4.953527405032723e-06,
373
+ "loss": 0.5558,
374
+ "step": 520
375
+ },
376
+ {
377
+ "epoch": 1.6944,
378
+ "grad_norm": 7.246820656990819,
379
+ "learning_rate": 4.767717313182611e-06,
380
+ "loss": 0.5946,
381
+ "step": 530
382
+ },
383
+ {
384
+ "epoch": 1.7264,
385
+ "grad_norm": 7.524427037650954,
386
+ "learning_rate": 4.582228256894093e-06,
387
+ "loss": 0.5453,
388
+ "step": 540
389
+ },
390
+ {
391
+ "epoch": 1.7584,
392
+ "grad_norm": 6.417637171346657,
393
+ "learning_rate": 4.397316598723385e-06,
394
+ "loss": 0.5631,
395
+ "step": 550
396
+ },
397
+ {
398
+ "epoch": 1.7904,
399
+ "grad_norm": 6.044008702275205,
400
+ "learning_rate": 4.2132379032105695e-06,
401
+ "loss": 0.5471,
402
+ "step": 560
403
+ },
404
+ {
405
+ "epoch": 1.8224,
406
+ "grad_norm": 7.703075980725136,
407
+ "learning_rate": 4.030246583666437e-06,
408
+ "loss": 0.5358,
409
+ "step": 570
410
+ },
411
+ {
412
+ "epoch": 1.8544,
413
+ "grad_norm": 6.933710387117784,
414
+ "learning_rate": 3.848595550550401e-06,
415
+ "loss": 0.5729,
416
+ "step": 580
417
+ },
418
+ {
419
+ "epoch": 1.8864,
420
+ "grad_norm": 5.836392891956106,
421
+ "learning_rate": 3.668535861925509e-06,
422
+ "loss": 0.6172,
423
+ "step": 590
424
+ },
425
+ {
426
+ "epoch": 1.9184,
427
+ "grad_norm": 4.850237169162472,
428
+ "learning_rate": 3.4903163764736104e-06,
429
+ "loss": 0.5838,
430
+ "step": 600
431
+ },
432
+ {
433
+ "epoch": 1.9504000000000001,
434
+ "grad_norm": 7.529837809243237,
435
+ "learning_rate": 3.314183409550293e-06,
436
+ "loss": 0.5265,
437
+ "step": 610
438
+ },
439
+ {
440
+ "epoch": 1.9824000000000002,
441
+ "grad_norm": 6.026318906361559,
442
+ "learning_rate": 3.140380392754901e-06,
443
+ "loss": 0.5766,
444
+ "step": 620
445
+ },
446
+ {
447
+ "epoch": 2.0128,
448
+ "grad_norm": 3.474380962484068,
449
+ "learning_rate": 2.969147537486175e-06,
450
+ "loss": 0.3757,
451
+ "step": 630
452
+ },
453
+ {
454
+ "epoch": 2.0448,
455
+ "grad_norm": 4.406661780903807,
456
+ "learning_rate": 2.800721502948506e-06,
457
+ "loss": 0.199,
458
+ "step": 640
459
+ },
460
+ {
461
+ "epoch": 2.0768,
462
+ "grad_norm": 3.7879901535036704,
463
+ "learning_rate": 2.635335069067617e-06,
464
+ "loss": 0.2483,
465
+ "step": 650
466
+ },
467
+ {
468
+ "epoch": 2.1088,
469
+ "grad_norm": 3.88954400445382,
470
+ "learning_rate": 2.4732168147677927e-06,
471
+ "loss": 0.1854,
472
+ "step": 660
473
+ },
474
+ {
475
+ "epoch": 2.1408,
476
+ "grad_norm": 4.9291108370074355,
477
+ "learning_rate": 2.314590802055232e-06,
478
+ "loss": 0.2256,
479
+ "step": 670
480
+ },
481
+ {
482
+ "epoch": 2.1728,
483
+ "grad_norm": 4.537416650941455,
484
+ "learning_rate": 2.159676266344222e-06,
485
+ "loss": 0.2417,
486
+ "step": 680
487
+ },
488
+ {
489
+ "epoch": 2.2048,
490
+ "grad_norm": 4.589342012607399,
491
+ "learning_rate": 2.0086873134540626e-06,
492
+ "loss": 0.2397,
493
+ "step": 690
494
+ },
495
+ {
496
+ "epoch": 2.2368,
497
+ "grad_norm": 5.00572141950068,
498
+ "learning_rate": 1.8618326236955908e-06,
499
+ "loss": 0.2253,
500
+ "step": 700
501
+ },
502
+ {
503
+ "epoch": 2.2688,
504
+ "grad_norm": 6.384579288199291,
505
+ "learning_rate": 1.7193151634562071e-06,
506
+ "loss": 0.2245,
507
+ "step": 710
508
+ },
509
+ {
510
+ "epoch": 2.3008,
511
+ "grad_norm": 3.5014444758775705,
512
+ "learning_rate": 1.581331904682089e-06,
513
+ "loss": 0.1615,
514
+ "step": 720
515
+ },
516
+ {
517
+ "epoch": 2.3327999999999998,
518
+ "grad_norm": 5.184650365658737,
519
+ "learning_rate": 1.4480735526452427e-06,
520
+ "loss": 0.2104,
521
+ "step": 730
522
+ },
523
+ {
524
+ "epoch": 2.3648,
525
+ "grad_norm": 4.482763808859523,
526
+ "learning_rate": 1.319724282371664e-06,
527
+ "loss": 0.1997,
528
+ "step": 740
529
+ },
530
+ {
531
+ "epoch": 2.3968,
532
+ "grad_norm": 7.081210646411153,
533
+ "learning_rate": 1.1964614840949002e-06,
534
+ "loss": 0.2237,
535
+ "step": 750
536
+ },
537
+ {
538
+ "epoch": 2.4288,
539
+ "grad_norm": 4.862708954606646,
540
+ "learning_rate": 1.078455518086784e-06,
541
+ "loss": 0.1726,
542
+ "step": 760
543
+ },
544
+ {
545
+ "epoch": 2.4608,
546
+ "grad_norm": 4.664882409595536,
547
+ "learning_rate": 9.658694792042284e-07,
548
+ "loss": 0.1998,
549
+ "step": 770
550
+ },
551
+ {
552
+ "epoch": 2.4928,
553
+ "grad_norm": 5.291912466864956,
554
+ "learning_rate": 8.58858971477457e-07,
555
+ "loss": 0.206,
556
+ "step": 780
557
+ },
558
+ {
559
+ "epoch": 2.5248,
560
+ "grad_norm": 4.565105616724458,
561
+ "learning_rate": 7.575718930512516e-07,
562
+ "loss": 0.2173,
563
+ "step": 790
564
+ },
565
+ {
566
+ "epoch": 2.5568,
567
+ "grad_norm": 4.915852306375132,
568
+ "learning_rate": 6.621482317764105e-07,
569
+ "loss": 0.2037,
570
+ "step": 800
571
+ },
572
+ {
573
+ "epoch": 2.5888,
574
+ "grad_norm": 4.505761751431015,
575
+ "learning_rate": 5.727198717339511e-07,
576
+ "loss": 0.2314,
577
+ "step": 810
578
+ },
579
+ {
580
+ "epoch": 2.6208,
581
+ "grad_norm": 3.909039745871737,
582
+ "learning_rate": 4.894104109594466e-07,
583
+ "loss": 0.2208,
584
+ "step": 820
585
+ },
586
+ {
587
+ "epoch": 2.6528,
588
+ "grad_norm": 6.617088897118751,
589
+ "learning_rate": 4.123349906194357e-07,
590
+ "loss": 0.2099,
591
+ "step": 830
592
+ },
593
+ {
594
+ "epoch": 2.6848,
595
+ "grad_norm": 5.4835386407124505,
596
+ "learning_rate": 3.416001358759635e-07,
597
+ "loss": 0.2293,
598
+ "step": 840
599
+ },
600
+ {
601
+ "epoch": 2.7168,
602
+ "grad_norm": 7.402876759701894,
603
+ "learning_rate": 2.7730360865923954e-07,
604
+ "loss": 0.2014,
605
+ "step": 850
606
+ },
607
+ {
608
+ "epoch": 2.7488,
609
+ "grad_norm": 4.87925788712136,
610
+ "learning_rate": 2.1953427255185122e-07,
611
+ "loss": 0.2449,
612
+ "step": 860
613
+ },
614
+ {
615
+ "epoch": 2.7808,
616
+ "grad_norm": 3.9779689426340785,
617
+ "learning_rate": 1.6837196997130434e-07,
618
+ "loss": 0.2224,
619
+ "step": 870
620
+ },
621
+ {
622
+ "epoch": 2.8128,
623
+ "grad_norm": 2.7382344696771734,
624
+ "learning_rate": 1.2388741182062348e-07,
625
+ "loss": 0.1969,
626
+ "step": 880
627
+ },
628
+ {
629
+ "epoch": 2.8448,
630
+ "grad_norm": 3.6699038732318003,
631
+ "learning_rate": 8.614207975952083e-08,
632
+ "loss": 0.2248,
633
+ "step": 890
634
+ },
635
+ {
636
+ "epoch": 2.8768000000000002,
637
+ "grad_norm": 9.495996724139596,
638
+ "learning_rate": 5.518814123121885e-08,
639
+ "loss": 0.1906,
640
+ "step": 900
641
+ },
642
+ {
643
+ "epoch": 2.9088000000000003,
644
+ "grad_norm": 8.699670035212591,
645
+ "learning_rate": 3.10683773623488e-08,
646
+ "loss": 0.1928,
647
+ "step": 910
648
+ },
649
+ {
650
+ "epoch": 2.9408,
651
+ "grad_norm": 4.133938457429926,
652
+ "learning_rate": 1.3816123835588835e-08,
653
+ "loss": 0.2329,
654
+ "step": 920
655
+ },
656
+ {
657
+ "epoch": 2.9728,
658
+ "grad_norm": 5.0989063447454175,
659
+ "learning_rate": 3.4552248167507576e-09,
660
+ "loss": 0.2185,
661
+ "step": 930
662
+ },
663
+ {
664
+ "epoch": 3.0,
665
+ "step": 939,
666
+ "total_flos": 3418309214208.0,
667
+ "train_loss": 0.6678480519273411,
668
+ "train_runtime": 1116.3574,
669
+ "train_samples_per_second": 13.437,
670
+ "train_steps_per_second": 0.841
671
+ }
672
+ ],
673
+ "logging_steps": 10,
674
+ "max_steps": 939,
675
+ "num_input_tokens_seen": 0,
676
+ "num_train_epochs": 3,
677
+ "save_steps": 500,
678
+ "stateful_callbacks": {
679
+ "TrainerControl": {
680
+ "args": {
681
+ "should_epoch_stop": false,
682
+ "should_evaluate": false,
683
+ "should_log": false,
684
+ "should_save": true,
685
+ "should_training_stop": true
686
+ },
687
+ "attributes": {}
688
+ }
689
+ },
690
+ "total_flos": 3418309214208.0,
691
+ "train_batch_size": 1,
692
+ "trial_name": null,
693
+ "trial_params": null
694
+ }
training_loss.png ADDED