chancharikm commited on
Commit
dff4ca7
·
verified ·
1 Parent(s): 5880df8

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 +882 -0
  5. training_loss.png +0 -0
README.md CHANGED
@@ -4,6 +4,7 @@ license: apache-2.0
4
  base_model: chancharikm/qwen2.5_saves_7b_1_3_6
5
  tags:
6
  - llama-factory
 
7
  - generated_from_trainer
8
  model-index:
9
  - name: saves_7b_1_3_6_final
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
15
 
16
  # saves_7b_1_3_6_final
17
 
18
- This model is a fine-tuned version of [chancharikm/qwen2.5_saves_7b_1_3_6](https://huggingface.co/chancharikm/qwen2.5_saves_7b_1_3_6) on the None dataset.
19
 
20
  ## Model description
21
 
 
4
  base_model: chancharikm/qwen2.5_saves_7b_1_3_6
5
  tags:
6
  - llama-factory
7
+ - full
8
  - generated_from_trainer
9
  model-index:
10
  - name: saves_7b_1_3_6_final
 
16
 
17
  # saves_7b_1_3_6_final
18
 
19
+ This model is a fine-tuned version of [chancharikm/qwen2.5_saves_7b_1_3_6](https://huggingface.co/chancharikm/qwen2.5_saves_7b_1_3_6) on the cam_motion_balanced_vqa_ORCHARD, the cam_motion_caption_ORCHARD and the cam_motion_imb_raw_ORCHARD datasets.
20
 
21
  ## Model description
22
 
all_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 1.0,
3
+ "total_flos": 901340604923904.0,
4
+ "train_loss": 0.0009180365409025791,
5
+ "train_runtime": 38806.1809,
6
+ "train_samples_per_second": 5.96,
7
+ "train_steps_per_second": 0.031
8
+ }
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 1.0,
3
+ "total_flos": 901340604923904.0,
4
+ "train_loss": 0.0009180365409025791,
5
+ "train_runtime": 38806.1809,
6
+ "train_samples_per_second": 5.96,
7
+ "train_steps_per_second": 0.031
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,882 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 1.0,
5
+ "eval_steps": 500,
6
+ "global_step": 1205,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.008298755186721992,
13
+ "grad_norm": 0.01952634189028208,
14
+ "learning_rate": 8.196721311475409e-07,
15
+ "loss": 0.0002,
16
+ "step": 10
17
+ },
18
+ {
19
+ "epoch": 0.016597510373443983,
20
+ "grad_norm": 0.2545911846047496,
21
+ "learning_rate": 1.6393442622950819e-06,
22
+ "loss": 0.0004,
23
+ "step": 20
24
+ },
25
+ {
26
+ "epoch": 0.024896265560165973,
27
+ "grad_norm": 0.013501883610251985,
28
+ "learning_rate": 2.459016393442623e-06,
29
+ "loss": 0.0002,
30
+ "step": 30
31
+ },
32
+ {
33
+ "epoch": 0.03319502074688797,
34
+ "grad_norm": 0.01295761896054431,
35
+ "learning_rate": 3.2786885245901638e-06,
36
+ "loss": 0.0004,
37
+ "step": 40
38
+ },
39
+ {
40
+ "epoch": 0.04149377593360996,
41
+ "grad_norm": 0.3035265968896415,
42
+ "learning_rate": 4.098360655737705e-06,
43
+ "loss": 0.0003,
44
+ "step": 50
45
+ },
46
+ {
47
+ "epoch": 0.04979253112033195,
48
+ "grad_norm": 0.0017646037422956603,
49
+ "learning_rate": 4.918032786885246e-06,
50
+ "loss": 0.0012,
51
+ "step": 60
52
+ },
53
+ {
54
+ "epoch": 0.058091286307053944,
55
+ "grad_norm": 0.017035062957617197,
56
+ "learning_rate": 4.999236479642602e-06,
57
+ "loss": 0.0025,
58
+ "step": 70
59
+ },
60
+ {
61
+ "epoch": 0.06639004149377593,
62
+ "grad_norm": 0.004209789578646451,
63
+ "learning_rate": 4.996597748756039e-06,
64
+ "loss": 0.0003,
65
+ "step": 80
66
+ },
67
+ {
68
+ "epoch": 0.07468879668049792,
69
+ "grad_norm": 0.016889199255074427,
70
+ "learning_rate": 4.992076370516985e-06,
71
+ "loss": 0.0029,
72
+ "step": 90
73
+ },
74
+ {
75
+ "epoch": 0.08298755186721991,
76
+ "grad_norm": 0.49972929296558655,
77
+ "learning_rate": 4.985675754429744e-06,
78
+ "loss": 0.0012,
79
+ "step": 100
80
+ },
81
+ {
82
+ "epoch": 0.0912863070539419,
83
+ "grad_norm": 0.05679961395868942,
84
+ "learning_rate": 4.977400727104024e-06,
85
+ "loss": 0.0013,
86
+ "step": 110
87
+ },
88
+ {
89
+ "epoch": 0.0995850622406639,
90
+ "grad_norm": 0.3163904422156924,
91
+ "learning_rate": 4.967257528615261e-06,
92
+ "loss": 0.001,
93
+ "step": 120
94
+ },
95
+ {
96
+ "epoch": 0.1078838174273859,
97
+ "grad_norm": 0.02845684920996449,
98
+ "learning_rate": 4.955253807799073e-06,
99
+ "loss": 0.0059,
100
+ "step": 130
101
+ },
102
+ {
103
+ "epoch": 0.11618257261410789,
104
+ "grad_norm": 0.1724673205403095,
105
+ "learning_rate": 4.941398616483382e-06,
106
+ "loss": 0.0016,
107
+ "step": 140
108
+ },
109
+ {
110
+ "epoch": 0.12448132780082988,
111
+ "grad_norm": 0.0447393567131259,
112
+ "learning_rate": 4.925702402662568e-06,
113
+ "loss": 0.001,
114
+ "step": 150
115
+ },
116
+ {
117
+ "epoch": 0.13278008298755187,
118
+ "grad_norm": 0.10461493774492425,
119
+ "learning_rate": 4.9081770026187915e-06,
120
+ "loss": 0.0012,
121
+ "step": 160
122
+ },
123
+ {
124
+ "epoch": 0.14107883817427386,
125
+ "grad_norm": 0.1340115340660757,
126
+ "learning_rate": 4.888835631996426e-06,
127
+ "loss": 0.0008,
128
+ "step": 170
129
+ },
130
+ {
131
+ "epoch": 0.14937759336099585,
132
+ "grad_norm": 0.025774190164302806,
133
+ "learning_rate": 4.86769287583634e-06,
134
+ "loss": 0.0018,
135
+ "step": 180
136
+ },
137
+ {
138
+ "epoch": 0.15767634854771784,
139
+ "grad_norm": 0.6750469917729486,
140
+ "learning_rate": 4.844764677577531e-06,
141
+ "loss": 0.0024,
142
+ "step": 190
143
+ },
144
+ {
145
+ "epoch": 0.16597510373443983,
146
+ "grad_norm": 0.5121699739264448,
147
+ "learning_rate": 4.820068327034416e-06,
148
+ "loss": 0.0025,
149
+ "step": 200
150
+ },
151
+ {
152
+ "epoch": 0.17427385892116182,
153
+ "grad_norm": 0.005162235007018351,
154
+ "learning_rate": 4.793622447358837e-06,
155
+ "loss": 0.0013,
156
+ "step": 210
157
+ },
158
+ {
159
+ "epoch": 0.1825726141078838,
160
+ "grad_norm": 0.0065204757813555,
161
+ "learning_rate": 4.765446980996622e-06,
162
+ "loss": 0.0025,
163
+ "step": 220
164
+ },
165
+ {
166
+ "epoch": 0.1908713692946058,
167
+ "grad_norm": 0.1834554236512134,
168
+ "learning_rate": 4.735563174649278e-06,
169
+ "loss": 0.0015,
170
+ "step": 230
171
+ },
172
+ {
173
+ "epoch": 0.1991701244813278,
174
+ "grad_norm": 1.0765680707456238,
175
+ "learning_rate": 4.703993563252172e-06,
176
+ "loss": 0.0034,
177
+ "step": 240
178
+ },
179
+ {
180
+ "epoch": 0.2074688796680498,
181
+ "grad_norm": 0.34800167205451604,
182
+ "learning_rate": 4.670761952981264e-06,
183
+ "loss": 0.0016,
184
+ "step": 250
185
+ },
186
+ {
187
+ "epoch": 0.2157676348547718,
188
+ "grad_norm": 0.12237972347896162,
189
+ "learning_rate": 4.635893403301233e-06,
190
+ "loss": 0.0014,
191
+ "step": 260
192
+ },
193
+ {
194
+ "epoch": 0.22406639004149378,
195
+ "grad_norm": 0.2737166694487476,
196
+ "learning_rate": 4.5994142080684956e-06,
197
+ "loss": 0.0025,
198
+ "step": 270
199
+ },
200
+ {
201
+ "epoch": 0.23236514522821577,
202
+ "grad_norm": 0.04068916670988713,
203
+ "learning_rate": 4.561351875703406e-06,
204
+ "loss": 0.0014,
205
+ "step": 280
206
+ },
207
+ {
208
+ "epoch": 0.24066390041493776,
209
+ "grad_norm": 0.3245341216908413,
210
+ "learning_rate": 4.52173510844656e-06,
211
+ "loss": 0.0012,
212
+ "step": 290
213
+ },
214
+ {
215
+ "epoch": 0.24896265560165975,
216
+ "grad_norm": 0.03783916464770265,
217
+ "learning_rate": 4.480593780714863e-06,
218
+ "loss": 0.0026,
219
+ "step": 300
220
+ },
221
+ {
222
+ "epoch": 0.2572614107883817,
223
+ "grad_norm": 0.4859003212418197,
224
+ "learning_rate": 4.4379589165736736e-06,
225
+ "loss": 0.0035,
226
+ "step": 310
227
+ },
228
+ {
229
+ "epoch": 0.26556016597510373,
230
+ "grad_norm": 0.43809699541809793,
231
+ "learning_rate": 4.393862666342017e-06,
232
+ "loss": 0.0006,
233
+ "step": 320
234
+ },
235
+ {
236
+ "epoch": 0.27385892116182575,
237
+ "grad_norm": 0.18772242388214733,
238
+ "learning_rate": 4.348338282348504e-06,
239
+ "loss": 0.0023,
240
+ "step": 330
241
+ },
242
+ {
243
+ "epoch": 0.2821576763485477,
244
+ "grad_norm": 0.0908936575905787,
245
+ "learning_rate": 4.301420093856247e-06,
246
+ "loss": 0.0011,
247
+ "step": 340
248
+ },
249
+ {
250
+ "epoch": 0.29045643153526973,
251
+ "grad_norm": 0.4647945053304386,
252
+ "learning_rate": 4.2531434811756675e-06,
253
+ "loss": 0.0008,
254
+ "step": 350
255
+ },
256
+ {
257
+ "epoch": 0.2987551867219917,
258
+ "grad_norm": 0.009093547777615585,
259
+ "learning_rate": 4.203544848984729e-06,
260
+ "loss": 0.0024,
261
+ "step": 360
262
+ },
263
+ {
264
+ "epoch": 0.3070539419087137,
265
+ "grad_norm": 0.06847326021423142,
266
+ "learning_rate": 4.15266159887671e-06,
267
+ "loss": 0.002,
268
+ "step": 370
269
+ },
270
+ {
271
+ "epoch": 0.3153526970954357,
272
+ "grad_norm": 0.09182889879413814,
273
+ "learning_rate": 4.10053210115622e-06,
274
+ "loss": 0.0016,
275
+ "step": 380
276
+ },
277
+ {
278
+ "epoch": 0.3236514522821577,
279
+ "grad_norm": 0.13339020629841652,
280
+ "learning_rate": 4.0471956659047166e-06,
281
+ "loss": 0.001,
282
+ "step": 390
283
+ },
284
+ {
285
+ "epoch": 0.33195020746887965,
286
+ "grad_norm": 0.008092316624748167,
287
+ "learning_rate": 3.9926925133373565e-06,
288
+ "loss": 0.0023,
289
+ "step": 400
290
+ },
291
+ {
292
+ "epoch": 0.34024896265560167,
293
+ "grad_norm": 0.023997101221404513,
294
+ "learning_rate": 3.93706374347353e-06,
295
+ "loss": 0.0007,
296
+ "step": 410
297
+ },
298
+ {
299
+ "epoch": 0.34854771784232363,
300
+ "grad_norm": 0.02702823530323454,
301
+ "learning_rate": 3.880351305143945e-06,
302
+ "loss": 0.0016,
303
+ "step": 420
304
+ },
305
+ {
306
+ "epoch": 0.35684647302904565,
307
+ "grad_norm": 0.14873455961870954,
308
+ "learning_rate": 3.822597964357632e-06,
309
+ "loss": 0.0014,
310
+ "step": 430
311
+ },
312
+ {
313
+ "epoch": 0.3651452282157676,
314
+ "grad_norm": 0.030315999061406942,
315
+ "learning_rate": 3.76384727205274e-06,
316
+ "loss": 0.0012,
317
+ "step": 440
318
+ },
319
+ {
320
+ "epoch": 0.37344398340248963,
321
+ "grad_norm": 0.011605294695610434,
322
+ "learning_rate": 3.7041435312554053e-06,
323
+ "loss": 0.0004,
324
+ "step": 450
325
+ },
326
+ {
327
+ "epoch": 0.3817427385892116,
328
+ "grad_norm": 0.009516457718031966,
329
+ "learning_rate": 3.6435317636715104e-06,
330
+ "loss": 0.001,
331
+ "step": 460
332
+ },
333
+ {
334
+ "epoch": 0.3900414937759336,
335
+ "grad_norm": 0.24625371958398262,
336
+ "learning_rate": 3.5820576757364743e-06,
337
+ "loss": 0.0005,
338
+ "step": 470
339
+ },
340
+ {
341
+ "epoch": 0.3983402489626556,
342
+ "grad_norm": 0.011967039603122181,
343
+ "learning_rate": 3.51976762414871e-06,
344
+ "loss": 0.0016,
345
+ "step": 480
346
+ },
347
+ {
348
+ "epoch": 0.4066390041493776,
349
+ "grad_norm": 0.03929185536463483,
350
+ "learning_rate": 3.4567085809127247e-06,
351
+ "loss": 0.0009,
352
+ "step": 490
353
+ },
354
+ {
355
+ "epoch": 0.4149377593360996,
356
+ "grad_norm": 0.1491795016675303,
357
+ "learning_rate": 3.392928097918229e-06,
358
+ "loss": 0.0011,
359
+ "step": 500
360
+ },
361
+ {
362
+ "epoch": 0.42323651452282157,
363
+ "grad_norm": 0.030419957346452066,
364
+ "learning_rate": 3.328474271081959e-06,
365
+ "loss": 0.0007,
366
+ "step": 510
367
+ },
368
+ {
369
+ "epoch": 0.4315352697095436,
370
+ "grad_norm": 0.011487360993853078,
371
+ "learning_rate": 3.2633957040792624e-06,
372
+ "loss": 0.0027,
373
+ "step": 520
374
+ },
375
+ {
376
+ "epoch": 0.43983402489626555,
377
+ "grad_norm": 0.10771152282716176,
378
+ "learning_rate": 3.1977414716927837e-06,
379
+ "loss": 0.0009,
380
+ "step": 530
381
+ },
382
+ {
383
+ "epoch": 0.44813278008298757,
384
+ "grad_norm": 0.011123547297225648,
385
+ "learning_rate": 3.131561082805898e-06,
386
+ "loss": 0.0012,
387
+ "step": 540
388
+ },
389
+ {
390
+ "epoch": 0.45643153526970953,
391
+ "grad_norm": 0.04158374445810553,
392
+ "learning_rate": 3.0649044430688017e-06,
393
+ "loss": 0.0005,
394
+ "step": 550
395
+ },
396
+ {
397
+ "epoch": 0.46473029045643155,
398
+ "grad_norm": 0.20382791628540436,
399
+ "learning_rate": 2.997821817265394e-06,
400
+ "loss": 0.0015,
401
+ "step": 560
402
+ },
403
+ {
404
+ "epoch": 0.4730290456431535,
405
+ "grad_norm": 0.10253541306876046,
406
+ "learning_rate": 2.9303637914093532e-06,
407
+ "loss": 0.0002,
408
+ "step": 570
409
+ },
410
+ {
411
+ "epoch": 0.48132780082987553,
412
+ "grad_norm": 0.0368908683192571,
413
+ "learning_rate": 2.8625812345979716e-06,
414
+ "loss": 0.0004,
415
+ "step": 580
416
+ },
417
+ {
418
+ "epoch": 0.4896265560165975,
419
+ "grad_norm": 0.4132164259019655,
420
+ "learning_rate": 2.7945252606525244e-06,
421
+ "loss": 0.0009,
422
+ "step": 590
423
+ },
424
+ {
425
+ "epoch": 0.4979253112033195,
426
+ "grad_norm": 0.02180157304973387,
427
+ "learning_rate": 2.726247189574095e-06,
428
+ "loss": 0.0007,
429
+ "step": 600
430
+ },
431
+ {
432
+ "epoch": 0.5062240663900415,
433
+ "grad_norm": 0.023273145386899886,
434
+ "learning_rate": 2.6577985088439212e-06,
435
+ "loss": 0.0013,
436
+ "step": 610
437
+ },
438
+ {
439
+ "epoch": 0.5145228215767634,
440
+ "grad_norm": 0.03091350937162301,
441
+ "learning_rate": 2.5892308345974517e-06,
442
+ "loss": 0.0003,
443
+ "step": 620
444
+ },
445
+ {
446
+ "epoch": 0.5228215767634855,
447
+ "grad_norm": 0.002613622106717658,
448
+ "learning_rate": 2.520595872701379e-06,
449
+ "loss": 0.0001,
450
+ "step": 630
451
+ },
452
+ {
453
+ "epoch": 0.5311203319502075,
454
+ "grad_norm": 0.06951997008434906,
455
+ "learning_rate": 2.451945379763019e-06,
456
+ "loss": 0.0002,
457
+ "step": 640
458
+ },
459
+ {
460
+ "epoch": 0.5394190871369294,
461
+ "grad_norm": 0.0012440442823388085,
462
+ "learning_rate": 2.383331124101412e-06,
463
+ "loss": 0.0004,
464
+ "step": 650
465
+ },
466
+ {
467
+ "epoch": 0.5477178423236515,
468
+ "grad_norm": 0.017965233689354338,
469
+ "learning_rate": 2.3148048467096076e-06,
470
+ "loss": 0.0005,
471
+ "step": 660
472
+ },
473
+ {
474
+ "epoch": 0.5560165975103735,
475
+ "grad_norm": 0.0017474955006528351,
476
+ "learning_rate": 2.2464182222375466e-06,
477
+ "loss": 0.0009,
478
+ "step": 670
479
+ },
480
+ {
481
+ "epoch": 0.5643153526970954,
482
+ "grad_norm": 0.08853303234184502,
483
+ "learning_rate": 2.1782228200249705e-06,
484
+ "loss": 0.0002,
485
+ "step": 680
486
+ },
487
+ {
488
+ "epoch": 0.5726141078838174,
489
+ "grad_norm": 0.055082873747902855,
490
+ "learning_rate": 2.1102700652137527e-06,
491
+ "loss": 0.0004,
492
+ "step": 690
493
+ },
494
+ {
495
+ "epoch": 0.5809128630705395,
496
+ "grad_norm": 0.0989162087632671,
497
+ "learning_rate": 2.0426111999689568e-06,
498
+ "loss": 0.001,
499
+ "step": 700
500
+ },
501
+ {
502
+ "epoch": 0.5892116182572614,
503
+ "grad_norm": 0.002211178671875013,
504
+ "learning_rate": 1.9752972448378817e-06,
505
+ "loss": 0.0004,
506
+ "step": 710
507
+ },
508
+ {
509
+ "epoch": 0.5975103734439834,
510
+ "grad_norm": 0.05298706307817462,
511
+ "learning_rate": 1.908378960276221e-06,
512
+ "loss": 0.0003,
513
+ "step": 720
514
+ },
515
+ {
516
+ "epoch": 0.6058091286307054,
517
+ "grad_norm": 0.0037145084040453163,
518
+ "learning_rate": 1.84190680837036e-06,
519
+ "loss": 0.0003,
520
+ "step": 730
521
+ },
522
+ {
523
+ "epoch": 0.6141078838174274,
524
+ "grad_norm": 0.0020591082549731407,
525
+ "learning_rate": 1.775930914784659e-06,
526
+ "loss": 0.0001,
527
+ "step": 740
528
+ },
529
+ {
530
+ "epoch": 0.6224066390041494,
531
+ "grad_norm": 0.002395583907814531,
532
+ "learning_rate": 1.7105010309624381e-06,
533
+ "loss": 0.001,
534
+ "step": 750
535
+ },
536
+ {
537
+ "epoch": 0.6307053941908713,
538
+ "grad_norm": 0.010281137035125171,
539
+ "learning_rate": 1.6456664966091463e-06,
540
+ "loss": 0.0001,
541
+ "step": 760
542
+ },
543
+ {
544
+ "epoch": 0.6390041493775933,
545
+ "grad_norm": 0.005408958318286851,
546
+ "learning_rate": 1.5814762024860261e-06,
547
+ "loss": 0.001,
548
+ "step": 770
549
+ },
550
+ {
551
+ "epoch": 0.6473029045643154,
552
+ "grad_norm": 0.016316274316565883,
553
+ "learning_rate": 1.5179785535423109e-06,
554
+ "loss": 0.0002,
555
+ "step": 780
556
+ },
557
+ {
558
+ "epoch": 0.6556016597510373,
559
+ "grad_norm": 0.025131773659064247,
560
+ "learning_rate": 1.4552214324137744e-06,
561
+ "loss": 0.0003,
562
+ "step": 790
563
+ },
564
+ {
565
+ "epoch": 0.6639004149377593,
566
+ "grad_norm": 0.005562868034328769,
567
+ "learning_rate": 1.393252163315138e-06,
568
+ "loss": 0.0003,
569
+ "step": 800
570
+ },
571
+ {
572
+ "epoch": 0.6721991701244814,
573
+ "grad_norm": 0.014346694616478282,
574
+ "learning_rate": 1.3321174763535926e-06,
575
+ "loss": 0.0005,
576
+ "step": 810
577
+ },
578
+ {
579
+ "epoch": 0.6804979253112033,
580
+ "grad_norm": 0.009721758111534043,
581
+ "learning_rate": 1.2718634722903073e-06,
582
+ "loss": 0.0011,
583
+ "step": 820
584
+ },
585
+ {
586
+ "epoch": 0.6887966804979253,
587
+ "grad_norm": 0.06104884728923139,
588
+ "learning_rate": 1.2125355877765336e-06,
589
+ "loss": 0.0003,
590
+ "step": 830
591
+ },
592
+ {
593
+ "epoch": 0.6970954356846473,
594
+ "grad_norm": 0.008343792515062282,
595
+ "learning_rate": 1.1541785610904995e-06,
596
+ "loss": 0.0,
597
+ "step": 840
598
+ },
599
+ {
600
+ "epoch": 0.7053941908713693,
601
+ "grad_norm": 0.7020599008202609,
602
+ "learning_rate": 1.0968363984009402e-06,
603
+ "loss": 0.001,
604
+ "step": 850
605
+ },
606
+ {
607
+ "epoch": 0.7136929460580913,
608
+ "grad_norm": 0.0019620049947002543,
609
+ "learning_rate": 1.0405523405826917e-06,
610
+ "loss": 0.0005,
611
+ "step": 860
612
+ },
613
+ {
614
+ "epoch": 0.7219917012448133,
615
+ "grad_norm": 0.0018741347835914191,
616
+ "learning_rate": 9.853688306093935e-07,
617
+ "loss": 0.0001,
618
+ "step": 870
619
+ },
620
+ {
621
+ "epoch": 0.7302904564315352,
622
+ "grad_norm": 0.0030367972735562128,
623
+ "learning_rate": 9.313274815478698e-07,
624
+ "loss": 0.0001,
625
+ "step": 880
626
+ },
627
+ {
628
+ "epoch": 0.7385892116182573,
629
+ "grad_norm": 0.06025337850459377,
630
+ "learning_rate": 8.784690451783337e-07,
631
+ "loss": 0.0012,
632
+ "step": 890
633
+ },
634
+ {
635
+ "epoch": 0.7468879668049793,
636
+ "grad_norm": 0.0008554758999917974,
637
+ "learning_rate": 8.268333812640717e-07,
638
+ "loss": 0.0013,
639
+ "step": 900
640
+ },
641
+ {
642
+ "epoch": 0.7551867219917012,
643
+ "grad_norm": 0.156929704775561,
644
+ "learning_rate": 7.764594274937884e-07,
645
+ "loss": 0.0006,
646
+ "step": 910
647
+ },
648
+ {
649
+ "epoch": 0.7634854771784232,
650
+ "grad_norm": 0.0015978578008091497,
651
+ "learning_rate": 7.27385170119273e-07,
652
+ "loss": 0.0001,
653
+ "step": 920
654
+ },
655
+ {
656
+ "epoch": 0.7717842323651453,
657
+ "grad_norm": 0.04001419658127684,
658
+ "learning_rate": 6.796476153105294e-07,
659
+ "loss": 0.0002,
660
+ "step": 930
661
+ },
662
+ {
663
+ "epoch": 0.7800829875518672,
664
+ "grad_norm": 0.0031656530089931665,
665
+ "learning_rate": 6.332827612499718e-07,
666
+ "loss": 0.0001,
667
+ "step": 940
668
+ },
669
+ {
670
+ "epoch": 0.7883817427385892,
671
+ "grad_norm": 0.0136750474203667,
672
+ "learning_rate": 5.883255709867288e-07,
673
+ "loss": 0.0001,
674
+ "step": 950
675
+ },
676
+ {
677
+ "epoch": 0.7966804979253111,
678
+ "grad_norm": 0.006491577024871877,
679
+ "learning_rate": 5.448099460715289e-07,
680
+ "loss": 0.0003,
681
+ "step": 960
682
+ },
683
+ {
684
+ "epoch": 0.8049792531120332,
685
+ "grad_norm": 0.03313707798793249,
686
+ "learning_rate": 5.027687009920393e-07,
687
+ "loss": 0.0003,
688
+ "step": 970
689
+ },
690
+ {
691
+ "epoch": 0.8132780082987552,
692
+ "grad_norm": 0.011464450089201293,
693
+ "learning_rate": 4.6223353842795053e-07,
694
+ "loss": 0.0003,
695
+ "step": 980
696
+ },
697
+ {
698
+ "epoch": 0.8215767634854771,
699
+ "grad_norm": 0.001030709037873557,
700
+ "learning_rate": 4.232350253444539e-07,
701
+ "loss": 0.0003,
702
+ "step": 990
703
+ },
704
+ {
705
+ "epoch": 0.8298755186721992,
706
+ "grad_norm": 0.01249193689623948,
707
+ "learning_rate": 3.8580256994214805e-07,
708
+ "loss": 0.0003,
709
+ "step": 1000
710
+ },
711
+ {
712
+ "epoch": 0.8381742738589212,
713
+ "grad_norm": 0.003310230202843625,
714
+ "learning_rate": 3.499643994807486e-07,
715
+ "loss": 0.0003,
716
+ "step": 1010
717
+ },
718
+ {
719
+ "epoch": 0.8464730290456431,
720
+ "grad_norm": 0.007029766220764641,
721
+ "learning_rate": 3.1574753899333176e-07,
722
+ "loss": 0.0001,
723
+ "step": 1020
724
+ },
725
+ {
726
+ "epoch": 0.8547717842323651,
727
+ "grad_norm": 0.015014047974484341,
728
+ "learning_rate": 2.8317779090715593e-07,
729
+ "loss": 0.0015,
730
+ "step": 1030
731
+ },
732
+ {
733
+ "epoch": 0.8630705394190872,
734
+ "grad_norm": 0.028295865852545128,
735
+ "learning_rate": 2.522797155864354e-07,
736
+ "loss": 0.0001,
737
+ "step": 1040
738
+ },
739
+ {
740
+ "epoch": 0.8713692946058091,
741
+ "grad_norm": 0.031063791170280682,
742
+ "learning_rate": 2.2307661281173314e-07,
743
+ "loss": 0.0001,
744
+ "step": 1050
745
+ },
746
+ {
747
+ "epoch": 0.8796680497925311,
748
+ "grad_norm": 0.016605299497541786,
749
+ "learning_rate": 1.9559050420994124e-07,
750
+ "loss": 0.0001,
751
+ "step": 1060
752
+ },
753
+ {
754
+ "epoch": 0.8879668049792531,
755
+ "grad_norm": 0.02594585316898821,
756
+ "learning_rate": 1.6984211664809957e-07,
757
+ "loss": 0.0002,
758
+ "step": 1070
759
+ },
760
+ {
761
+ "epoch": 0.8962655601659751,
762
+ "grad_norm": 0.0010463749820481666,
763
+ "learning_rate": 1.4585086660357155e-07,
764
+ "loss": 0.0,
765
+ "step": 1080
766
+ },
767
+ {
768
+ "epoch": 0.9045643153526971,
769
+ "grad_norm": 0.014847962234540494,
770
+ "learning_rate": 1.2363484552236654e-07,
771
+ "loss": 0.0005,
772
+ "step": 1090
773
+ },
774
+ {
775
+ "epoch": 0.9128630705394191,
776
+ "grad_norm": 0.0031223430965961747,
777
+ "learning_rate": 1.0321080617664846e-07,
778
+ "loss": 0.0001,
779
+ "step": 1100
780
+ },
781
+ {
782
+ "epoch": 0.921161825726141,
783
+ "grad_norm": 0.0014872131531329376,
784
+ "learning_rate": 8.4594150031721e-08,
785
+ "loss": 0.0001,
786
+ "step": 1110
787
+ },
788
+ {
789
+ "epoch": 0.9294605809128631,
790
+ "grad_norm": 0.0008679571388583524,
791
+ "learning_rate": 6.779891563200664e-08,
792
+ "loss": 0.0001,
793
+ "step": 1120
794
+ },
795
+ {
796
+ "epoch": 0.9377593360995851,
797
+ "grad_norm": 0.0028433062789473456,
798
+ "learning_rate": 5.2837768014792066e-08,
799
+ "loss": 0.0001,
800
+ "step": 1130
801
+ },
802
+ {
803
+ "epoch": 0.946058091286307,
804
+ "grad_norm": 0.01390940083993496,
805
+ "learning_rate": 3.972198915970976e-08,
806
+ "loss": 0.0004,
807
+ "step": 1140
808
+ },
809
+ {
810
+ "epoch": 0.9543568464730291,
811
+ "grad_norm": 0.013627677425191088,
812
+ "learning_rate": 2.8461469481164682e-08,
813
+ "loss": 0.0002,
814
+ "step": 1150
815
+ },
816
+ {
817
+ "epoch": 0.9626556016597511,
818
+ "grad_norm": 0.0006225638406747092,
819
+ "learning_rate": 1.9064700370118593e-08,
820
+ "loss": 0.001,
821
+ "step": 1160
822
+ },
823
+ {
824
+ "epoch": 0.970954356846473,
825
+ "grad_norm": 0.06472085081508153,
826
+ "learning_rate": 1.1538767790859885e-08,
827
+ "loss": 0.0009,
828
+ "step": 1170
829
+ },
830
+ {
831
+ "epoch": 0.979253112033195,
832
+ "grad_norm": 0.03204457354498749,
833
+ "learning_rate": 5.889346937581475e-09,
834
+ "loss": 0.0002,
835
+ "step": 1180
836
+ },
837
+ {
838
+ "epoch": 0.9875518672199171,
839
+ "grad_norm": 0.19770717466243257,
840
+ "learning_rate": 2.120697954800799e-09,
841
+ "loss": 0.0003,
842
+ "step": 1190
843
+ },
844
+ {
845
+ "epoch": 0.995850622406639,
846
+ "grad_norm": 0.0013634677194990677,
847
+ "learning_rate": 2.35662724848984e-10,
848
+ "loss": 0.0001,
849
+ "step": 1200
850
+ },
851
+ {
852
+ "epoch": 1.0,
853
+ "step": 1205,
854
+ "total_flos": 901340604923904.0,
855
+ "train_loss": 0.0009180365409025791,
856
+ "train_runtime": 38806.1809,
857
+ "train_samples_per_second": 5.96,
858
+ "train_steps_per_second": 0.031
859
+ }
860
+ ],
861
+ "logging_steps": 10,
862
+ "max_steps": 1205,
863
+ "num_input_tokens_seen": 0,
864
+ "num_train_epochs": 1,
865
+ "save_steps": 800,
866
+ "stateful_callbacks": {
867
+ "TrainerControl": {
868
+ "args": {
869
+ "should_epoch_stop": false,
870
+ "should_evaluate": false,
871
+ "should_log": false,
872
+ "should_save": true,
873
+ "should_training_stop": true
874
+ },
875
+ "attributes": {}
876
+ }
877
+ },
878
+ "total_flos": 901340604923904.0,
879
+ "train_batch_size": 4,
880
+ "trial_name": null,
881
+ "trial_params": null
882
+ }
training_loss.png ADDED