martimfasantos commited on
Commit
a776c48
1 Parent(s): 558e792

Model save

Browse files
README.md CHANGED
@@ -2,15 +2,11 @@
2
  license: apache-2.0
3
  base_model: TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T
4
  tags:
5
- - alignment-handbook
6
- - trl
7
- - sft
8
- - generated_from_trainer
9
  - trl
10
  - sft
11
  - generated_from_trainer
12
  datasets:
13
- - haoranxu/ALMA-Human-Parallel
14
  model-index:
15
  - name: tinyllama-1.1b-mt-sft-full
16
  results: []
@@ -21,9 +17,9 @@ should probably proofread and complete it, then remove this comment. -->
21
 
22
  # tinyllama-1.1b-mt-sft-full
23
 
24
- This model is a fine-tuned version of [TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T](https://huggingface.co/TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T) on the haoranxu/ALMA-Human-Parallel dataset.
25
  It achieves the following results on the evaluation set:
26
- - Loss: 1.6971
27
 
28
  ## Model description
29
 
@@ -43,12 +39,12 @@ More information needed
43
 
44
  The following hyperparameters were used during training:
45
  - learning_rate: 2e-05
46
- - train_batch_size: 8
47
  - eval_batch_size: 8
48
  - seed: 42
49
  - distributed_type: multi-GPU
50
- - gradient_accumulation_steps: 4
51
- - total_train_batch_size: 32
52
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
53
  - lr_scheduler_type: cosine
54
  - lr_scheduler_warmup_ratio: 0.1
@@ -56,10 +52,10 @@ The following hyperparameters were used during training:
56
 
57
  ### Training results
58
 
59
- | Training Loss | Epoch | Step | Validation Loss |
60
- |:-------------:|:------:|:----:|:---------------:|
61
- | 1.601 | 0.9959 | 184 | 1.6972 |
62
- | 1.5526 | 1.9919 | 368 | 1.6971 |
63
 
64
 
65
  ### Framework versions
 
2
  license: apache-2.0
3
  base_model: TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T
4
  tags:
 
 
 
 
5
  - trl
6
  - sft
7
  - generated_from_trainer
8
  datasets:
9
+ - generator
10
  model-index:
11
  - name: tinyllama-1.1b-mt-sft-full
12
  results: []
 
17
 
18
  # tinyllama-1.1b-mt-sft-full
19
 
20
+ This model is a fine-tuned version of [TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T](https://huggingface.co/TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T) on the generator dataset.
21
  It achieves the following results on the evaluation set:
22
+ - Loss: 1.6920
23
 
24
  ## Model description
25
 
 
39
 
40
  The following hyperparameters were used during training:
41
  - learning_rate: 2e-05
42
+ - train_batch_size: 4
43
  - eval_batch_size: 8
44
  - seed: 42
45
  - distributed_type: multi-GPU
46
+ - gradient_accumulation_steps: 2
47
+ - total_train_batch_size: 8
48
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
49
  - lr_scheduler_type: cosine
50
  - lr_scheduler_warmup_ratio: 0.1
 
52
 
53
  ### Training results
54
 
55
+ | Training Loss | Epoch | Step | Validation Loss |
56
+ |:-------------:|:-----:|:----:|:---------------:|
57
+ | 1.5838 | 1.0 | 739 | 1.6892 |
58
+ | 1.5051 | 2.0 | 1478 | 1.6920 |
59
 
60
 
61
  ### Framework versions
all_results.json CHANGED
@@ -1,14 +1,9 @@
1
  {
2
- "epoch": 1.9918809201623815,
3
- "eval_loss": 1.697079062461853,
4
- "eval_runtime": 28.4431,
5
- "eval_samples": 8016,
6
- "eval_samples_per_second": 15.294,
7
- "eval_steps_per_second": 1.934,
8
- "total_flos": 1.4968483831454106e+17,
9
- "train_loss": 1.6339908747569374,
10
- "train_runtime": 3216.6871,
11
  "train_samples": 117404,
12
- "train_samples_per_second": 3.675,
13
- "train_steps_per_second": 0.114
14
  }
 
1
  {
2
+ "epoch": 2.0,
3
+ "total_flos": 1.502823063832494e+17,
4
+ "train_loss": 1.5822677802652403,
5
+ "train_runtime": 3974.9532,
 
 
 
 
 
6
  "train_samples": 117404,
7
+ "train_samples_per_second": 2.974,
8
+ "train_steps_per_second": 0.372
9
  }
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:8c9e4694cfd0a160e827ab9195920fa7e575c038f71807d772c119ec1a088bfd
3
  size 2200119864
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:304ebde9357f88524cacc379e54607dd078ba448abd4d2c95eba37ed9410ffe3
3
  size 2200119864
runs/May30_13-37-43_poseidon/events.out.tfevents.1717076296.poseidon.2020252.0 CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:e4994d7216f1238d76444860608ec7ca8e4b9af1ce9769bf101d6487a879003b
3
- size 64485
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b947905388b1a520b480679f89094b007b4fd81de4f5ebd5b794bee9b4bd2007
3
+ size 68275
train_results.json CHANGED
@@ -1,9 +1,9 @@
1
  {
2
- "epoch": 1.9918809201623815,
3
- "total_flos": 1.4968483831454106e+17,
4
- "train_loss": 1.6339908747569374,
5
- "train_runtime": 3216.6871,
6
  "train_samples": 117404,
7
- "train_samples_per_second": 3.675,
8
- "train_steps_per_second": 0.114
9
  }
 
1
  {
2
+ "epoch": 2.0,
3
+ "total_flos": 1.502823063832494e+17,
4
+ "train_loss": 1.5822677802652403,
5
+ "train_runtime": 3974.9532,
6
  "train_samples": 117404,
7
+ "train_samples_per_second": 2.974,
8
+ "train_steps_per_second": 0.372
9
  }
trainer_state.json CHANGED
@@ -1,559 +1,2113 @@
1
  {
2
  "best_metric": null,
3
  "best_model_checkpoint": null,
4
- "epoch": 1.9918809201623815,
5
  "eval_steps": 500,
6
- "global_step": 368,
7
  "is_hyper_param_search": false,
8
  "is_local_process_zero": true,
9
  "is_world_process_zero": true,
10
  "log_history": [
11
  {
12
- "epoch": 0.005412719891745603,
13
- "grad_norm": 23.5,
14
- "learning_rate": 5.405405405405406e-07,
15
- "loss": 2.2045,
16
  "step": 1
17
  },
18
  {
19
- "epoch": 0.02706359945872801,
20
- "grad_norm": 19.375,
21
- "learning_rate": 2.702702702702703e-06,
22
- "loss": 2.2384,
23
  "step": 5
24
  },
25
  {
26
- "epoch": 0.05412719891745602,
27
- "grad_norm": 6.6875,
28
- "learning_rate": 5.405405405405406e-06,
29
- "loss": 2.1809,
30
  "step": 10
31
  },
32
  {
33
- "epoch": 0.08119079837618404,
34
- "grad_norm": 3.921875,
35
- "learning_rate": 8.108108108108109e-06,
36
- "loss": 2.0942,
37
  "step": 15
38
  },
39
  {
40
- "epoch": 0.10825439783491204,
41
- "grad_norm": 2.765625,
42
- "learning_rate": 1.0810810810810812e-05,
43
- "loss": 2.0225,
44
  "step": 20
45
  },
46
  {
47
- "epoch": 0.13531799729364005,
48
- "grad_norm": 3.40625,
49
- "learning_rate": 1.3513513513513515e-05,
50
- "loss": 1.9498,
51
  "step": 25
52
  },
53
  {
54
- "epoch": 0.16238159675236807,
55
- "grad_norm": 2.703125,
56
- "learning_rate": 1.6216216216216218e-05,
57
- "loss": 1.8531,
58
  "step": 30
59
  },
60
  {
61
- "epoch": 0.18944519621109607,
62
- "grad_norm": 1.78125,
63
- "learning_rate": 1.891891891891892e-05,
64
- "loss": 1.7903,
65
  "step": 35
66
  },
67
  {
68
- "epoch": 0.2165087956698241,
69
- "grad_norm": 1.703125,
70
- "learning_rate": 1.9995946530314384e-05,
71
- "loss": 1.7538,
72
  "step": 40
73
  },
74
  {
75
- "epoch": 0.2435723951285521,
76
- "grad_norm": 1.5078125,
77
- "learning_rate": 1.9971187226043746e-05,
78
- "loss": 1.7264,
79
  "step": 45
80
  },
81
  {
82
- "epoch": 0.2706359945872801,
83
- "grad_norm": 1.3515625,
84
- "learning_rate": 1.9923976226947417e-05,
85
- "loss": 1.6995,
86
  "step": 50
87
  },
88
  {
89
- "epoch": 0.2976995940460081,
90
- "grad_norm": 1.28125,
91
- "learning_rate": 1.985441983600819e-05,
92
- "loss": 1.6901,
93
  "step": 55
94
  },
95
  {
96
- "epoch": 0.32476319350473615,
97
- "grad_norm": 1.34375,
98
- "learning_rate": 1.9762674670369757e-05,
99
- "loss": 1.6784,
100
  "step": 60
101
  },
102
  {
103
- "epoch": 0.35182679296346414,
104
- "grad_norm": 1.2890625,
105
- "learning_rate": 1.9648947308688594e-05,
106
- "loss": 1.6738,
107
  "step": 65
108
  },
109
  {
110
- "epoch": 0.37889039242219213,
111
- "grad_norm": 1.328125,
112
- "learning_rate": 1.9513493825989664e-05,
113
- "loss": 1.6719,
114
  "step": 70
115
  },
116
  {
117
- "epoch": 0.4059539918809202,
118
- "grad_norm": 1.2734375,
119
- "learning_rate": 1.9356619217073252e-05,
120
- "loss": 1.6617,
121
  "step": 75
122
  },
123
  {
124
- "epoch": 0.4330175913396482,
125
- "grad_norm": 1.2734375,
126
- "learning_rate": 1.917867670977126e-05,
127
- "loss": 1.6447,
128
  "step": 80
129
  },
130
  {
131
- "epoch": 0.46008119079837617,
132
- "grad_norm": 1.296875,
133
- "learning_rate": 1.8980066969599216e-05,
134
- "loss": 1.6337,
135
  "step": 85
136
  },
137
  {
138
- "epoch": 0.4871447902571042,
139
- "grad_norm": 1.25,
140
- "learning_rate": 1.8761237197594945e-05,
141
- "loss": 1.6549,
142
  "step": 90
143
  },
144
  {
145
- "epoch": 0.5142083897158322,
146
- "grad_norm": 1.25,
147
- "learning_rate": 1.852268012337514e-05,
148
- "loss": 1.6334,
149
  "step": 95
150
  },
151
  {
152
- "epoch": 0.5412719891745602,
153
- "grad_norm": 1.265625,
154
- "learning_rate": 1.8264932895677195e-05,
155
- "loss": 1.6276,
156
  "step": 100
157
  },
158
  {
159
- "epoch": 0.5683355886332883,
160
- "grad_norm": 1.2421875,
161
- "learning_rate": 1.798857587288445e-05,
162
- "loss": 1.6326,
163
  "step": 105
164
  },
165
  {
166
- "epoch": 0.5953991880920162,
167
- "grad_norm": 1.203125,
168
- "learning_rate": 1.769423131625808e-05,
169
- "loss": 1.6334,
170
  "step": 110
171
  },
172
  {
173
- "epoch": 0.6224627875507442,
174
- "grad_norm": 1.234375,
175
- "learning_rate": 1.738256198881809e-05,
176
- "loss": 1.6327,
177
  "step": 115
178
  },
179
  {
180
- "epoch": 0.6495263870094723,
181
- "grad_norm": 1.3125,
182
- "learning_rate": 1.7054269663028232e-05,
183
- "loss": 1.6271,
184
  "step": 120
185
  },
186
  {
187
- "epoch": 0.6765899864682002,
188
- "grad_norm": 1.25,
189
- "learning_rate": 1.6710093540645056e-05,
190
- "loss": 1.6247,
191
  "step": 125
192
  },
193
  {
194
- "epoch": 0.7036535859269283,
195
- "grad_norm": 1.296875,
196
- "learning_rate": 1.6350808588288964e-05,
197
- "loss": 1.6255,
198
  "step": 130
199
  },
200
  {
201
- "epoch": 0.7307171853856563,
202
- "grad_norm": 1.234375,
203
- "learning_rate": 1.597722379248512e-05,
204
- "loss": 1.6155,
205
  "step": 135
206
  },
207
  {
208
- "epoch": 0.7577807848443843,
209
- "grad_norm": 1.2421875,
210
- "learning_rate": 1.559018033810316e-05,
211
- "loss": 1.6162,
212
  "step": 140
213
  },
214
  {
215
- "epoch": 0.7848443843031123,
216
- "grad_norm": 1.21875,
217
- "learning_rate": 1.5190549714297303e-05,
218
- "loss": 1.6081,
219
  "step": 145
220
  },
221
  {
222
- "epoch": 0.8119079837618404,
223
- "grad_norm": 1.21875,
224
- "learning_rate": 1.4779231752211546e-05,
225
- "loss": 1.6031,
226
  "step": 150
227
  },
228
  {
229
- "epoch": 0.8389715832205683,
230
- "grad_norm": 1.28125,
231
- "learning_rate": 1.4357152598868478e-05,
232
- "loss": 1.6155,
233
  "step": 155
234
  },
235
  {
236
- "epoch": 0.8660351826792964,
237
- "grad_norm": 1.2109375,
238
- "learning_rate": 1.3925262631803722e-05,
239
- "loss": 1.6039,
240
  "step": 160
241
  },
242
  {
243
- "epoch": 0.8930987821380244,
244
- "grad_norm": 1.2265625,
245
- "learning_rate": 1.3484534319141592e-05,
246
- "loss": 1.608,
247
  "step": 165
248
  },
249
  {
250
- "epoch": 0.9201623815967523,
251
- "grad_norm": 1.1875,
252
- "learning_rate": 1.303596002993028e-05,
253
- "loss": 1.6036,
254
  "step": 170
255
  },
256
  {
257
- "epoch": 0.9472259810554804,
258
- "grad_norm": 1.2578125,
259
- "learning_rate": 1.2580549799667034e-05,
260
- "loss": 1.6157,
261
  "step": 175
262
  },
263
  {
264
- "epoch": 0.9742895805142084,
265
- "grad_norm": 1.2734375,
266
- "learning_rate": 1.2119329056044533e-05,
267
- "loss": 1.601,
268
  "step": 180
269
  },
270
  {
271
- "epoch": 0.9959404600811907,
272
- "eval_loss": 1.6971594095230103,
273
- "eval_runtime": 28.8094,
274
- "eval_samples_per_second": 15.099,
275
- "eval_steps_per_second": 1.909,
276
- "step": 184
277
- },
278
- {
279
- "epoch": 1.0013531799729365,
280
- "grad_norm": 1.21875,
281
- "learning_rate": 1.165333631003928e-05,
282
- "loss": 1.5923,
283
  "step": 185
284
  },
285
  {
286
- "epoch": 1.0284167794316643,
287
- "grad_norm": 1.2109375,
288
- "learning_rate": 1.1183620817540985e-05,
289
- "loss": 1.5652,
290
  "step": 190
291
  },
292
  {
293
- "epoch": 1.0554803788903924,
294
- "grad_norm": 1.2265625,
295
- "learning_rate": 1.0711240216788036e-05,
296
- "loss": 1.5483,
297
  "step": 195
298
  },
299
  {
300
- "epoch": 1.0825439783491204,
301
- "grad_norm": 1.2578125,
302
- "learning_rate": 1.0237258146928849e-05,
303
- "loss": 1.5504,
304
  "step": 200
305
  },
306
  {
307
- "epoch": 1.1096075778078485,
308
- "grad_norm": 1.21875,
309
- "learning_rate": 9.762741853071153e-06,
310
- "loss": 1.555,
311
  "step": 205
312
  },
313
  {
314
- "epoch": 1.1366711772665765,
315
- "grad_norm": 1.1796875,
316
- "learning_rate": 9.288759783211967e-06,
317
- "loss": 1.5659,
318
  "step": 210
319
  },
320
  {
321
- "epoch": 1.1637347767253043,
322
- "grad_norm": 1.2109375,
323
- "learning_rate": 8.81637918245902e-06,
324
- "loss": 1.5601,
325
  "step": 215
326
  },
327
  {
328
- "epoch": 1.1907983761840324,
329
- "grad_norm": 1.1640625,
330
- "learning_rate": 8.346663689960724e-06,
331
- "loss": 1.5516,
332
  "step": 220
333
  },
334
  {
335
- "epoch": 1.2178619756427604,
336
- "grad_norm": 1.171875,
337
- "learning_rate": 7.880670943955467e-06,
338
- "loss": 1.5542,
339
  "step": 225
340
  },
341
  {
342
- "epoch": 1.2449255751014885,
343
- "grad_norm": 1.203125,
344
- "learning_rate": 7.419450200332965e-06,
345
- "loss": 1.5491,
346
  "step": 230
347
  },
348
  {
349
- "epoch": 1.2719891745602165,
350
- "grad_norm": 1.1640625,
351
- "learning_rate": 6.964039970069722e-06,
352
- "loss": 1.5564,
353
  "step": 235
354
  },
355
  {
356
- "epoch": 1.2990527740189446,
357
- "grad_norm": 1.1796875,
358
- "learning_rate": 6.515465680858412e-06,
359
- "loss": 1.5584,
360
  "step": 240
361
  },
362
  {
363
- "epoch": 1.3261163734776726,
364
- "grad_norm": 1.21875,
365
- "learning_rate": 6.074737368196279e-06,
366
- "loss": 1.5534,
367
  "step": 245
368
  },
369
  {
370
- "epoch": 1.3531799729364005,
371
- "grad_norm": 1.2421875,
372
- "learning_rate": 5.642847401131526e-06,
373
- "loss": 1.5588,
374
  "step": 250
375
  },
376
  {
377
- "epoch": 1.3802435723951285,
378
- "grad_norm": 1.21875,
379
- "learning_rate": 5.220768247788458e-06,
380
- "loss": 1.552,
381
  "step": 255
382
  },
383
  {
384
- "epoch": 1.4073071718538566,
385
- "grad_norm": 1.1875,
386
- "learning_rate": 4.809450285702697e-06,
387
- "loss": 1.5462,
388
  "step": 260
389
  },
390
  {
391
- "epoch": 1.4343707713125846,
392
- "grad_norm": 1.203125,
393
- "learning_rate": 4.409819661896839e-06,
394
- "loss": 1.5623,
395
  "step": 265
396
  },
397
  {
398
- "epoch": 1.4614343707713127,
399
- "grad_norm": 1.2421875,
400
- "learning_rate": 4.022776207514885e-06,
401
- "loss": 1.5605,
402
  "step": 270
403
  },
404
  {
405
- "epoch": 1.4884979702300405,
406
- "grad_norm": 1.2265625,
407
- "learning_rate": 3.6491914117110405e-06,
408
- "loss": 1.5616,
409
  "step": 275
410
  },
411
  {
412
- "epoch": 1.5155615696887685,
413
- "grad_norm": 1.1875,
414
- "learning_rate": 3.2899064593549477e-06,
415
- "loss": 1.5578,
416
  "step": 280
417
  },
418
  {
419
- "epoch": 1.5426251691474966,
420
- "grad_norm": 1.1953125,
421
- "learning_rate": 2.945730336971767e-06,
422
- "loss": 1.5482,
423
  "step": 285
424
  },
425
  {
426
- "epoch": 1.5696887686062246,
427
- "grad_norm": 1.21875,
428
- "learning_rate": 2.6174380111819144e-06,
429
- "loss": 1.5559,
430
  "step": 290
431
  },
432
  {
433
- "epoch": 1.5967523680649527,
434
- "grad_norm": 1.21875,
435
- "learning_rate": 2.3057686837419246e-06,
436
- "loss": 1.5481,
437
  "step": 295
438
  },
439
  {
440
- "epoch": 1.6238159675236807,
441
- "grad_norm": 1.1640625,
442
- "learning_rate": 2.011424127115552e-06,
443
- "loss": 1.5412,
444
  "step": 300
445
  },
446
  {
447
- "epoch": 1.6508795669824088,
448
- "grad_norm": 1.234375,
449
- "learning_rate": 1.7350671043228072e-06,
450
- "loss": 1.5611,
451
  "step": 305
452
  },
453
  {
454
- "epoch": 1.6779431664411368,
455
- "grad_norm": 1.203125,
456
- "learning_rate": 1.4773198766248642e-06,
457
- "loss": 1.5669,
458
  "step": 310
459
  },
460
  {
461
- "epoch": 1.7050067658998647,
462
- "grad_norm": 1.2109375,
463
- "learning_rate": 1.2387628024050557e-06,
464
- "loss": 1.5515,
465
  "step": 315
466
  },
467
  {
468
- "epoch": 1.7320703653585927,
469
- "grad_norm": 1.2109375,
470
- "learning_rate": 1.0199330304007858e-06,
471
- "loss": 1.5633,
472
  "step": 320
473
  },
474
  {
475
- "epoch": 1.7591339648173205,
476
- "grad_norm": 1.234375,
477
- "learning_rate": 8.213232902287438e-07,
478
- "loss": 1.5542,
479
  "step": 325
480
  },
481
  {
482
- "epoch": 1.7861975642760486,
483
- "grad_norm": 1.2421875,
484
- "learning_rate": 6.433807829267491e-07,
485
- "loss": 1.5575,
486
  "step": 330
487
  },
488
  {
489
- "epoch": 1.8132611637347766,
490
- "grad_norm": 1.1953125,
491
- "learning_rate": 4.865061740103361e-07,
492
- "loss": 1.5532,
493
  "step": 335
494
  },
495
  {
496
- "epoch": 1.8403247631935047,
497
- "grad_norm": 1.21875,
498
- "learning_rate": 3.510526913114065e-07,
499
- "loss": 1.5686,
500
  "step": 340
501
  },
502
  {
503
- "epoch": 1.8673883626522327,
504
- "grad_norm": 1.1953125,
505
- "learning_rate": 2.3732532963024468e-07,
506
- "loss": 1.5484,
507
  "step": 345
508
  },
509
  {
510
- "epoch": 1.8944519621109608,
511
- "grad_norm": 1.1953125,
512
- "learning_rate": 1.4558016399181086e-07,
513
- "loss": 1.56,
514
  "step": 350
515
  },
516
  {
517
- "epoch": 1.9215155615696888,
518
- "grad_norm": 1.1953125,
519
- "learning_rate": 7.602377305258479e-08,
520
- "loss": 1.5479,
521
  "step": 355
522
  },
523
  {
524
- "epoch": 1.9485791610284169,
525
- "grad_norm": 1.1796875,
526
- "learning_rate": 2.8812773956256034e-08,
527
- "loss": 1.5456,
528
  "step": 360
529
  },
530
  {
531
- "epoch": 1.975642760487145,
532
- "grad_norm": 1.2265625,
533
- "learning_rate": 4.053469685617595e-09,
534
- "loss": 1.5526,
535
  "step": 365
536
  },
537
  {
538
- "epoch": 1.9918809201623815,
539
- "eval_loss": 1.697079062461853,
540
- "eval_runtime": 28.912,
541
- "eval_samples_per_second": 15.046,
542
- "eval_steps_per_second": 1.902,
543
- "step": 368
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
544
  },
545
  {
546
- "epoch": 1.9918809201623815,
547
- "step": 368,
548
- "total_flos": 1.4968483831454106e+17,
549
- "train_loss": 1.6339908747569374,
550
- "train_runtime": 3216.6871,
551
- "train_samples_per_second": 3.675,
552
- "train_steps_per_second": 0.114
553
  }
554
  ],
555
  "logging_steps": 5,
556
- "max_steps": 368,
557
  "num_input_tokens_seen": 0,
558
  "num_train_epochs": 2,
559
  "save_steps": 100,
@@ -569,8 +2123,8 @@
569
  "attributes": {}
570
  }
571
  },
572
- "total_flos": 1.4968483831454106e+17,
573
- "train_batch_size": 8,
574
  "trial_name": null,
575
  "trial_params": null
576
  }
 
1
  {
2
  "best_metric": null,
3
  "best_model_checkpoint": null,
4
+ "epoch": 2.0,
5
  "eval_steps": 500,
6
+ "global_step": 1478,
7
  "is_hyper_param_search": false,
8
  "is_local_process_zero": true,
9
  "is_world_process_zero": true,
10
  "log_history": [
11
  {
12
+ "epoch": 0.0013531799729364006,
13
+ "grad_norm": 22.625,
14
+ "learning_rate": 1.3513513513513515e-07,
15
+ "loss": 2.1988,
16
  "step": 1
17
  },
18
  {
19
+ "epoch": 0.006765899864682003,
20
+ "grad_norm": 21.875,
21
+ "learning_rate": 6.756756756756758e-07,
22
+ "loss": 2.2059,
23
  "step": 5
24
  },
25
  {
26
+ "epoch": 0.013531799729364006,
27
+ "grad_norm": 22.5,
28
+ "learning_rate": 1.3513513513513515e-06,
29
+ "loss": 2.2336,
30
  "step": 10
31
  },
32
  {
33
+ "epoch": 0.02029769959404601,
34
+ "grad_norm": 16.25,
35
+ "learning_rate": 2.0270270270270273e-06,
36
+ "loss": 2.244,
37
  "step": 15
38
  },
39
  {
40
+ "epoch": 0.02706359945872801,
41
+ "grad_norm": 8.6875,
42
+ "learning_rate": 2.702702702702703e-06,
43
+ "loss": 2.181,
44
  "step": 20
45
  },
46
  {
47
+ "epoch": 0.03382949932341001,
48
+ "grad_norm": 6.03125,
49
+ "learning_rate": 3.3783783783783788e-06,
50
+ "loss": 2.1568,
51
  "step": 25
52
  },
53
  {
54
+ "epoch": 0.04059539918809202,
55
+ "grad_norm": 5.15625,
56
+ "learning_rate": 4.0540540540540545e-06,
57
+ "loss": 2.1029,
58
  "step": 30
59
  },
60
  {
61
+ "epoch": 0.04736129905277402,
62
+ "grad_norm": 4.4375,
63
+ "learning_rate": 4.72972972972973e-06,
64
+ "loss": 2.0545,
65
  "step": 35
66
  },
67
  {
68
+ "epoch": 0.05412719891745602,
69
+ "grad_norm": 3.734375,
70
+ "learning_rate": 5.405405405405406e-06,
71
+ "loss": 2.0671,
72
  "step": 40
73
  },
74
  {
75
+ "epoch": 0.06089309878213803,
76
+ "grad_norm": 3.546875,
77
+ "learning_rate": 6.081081081081082e-06,
78
+ "loss": 2.0335,
79
  "step": 45
80
  },
81
  {
82
+ "epoch": 0.06765899864682003,
83
+ "grad_norm": 3.671875,
84
+ "learning_rate": 6.7567567567567575e-06,
85
+ "loss": 1.9884,
86
  "step": 50
87
  },
88
  {
89
+ "epoch": 0.07442489851150202,
90
+ "grad_norm": 4.84375,
91
+ "learning_rate": 7.4324324324324324e-06,
92
+ "loss": 1.9429,
93
  "step": 55
94
  },
95
  {
96
+ "epoch": 0.08119079837618404,
97
+ "grad_norm": 3.609375,
98
+ "learning_rate": 8.108108108108109e-06,
99
+ "loss": 1.8937,
100
  "step": 60
101
  },
102
  {
103
+ "epoch": 0.08795669824086604,
104
+ "grad_norm": 3.21875,
105
+ "learning_rate": 8.783783783783785e-06,
106
+ "loss": 1.8555,
107
  "step": 65
108
  },
109
  {
110
+ "epoch": 0.09472259810554803,
111
+ "grad_norm": 3.015625,
112
+ "learning_rate": 9.45945945945946e-06,
113
+ "loss": 1.8032,
114
  "step": 70
115
  },
116
  {
117
+ "epoch": 0.10148849797023005,
118
+ "grad_norm": 2.921875,
119
+ "learning_rate": 1.0135135135135136e-05,
120
+ "loss": 1.8064,
121
  "step": 75
122
  },
123
  {
124
+ "epoch": 0.10825439783491204,
125
+ "grad_norm": 2.90625,
126
+ "learning_rate": 1.0810810810810812e-05,
127
+ "loss": 1.8009,
128
  "step": 80
129
  },
130
  {
131
+ "epoch": 0.11502029769959404,
132
+ "grad_norm": 2.859375,
133
+ "learning_rate": 1.1486486486486488e-05,
134
+ "loss": 1.7624,
135
  "step": 85
136
  },
137
  {
138
+ "epoch": 0.12178619756427606,
139
+ "grad_norm": 2.796875,
140
+ "learning_rate": 1.2162162162162164e-05,
141
+ "loss": 1.7737,
142
  "step": 90
143
  },
144
  {
145
+ "epoch": 0.12855209742895804,
146
+ "grad_norm": 2.8125,
147
+ "learning_rate": 1.283783783783784e-05,
148
+ "loss": 1.7472,
149
  "step": 95
150
  },
151
  {
152
+ "epoch": 0.13531799729364005,
153
+ "grad_norm": 2.609375,
154
+ "learning_rate": 1.3513513513513515e-05,
155
+ "loss": 1.7465,
156
  "step": 100
157
  },
158
  {
159
+ "epoch": 0.14208389715832206,
160
+ "grad_norm": 2.75,
161
+ "learning_rate": 1.4189189189189189e-05,
162
+ "loss": 1.7326,
163
  "step": 105
164
  },
165
  {
166
+ "epoch": 0.14884979702300405,
167
+ "grad_norm": 2.5625,
168
+ "learning_rate": 1.4864864864864865e-05,
169
+ "loss": 1.7463,
170
  "step": 110
171
  },
172
  {
173
+ "epoch": 0.15561569688768606,
174
+ "grad_norm": 2.609375,
175
+ "learning_rate": 1.554054054054054e-05,
176
+ "loss": 1.7175,
177
  "step": 115
178
  },
179
  {
180
+ "epoch": 0.16238159675236807,
181
+ "grad_norm": 2.671875,
182
+ "learning_rate": 1.6216216216216218e-05,
183
+ "loss": 1.6941,
184
  "step": 120
185
  },
186
  {
187
+ "epoch": 0.16914749661705006,
188
+ "grad_norm": 2.609375,
189
+ "learning_rate": 1.6891891891891896e-05,
190
+ "loss": 1.6953,
191
  "step": 125
192
  },
193
  {
194
+ "epoch": 0.17591339648173207,
195
+ "grad_norm": 2.578125,
196
+ "learning_rate": 1.756756756756757e-05,
197
+ "loss": 1.7159,
198
  "step": 130
199
  },
200
  {
201
+ "epoch": 0.18267929634641408,
202
+ "grad_norm": 2.671875,
203
+ "learning_rate": 1.8243243243243244e-05,
204
+ "loss": 1.7148,
205
  "step": 135
206
  },
207
  {
208
+ "epoch": 0.18944519621109607,
209
+ "grad_norm": 2.6875,
210
+ "learning_rate": 1.891891891891892e-05,
211
+ "loss": 1.6967,
212
  "step": 140
213
  },
214
  {
215
+ "epoch": 0.19621109607577808,
216
+ "grad_norm": 2.578125,
217
+ "learning_rate": 1.9594594594594595e-05,
218
+ "loss": 1.7066,
219
  "step": 145
220
  },
221
  {
222
+ "epoch": 0.2029769959404601,
223
+ "grad_norm": 2.578125,
224
+ "learning_rate": 1.9999888409903948e-05,
225
+ "loss": 1.7105,
226
  "step": 150
227
  },
228
  {
229
+ "epoch": 0.20974289580514208,
230
+ "grad_norm": 2.65625,
231
+ "learning_rate": 1.9998633049924693e-05,
232
+ "loss": 1.6862,
233
  "step": 155
234
  },
235
  {
236
+ "epoch": 0.2165087956698241,
237
+ "grad_norm": 2.46875,
238
+ "learning_rate": 1.999598301803528e-05,
239
+ "loss": 1.6791,
240
  "step": 160
241
  },
242
  {
243
+ "epoch": 0.2232746955345061,
244
+ "grad_norm": 2.8125,
245
+ "learning_rate": 1.9991938683878746e-05,
246
+ "loss": 1.6925,
247
  "step": 165
248
  },
249
  {
250
+ "epoch": 0.23004059539918809,
251
+ "grad_norm": 2.515625,
252
+ "learning_rate": 1.9986500611584133e-05,
253
+ "loss": 1.6846,
254
  "step": 170
255
  },
256
  {
257
+ "epoch": 0.2368064952638701,
258
+ "grad_norm": 2.421875,
259
+ "learning_rate": 1.997966955968779e-05,
260
+ "loss": 1.6782,
261
  "step": 175
262
  },
263
  {
264
+ "epoch": 0.2435723951285521,
265
+ "grad_norm": 2.734375,
266
+ "learning_rate": 1.997144648102759e-05,
267
+ "loss": 1.6852,
268
  "step": 180
269
  },
270
  {
271
+ "epoch": 0.2503382949932341,
272
+ "grad_norm": 2.53125,
273
+ "learning_rate": 1.9961832522610004e-05,
274
+ "loss": 1.7028,
 
 
 
 
 
 
 
 
275
  "step": 185
276
  },
277
  {
278
+ "epoch": 0.2571041948579161,
279
+ "grad_norm": 2.59375,
280
+ "learning_rate": 1.9950829025450116e-05,
281
+ "loss": 1.6483,
282
  "step": 190
283
  },
284
  {
285
+ "epoch": 0.2638700947225981,
286
+ "grad_norm": 2.453125,
287
+ "learning_rate": 1.9938437524384572e-05,
288
+ "loss": 1.6498,
289
  "step": 195
290
  },
291
  {
292
+ "epoch": 0.2706359945872801,
293
+ "grad_norm": 2.59375,
294
+ "learning_rate": 1.9924659747857485e-05,
295
+ "loss": 1.6601,
296
  "step": 200
297
  },
298
  {
299
+ "epoch": 0.2774018944519621,
300
+ "grad_norm": 2.5,
301
+ "learning_rate": 1.990949761767935e-05,
302
+ "loss": 1.6657,
303
  "step": 205
304
  },
305
  {
306
+ "epoch": 0.28416779431664413,
307
+ "grad_norm": 2.609375,
308
+ "learning_rate": 1.989295324875897e-05,
309
+ "loss": 1.676,
310
  "step": 210
311
  },
312
  {
313
+ "epoch": 0.29093369418132614,
314
+ "grad_norm": 2.515625,
315
+ "learning_rate": 1.9875028948808457e-05,
316
+ "loss": 1.65,
317
  "step": 215
318
  },
319
  {
320
+ "epoch": 0.2976995940460081,
321
+ "grad_norm": 2.484375,
322
+ "learning_rate": 1.985572721802134e-05,
323
+ "loss": 1.648,
324
  "step": 220
325
  },
326
  {
327
+ "epoch": 0.3044654939106901,
328
+ "grad_norm": 2.46875,
329
+ "learning_rate": 1.9835050748723826e-05,
330
+ "loss": 1.6537,
331
  "step": 225
332
  },
333
  {
334
+ "epoch": 0.3112313937753721,
335
+ "grad_norm": 2.453125,
336
+ "learning_rate": 1.981300242499924e-05,
337
+ "loss": 1.6459,
338
  "step": 230
339
  },
340
  {
341
+ "epoch": 0.31799729364005414,
342
+ "grad_norm": 2.5625,
343
+ "learning_rate": 1.978958532228576e-05,
344
+ "loss": 1.6383,
345
  "step": 235
346
  },
347
  {
348
+ "epoch": 0.32476319350473615,
349
+ "grad_norm": 2.578125,
350
+ "learning_rate": 1.9764802706947423e-05,
351
+ "loss": 1.6624,
352
  "step": 240
353
  },
354
  {
355
+ "epoch": 0.33152909336941816,
356
+ "grad_norm": 2.53125,
357
+ "learning_rate": 1.9738658035818495e-05,
358
+ "loss": 1.6607,
359
  "step": 245
360
  },
361
  {
362
+ "epoch": 0.3382949932341001,
363
+ "grad_norm": 2.375,
364
+ "learning_rate": 1.9711154955721338e-05,
365
+ "loss": 1.6527,
366
  "step": 250
367
  },
368
  {
369
+ "epoch": 0.34506089309878213,
370
+ "grad_norm": 2.59375,
371
+ "learning_rate": 1.9682297302957666e-05,
372
+ "loss": 1.6423,
373
  "step": 255
374
  },
375
  {
376
+ "epoch": 0.35182679296346414,
377
+ "grad_norm": 2.5,
378
+ "learning_rate": 1.9652089102773487e-05,
379
+ "loss": 1.6288,
380
  "step": 260
381
  },
382
  {
383
+ "epoch": 0.35859269282814615,
384
+ "grad_norm": 2.4375,
385
+ "learning_rate": 1.962053456879761e-05,
386
+ "loss": 1.6436,
387
  "step": 265
388
  },
389
  {
390
+ "epoch": 0.36535859269282817,
391
+ "grad_norm": 2.46875,
392
+ "learning_rate": 1.95876381024539e-05,
393
+ "loss": 1.6378,
394
  "step": 270
395
  },
396
  {
397
+ "epoch": 0.3721244925575101,
398
+ "grad_norm": 2.59375,
399
+ "learning_rate": 1.9553404292347356e-05,
400
+ "loss": 1.6508,
401
  "step": 275
402
  },
403
  {
404
+ "epoch": 0.37889039242219213,
405
+ "grad_norm": 2.578125,
406
+ "learning_rate": 1.9517837913624048e-05,
407
+ "loss": 1.6475,
408
  "step": 280
409
  },
410
  {
411
+ "epoch": 0.38565629228687415,
412
+ "grad_norm": 2.546875,
413
+ "learning_rate": 1.948094392730506e-05,
414
+ "loss": 1.6353,
415
  "step": 285
416
  },
417
  {
418
+ "epoch": 0.39242219215155616,
419
+ "grad_norm": 2.40625,
420
+ "learning_rate": 1.9442727479594486e-05,
421
+ "loss": 1.6464,
422
  "step": 290
423
  },
424
  {
425
+ "epoch": 0.39918809201623817,
426
+ "grad_norm": 2.328125,
427
+ "learning_rate": 1.9403193901161614e-05,
428
+ "loss": 1.6157,
429
  "step": 295
430
  },
431
  {
432
+ "epoch": 0.4059539918809202,
433
+ "grad_norm": 2.5625,
434
+ "learning_rate": 1.9362348706397374e-05,
435
+ "loss": 1.643,
436
  "step": 300
437
  },
438
  {
439
+ "epoch": 0.41271989174560214,
440
+ "grad_norm": 2.5,
441
+ "learning_rate": 1.932019759264514e-05,
442
+ "loss": 1.6225,
443
  "step": 305
444
  },
445
  {
446
+ "epoch": 0.41948579161028415,
447
+ "grad_norm": 2.625,
448
+ "learning_rate": 1.9276746439406046e-05,
449
+ "loss": 1.6139,
450
  "step": 310
451
  },
452
  {
453
+ "epoch": 0.42625169147496617,
454
+ "grad_norm": 2.59375,
455
+ "learning_rate": 1.923200130751887e-05,
456
+ "loss": 1.6127,
457
  "step": 315
458
  },
459
  {
460
+ "epoch": 0.4330175913396482,
461
+ "grad_norm": 2.53125,
462
+ "learning_rate": 1.918596843831462e-05,
463
+ "loss": 1.6245,
464
  "step": 320
465
  },
466
  {
467
+ "epoch": 0.4397834912043302,
468
+ "grad_norm": 2.84375,
469
+ "learning_rate": 1.913865425274597e-05,
470
+ "loss": 1.6153,
471
  "step": 325
472
  },
473
  {
474
+ "epoch": 0.4465493910690122,
475
+ "grad_norm": 2.484375,
476
+ "learning_rate": 1.909006535049163e-05,
477
+ "loss": 1.6101,
478
  "step": 330
479
  },
480
  {
481
+ "epoch": 0.45331529093369416,
482
+ "grad_norm": 2.453125,
483
+ "learning_rate": 1.9040208509035745e-05,
484
+ "loss": 1.5968,
485
  "step": 335
486
  },
487
  {
488
+ "epoch": 0.46008119079837617,
489
+ "grad_norm": 2.421875,
490
+ "learning_rate": 1.8989090682722583e-05,
491
+ "loss": 1.6055,
492
  "step": 340
493
  },
494
  {
495
+ "epoch": 0.4668470906630582,
496
+ "grad_norm": 2.484375,
497
+ "learning_rate": 1.8936719001786453e-05,
498
+ "loss": 1.6253,
499
  "step": 345
500
  },
501
  {
502
+ "epoch": 0.4736129905277402,
503
+ "grad_norm": 2.625,
504
+ "learning_rate": 1.888310077135716e-05,
505
+ "loss": 1.6369,
506
  "step": 350
507
  },
508
  {
509
+ "epoch": 0.4803788903924222,
510
+ "grad_norm": 2.609375,
511
+ "learning_rate": 1.8828243470441026e-05,
512
+ "loss": 1.6359,
513
  "step": 355
514
  },
515
  {
516
+ "epoch": 0.4871447902571042,
517
+ "grad_norm": 2.453125,
518
+ "learning_rate": 1.8772154750877696e-05,
519
+ "loss": 1.613,
520
  "step": 360
521
  },
522
  {
523
+ "epoch": 0.4939106901217862,
524
+ "grad_norm": 2.421875,
525
+ "learning_rate": 1.8714842436272774e-05,
526
+ "loss": 1.603,
527
  "step": 365
528
  },
529
  {
530
+ "epoch": 0.5006765899864682,
531
+ "grad_norm": 2.484375,
532
+ "learning_rate": 1.865631452090657e-05,
533
+ "loss": 1.5804,
534
+ "step": 370
535
+ },
536
+ {
537
+ "epoch": 0.5074424898511503,
538
+ "grad_norm": 2.953125,
539
+ "learning_rate": 1.859657916861899e-05,
540
+ "loss": 1.62,
541
+ "step": 375
542
+ },
543
+ {
544
+ "epoch": 0.5142083897158322,
545
+ "grad_norm": 2.390625,
546
+ "learning_rate": 1.8535644711670804e-05,
547
+ "loss": 1.6201,
548
+ "step": 380
549
+ },
550
+ {
551
+ "epoch": 0.5209742895805142,
552
+ "grad_norm": 2.421875,
553
+ "learning_rate": 1.8473519649581396e-05,
554
+ "loss": 1.6105,
555
+ "step": 385
556
+ },
557
+ {
558
+ "epoch": 0.5277401894451962,
559
+ "grad_norm": 2.640625,
560
+ "learning_rate": 1.8410212647943215e-05,
561
+ "loss": 1.6231,
562
+ "step": 390
563
+ },
564
+ {
565
+ "epoch": 0.5345060893098782,
566
+ "grad_norm": 2.46875,
567
+ "learning_rate": 1.834573253721303e-05,
568
+ "loss": 1.5854,
569
+ "step": 395
570
+ },
571
+ {
572
+ "epoch": 0.5412719891745602,
573
+ "grad_norm": 2.46875,
574
+ "learning_rate": 1.8280088311480203e-05,
575
+ "loss": 1.5792,
576
+ "step": 400
577
+ },
578
+ {
579
+ "epoch": 0.5480378890392422,
580
+ "grad_norm": 2.546875,
581
+ "learning_rate": 1.8213289127212152e-05,
582
+ "loss": 1.6033,
583
+ "step": 405
584
+ },
585
+ {
586
+ "epoch": 0.5548037889039242,
587
+ "grad_norm": 2.53125,
588
+ "learning_rate": 1.8145344301977126e-05,
589
+ "loss": 1.5986,
590
+ "step": 410
591
+ },
592
+ {
593
+ "epoch": 0.5615696887686062,
594
+ "grad_norm": 2.484375,
595
+ "learning_rate": 1.8076263313144568e-05,
596
+ "loss": 1.6238,
597
+ "step": 415
598
+ },
599
+ {
600
+ "epoch": 0.5683355886332883,
601
+ "grad_norm": 2.53125,
602
+ "learning_rate": 1.8006055796563103e-05,
603
+ "loss": 1.5919,
604
+ "step": 420
605
+ },
606
+ {
607
+ "epoch": 0.5751014884979703,
608
+ "grad_norm": 2.390625,
609
+ "learning_rate": 1.7934731545216515e-05,
610
+ "loss": 1.6125,
611
+ "step": 425
612
+ },
613
+ {
614
+ "epoch": 0.5818673883626523,
615
+ "grad_norm": 2.46875,
616
+ "learning_rate": 1.7862300507857733e-05,
617
+ "loss": 1.6184,
618
+ "step": 430
619
+ },
620
+ {
621
+ "epoch": 0.5886332882273342,
622
+ "grad_norm": 2.4375,
623
+ "learning_rate": 1.7788772787621126e-05,
624
+ "loss": 1.6034,
625
+ "step": 435
626
+ },
627
+ {
628
+ "epoch": 0.5953991880920162,
629
+ "grad_norm": 2.53125,
630
+ "learning_rate": 1.771415864061326e-05,
631
+ "loss": 1.5829,
632
+ "step": 440
633
+ },
634
+ {
635
+ "epoch": 0.6021650879566982,
636
+ "grad_norm": 2.421875,
637
+ "learning_rate": 1.7638468474482297e-05,
638
+ "loss": 1.6085,
639
+ "step": 445
640
+ },
641
+ {
642
+ "epoch": 0.6089309878213802,
643
+ "grad_norm": 2.515625,
644
+ "learning_rate": 1.756171284696629e-05,
645
+ "loss": 1.6123,
646
+ "step": 450
647
+ },
648
+ {
649
+ "epoch": 0.6156968876860622,
650
+ "grad_norm": 2.46875,
651
+ "learning_rate": 1.7483902464420507e-05,
652
+ "loss": 1.6034,
653
+ "step": 455
654
+ },
655
+ {
656
+ "epoch": 0.6224627875507442,
657
+ "grad_norm": 2.46875,
658
+ "learning_rate": 1.7405048180324046e-05,
659
+ "loss": 1.5858,
660
+ "step": 460
661
+ },
662
+ {
663
+ "epoch": 0.6292286874154263,
664
+ "grad_norm": 2.5,
665
+ "learning_rate": 1.7325160993765934e-05,
666
+ "loss": 1.6065,
667
+ "step": 465
668
+ },
669
+ {
670
+ "epoch": 0.6359945872801083,
671
+ "grad_norm": 2.515625,
672
+ "learning_rate": 1.7244252047910893e-05,
673
+ "loss": 1.5963,
674
+ "step": 470
675
+ },
676
+ {
677
+ "epoch": 0.6427604871447903,
678
+ "grad_norm": 2.484375,
679
+ "learning_rate": 1.7162332628445024e-05,
680
+ "loss": 1.594,
681
+ "step": 475
682
+ },
683
+ {
684
+ "epoch": 0.6495263870094723,
685
+ "grad_norm": 2.484375,
686
+ "learning_rate": 1.7079414162001617e-05,
687
+ "loss": 1.5899,
688
+ "step": 480
689
+ },
690
+ {
691
+ "epoch": 0.6562922868741543,
692
+ "grad_norm": 2.484375,
693
+ "learning_rate": 1.6995508214567275e-05,
694
+ "loss": 1.5786,
695
+ "step": 485
696
+ },
697
+ {
698
+ "epoch": 0.6630581867388363,
699
+ "grad_norm": 2.484375,
700
+ "learning_rate": 1.691062648986865e-05,
701
+ "loss": 1.5962,
702
+ "step": 490
703
+ },
704
+ {
705
+ "epoch": 0.6698240866035182,
706
+ "grad_norm": 2.625,
707
+ "learning_rate": 1.682478082773989e-05,
708
+ "loss": 1.5961,
709
+ "step": 495
710
+ },
711
+ {
712
+ "epoch": 0.6765899864682002,
713
+ "grad_norm": 2.46875,
714
+ "learning_rate": 1.673798320247118e-05,
715
+ "loss": 1.6023,
716
+ "step": 500
717
+ },
718
+ {
719
+ "epoch": 0.6833558863328822,
720
+ "grad_norm": 2.40625,
721
+ "learning_rate": 1.6650245721138483e-05,
722
+ "loss": 1.6074,
723
+ "step": 505
724
+ },
725
+ {
726
+ "epoch": 0.6901217861975643,
727
+ "grad_norm": 2.421875,
728
+ "learning_rate": 1.6561580621914764e-05,
729
+ "loss": 1.5767,
730
+ "step": 510
731
+ },
732
+ {
733
+ "epoch": 0.6968876860622463,
734
+ "grad_norm": 2.46875,
735
+ "learning_rate": 1.6472000272362937e-05,
736
+ "loss": 1.6047,
737
+ "step": 515
738
+ },
739
+ {
740
+ "epoch": 0.7036535859269283,
741
+ "grad_norm": 2.53125,
742
+ "learning_rate": 1.6381517167710757e-05,
743
+ "loss": 1.5846,
744
+ "step": 520
745
+ },
746
+ {
747
+ "epoch": 0.7104194857916103,
748
+ "grad_norm": 2.5625,
749
+ "learning_rate": 1.629014392910791e-05,
750
+ "loss": 1.5756,
751
+ "step": 525
752
+ },
753
+ {
754
+ "epoch": 0.7171853856562923,
755
+ "grad_norm": 2.453125,
756
+ "learning_rate": 1.619789330186555e-05,
757
+ "loss": 1.5966,
758
+ "step": 530
759
+ },
760
+ {
761
+ "epoch": 0.7239512855209743,
762
+ "grad_norm": 2.53125,
763
+ "learning_rate": 1.6104778153678467e-05,
764
+ "loss": 1.5901,
765
+ "step": 535
766
+ },
767
+ {
768
+ "epoch": 0.7307171853856563,
769
+ "grad_norm": 2.515625,
770
+ "learning_rate": 1.6010811472830253e-05,
771
+ "loss": 1.5712,
772
+ "step": 540
773
+ },
774
+ {
775
+ "epoch": 0.7374830852503383,
776
+ "grad_norm": 2.5,
777
+ "learning_rate": 1.591600636638161e-05,
778
+ "loss": 1.5877,
779
+ "step": 545
780
+ },
781
+ {
782
+ "epoch": 0.7442489851150202,
783
+ "grad_norm": 2.453125,
784
+ "learning_rate": 1.5820376058342077e-05,
785
+ "loss": 1.591,
786
+ "step": 550
787
+ },
788
+ {
789
+ "epoch": 0.7510148849797023,
790
+ "grad_norm": 2.453125,
791
+ "learning_rate": 1.5723933887825492e-05,
792
+ "loss": 1.5814,
793
+ "step": 555
794
+ },
795
+ {
796
+ "epoch": 0.7577807848443843,
797
+ "grad_norm": 2.609375,
798
+ "learning_rate": 1.5626693307189334e-05,
799
+ "loss": 1.5764,
800
+ "step": 560
801
+ },
802
+ {
803
+ "epoch": 0.7645466847090663,
804
+ "grad_norm": 2.484375,
805
+ "learning_rate": 1.5528667880158338e-05,
806
+ "loss": 1.5726,
807
+ "step": 565
808
+ },
809
+ {
810
+ "epoch": 0.7713125845737483,
811
+ "grad_norm": 2.484375,
812
+ "learning_rate": 1.5429871279932514e-05,
813
+ "loss": 1.5913,
814
+ "step": 570
815
+ },
816
+ {
817
+ "epoch": 0.7780784844384303,
818
+ "grad_norm": 2.5,
819
+ "learning_rate": 1.533031728727994e-05,
820
+ "loss": 1.5674,
821
+ "step": 575
822
+ },
823
+ {
824
+ "epoch": 0.7848443843031123,
825
+ "grad_norm": 2.421875,
826
+ "learning_rate": 1.5230019788614527e-05,
827
+ "loss": 1.5727,
828
+ "step": 580
829
+ },
830
+ {
831
+ "epoch": 0.7916102841677943,
832
+ "grad_norm": 2.4375,
833
+ "learning_rate": 1.5128992774059063e-05,
834
+ "loss": 1.5762,
835
+ "step": 585
836
+ },
837
+ {
838
+ "epoch": 0.7983761840324763,
839
+ "grad_norm": 2.484375,
840
+ "learning_rate": 1.5027250335493771e-05,
841
+ "loss": 1.5678,
842
+ "step": 590
843
+ },
844
+ {
845
+ "epoch": 0.8051420838971584,
846
+ "grad_norm": 2.5,
847
+ "learning_rate": 1.4924806664590702e-05,
848
+ "loss": 1.5602,
849
+ "step": 595
850
+ },
851
+ {
852
+ "epoch": 0.8119079837618404,
853
+ "grad_norm": 2.5625,
854
+ "learning_rate": 1.4821676050834166e-05,
855
+ "loss": 1.5761,
856
+ "step": 600
857
+ },
858
+ {
859
+ "epoch": 0.8186738836265224,
860
+ "grad_norm": 2.4375,
861
+ "learning_rate": 1.4717872879527578e-05,
862
+ "loss": 1.5751,
863
+ "step": 605
864
+ },
865
+ {
866
+ "epoch": 0.8254397834912043,
867
+ "grad_norm": 3.96875,
868
+ "learning_rate": 1.461341162978688e-05,
869
+ "loss": 1.5783,
870
+ "step": 610
871
+ },
872
+ {
873
+ "epoch": 0.8322056833558863,
874
+ "grad_norm": 2.46875,
875
+ "learning_rate": 1.4508306872520912e-05,
876
+ "loss": 1.5859,
877
+ "step": 615
878
+ },
879
+ {
880
+ "epoch": 0.8389715832205683,
881
+ "grad_norm": 2.515625,
882
+ "learning_rate": 1.4402573268398969e-05,
883
+ "loss": 1.5898,
884
+ "step": 620
885
+ },
886
+ {
887
+ "epoch": 0.8457374830852503,
888
+ "grad_norm": 2.421875,
889
+ "learning_rate": 1.4296225565805854e-05,
890
+ "loss": 1.5665,
891
+ "step": 625
892
+ },
893
+ {
894
+ "epoch": 0.8525033829499323,
895
+ "grad_norm": 2.421875,
896
+ "learning_rate": 1.4189278598784648e-05,
897
+ "loss": 1.5729,
898
+ "step": 630
899
+ },
900
+ {
901
+ "epoch": 0.8592692828146143,
902
+ "grad_norm": 2.421875,
903
+ "learning_rate": 1.4081747284967602e-05,
904
+ "loss": 1.5769,
905
+ "step": 635
906
+ },
907
+ {
908
+ "epoch": 0.8660351826792964,
909
+ "grad_norm": 2.34375,
910
+ "learning_rate": 1.3973646623495305e-05,
911
+ "loss": 1.5618,
912
+ "step": 640
913
+ },
914
+ {
915
+ "epoch": 0.8728010825439784,
916
+ "grad_norm": 2.4375,
917
+ "learning_rate": 1.3864991692924524e-05,
918
+ "loss": 1.589,
919
+ "step": 645
920
+ },
921
+ {
922
+ "epoch": 0.8795669824086604,
923
+ "grad_norm": 2.484375,
924
+ "learning_rate": 1.3755797649124944e-05,
925
+ "loss": 1.5733,
926
+ "step": 650
927
+ },
928
+ {
929
+ "epoch": 0.8863328822733424,
930
+ "grad_norm": 2.453125,
931
+ "learning_rate": 1.3646079723165148e-05,
932
+ "loss": 1.5596,
933
+ "step": 655
934
+ },
935
+ {
936
+ "epoch": 0.8930987821380244,
937
+ "grad_norm": 2.5,
938
+ "learning_rate": 1.3535853219188064e-05,
939
+ "loss": 1.5722,
940
+ "step": 660
941
+ },
942
+ {
943
+ "epoch": 0.8998646820027063,
944
+ "grad_norm": 2.484375,
945
+ "learning_rate": 1.3425133512276284e-05,
946
+ "loss": 1.5554,
947
+ "step": 665
948
+ },
949
+ {
950
+ "epoch": 0.9066305818673883,
951
+ "grad_norm": 2.453125,
952
+ "learning_rate": 1.3313936046307411e-05,
953
+ "loss": 1.576,
954
+ "step": 670
955
+ },
956
+ {
957
+ "epoch": 0.9133964817320703,
958
+ "grad_norm": 2.453125,
959
+ "learning_rate": 1.320227633179989e-05,
960
+ "loss": 1.5668,
961
+ "step": 675
962
+ },
963
+ {
964
+ "epoch": 0.9201623815967523,
965
+ "grad_norm": 2.25,
966
+ "learning_rate": 1.3090169943749475e-05,
967
+ "loss": 1.5757,
968
+ "step": 680
969
+ },
970
+ {
971
+ "epoch": 0.9269282814614344,
972
+ "grad_norm": 2.515625,
973
+ "learning_rate": 1.2977632519456745e-05,
974
+ "loss": 1.5817,
975
+ "step": 685
976
+ },
977
+ {
978
+ "epoch": 0.9336941813261164,
979
+ "grad_norm": 2.359375,
980
+ "learning_rate": 1.2864679756345905e-05,
981
+ "loss": 1.5847,
982
+ "step": 690
983
+ },
984
+ {
985
+ "epoch": 0.9404600811907984,
986
+ "grad_norm": 2.484375,
987
+ "learning_rate": 1.2751327409775227e-05,
988
+ "loss": 1.5847,
989
+ "step": 695
990
+ },
991
+ {
992
+ "epoch": 0.9472259810554804,
993
+ "grad_norm": 2.625,
994
+ "learning_rate": 1.2637591290839377e-05,
995
+ "loss": 1.5695,
996
+ "step": 700
997
+ },
998
+ {
999
+ "epoch": 0.9539918809201624,
1000
+ "grad_norm": 2.4375,
1001
+ "learning_rate": 1.2523487264163997e-05,
1002
+ "loss": 1.566,
1003
+ "step": 705
1004
+ },
1005
+ {
1006
+ "epoch": 0.9607577807848444,
1007
+ "grad_norm": 2.34375,
1008
+ "learning_rate": 1.2409031245692798e-05,
1009
+ "loss": 1.5487,
1010
+ "step": 710
1011
+ },
1012
+ {
1013
+ "epoch": 0.9675236806495264,
1014
+ "grad_norm": 2.453125,
1015
+ "learning_rate": 1.2294239200467516e-05,
1016
+ "loss": 1.5825,
1017
+ "step": 715
1018
+ },
1019
+ {
1020
+ "epoch": 0.9742895805142084,
1021
+ "grad_norm": 2.546875,
1022
+ "learning_rate": 1.2179127140400997e-05,
1023
+ "loss": 1.5651,
1024
+ "step": 720
1025
+ },
1026
+ {
1027
+ "epoch": 0.9810554803788903,
1028
+ "grad_norm": 2.40625,
1029
+ "learning_rate": 1.2063711122043759e-05,
1030
+ "loss": 1.5487,
1031
+ "step": 725
1032
+ },
1033
+ {
1034
+ "epoch": 0.9878213802435724,
1035
+ "grad_norm": 2.46875,
1036
+ "learning_rate": 1.1948007244344334e-05,
1037
+ "loss": 1.5604,
1038
+ "step": 730
1039
+ },
1040
+ {
1041
+ "epoch": 0.9945872801082544,
1042
+ "grad_norm": 2.375,
1043
+ "learning_rate": 1.1832031646403654e-05,
1044
+ "loss": 1.5838,
1045
+ "step": 735
1046
+ },
1047
+ {
1048
+ "epoch": 1.0,
1049
+ "eval_loss": 1.6891789436340332,
1050
+ "eval_runtime": 28.8376,
1051
+ "eval_samples_per_second": 15.084,
1052
+ "eval_steps_per_second": 1.907,
1053
+ "step": 739
1054
+ },
1055
+ {
1056
+ "epoch": 1.0013531799729365,
1057
+ "grad_norm": 2.265625,
1058
+ "learning_rate": 1.1715800505223918e-05,
1059
+ "loss": 1.5323,
1060
+ "step": 740
1061
+ },
1062
+ {
1063
+ "epoch": 1.0081190798376185,
1064
+ "grad_norm": 2.265625,
1065
+ "learning_rate": 1.1599330033452078e-05,
1066
+ "loss": 1.4855,
1067
+ "step": 745
1068
+ },
1069
+ {
1070
+ "epoch": 1.0148849797023005,
1071
+ "grad_norm": 2.4375,
1072
+ "learning_rate": 1.148263647711842e-05,
1073
+ "loss": 1.531,
1074
+ "step": 750
1075
+ },
1076
+ {
1077
+ "epoch": 1.0216508795669823,
1078
+ "grad_norm": 2.390625,
1079
+ "learning_rate": 1.1365736113370463e-05,
1080
+ "loss": 1.5063,
1081
+ "step": 755
1082
+ },
1083
+ {
1084
+ "epoch": 1.0284167794316643,
1085
+ "grad_norm": 2.515625,
1086
+ "learning_rate": 1.124864524820251e-05,
1087
+ "loss": 1.5219,
1088
+ "step": 760
1089
+ },
1090
+ {
1091
+ "epoch": 1.0351826792963463,
1092
+ "grad_norm": 2.515625,
1093
+ "learning_rate": 1.1131380214181205e-05,
1094
+ "loss": 1.4972,
1095
+ "step": 765
1096
+ },
1097
+ {
1098
+ "epoch": 1.0419485791610283,
1099
+ "grad_norm": 2.328125,
1100
+ "learning_rate": 1.1013957368167343e-05,
1101
+ "loss": 1.5092,
1102
+ "step": 770
1103
+ },
1104
+ {
1105
+ "epoch": 1.0487144790257104,
1106
+ "grad_norm": 2.4375,
1107
+ "learning_rate": 1.0896393089034336e-05,
1108
+ "loss": 1.4743,
1109
+ "step": 775
1110
+ },
1111
+ {
1112
+ "epoch": 1.0554803788903924,
1113
+ "grad_norm": 2.3125,
1114
+ "learning_rate": 1.0778703775383559e-05,
1115
+ "loss": 1.4991,
1116
+ "step": 780
1117
+ },
1118
+ {
1119
+ "epoch": 1.0622462787550744,
1120
+ "grad_norm": 2.328125,
1121
+ "learning_rate": 1.0660905843256995e-05,
1122
+ "loss": 1.4727,
1123
+ "step": 785
1124
+ },
1125
+ {
1126
+ "epoch": 1.0690121786197564,
1127
+ "grad_norm": 2.390625,
1128
+ "learning_rate": 1.0543015723847402e-05,
1129
+ "loss": 1.4992,
1130
+ "step": 790
1131
+ },
1132
+ {
1133
+ "epoch": 1.0757780784844384,
1134
+ "grad_norm": 2.375,
1135
+ "learning_rate": 1.042504986120641e-05,
1136
+ "loss": 1.5143,
1137
+ "step": 795
1138
+ },
1139
+ {
1140
+ "epoch": 1.0825439783491204,
1141
+ "grad_norm": 2.453125,
1142
+ "learning_rate": 1.0307024709950775e-05,
1143
+ "loss": 1.5027,
1144
+ "step": 800
1145
+ },
1146
+ {
1147
+ "epoch": 1.0893098782138024,
1148
+ "grad_norm": 2.25,
1149
+ "learning_rate": 1.0188956732967208e-05,
1150
+ "loss": 1.4863,
1151
+ "step": 805
1152
+ },
1153
+ {
1154
+ "epoch": 1.0960757780784844,
1155
+ "grad_norm": 2.34375,
1156
+ "learning_rate": 1.0070862399116016e-05,
1157
+ "loss": 1.5161,
1158
+ "step": 810
1159
+ },
1160
+ {
1161
+ "epoch": 1.1028416779431665,
1162
+ "grad_norm": 2.390625,
1163
+ "learning_rate": 9.952758180933933e-06,
1164
+ "loss": 1.5031,
1165
+ "step": 815
1166
+ },
1167
+ {
1168
+ "epoch": 1.1096075778078485,
1169
+ "grad_norm": 2.40625,
1170
+ "learning_rate": 9.834660552336415e-06,
1171
+ "loss": 1.5017,
1172
+ "step": 820
1173
+ },
1174
+ {
1175
+ "epoch": 1.1163734776725305,
1176
+ "grad_norm": 2.359375,
1177
+ "learning_rate": 9.716585986319769e-06,
1178
+ "loss": 1.4958,
1179
+ "step": 825
1180
+ },
1181
+ {
1182
+ "epoch": 1.1231393775372125,
1183
+ "grad_norm": 2.453125,
1184
+ "learning_rate": 9.598550952663383e-06,
1185
+ "loss": 1.5104,
1186
+ "step": 830
1187
+ },
1188
+ {
1189
+ "epoch": 1.1299052774018945,
1190
+ "grad_norm": 2.46875,
1191
+ "learning_rate": 9.480571915632422e-06,
1192
+ "loss": 1.5208,
1193
+ "step": 835
1194
+ },
1195
+ {
1196
+ "epoch": 1.1366711772665765,
1197
+ "grad_norm": 2.40625,
1198
+ "learning_rate": 9.362665331681294e-06,
1199
+ "loss": 1.5244,
1200
+ "step": 840
1201
+ },
1202
+ {
1203
+ "epoch": 1.1434370771312585,
1204
+ "grad_norm": 2.359375,
1205
+ "learning_rate": 9.244847647158203e-06,
1206
+ "loss": 1.5114,
1207
+ "step": 845
1208
+ },
1209
+ {
1210
+ "epoch": 1.1502029769959405,
1211
+ "grad_norm": 2.3125,
1212
+ "learning_rate": 9.127135296011102e-06,
1213
+ "loss": 1.5153,
1214
+ "step": 850
1215
+ },
1216
+ {
1217
+ "epoch": 1.1569688768606226,
1218
+ "grad_norm": 2.359375,
1219
+ "learning_rate": 9.009544697495373e-06,
1220
+ "loss": 1.4826,
1221
+ "step": 855
1222
+ },
1223
+ {
1224
+ "epoch": 1.1637347767253043,
1225
+ "grad_norm": 2.3125,
1226
+ "learning_rate": 8.892092253883602e-06,
1227
+ "loss": 1.5164,
1228
+ "step": 860
1229
+ },
1230
+ {
1231
+ "epoch": 1.1705006765899864,
1232
+ "grad_norm": 2.59375,
1233
+ "learning_rate": 8.774794348177641e-06,
1234
+ "loss": 1.4995,
1235
+ "step": 865
1236
+ },
1237
+ {
1238
+ "epoch": 1.1772665764546684,
1239
+ "grad_norm": 2.4375,
1240
+ "learning_rate": 8.657667341823449e-06,
1241
+ "loss": 1.5047,
1242
+ "step": 870
1243
+ },
1244
+ {
1245
+ "epoch": 1.1840324763193504,
1246
+ "grad_norm": 2.46875,
1247
+ "learning_rate": 8.540727572428854e-06,
1248
+ "loss": 1.5063,
1249
+ "step": 875
1250
+ },
1251
+ {
1252
+ "epoch": 1.1907983761840324,
1253
+ "grad_norm": 2.296875,
1254
+ "learning_rate": 8.423991351484715e-06,
1255
+ "loss": 1.4832,
1256
+ "step": 880
1257
+ },
1258
+ {
1259
+ "epoch": 1.1975642760487144,
1260
+ "grad_norm": 2.359375,
1261
+ "learning_rate": 8.307474962089676e-06,
1262
+ "loss": 1.4901,
1263
+ "step": 885
1264
+ },
1265
+ {
1266
+ "epoch": 1.2043301759133964,
1267
+ "grad_norm": 2.265625,
1268
+ "learning_rate": 8.191194656678905e-06,
1269
+ "loss": 1.5061,
1270
+ "step": 890
1271
+ },
1272
+ {
1273
+ "epoch": 1.2110960757780784,
1274
+ "grad_norm": 2.46875,
1275
+ "learning_rate": 8.07516665475708e-06,
1276
+ "loss": 1.4967,
1277
+ "step": 895
1278
+ },
1279
+ {
1280
+ "epoch": 1.2178619756427604,
1281
+ "grad_norm": 2.46875,
1282
+ "learning_rate": 7.959407140636034e-06,
1283
+ "loss": 1.5082,
1284
+ "step": 900
1285
+ },
1286
+ {
1287
+ "epoch": 1.2246278755074425,
1288
+ "grad_norm": 2.390625,
1289
+ "learning_rate": 7.843932261177224e-06,
1290
+ "loss": 1.485,
1291
+ "step": 905
1292
+ },
1293
+ {
1294
+ "epoch": 1.2313937753721245,
1295
+ "grad_norm": 2.328125,
1296
+ "learning_rate": 7.728758123539498e-06,
1297
+ "loss": 1.5061,
1298
+ "step": 910
1299
+ },
1300
+ {
1301
+ "epoch": 1.2381596752368065,
1302
+ "grad_norm": 2.375,
1303
+ "learning_rate": 7.613900792932331e-06,
1304
+ "loss": 1.4953,
1305
+ "step": 915
1306
+ },
1307
+ {
1308
+ "epoch": 1.2449255751014885,
1309
+ "grad_norm": 2.421875,
1310
+ "learning_rate": 7.499376290374994e-06,
1311
+ "loss": 1.493,
1312
+ "step": 920
1313
+ },
1314
+ {
1315
+ "epoch": 1.2516914749661705,
1316
+ "grad_norm": 2.53125,
1317
+ "learning_rate": 7.385200590461803e-06,
1318
+ "loss": 1.506,
1319
+ "step": 925
1320
+ },
1321
+ {
1322
+ "epoch": 1.2584573748308525,
1323
+ "grad_norm": 2.40625,
1324
+ "learning_rate": 7.271389619133908e-06,
1325
+ "loss": 1.5135,
1326
+ "step": 930
1327
+ },
1328
+ {
1329
+ "epoch": 1.2652232746955345,
1330
+ "grad_norm": 2.34375,
1331
+ "learning_rate": 7.1579592514578234e-06,
1332
+ "loss": 1.5076,
1333
+ "step": 935
1334
+ },
1335
+ {
1336
+ "epoch": 1.2719891745602165,
1337
+ "grad_norm": 2.359375,
1338
+ "learning_rate": 7.0449253094110925e-06,
1339
+ "loss": 1.486,
1340
+ "step": 940
1341
+ },
1342
+ {
1343
+ "epoch": 1.2787550744248986,
1344
+ "grad_norm": 2.4375,
1345
+ "learning_rate": 6.93230355967533e-06,
1346
+ "loss": 1.5115,
1347
+ "step": 945
1348
+ },
1349
+ {
1350
+ "epoch": 1.2855209742895806,
1351
+ "grad_norm": 2.484375,
1352
+ "learning_rate": 6.820109711436989e-06,
1353
+ "loss": 1.4928,
1354
+ "step": 950
1355
+ },
1356
+ {
1357
+ "epoch": 1.2922868741542626,
1358
+ "grad_norm": 2.3125,
1359
+ "learning_rate": 6.7083594141961326e-06,
1360
+ "loss": 1.5128,
1361
+ "step": 955
1362
+ },
1363
+ {
1364
+ "epoch": 1.2990527740189446,
1365
+ "grad_norm": 2.28125,
1366
+ "learning_rate": 6.59706825558357e-06,
1367
+ "loss": 1.5023,
1368
+ "step": 960
1369
+ },
1370
+ {
1371
+ "epoch": 1.3058186738836266,
1372
+ "grad_norm": 2.3125,
1373
+ "learning_rate": 6.486251759186573e-06,
1374
+ "loss": 1.489,
1375
+ "step": 965
1376
+ },
1377
+ {
1378
+ "epoch": 1.3125845737483086,
1379
+ "grad_norm": 2.328125,
1380
+ "learning_rate": 6.375925382383561e-06,
1381
+ "loss": 1.4976,
1382
+ "step": 970
1383
+ },
1384
+ {
1385
+ "epoch": 1.3193504736129906,
1386
+ "grad_norm": 2.546875,
1387
+ "learning_rate": 6.266104514187997e-06,
1388
+ "loss": 1.5049,
1389
+ "step": 975
1390
+ },
1391
+ {
1392
+ "epoch": 1.3261163734776726,
1393
+ "grad_norm": 2.3125,
1394
+ "learning_rate": 6.156804473101852e-06,
1395
+ "loss": 1.506,
1396
+ "step": 980
1397
+ },
1398
+ {
1399
+ "epoch": 1.3328822733423547,
1400
+ "grad_norm": 2.4375,
1401
+ "learning_rate": 6.048040504978861e-06,
1402
+ "loss": 1.5012,
1403
+ "step": 985
1404
+ },
1405
+ {
1406
+ "epoch": 1.3396481732070367,
1407
+ "grad_norm": 2.375,
1408
+ "learning_rate": 5.93982778089796e-06,
1409
+ "loss": 1.5174,
1410
+ "step": 990
1411
+ },
1412
+ {
1413
+ "epoch": 1.3464140730717187,
1414
+ "grad_norm": 2.328125,
1415
+ "learning_rate": 5.832181395047099e-06,
1416
+ "loss": 1.5131,
1417
+ "step": 995
1418
+ },
1419
+ {
1420
+ "epoch": 1.3531799729364005,
1421
+ "grad_norm": 2.515625,
1422
+ "learning_rate": 5.72511636261784e-06,
1423
+ "loss": 1.4879,
1424
+ "step": 1000
1425
+ },
1426
+ {
1427
+ "epoch": 1.3599458728010825,
1428
+ "grad_norm": 2.390625,
1429
+ "learning_rate": 5.618647617710935e-06,
1430
+ "loss": 1.4899,
1431
+ "step": 1005
1432
+ },
1433
+ {
1434
+ "epoch": 1.3667117726657645,
1435
+ "grad_norm": 2.453125,
1436
+ "learning_rate": 5.512790011253211e-06,
1437
+ "loss": 1.4927,
1438
+ "step": 1010
1439
+ },
1440
+ {
1441
+ "epoch": 1.3734776725304465,
1442
+ "grad_norm": 2.375,
1443
+ "learning_rate": 5.407558308926083e-06,
1444
+ "loss": 1.5084,
1445
+ "step": 1015
1446
+ },
1447
+ {
1448
+ "epoch": 1.3802435723951285,
1449
+ "grad_norm": 2.53125,
1450
+ "learning_rate": 5.302967189105941e-06,
1451
+ "loss": 1.4999,
1452
+ "step": 1020
1453
+ },
1454
+ {
1455
+ "epoch": 1.3870094722598105,
1456
+ "grad_norm": 2.3125,
1457
+ "learning_rate": 5.199031240816715e-06,
1458
+ "loss": 1.4828,
1459
+ "step": 1025
1460
+ },
1461
+ {
1462
+ "epoch": 1.3937753721244925,
1463
+ "grad_norm": 2.4375,
1464
+ "learning_rate": 5.095764961694923e-06,
1465
+ "loss": 1.4967,
1466
+ "step": 1030
1467
+ },
1468
+ {
1469
+ "epoch": 1.4005412719891746,
1470
+ "grad_norm": 2.34375,
1471
+ "learning_rate": 4.993182755967412e-06,
1472
+ "loss": 1.4959,
1473
+ "step": 1035
1474
+ },
1475
+ {
1476
+ "epoch": 1.4073071718538566,
1477
+ "grad_norm": 2.421875,
1478
+ "learning_rate": 4.8912989324422164e-06,
1479
+ "loss": 1.4905,
1480
+ "step": 1040
1481
+ },
1482
+ {
1483
+ "epoch": 1.4140730717185386,
1484
+ "grad_norm": 2.359375,
1485
+ "learning_rate": 4.7901277025126345e-06,
1486
+ "loss": 1.5105,
1487
+ "step": 1045
1488
+ },
1489
+ {
1490
+ "epoch": 1.4208389715832206,
1491
+ "grad_norm": 2.390625,
1492
+ "learning_rate": 4.689683178174964e-06,
1493
+ "loss": 1.5236,
1494
+ "step": 1050
1495
+ },
1496
+ {
1497
+ "epoch": 1.4276048714479026,
1498
+ "grad_norm": 2.328125,
1499
+ "learning_rate": 4.589979370060037e-06,
1500
+ "loss": 1.4801,
1501
+ "step": 1055
1502
+ },
1503
+ {
1504
+ "epoch": 1.4343707713125846,
1505
+ "grad_norm": 2.28125,
1506
+ "learning_rate": 4.491030185478976e-06,
1507
+ "loss": 1.5219,
1508
+ "step": 1060
1509
+ },
1510
+ {
1511
+ "epoch": 1.4411366711772666,
1512
+ "grad_norm": 2.375,
1513
+ "learning_rate": 4.392849426483275e-06,
1514
+ "loss": 1.5072,
1515
+ "step": 1065
1516
+ },
1517
+ {
1518
+ "epoch": 1.4479025710419486,
1519
+ "grad_norm": 2.3125,
1520
+ "learning_rate": 4.295450787939622e-06,
1521
+ "loss": 1.4897,
1522
+ "step": 1070
1523
+ },
1524
+ {
1525
+ "epoch": 1.4546684709066307,
1526
+ "grad_norm": 2.46875,
1527
+ "learning_rate": 4.198847855619652e-06,
1528
+ "loss": 1.5098,
1529
+ "step": 1075
1530
+ },
1531
+ {
1532
+ "epoch": 1.4614343707713127,
1533
+ "grad_norm": 2.515625,
1534
+ "learning_rate": 4.1030541043049125e-06,
1535
+ "loss": 1.5226,
1536
+ "step": 1080
1537
+ },
1538
+ {
1539
+ "epoch": 1.4682002706359945,
1540
+ "grad_norm": 2.296875,
1541
+ "learning_rate": 4.008082895907326e-06,
1542
+ "loss": 1.4999,
1543
+ "step": 1085
1544
+ },
1545
+ {
1546
+ "epoch": 1.4749661705006765,
1547
+ "grad_norm": 2.453125,
1548
+ "learning_rate": 3.913947477605378e-06,
1549
+ "loss": 1.5023,
1550
+ "step": 1090
1551
+ },
1552
+ {
1553
+ "epoch": 1.4817320703653585,
1554
+ "grad_norm": 2.5,
1555
+ "learning_rate": 3.8206609799963205e-06,
1556
+ "loss": 1.5144,
1557
+ "step": 1095
1558
+ },
1559
+ {
1560
+ "epoch": 1.4884979702300405,
1561
+ "grad_norm": 2.375,
1562
+ "learning_rate": 3.72823641526463e-06,
1563
+ "loss": 1.5185,
1564
+ "step": 1100
1565
+ },
1566
+ {
1567
+ "epoch": 1.4952638700947225,
1568
+ "grad_norm": 2.390625,
1569
+ "learning_rate": 3.636686675367006e-06,
1570
+ "loss": 1.4987,
1571
+ "step": 1105
1572
+ },
1573
+ {
1574
+ "epoch": 1.5020297699594045,
1575
+ "grad_norm": 2.484375,
1576
+ "learning_rate": 3.5460245302340914e-06,
1577
+ "loss": 1.5168,
1578
+ "step": 1110
1579
+ },
1580
+ {
1581
+ "epoch": 1.5087956698240865,
1582
+ "grad_norm": 2.34375,
1583
+ "learning_rate": 3.4562626259892605e-06,
1584
+ "loss": 1.4996,
1585
+ "step": 1115
1586
+ },
1587
+ {
1588
+ "epoch": 1.5155615696887685,
1589
+ "grad_norm": 2.328125,
1590
+ "learning_rate": 3.367413483184654e-06,
1591
+ "loss": 1.5008,
1592
+ "step": 1120
1593
+ },
1594
+ {
1595
+ "epoch": 1.5223274695534506,
1596
+ "grad_norm": 2.359375,
1597
+ "learning_rate": 3.279489495054742e-06,
1598
+ "loss": 1.4882,
1599
+ "step": 1125
1600
+ },
1601
+ {
1602
+ "epoch": 1.5290933694181326,
1603
+ "grad_norm": 2.296875,
1604
+ "learning_rate": 3.1925029257876206e-06,
1605
+ "loss": 1.4648,
1606
+ "step": 1130
1607
+ },
1608
+ {
1609
+ "epoch": 1.5358592692828146,
1610
+ "grad_norm": 2.5,
1611
+ "learning_rate": 3.1064659088143424e-06,
1612
+ "loss": 1.5048,
1613
+ "step": 1135
1614
+ },
1615
+ {
1616
+ "epoch": 1.5426251691474966,
1617
+ "grad_norm": 2.359375,
1618
+ "learning_rate": 3.0213904451164623e-06,
1619
+ "loss": 1.5161,
1620
+ "step": 1140
1621
+ },
1622
+ {
1623
+ "epoch": 1.5493910690121786,
1624
+ "grad_norm": 2.421875,
1625
+ "learning_rate": 2.937288401552063e-06,
1626
+ "loss": 1.5037,
1627
+ "step": 1145
1628
+ },
1629
+ {
1630
+ "epoch": 1.5561569688768606,
1631
+ "grad_norm": 2.296875,
1632
+ "learning_rate": 2.8541715092005097e-06,
1633
+ "loss": 1.5058,
1634
+ "step": 1150
1635
+ },
1636
+ {
1637
+ "epoch": 1.5629228687415426,
1638
+ "grad_norm": 2.421875,
1639
+ "learning_rate": 2.7720513617260857e-06,
1640
+ "loss": 1.5019,
1641
+ "step": 1155
1642
+ },
1643
+ {
1644
+ "epoch": 1.5696887686062246,
1645
+ "grad_norm": 2.859375,
1646
+ "learning_rate": 2.690939413760887e-06,
1647
+ "loss": 1.5006,
1648
+ "step": 1160
1649
+ },
1650
+ {
1651
+ "epoch": 1.5764546684709067,
1652
+ "grad_norm": 2.40625,
1653
+ "learning_rate": 2.610846979307016e-06,
1654
+ "loss": 1.5078,
1655
+ "step": 1165
1656
+ },
1657
+ {
1658
+ "epoch": 1.5832205683355887,
1659
+ "grad_norm": 2.546875,
1660
+ "learning_rate": 2.5317852301584642e-06,
1661
+ "loss": 1.4985,
1662
+ "step": 1170
1663
+ },
1664
+ {
1665
+ "epoch": 1.5899864682002707,
1666
+ "grad_norm": 2.34375,
1667
+ "learning_rate": 2.4537651943427666e-06,
1668
+ "loss": 1.4675,
1669
+ "step": 1175
1670
+ },
1671
+ {
1672
+ "epoch": 1.5967523680649527,
1673
+ "grad_norm": 2.4375,
1674
+ "learning_rate": 2.376797754582785e-06,
1675
+ "loss": 1.4989,
1676
+ "step": 1180
1677
+ },
1678
+ {
1679
+ "epoch": 1.6035182679296347,
1680
+ "grad_norm": 2.359375,
1681
+ "learning_rate": 2.3008936467786815e-06,
1682
+ "loss": 1.4755,
1683
+ "step": 1185
1684
+ },
1685
+ {
1686
+ "epoch": 1.6102841677943167,
1687
+ "grad_norm": 2.359375,
1688
+ "learning_rate": 2.226063458510428e-06,
1689
+ "loss": 1.4977,
1690
+ "step": 1190
1691
+ },
1692
+ {
1693
+ "epoch": 1.6170500676589987,
1694
+ "grad_norm": 2.3125,
1695
+ "learning_rate": 2.152317627560979e-06,
1696
+ "loss": 1.4965,
1697
+ "step": 1195
1698
+ },
1699
+ {
1700
+ "epoch": 1.6238159675236807,
1701
+ "grad_norm": 2.34375,
1702
+ "learning_rate": 2.0796664404603416e-06,
1703
+ "loss": 1.4801,
1704
+ "step": 1200
1705
+ },
1706
+ {
1707
+ "epoch": 1.6305818673883627,
1708
+ "grad_norm": 2.421875,
1709
+ "learning_rate": 2.008120031050753e-06,
1710
+ "loss": 1.5145,
1711
+ "step": 1205
1712
+ },
1713
+ {
1714
+ "epoch": 1.6373477672530448,
1715
+ "grad_norm": 2.46875,
1716
+ "learning_rate": 1.9376883790731417e-06,
1717
+ "loss": 1.5032,
1718
+ "step": 1210
1719
+ },
1720
+ {
1721
+ "epoch": 1.6441136671177268,
1722
+ "grad_norm": 2.484375,
1723
+ "learning_rate": 1.8683813087750901e-06,
1724
+ "loss": 1.4968,
1725
+ "step": 1215
1726
+ },
1727
+ {
1728
+ "epoch": 1.6508795669824088,
1729
+ "grad_norm": 2.375,
1730
+ "learning_rate": 1.8002084875404935e-06,
1731
+ "loss": 1.512,
1732
+ "step": 1220
1733
+ },
1734
+ {
1735
+ "epoch": 1.6576454668470908,
1736
+ "grad_norm": 2.40625,
1737
+ "learning_rate": 1.7331794245410926e-06,
1738
+ "loss": 1.5059,
1739
+ "step": 1225
1740
+ },
1741
+ {
1742
+ "epoch": 1.6644113667117728,
1743
+ "grad_norm": 2.484375,
1744
+ "learning_rate": 1.6673034694100655e-06,
1745
+ "loss": 1.5124,
1746
+ "step": 1230
1747
+ },
1748
+ {
1749
+ "epoch": 1.6711772665764548,
1750
+ "grad_norm": 2.296875,
1751
+ "learning_rate": 1.6025898109378967e-06,
1752
+ "loss": 1.5248,
1753
+ "step": 1235
1754
+ },
1755
+ {
1756
+ "epoch": 1.6779431664411368,
1757
+ "grad_norm": 2.265625,
1758
+ "learning_rate": 1.5390474757906449e-06,
1759
+ "loss": 1.5044,
1760
+ "step": 1240
1761
+ },
1762
+ {
1763
+ "epoch": 1.6847090663058186,
1764
+ "grad_norm": 2.40625,
1765
+ "learning_rate": 1.4766853272508785e-06,
1766
+ "loss": 1.5081,
1767
+ "step": 1245
1768
+ },
1769
+ {
1770
+ "epoch": 1.6914749661705006,
1771
+ "grad_norm": 2.375,
1772
+ "learning_rate": 1.4155120639813392e-06,
1773
+ "loss": 1.4982,
1774
+ "step": 1250
1775
+ },
1776
+ {
1777
+ "epoch": 1.6982408660351827,
1778
+ "grad_norm": 2.375,
1779
+ "learning_rate": 1.3555362188116173e-06,
1780
+ "loss": 1.4716,
1781
+ "step": 1255
1782
+ },
1783
+ {
1784
+ "epoch": 1.7050067658998647,
1785
+ "grad_norm": 2.484375,
1786
+ "learning_rate": 1.2967661575479318e-06,
1787
+ "loss": 1.5055,
1788
+ "step": 1260
1789
+ },
1790
+ {
1791
+ "epoch": 1.7117726657645467,
1792
+ "grad_norm": 2.328125,
1793
+ "learning_rate": 1.2392100778062123e-06,
1794
+ "loss": 1.5066,
1795
+ "step": 1265
1796
+ },
1797
+ {
1798
+ "epoch": 1.7185385656292287,
1799
+ "grad_norm": 2.359375,
1800
+ "learning_rate": 1.1828760078686563e-06,
1801
+ "loss": 1.5146,
1802
+ "step": 1270
1803
+ },
1804
+ {
1805
+ "epoch": 1.7253044654939107,
1806
+ "grad_norm": 2.34375,
1807
+ "learning_rate": 1.127771805563882e-06,
1808
+ "loss": 1.5009,
1809
+ "step": 1275
1810
+ },
1811
+ {
1812
+ "epoch": 1.7320703653585927,
1813
+ "grad_norm": 2.4375,
1814
+ "learning_rate": 1.0739051571708736e-06,
1815
+ "loss": 1.5149,
1816
+ "step": 1280
1817
+ },
1818
+ {
1819
+ "epoch": 1.7388362652232747,
1820
+ "grad_norm": 2.359375,
1821
+ "learning_rate": 1.0212835763468488e-06,
1822
+ "loss": 1.5055,
1823
+ "step": 1285
1824
+ },
1825
+ {
1826
+ "epoch": 1.7456021650879567,
1827
+ "grad_norm": 2.328125,
1828
+ "learning_rate": 9.699144030792163e-07,
1829
+ "loss": 1.5028,
1830
+ "step": 1290
1831
+ },
1832
+ {
1833
+ "epoch": 1.7523680649526387,
1834
+ "grad_norm": 2.34375,
1835
+ "learning_rate": 9.198048026617323e-07,
1836
+ "loss": 1.4877,
1837
+ "step": 1295
1838
+ },
1839
+ {
1840
+ "epoch": 1.7591339648173205,
1841
+ "grad_norm": 2.359375,
1842
+ "learning_rate": 8.709617646950563e-07,
1843
+ "loss": 1.5085,
1844
+ "step": 1300
1845
+ },
1846
+ {
1847
+ "epoch": 1.7658998646820026,
1848
+ "grad_norm": 2.203125,
1849
+ "learning_rate": 8.233921021117863e-07,
1850
+ "loss": 1.4859,
1851
+ "step": 1305
1852
+ },
1853
+ {
1854
+ "epoch": 1.7726657645466846,
1855
+ "grad_norm": 2.40625,
1856
+ "learning_rate": 7.771024502261526e-07,
1857
+ "loss": 1.5006,
1858
+ "step": 1310
1859
+ },
1860
+ {
1861
+ "epoch": 1.7794316644113666,
1862
+ "grad_norm": 2.359375,
1863
+ "learning_rate": 7.320992658084891e-07,
1864
+ "loss": 1.5159,
1865
+ "step": 1315
1866
+ },
1867
+ {
1868
+ "epoch": 1.7861975642760486,
1869
+ "grad_norm": 2.421875,
1870
+ "learning_rate": 6.883888261845917e-07,
1871
+ "loss": 1.5117,
1872
+ "step": 1320
1873
+ },
1874
+ {
1875
+ "epoch": 1.7929634641407306,
1876
+ "grad_norm": 2.4375,
1877
+ "learning_rate": 6.459772283601218e-07,
1878
+ "loss": 1.5026,
1879
+ "step": 1325
1880
+ },
1881
+ {
1882
+ "epoch": 1.7997293640054126,
1883
+ "grad_norm": 2.46875,
1884
+ "learning_rate": 6.048703881701579e-07,
1885
+ "loss": 1.4971,
1886
+ "step": 1330
1887
+ },
1888
+ {
1889
+ "epoch": 1.8064952638700946,
1890
+ "grad_norm": 2.40625,
1891
+ "learning_rate": 5.650740394540255e-07,
1892
+ "loss": 1.4969,
1893
+ "step": 1335
1894
+ },
1895
+ {
1896
+ "epoch": 1.8132611637347766,
1897
+ "grad_norm": 2.359375,
1898
+ "learning_rate": 5.265937332554849e-07,
1899
+ "loss": 1.4996,
1900
+ "step": 1340
1901
+ },
1902
+ {
1903
+ "epoch": 1.8200270635994586,
1904
+ "grad_norm": 2.375,
1905
+ "learning_rate": 4.894348370484648e-07,
1906
+ "loss": 1.5124,
1907
+ "step": 1345
1908
+ },
1909
+ {
1910
+ "epoch": 1.8267929634641407,
1911
+ "grad_norm": 2.5,
1912
+ "learning_rate": 4.5360253398834765e-07,
1913
+ "loss": 1.5047,
1914
+ "step": 1350
1915
+ },
1916
+ {
1917
+ "epoch": 1.8335588633288227,
1918
+ "grad_norm": 2.359375,
1919
+ "learning_rate": 4.1910182218900977e-07,
1920
+ "loss": 1.5211,
1921
+ "step": 1355
1922
+ },
1923
+ {
1924
+ "epoch": 1.8403247631935047,
1925
+ "grad_norm": 2.34375,
1926
+ "learning_rate": 3.8593751402563715e-07,
1927
+ "loss": 1.5189,
1928
+ "step": 1360
1929
+ },
1930
+ {
1931
+ "epoch": 1.8470906630581867,
1932
+ "grad_norm": 2.40625,
1933
+ "learning_rate": 3.541142354634808e-07,
1934
+ "loss": 1.4941,
1935
+ "step": 1365
1936
+ },
1937
+ {
1938
+ "epoch": 1.8538565629228687,
1939
+ "grad_norm": 2.5,
1940
+ "learning_rate": 3.236364254125868e-07,
1941
+ "loss": 1.47,
1942
+ "step": 1370
1943
+ },
1944
+ {
1945
+ "epoch": 1.8606224627875507,
1946
+ "grad_norm": 2.421875,
1947
+ "learning_rate": 2.945083351086331e-07,
1948
+ "loss": 1.5138,
1949
+ "step": 1375
1950
+ },
1951
+ {
1952
+ "epoch": 1.8673883626522327,
1953
+ "grad_norm": 2.328125,
1954
+ "learning_rate": 2.667340275199426e-07,
1955
+ "loss": 1.5025,
1956
+ "step": 1380
1957
+ },
1958
+ {
1959
+ "epoch": 1.8741542625169147,
1960
+ "grad_norm": 2.5,
1961
+ "learning_rate": 2.4031737678074985e-07,
1962
+ "loss": 1.5216,
1963
+ "step": 1385
1964
+ },
1965
+ {
1966
+ "epoch": 1.8809201623815968,
1967
+ "grad_norm": 2.296875,
1968
+ "learning_rate": 2.1526206765082104e-07,
1969
+ "loss": 1.5059,
1970
+ "step": 1390
1971
+ },
1972
+ {
1973
+ "epoch": 1.8876860622462788,
1974
+ "grad_norm": 2.390625,
1975
+ "learning_rate": 1.915715950014696e-07,
1976
+ "loss": 1.4874,
1977
+ "step": 1395
1978
+ },
1979
+ {
1980
+ "epoch": 1.8944519621109608,
1981
+ "grad_norm": 2.515625,
1982
+ "learning_rate": 1.6924926332807956e-07,
1983
+ "loss": 1.5109,
1984
+ "step": 1400
1985
+ },
1986
+ {
1987
+ "epoch": 1.9012178619756428,
1988
+ "grad_norm": 2.3125,
1989
+ "learning_rate": 1.4829818628916525e-07,
1990
+ "loss": 1.4955,
1991
+ "step": 1405
1992
+ },
1993
+ {
1994
+ "epoch": 1.9079837618403248,
1995
+ "grad_norm": 2.25,
1996
+ "learning_rate": 1.287212862720677e-07,
1997
+ "loss": 1.5023,
1998
+ "step": 1410
1999
+ },
2000
+ {
2001
+ "epoch": 1.9147496617050068,
2002
+ "grad_norm": 2.390625,
2003
+ "learning_rate": 1.1052129398531508e-07,
2004
+ "loss": 1.4818,
2005
+ "step": 1415
2006
+ },
2007
+ {
2008
+ "epoch": 1.9215155615696888,
2009
+ "grad_norm": 2.25,
2010
+ "learning_rate": 9.370074807772966e-08,
2011
+ "loss": 1.4966,
2012
+ "step": 1420
2013
+ },
2014
+ {
2015
+ "epoch": 1.9282814614343708,
2016
+ "grad_norm": 2.375,
2017
+ "learning_rate": 7.826199478431551e-08,
2018
+ "loss": 1.4801,
2019
+ "step": 1425
2020
+ },
2021
+ {
2022
+ "epoch": 1.9350473612990529,
2023
+ "grad_norm": 2.390625,
2024
+ "learning_rate": 6.420718759900357e-08,
2025
+ "loss": 1.5037,
2026
+ "step": 1430
2027
+ },
2028
+ {
2029
+ "epoch": 1.9418132611637349,
2030
+ "grad_norm": 2.28125,
2031
+ "learning_rate": 5.1538286974254225e-08,
2032
+ "loss": 1.4796,
2033
+ "step": 1435
2034
+ },
2035
+ {
2036
+ "epoch": 1.9485791610284169,
2037
+ "grad_norm": 2.421875,
2038
+ "learning_rate": 4.025706004760932e-08,
2039
+ "loss": 1.5041,
2040
+ "step": 1440
2041
+ },
2042
+ {
2043
+ "epoch": 1.955345060893099,
2044
+ "grad_norm": 2.359375,
2045
+ "learning_rate": 3.036508039520047e-08,
2046
+ "loss": 1.4933,
2047
+ "step": 1445
2048
+ },
2049
+ {
2050
+ "epoch": 1.962110960757781,
2051
+ "grad_norm": 2.375,
2052
+ "learning_rate": 2.1863727812254653e-08,
2053
+ "loss": 1.4872,
2054
+ "step": 1450
2055
+ },
2056
+ {
2057
+ "epoch": 1.968876860622463,
2058
+ "grad_norm": 2.4375,
2059
+ "learning_rate": 1.4754188120631453e-08,
2060
+ "loss": 1.4953,
2061
+ "step": 1455
2062
+ },
2063
+ {
2064
+ "epoch": 1.975642760487145,
2065
+ "grad_norm": 2.296875,
2066
+ "learning_rate": 9.03745300341874e-09,
2067
+ "loss": 1.5199,
2068
+ "step": 1460
2069
+ },
2070
+ {
2071
+ "epoch": 1.982408660351827,
2072
+ "grad_norm": 2.328125,
2073
+ "learning_rate": 4.714319866607753e-09,
2074
+ "loss": 1.519,
2075
+ "step": 1465
2076
+ },
2077
+ {
2078
+ "epoch": 1.989174560216509,
2079
+ "grad_norm": 2.375,
2080
+ "learning_rate": 1.7853917278631838e-09,
2081
+ "loss": 1.5044,
2082
+ "step": 1470
2083
+ },
2084
+ {
2085
+ "epoch": 1.9959404600811907,
2086
+ "grad_norm": 2.4375,
2087
+ "learning_rate": 2.5107713241045994e-10,
2088
+ "loss": 1.5051,
2089
+ "step": 1475
2090
+ },
2091
+ {
2092
+ "epoch": 2.0,
2093
+ "eval_loss": 1.691962480545044,
2094
+ "eval_runtime": 28.7807,
2095
+ "eval_samples_per_second": 15.114,
2096
+ "eval_steps_per_second": 1.911,
2097
+ "step": 1478
2098
  },
2099
  {
2100
+ "epoch": 2.0,
2101
+ "step": 1478,
2102
+ "total_flos": 1.502823063832494e+17,
2103
+ "train_loss": 1.5822677802652403,
2104
+ "train_runtime": 3974.9532,
2105
+ "train_samples_per_second": 2.974,
2106
+ "train_steps_per_second": 0.372
2107
  }
2108
  ],
2109
  "logging_steps": 5,
2110
+ "max_steps": 1478,
2111
  "num_input_tokens_seen": 0,
2112
  "num_train_epochs": 2,
2113
  "save_steps": 100,
 
2123
  "attributes": {}
2124
  }
2125
  },
2126
+ "total_flos": 1.502823063832494e+17,
2127
+ "train_batch_size": 4,
2128
  "trial_name": null,
2129
  "trial_params": null
2130
  }