sedrickkeh commited on
Commit
824cfb3
·
verified ·
1 Parent(s): 638de3c

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: oh-dcft-v3.1-gemini-1.5-flash
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
15
 
16
  # oh-dcft-v3.1-gemini-1.5-flash
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: 0.5841
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: oh-dcft-v3.1-gemini-1.5-flash
 
16
 
17
  # oh-dcft-v3.1-gemini-1.5-flash
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 mlfoundations-dev/oh-dcft-v3.1-gemini-1.5-flash dataset.
20
  It achieves the following results on the evaluation set:
21
  - Loss: 0.5841
22
 
all_results.json ADDED
@@ -0,0 +1,12 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.9944232462577047,
3
+ "eval_loss": 0.5840933322906494,
4
+ "eval_runtime": 287.6095,
5
+ "eval_samples_per_second": 39.894,
6
+ "eval_steps_per_second": 0.626,
7
+ "total_flos": 2135463108280320.0,
8
+ "train_loss": 0.5660238438026578,
9
+ "train_runtime": 41769.8593,
10
+ "train_samples_per_second": 15.658,
11
+ "train_steps_per_second": 0.031
12
+ }
eval_results.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.9944232462577047,
3
+ "eval_loss": 0.5840933322906494,
4
+ "eval_runtime": 287.6095,
5
+ "eval_samples_per_second": 39.894,
6
+ "eval_steps_per_second": 0.626
7
+ }
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.9944232462577047,
3
+ "total_flos": 2135463108280320.0,
4
+ "train_loss": 0.5660238438026578,
5
+ "train_runtime": 41769.8593,
6
+ "train_samples_per_second": 15.658,
7
+ "train_steps_per_second": 0.031
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,955 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 2.9944232462577047,
5
+ "eval_steps": 500,
6
+ "global_step": 1275,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.023481068388611682,
13
+ "grad_norm": 17.85382376485096,
14
+ "learning_rate": 5e-06,
15
+ "loss": 0.9488,
16
+ "step": 10
17
+ },
18
+ {
19
+ "epoch": 0.046962136777223364,
20
+ "grad_norm": 3.177756281658111,
21
+ "learning_rate": 5e-06,
22
+ "loss": 0.8343,
23
+ "step": 20
24
+ },
25
+ {
26
+ "epoch": 0.07044320516583505,
27
+ "grad_norm": 2.950019754054781,
28
+ "learning_rate": 5e-06,
29
+ "loss": 0.7651,
30
+ "step": 30
31
+ },
32
+ {
33
+ "epoch": 0.09392427355444673,
34
+ "grad_norm": 1.450856517854765,
35
+ "learning_rate": 5e-06,
36
+ "loss": 0.7374,
37
+ "step": 40
38
+ },
39
+ {
40
+ "epoch": 0.1174053419430584,
41
+ "grad_norm": 0.8091667434976095,
42
+ "learning_rate": 5e-06,
43
+ "loss": 0.7161,
44
+ "step": 50
45
+ },
46
+ {
47
+ "epoch": 0.1408864103316701,
48
+ "grad_norm": 0.8253122020905036,
49
+ "learning_rate": 5e-06,
50
+ "loss": 0.6935,
51
+ "step": 60
52
+ },
53
+ {
54
+ "epoch": 0.16436747872028176,
55
+ "grad_norm": 0.6338190868455624,
56
+ "learning_rate": 5e-06,
57
+ "loss": 0.6718,
58
+ "step": 70
59
+ },
60
+ {
61
+ "epoch": 0.18784854710889345,
62
+ "grad_norm": 0.8143184435680945,
63
+ "learning_rate": 5e-06,
64
+ "loss": 0.6635,
65
+ "step": 80
66
+ },
67
+ {
68
+ "epoch": 0.21132961549750515,
69
+ "grad_norm": 0.7202907234528798,
70
+ "learning_rate": 5e-06,
71
+ "loss": 0.6581,
72
+ "step": 90
73
+ },
74
+ {
75
+ "epoch": 0.2348106838861168,
76
+ "grad_norm": 0.9812703695297975,
77
+ "learning_rate": 5e-06,
78
+ "loss": 0.6595,
79
+ "step": 100
80
+ },
81
+ {
82
+ "epoch": 0.2582917522747285,
83
+ "grad_norm": 0.6403918144312425,
84
+ "learning_rate": 5e-06,
85
+ "loss": 0.6511,
86
+ "step": 110
87
+ },
88
+ {
89
+ "epoch": 0.2817728206633402,
90
+ "grad_norm": 0.6355110618738598,
91
+ "learning_rate": 5e-06,
92
+ "loss": 0.6443,
93
+ "step": 120
94
+ },
95
+ {
96
+ "epoch": 0.3052538890519519,
97
+ "grad_norm": 0.8767568074879388,
98
+ "learning_rate": 5e-06,
99
+ "loss": 0.6457,
100
+ "step": 130
101
+ },
102
+ {
103
+ "epoch": 0.3287349574405635,
104
+ "grad_norm": 0.7912634151639979,
105
+ "learning_rate": 5e-06,
106
+ "loss": 0.6352,
107
+ "step": 140
108
+ },
109
+ {
110
+ "epoch": 0.3522160258291752,
111
+ "grad_norm": 0.6346769190072642,
112
+ "learning_rate": 5e-06,
113
+ "loss": 0.642,
114
+ "step": 150
115
+ },
116
+ {
117
+ "epoch": 0.3756970942177869,
118
+ "grad_norm": 0.6538804073841475,
119
+ "learning_rate": 5e-06,
120
+ "loss": 0.6453,
121
+ "step": 160
122
+ },
123
+ {
124
+ "epoch": 0.3991781626063986,
125
+ "grad_norm": 0.524677402567732,
126
+ "learning_rate": 5e-06,
127
+ "loss": 0.6375,
128
+ "step": 170
129
+ },
130
+ {
131
+ "epoch": 0.4226592309950103,
132
+ "grad_norm": 0.6983299832658841,
133
+ "learning_rate": 5e-06,
134
+ "loss": 0.6404,
135
+ "step": 180
136
+ },
137
+ {
138
+ "epoch": 0.44614029938362193,
139
+ "grad_norm": 0.5730458718423643,
140
+ "learning_rate": 5e-06,
141
+ "loss": 0.6282,
142
+ "step": 190
143
+ },
144
+ {
145
+ "epoch": 0.4696213677722336,
146
+ "grad_norm": 0.5795935762469521,
147
+ "learning_rate": 5e-06,
148
+ "loss": 0.6258,
149
+ "step": 200
150
+ },
151
+ {
152
+ "epoch": 0.4931024361608453,
153
+ "grad_norm": 0.6568013132353124,
154
+ "learning_rate": 5e-06,
155
+ "loss": 0.633,
156
+ "step": 210
157
+ },
158
+ {
159
+ "epoch": 0.516583504549457,
160
+ "grad_norm": 0.625002257246399,
161
+ "learning_rate": 5e-06,
162
+ "loss": 0.6215,
163
+ "step": 220
164
+ },
165
+ {
166
+ "epoch": 0.5400645729380686,
167
+ "grad_norm": 0.5866557972006909,
168
+ "learning_rate": 5e-06,
169
+ "loss": 0.6192,
170
+ "step": 230
171
+ },
172
+ {
173
+ "epoch": 0.5635456413266804,
174
+ "grad_norm": 0.5523814224922182,
175
+ "learning_rate": 5e-06,
176
+ "loss": 0.6225,
177
+ "step": 240
178
+ },
179
+ {
180
+ "epoch": 0.587026709715292,
181
+ "grad_norm": 0.621565965520602,
182
+ "learning_rate": 5e-06,
183
+ "loss": 0.6176,
184
+ "step": 250
185
+ },
186
+ {
187
+ "epoch": 0.6105077781039038,
188
+ "grad_norm": 0.5699863716830831,
189
+ "learning_rate": 5e-06,
190
+ "loss": 0.6184,
191
+ "step": 260
192
+ },
193
+ {
194
+ "epoch": 0.6339888464925154,
195
+ "grad_norm": 0.5867908049315175,
196
+ "learning_rate": 5e-06,
197
+ "loss": 0.6214,
198
+ "step": 270
199
+ },
200
+ {
201
+ "epoch": 0.657469914881127,
202
+ "grad_norm": 1.0031834057307554,
203
+ "learning_rate": 5e-06,
204
+ "loss": 0.6204,
205
+ "step": 280
206
+ },
207
+ {
208
+ "epoch": 0.6809509832697388,
209
+ "grad_norm": 0.5739527115683025,
210
+ "learning_rate": 5e-06,
211
+ "loss": 0.6166,
212
+ "step": 290
213
+ },
214
+ {
215
+ "epoch": 0.7044320516583504,
216
+ "grad_norm": 0.6047949935557967,
217
+ "learning_rate": 5e-06,
218
+ "loss": 0.6096,
219
+ "step": 300
220
+ },
221
+ {
222
+ "epoch": 0.7279131200469622,
223
+ "grad_norm": 0.5141588768585372,
224
+ "learning_rate": 5e-06,
225
+ "loss": 0.6076,
226
+ "step": 310
227
+ },
228
+ {
229
+ "epoch": 0.7513941884355738,
230
+ "grad_norm": 0.689840031854037,
231
+ "learning_rate": 5e-06,
232
+ "loss": 0.6115,
233
+ "step": 320
234
+ },
235
+ {
236
+ "epoch": 0.7748752568241855,
237
+ "grad_norm": 0.7551629396423358,
238
+ "learning_rate": 5e-06,
239
+ "loss": 0.6138,
240
+ "step": 330
241
+ },
242
+ {
243
+ "epoch": 0.7983563252127972,
244
+ "grad_norm": 0.5356794354442618,
245
+ "learning_rate": 5e-06,
246
+ "loss": 0.611,
247
+ "step": 340
248
+ },
249
+ {
250
+ "epoch": 0.8218373936014088,
251
+ "grad_norm": 0.6083117103787592,
252
+ "learning_rate": 5e-06,
253
+ "loss": 0.611,
254
+ "step": 350
255
+ },
256
+ {
257
+ "epoch": 0.8453184619900206,
258
+ "grad_norm": 0.7034174178117553,
259
+ "learning_rate": 5e-06,
260
+ "loss": 0.6104,
261
+ "step": 360
262
+ },
263
+ {
264
+ "epoch": 0.8687995303786322,
265
+ "grad_norm": 0.5650614816783488,
266
+ "learning_rate": 5e-06,
267
+ "loss": 0.6047,
268
+ "step": 370
269
+ },
270
+ {
271
+ "epoch": 0.8922805987672439,
272
+ "grad_norm": 0.6449034585856688,
273
+ "learning_rate": 5e-06,
274
+ "loss": 0.613,
275
+ "step": 380
276
+ },
277
+ {
278
+ "epoch": 0.9157616671558556,
279
+ "grad_norm": 0.6460327839123797,
280
+ "learning_rate": 5e-06,
281
+ "loss": 0.6104,
282
+ "step": 390
283
+ },
284
+ {
285
+ "epoch": 0.9392427355444672,
286
+ "grad_norm": 0.6977465772968938,
287
+ "learning_rate": 5e-06,
288
+ "loss": 0.6124,
289
+ "step": 400
290
+ },
291
+ {
292
+ "epoch": 0.962723803933079,
293
+ "grad_norm": 0.6646492018550946,
294
+ "learning_rate": 5e-06,
295
+ "loss": 0.5957,
296
+ "step": 410
297
+ },
298
+ {
299
+ "epoch": 0.9862048723216906,
300
+ "grad_norm": 0.48972085408134464,
301
+ "learning_rate": 5e-06,
302
+ "loss": 0.5979,
303
+ "step": 420
304
+ },
305
+ {
306
+ "epoch": 0.9979454065159965,
307
+ "eval_loss": 0.6003177165985107,
308
+ "eval_runtime": 300.5123,
309
+ "eval_samples_per_second": 38.181,
310
+ "eval_steps_per_second": 0.599,
311
+ "step": 425
312
+ },
313
+ {
314
+ "epoch": 1.0099794540651599,
315
+ "grad_norm": 0.8929971915551214,
316
+ "learning_rate": 5e-06,
317
+ "loss": 0.6046,
318
+ "step": 430
319
+ },
320
+ {
321
+ "epoch": 1.0334605224537716,
322
+ "grad_norm": 0.6204797689157823,
323
+ "learning_rate": 5e-06,
324
+ "loss": 0.5525,
325
+ "step": 440
326
+ },
327
+ {
328
+ "epoch": 1.0569415908423834,
329
+ "grad_norm": 0.629777186945133,
330
+ "learning_rate": 5e-06,
331
+ "loss": 0.5507,
332
+ "step": 450
333
+ },
334
+ {
335
+ "epoch": 1.0804226592309951,
336
+ "grad_norm": 0.6177917289848706,
337
+ "learning_rate": 5e-06,
338
+ "loss": 0.5488,
339
+ "step": 460
340
+ },
341
+ {
342
+ "epoch": 1.1039037276196066,
343
+ "grad_norm": 0.616801854556511,
344
+ "learning_rate": 5e-06,
345
+ "loss": 0.5553,
346
+ "step": 470
347
+ },
348
+ {
349
+ "epoch": 1.1273847960082184,
350
+ "grad_norm": 0.6202115496799256,
351
+ "learning_rate": 5e-06,
352
+ "loss": 0.5517,
353
+ "step": 480
354
+ },
355
+ {
356
+ "epoch": 1.1508658643968301,
357
+ "grad_norm": 0.6869810211842367,
358
+ "learning_rate": 5e-06,
359
+ "loss": 0.5513,
360
+ "step": 490
361
+ },
362
+ {
363
+ "epoch": 1.1743469327854417,
364
+ "grad_norm": 0.6134525997635181,
365
+ "learning_rate": 5e-06,
366
+ "loss": 0.5565,
367
+ "step": 500
368
+ },
369
+ {
370
+ "epoch": 1.1978280011740534,
371
+ "grad_norm": 0.6533976679541116,
372
+ "learning_rate": 5e-06,
373
+ "loss": 0.5508,
374
+ "step": 510
375
+ },
376
+ {
377
+ "epoch": 1.2213090695626652,
378
+ "grad_norm": 0.7384172080803894,
379
+ "learning_rate": 5e-06,
380
+ "loss": 0.5513,
381
+ "step": 520
382
+ },
383
+ {
384
+ "epoch": 1.2447901379512767,
385
+ "grad_norm": 0.6446266722666449,
386
+ "learning_rate": 5e-06,
387
+ "loss": 0.5538,
388
+ "step": 530
389
+ },
390
+ {
391
+ "epoch": 1.2682712063398884,
392
+ "grad_norm": 0.6920463607543151,
393
+ "learning_rate": 5e-06,
394
+ "loss": 0.5442,
395
+ "step": 540
396
+ },
397
+ {
398
+ "epoch": 1.2917522747285002,
399
+ "grad_norm": 0.7090350057008448,
400
+ "learning_rate": 5e-06,
401
+ "loss": 0.5566,
402
+ "step": 550
403
+ },
404
+ {
405
+ "epoch": 1.3152333431171117,
406
+ "grad_norm": 0.7226606579176621,
407
+ "learning_rate": 5e-06,
408
+ "loss": 0.5494,
409
+ "step": 560
410
+ },
411
+ {
412
+ "epoch": 1.3387144115057235,
413
+ "grad_norm": 0.5484819340890661,
414
+ "learning_rate": 5e-06,
415
+ "loss": 0.5471,
416
+ "step": 570
417
+ },
418
+ {
419
+ "epoch": 1.3621954798943352,
420
+ "grad_norm": 0.6982064642691793,
421
+ "learning_rate": 5e-06,
422
+ "loss": 0.552,
423
+ "step": 580
424
+ },
425
+ {
426
+ "epoch": 1.385676548282947,
427
+ "grad_norm": 0.6147097813447182,
428
+ "learning_rate": 5e-06,
429
+ "loss": 0.5505,
430
+ "step": 590
431
+ },
432
+ {
433
+ "epoch": 1.4091576166715585,
434
+ "grad_norm": 0.6598577111107449,
435
+ "learning_rate": 5e-06,
436
+ "loss": 0.5498,
437
+ "step": 600
438
+ },
439
+ {
440
+ "epoch": 1.4326386850601702,
441
+ "grad_norm": 0.5603317744558156,
442
+ "learning_rate": 5e-06,
443
+ "loss": 0.5562,
444
+ "step": 610
445
+ },
446
+ {
447
+ "epoch": 1.456119753448782,
448
+ "grad_norm": 0.6373670868271739,
449
+ "learning_rate": 5e-06,
450
+ "loss": 0.5555,
451
+ "step": 620
452
+ },
453
+ {
454
+ "epoch": 1.4796008218373937,
455
+ "grad_norm": 0.5932324800613761,
456
+ "learning_rate": 5e-06,
457
+ "loss": 0.544,
458
+ "step": 630
459
+ },
460
+ {
461
+ "epoch": 1.5030818902260052,
462
+ "grad_norm": 0.5227721794518003,
463
+ "learning_rate": 5e-06,
464
+ "loss": 0.5451,
465
+ "step": 640
466
+ },
467
+ {
468
+ "epoch": 1.526562958614617,
469
+ "grad_norm": 0.6093734754379218,
470
+ "learning_rate": 5e-06,
471
+ "loss": 0.5587,
472
+ "step": 650
473
+ },
474
+ {
475
+ "epoch": 1.5500440270032287,
476
+ "grad_norm": 0.6287567431322201,
477
+ "learning_rate": 5e-06,
478
+ "loss": 0.5518,
479
+ "step": 660
480
+ },
481
+ {
482
+ "epoch": 1.5735250953918403,
483
+ "grad_norm": 0.6541153306269628,
484
+ "learning_rate": 5e-06,
485
+ "loss": 0.5526,
486
+ "step": 670
487
+ },
488
+ {
489
+ "epoch": 1.597006163780452,
490
+ "grad_norm": 0.6150418746443832,
491
+ "learning_rate": 5e-06,
492
+ "loss": 0.5478,
493
+ "step": 680
494
+ },
495
+ {
496
+ "epoch": 1.6204872321690638,
497
+ "grad_norm": 0.5475396144251581,
498
+ "learning_rate": 5e-06,
499
+ "loss": 0.5481,
500
+ "step": 690
501
+ },
502
+ {
503
+ "epoch": 1.6439683005576753,
504
+ "grad_norm": 0.6160254516028266,
505
+ "learning_rate": 5e-06,
506
+ "loss": 0.5515,
507
+ "step": 700
508
+ },
509
+ {
510
+ "epoch": 1.667449368946287,
511
+ "grad_norm": 0.843962487015559,
512
+ "learning_rate": 5e-06,
513
+ "loss": 0.5505,
514
+ "step": 710
515
+ },
516
+ {
517
+ "epoch": 1.6909304373348988,
518
+ "grad_norm": 0.5347960520186497,
519
+ "learning_rate": 5e-06,
520
+ "loss": 0.5496,
521
+ "step": 720
522
+ },
523
+ {
524
+ "epoch": 1.7144115057235103,
525
+ "grad_norm": 0.7323294593212619,
526
+ "learning_rate": 5e-06,
527
+ "loss": 0.5484,
528
+ "step": 730
529
+ },
530
+ {
531
+ "epoch": 1.737892574112122,
532
+ "grad_norm": 0.5476336095439653,
533
+ "learning_rate": 5e-06,
534
+ "loss": 0.5483,
535
+ "step": 740
536
+ },
537
+ {
538
+ "epoch": 1.7613736425007338,
539
+ "grad_norm": 0.6011568410465314,
540
+ "learning_rate": 5e-06,
541
+ "loss": 0.5531,
542
+ "step": 750
543
+ },
544
+ {
545
+ "epoch": 1.7848547108893453,
546
+ "grad_norm": 0.6797212856687878,
547
+ "learning_rate": 5e-06,
548
+ "loss": 0.5487,
549
+ "step": 760
550
+ },
551
+ {
552
+ "epoch": 1.8083357792779573,
553
+ "grad_norm": 0.5819194256052239,
554
+ "learning_rate": 5e-06,
555
+ "loss": 0.5485,
556
+ "step": 770
557
+ },
558
+ {
559
+ "epoch": 1.8318168476665688,
560
+ "grad_norm": 0.5292485271089171,
561
+ "learning_rate": 5e-06,
562
+ "loss": 0.5518,
563
+ "step": 780
564
+ },
565
+ {
566
+ "epoch": 1.8552979160551804,
567
+ "grad_norm": 0.5799939542823142,
568
+ "learning_rate": 5e-06,
569
+ "loss": 0.5467,
570
+ "step": 790
571
+ },
572
+ {
573
+ "epoch": 1.8787789844437923,
574
+ "grad_norm": 0.5570487573376319,
575
+ "learning_rate": 5e-06,
576
+ "loss": 0.5468,
577
+ "step": 800
578
+ },
579
+ {
580
+ "epoch": 1.9022600528324038,
581
+ "grad_norm": 0.6174505205210491,
582
+ "learning_rate": 5e-06,
583
+ "loss": 0.5488,
584
+ "step": 810
585
+ },
586
+ {
587
+ "epoch": 1.9257411212210156,
588
+ "grad_norm": 0.5693760728585585,
589
+ "learning_rate": 5e-06,
590
+ "loss": 0.5458,
591
+ "step": 820
592
+ },
593
+ {
594
+ "epoch": 1.9492221896096273,
595
+ "grad_norm": 0.5189329244004007,
596
+ "learning_rate": 5e-06,
597
+ "loss": 0.5447,
598
+ "step": 830
599
+ },
600
+ {
601
+ "epoch": 1.9727032579982389,
602
+ "grad_norm": 0.6238194647054416,
603
+ "learning_rate": 5e-06,
604
+ "loss": 0.5398,
605
+ "step": 840
606
+ },
607
+ {
608
+ "epoch": 1.9961843263868506,
609
+ "grad_norm": 0.5651058414473042,
610
+ "learning_rate": 5e-06,
611
+ "loss": 0.547,
612
+ "step": 850
613
+ },
614
+ {
615
+ "epoch": 1.9985324332257117,
616
+ "eval_loss": 0.5846751928329468,
617
+ "eval_runtime": 295.576,
618
+ "eval_samples_per_second": 38.819,
619
+ "eval_steps_per_second": 0.609,
620
+ "step": 851
621
+ },
622
+ {
623
+ "epoch": 2.0199589081303198,
624
+ "grad_norm": 0.6946392563773179,
625
+ "learning_rate": 5e-06,
626
+ "loss": 0.523,
627
+ "step": 860
628
+ },
629
+ {
630
+ "epoch": 2.0434399765189317,
631
+ "grad_norm": 0.8362364731068118,
632
+ "learning_rate": 5e-06,
633
+ "loss": 0.4924,
634
+ "step": 870
635
+ },
636
+ {
637
+ "epoch": 2.0669210449075432,
638
+ "grad_norm": 0.7772390407263537,
639
+ "learning_rate": 5e-06,
640
+ "loss": 0.4902,
641
+ "step": 880
642
+ },
643
+ {
644
+ "epoch": 2.0904021132961548,
645
+ "grad_norm": 0.5219899057565048,
646
+ "learning_rate": 5e-06,
647
+ "loss": 0.4955,
648
+ "step": 890
649
+ },
650
+ {
651
+ "epoch": 2.1138831816847667,
652
+ "grad_norm": 0.6541702370902746,
653
+ "learning_rate": 5e-06,
654
+ "loss": 0.4969,
655
+ "step": 900
656
+ },
657
+ {
658
+ "epoch": 2.1373642500733783,
659
+ "grad_norm": 0.5786919167953266,
660
+ "learning_rate": 5e-06,
661
+ "loss": 0.4865,
662
+ "step": 910
663
+ },
664
+ {
665
+ "epoch": 2.1608453184619902,
666
+ "grad_norm": 0.6257190526717714,
667
+ "learning_rate": 5e-06,
668
+ "loss": 0.4966,
669
+ "step": 920
670
+ },
671
+ {
672
+ "epoch": 2.1843263868506018,
673
+ "grad_norm": 0.6070724867911851,
674
+ "learning_rate": 5e-06,
675
+ "loss": 0.497,
676
+ "step": 930
677
+ },
678
+ {
679
+ "epoch": 2.2078074552392133,
680
+ "grad_norm": 0.7626194518667021,
681
+ "learning_rate": 5e-06,
682
+ "loss": 0.498,
683
+ "step": 940
684
+ },
685
+ {
686
+ "epoch": 2.2312885236278253,
687
+ "grad_norm": 0.66208678281747,
688
+ "learning_rate": 5e-06,
689
+ "loss": 0.4978,
690
+ "step": 950
691
+ },
692
+ {
693
+ "epoch": 2.254769592016437,
694
+ "grad_norm": 0.597091889958803,
695
+ "learning_rate": 5e-06,
696
+ "loss": 0.4954,
697
+ "step": 960
698
+ },
699
+ {
700
+ "epoch": 2.2782506604050483,
701
+ "grad_norm": 0.648113572619813,
702
+ "learning_rate": 5e-06,
703
+ "loss": 0.5024,
704
+ "step": 970
705
+ },
706
+ {
707
+ "epoch": 2.3017317287936603,
708
+ "grad_norm": 0.5800444078314209,
709
+ "learning_rate": 5e-06,
710
+ "loss": 0.4982,
711
+ "step": 980
712
+ },
713
+ {
714
+ "epoch": 2.325212797182272,
715
+ "grad_norm": 0.5914333960587344,
716
+ "learning_rate": 5e-06,
717
+ "loss": 0.4994,
718
+ "step": 990
719
+ },
720
+ {
721
+ "epoch": 2.3486938655708833,
722
+ "grad_norm": 0.7924399848654731,
723
+ "learning_rate": 5e-06,
724
+ "loss": 0.5002,
725
+ "step": 1000
726
+ },
727
+ {
728
+ "epoch": 2.3721749339594953,
729
+ "grad_norm": 0.624856284950558,
730
+ "learning_rate": 5e-06,
731
+ "loss": 0.4994,
732
+ "step": 1010
733
+ },
734
+ {
735
+ "epoch": 2.395656002348107,
736
+ "grad_norm": 0.6655834278053183,
737
+ "learning_rate": 5e-06,
738
+ "loss": 0.4906,
739
+ "step": 1020
740
+ },
741
+ {
742
+ "epoch": 2.4191370707367184,
743
+ "grad_norm": 0.5518645145037842,
744
+ "learning_rate": 5e-06,
745
+ "loss": 0.4975,
746
+ "step": 1030
747
+ },
748
+ {
749
+ "epoch": 2.4426181391253303,
750
+ "grad_norm": 0.5785599308848968,
751
+ "learning_rate": 5e-06,
752
+ "loss": 0.4976,
753
+ "step": 1040
754
+ },
755
+ {
756
+ "epoch": 2.466099207513942,
757
+ "grad_norm": 0.5834274439157257,
758
+ "learning_rate": 5e-06,
759
+ "loss": 0.5052,
760
+ "step": 1050
761
+ },
762
+ {
763
+ "epoch": 2.4895802759025534,
764
+ "grad_norm": 0.777173362933167,
765
+ "learning_rate": 5e-06,
766
+ "loss": 0.4989,
767
+ "step": 1060
768
+ },
769
+ {
770
+ "epoch": 2.5130613442911653,
771
+ "grad_norm": 0.929125527693667,
772
+ "learning_rate": 5e-06,
773
+ "loss": 0.5059,
774
+ "step": 1070
775
+ },
776
+ {
777
+ "epoch": 2.536542412679777,
778
+ "grad_norm": 0.6950708275354937,
779
+ "learning_rate": 5e-06,
780
+ "loss": 0.4946,
781
+ "step": 1080
782
+ },
783
+ {
784
+ "epoch": 2.560023481068389,
785
+ "grad_norm": 0.5542934300294031,
786
+ "learning_rate": 5e-06,
787
+ "loss": 0.5086,
788
+ "step": 1090
789
+ },
790
+ {
791
+ "epoch": 2.5835045494570004,
792
+ "grad_norm": 0.7961833486998505,
793
+ "learning_rate": 5e-06,
794
+ "loss": 0.498,
795
+ "step": 1100
796
+ },
797
+ {
798
+ "epoch": 2.606985617845612,
799
+ "grad_norm": 0.6237039054057065,
800
+ "learning_rate": 5e-06,
801
+ "loss": 0.4998,
802
+ "step": 1110
803
+ },
804
+ {
805
+ "epoch": 2.6304666862342234,
806
+ "grad_norm": 0.5833433757118394,
807
+ "learning_rate": 5e-06,
808
+ "loss": 0.4944,
809
+ "step": 1120
810
+ },
811
+ {
812
+ "epoch": 2.6539477546228354,
813
+ "grad_norm": 0.5868493372745182,
814
+ "learning_rate": 5e-06,
815
+ "loss": 0.5028,
816
+ "step": 1130
817
+ },
818
+ {
819
+ "epoch": 2.677428823011447,
820
+ "grad_norm": 0.6399452879323836,
821
+ "learning_rate": 5e-06,
822
+ "loss": 0.4982,
823
+ "step": 1140
824
+ },
825
+ {
826
+ "epoch": 2.700909891400059,
827
+ "grad_norm": 0.6458178037380802,
828
+ "learning_rate": 5e-06,
829
+ "loss": 0.5009,
830
+ "step": 1150
831
+ },
832
+ {
833
+ "epoch": 2.7243909597886704,
834
+ "grad_norm": 0.5439040913583559,
835
+ "learning_rate": 5e-06,
836
+ "loss": 0.5012,
837
+ "step": 1160
838
+ },
839
+ {
840
+ "epoch": 2.747872028177282,
841
+ "grad_norm": 0.735794966069211,
842
+ "learning_rate": 5e-06,
843
+ "loss": 0.5032,
844
+ "step": 1170
845
+ },
846
+ {
847
+ "epoch": 2.771353096565894,
848
+ "grad_norm": 0.601851566145536,
849
+ "learning_rate": 5e-06,
850
+ "loss": 0.5033,
851
+ "step": 1180
852
+ },
853
+ {
854
+ "epoch": 2.7948341649545054,
855
+ "grad_norm": 0.5929336883777777,
856
+ "learning_rate": 5e-06,
857
+ "loss": 0.5037,
858
+ "step": 1190
859
+ },
860
+ {
861
+ "epoch": 2.818315233343117,
862
+ "grad_norm": 0.5778177618943612,
863
+ "learning_rate": 5e-06,
864
+ "loss": 0.508,
865
+ "step": 1200
866
+ },
867
+ {
868
+ "epoch": 2.841796301731729,
869
+ "grad_norm": 0.5990753577226628,
870
+ "learning_rate": 5e-06,
871
+ "loss": 0.4957,
872
+ "step": 1210
873
+ },
874
+ {
875
+ "epoch": 2.8652773701203404,
876
+ "grad_norm": 0.7825195485986434,
877
+ "learning_rate": 5e-06,
878
+ "loss": 0.499,
879
+ "step": 1220
880
+ },
881
+ {
882
+ "epoch": 2.888758438508952,
883
+ "grad_norm": 0.5698945641530911,
884
+ "learning_rate": 5e-06,
885
+ "loss": 0.4982,
886
+ "step": 1230
887
+ },
888
+ {
889
+ "epoch": 2.912239506897564,
890
+ "grad_norm": 0.5941997062897303,
891
+ "learning_rate": 5e-06,
892
+ "loss": 0.5006,
893
+ "step": 1240
894
+ },
895
+ {
896
+ "epoch": 2.9357205752861755,
897
+ "grad_norm": 0.5494008699329855,
898
+ "learning_rate": 5e-06,
899
+ "loss": 0.494,
900
+ "step": 1250
901
+ },
902
+ {
903
+ "epoch": 2.9592016436747874,
904
+ "grad_norm": 0.6295305801304111,
905
+ "learning_rate": 5e-06,
906
+ "loss": 0.5001,
907
+ "step": 1260
908
+ },
909
+ {
910
+ "epoch": 2.982682712063399,
911
+ "grad_norm": 0.6627594327431557,
912
+ "learning_rate": 5e-06,
913
+ "loss": 0.5003,
914
+ "step": 1270
915
+ },
916
+ {
917
+ "epoch": 2.9944232462577047,
918
+ "eval_loss": 0.5840933322906494,
919
+ "eval_runtime": 292.1509,
920
+ "eval_samples_per_second": 39.274,
921
+ "eval_steps_per_second": 0.616,
922
+ "step": 1275
923
+ },
924
+ {
925
+ "epoch": 2.9944232462577047,
926
+ "step": 1275,
927
+ "total_flos": 2135463108280320.0,
928
+ "train_loss": 0.5660238438026578,
929
+ "train_runtime": 41769.8593,
930
+ "train_samples_per_second": 15.658,
931
+ "train_steps_per_second": 0.031
932
+ }
933
+ ],
934
+ "logging_steps": 10,
935
+ "max_steps": 1275,
936
+ "num_input_tokens_seen": 0,
937
+ "num_train_epochs": 3,
938
+ "save_steps": 500,
939
+ "stateful_callbacks": {
940
+ "TrainerControl": {
941
+ "args": {
942
+ "should_epoch_stop": false,
943
+ "should_evaluate": false,
944
+ "should_log": false,
945
+ "should_save": true,
946
+ "should_training_stop": true
947
+ },
948
+ "attributes": {}
949
+ }
950
+ },
951
+ "total_flos": 2135463108280320.0,
952
+ "train_batch_size": 8,
953
+ "trial_name": null,
954
+ "trial_params": null
955
+ }
training_eval_loss.png ADDED
training_loss.png ADDED