lewtun HF staff commited on
Commit
c3160e9
1 Parent(s): 10ee951

Model save

Browse files
README.md CHANGED
@@ -2,13 +2,14 @@
2
  license: apache-2.0
3
  base_model: mistralai/Mistral-7B-v0.1
4
  tags:
 
 
5
  - generated_from_trainer
6
- - alignment-handbook
 
7
  model-index:
8
  - name: zephyr-7b-sft-full
9
  results: []
10
- datasets:
11
- - HuggingFaceH4/ultrachat_200k
12
  ---
13
 
14
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -16,9 +17,9 @@ should probably proofread and complete it, then remove this comment. -->
16
 
17
  # zephyr-7b-sft-full
18
 
19
- This model is a fine-tuned version of [mistralai/Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1) on the [HuggingFaceH4/ultrachat_200k](https://huggingface.co/datasets/HuggingFaceH4/ultrachat_200k) dataset.
20
  It achieves the following results on the evaluation set:
21
- - Loss: 0.9324
22
 
23
  ## Model description
24
 
@@ -38,28 +39,28 @@ More information needed
38
 
39
  The following hyperparameters were used during training:
40
  - learning_rate: 2e-05
41
- - train_batch_size: 32
42
- - eval_batch_size: 16
43
  - seed: 42
44
  - distributed_type: multi-GPU
45
  - num_devices: 8
46
- - gradient_accumulation_steps: 2
47
- - total_train_batch_size: 512
48
- - total_eval_batch_size: 128
49
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
50
  - lr_scheduler_type: cosine
 
51
  - num_epochs: 1
52
 
53
  ### Training results
54
 
55
  | Training Loss | Epoch | Step | Validation Loss |
56
  |:-------------:|:-----:|:----:|:---------------:|
57
- | 0.9292 | 0.67 | 272 | 0.9323 |
58
 
59
 
60
  ### Framework versions
61
 
62
- - Transformers 4.35.0
63
- - Pytorch 2.1.0+cu118
64
  - Datasets 2.14.6
65
- - Tokenizers 0.14.1
 
2
  license: apache-2.0
3
  base_model: mistralai/Mistral-7B-v0.1
4
  tags:
5
+ - trl
6
+ - sft
7
  - generated_from_trainer
8
+ datasets:
9
+ - generator
10
  model-index:
11
  - name: zephyr-7b-sft-full
12
  results: []
 
 
13
  ---
14
 
15
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
 
17
 
18
  # zephyr-7b-sft-full
19
 
20
+ This model is a fine-tuned version of [mistralai/Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1) on the generator dataset.
21
  It achieves the following results on the evaluation set:
22
+ - Loss: 0.9553
23
 
24
  ## Model description
25
 
 
39
 
40
  The following hyperparameters were used during training:
41
  - learning_rate: 2e-05
42
+ - train_batch_size: 16
43
+ - eval_batch_size: 8
44
  - seed: 42
45
  - distributed_type: multi-GPU
46
  - num_devices: 8
47
+ - total_train_batch_size: 128
48
+ - total_eval_batch_size: 64
 
49
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
50
  - lr_scheduler_type: cosine
51
+ - lr_scheduler_warmup_ratio: 0.1
52
  - num_epochs: 1
53
 
54
  ### Training results
55
 
56
  | Training Loss | Epoch | Step | Validation Loss |
57
  |:-------------:|:-----:|:----:|:---------------:|
58
+ | 0.9511 | 1.0 | 1068 | 0.9553 |
59
 
60
 
61
  ### Framework versions
62
 
63
+ - Transformers 4.36.2
64
+ - Pytorch 2.1.2+cu121
65
  - Datasets 2.14.6
66
+ - Tokenizers 0.15.0
all_results.json CHANGED
@@ -1,13 +1,13 @@
1
  {
2
- "epoch": 0.67,
3
- "eval_loss": 0.932437002658844,
4
- "eval_runtime": 331.0457,
5
  "eval_samples": 23110,
6
- "eval_samples_per_second": 69.809,
7
- "eval_steps_per_second": 0.547,
8
- "train_loss": 0.9717322877224754,
9
- "train_runtime": 11645.5121,
10
  "train_samples": 207865,
11
- "train_samples_per_second": 17.849,
12
- "train_steps_per_second": 0.035
13
  }
 
1
  {
2
+ "epoch": 1.0,
3
+ "eval_loss": 0.9552798271179199,
4
+ "eval_runtime": 138.0032,
5
  "eval_samples": 23110,
6
+ "eval_samples_per_second": 109.592,
7
+ "eval_steps_per_second": 1.717,
8
+ "train_loss": 0.99920238649354,
9
+ "train_runtime": 5476.1677,
10
  "train_samples": 207865,
11
+ "train_samples_per_second": 24.954,
12
+ "train_steps_per_second": 0.195
13
  }
eval_results.json CHANGED
@@ -1,8 +1,8 @@
1
  {
2
- "epoch": 0.67,
3
- "eval_loss": 0.932437002658844,
4
- "eval_runtime": 331.0457,
5
  "eval_samples": 23110,
6
- "eval_samples_per_second": 69.809,
7
- "eval_steps_per_second": 0.547
8
  }
 
1
  {
2
+ "epoch": 1.0,
3
+ "eval_loss": 0.9552798271179199,
4
+ "eval_runtime": 138.0032,
5
  "eval_samples": 23110,
6
+ "eval_samples_per_second": 109.592,
7
+ "eval_steps_per_second": 1.717
8
  }
generation_config.json CHANGED
@@ -2,5 +2,5 @@
2
  "_from_model_config": true,
3
  "bos_token_id": 1,
4
  "eos_token_id": 2,
5
- "transformers_version": "4.35.0"
6
  }
 
2
  "_from_model_config": true,
3
  "bos_token_id": 1,
4
  "eos_token_id": 2,
5
+ "transformers_version": "4.36.2"
6
  }
model-00001-of-00003.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:68b1e135ad66cee90fef5c9335e8c80b8e60b16254ff5f4e88d3369ebcce96a4
3
  size 4943162336
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:790beac063dfce40753797199b3ecddb35a304cb429b52224980783d03bc5ce2
3
  size 4943162336
model-00002-of-00003.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:8df5a386b05f0393662aaa5d39d8dd052a22a366fe6f5cd42a0c7bc940898d6e
3
  size 4999819336
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:16e773312de757cd06bfde84adf7d13dde3e46542b01f4350b2175a100aa52a2
3
  size 4999819336
model-00003-of-00003.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:202c8e90c2ac2219a17c1bce35622a063977658808c635d326433237022d2ea3
3
  size 4540516344
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:52b9fc1b620a7b3d326852ee236af0836310b5c8276a1f845c00f752419e5497
3
  size 4540516344
runs/Jan08_12-40-07_ip-26-0-175-170/events.out.tfevents.1704718278.ip-26-0-175-170.1661731.0 CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:3094004129e89805d59299de3480c0eda067bb5948266f132fa149403d67f384
3
- size 35929
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:44acf8a306a7de7eb650f833265fd70cd76124342f74f5283e316a8b38f89fc4
3
+ size 38595
runs/Jan08_12-40-07_ip-26-0-175-170/events.out.tfevents.1704723892.ip-26-0-175-170.1661731.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d5474ae84796dd04fe54b37e7a66acbcd9b7ad047dac5d9d3096b2abc43536ef
3
+ size 359
train_results.json CHANGED
@@ -1,8 +1,8 @@
1
  {
2
- "epoch": 0.67,
3
- "train_loss": 0.9717322877224754,
4
- "train_runtime": 11645.5121,
5
  "train_samples": 207865,
6
- "train_samples_per_second": 17.849,
7
- "train_steps_per_second": 0.035
8
  }
 
1
  {
2
+ "epoch": 1.0,
3
+ "train_loss": 0.99920238649354,
4
+ "train_runtime": 5476.1677,
5
  "train_samples": 207865,
6
+ "train_samples_per_second": 24.954,
7
+ "train_steps_per_second": 0.195
8
  }
trainer_state.json CHANGED
@@ -1,366 +1,1322 @@
1
  {
2
  "best_metric": null,
3
  "best_model_checkpoint": null,
4
- "epoch": 0.6699507389162561,
5
  "eval_steps": 500,
6
- "global_step": 272,
7
  "is_hyper_param_search": false,
8
  "is_local_process_zero": true,
9
  "is_world_process_zero": true,
10
  "log_history": [
11
  {
12
  "epoch": 0.0,
13
- "learning_rate": 1.9999700625010444e-05,
14
- "loss": 1.1295,
15
  "step": 1
16
  },
17
  {
18
- "epoch": 0.01,
19
- "learning_rate": 1.999251652147735e-05,
20
- "loss": 1.6995,
21
  "step": 5
22
  },
23
  {
24
- "epoch": 0.02,
25
- "learning_rate": 1.997007728639956e-05,
26
- "loss": 1.1502,
27
  "step": 10
28
  },
29
  {
30
- "epoch": 0.04,
31
- "learning_rate": 1.9932715879473385e-05,
32
- "loss": 1.0714,
33
  "step": 15
34
  },
35
  {
36
- "epoch": 0.05,
37
- "learning_rate": 1.9880488219356086e-05,
38
- "loss": 1.0487,
39
  "step": 20
40
  },
41
  {
42
- "epoch": 0.06,
43
- "learning_rate": 1.981347247496222e-05,
44
- "loss": 1.0138,
45
  "step": 25
46
  },
47
  {
48
- "epoch": 0.07,
49
- "learning_rate": 1.973176894846855e-05,
50
- "loss": 0.9986,
51
  "step": 30
52
  },
53
  {
54
- "epoch": 0.09,
55
- "learning_rate": 1.963549992519223e-05,
56
- "loss": 0.9962,
57
  "step": 35
58
  },
59
  {
60
- "epoch": 0.1,
61
- "learning_rate": 1.9524809490566878e-05,
62
- "loss": 0.9805,
63
  "step": 40
64
  },
65
  {
66
- "epoch": 0.11,
67
- "learning_rate": 1.939986331449053e-05,
68
- "loss": 0.9783,
69
  "step": 45
70
  },
71
  {
72
- "epoch": 0.12,
73
- "learning_rate": 1.926084840336821e-05,
74
- "loss": 0.9814,
75
  "step": 50
76
  },
77
  {
78
- "epoch": 0.14,
79
- "learning_rate": 1.910797282022027e-05,
80
- "loss": 0.9725,
81
  "step": 55
82
  },
83
  {
84
- "epoch": 0.15,
85
- "learning_rate": 1.894146537327533e-05,
86
- "loss": 0.9655,
87
  "step": 60
88
  },
89
  {
90
- "epoch": 0.16,
91
- "learning_rate": 1.8761575273514005e-05,
92
- "loss": 0.9601,
93
  "step": 65
94
  },
95
  {
96
- "epoch": 0.17,
97
- "learning_rate": 1.8568571761675893e-05,
98
- "loss": 0.9532,
99
  "step": 70
100
  },
101
  {
102
- "epoch": 0.18,
103
- "learning_rate": 1.8362743705288127e-05,
104
- "loss": 0.9606,
105
  "step": 75
106
  },
107
  {
108
- "epoch": 0.2,
109
- "learning_rate": 1.814439916631857e-05,
110
- "loss": 0.9682,
111
  "step": 80
112
  },
113
  {
114
- "epoch": 0.21,
115
- "learning_rate": 1.791386494010081e-05,
116
- "loss": 0.9642,
117
  "step": 85
118
  },
119
  {
120
- "epoch": 0.22,
121
- "learning_rate": 1.7671486066220965e-05,
122
- "loss": 0.9549,
123
  "step": 90
124
  },
125
  {
126
- "epoch": 0.23,
127
- "learning_rate": 1.7417625312098453e-05,
128
- "loss": 0.9444,
129
  "step": 95
130
  },
131
  {
132
- "epoch": 0.25,
133
- "learning_rate": 1.7152662630033506e-05,
134
- "loss": 0.9562,
135
  "step": 100
136
  },
137
  {
138
- "epoch": 0.26,
139
- "learning_rate": 1.6876994588534234e-05,
140
- "loss": 0.9707,
141
  "step": 105
142
  },
143
  {
144
- "epoch": 0.27,
145
- "learning_rate": 1.659103377877423e-05,
146
- "loss": 0.9498,
147
  "step": 110
148
  },
149
  {
150
- "epoch": 0.28,
151
- "learning_rate": 1.629520819706912e-05,
152
- "loss": 0.9464,
153
  "step": 115
154
  },
155
  {
156
- "epoch": 0.3,
157
- "learning_rate": 1.598996060429634e-05,
158
- "loss": 0.9516,
159
  "step": 120
160
  },
161
  {
162
- "epoch": 0.31,
163
- "learning_rate": 1.56757478632168e-05,
164
- "loss": 0.9486,
165
  "step": 125
166
  },
167
  {
168
- "epoch": 0.32,
169
- "learning_rate": 1.5353040254690396e-05,
170
- "loss": 0.9566,
171
  "step": 130
172
  },
173
  {
174
- "epoch": 0.33,
175
- "learning_rate": 1.5022320773808612e-05,
176
- "loss": 0.9569,
177
  "step": 135
178
  },
179
  {
180
- "epoch": 0.34,
181
- "learning_rate": 1.4684084406997903e-05,
182
- "loss": 0.9427,
183
  "step": 140
184
  },
185
  {
186
- "epoch": 0.36,
187
- "learning_rate": 1.4338837391175582e-05,
188
- "loss": 0.9485,
189
  "step": 145
190
  },
191
  {
192
- "epoch": 0.37,
193
- "learning_rate": 1.3987096456067236e-05,
194
- "loss": 0.9444,
195
  "step": 150
196
  },
197
  {
198
- "epoch": 0.38,
199
- "learning_rate": 1.3629388050819547e-05,
200
- "loss": 0.9396,
201
  "step": 155
202
  },
203
  {
204
- "epoch": 0.39,
205
- "learning_rate": 1.3266247556066122e-05,
206
- "loss": 0.9289,
207
  "step": 160
208
  },
209
  {
210
- "epoch": 0.41,
211
- "learning_rate": 1.2898218482625606e-05,
212
- "loss": 0.9454,
213
  "step": 165
214
  },
215
  {
216
- "epoch": 0.42,
217
- "learning_rate": 1.252585165803135e-05,
218
- "loss": 0.9463,
219
  "step": 170
220
  },
221
  {
222
- "epoch": 0.43,
223
- "learning_rate": 1.2149704402110243e-05,
224
- "loss": 0.9323,
225
  "step": 175
226
  },
227
  {
228
- "epoch": 0.44,
229
- "learning_rate": 1.1770339692844484e-05,
230
- "loss": 0.9382,
231
  "step": 180
232
  },
233
  {
234
- "epoch": 0.46,
235
- "learning_rate": 1.1388325323764889e-05,
236
- "loss": 0.9476,
237
  "step": 185
238
  },
239
  {
240
- "epoch": 0.47,
241
- "learning_rate": 1.1004233054136726e-05,
242
- "loss": 0.9421,
243
  "step": 190
244
  },
245
  {
246
- "epoch": 0.48,
247
- "learning_rate": 1.0618637753210086e-05,
248
- "loss": 0.9421,
249
  "step": 195
250
  },
251
  {
252
- "epoch": 0.49,
253
- "learning_rate": 1.0232116539815558e-05,
254
- "loss": 0.9324,
255
  "step": 200
256
  },
257
  {
258
- "epoch": 0.5,
259
- "learning_rate": 9.845247918592937e-06,
260
- "loss": 0.937,
261
  "step": 205
262
  },
263
  {
264
- "epoch": 0.52,
265
- "learning_rate": 9.458610914145826e-06,
266
- "loss": 0.935,
267
  "step": 210
268
  },
269
  {
270
- "epoch": 0.53,
271
- "learning_rate": 9.072784204417995e-06,
272
- "loss": 0.9461,
273
  "step": 215
274
  },
275
  {
276
- "epoch": 0.54,
277
- "learning_rate": 8.688345254588579e-06,
278
- "loss": 0.9346,
279
  "step": 220
280
  },
281
  {
282
- "epoch": 0.55,
283
- "learning_rate": 8.305869452782446e-06,
284
- "loss": 0.9384,
285
  "step": 225
286
  },
287
  {
288
- "epoch": 0.57,
289
- "learning_rate": 7.92592924888925e-06,
290
- "loss": 0.9362,
291
  "step": 230
292
  },
293
  {
294
- "epoch": 0.58,
295
- "learning_rate": 7.549093297780133e-06,
296
- "loss": 0.9331,
297
  "step": 235
298
  },
299
  {
300
- "epoch": 0.59,
301
- "learning_rate": 7.175925608204428e-06,
302
- "loss": 0.9396,
303
  "step": 240
304
  },
305
  {
306
- "epoch": 0.6,
307
- "learning_rate": 6.806984698640202e-06,
308
- "loss": 0.9421,
309
  "step": 245
310
  },
311
  {
312
- "epoch": 0.62,
313
- "learning_rate": 6.442822761362015e-06,
314
- "loss": 0.9285,
315
  "step": 250
316
  },
317
  {
318
- "epoch": 0.63,
319
- "learning_rate": 6.083984835977154e-06,
320
- "loss": 0.9226,
321
  "step": 255
322
  },
323
  {
324
- "epoch": 0.64,
325
- "learning_rate": 5.731007993667155e-06,
326
- "loss": 0.9372,
327
  "step": 260
328
  },
329
  {
330
- "epoch": 0.65,
331
- "learning_rate": 5.38442053335571e-06,
332
- "loss": 0.9264,
333
  "step": 265
334
  },
335
  {
336
- "epoch": 0.67,
337
- "learning_rate": 5.044741191005908e-06,
338
- "loss": 0.9292,
339
  "step": 270
340
  },
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
341
  {
342
  "epoch": 0.67,
343
- "eval_loss": 0.9322898387908936,
344
- "eval_runtime": 337.0923,
345
- "eval_samples_per_second": 68.557,
346
- "eval_steps_per_second": 0.537,
347
- "step": 272
348
  },
349
  {
350
  "epoch": 0.67,
351
- "step": 272,
352
- "total_flos": 455322233733120.0,
353
- "train_loss": 0.9717322877224754,
354
- "train_runtime": 11645.5121,
355
- "train_samples_per_second": 17.849,
356
- "train_steps_per_second": 0.035
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
357
  }
358
  ],
359
  "logging_steps": 5,
360
- "max_steps": 406,
 
361
  "num_train_epochs": 1,
362
- "save_steps": 500,
363
- "total_flos": 455322233733120.0,
 
364
  "trial_name": null,
365
  "trial_params": null
366
  }
 
1
  {
2
  "best_metric": null,
3
  "best_model_checkpoint": null,
4
+ "epoch": 1.0,
5
  "eval_steps": 500,
6
+ "global_step": 1068,
7
  "is_hyper_param_search": false,
8
  "is_local_process_zero": true,
9
  "is_world_process_zero": true,
10
  "log_history": [
11
  {
12
  "epoch": 0.0,
13
+ "learning_rate": 1.8691588785046729e-07,
14
+ "loss": 1.862,
15
  "step": 1
16
  },
17
  {
18
+ "epoch": 0.0,
19
+ "learning_rate": 9.345794392523365e-07,
20
+ "loss": 1.7887,
21
  "step": 5
22
  },
23
  {
24
+ "epoch": 0.01,
25
+ "learning_rate": 1.869158878504673e-06,
26
+ "loss": 1.5242,
27
  "step": 10
28
  },
29
  {
30
+ "epoch": 0.01,
31
+ "learning_rate": 2.8037383177570094e-06,
32
+ "loss": 1.2751,
33
  "step": 15
34
  },
35
  {
36
+ "epoch": 0.02,
37
+ "learning_rate": 3.738317757009346e-06,
38
+ "loss": 1.208,
39
  "step": 20
40
  },
41
  {
42
+ "epoch": 0.02,
43
+ "learning_rate": 4.6728971962616825e-06,
44
+ "loss": 1.1479,
45
  "step": 25
46
  },
47
  {
48
+ "epoch": 0.03,
49
+ "learning_rate": 5.607476635514019e-06,
50
+ "loss": 1.1126,
51
  "step": 30
52
  },
53
  {
54
+ "epoch": 0.03,
55
+ "learning_rate": 6.542056074766355e-06,
56
+ "loss": 1.0975,
57
  "step": 35
58
  },
59
  {
60
+ "epoch": 0.04,
61
+ "learning_rate": 7.476635514018692e-06,
62
+ "loss": 1.0781,
63
  "step": 40
64
  },
65
  {
66
+ "epoch": 0.04,
67
+ "learning_rate": 8.411214953271028e-06,
68
+ "loss": 1.0466,
69
  "step": 45
70
  },
71
  {
72
+ "epoch": 0.05,
73
+ "learning_rate": 9.345794392523365e-06,
74
+ "loss": 1.0434,
75
  "step": 50
76
  },
77
  {
78
+ "epoch": 0.05,
79
+ "learning_rate": 1.02803738317757e-05,
80
+ "loss": 1.057,
81
  "step": 55
82
  },
83
  {
84
+ "epoch": 0.06,
85
+ "learning_rate": 1.1214953271028037e-05,
86
+ "loss": 1.0466,
87
  "step": 60
88
  },
89
  {
90
+ "epoch": 0.06,
91
+ "learning_rate": 1.2149532710280374e-05,
92
+ "loss": 1.0452,
93
  "step": 65
94
  },
95
  {
96
+ "epoch": 0.07,
97
+ "learning_rate": 1.308411214953271e-05,
98
+ "loss": 1.0368,
99
  "step": 70
100
  },
101
  {
102
+ "epoch": 0.07,
103
+ "learning_rate": 1.4018691588785047e-05,
104
+ "loss": 1.0308,
105
  "step": 75
106
  },
107
  {
108
+ "epoch": 0.07,
109
+ "learning_rate": 1.4953271028037384e-05,
110
+ "loss": 1.0298,
111
  "step": 80
112
  },
113
  {
114
+ "epoch": 0.08,
115
+ "learning_rate": 1.588785046728972e-05,
116
+ "loss": 1.0334,
117
  "step": 85
118
  },
119
  {
120
+ "epoch": 0.08,
121
+ "learning_rate": 1.6822429906542056e-05,
122
+ "loss": 1.0297,
123
  "step": 90
124
  },
125
  {
126
+ "epoch": 0.09,
127
+ "learning_rate": 1.7757009345794395e-05,
128
+ "loss": 1.0464,
129
  "step": 95
130
  },
131
  {
132
+ "epoch": 0.09,
133
+ "learning_rate": 1.869158878504673e-05,
134
+ "loss": 1.0239,
135
  "step": 100
136
  },
137
  {
138
+ "epoch": 0.1,
139
+ "learning_rate": 1.9626168224299065e-05,
140
+ "loss": 1.06,
141
  "step": 105
142
  },
143
  {
144
+ "epoch": 0.1,
145
+ "learning_rate": 1.9999519091998718e-05,
146
+ "loss": 1.0427,
147
  "step": 110
148
  },
149
  {
150
+ "epoch": 0.11,
151
+ "learning_rate": 1.9996580377272997e-05,
152
+ "loss": 1.0744,
153
  "step": 115
154
  },
155
  {
156
+ "epoch": 0.11,
157
+ "learning_rate": 1.9990970903110117e-05,
158
+ "loss": 1.0433,
159
  "step": 120
160
  },
161
  {
162
+ "epoch": 0.12,
163
+ "learning_rate": 1.9982692168178177e-05,
164
+ "loss": 1.0239,
165
  "step": 125
166
  },
167
  {
168
+ "epoch": 0.12,
169
+ "learning_rate": 1.997174638428448e-05,
170
+ "loss": 1.0644,
171
  "step": 130
172
  },
173
  {
174
+ "epoch": 0.13,
175
+ "learning_rate": 1.9958136475784596e-05,
176
+ "loss": 1.0385,
177
  "step": 135
178
  },
179
  {
180
+ "epoch": 0.13,
181
+ "learning_rate": 1.9941866078801108e-05,
182
+ "loss": 1.0404,
183
  "step": 140
184
  },
185
  {
186
+ "epoch": 0.14,
187
+ "learning_rate": 1.9922939540252124e-05,
188
+ "loss": 1.0414,
189
  "step": 145
190
  },
191
  {
192
+ "epoch": 0.14,
193
+ "learning_rate": 1.9901361916689937e-05,
194
+ "loss": 1.0151,
195
  "step": 150
196
  },
197
  {
198
+ "epoch": 0.15,
199
+ "learning_rate": 1.9877138972950084e-05,
200
+ "loss": 1.0413,
201
  "step": 155
202
  },
203
  {
204
+ "epoch": 0.15,
205
+ "learning_rate": 1.985027718061116e-05,
206
+ "loss": 1.0299,
207
  "step": 160
208
  },
209
  {
210
+ "epoch": 0.15,
211
+ "learning_rate": 1.982078371626582e-05,
212
+ "loss": 1.0269,
213
  "step": 165
214
  },
215
  {
216
+ "epoch": 0.16,
217
+ "learning_rate": 1.9788666459603448e-05,
218
+ "loss": 1.0227,
219
  "step": 170
220
  },
221
  {
222
+ "epoch": 0.16,
223
+ "learning_rate": 1.975393399130495e-05,
224
+ "loss": 1.0139,
225
  "step": 175
226
  },
227
  {
228
+ "epoch": 0.17,
229
+ "learning_rate": 1.9716595590750264e-05,
230
+ "loss": 1.0407,
231
  "step": 180
232
  },
233
  {
234
+ "epoch": 0.17,
235
+ "learning_rate": 1.967666123353924e-05,
236
+ "loss": 1.0224,
237
  "step": 185
238
  },
239
  {
240
+ "epoch": 0.18,
241
+ "learning_rate": 1.9634141588826464e-05,
242
+ "loss": 1.0385,
243
  "step": 190
244
  },
245
  {
246
+ "epoch": 0.18,
247
+ "learning_rate": 1.9589048016470816e-05,
248
+ "loss": 1.0272,
249
  "step": 195
250
  },
251
  {
252
+ "epoch": 0.19,
253
+ "learning_rate": 1.954139256400049e-05,
254
+ "loss": 1.0278,
255
  "step": 200
256
  },
257
  {
258
+ "epoch": 0.19,
259
+ "learning_rate": 1.9491187963394283e-05,
260
+ "loss": 1.0096,
261
  "step": 205
262
  },
263
  {
264
+ "epoch": 0.2,
265
+ "learning_rate": 1.9438447627680032e-05,
266
+ "loss": 1.0518,
267
  "step": 210
268
  },
269
  {
270
+ "epoch": 0.2,
271
+ "learning_rate": 1.9383185647351083e-05,
272
+ "loss": 1.0361,
273
  "step": 215
274
  },
275
  {
276
+ "epoch": 0.21,
277
+ "learning_rate": 1.9325416786601785e-05,
278
+ "loss": 1.0256,
279
  "step": 220
280
  },
281
  {
282
+ "epoch": 0.21,
283
+ "learning_rate": 1.9265156479382968e-05,
284
+ "loss": 1.024,
285
  "step": 225
286
  },
287
  {
288
+ "epoch": 0.22,
289
+ "learning_rate": 1.9202420825278497e-05,
290
+ "loss": 1.0309,
291
  "step": 230
292
  },
293
  {
294
+ "epoch": 0.22,
295
+ "learning_rate": 1.9137226585204006e-05,
296
+ "loss": 1.0322,
297
  "step": 235
298
  },
299
  {
300
+ "epoch": 0.22,
301
+ "learning_rate": 1.9069591176928903e-05,
302
+ "loss": 1.0342,
303
  "step": 240
304
  },
305
  {
306
+ "epoch": 0.23,
307
+ "learning_rate": 1.899953267042293e-05,
308
+ "loss": 1.041,
309
  "step": 245
310
  },
311
  {
312
+ "epoch": 0.23,
313
+ "learning_rate": 1.8927069783028447e-05,
314
+ "loss": 1.0405,
315
  "step": 250
316
  },
317
  {
318
+ "epoch": 0.24,
319
+ "learning_rate": 1.885222187445979e-05,
320
+ "loss": 1.0345,
321
  "step": 255
322
  },
323
  {
324
+ "epoch": 0.24,
325
+ "learning_rate": 1.877500894163096e-05,
326
+ "loss": 1.0386,
327
  "step": 260
328
  },
329
  {
330
+ "epoch": 0.25,
331
+ "learning_rate": 1.8695451613313125e-05,
332
+ "loss": 1.0244,
333
  "step": 265
334
  },
335
  {
336
+ "epoch": 0.25,
337
+ "learning_rate": 1.8613571144623267e-05,
338
+ "loss": 1.041,
339
  "step": 270
340
  },
341
+ {
342
+ "epoch": 0.26,
343
+ "learning_rate": 1.8529389411345534e-05,
344
+ "loss": 1.0224,
345
+ "step": 275
346
+ },
347
+ {
348
+ "epoch": 0.26,
349
+ "learning_rate": 1.8442928904086717e-05,
350
+ "loss": 1.0206,
351
+ "step": 280
352
+ },
353
+ {
354
+ "epoch": 0.27,
355
+ "learning_rate": 1.8354212722267516e-05,
356
+ "loss": 1.0208,
357
+ "step": 285
358
+ },
359
+ {
360
+ "epoch": 0.27,
361
+ "learning_rate": 1.8263264567951132e-05,
362
+ "loss": 1.0128,
363
+ "step": 290
364
+ },
365
+ {
366
+ "epoch": 0.28,
367
+ "learning_rate": 1.8170108739510825e-05,
368
+ "loss": 1.0328,
369
+ "step": 295
370
+ },
371
+ {
372
+ "epoch": 0.28,
373
+ "learning_rate": 1.807477012513822e-05,
374
+ "loss": 1.0304,
375
+ "step": 300
376
+ },
377
+ {
378
+ "epoch": 0.29,
379
+ "learning_rate": 1.7977274196193966e-05,
380
+ "loss": 1.0057,
381
+ "step": 305
382
+ },
383
+ {
384
+ "epoch": 0.29,
385
+ "learning_rate": 1.787764700040263e-05,
386
+ "loss": 1.0193,
387
+ "step": 310
388
+ },
389
+ {
390
+ "epoch": 0.29,
391
+ "learning_rate": 1.77759151548936e-05,
392
+ "loss": 1.0387,
393
+ "step": 315
394
+ },
395
+ {
396
+ "epoch": 0.3,
397
+ "learning_rate": 1.7672105839089862e-05,
398
+ "loss": 1.0062,
399
+ "step": 320
400
+ },
401
+ {
402
+ "epoch": 0.3,
403
+ "learning_rate": 1.7566246787446538e-05,
404
+ "loss": 0.9987,
405
+ "step": 325
406
+ },
407
+ {
408
+ "epoch": 0.31,
409
+ "learning_rate": 1.7458366282041154e-05,
410
+ "loss": 1.0177,
411
+ "step": 330
412
+ },
413
+ {
414
+ "epoch": 0.31,
415
+ "learning_rate": 1.7348493145017576e-05,
416
+ "loss": 1.0183,
417
+ "step": 335
418
+ },
419
+ {
420
+ "epoch": 0.32,
421
+ "learning_rate": 1.7236656730885703e-05,
422
+ "loss": 1.0134,
423
+ "step": 340
424
+ },
425
+ {
426
+ "epoch": 0.32,
427
+ "learning_rate": 1.7122886918678887e-05,
428
+ "loss": 1.022,
429
+ "step": 345
430
+ },
431
+ {
432
+ "epoch": 0.33,
433
+ "learning_rate": 1.7007214103971224e-05,
434
+ "loss": 1.0,
435
+ "step": 350
436
+ },
437
+ {
438
+ "epoch": 0.33,
439
+ "learning_rate": 1.688966919075687e-05,
440
+ "loss": 1.0061,
441
+ "step": 355
442
+ },
443
+ {
444
+ "epoch": 0.34,
445
+ "learning_rate": 1.6770283583193478e-05,
446
+ "loss": 1.0263,
447
+ "step": 360
448
+ },
449
+ {
450
+ "epoch": 0.34,
451
+ "learning_rate": 1.664908917721207e-05,
452
+ "loss": 1.0131,
453
+ "step": 365
454
+ },
455
+ {
456
+ "epoch": 0.35,
457
+ "learning_rate": 1.6526118351995444e-05,
458
+ "loss": 1.009,
459
+ "step": 370
460
+ },
461
+ {
462
+ "epoch": 0.35,
463
+ "learning_rate": 1.6401403961327558e-05,
464
+ "loss": 1.0104,
465
+ "step": 375
466
+ },
467
+ {
468
+ "epoch": 0.36,
469
+ "learning_rate": 1.6274979324816043e-05,
470
+ "loss": 1.0029,
471
+ "step": 380
472
+ },
473
+ {
474
+ "epoch": 0.36,
475
+ "learning_rate": 1.6146878218990302e-05,
476
+ "loss": 1.0032,
477
+ "step": 385
478
+ },
479
+ {
480
+ "epoch": 0.37,
481
+ "learning_rate": 1.6017134868277522e-05,
482
+ "loss": 0.9954,
483
+ "step": 390
484
+ },
485
+ {
486
+ "epoch": 0.37,
487
+ "learning_rate": 1.588578393585904e-05,
488
+ "loss": 1.0294,
489
+ "step": 395
490
+ },
491
+ {
492
+ "epoch": 0.37,
493
+ "learning_rate": 1.5752860514409462e-05,
494
+ "loss": 1.0063,
495
+ "step": 400
496
+ },
497
+ {
498
+ "epoch": 0.38,
499
+ "learning_rate": 1.5618400116721045e-05,
500
+ "loss": 1.021,
501
+ "step": 405
502
+ },
503
+ {
504
+ "epoch": 0.38,
505
+ "learning_rate": 1.5482438666215848e-05,
506
+ "loss": 1.0124,
507
+ "step": 410
508
+ },
509
+ {
510
+ "epoch": 0.39,
511
+ "learning_rate": 1.534501248734817e-05,
512
+ "loss": 1.0082,
513
+ "step": 415
514
+ },
515
+ {
516
+ "epoch": 0.39,
517
+ "learning_rate": 1.5206158295899837e-05,
518
+ "loss": 0.9861,
519
+ "step": 420
520
+ },
521
+ {
522
+ "epoch": 0.4,
523
+ "learning_rate": 1.5065913189170944e-05,
524
+ "loss": 1.0259,
525
+ "step": 425
526
+ },
527
+ {
528
+ "epoch": 0.4,
529
+ "learning_rate": 1.4924314636068675e-05,
530
+ "loss": 1.0183,
531
+ "step": 430
532
+ },
533
+ {
534
+ "epoch": 0.41,
535
+ "learning_rate": 1.4781400467096834e-05,
536
+ "loss": 0.9975,
537
+ "step": 435
538
+ },
539
+ {
540
+ "epoch": 0.41,
541
+ "learning_rate": 1.4637208864248766e-05,
542
+ "loss": 1.0088,
543
+ "step": 440
544
+ },
545
+ {
546
+ "epoch": 0.42,
547
+ "learning_rate": 1.4491778350806392e-05,
548
+ "loss": 1.0142,
549
+ "step": 445
550
+ },
551
+ {
552
+ "epoch": 0.42,
553
+ "learning_rate": 1.434514778104803e-05,
554
+ "loss": 1.0055,
555
+ "step": 450
556
+ },
557
+ {
558
+ "epoch": 0.43,
559
+ "learning_rate": 1.4197356329867813e-05,
560
+ "loss": 0.9996,
561
+ "step": 455
562
+ },
563
+ {
564
+ "epoch": 0.43,
565
+ "learning_rate": 1.4048443482309429e-05,
566
+ "loss": 1.0101,
567
+ "step": 460
568
+ },
569
+ {
570
+ "epoch": 0.44,
571
+ "learning_rate": 1.3898449023017024e-05,
572
+ "loss": 1.0033,
573
+ "step": 465
574
+ },
575
+ {
576
+ "epoch": 0.44,
577
+ "learning_rate": 1.3747413025606027e-05,
578
+ "loss": 0.9942,
579
+ "step": 470
580
+ },
581
+ {
582
+ "epoch": 0.44,
583
+ "learning_rate": 1.3595375841956804e-05,
584
+ "loss": 1.0043,
585
+ "step": 475
586
+ },
587
+ {
588
+ "epoch": 0.45,
589
+ "learning_rate": 1.3442378091433947e-05,
590
+ "loss": 0.9997,
591
+ "step": 480
592
+ },
593
+ {
594
+ "epoch": 0.45,
595
+ "learning_rate": 1.3288460650034111e-05,
596
+ "loss": 0.9995,
597
+ "step": 485
598
+ },
599
+ {
600
+ "epoch": 0.46,
601
+ "learning_rate": 1.3133664639465287e-05,
602
+ "loss": 0.9904,
603
+ "step": 490
604
+ },
605
+ {
606
+ "epoch": 0.46,
607
+ "learning_rate": 1.297803141616042e-05,
608
+ "loss": 0.982,
609
+ "step": 495
610
+ },
611
+ {
612
+ "epoch": 0.47,
613
+ "learning_rate": 1.2821602560228326e-05,
614
+ "loss": 1.0035,
615
+ "step": 500
616
+ },
617
+ {
618
+ "epoch": 0.47,
619
+ "learning_rate": 1.2664419864344843e-05,
620
+ "loss": 1.0017,
621
+ "step": 505
622
+ },
623
+ {
624
+ "epoch": 0.48,
625
+ "learning_rate": 1.2506525322587207e-05,
626
+ "loss": 0.9985,
627
+ "step": 510
628
+ },
629
+ {
630
+ "epoch": 0.48,
631
+ "learning_rate": 1.2347961119214584e-05,
632
+ "loss": 0.9967,
633
+ "step": 515
634
+ },
635
+ {
636
+ "epoch": 0.49,
637
+ "learning_rate": 1.2188769617397839e-05,
638
+ "loss": 0.9898,
639
+ "step": 520
640
+ },
641
+ {
642
+ "epoch": 0.49,
643
+ "learning_rate": 1.2028993347901467e-05,
644
+ "loss": 0.9834,
645
+ "step": 525
646
+ },
647
+ {
648
+ "epoch": 0.5,
649
+ "learning_rate": 1.1868674997720767e-05,
650
+ "loss": 0.9836,
651
+ "step": 530
652
+ },
653
+ {
654
+ "epoch": 0.5,
655
+ "learning_rate": 1.1707857398677277e-05,
656
+ "loss": 0.9866,
657
+ "step": 535
658
+ },
659
+ {
660
+ "epoch": 0.51,
661
+ "learning_rate": 1.1546583515975511e-05,
662
+ "loss": 0.9989,
663
+ "step": 540
664
+ },
665
+ {
666
+ "epoch": 0.51,
667
+ "learning_rate": 1.1384896436724047e-05,
668
+ "loss": 0.9999,
669
+ "step": 545
670
+ },
671
+ {
672
+ "epoch": 0.51,
673
+ "learning_rate": 1.122283935842408e-05,
674
+ "loss": 0.9808,
675
+ "step": 550
676
+ },
677
+ {
678
+ "epoch": 0.52,
679
+ "learning_rate": 1.1060455577428454e-05,
680
+ "loss": 0.98,
681
+ "step": 555
682
+ },
683
+ {
684
+ "epoch": 0.52,
685
+ "learning_rate": 1.0897788477374289e-05,
686
+ "loss": 0.9846,
687
+ "step": 560
688
+ },
689
+ {
690
+ "epoch": 0.53,
691
+ "learning_rate": 1.0734881517592297e-05,
692
+ "loss": 0.9847,
693
+ "step": 565
694
+ },
695
+ {
696
+ "epoch": 0.53,
697
+ "learning_rate": 1.057177822149586e-05,
698
+ "loss": 0.9809,
699
+ "step": 570
700
+ },
701
+ {
702
+ "epoch": 0.54,
703
+ "learning_rate": 1.0408522164953013e-05,
704
+ "loss": 0.9947,
705
+ "step": 575
706
+ },
707
+ {
708
+ "epoch": 0.54,
709
+ "learning_rate": 1.0245156964644381e-05,
710
+ "loss": 0.9867,
711
+ "step": 580
712
+ },
713
+ {
714
+ "epoch": 0.55,
715
+ "learning_rate": 1.0081726266410239e-05,
716
+ "loss": 0.981,
717
+ "step": 585
718
+ },
719
+ {
720
+ "epoch": 0.55,
721
+ "learning_rate": 9.918273733589764e-06,
722
+ "loss": 0.9738,
723
+ "step": 590
724
+ },
725
+ {
726
+ "epoch": 0.56,
727
+ "learning_rate": 9.754843035355624e-06,
728
+ "loss": 0.9979,
729
+ "step": 595
730
+ },
731
+ {
732
+ "epoch": 0.56,
733
+ "learning_rate": 9.59147783504699e-06,
734
+ "loss": 0.968,
735
+ "step": 600
736
+ },
737
+ {
738
+ "epoch": 0.57,
739
+ "learning_rate": 9.428221778504143e-06,
740
+ "loss": 0.9837,
741
+ "step": 605
742
+ },
743
+ {
744
+ "epoch": 0.57,
745
+ "learning_rate": 9.265118482407708e-06,
746
+ "loss": 0.9787,
747
+ "step": 610
748
+ },
749
+ {
750
+ "epoch": 0.58,
751
+ "learning_rate": 9.102211522625714e-06,
752
+ "loss": 0.9646,
753
+ "step": 615
754
+ },
755
+ {
756
+ "epoch": 0.58,
757
+ "learning_rate": 8.939544422571547e-06,
758
+ "loss": 0.9826,
759
+ "step": 620
760
+ },
761
+ {
762
+ "epoch": 0.59,
763
+ "learning_rate": 8.777160641575923e-06,
764
+ "loss": 0.9773,
765
+ "step": 625
766
+ },
767
+ {
768
+ "epoch": 0.59,
769
+ "learning_rate": 8.615103563275958e-06,
770
+ "loss": 0.968,
771
+ "step": 630
772
+ },
773
+ {
774
+ "epoch": 0.59,
775
+ "learning_rate": 8.453416484024492e-06,
776
+ "loss": 0.9746,
777
+ "step": 635
778
+ },
779
+ {
780
+ "epoch": 0.6,
781
+ "learning_rate": 8.292142601322724e-06,
782
+ "loss": 0.9642,
783
+ "step": 640
784
+ },
785
+ {
786
+ "epoch": 0.6,
787
+ "learning_rate": 8.131325002279235e-06,
788
+ "loss": 0.9647,
789
+ "step": 645
790
+ },
791
+ {
792
+ "epoch": 0.61,
793
+ "learning_rate": 7.971006652098535e-06,
794
+ "loss": 0.9784,
795
+ "step": 650
796
+ },
797
+ {
798
+ "epoch": 0.61,
799
+ "learning_rate": 7.811230382602163e-06,
800
+ "loss": 0.9615,
801
+ "step": 655
802
+ },
803
+ {
804
+ "epoch": 0.62,
805
+ "learning_rate": 7.652038880785419e-06,
806
+ "loss": 0.9641,
807
+ "step": 660
808
+ },
809
+ {
810
+ "epoch": 0.62,
811
+ "learning_rate": 7.493474677412795e-06,
812
+ "loss": 0.9619,
813
+ "step": 665
814
+ },
815
+ {
816
+ "epoch": 0.63,
817
+ "learning_rate": 7.335580135655157e-06,
818
+ "loss": 0.9677,
819
+ "step": 670
820
+ },
821
+ {
822
+ "epoch": 0.63,
823
+ "learning_rate": 7.178397439771676e-06,
824
+ "loss": 0.9657,
825
+ "step": 675
826
+ },
827
+ {
828
+ "epoch": 0.64,
829
+ "learning_rate": 7.021968583839585e-06,
830
+ "loss": 0.963,
831
+ "step": 680
832
+ },
833
+ {
834
+ "epoch": 0.64,
835
+ "learning_rate": 6.8663353605347185e-06,
836
+ "loss": 0.9524,
837
+ "step": 685
838
+ },
839
+ {
840
+ "epoch": 0.65,
841
+ "learning_rate": 6.711539349965892e-06,
842
+ "loss": 0.9721,
843
+ "step": 690
844
+ },
845
+ {
846
+ "epoch": 0.65,
847
+ "learning_rate": 6.557621908566058e-06,
848
+ "loss": 0.963,
849
+ "step": 695
850
+ },
851
+ {
852
+ "epoch": 0.66,
853
+ "learning_rate": 6.4046241580432e-06,
854
+ "loss": 0.9782,
855
+ "step": 700
856
+ },
857
+ {
858
+ "epoch": 0.66,
859
+ "learning_rate": 6.252586974393975e-06,
860
+ "loss": 0.9578,
861
+ "step": 705
862
+ },
863
+ {
864
+ "epoch": 0.66,
865
+ "learning_rate": 6.101550976982979e-06,
866
+ "loss": 0.9607,
867
+ "step": 710
868
+ },
869
  {
870
  "epoch": 0.67,
871
+ "learning_rate": 5.951556517690573e-06,
872
+ "loss": 0.9819,
873
+ "step": 715
 
 
874
  },
875
  {
876
  "epoch": 0.67,
877
+ "learning_rate": 5.802643670132193e-06,
878
+ "loss": 0.9656,
879
+ "step": 720
880
+ },
881
+ {
882
+ "epoch": 0.68,
883
+ "learning_rate": 5.654852218951973e-06,
884
+ "loss": 0.9672,
885
+ "step": 725
886
+ },
887
+ {
888
+ "epoch": 0.68,
889
+ "learning_rate": 5.508221649193609e-06,
890
+ "loss": 0.9486,
891
+ "step": 730
892
+ },
893
+ {
894
+ "epoch": 0.69,
895
+ "learning_rate": 5.362791135751237e-06,
896
+ "loss": 0.9557,
897
+ "step": 735
898
+ },
899
+ {
900
+ "epoch": 0.69,
901
+ "learning_rate": 5.21859953290317e-06,
902
+ "loss": 0.9569,
903
+ "step": 740
904
+ },
905
+ {
906
+ "epoch": 0.7,
907
+ "learning_rate": 5.0756853639313245e-06,
908
+ "loss": 0.9667,
909
+ "step": 745
910
+ },
911
+ {
912
+ "epoch": 0.7,
913
+ "learning_rate": 4.934086810829057e-06,
914
+ "loss": 0.9662,
915
+ "step": 750
916
+ },
917
+ {
918
+ "epoch": 0.71,
919
+ "learning_rate": 4.793841704100162e-06,
920
+ "loss": 0.9721,
921
+ "step": 755
922
+ },
923
+ {
924
+ "epoch": 0.71,
925
+ "learning_rate": 4.654987512651827e-06,
926
+ "loss": 0.9644,
927
+ "step": 760
928
+ },
929
+ {
930
+ "epoch": 0.72,
931
+ "learning_rate": 4.517561333784154e-06,
932
+ "loss": 0.9519,
933
+ "step": 765
934
+ },
935
+ {
936
+ "epoch": 0.72,
937
+ "learning_rate": 4.381599883278962e-06,
938
+ "loss": 0.9523,
939
+ "step": 770
940
+ },
941
+ {
942
+ "epoch": 0.73,
943
+ "learning_rate": 4.247139485590543e-06,
944
+ "loss": 0.9613,
945
+ "step": 775
946
+ },
947
+ {
948
+ "epoch": 0.73,
949
+ "learning_rate": 4.1142160641409626e-06,
950
+ "loss": 0.9515,
951
+ "step": 780
952
+ },
953
+ {
954
+ "epoch": 0.74,
955
+ "learning_rate": 3.982865131722483e-06,
956
+ "loss": 0.9638,
957
+ "step": 785
958
+ },
959
+ {
960
+ "epoch": 0.74,
961
+ "learning_rate": 3.853121781009703e-06,
962
+ "loss": 0.9749,
963
+ "step": 790
964
+ },
965
+ {
966
+ "epoch": 0.74,
967
+ "learning_rate": 3.725020675183961e-06,
968
+ "loss": 0.9509,
969
+ "step": 795
970
+ },
971
+ {
972
+ "epoch": 0.75,
973
+ "learning_rate": 3.5985960386724438e-06,
974
+ "loss": 0.9403,
975
+ "step": 800
976
+ },
977
+ {
978
+ "epoch": 0.75,
979
+ "learning_rate": 3.4738816480045577e-06,
980
+ "loss": 0.9602,
981
+ "step": 805
982
+ },
983
+ {
984
+ "epoch": 0.76,
985
+ "learning_rate": 3.350910822787937e-06,
986
+ "loss": 0.9503,
987
+ "step": 810
988
+ },
989
+ {
990
+ "epoch": 0.76,
991
+ "learning_rate": 3.2297164168065253e-06,
992
+ "loss": 0.9558,
993
+ "step": 815
994
+ },
995
+ {
996
+ "epoch": 0.77,
997
+ "learning_rate": 3.110330809243134e-06,
998
+ "loss": 0.9427,
999
+ "step": 820
1000
+ },
1001
+ {
1002
+ "epoch": 0.77,
1003
+ "learning_rate": 2.9927858960287793e-06,
1004
+ "loss": 0.941,
1005
+ "step": 825
1006
+ },
1007
+ {
1008
+ "epoch": 0.78,
1009
+ "learning_rate": 2.8771130813211147e-06,
1010
+ "loss": 0.9546,
1011
+ "step": 830
1012
+ },
1013
+ {
1014
+ "epoch": 0.78,
1015
+ "learning_rate": 2.7633432691142957e-06,
1016
+ "loss": 0.9587,
1017
+ "step": 835
1018
+ },
1019
+ {
1020
+ "epoch": 0.79,
1021
+ "learning_rate": 2.6515068549824274e-06,
1022
+ "loss": 0.9659,
1023
+ "step": 840
1024
+ },
1025
+ {
1026
+ "epoch": 0.79,
1027
+ "learning_rate": 2.541633717958849e-06,
1028
+ "loss": 0.9557,
1029
+ "step": 845
1030
+ },
1031
+ {
1032
+ "epoch": 0.8,
1033
+ "learning_rate": 2.4337532125534602e-06,
1034
+ "loss": 0.9352,
1035
+ "step": 850
1036
+ },
1037
+ {
1038
+ "epoch": 0.8,
1039
+ "learning_rate": 2.327894160910139e-06,
1040
+ "loss": 0.9666,
1041
+ "step": 855
1042
+ },
1043
+ {
1044
+ "epoch": 0.81,
1045
+ "learning_rate": 2.2240848451064047e-06,
1046
+ "loss": 0.9426,
1047
+ "step": 860
1048
+ },
1049
+ {
1050
+ "epoch": 0.81,
1051
+ "learning_rate": 2.1223529995973747e-06,
1052
+ "loss": 0.9444,
1053
+ "step": 865
1054
+ },
1055
+ {
1056
+ "epoch": 0.81,
1057
+ "learning_rate": 2.0227258038060372e-06,
1058
+ "loss": 0.9429,
1059
+ "step": 870
1060
+ },
1061
+ {
1062
+ "epoch": 0.82,
1063
+ "learning_rate": 1.9252298748617827e-06,
1064
+ "loss": 0.9449,
1065
+ "step": 875
1066
+ },
1067
+ {
1068
+ "epoch": 0.82,
1069
+ "learning_rate": 1.829891260489176e-06,
1070
+ "loss": 0.9336,
1071
+ "step": 880
1072
+ },
1073
+ {
1074
+ "epoch": 0.83,
1075
+ "learning_rate": 1.7367354320488704e-06,
1076
+ "loss": 0.9292,
1077
+ "step": 885
1078
+ },
1079
+ {
1080
+ "epoch": 0.83,
1081
+ "learning_rate": 1.6457872777324856e-06,
1082
+ "loss": 0.96,
1083
+ "step": 890
1084
+ },
1085
+ {
1086
+ "epoch": 0.84,
1087
+ "learning_rate": 1.5570710959132861e-06,
1088
+ "loss": 0.9504,
1089
+ "step": 895
1090
+ },
1091
+ {
1092
+ "epoch": 0.84,
1093
+ "learning_rate": 1.4706105886544674e-06,
1094
+ "loss": 0.9474,
1095
+ "step": 900
1096
+ },
1097
+ {
1098
+ "epoch": 0.85,
1099
+ "learning_rate": 1.3864288553767335e-06,
1100
+ "loss": 0.9474,
1101
+ "step": 905
1102
+ },
1103
+ {
1104
+ "epoch": 0.85,
1105
+ "learning_rate": 1.3045483866868757e-06,
1106
+ "loss": 0.9581,
1107
+ "step": 910
1108
+ },
1109
+ {
1110
+ "epoch": 0.86,
1111
+ "learning_rate": 1.2249910583690394e-06,
1112
+ "loss": 0.9533,
1113
+ "step": 915
1114
+ },
1115
+ {
1116
+ "epoch": 0.86,
1117
+ "learning_rate": 1.147778125540212e-06,
1118
+ "loss": 0.9414,
1119
+ "step": 920
1120
+ },
1121
+ {
1122
+ "epoch": 0.87,
1123
+ "learning_rate": 1.0729302169715538e-06,
1124
+ "loss": 0.935,
1125
+ "step": 925
1126
+ },
1127
+ {
1128
+ "epoch": 0.87,
1129
+ "learning_rate": 1.000467329577074e-06,
1130
+ "loss": 0.9579,
1131
+ "step": 930
1132
+ },
1133
+ {
1134
+ "epoch": 0.88,
1135
+ "learning_rate": 9.304088230711006e-07,
1136
+ "loss": 0.9589,
1137
+ "step": 935
1138
+ },
1139
+ {
1140
+ "epoch": 0.88,
1141
+ "learning_rate": 8.627734147959954e-07,
1142
+ "loss": 0.9434,
1143
+ "step": 940
1144
+ },
1145
+ {
1146
+ "epoch": 0.88,
1147
+ "learning_rate": 7.975791747215023e-07,
1148
+ "loss": 0.9371,
1149
+ "step": 945
1150
+ },
1151
+ {
1152
+ "epoch": 0.89,
1153
+ "learning_rate": 7.348435206170357e-07,
1154
+ "loss": 0.9266,
1155
+ "step": 950
1156
+ },
1157
+ {
1158
+ "epoch": 0.89,
1159
+ "learning_rate": 6.745832133982188e-07,
1160
+ "loss": 0.9509,
1161
+ "step": 955
1162
+ },
1163
+ {
1164
+ "epoch": 0.9,
1165
+ "learning_rate": 6.168143526489201e-07,
1166
+ "loss": 0.9363,
1167
+ "step": 960
1168
+ },
1169
+ {
1170
+ "epoch": 0.9,
1171
+ "learning_rate": 5.615523723199723e-07,
1172
+ "loss": 0.9496,
1173
+ "step": 965
1174
+ },
1175
+ {
1176
+ "epoch": 0.91,
1177
+ "learning_rate": 5.088120366057192e-07,
1178
+ "loss": 0.9388,
1179
+ "step": 970
1180
+ },
1181
+ {
1182
+ "epoch": 0.91,
1183
+ "learning_rate": 4.5860743599951186e-07,
1184
+ "loss": 0.9501,
1185
+ "step": 975
1186
+ },
1187
+ {
1188
+ "epoch": 0.92,
1189
+ "learning_rate": 4.1095198352918643e-07,
1190
+ "loss": 0.9446,
1191
+ "step": 980
1192
+ },
1193
+ {
1194
+ "epoch": 0.92,
1195
+ "learning_rate": 3.658584111735375e-07,
1196
+ "loss": 0.9444,
1197
+ "step": 985
1198
+ },
1199
+ {
1200
+ "epoch": 0.93,
1201
+ "learning_rate": 3.2333876646076126e-07,
1202
+ "loss": 0.9354,
1203
+ "step": 990
1204
+ },
1205
+ {
1206
+ "epoch": 0.93,
1207
+ "learning_rate": 2.834044092497379e-07,
1208
+ "loss": 0.9361,
1209
+ "step": 995
1210
+ },
1211
+ {
1212
+ "epoch": 0.94,
1213
+ "learning_rate": 2.460660086950539e-07,
1214
+ "loss": 0.9314,
1215
+ "step": 1000
1216
+ },
1217
+ {
1218
+ "epoch": 0.94,
1219
+ "learning_rate": 2.113335403965522e-07,
1220
+ "loss": 0.9387,
1221
+ "step": 1005
1222
+ },
1223
+ {
1224
+ "epoch": 0.95,
1225
+ "learning_rate": 1.7921628373418264e-07,
1226
+ "loss": 0.9472,
1227
+ "step": 1010
1228
+ },
1229
+ {
1230
+ "epoch": 0.95,
1231
+ "learning_rate": 1.4972281938884337e-07,
1232
+ "loss": 0.9493,
1233
+ "step": 1015
1234
+ },
1235
+ {
1236
+ "epoch": 0.96,
1237
+ "learning_rate": 1.228610270499153e-07,
1238
+ "loss": 0.9297,
1239
+ "step": 1020
1240
+ },
1241
+ {
1242
+ "epoch": 0.96,
1243
+ "learning_rate": 9.863808331006197e-08,
1244
+ "loss": 0.9439,
1245
+ "step": 1025
1246
+ },
1247
+ {
1248
+ "epoch": 0.96,
1249
+ "learning_rate": 7.706045974787746e-08,
1250
+ "loss": 0.9522,
1251
+ "step": 1030
1252
+ },
1253
+ {
1254
+ "epoch": 0.97,
1255
+ "learning_rate": 5.813392119889405e-08,
1256
+ "loss": 0.9491,
1257
+ "step": 1035
1258
+ },
1259
+ {
1260
+ "epoch": 0.97,
1261
+ "learning_rate": 4.1863524215406405e-08,
1262
+ "loss": 0.9329,
1263
+ "step": 1040
1264
+ },
1265
+ {
1266
+ "epoch": 0.98,
1267
+ "learning_rate": 2.825361571552554e-08,
1268
+ "loss": 0.9478,
1269
+ "step": 1045
1270
+ },
1271
+ {
1272
+ "epoch": 0.98,
1273
+ "learning_rate": 1.7307831821823385e-08,
1274
+ "loss": 0.9417,
1275
+ "step": 1050
1276
+ },
1277
+ {
1278
+ "epoch": 0.99,
1279
+ "learning_rate": 9.029096889884293e-09,
1280
+ "loss": 0.9452,
1281
+ "step": 1055
1282
+ },
1283
+ {
1284
+ "epoch": 0.99,
1285
+ "learning_rate": 3.4196227270055783e-09,
1286
+ "loss": 0.9355,
1287
+ "step": 1060
1288
+ },
1289
+ {
1290
+ "epoch": 1.0,
1291
+ "learning_rate": 4.80908001283531e-10,
1292
+ "loss": 0.9511,
1293
+ "step": 1065
1294
+ },
1295
+ {
1296
+ "epoch": 1.0,
1297
+ "eval_loss": 0.9552798271179199,
1298
+ "eval_runtime": 138.8736,
1299
+ "eval_samples_per_second": 108.905,
1300
+ "eval_steps_per_second": 1.707,
1301
+ "step": 1068
1302
+ },
1303
+ {
1304
+ "epoch": 1.0,
1305
+ "step": 1068,
1306
+ "total_flos": 447234944532480.0,
1307
+ "train_loss": 0.99920238649354,
1308
+ "train_runtime": 5476.1677,
1309
+ "train_samples_per_second": 24.954,
1310
+ "train_steps_per_second": 0.195
1311
  }
1312
  ],
1313
  "logging_steps": 5,
1314
+ "max_steps": 1068,
1315
+ "num_input_tokens_seen": 0,
1316
  "num_train_epochs": 1,
1317
+ "save_steps": 100,
1318
+ "total_flos": 447234944532480.0,
1319
+ "train_batch_size": 16,
1320
  "trial_name": null,
1321
  "trial_params": null
1322
  }