sedrickkeh commited on
Commit
45fba3a
·
verified ·
1 Parent(s): 30c72bd

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-claude-3-5-haiku-20241022
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
15
 
16
  # oh-dcft-v3.1-claude-3-5-haiku-20241022
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: 0.4348
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-claude-3-5-haiku-20241022
 
16
 
17
  # oh-dcft-v3.1-claude-3-5-haiku-20241022
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-claude-3-5-haiku-20241022 dataset.
20
  It achieves the following results on the evaluation set:
21
  - Loss: 0.4348
22
 
all_results.json ADDED
@@ -0,0 +1,12 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.9973206311402203,
3
+ "eval_loss": 0.4348444640636444,
4
+ "eval_runtime": 283.2215,
5
+ "eval_samples_per_second": 39.951,
6
+ "eval_steps_per_second": 0.625,
7
+ "total_flos": 2105312437862400.0,
8
+ "train_loss": 0.41854361285268077,
9
+ "train_runtime": 41355.0058,
10
+ "train_samples_per_second": 15.594,
11
+ "train_steps_per_second": 0.03
12
+ }
eval_results.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.9973206311402203,
3
+ "eval_loss": 0.4348444640636444,
4
+ "eval_runtime": 283.2215,
5
+ "eval_samples_per_second": 39.951,
6
+ "eval_steps_per_second": 0.625
7
+ }
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.9973206311402203,
3
+ "total_flos": 2105312437862400.0,
4
+ "train_loss": 0.41854361285268077,
5
+ "train_runtime": 41355.0058,
6
+ "train_samples_per_second": 15.594,
7
+ "train_steps_per_second": 0.03
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,941 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 2.9973206311402203,
5
+ "eval_steps": 500,
6
+ "global_step": 1257,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.023816612086930634,
13
+ "grad_norm": 10.820662608896704,
14
+ "learning_rate": 5e-06,
15
+ "loss": 0.7962,
16
+ "step": 10
17
+ },
18
+ {
19
+ "epoch": 0.04763322417386127,
20
+ "grad_norm": 1.3002290223711226,
21
+ "learning_rate": 5e-06,
22
+ "loss": 0.6506,
23
+ "step": 20
24
+ },
25
+ {
26
+ "epoch": 0.0714498362607919,
27
+ "grad_norm": 2.274364660793754,
28
+ "learning_rate": 5e-06,
29
+ "loss": 0.5997,
30
+ "step": 30
31
+ },
32
+ {
33
+ "epoch": 0.09526644834772253,
34
+ "grad_norm": 0.91227766008184,
35
+ "learning_rate": 5e-06,
36
+ "loss": 0.5732,
37
+ "step": 40
38
+ },
39
+ {
40
+ "epoch": 0.11908306043465317,
41
+ "grad_norm": 0.8965545265977676,
42
+ "learning_rate": 5e-06,
43
+ "loss": 0.5522,
44
+ "step": 50
45
+ },
46
+ {
47
+ "epoch": 0.1428996725215838,
48
+ "grad_norm": 0.7367789232478065,
49
+ "learning_rate": 5e-06,
50
+ "loss": 0.5413,
51
+ "step": 60
52
+ },
53
+ {
54
+ "epoch": 0.16671628460851443,
55
+ "grad_norm": 1.959786214629393,
56
+ "learning_rate": 5e-06,
57
+ "loss": 0.5254,
58
+ "step": 70
59
+ },
60
+ {
61
+ "epoch": 0.19053289669544507,
62
+ "grad_norm": 1.1527597962719172,
63
+ "learning_rate": 5e-06,
64
+ "loss": 0.5263,
65
+ "step": 80
66
+ },
67
+ {
68
+ "epoch": 0.21434950878237571,
69
+ "grad_norm": 0.8809156142645043,
70
+ "learning_rate": 5e-06,
71
+ "loss": 0.5139,
72
+ "step": 90
73
+ },
74
+ {
75
+ "epoch": 0.23816612086930633,
76
+ "grad_norm": 0.6829627408546641,
77
+ "learning_rate": 5e-06,
78
+ "loss": 0.5047,
79
+ "step": 100
80
+ },
81
+ {
82
+ "epoch": 0.261982732956237,
83
+ "grad_norm": 0.5878115863496259,
84
+ "learning_rate": 5e-06,
85
+ "loss": 0.4974,
86
+ "step": 110
87
+ },
88
+ {
89
+ "epoch": 0.2857993450431676,
90
+ "grad_norm": 0.7963943568799848,
91
+ "learning_rate": 5e-06,
92
+ "loss": 0.4997,
93
+ "step": 120
94
+ },
95
+ {
96
+ "epoch": 0.30961595713009826,
97
+ "grad_norm": 0.8168312496362183,
98
+ "learning_rate": 5e-06,
99
+ "loss": 0.4909,
100
+ "step": 130
101
+ },
102
+ {
103
+ "epoch": 0.33343256921702885,
104
+ "grad_norm": 0.5860607753114995,
105
+ "learning_rate": 5e-06,
106
+ "loss": 0.4881,
107
+ "step": 140
108
+ },
109
+ {
110
+ "epoch": 0.3572491813039595,
111
+ "grad_norm": 0.6645019696194568,
112
+ "learning_rate": 5e-06,
113
+ "loss": 0.4891,
114
+ "step": 150
115
+ },
116
+ {
117
+ "epoch": 0.38106579339089014,
118
+ "grad_norm": 1.1349953393684589,
119
+ "learning_rate": 5e-06,
120
+ "loss": 0.4886,
121
+ "step": 160
122
+ },
123
+ {
124
+ "epoch": 0.4048824054778208,
125
+ "grad_norm": 0.842239450444899,
126
+ "learning_rate": 5e-06,
127
+ "loss": 0.486,
128
+ "step": 170
129
+ },
130
+ {
131
+ "epoch": 0.42869901756475143,
132
+ "grad_norm": 0.8029347648800014,
133
+ "learning_rate": 5e-06,
134
+ "loss": 0.4794,
135
+ "step": 180
136
+ },
137
+ {
138
+ "epoch": 0.45251562965168207,
139
+ "grad_norm": 0.7453080201307035,
140
+ "learning_rate": 5e-06,
141
+ "loss": 0.4779,
142
+ "step": 190
143
+ },
144
+ {
145
+ "epoch": 0.47633224173861266,
146
+ "grad_norm": 0.7560106227106046,
147
+ "learning_rate": 5e-06,
148
+ "loss": 0.4755,
149
+ "step": 200
150
+ },
151
+ {
152
+ "epoch": 0.5001488538255433,
153
+ "grad_norm": 0.6803104426449306,
154
+ "learning_rate": 5e-06,
155
+ "loss": 0.4738,
156
+ "step": 210
157
+ },
158
+ {
159
+ "epoch": 0.523965465912474,
160
+ "grad_norm": 0.5983251461773578,
161
+ "learning_rate": 5e-06,
162
+ "loss": 0.4765,
163
+ "step": 220
164
+ },
165
+ {
166
+ "epoch": 0.5477820779994046,
167
+ "grad_norm": 0.9213163769902233,
168
+ "learning_rate": 5e-06,
169
+ "loss": 0.466,
170
+ "step": 230
171
+ },
172
+ {
173
+ "epoch": 0.5715986900863352,
174
+ "grad_norm": 0.7194449280323925,
175
+ "learning_rate": 5e-06,
176
+ "loss": 0.4771,
177
+ "step": 240
178
+ },
179
+ {
180
+ "epoch": 0.5954153021732659,
181
+ "grad_norm": 0.5686308646341993,
182
+ "learning_rate": 5e-06,
183
+ "loss": 0.472,
184
+ "step": 250
185
+ },
186
+ {
187
+ "epoch": 0.6192319142601965,
188
+ "grad_norm": 1.0813640818776593,
189
+ "learning_rate": 5e-06,
190
+ "loss": 0.4661,
191
+ "step": 260
192
+ },
193
+ {
194
+ "epoch": 0.6430485263471272,
195
+ "grad_norm": 0.7358980285755272,
196
+ "learning_rate": 5e-06,
197
+ "loss": 0.4659,
198
+ "step": 270
199
+ },
200
+ {
201
+ "epoch": 0.6668651384340577,
202
+ "grad_norm": 0.6604585093191498,
203
+ "learning_rate": 5e-06,
204
+ "loss": 0.4679,
205
+ "step": 280
206
+ },
207
+ {
208
+ "epoch": 0.6906817505209883,
209
+ "grad_norm": 0.7044805358584626,
210
+ "learning_rate": 5e-06,
211
+ "loss": 0.4627,
212
+ "step": 290
213
+ },
214
+ {
215
+ "epoch": 0.714498362607919,
216
+ "grad_norm": 0.7424204761656356,
217
+ "learning_rate": 5e-06,
218
+ "loss": 0.4597,
219
+ "step": 300
220
+ },
221
+ {
222
+ "epoch": 0.7383149746948496,
223
+ "grad_norm": 0.6489446693829575,
224
+ "learning_rate": 5e-06,
225
+ "loss": 0.4596,
226
+ "step": 310
227
+ },
228
+ {
229
+ "epoch": 0.7621315867817803,
230
+ "grad_norm": 0.7368542344749955,
231
+ "learning_rate": 5e-06,
232
+ "loss": 0.4645,
233
+ "step": 320
234
+ },
235
+ {
236
+ "epoch": 0.7859481988687109,
237
+ "grad_norm": 0.7920469169588614,
238
+ "learning_rate": 5e-06,
239
+ "loss": 0.4532,
240
+ "step": 330
241
+ },
242
+ {
243
+ "epoch": 0.8097648109556416,
244
+ "grad_norm": 0.4836289408504863,
245
+ "learning_rate": 5e-06,
246
+ "loss": 0.4546,
247
+ "step": 340
248
+ },
249
+ {
250
+ "epoch": 0.8335814230425722,
251
+ "grad_norm": 0.4793593414506298,
252
+ "learning_rate": 5e-06,
253
+ "loss": 0.455,
254
+ "step": 350
255
+ },
256
+ {
257
+ "epoch": 0.8573980351295029,
258
+ "grad_norm": 0.6323515901480585,
259
+ "learning_rate": 5e-06,
260
+ "loss": 0.4562,
261
+ "step": 360
262
+ },
263
+ {
264
+ "epoch": 0.8812146472164335,
265
+ "grad_norm": 0.6239832843962634,
266
+ "learning_rate": 5e-06,
267
+ "loss": 0.4613,
268
+ "step": 370
269
+ },
270
+ {
271
+ "epoch": 0.9050312593033641,
272
+ "grad_norm": 0.545469740506038,
273
+ "learning_rate": 5e-06,
274
+ "loss": 0.4505,
275
+ "step": 380
276
+ },
277
+ {
278
+ "epoch": 0.9288478713902947,
279
+ "grad_norm": 0.7122693018169797,
280
+ "learning_rate": 5e-06,
281
+ "loss": 0.4548,
282
+ "step": 390
283
+ },
284
+ {
285
+ "epoch": 0.9526644834772253,
286
+ "grad_norm": 0.7239717244861085,
287
+ "learning_rate": 5e-06,
288
+ "loss": 0.4482,
289
+ "step": 400
290
+ },
291
+ {
292
+ "epoch": 0.976481095564156,
293
+ "grad_norm": 0.8780808883400644,
294
+ "learning_rate": 5e-06,
295
+ "loss": 0.4531,
296
+ "step": 410
297
+ },
298
+ {
299
+ "epoch": 0.9979160464423936,
300
+ "eval_loss": 0.45424818992614746,
301
+ "eval_runtime": 286.0837,
302
+ "eval_samples_per_second": 39.551,
303
+ "eval_steps_per_second": 0.619,
304
+ "step": 419
305
+ },
306
+ {
307
+ "epoch": 1.0020839535576065,
308
+ "grad_norm": 1.5666613895216177,
309
+ "learning_rate": 5e-06,
310
+ "loss": 0.4859,
311
+ "step": 420
312
+ },
313
+ {
314
+ "epoch": 1.025900565644537,
315
+ "grad_norm": 0.8242103660038579,
316
+ "learning_rate": 5e-06,
317
+ "loss": 0.4102,
318
+ "step": 430
319
+ },
320
+ {
321
+ "epoch": 1.0497171777314678,
322
+ "grad_norm": 0.7452119219276494,
323
+ "learning_rate": 5e-06,
324
+ "loss": 0.4092,
325
+ "step": 440
326
+ },
327
+ {
328
+ "epoch": 1.0735337898183983,
329
+ "grad_norm": 0.6691470793599864,
330
+ "learning_rate": 5e-06,
331
+ "loss": 0.4052,
332
+ "step": 450
333
+ },
334
+ {
335
+ "epoch": 1.097350401905329,
336
+ "grad_norm": 0.7840952133237613,
337
+ "learning_rate": 5e-06,
338
+ "loss": 0.4105,
339
+ "step": 460
340
+ },
341
+ {
342
+ "epoch": 1.1211670139922596,
343
+ "grad_norm": 0.6028706220060683,
344
+ "learning_rate": 5e-06,
345
+ "loss": 0.4084,
346
+ "step": 470
347
+ },
348
+ {
349
+ "epoch": 1.1449836260791901,
350
+ "grad_norm": 0.6295729577533348,
351
+ "learning_rate": 5e-06,
352
+ "loss": 0.411,
353
+ "step": 480
354
+ },
355
+ {
356
+ "epoch": 1.1688002381661209,
357
+ "grad_norm": 0.5614428880763974,
358
+ "learning_rate": 5e-06,
359
+ "loss": 0.4035,
360
+ "step": 490
361
+ },
362
+ {
363
+ "epoch": 1.1926168502530514,
364
+ "grad_norm": 0.7354574620421201,
365
+ "learning_rate": 5e-06,
366
+ "loss": 0.4128,
367
+ "step": 500
368
+ },
369
+ {
370
+ "epoch": 1.2164334623399822,
371
+ "grad_norm": 0.6533242864715311,
372
+ "learning_rate": 5e-06,
373
+ "loss": 0.409,
374
+ "step": 510
375
+ },
376
+ {
377
+ "epoch": 1.2402500744269127,
378
+ "grad_norm": 0.5341268826400728,
379
+ "learning_rate": 5e-06,
380
+ "loss": 0.4069,
381
+ "step": 520
382
+ },
383
+ {
384
+ "epoch": 1.2640666865138435,
385
+ "grad_norm": 0.595790957629028,
386
+ "learning_rate": 5e-06,
387
+ "loss": 0.403,
388
+ "step": 530
389
+ },
390
+ {
391
+ "epoch": 1.287883298600774,
392
+ "grad_norm": 0.5757459346899625,
393
+ "learning_rate": 5e-06,
394
+ "loss": 0.4043,
395
+ "step": 540
396
+ },
397
+ {
398
+ "epoch": 1.3116999106877048,
399
+ "grad_norm": 0.6014475432063849,
400
+ "learning_rate": 5e-06,
401
+ "loss": 0.4063,
402
+ "step": 550
403
+ },
404
+ {
405
+ "epoch": 1.3355165227746353,
406
+ "grad_norm": 0.6816804823593294,
407
+ "learning_rate": 5e-06,
408
+ "loss": 0.4054,
409
+ "step": 560
410
+ },
411
+ {
412
+ "epoch": 1.3593331348615658,
413
+ "grad_norm": 0.8369295577084369,
414
+ "learning_rate": 5e-06,
415
+ "loss": 0.4038,
416
+ "step": 570
417
+ },
418
+ {
419
+ "epoch": 1.3831497469484966,
420
+ "grad_norm": 0.5774046033105074,
421
+ "learning_rate": 5e-06,
422
+ "loss": 0.4041,
423
+ "step": 580
424
+ },
425
+ {
426
+ "epoch": 1.4069663590354273,
427
+ "grad_norm": 0.6188803841652969,
428
+ "learning_rate": 5e-06,
429
+ "loss": 0.4064,
430
+ "step": 590
431
+ },
432
+ {
433
+ "epoch": 1.4307829711223579,
434
+ "grad_norm": 0.6067419957695976,
435
+ "learning_rate": 5e-06,
436
+ "loss": 0.3998,
437
+ "step": 600
438
+ },
439
+ {
440
+ "epoch": 1.4545995832092884,
441
+ "grad_norm": 0.7114287172431103,
442
+ "learning_rate": 5e-06,
443
+ "loss": 0.4024,
444
+ "step": 610
445
+ },
446
+ {
447
+ "epoch": 1.4784161952962191,
448
+ "grad_norm": 0.6126178425690488,
449
+ "learning_rate": 5e-06,
450
+ "loss": 0.4026,
451
+ "step": 620
452
+ },
453
+ {
454
+ "epoch": 1.50223280738315,
455
+ "grad_norm": 0.5281312737413801,
456
+ "learning_rate": 5e-06,
457
+ "loss": 0.4027,
458
+ "step": 630
459
+ },
460
+ {
461
+ "epoch": 1.5260494194700804,
462
+ "grad_norm": 0.7216431853686169,
463
+ "learning_rate": 5e-06,
464
+ "loss": 0.4054,
465
+ "step": 640
466
+ },
467
+ {
468
+ "epoch": 1.549866031557011,
469
+ "grad_norm": 0.6037953128159437,
470
+ "learning_rate": 5e-06,
471
+ "loss": 0.402,
472
+ "step": 650
473
+ },
474
+ {
475
+ "epoch": 1.5736826436439415,
476
+ "grad_norm": 0.7237089947239425,
477
+ "learning_rate": 5e-06,
478
+ "loss": 0.3977,
479
+ "step": 660
480
+ },
481
+ {
482
+ "epoch": 1.5974992557308723,
483
+ "grad_norm": 0.6225585837795767,
484
+ "learning_rate": 5e-06,
485
+ "loss": 0.3986,
486
+ "step": 670
487
+ },
488
+ {
489
+ "epoch": 1.621315867817803,
490
+ "grad_norm": 0.6880389181478974,
491
+ "learning_rate": 5e-06,
492
+ "loss": 0.4037,
493
+ "step": 680
494
+ },
495
+ {
496
+ "epoch": 1.6451324799047335,
497
+ "grad_norm": 0.6033306584926863,
498
+ "learning_rate": 5e-06,
499
+ "loss": 0.4022,
500
+ "step": 690
501
+ },
502
+ {
503
+ "epoch": 1.668949091991664,
504
+ "grad_norm": 0.635681818377172,
505
+ "learning_rate": 5e-06,
506
+ "loss": 0.4068,
507
+ "step": 700
508
+ },
509
+ {
510
+ "epoch": 1.6927657040785948,
511
+ "grad_norm": 0.6212447265932675,
512
+ "learning_rate": 5e-06,
513
+ "loss": 0.4003,
514
+ "step": 710
515
+ },
516
+ {
517
+ "epoch": 1.7165823161655256,
518
+ "grad_norm": 0.8760196584827,
519
+ "learning_rate": 5e-06,
520
+ "loss": 0.3978,
521
+ "step": 720
522
+ },
523
+ {
524
+ "epoch": 1.7403989282524561,
525
+ "grad_norm": 0.6660720739374277,
526
+ "learning_rate": 5e-06,
527
+ "loss": 0.4004,
528
+ "step": 730
529
+ },
530
+ {
531
+ "epoch": 1.7642155403393867,
532
+ "grad_norm": 0.6326799745725241,
533
+ "learning_rate": 5e-06,
534
+ "loss": 0.4007,
535
+ "step": 740
536
+ },
537
+ {
538
+ "epoch": 1.7880321524263174,
539
+ "grad_norm": 0.6030163934725129,
540
+ "learning_rate": 5e-06,
541
+ "loss": 0.4011,
542
+ "step": 750
543
+ },
544
+ {
545
+ "epoch": 1.811848764513248,
546
+ "grad_norm": 0.6024754641341044,
547
+ "learning_rate": 5e-06,
548
+ "loss": 0.3941,
549
+ "step": 760
550
+ },
551
+ {
552
+ "epoch": 1.8356653766001787,
553
+ "grad_norm": 0.5552872150124285,
554
+ "learning_rate": 5e-06,
555
+ "loss": 0.4042,
556
+ "step": 770
557
+ },
558
+ {
559
+ "epoch": 1.8594819886871092,
560
+ "grad_norm": 0.5054694344878702,
561
+ "learning_rate": 5e-06,
562
+ "loss": 0.4029,
563
+ "step": 780
564
+ },
565
+ {
566
+ "epoch": 1.8832986007740398,
567
+ "grad_norm": 0.7189726869611571,
568
+ "learning_rate": 5e-06,
569
+ "loss": 0.3992,
570
+ "step": 790
571
+ },
572
+ {
573
+ "epoch": 1.9071152128609705,
574
+ "grad_norm": 0.5495927800255059,
575
+ "learning_rate": 5e-06,
576
+ "loss": 0.3969,
577
+ "step": 800
578
+ },
579
+ {
580
+ "epoch": 1.9309318249479013,
581
+ "grad_norm": 0.5757953006798125,
582
+ "learning_rate": 5e-06,
583
+ "loss": 0.3995,
584
+ "step": 810
585
+ },
586
+ {
587
+ "epoch": 1.9547484370348318,
588
+ "grad_norm": 0.5635263667269701,
589
+ "learning_rate": 5e-06,
590
+ "loss": 0.402,
591
+ "step": 820
592
+ },
593
+ {
594
+ "epoch": 1.9785650491217623,
595
+ "grad_norm": 0.6248661080671392,
596
+ "learning_rate": 5e-06,
597
+ "loss": 0.3953,
598
+ "step": 830
599
+ },
600
+ {
601
+ "epoch": 1.997618338791307,
602
+ "eval_loss": 0.43693724274635315,
603
+ "eval_runtime": 286.32,
604
+ "eval_samples_per_second": 39.519,
605
+ "eval_steps_per_second": 0.618,
606
+ "step": 838
607
+ },
608
+ {
609
+ "epoch": 2.004167907115213,
610
+ "grad_norm": 0.8252136402016477,
611
+ "learning_rate": 5e-06,
612
+ "loss": 0.4321,
613
+ "step": 840
614
+ },
615
+ {
616
+ "epoch": 2.0279845192021435,
617
+ "grad_norm": 0.8217614405447083,
618
+ "learning_rate": 5e-06,
619
+ "loss": 0.3538,
620
+ "step": 850
621
+ },
622
+ {
623
+ "epoch": 2.051801131289074,
624
+ "grad_norm": 0.597126610910981,
625
+ "learning_rate": 5e-06,
626
+ "loss": 0.353,
627
+ "step": 860
628
+ },
629
+ {
630
+ "epoch": 2.0756177433760046,
631
+ "grad_norm": 0.5891946978445731,
632
+ "learning_rate": 5e-06,
633
+ "loss": 0.3476,
634
+ "step": 870
635
+ },
636
+ {
637
+ "epoch": 2.0994343554629356,
638
+ "grad_norm": 0.7349012751558448,
639
+ "learning_rate": 5e-06,
640
+ "loss": 0.3513,
641
+ "step": 880
642
+ },
643
+ {
644
+ "epoch": 2.123250967549866,
645
+ "grad_norm": 0.6562744496253617,
646
+ "learning_rate": 5e-06,
647
+ "loss": 0.352,
648
+ "step": 890
649
+ },
650
+ {
651
+ "epoch": 2.1470675796367966,
652
+ "grad_norm": 0.7348610922149896,
653
+ "learning_rate": 5e-06,
654
+ "loss": 0.3509,
655
+ "step": 900
656
+ },
657
+ {
658
+ "epoch": 2.170884191723727,
659
+ "grad_norm": 0.5550207440255053,
660
+ "learning_rate": 5e-06,
661
+ "loss": 0.3517,
662
+ "step": 910
663
+ },
664
+ {
665
+ "epoch": 2.194700803810658,
666
+ "grad_norm": 0.7725877741803382,
667
+ "learning_rate": 5e-06,
668
+ "loss": 0.3547,
669
+ "step": 920
670
+ },
671
+ {
672
+ "epoch": 2.2185174158975887,
673
+ "grad_norm": 0.7254370201802991,
674
+ "learning_rate": 5e-06,
675
+ "loss": 0.3513,
676
+ "step": 930
677
+ },
678
+ {
679
+ "epoch": 2.242334027984519,
680
+ "grad_norm": 0.7099191501450315,
681
+ "learning_rate": 5e-06,
682
+ "loss": 0.3534,
683
+ "step": 940
684
+ },
685
+ {
686
+ "epoch": 2.2661506400714497,
687
+ "grad_norm": 0.6217727164056416,
688
+ "learning_rate": 5e-06,
689
+ "loss": 0.3525,
690
+ "step": 950
691
+ },
692
+ {
693
+ "epoch": 2.2899672521583803,
694
+ "grad_norm": 0.7759463308703756,
695
+ "learning_rate": 5e-06,
696
+ "loss": 0.3561,
697
+ "step": 960
698
+ },
699
+ {
700
+ "epoch": 2.3137838642453112,
701
+ "grad_norm": 0.8023649875600256,
702
+ "learning_rate": 5e-06,
703
+ "loss": 0.3574,
704
+ "step": 970
705
+ },
706
+ {
707
+ "epoch": 2.3376004763322418,
708
+ "grad_norm": 0.6363742954254966,
709
+ "learning_rate": 5e-06,
710
+ "loss": 0.3568,
711
+ "step": 980
712
+ },
713
+ {
714
+ "epoch": 2.3614170884191723,
715
+ "grad_norm": 0.6522324898510586,
716
+ "learning_rate": 5e-06,
717
+ "loss": 0.3545,
718
+ "step": 990
719
+ },
720
+ {
721
+ "epoch": 2.385233700506103,
722
+ "grad_norm": 0.6816643137692279,
723
+ "learning_rate": 5e-06,
724
+ "loss": 0.3503,
725
+ "step": 1000
726
+ },
727
+ {
728
+ "epoch": 2.409050312593034,
729
+ "grad_norm": 0.616355784268503,
730
+ "learning_rate": 5e-06,
731
+ "loss": 0.3539,
732
+ "step": 1010
733
+ },
734
+ {
735
+ "epoch": 2.4328669246799643,
736
+ "grad_norm": 0.8616495256742727,
737
+ "learning_rate": 5e-06,
738
+ "loss": 0.3567,
739
+ "step": 1020
740
+ },
741
+ {
742
+ "epoch": 2.456683536766895,
743
+ "grad_norm": 0.6281577543660147,
744
+ "learning_rate": 5e-06,
745
+ "loss": 0.355,
746
+ "step": 1030
747
+ },
748
+ {
749
+ "epoch": 2.4805001488538254,
750
+ "grad_norm": 0.5605730765386269,
751
+ "learning_rate": 5e-06,
752
+ "loss": 0.3551,
753
+ "step": 1040
754
+ },
755
+ {
756
+ "epoch": 2.504316760940756,
757
+ "grad_norm": 0.6217913711967288,
758
+ "learning_rate": 5e-06,
759
+ "loss": 0.3552,
760
+ "step": 1050
761
+ },
762
+ {
763
+ "epoch": 2.528133373027687,
764
+ "grad_norm": 0.6599048002895368,
765
+ "learning_rate": 5e-06,
766
+ "loss": 0.3532,
767
+ "step": 1060
768
+ },
769
+ {
770
+ "epoch": 2.5519499851146175,
771
+ "grad_norm": 0.7340133449366499,
772
+ "learning_rate": 5e-06,
773
+ "loss": 0.3546,
774
+ "step": 1070
775
+ },
776
+ {
777
+ "epoch": 2.575766597201548,
778
+ "grad_norm": 0.7470486435804491,
779
+ "learning_rate": 5e-06,
780
+ "loss": 0.356,
781
+ "step": 1080
782
+ },
783
+ {
784
+ "epoch": 2.599583209288479,
785
+ "grad_norm": 0.5596330861321896,
786
+ "learning_rate": 5e-06,
787
+ "loss": 0.3593,
788
+ "step": 1090
789
+ },
790
+ {
791
+ "epoch": 2.6233998213754095,
792
+ "grad_norm": 0.655258490698248,
793
+ "learning_rate": 5e-06,
794
+ "loss": 0.3574,
795
+ "step": 1100
796
+ },
797
+ {
798
+ "epoch": 2.64721643346234,
799
+ "grad_norm": 0.6634626122990018,
800
+ "learning_rate": 5e-06,
801
+ "loss": 0.3582,
802
+ "step": 1110
803
+ },
804
+ {
805
+ "epoch": 2.6710330455492706,
806
+ "grad_norm": 0.6875550380429598,
807
+ "learning_rate": 5e-06,
808
+ "loss": 0.3567,
809
+ "step": 1120
810
+ },
811
+ {
812
+ "epoch": 2.694849657636201,
813
+ "grad_norm": 0.6775433440528874,
814
+ "learning_rate": 5e-06,
815
+ "loss": 0.3574,
816
+ "step": 1130
817
+ },
818
+ {
819
+ "epoch": 2.7186662697231316,
820
+ "grad_norm": 0.6304790171059349,
821
+ "learning_rate": 5e-06,
822
+ "loss": 0.3547,
823
+ "step": 1140
824
+ },
825
+ {
826
+ "epoch": 2.7424828818100626,
827
+ "grad_norm": 0.6868875377983221,
828
+ "learning_rate": 5e-06,
829
+ "loss": 0.3547,
830
+ "step": 1150
831
+ },
832
+ {
833
+ "epoch": 2.766299493896993,
834
+ "grad_norm": 0.8782114193263775,
835
+ "learning_rate": 5e-06,
836
+ "loss": 0.3586,
837
+ "step": 1160
838
+ },
839
+ {
840
+ "epoch": 2.7901161059839237,
841
+ "grad_norm": 0.6003211481767972,
842
+ "learning_rate": 5e-06,
843
+ "loss": 0.3498,
844
+ "step": 1170
845
+ },
846
+ {
847
+ "epoch": 2.8139327180708547,
848
+ "grad_norm": 0.6181739427734269,
849
+ "learning_rate": 5e-06,
850
+ "loss": 0.3611,
851
+ "step": 1180
852
+ },
853
+ {
854
+ "epoch": 2.837749330157785,
855
+ "grad_norm": 0.5812515280780504,
856
+ "learning_rate": 5e-06,
857
+ "loss": 0.3593,
858
+ "step": 1190
859
+ },
860
+ {
861
+ "epoch": 2.8615659422447157,
862
+ "grad_norm": 0.6647885239726261,
863
+ "learning_rate": 5e-06,
864
+ "loss": 0.357,
865
+ "step": 1200
866
+ },
867
+ {
868
+ "epoch": 2.8853825543316463,
869
+ "grad_norm": 1.1264669697440912,
870
+ "learning_rate": 5e-06,
871
+ "loss": 0.356,
872
+ "step": 1210
873
+ },
874
+ {
875
+ "epoch": 2.909199166418577,
876
+ "grad_norm": 0.7668811596236461,
877
+ "learning_rate": 5e-06,
878
+ "loss": 0.357,
879
+ "step": 1220
880
+ },
881
+ {
882
+ "epoch": 2.9330157785055073,
883
+ "grad_norm": 0.5719772302769721,
884
+ "learning_rate": 5e-06,
885
+ "loss": 0.3531,
886
+ "step": 1230
887
+ },
888
+ {
889
+ "epoch": 2.9568323905924383,
890
+ "grad_norm": 0.6110386348906062,
891
+ "learning_rate": 5e-06,
892
+ "loss": 0.3583,
893
+ "step": 1240
894
+ },
895
+ {
896
+ "epoch": 2.980649002679369,
897
+ "grad_norm": 0.5843814268513303,
898
+ "learning_rate": 5e-06,
899
+ "loss": 0.3539,
900
+ "step": 1250
901
+ },
902
+ {
903
+ "epoch": 2.9973206311402203,
904
+ "eval_loss": 0.4348444640636444,
905
+ "eval_runtime": 294.9931,
906
+ "eval_samples_per_second": 38.357,
907
+ "eval_steps_per_second": 0.6,
908
+ "step": 1257
909
+ },
910
+ {
911
+ "epoch": 2.9973206311402203,
912
+ "step": 1257,
913
+ "total_flos": 2105312437862400.0,
914
+ "train_loss": 0.41854361285268077,
915
+ "train_runtime": 41355.0058,
916
+ "train_samples_per_second": 15.594,
917
+ "train_steps_per_second": 0.03
918
+ }
919
+ ],
920
+ "logging_steps": 10,
921
+ "max_steps": 1257,
922
+ "num_input_tokens_seen": 0,
923
+ "num_train_epochs": 3,
924
+ "save_steps": 500,
925
+ "stateful_callbacks": {
926
+ "TrainerControl": {
927
+ "args": {
928
+ "should_epoch_stop": false,
929
+ "should_evaluate": false,
930
+ "should_log": false,
931
+ "should_save": true,
932
+ "should_training_stop": true
933
+ },
934
+ "attributes": {}
935
+ }
936
+ },
937
+ "total_flos": 2105312437862400.0,
938
+ "train_batch_size": 8,
939
+ "trial_name": null,
940
+ "trial_params": null
941
+ }
training_eval_loss.png ADDED
training_loss.png ADDED