tangken333 commited on
Commit
bbc1184
·
verified ·
1 Parent(s): 078eafc

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 +672 -0
  5. training_loss.png +0 -0
README.md CHANGED
@@ -4,6 +4,7 @@ license: apache-2.0
4
  base_model: Qwen/Qwen2.5-Coder-14B-Instruct
5
  tags:
6
  - llama-factory
 
7
  - generated_from_trainer
8
  model-index:
9
  - name: SWE-BENCH-5k-first-2000-claude-search-replace-generation_qwen_code_14b_5k_first_2000_generation
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
15
 
16
  # SWE-BENCH-5k-first-2000-claude-search-replace-generation_qwen_code_14b_5k_first_2000_generation
17
 
18
- This model is a fine-tuned version of [Qwen/Qwen2.5-Coder-14B-Instruct](https://huggingface.co/Qwen/Qwen2.5-Coder-14B-Instruct) on an unknown dataset.
19
 
20
  ## Model description
21
 
 
4
  base_model: Qwen/Qwen2.5-Coder-14B-Instruct
5
  tags:
6
  - llama-factory
7
+ - full
8
  - generated_from_trainer
9
  model-index:
10
  - name: SWE-BENCH-5k-first-2000-claude-search-replace-generation_qwen_code_14b_5k_first_2000_generation
 
16
 
17
  # SWE-BENCH-5k-first-2000-claude-search-replace-generation_qwen_code_14b_5k_first_2000_generation
18
 
19
+ This model is a fine-tuned version of [Qwen/Qwen2.5-Coder-14B-Instruct](https://huggingface.co/Qwen/Qwen2.5-Coder-14B-Instruct) on the SWE-BENCH-5k-first-2000-claude-search-replace-generation dataset.
20
 
21
  ## Model description
22
 
all_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.9916897506925206,
3
+ "total_flos": 6.786493896982528e+16,
4
+ "train_loss": 0.4722738567325804,
5
+ "train_runtime": 13326.5194,
6
+ "train_samples_per_second": 0.325,
7
+ "train_steps_per_second": 0.007
8
+ }
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.9916897506925206,
3
+ "total_flos": 6.786493896982528e+16,
4
+ "train_loss": 0.4722738567325804,
5
+ "train_runtime": 13326.5194,
6
+ "train_samples_per_second": 0.325,
7
+ "train_steps_per_second": 0.007
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,672 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 2.9916897506925206,
5
+ "eval_steps": 500,
6
+ "global_step": 90,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.0332409972299169,
13
+ "grad_norm": 1.3523818254470825,
14
+ "learning_rate": 1.111111111111111e-06,
15
+ "loss": 0.6805,
16
+ "step": 1
17
+ },
18
+ {
19
+ "epoch": 0.0664819944598338,
20
+ "grad_norm": 1.467878818511963,
21
+ "learning_rate": 2.222222222222222e-06,
22
+ "loss": 0.7126,
23
+ "step": 2
24
+ },
25
+ {
26
+ "epoch": 0.0997229916897507,
27
+ "grad_norm": 1.3939130306243896,
28
+ "learning_rate": 3.3333333333333333e-06,
29
+ "loss": 0.6845,
30
+ "step": 3
31
+ },
32
+ {
33
+ "epoch": 0.1329639889196676,
34
+ "grad_norm": 1.1881986856460571,
35
+ "learning_rate": 4.444444444444444e-06,
36
+ "loss": 0.6266,
37
+ "step": 4
38
+ },
39
+ {
40
+ "epoch": 0.16620498614958448,
41
+ "grad_norm": 1.1864044666290283,
42
+ "learning_rate": 5.555555555555557e-06,
43
+ "loss": 0.6882,
44
+ "step": 5
45
+ },
46
+ {
47
+ "epoch": 0.1994459833795014,
48
+ "grad_norm": 0.8443530201911926,
49
+ "learning_rate": 6.666666666666667e-06,
50
+ "loss": 0.6415,
51
+ "step": 6
52
+ },
53
+ {
54
+ "epoch": 0.23268698060941828,
55
+ "grad_norm": 0.739067018032074,
56
+ "learning_rate": 7.77777777777778e-06,
57
+ "loss": 0.6089,
58
+ "step": 7
59
+ },
60
+ {
61
+ "epoch": 0.2659279778393352,
62
+ "grad_norm": 0.8895483613014221,
63
+ "learning_rate": 8.888888888888888e-06,
64
+ "loss": 0.6387,
65
+ "step": 8
66
+ },
67
+ {
68
+ "epoch": 0.29916897506925205,
69
+ "grad_norm": 0.7306845188140869,
70
+ "learning_rate": 1e-05,
71
+ "loss": 0.6311,
72
+ "step": 9
73
+ },
74
+ {
75
+ "epoch": 0.33240997229916897,
76
+ "grad_norm": 0.8551097512245178,
77
+ "learning_rate": 9.996239762521152e-06,
78
+ "loss": 0.5691,
79
+ "step": 10
80
+ },
81
+ {
82
+ "epoch": 0.3656509695290859,
83
+ "grad_norm": 0.7852718234062195,
84
+ "learning_rate": 9.98496470583896e-06,
85
+ "loss": 0.5585,
86
+ "step": 11
87
+ },
88
+ {
89
+ "epoch": 0.3988919667590028,
90
+ "grad_norm": 0.6744587421417236,
91
+ "learning_rate": 9.966191788709716e-06,
92
+ "loss": 0.5555,
93
+ "step": 12
94
+ },
95
+ {
96
+ "epoch": 0.43213296398891965,
97
+ "grad_norm": 0.5322245359420776,
98
+ "learning_rate": 9.939949247384046e-06,
99
+ "loss": 0.5708,
100
+ "step": 13
101
+ },
102
+ {
103
+ "epoch": 0.46537396121883656,
104
+ "grad_norm": 0.634617805480957,
105
+ "learning_rate": 9.906276553136924e-06,
106
+ "loss": 0.5744,
107
+ "step": 14
108
+ },
109
+ {
110
+ "epoch": 0.4986149584487535,
111
+ "grad_norm": 0.5223730206489563,
112
+ "learning_rate": 9.86522435289912e-06,
113
+ "loss": 0.5239,
114
+ "step": 15
115
+ },
116
+ {
117
+ "epoch": 0.5318559556786704,
118
+ "grad_norm": 0.5203894972801208,
119
+ "learning_rate": 9.816854393079402e-06,
120
+ "loss": 0.5386,
121
+ "step": 16
122
+ },
123
+ {
124
+ "epoch": 0.5650969529085873,
125
+ "grad_norm": 0.4948791563510895,
126
+ "learning_rate": 9.761239426692077e-06,
127
+ "loss": 0.5192,
128
+ "step": 17
129
+ },
130
+ {
131
+ "epoch": 0.5983379501385041,
132
+ "grad_norm": 0.47360560297966003,
133
+ "learning_rate": 9.698463103929542e-06,
134
+ "loss": 0.5328,
135
+ "step": 18
136
+ },
137
+ {
138
+ "epoch": 0.631578947368421,
139
+ "grad_norm": 0.47556865215301514,
140
+ "learning_rate": 9.628619846344453e-06,
141
+ "loss": 0.5252,
142
+ "step": 19
143
+ },
144
+ {
145
+ "epoch": 0.6648199445983379,
146
+ "grad_norm": 0.46637672185897827,
147
+ "learning_rate": 9.551814704830734e-06,
148
+ "loss": 0.5394,
149
+ "step": 20
150
+ },
151
+ {
152
+ "epoch": 0.6980609418282548,
153
+ "grad_norm": 0.408130407333374,
154
+ "learning_rate": 9.468163201617063e-06,
155
+ "loss": 0.4791,
156
+ "step": 21
157
+ },
158
+ {
159
+ "epoch": 0.7313019390581718,
160
+ "grad_norm": 0.4232206642627716,
161
+ "learning_rate": 9.377791156510456e-06,
162
+ "loss": 0.5604,
163
+ "step": 22
164
+ },
165
+ {
166
+ "epoch": 0.7645429362880887,
167
+ "grad_norm": 0.4065282642841339,
168
+ "learning_rate": 9.280834497651334e-06,
169
+ "loss": 0.5219,
170
+ "step": 23
171
+ },
172
+ {
173
+ "epoch": 0.7977839335180056,
174
+ "grad_norm": 0.42849838733673096,
175
+ "learning_rate": 9.177439057064684e-06,
176
+ "loss": 0.5247,
177
+ "step": 24
178
+ },
179
+ {
180
+ "epoch": 0.8310249307479224,
181
+ "grad_norm": 0.4445164203643799,
182
+ "learning_rate": 9.067760351314838e-06,
183
+ "loss": 0.5421,
184
+ "step": 25
185
+ },
186
+ {
187
+ "epoch": 0.8642659279778393,
188
+ "grad_norm": 0.4322611689567566,
189
+ "learning_rate": 8.951963347593797e-06,
190
+ "loss": 0.5349,
191
+ "step": 26
192
+ },
193
+ {
194
+ "epoch": 0.8975069252077562,
195
+ "grad_norm": 0.42110881209373474,
196
+ "learning_rate": 8.83022221559489e-06,
197
+ "loss": 0.5179,
198
+ "step": 27
199
+ },
200
+ {
201
+ "epoch": 0.9307479224376731,
202
+ "grad_norm": 0.44587212800979614,
203
+ "learning_rate": 8.702720065545024e-06,
204
+ "loss": 0.5687,
205
+ "step": 28
206
+ },
207
+ {
208
+ "epoch": 0.96398891966759,
209
+ "grad_norm": 0.40205827355384827,
210
+ "learning_rate": 8.569648672789496e-06,
211
+ "loss": 0.5441,
212
+ "step": 29
213
+ },
214
+ {
215
+ "epoch": 0.997229916897507,
216
+ "grad_norm": 0.4229365885257721,
217
+ "learning_rate": 8.43120818934367e-06,
218
+ "loss": 0.5823,
219
+ "step": 30
220
+ },
221
+ {
222
+ "epoch": 1.0304709141274238,
223
+ "grad_norm": 1.2778958082199097,
224
+ "learning_rate": 8.28760684284532e-06,
225
+ "loss": 0.8967,
226
+ "step": 31
227
+ },
228
+ {
229
+ "epoch": 1.0637119113573408,
230
+ "grad_norm": 0.39860862493515015,
231
+ "learning_rate": 8.139060623360494e-06,
232
+ "loss": 0.4919,
233
+ "step": 32
234
+ },
235
+ {
236
+ "epoch": 1.0969529085872576,
237
+ "grad_norm": 0.3812428116798401,
238
+ "learning_rate": 7.985792958513932e-06,
239
+ "loss": 0.4475,
240
+ "step": 33
241
+ },
242
+ {
243
+ "epoch": 1.1301939058171746,
244
+ "grad_norm": 0.38544097542762756,
245
+ "learning_rate": 7.828034377432694e-06,
246
+ "loss": 0.4702,
247
+ "step": 34
248
+ },
249
+ {
250
+ "epoch": 1.1634349030470914,
251
+ "grad_norm": 0.38543078303337097,
252
+ "learning_rate": 7.666022164008458e-06,
253
+ "loss": 0.4492,
254
+ "step": 35
255
+ },
256
+ {
257
+ "epoch": 1.1966759002770084,
258
+ "grad_norm": 0.36279022693634033,
259
+ "learning_rate": 7.500000000000001e-06,
260
+ "loss": 0.4394,
261
+ "step": 36
262
+ },
263
+ {
264
+ "epoch": 1.2299168975069252,
265
+ "grad_norm": 0.4173636734485626,
266
+ "learning_rate": 7.330217598512696e-06,
267
+ "loss": 0.4573,
268
+ "step": 37
269
+ },
270
+ {
271
+ "epoch": 1.263157894736842,
272
+ "grad_norm": 0.3636619448661804,
273
+ "learning_rate": 7.156930328406268e-06,
274
+ "loss": 0.4374,
275
+ "step": 38
276
+ },
277
+ {
278
+ "epoch": 1.296398891966759,
279
+ "grad_norm": 0.44060707092285156,
280
+ "learning_rate": 6.980398830195785e-06,
281
+ "loss": 0.4401,
282
+ "step": 39
283
+ },
284
+ {
285
+ "epoch": 1.3296398891966759,
286
+ "grad_norm": 0.33170637488365173,
287
+ "learning_rate": 6.800888624023552e-06,
288
+ "loss": 0.3893,
289
+ "step": 40
290
+ },
291
+ {
292
+ "epoch": 1.3628808864265927,
293
+ "grad_norm": 0.3763296902179718,
294
+ "learning_rate": 6.618669710291607e-06,
295
+ "loss": 0.4629,
296
+ "step": 41
297
+ },
298
+ {
299
+ "epoch": 1.3961218836565097,
300
+ "grad_norm": 0.4101507365703583,
301
+ "learning_rate": 6.434016163555452e-06,
302
+ "loss": 0.4972,
303
+ "step": 42
304
+ },
305
+ {
306
+ "epoch": 1.4293628808864267,
307
+ "grad_norm": 0.38400354981422424,
308
+ "learning_rate": 6.247205720289907e-06,
309
+ "loss": 0.4663,
310
+ "step": 43
311
+ },
312
+ {
313
+ "epoch": 1.4626038781163435,
314
+ "grad_norm": 0.39262333512306213,
315
+ "learning_rate": 6.058519361147055e-06,
316
+ "loss": 0.4388,
317
+ "step": 44
318
+ },
319
+ {
320
+ "epoch": 1.4958448753462603,
321
+ "grad_norm": 0.3506338894367218,
322
+ "learning_rate": 5.8682408883346535e-06,
323
+ "loss": 0.3527,
324
+ "step": 45
325
+ },
326
+ {
327
+ "epoch": 1.5290858725761773,
328
+ "grad_norm": 0.35712841153144836,
329
+ "learning_rate": 5.6766564987506564e-06,
330
+ "loss": 0.4269,
331
+ "step": 46
332
+ },
333
+ {
334
+ "epoch": 1.5623268698060941,
335
+ "grad_norm": 0.35582512617111206,
336
+ "learning_rate": 5.484054353515896e-06,
337
+ "loss": 0.3959,
338
+ "step": 47
339
+ },
340
+ {
341
+ "epoch": 1.595567867036011,
342
+ "grad_norm": 0.3706922233104706,
343
+ "learning_rate": 5.290724144552379e-06,
344
+ "loss": 0.4267,
345
+ "step": 48
346
+ },
347
+ {
348
+ "epoch": 1.628808864265928,
349
+ "grad_norm": 0.36104509234428406,
350
+ "learning_rate": 5.096956658859122e-06,
351
+ "loss": 0.4238,
352
+ "step": 49
353
+ },
354
+ {
355
+ "epoch": 1.662049861495845,
356
+ "grad_norm": 0.38358086347579956,
357
+ "learning_rate": 4.903043341140879e-06,
358
+ "loss": 0.4498,
359
+ "step": 50
360
+ },
361
+ {
362
+ "epoch": 1.6952908587257618,
363
+ "grad_norm": 0.32370346784591675,
364
+ "learning_rate": 4.7092758554476215e-06,
365
+ "loss": 0.379,
366
+ "step": 51
367
+ },
368
+ {
369
+ "epoch": 1.7285318559556786,
370
+ "grad_norm": 0.390703946352005,
371
+ "learning_rate": 4.515945646484105e-06,
372
+ "loss": 0.4589,
373
+ "step": 52
374
+ },
375
+ {
376
+ "epoch": 1.7617728531855956,
377
+ "grad_norm": 0.35574859380722046,
378
+ "learning_rate": 4.323343501249346e-06,
379
+ "loss": 0.4284,
380
+ "step": 53
381
+ },
382
+ {
383
+ "epoch": 1.7950138504155124,
384
+ "grad_norm": 0.3821960985660553,
385
+ "learning_rate": 4.131759111665349e-06,
386
+ "loss": 0.4464,
387
+ "step": 54
388
+ },
389
+ {
390
+ "epoch": 1.8282548476454292,
391
+ "grad_norm": 0.3628939688205719,
392
+ "learning_rate": 3.941480638852948e-06,
393
+ "loss": 0.4127,
394
+ "step": 55
395
+ },
396
+ {
397
+ "epoch": 1.8614958448753463,
398
+ "grad_norm": 0.3751230537891388,
399
+ "learning_rate": 3.752794279710094e-06,
400
+ "loss": 0.4308,
401
+ "step": 56
402
+ },
403
+ {
404
+ "epoch": 1.8947368421052633,
405
+ "grad_norm": 0.36090031266212463,
406
+ "learning_rate": 3.5659838364445505e-06,
407
+ "loss": 0.426,
408
+ "step": 57
409
+ },
410
+ {
411
+ "epoch": 1.92797783933518,
412
+ "grad_norm": 0.37130820751190186,
413
+ "learning_rate": 3.3813302897083955e-06,
414
+ "loss": 0.4299,
415
+ "step": 58
416
+ },
417
+ {
418
+ "epoch": 1.9612188365650969,
419
+ "grad_norm": 0.3735775053501129,
420
+ "learning_rate": 3.1991113759764493e-06,
421
+ "loss": 0.4212,
422
+ "step": 59
423
+ },
424
+ {
425
+ "epoch": 1.994459833795014,
426
+ "grad_norm": 0.37526699900627136,
427
+ "learning_rate": 3.019601169804216e-06,
428
+ "loss": 0.4594,
429
+ "step": 60
430
+ },
431
+ {
432
+ "epoch": 2.027700831024931,
433
+ "grad_norm": 1.2472105026245117,
434
+ "learning_rate": 2.843069671593734e-06,
435
+ "loss": 0.6625,
436
+ "step": 61
437
+ },
438
+ {
439
+ "epoch": 2.0609418282548475,
440
+ "grad_norm": 0.5256877541542053,
441
+ "learning_rate": 2.6697824014873076e-06,
442
+ "loss": 0.3823,
443
+ "step": 62
444
+ },
445
+ {
446
+ "epoch": 2.0941828254847645,
447
+ "grad_norm": 0.35856226086616516,
448
+ "learning_rate": 2.5000000000000015e-06,
449
+ "loss": 0.3802,
450
+ "step": 63
451
+ },
452
+ {
453
+ "epoch": 2.1274238227146816,
454
+ "grad_norm": 0.34101545810699463,
455
+ "learning_rate": 2.333977835991545e-06,
456
+ "loss": 0.3445,
457
+ "step": 64
458
+ },
459
+ {
460
+ "epoch": 2.160664819944598,
461
+ "grad_norm": 0.3762127757072449,
462
+ "learning_rate": 2.171965622567308e-06,
463
+ "loss": 0.3903,
464
+ "step": 65
465
+ },
466
+ {
467
+ "epoch": 2.193905817174515,
468
+ "grad_norm": 0.36153438687324524,
469
+ "learning_rate": 2.0142070414860704e-06,
470
+ "loss": 0.3968,
471
+ "step": 66
472
+ },
473
+ {
474
+ "epoch": 2.227146814404432,
475
+ "grad_norm": 0.3704725503921509,
476
+ "learning_rate": 1.8609393766395083e-06,
477
+ "loss": 0.4023,
478
+ "step": 67
479
+ },
480
+ {
481
+ "epoch": 2.260387811634349,
482
+ "grad_norm": 0.345738023519516,
483
+ "learning_rate": 1.7123931571546826e-06,
484
+ "loss": 0.3728,
485
+ "step": 68
486
+ },
487
+ {
488
+ "epoch": 2.293628808864266,
489
+ "grad_norm": 0.35163000226020813,
490
+ "learning_rate": 1.5687918106563326e-06,
491
+ "loss": 0.3588,
492
+ "step": 69
493
+ },
494
+ {
495
+ "epoch": 2.326869806094183,
496
+ "grad_norm": 0.3672683537006378,
497
+ "learning_rate": 1.4303513272105057e-06,
498
+ "loss": 0.4005,
499
+ "step": 70
500
+ },
501
+ {
502
+ "epoch": 2.3601108033241,
503
+ "grad_norm": 0.33104199171066284,
504
+ "learning_rate": 1.297279934454978e-06,
505
+ "loss": 0.3324,
506
+ "step": 71
507
+ },
508
+ {
509
+ "epoch": 2.393351800554017,
510
+ "grad_norm": 0.35979458689689636,
511
+ "learning_rate": 1.1697777844051105e-06,
512
+ "loss": 0.3939,
513
+ "step": 72
514
+ },
515
+ {
516
+ "epoch": 2.4265927977839334,
517
+ "grad_norm": 0.34565168619155884,
518
+ "learning_rate": 1.0480366524062041e-06,
519
+ "loss": 0.3498,
520
+ "step": 73
521
+ },
522
+ {
523
+ "epoch": 2.4598337950138505,
524
+ "grad_norm": 0.3453231453895569,
525
+ "learning_rate": 9.322396486851626e-07,
526
+ "loss": 0.3846,
527
+ "step": 74
528
+ },
529
+ {
530
+ "epoch": 2.4930747922437675,
531
+ "grad_norm": 0.3575272560119629,
532
+ "learning_rate": 8.225609429353187e-07,
533
+ "loss": 0.3663,
534
+ "step": 75
535
+ },
536
+ {
537
+ "epoch": 2.526315789473684,
538
+ "grad_norm": 0.3252071738243103,
539
+ "learning_rate": 7.191655023486682e-07,
540
+ "loss": 0.3424,
541
+ "step": 76
542
+ },
543
+ {
544
+ "epoch": 2.559556786703601,
545
+ "grad_norm": 0.36309072375297546,
546
+ "learning_rate": 6.222088434895462e-07,
547
+ "loss": 0.372,
548
+ "step": 77
549
+ },
550
+ {
551
+ "epoch": 2.592797783933518,
552
+ "grad_norm": 0.3643971383571625,
553
+ "learning_rate": 5.318367983829393e-07,
554
+ "loss": 0.3895,
555
+ "step": 78
556
+ },
557
+ {
558
+ "epoch": 2.6260387811634347,
559
+ "grad_norm": 0.33617454767227173,
560
+ "learning_rate": 4.481852951692672e-07,
561
+ "loss": 0.3742,
562
+ "step": 79
563
+ },
564
+ {
565
+ "epoch": 2.6592797783933517,
566
+ "grad_norm": 0.3941975235939026,
567
+ "learning_rate": 3.7138015365554834e-07,
568
+ "loss": 0.4365,
569
+ "step": 80
570
+ },
571
+ {
572
+ "epoch": 2.6925207756232687,
573
+ "grad_norm": 0.335756778717041,
574
+ "learning_rate": 3.015368960704584e-07,
575
+ "loss": 0.3664,
576
+ "step": 81
577
+ },
578
+ {
579
+ "epoch": 2.7257617728531853,
580
+ "grad_norm": 0.3415193557739258,
581
+ "learning_rate": 2.3876057330792344e-07,
582
+ "loss": 0.3506,
583
+ "step": 82
584
+ },
585
+ {
586
+ "epoch": 2.7590027700831024,
587
+ "grad_norm": 0.3626050651073456,
588
+ "learning_rate": 1.8314560692059836e-07,
589
+ "loss": 0.3794,
590
+ "step": 83
591
+ },
592
+ {
593
+ "epoch": 2.7922437673130194,
594
+ "grad_norm": 0.3612777590751648,
595
+ "learning_rate": 1.3477564710088097e-07,
596
+ "loss": 0.3675,
597
+ "step": 84
598
+ },
599
+ {
600
+ "epoch": 2.8254847645429364,
601
+ "grad_norm": 0.38328835368156433,
602
+ "learning_rate": 9.372344686307655e-08,
603
+ "loss": 0.4047,
604
+ "step": 85
605
+ },
606
+ {
607
+ "epoch": 2.8587257617728534,
608
+ "grad_norm": 0.36719197034835815,
609
+ "learning_rate": 6.005075261595495e-08,
610
+ "loss": 0.3681,
611
+ "step": 86
612
+ },
613
+ {
614
+ "epoch": 2.89196675900277,
615
+ "grad_norm": 0.3810074031352997,
616
+ "learning_rate": 3.3808211290284886e-08,
617
+ "loss": 0.4046,
618
+ "step": 87
619
+ },
620
+ {
621
+ "epoch": 2.925207756232687,
622
+ "grad_norm": 0.36970335245132446,
623
+ "learning_rate": 1.5035294161039882e-08,
624
+ "loss": 0.3916,
625
+ "step": 88
626
+ },
627
+ {
628
+ "epoch": 2.958448753462604,
629
+ "grad_norm": 0.4210459291934967,
630
+ "learning_rate": 3.760237478849793e-09,
631
+ "loss": 0.4253,
632
+ "step": 89
633
+ },
634
+ {
635
+ "epoch": 2.9916897506925206,
636
+ "grad_norm": 0.3731752634048462,
637
+ "learning_rate": 0.0,
638
+ "loss": 0.3648,
639
+ "step": 90
640
+ },
641
+ {
642
+ "epoch": 2.9916897506925206,
643
+ "step": 90,
644
+ "total_flos": 6.786493896982528e+16,
645
+ "train_loss": 0.4722738567325804,
646
+ "train_runtime": 13326.5194,
647
+ "train_samples_per_second": 0.325,
648
+ "train_steps_per_second": 0.007
649
+ }
650
+ ],
651
+ "logging_steps": 1,
652
+ "max_steps": 90,
653
+ "num_input_tokens_seen": 0,
654
+ "num_train_epochs": 3,
655
+ "save_steps": 200,
656
+ "stateful_callbacks": {
657
+ "TrainerControl": {
658
+ "args": {
659
+ "should_epoch_stop": false,
660
+ "should_evaluate": false,
661
+ "should_log": false,
662
+ "should_save": true,
663
+ "should_training_stop": true
664
+ },
665
+ "attributes": {}
666
+ }
667
+ },
668
+ "total_flos": 6.786493896982528e+16,
669
+ "train_batch_size": 1,
670
+ "trial_name": null,
671
+ "trial_params": null
672
+ }
training_loss.png ADDED