Serega6678 commited on
Commit
e96cfad
1 Parent(s): 66493f1

Model save

Browse files
README.md ADDED
@@ -0,0 +1,67 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: apache-2.0
3
+ library_name: peft
4
+ tags:
5
+ - trl
6
+ - sft
7
+ - generated_from_trainer
8
+ base_model: mistralai/Mistral-7B-v0.1
9
+ datasets:
10
+ - generator
11
+ model-index:
12
+ - name: M7Bv01_SFT_50pct_QLORA_2H100
13
+ results: []
14
+ ---
15
+
16
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
17
+ should probably proofread and complete it, then remove this comment. -->
18
+
19
+ # M7Bv01_SFT_50pct_QLORA_2H100
20
+
21
+ 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.
22
+ It achieves the following results on the evaluation set:
23
+ - Loss: 0.9550
24
+
25
+ ## Model description
26
+
27
+ More information needed
28
+
29
+ ## Intended uses & limitations
30
+
31
+ More information needed
32
+
33
+ ## Training and evaluation data
34
+
35
+ More information needed
36
+
37
+ ## Training procedure
38
+
39
+ ### Training hyperparameters
40
+
41
+ The following hyperparameters were used during training:
42
+ - learning_rate: 0.0002
43
+ - train_batch_size: 8
44
+ - eval_batch_size: 8
45
+ - seed: 42
46
+ - distributed_type: multi-GPU
47
+ - gradient_accumulation_steps: 8
48
+ - total_train_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
+ | 0.9504 | 1.0 | 1088 | 0.9550 |
59
+
60
+
61
+ ### Framework versions
62
+
63
+ - PEFT 0.7.1
64
+ - Transformers 4.36.2
65
+ - Pytorch 2.1.0+cu118
66
+ - Datasets 2.14.6
67
+ - Tokenizers 0.15.2
adapter_model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:d6e127e4d6e28ac15378e0f1d740eb387f6b9ac2bd052cafc5b2cde5d879c8b3
3
  size 83946192
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1e73639033fa421cdb51b052332a31f9e9a0210134c57b5329e6169dbd1f696a
3
  size 83946192
all_results.json ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 1.0,
3
+ "eval_loss": 0.9550448656082153,
4
+ "eval_runtime": 1442.0062,
5
+ "eval_samples": 11555,
6
+ "eval_samples_per_second": 5.333,
7
+ "eval_steps_per_second": 0.667,
8
+ "train_loss": 0.9663942014688954,
9
+ "train_runtime": 46060.5946,
10
+ "train_samples": 103932,
11
+ "train_samples_per_second": 1.512,
12
+ "train_steps_per_second": 0.024
13
+ }
eval_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 1.0,
3
+ "eval_loss": 0.9550448656082153,
4
+ "eval_runtime": 1442.0062,
5
+ "eval_samples": 11555,
6
+ "eval_samples_per_second": 5.333,
7
+ "eval_steps_per_second": 0.667
8
+ }
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 1.0,
3
+ "train_loss": 0.9663942014688954,
4
+ "train_runtime": 46060.5946,
5
+ "train_samples": 103932,
6
+ "train_samples_per_second": 1.512,
7
+ "train_steps_per_second": 0.024
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,1346 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 0.9996554496382222,
5
+ "eval_steps": 500,
6
+ "global_step": 1088,
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": 1.8348623853211011e-06,
14
+ "loss": 1.166,
15
+ "step": 1
16
+ },
17
+ {
18
+ "epoch": 0.0,
19
+ "learning_rate": 9.174311926605506e-06,
20
+ "loss": 1.1533,
21
+ "step": 5
22
+ },
23
+ {
24
+ "epoch": 0.01,
25
+ "learning_rate": 1.834862385321101e-05,
26
+ "loss": 1.1332,
27
+ "step": 10
28
+ },
29
+ {
30
+ "epoch": 0.01,
31
+ "learning_rate": 2.7522935779816515e-05,
32
+ "loss": 1.1124,
33
+ "step": 15
34
+ },
35
+ {
36
+ "epoch": 0.02,
37
+ "learning_rate": 3.669724770642202e-05,
38
+ "loss": 1.0896,
39
+ "step": 20
40
+ },
41
+ {
42
+ "epoch": 0.02,
43
+ "learning_rate": 4.587155963302753e-05,
44
+ "loss": 1.0765,
45
+ "step": 25
46
+ },
47
+ {
48
+ "epoch": 0.03,
49
+ "learning_rate": 5.504587155963303e-05,
50
+ "loss": 1.0314,
51
+ "step": 30
52
+ },
53
+ {
54
+ "epoch": 0.03,
55
+ "learning_rate": 6.422018348623854e-05,
56
+ "loss": 1.0489,
57
+ "step": 35
58
+ },
59
+ {
60
+ "epoch": 0.04,
61
+ "learning_rate": 7.339449541284404e-05,
62
+ "loss": 1.0039,
63
+ "step": 40
64
+ },
65
+ {
66
+ "epoch": 0.04,
67
+ "learning_rate": 8.256880733944955e-05,
68
+ "loss": 0.9825,
69
+ "step": 45
70
+ },
71
+ {
72
+ "epoch": 0.05,
73
+ "learning_rate": 9.174311926605506e-05,
74
+ "loss": 0.9929,
75
+ "step": 50
76
+ },
77
+ {
78
+ "epoch": 0.05,
79
+ "learning_rate": 0.00010091743119266055,
80
+ "loss": 0.9923,
81
+ "step": 55
82
+ },
83
+ {
84
+ "epoch": 0.06,
85
+ "learning_rate": 0.00011009174311926606,
86
+ "loss": 1.0024,
87
+ "step": 60
88
+ },
89
+ {
90
+ "epoch": 0.06,
91
+ "learning_rate": 0.00011926605504587157,
92
+ "loss": 1.0202,
93
+ "step": 65
94
+ },
95
+ {
96
+ "epoch": 0.06,
97
+ "learning_rate": 0.00012844036697247707,
98
+ "loss": 1.0064,
99
+ "step": 70
100
+ },
101
+ {
102
+ "epoch": 0.07,
103
+ "learning_rate": 0.00013761467889908258,
104
+ "loss": 0.9896,
105
+ "step": 75
106
+ },
107
+ {
108
+ "epoch": 0.07,
109
+ "learning_rate": 0.0001467889908256881,
110
+ "loss": 0.9902,
111
+ "step": 80
112
+ },
113
+ {
114
+ "epoch": 0.08,
115
+ "learning_rate": 0.0001559633027522936,
116
+ "loss": 0.9856,
117
+ "step": 85
118
+ },
119
+ {
120
+ "epoch": 0.08,
121
+ "learning_rate": 0.0001651376146788991,
122
+ "loss": 0.9789,
123
+ "step": 90
124
+ },
125
+ {
126
+ "epoch": 0.09,
127
+ "learning_rate": 0.00017431192660550458,
128
+ "loss": 0.9845,
129
+ "step": 95
130
+ },
131
+ {
132
+ "epoch": 0.09,
133
+ "learning_rate": 0.00018348623853211012,
134
+ "loss": 0.9741,
135
+ "step": 100
136
+ },
137
+ {
138
+ "epoch": 0.1,
139
+ "learning_rate": 0.0001926605504587156,
140
+ "loss": 0.9714,
141
+ "step": 105
142
+ },
143
+ {
144
+ "epoch": 0.1,
145
+ "learning_rate": 0.00019999948512240548,
146
+ "loss": 1.0182,
147
+ "step": 110
148
+ },
149
+ {
150
+ "epoch": 0.11,
151
+ "learning_rate": 0.00019998146496329837,
152
+ "loss": 0.9622,
153
+ "step": 115
154
+ },
155
+ {
156
+ "epoch": 0.11,
157
+ "learning_rate": 0.00019993770622619782,
158
+ "loss": 0.9725,
159
+ "step": 120
160
+ },
161
+ {
162
+ "epoch": 0.11,
163
+ "learning_rate": 0.00019986822017606848,
164
+ "loss": 0.9853,
165
+ "step": 125
166
+ },
167
+ {
168
+ "epoch": 0.12,
169
+ "learning_rate": 0.00019977302470094708,
170
+ "loss": 0.981,
171
+ "step": 130
172
+ },
173
+ {
174
+ "epoch": 0.12,
175
+ "learning_rate": 0.00019965214430733754,
176
+ "loss": 0.9773,
177
+ "step": 135
178
+ },
179
+ {
180
+ "epoch": 0.13,
181
+ "learning_rate": 0.00019950561011390213,
182
+ "loss": 0.9749,
183
+ "step": 140
184
+ },
185
+ {
186
+ "epoch": 0.13,
187
+ "learning_rate": 0.00019933345984345037,
188
+ "loss": 0.988,
189
+ "step": 145
190
+ },
191
+ {
192
+ "epoch": 0.14,
193
+ "learning_rate": 0.00019913573781322818,
194
+ "loss": 0.9992,
195
+ "step": 150
196
+ },
197
+ {
198
+ "epoch": 0.14,
199
+ "learning_rate": 0.00019891249492350887,
200
+ "loss": 0.9706,
201
+ "step": 155
202
+ },
203
+ {
204
+ "epoch": 0.15,
205
+ "learning_rate": 0.00019866378864448985,
206
+ "loss": 0.9663,
207
+ "step": 160
208
+ },
209
+ {
210
+ "epoch": 0.15,
211
+ "learning_rate": 0.00019838968300149782,
212
+ "loss": 0.974,
213
+ "step": 165
214
+ },
215
+ {
216
+ "epoch": 0.16,
217
+ "learning_rate": 0.00019809024855850662,
218
+ "loss": 0.9755,
219
+ "step": 170
220
+ },
221
+ {
222
+ "epoch": 0.16,
223
+ "learning_rate": 0.00019776556239997146,
224
+ "loss": 0.9621,
225
+ "step": 175
226
+ },
227
+ {
228
+ "epoch": 0.17,
229
+ "learning_rate": 0.000197415708110985,
230
+ "loss": 0.9749,
231
+ "step": 180
232
+ },
233
+ {
234
+ "epoch": 0.17,
235
+ "learning_rate": 0.00019704077575575978,
236
+ "loss": 0.958,
237
+ "step": 185
238
+ },
239
+ {
240
+ "epoch": 0.17,
241
+ "learning_rate": 0.00019664086185444246,
242
+ "loss": 0.9863,
243
+ "step": 190
244
+ },
245
+ {
246
+ "epoch": 0.18,
247
+ "learning_rate": 0.0001962160693582665,
248
+ "loss": 1.0012,
249
+ "step": 195
250
+ },
251
+ {
252
+ "epoch": 0.18,
253
+ "learning_rate": 0.00019576650762304903,
254
+ "loss": 0.9802,
255
+ "step": 200
256
+ },
257
+ {
258
+ "epoch": 0.19,
259
+ "learning_rate": 0.00019529229238103883,
260
+ "loss": 0.9748,
261
+ "step": 205
262
+ },
263
+ {
264
+ "epoch": 0.19,
265
+ "learning_rate": 0.00019479354571112323,
266
+ "loss": 1.012,
267
+ "step": 210
268
+ },
269
+ {
270
+ "epoch": 0.2,
271
+ "learning_rate": 0.00019427039600740072,
272
+ "loss": 0.9702,
273
+ "step": 215
274
+ },
275
+ {
276
+ "epoch": 0.2,
277
+ "learning_rate": 0.00019372297794612817,
278
+ "loss": 0.9607,
279
+ "step": 220
280
+ },
281
+ {
282
+ "epoch": 0.21,
283
+ "learning_rate": 0.00019315143245105047,
284
+ "loss": 0.9543,
285
+ "step": 225
286
+ },
287
+ {
288
+ "epoch": 0.21,
289
+ "learning_rate": 0.00019255590665712214,
290
+ "loss": 0.9941,
291
+ "step": 230
292
+ },
293
+ {
294
+ "epoch": 0.22,
295
+ "learning_rate": 0.00019193655387262984,
296
+ "loss": 0.9915,
297
+ "step": 235
298
+ },
299
+ {
300
+ "epoch": 0.22,
301
+ "learning_rate": 0.00019129353353972581,
302
+ "loss": 0.98,
303
+ "step": 240
304
+ },
305
+ {
306
+ "epoch": 0.23,
307
+ "learning_rate": 0.00019062701119338185,
308
+ "loss": 0.9759,
309
+ "step": 245
310
+ },
311
+ {
312
+ "epoch": 0.23,
313
+ "learning_rate": 0.0001899371584187753,
314
+ "loss": 0.9784,
315
+ "step": 250
316
+ },
317
+ {
318
+ "epoch": 0.23,
319
+ "learning_rate": 0.00018922415280711716,
320
+ "loss": 0.9737,
321
+ "step": 255
322
+ },
323
+ {
324
+ "epoch": 0.24,
325
+ "learning_rate": 0.00018848817790993432,
326
+ "loss": 0.9737,
327
+ "step": 260
328
+ },
329
+ {
330
+ "epoch": 0.24,
331
+ "learning_rate": 0.00018772942319181696,
332
+ "loss": 0.9674,
333
+ "step": 265
334
+ },
335
+ {
336
+ "epoch": 0.25,
337
+ "learning_rate": 0.0001869480839816443,
338
+ "loss": 0.9572,
339
+ "step": 270
340
+ },
341
+ {
342
+ "epoch": 0.25,
343
+ "learning_rate": 0.0001861443614223002,
344
+ "loss": 0.9758,
345
+ "step": 275
346
+ },
347
+ {
348
+ "epoch": 0.26,
349
+ "learning_rate": 0.00018531846241889245,
350
+ "loss": 0.9611,
351
+ "step": 280
352
+ },
353
+ {
354
+ "epoch": 0.26,
355
+ "learning_rate": 0.0001844705995854882,
356
+ "loss": 0.9593,
357
+ "step": 285
358
+ },
359
+ {
360
+ "epoch": 0.27,
361
+ "learning_rate": 0.0001836009911903803,
362
+ "loss": 0.9847,
363
+ "step": 290
364
+ },
365
+ {
366
+ "epoch": 0.27,
367
+ "learning_rate": 0.00018270986109989744,
368
+ "loss": 0.9755,
369
+ "step": 295
370
+ },
371
+ {
372
+ "epoch": 0.28,
373
+ "learning_rate": 0.00018179743872077359,
374
+ "loss": 0.9491,
375
+ "step": 300
376
+ },
377
+ {
378
+ "epoch": 0.28,
379
+ "learning_rate": 0.000180863958941091,
380
+ "loss": 0.9872,
381
+ "step": 305
382
+ },
383
+ {
384
+ "epoch": 0.28,
385
+ "learning_rate": 0.00017990966206981224,
386
+ "loss": 0.9491,
387
+ "step": 310
388
+ },
389
+ {
390
+ "epoch": 0.29,
391
+ "learning_rate": 0.0001789347937749164,
392
+ "loss": 0.9579,
393
+ "step": 315
394
+ },
395
+ {
396
+ "epoch": 0.29,
397
+ "learning_rate": 0.00017793960502015613,
398
+ "loss": 0.9702,
399
+ "step": 320
400
+ },
401
+ {
402
+ "epoch": 0.3,
403
+ "learning_rate": 0.0001769243520004511,
404
+ "loss": 0.9749,
405
+ "step": 325
406
+ },
407
+ {
408
+ "epoch": 0.3,
409
+ "learning_rate": 0.0001758892960759348,
410
+ "loss": 0.9624,
411
+ "step": 330
412
+ },
413
+ {
414
+ "epoch": 0.31,
415
+ "learning_rate": 0.00017483470370467178,
416
+ "loss": 0.9547,
417
+ "step": 335
418
+ },
419
+ {
420
+ "epoch": 0.31,
421
+ "learning_rate": 0.00017376084637406222,
422
+ "loss": 0.9685,
423
+ "step": 340
424
+ },
425
+ {
426
+ "epoch": 0.32,
427
+ "learning_rate": 0.00017266800053095232,
428
+ "loss": 0.9455,
429
+ "step": 345
430
+ },
431
+ {
432
+ "epoch": 0.32,
433
+ "learning_rate": 0.0001715564475104673,
434
+ "loss": 0.9745,
435
+ "step": 350
436
+ },
437
+ {
438
+ "epoch": 0.33,
439
+ "learning_rate": 0.00017042647346358645,
440
+ "loss": 0.9906,
441
+ "step": 355
442
+ },
443
+ {
444
+ "epoch": 0.33,
445
+ "learning_rate": 0.00016927836928347826,
446
+ "loss": 0.9749,
447
+ "step": 360
448
+ },
449
+ {
450
+ "epoch": 0.34,
451
+ "learning_rate": 0.00016811243053061487,
452
+ "loss": 0.9674,
453
+ "step": 365
454
+ },
455
+ {
456
+ "epoch": 0.34,
457
+ "learning_rate": 0.00016692895735668476,
458
+ "loss": 0.9589,
459
+ "step": 370
460
+ },
461
+ {
462
+ "epoch": 0.34,
463
+ "learning_rate": 0.00016572825442732366,
464
+ "loss": 0.9703,
465
+ "step": 375
466
+ },
467
+ {
468
+ "epoch": 0.35,
469
+ "learning_rate": 0.0001645106308436836,
470
+ "loss": 0.9472,
471
+ "step": 380
472
+ },
473
+ {
474
+ "epoch": 0.35,
475
+ "learning_rate": 0.00016327640006285967,
476
+ "loss": 0.9647,
477
+ "step": 385
478
+ },
479
+ {
480
+ "epoch": 0.36,
481
+ "learning_rate": 0.00016202587981719584,
482
+ "loss": 0.9631,
483
+ "step": 390
484
+ },
485
+ {
486
+ "epoch": 0.36,
487
+ "learning_rate": 0.0001607593920324899,
488
+ "loss": 0.9784,
489
+ "step": 395
490
+ },
491
+ {
492
+ "epoch": 0.37,
493
+ "learning_rate": 0.00015947726274511908,
494
+ "loss": 0.9662,
495
+ "step": 400
496
+ },
497
+ {
498
+ "epoch": 0.37,
499
+ "learning_rate": 0.0001581798220181073,
500
+ "loss": 0.9433,
501
+ "step": 405
502
+ },
503
+ {
504
+ "epoch": 0.38,
505
+ "learning_rate": 0.00015686740385615586,
506
+ "loss": 0.9731,
507
+ "step": 410
508
+ },
509
+ {
510
+ "epoch": 0.38,
511
+ "learning_rate": 0.00015554034611965954,
512
+ "loss": 0.9593,
513
+ "step": 415
514
+ },
515
+ {
516
+ "epoch": 0.39,
517
+ "learning_rate": 0.00015419899043772994,
518
+ "loss": 0.9379,
519
+ "step": 420
520
+ },
521
+ {
522
+ "epoch": 0.39,
523
+ "learning_rate": 0.00015284368212024877,
524
+ "loss": 0.9453,
525
+ "step": 425
526
+ },
527
+ {
528
+ "epoch": 0.4,
529
+ "learning_rate": 0.00015147477006897337,
530
+ "loss": 0.9581,
531
+ "step": 430
532
+ },
533
+ {
534
+ "epoch": 0.4,
535
+ "learning_rate": 0.000150092606687718,
536
+ "loss": 0.9595,
537
+ "step": 435
538
+ },
539
+ {
540
+ "epoch": 0.4,
541
+ "learning_rate": 0.0001486975477916329,
542
+ "loss": 0.9726,
543
+ "step": 440
544
+ },
545
+ {
546
+ "epoch": 0.41,
547
+ "learning_rate": 0.00014728995251560596,
548
+ "loss": 0.9479,
549
+ "step": 445
550
+ },
551
+ {
552
+ "epoch": 0.41,
553
+ "learning_rate": 0.00014587018322180905,
554
+ "loss": 0.9698,
555
+ "step": 450
556
+ },
557
+ {
558
+ "epoch": 0.42,
559
+ "learning_rate": 0.00014443860540641406,
560
+ "loss": 0.9639,
561
+ "step": 455
562
+ },
563
+ {
564
+ "epoch": 0.42,
565
+ "learning_rate": 0.00014299558760550184,
566
+ "loss": 0.9803,
567
+ "step": 460
568
+ },
569
+ {
570
+ "epoch": 0.43,
571
+ "learning_rate": 0.00014154150130018866,
572
+ "loss": 0.9708,
573
+ "step": 465
574
+ },
575
+ {
576
+ "epoch": 0.43,
577
+ "learning_rate": 0.0001400767208209946,
578
+ "loss": 0.9624,
579
+ "step": 470
580
+ },
581
+ {
582
+ "epoch": 0.44,
583
+ "learning_rate": 0.0001386016232514784,
584
+ "loss": 0.9683,
585
+ "step": 475
586
+ },
587
+ {
588
+ "epoch": 0.44,
589
+ "learning_rate": 0.00013711658833116323,
590
+ "loss": 0.9472,
591
+ "step": 480
592
+ },
593
+ {
594
+ "epoch": 0.45,
595
+ "learning_rate": 0.00013562199835777934,
596
+ "loss": 0.9613,
597
+ "step": 485
598
+ },
599
+ {
600
+ "epoch": 0.45,
601
+ "learning_rate": 0.00013411823808884765,
602
+ "loss": 0.9535,
603
+ "step": 490
604
+ },
605
+ {
606
+ "epoch": 0.45,
607
+ "learning_rate": 0.00013260569464263036,
608
+ "loss": 0.9339,
609
+ "step": 495
610
+ },
611
+ {
612
+ "epoch": 0.46,
613
+ "learning_rate": 0.00013108475739847362,
614
+ "loss": 0.9634,
615
+ "step": 500
616
+ },
617
+ {
618
+ "epoch": 0.46,
619
+ "learning_rate": 0.00012955581789656843,
620
+ "loss": 0.9665,
621
+ "step": 505
622
+ },
623
+ {
624
+ "epoch": 0.47,
625
+ "learning_rate": 0.00012801926973715483,
626
+ "loss": 0.9415,
627
+ "step": 510
628
+ },
629
+ {
630
+ "epoch": 0.47,
631
+ "learning_rate": 0.0001264755084791963,
632
+ "loss": 0.9423,
633
+ "step": 515
634
+ },
635
+ {
636
+ "epoch": 0.48,
637
+ "learning_rate": 0.00012492493153854937,
638
+ "loss": 0.9653,
639
+ "step": 520
640
+ },
641
+ {
642
+ "epoch": 0.48,
643
+ "learning_rate": 0.0001233679380856557,
644
+ "loss": 0.9544,
645
+ "step": 525
646
+ },
647
+ {
648
+ "epoch": 0.49,
649
+ "learning_rate": 0.00012180492894278206,
650
+ "loss": 0.9639,
651
+ "step": 530
652
+ },
653
+ {
654
+ "epoch": 0.49,
655
+ "learning_rate": 0.00012023630648083528,
656
+ "loss": 0.9618,
657
+ "step": 535
658
+ },
659
+ {
660
+ "epoch": 0.5,
661
+ "learning_rate": 0.00011866247451577864,
662
+ "loss": 0.9411,
663
+ "step": 540
664
+ },
665
+ {
666
+ "epoch": 0.5,
667
+ "learning_rate": 0.00011708383820467595,
668
+ "loss": 0.9592,
669
+ "step": 545
670
+ },
671
+ {
672
+ "epoch": 0.51,
673
+ "learning_rate": 0.00011550080394139062,
674
+ "loss": 0.962,
675
+ "step": 550
676
+ },
677
+ {
678
+ "epoch": 0.51,
679
+ "learning_rate": 0.00011391377925196626,
680
+ "loss": 0.9857,
681
+ "step": 555
682
+ },
683
+ {
684
+ "epoch": 0.51,
685
+ "learning_rate": 0.00011232317268971585,
686
+ "loss": 0.9723,
687
+ "step": 560
688
+ },
689
+ {
690
+ "epoch": 0.52,
691
+ "learning_rate": 0.00011072939373004647,
692
+ "loss": 0.9561,
693
+ "step": 565
694
+ },
695
+ {
696
+ "epoch": 0.52,
697
+ "learning_rate": 0.00010913285266504636,
698
+ "loss": 0.9531,
699
+ "step": 570
700
+ },
701
+ {
702
+ "epoch": 0.53,
703
+ "learning_rate": 0.0001075339604978624,
704
+ "loss": 0.9512,
705
+ "step": 575
706
+ },
707
+ {
708
+ "epoch": 0.53,
709
+ "learning_rate": 0.0001059331288368938,
710
+ "loss": 0.959,
711
+ "step": 580
712
+ },
713
+ {
714
+ "epoch": 0.54,
715
+ "learning_rate": 0.0001043307697898305,
716
+ "loss": 0.9423,
717
+ "step": 585
718
+ },
719
+ {
720
+ "epoch": 0.54,
721
+ "learning_rate": 0.00010272729585756275,
722
+ "loss": 0.9656,
723
+ "step": 590
724
+ },
725
+ {
726
+ "epoch": 0.55,
727
+ "learning_rate": 0.00010112311982798959,
728
+ "loss": 0.9701,
729
+ "step": 595
730
+ },
731
+ {
732
+ "epoch": 0.55,
733
+ "learning_rate": 9.951865466975344e-05,
734
+ "loss": 0.9553,
735
+ "step": 600
736
+ },
737
+ {
738
+ "epoch": 0.56,
739
+ "learning_rate": 9.791431342592811e-05,
740
+ "loss": 0.9353,
741
+ "step": 605
742
+ },
743
+ {
744
+ "epoch": 0.56,
745
+ "learning_rate": 9.631050910768773e-05,
746
+ "loss": 0.9363,
747
+ "step": 610
748
+ },
749
+ {
750
+ "epoch": 0.57,
751
+ "learning_rate": 9.470765458798368e-05,
752
+ "loss": 0.9463,
753
+ "step": 615
754
+ },
755
+ {
756
+ "epoch": 0.57,
757
+ "learning_rate": 9.310616249525759e-05,
758
+ "loss": 0.9629,
759
+ "step": 620
760
+ },
761
+ {
762
+ "epoch": 0.57,
763
+ "learning_rate": 9.15064451072165e-05,
764
+ "loss": 0.9485,
765
+ "step": 625
766
+ },
767
+ {
768
+ "epoch": 0.58,
769
+ "learning_rate": 8.990891424469927e-05,
770
+ "loss": 0.9585,
771
+ "step": 630
772
+ },
773
+ {
774
+ "epoch": 0.58,
775
+ "learning_rate": 8.831398116565964e-05,
776
+ "loss": 0.9848,
777
+ "step": 635
778
+ },
779
+ {
780
+ "epoch": 0.59,
781
+ "learning_rate": 8.672205645929503e-05,
782
+ "loss": 0.9371,
783
+ "step": 640
784
+ },
785
+ {
786
+ "epoch": 0.59,
787
+ "learning_rate": 8.513354994034681e-05,
788
+ "loss": 0.9354,
789
+ "step": 645
790
+ },
791
+ {
792
+ "epoch": 0.6,
793
+ "learning_rate": 8.35488705436006e-05,
794
+ "loss": 0.9606,
795
+ "step": 650
796
+ },
797
+ {
798
+ "epoch": 0.6,
799
+ "learning_rate": 8.196842621861242e-05,
800
+ "loss": 0.9672,
801
+ "step": 655
802
+ },
803
+ {
804
+ "epoch": 0.61,
805
+ "learning_rate": 8.039262382468905e-05,
806
+ "loss": 0.9562,
807
+ "step": 660
808
+ },
809
+ {
810
+ "epoch": 0.61,
811
+ "learning_rate": 7.882186902614865e-05,
812
+ "loss": 0.9731,
813
+ "step": 665
814
+ },
815
+ {
816
+ "epoch": 0.62,
817
+ "learning_rate": 7.725656618788937e-05,
818
+ "loss": 0.9606,
819
+ "step": 670
820
+ },
821
+ {
822
+ "epoch": 0.62,
823
+ "learning_rate": 7.569711827129208e-05,
824
+ "loss": 0.9426,
825
+ "step": 675
826
+ },
827
+ {
828
+ "epoch": 0.62,
829
+ "learning_rate": 7.414392673048478e-05,
830
+ "loss": 0.9451,
831
+ "step": 680
832
+ },
833
+ {
834
+ "epoch": 0.63,
835
+ "learning_rate": 7.259739140899462e-05,
836
+ "loss": 0.95,
837
+ "step": 685
838
+ },
839
+ {
840
+ "epoch": 0.63,
841
+ "learning_rate": 7.105791043681519e-05,
842
+ "loss": 0.9636,
843
+ "step": 690
844
+ },
845
+ {
846
+ "epoch": 0.64,
847
+ "learning_rate": 6.952588012791405e-05,
848
+ "loss": 0.9472,
849
+ "step": 695
850
+ },
851
+ {
852
+ "epoch": 0.64,
853
+ "learning_rate": 6.80016948782086e-05,
854
+ "loss": 0.9516,
855
+ "step": 700
856
+ },
857
+ {
858
+ "epoch": 0.65,
859
+ "learning_rate": 6.648574706403522e-05,
860
+ "loss": 0.9558,
861
+ "step": 705
862
+ },
863
+ {
864
+ "epoch": 0.65,
865
+ "learning_rate": 6.497842694113842e-05,
866
+ "loss": 0.9741,
867
+ "step": 710
868
+ },
869
+ {
870
+ "epoch": 0.66,
871
+ "learning_rate": 6.348012254420606e-05,
872
+ "loss": 0.9626,
873
+ "step": 715
874
+ },
875
+ {
876
+ "epoch": 0.66,
877
+ "learning_rate": 6.199121958697604e-05,
878
+ "loss": 0.9499,
879
+ "step": 720
880
+ },
881
+ {
882
+ "epoch": 0.67,
883
+ "learning_rate": 6.051210136294089e-05,
884
+ "loss": 0.9557,
885
+ "step": 725
886
+ },
887
+ {
888
+ "epoch": 0.67,
889
+ "learning_rate": 5.904314864667497e-05,
890
+ "loss": 0.9562,
891
+ "step": 730
892
+ },
893
+ {
894
+ "epoch": 0.68,
895
+ "learning_rate": 5.758473959581061e-05,
896
+ "loss": 0.933,
897
+ "step": 735
898
+ },
899
+ {
900
+ "epoch": 0.68,
901
+ "learning_rate": 5.613724965368723e-05,
902
+ "loss": 0.9766,
903
+ "step": 740
904
+ },
905
+ {
906
+ "epoch": 0.68,
907
+ "learning_rate": 5.4701051452700245e-05,
908
+ "loss": 0.9381,
909
+ "step": 745
910
+ },
911
+ {
912
+ "epoch": 0.69,
913
+ "learning_rate": 5.327651471837242e-05,
914
+ "loss": 0.9397,
915
+ "step": 750
916
+ },
917
+ {
918
+ "epoch": 0.69,
919
+ "learning_rate": 5.1864006174174504e-05,
920
+ "loss": 0.9309,
921
+ "step": 755
922
+ },
923
+ {
924
+ "epoch": 0.7,
925
+ "learning_rate": 5.046388944711824e-05,
926
+ "loss": 0.9578,
927
+ "step": 760
928
+ },
929
+ {
930
+ "epoch": 0.7,
931
+ "learning_rate": 4.9076524974146507e-05,
932
+ "loss": 0.9472,
933
+ "step": 765
934
+ },
935
+ {
936
+ "epoch": 0.71,
937
+ "learning_rate": 4.7702269909344907e-05,
938
+ "loss": 0.9419,
939
+ "step": 770
940
+ },
941
+ {
942
+ "epoch": 0.71,
943
+ "learning_rate": 4.6341478031998265e-05,
944
+ "loss": 0.9791,
945
+ "step": 775
946
+ },
947
+ {
948
+ "epoch": 0.72,
949
+ "learning_rate": 4.4994499655515865e-05,
950
+ "loss": 0.9417,
951
+ "step": 780
952
+ },
953
+ {
954
+ "epoch": 0.72,
955
+ "learning_rate": 4.3661681537249455e-05,
956
+ "loss": 0.9433,
957
+ "step": 785
958
+ },
959
+ {
960
+ "epoch": 0.73,
961
+ "learning_rate": 4.234336678922569e-05,
962
+ "loss": 0.9308,
963
+ "step": 790
964
+ },
965
+ {
966
+ "epoch": 0.73,
967
+ "learning_rate": 4.103989478981827e-05,
968
+ "loss": 0.9556,
969
+ "step": 795
970
+ },
971
+ {
972
+ "epoch": 0.74,
973
+ "learning_rate": 3.975160109637992e-05,
974
+ "loss": 0.9535,
975
+ "step": 800
976
+ },
977
+ {
978
+ "epoch": 0.74,
979
+ "learning_rate": 3.847881735885918e-05,
980
+ "loss": 0.9496,
981
+ "step": 805
982
+ },
983
+ {
984
+ "epoch": 0.74,
985
+ "learning_rate": 3.722187123442249e-05,
986
+ "loss": 0.9417,
987
+ "step": 810
988
+ },
989
+ {
990
+ "epoch": 0.75,
991
+ "learning_rate": 3.598108630310399e-05,
992
+ "loss": 0.9336,
993
+ "step": 815
994
+ },
995
+ {
996
+ "epoch": 0.75,
997
+ "learning_rate": 3.475678198450555e-05,
998
+ "loss": 0.9492,
999
+ "step": 820
1000
+ },
1001
+ {
1002
+ "epoch": 0.76,
1003
+ "learning_rate": 3.354927345556723e-05,
1004
+ "loss": 0.963,
1005
+ "step": 825
1006
+ },
1007
+ {
1008
+ "epoch": 0.76,
1009
+ "learning_rate": 3.235887156943029e-05,
1010
+ "loss": 0.9489,
1011
+ "step": 830
1012
+ },
1013
+ {
1014
+ "epoch": 0.77,
1015
+ "learning_rate": 3.118588277541312e-05,
1016
+ "loss": 0.9551,
1017
+ "step": 835
1018
+ },
1019
+ {
1020
+ "epoch": 0.77,
1021
+ "learning_rate": 3.003060904012096e-05,
1022
+ "loss": 0.9497,
1023
+ "step": 840
1024
+ },
1025
+ {
1026
+ "epoch": 0.78,
1027
+ "learning_rate": 2.8893347769709476e-05,
1028
+ "loss": 0.927,
1029
+ "step": 845
1030
+ },
1031
+ {
1032
+ "epoch": 0.78,
1033
+ "learning_rate": 2.7774391733322713e-05,
1034
+ "loss": 0.964,
1035
+ "step": 850
1036
+ },
1037
+ {
1038
+ "epoch": 0.79,
1039
+ "learning_rate": 2.6674028987724163e-05,
1040
+ "loss": 0.9524,
1041
+ "step": 855
1042
+ },
1043
+ {
1044
+ "epoch": 0.79,
1045
+ "learning_rate": 2.559254280314156e-05,
1046
+ "loss": 0.9519,
1047
+ "step": 860
1048
+ },
1049
+ {
1050
+ "epoch": 0.79,
1051
+ "learning_rate": 2.4530211590343578e-05,
1052
+ "loss": 0.96,
1053
+ "step": 865
1054
+ },
1055
+ {
1056
+ "epoch": 0.8,
1057
+ "learning_rate": 2.3487308828967493e-05,
1058
+ "loss": 0.9603,
1059
+ "step": 870
1060
+ },
1061
+ {
1062
+ "epoch": 0.8,
1063
+ "learning_rate": 2.2464102997116475e-05,
1064
+ "loss": 0.9513,
1065
+ "step": 875
1066
+ },
1067
+ {
1068
+ "epoch": 0.81,
1069
+ "learning_rate": 2.1460857502244248e-05,
1070
+ "loss": 0.9621,
1071
+ "step": 880
1072
+ },
1073
+ {
1074
+ "epoch": 0.81,
1075
+ "learning_rate": 2.047783061334523e-05,
1076
+ "loss": 0.9459,
1077
+ "step": 885
1078
+ },
1079
+ {
1080
+ "epoch": 0.82,
1081
+ "learning_rate": 1.9515275394467446e-05,
1082
+ "loss": 0.9421,
1083
+ "step": 890
1084
+ },
1085
+ {
1086
+ "epoch": 0.82,
1087
+ "learning_rate": 1.8573439639565282e-05,
1088
+ "loss": 0.9757,
1089
+ "step": 895
1090
+ },
1091
+ {
1092
+ "epoch": 0.83,
1093
+ "learning_rate": 1.765256580870924e-05,
1094
+ "loss": 0.9392,
1095
+ "step": 900
1096
+ },
1097
+ {
1098
+ "epoch": 0.83,
1099
+ "learning_rate": 1.6752890965668266e-05,
1100
+ "loss": 0.9413,
1101
+ "step": 905
1102
+ },
1103
+ {
1104
+ "epoch": 0.84,
1105
+ "learning_rate": 1.587464671688187e-05,
1106
+ "loss": 0.9638,
1107
+ "step": 910
1108
+ },
1109
+ {
1110
+ "epoch": 0.84,
1111
+ "learning_rate": 1.501805915183685e-05,
1112
+ "loss": 0.9464,
1113
+ "step": 915
1114
+ },
1115
+ {
1116
+ "epoch": 0.85,
1117
+ "learning_rate": 1.4183348784864037e-05,
1118
+ "loss": 0.9909,
1119
+ "step": 920
1120
+ },
1121
+ {
1122
+ "epoch": 0.85,
1123
+ "learning_rate": 1.3370730498370831e-05,
1124
+ "loss": 0.9639,
1125
+ "step": 925
1126
+ },
1127
+ {
1128
+ "epoch": 0.85,
1129
+ "learning_rate": 1.258041348752308e-05,
1130
+ "loss": 0.9478,
1131
+ "step": 930
1132
+ },
1133
+ {
1134
+ "epoch": 0.86,
1135
+ "learning_rate": 1.1812601206391304e-05,
1136
+ "loss": 0.9451,
1137
+ "step": 935
1138
+ },
1139
+ {
1140
+ "epoch": 0.86,
1141
+ "learning_rate": 1.1067491315574797e-05,
1142
+ "loss": 0.9606,
1143
+ "step": 940
1144
+ },
1145
+ {
1146
+ "epoch": 0.87,
1147
+ "learning_rate": 1.0345275631317163e-05,
1148
+ "loss": 0.95,
1149
+ "step": 945
1150
+ },
1151
+ {
1152
+ "epoch": 0.87,
1153
+ "learning_rate": 9.64614007612633e-06,
1154
+ "loss": 0.917,
1155
+ "step": 950
1156
+ },
1157
+ {
1158
+ "epoch": 0.88,
1159
+ "learning_rate": 8.970264630912061e-06,
1160
+ "loss": 0.9498,
1161
+ "step": 955
1162
+ },
1163
+ {
1164
+ "epoch": 0.88,
1165
+ "learning_rate": 8.317823288652526e-06,
1166
+ "loss": 0.9526,
1167
+ "step": 960
1168
+ },
1169
+ {
1170
+ "epoch": 0.89,
1171
+ "learning_rate": 7.688984009603062e-06,
1172
+ "loss": 0.9482,
1173
+ "step": 965
1174
+ },
1175
+ {
1176
+ "epoch": 0.89,
1177
+ "learning_rate": 7.083908678057194e-06,
1178
+ "loss": 0.9614,
1179
+ "step": 970
1180
+ },
1181
+ {
1182
+ "epoch": 0.9,
1183
+ "learning_rate": 6.502753060672495e-06,
1184
+ "loss": 0.948,
1185
+ "step": 975
1186
+ },
1187
+ {
1188
+ "epoch": 0.9,
1189
+ "learning_rate": 5.945666766370861e-06,
1190
+ "loss": 0.9497,
1191
+ "step": 980
1192
+ },
1193
+ {
1194
+ "epoch": 0.91,
1195
+ "learning_rate": 5.412793207824252e-06,
1196
+ "loss": 0.958,
1197
+ "step": 985
1198
+ },
1199
+ {
1200
+ "epoch": 0.91,
1201
+ "learning_rate": 4.904269564535391e-06,
1202
+ "loss": 0.9649,
1203
+ "step": 990
1204
+ },
1205
+ {
1206
+ "epoch": 0.91,
1207
+ "learning_rate": 4.42022674752326e-06,
1208
+ "loss": 0.9505,
1209
+ "step": 995
1210
+ },
1211
+ {
1212
+ "epoch": 0.92,
1213
+ "learning_rate": 3.960789365622075e-06,
1214
+ "loss": 0.9659,
1215
+ "step": 1000
1216
+ },
1217
+ {
1218
+ "epoch": 0.92,
1219
+ "learning_rate": 3.526075693402986e-06,
1220
+ "loss": 0.9486,
1221
+ "step": 1005
1222
+ },
1223
+ {
1224
+ "epoch": 0.93,
1225
+ "learning_rate": 3.116197640726104e-06,
1226
+ "loss": 0.9495,
1227
+ "step": 1010
1228
+ },
1229
+ {
1230
+ "epoch": 0.93,
1231
+ "learning_rate": 2.7312607239311505e-06,
1232
+ "loss": 0.9771,
1233
+ "step": 1015
1234
+ },
1235
+ {
1236
+ "epoch": 0.94,
1237
+ "learning_rate": 2.3713640386741396e-06,
1238
+ "loss": 0.938,
1239
+ "step": 1020
1240
+ },
1241
+ {
1242
+ "epoch": 0.94,
1243
+ "learning_rate": 2.0366002344166745e-06,
1244
+ "loss": 0.953,
1245
+ "step": 1025
1246
+ },
1247
+ {
1248
+ "epoch": 0.95,
1249
+ "learning_rate": 1.7270554905750137e-06,
1250
+ "loss": 0.976,
1251
+ "step": 1030
1252
+ },
1253
+ {
1254
+ "epoch": 0.95,
1255
+ "learning_rate": 1.4428094943345294e-06,
1256
+ "loss": 0.9734,
1257
+ "step": 1035
1258
+ },
1259
+ {
1260
+ "epoch": 0.96,
1261
+ "learning_rate": 1.1839354201355513e-06,
1262
+ "loss": 0.9538,
1263
+ "step": 1040
1264
+ },
1265
+ {
1266
+ "epoch": 0.96,
1267
+ "learning_rate": 9.50499910835867e-07,
1268
+ "loss": 0.998,
1269
+ "step": 1045
1270
+ },
1271
+ {
1272
+ "epoch": 0.96,
1273
+ "learning_rate": 7.425630605545575e-07,
1274
+ "loss": 0.9474,
1275
+ "step": 1050
1276
+ },
1277
+ {
1278
+ "epoch": 0.97,
1279
+ "learning_rate": 5.60178399201805e-07,
1280
+ "loss": 0.957,
1281
+ "step": 1055
1282
+ },
1283
+ {
1284
+ "epoch": 0.97,
1285
+ "learning_rate": 4.0339287869847197e-07,
1286
+ "loss": 0.9426,
1287
+ "step": 1060
1288
+ },
1289
+ {
1290
+ "epoch": 0.98,
1291
+ "learning_rate": 2.7224686088910265e-07,
1292
+ "loss": 0.9386,
1293
+ "step": 1065
1294
+ },
1295
+ {
1296
+ "epoch": 0.98,
1297
+ "learning_rate": 1.6677410715149054e-07,
1298
+ "loss": 0.954,
1299
+ "step": 1070
1300
+ },
1301
+ {
1302
+ "epoch": 0.99,
1303
+ "learning_rate": 8.700176970527497e-08,
1304
+ "loss": 0.9449,
1305
+ "step": 1075
1306
+ },
1307
+ {
1308
+ "epoch": 0.99,
1309
+ "learning_rate": 3.29503846221213e-08,
1310
+ "loss": 0.9593,
1311
+ "step": 1080
1312
+ },
1313
+ {
1314
+ "epoch": 1.0,
1315
+ "learning_rate": 4.633866539005549e-09,
1316
+ "loss": 0.9504,
1317
+ "step": 1085
1318
+ },
1319
+ {
1320
+ "epoch": 1.0,
1321
+ "eval_loss": 0.9550448656082153,
1322
+ "eval_runtime": 1441.354,
1323
+ "eval_samples_per_second": 5.335,
1324
+ "eval_steps_per_second": 0.667,
1325
+ "step": 1088
1326
+ },
1327
+ {
1328
+ "epoch": 1.0,
1329
+ "step": 1088,
1330
+ "total_flos": 6.120039388450652e+18,
1331
+ "train_loss": 0.9663942014688954,
1332
+ "train_runtime": 46060.5946,
1333
+ "train_samples_per_second": 1.512,
1334
+ "train_steps_per_second": 0.024
1335
+ }
1336
+ ],
1337
+ "logging_steps": 5,
1338
+ "max_steps": 1088,
1339
+ "num_input_tokens_seen": 0,
1340
+ "num_train_epochs": 1,
1341
+ "save_steps": 100,
1342
+ "total_flos": 6.120039388450652e+18,
1343
+ "train_batch_size": 8,
1344
+ "trial_name": null,
1345
+ "trial_params": null
1346
+ }