YeRyeongLee commited on
Commit
4ceacd6
1 Parent(s): 07a7576

End of training

Browse files
Files changed (4) hide show
  1. all_results.json +8 -0
  2. train_results.json +8 -0
  3. train_results.txt +6 -0
  4. trainer_state.json +2371 -0
all_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 10.0,
3
+ "train_loss": 0.6484309974533823,
4
+ "train_runtime": 10912.4775,
5
+ "train_samples": 100102,
6
+ "train_samples_per_second": 91.732,
7
+ "train_steps_per_second": 0.358
8
+ }
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 10.0,
3
+ "train_loss": 0.6484309974533823,
4
+ "train_runtime": 10912.4775,
5
+ "train_samples": 100102,
6
+ "train_samples_per_second": 91.732,
7
+ "train_steps_per_second": 0.358
8
+ }
train_results.txt ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ epoch = 10.0
2
+ train_loss = 0.6484309974533823
3
+ train_runtime = 10912.4775
4
+ train_samples = 100102
5
+ train_samples_per_second = 91.732
6
+ train_steps_per_second = 0.358
trainer_state.json ADDED
@@ -0,0 +1,2371 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 9.999680409076383,
5
+ "global_step": 3910,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.03,
12
+ "learning_rate": 5.115089514066497e-07,
13
+ "loss": 2.3876,
14
+ "step": 10
15
+ },
16
+ {
17
+ "epoch": 0.05,
18
+ "learning_rate": 1.0230179028132994e-06,
19
+ "loss": 2.382,
20
+ "step": 20
21
+ },
22
+ {
23
+ "epoch": 0.08,
24
+ "learning_rate": 1.534526854219949e-06,
25
+ "loss": 2.3645,
26
+ "step": 30
27
+ },
28
+ {
29
+ "epoch": 0.1,
30
+ "learning_rate": 2.0460358056265987e-06,
31
+ "loss": 2.3333,
32
+ "step": 40
33
+ },
34
+ {
35
+ "epoch": 0.13,
36
+ "learning_rate": 2.5575447570332483e-06,
37
+ "loss": 2.2959,
38
+ "step": 50
39
+ },
40
+ {
41
+ "epoch": 0.15,
42
+ "learning_rate": 3.069053708439898e-06,
43
+ "loss": 2.2527,
44
+ "step": 60
45
+ },
46
+ {
47
+ "epoch": 0.18,
48
+ "learning_rate": 3.5805626598465474e-06,
49
+ "loss": 2.2021,
50
+ "step": 70
51
+ },
52
+ {
53
+ "epoch": 0.2,
54
+ "learning_rate": 4.092071611253197e-06,
55
+ "loss": 2.1613,
56
+ "step": 80
57
+ },
58
+ {
59
+ "epoch": 0.23,
60
+ "learning_rate": 4.603580562659847e-06,
61
+ "loss": 2.0852,
62
+ "step": 90
63
+ },
64
+ {
65
+ "epoch": 0.26,
66
+ "learning_rate": 5.1150895140664966e-06,
67
+ "loss": 2.0602,
68
+ "step": 100
69
+ },
70
+ {
71
+ "epoch": 0.28,
72
+ "learning_rate": 5.626598465473146e-06,
73
+ "loss": 2.0174,
74
+ "step": 110
75
+ },
76
+ {
77
+ "epoch": 0.31,
78
+ "learning_rate": 6.138107416879796e-06,
79
+ "loss": 1.9743,
80
+ "step": 120
81
+ },
82
+ {
83
+ "epoch": 0.33,
84
+ "learning_rate": 6.649616368286445e-06,
85
+ "loss": 1.8913,
86
+ "step": 130
87
+ },
88
+ {
89
+ "epoch": 0.36,
90
+ "learning_rate": 7.161125319693095e-06,
91
+ "loss": 1.8595,
92
+ "step": 140
93
+ },
94
+ {
95
+ "epoch": 0.38,
96
+ "learning_rate": 7.672634271099745e-06,
97
+ "loss": 1.7506,
98
+ "step": 150
99
+ },
100
+ {
101
+ "epoch": 0.41,
102
+ "learning_rate": 8.184143222506395e-06,
103
+ "loss": 1.6762,
104
+ "step": 160
105
+ },
106
+ {
107
+ "epoch": 0.43,
108
+ "learning_rate": 8.695652173913044e-06,
109
+ "loss": 1.6315,
110
+ "step": 170
111
+ },
112
+ {
113
+ "epoch": 0.46,
114
+ "learning_rate": 9.207161125319694e-06,
115
+ "loss": 1.5075,
116
+ "step": 180
117
+ },
118
+ {
119
+ "epoch": 0.49,
120
+ "learning_rate": 9.718670076726344e-06,
121
+ "loss": 1.522,
122
+ "step": 190
123
+ },
124
+ {
125
+ "epoch": 0.51,
126
+ "learning_rate": 1.0230179028132993e-05,
127
+ "loss": 1.4587,
128
+ "step": 200
129
+ },
130
+ {
131
+ "epoch": 0.54,
132
+ "learning_rate": 1.0741687979539643e-05,
133
+ "loss": 1.4112,
134
+ "step": 210
135
+ },
136
+ {
137
+ "epoch": 0.56,
138
+ "learning_rate": 1.1253196930946292e-05,
139
+ "loss": 1.4088,
140
+ "step": 220
141
+ },
142
+ {
143
+ "epoch": 0.59,
144
+ "learning_rate": 1.1764705882352942e-05,
145
+ "loss": 1.3279,
146
+ "step": 230
147
+ },
148
+ {
149
+ "epoch": 0.61,
150
+ "learning_rate": 1.2276214833759591e-05,
151
+ "loss": 1.3437,
152
+ "step": 240
153
+ },
154
+ {
155
+ "epoch": 0.64,
156
+ "learning_rate": 1.2787723785166241e-05,
157
+ "loss": 1.312,
158
+ "step": 250
159
+ },
160
+ {
161
+ "epoch": 0.66,
162
+ "learning_rate": 1.329923273657289e-05,
163
+ "loss": 1.2475,
164
+ "step": 260
165
+ },
166
+ {
167
+ "epoch": 0.69,
168
+ "learning_rate": 1.381074168797954e-05,
169
+ "loss": 1.2143,
170
+ "step": 270
171
+ },
172
+ {
173
+ "epoch": 0.72,
174
+ "learning_rate": 1.432225063938619e-05,
175
+ "loss": 1.2112,
176
+ "step": 280
177
+ },
178
+ {
179
+ "epoch": 0.74,
180
+ "learning_rate": 1.483375959079284e-05,
181
+ "loss": 1.1612,
182
+ "step": 290
183
+ },
184
+ {
185
+ "epoch": 0.77,
186
+ "learning_rate": 1.534526854219949e-05,
187
+ "loss": 1.1555,
188
+ "step": 300
189
+ },
190
+ {
191
+ "epoch": 0.79,
192
+ "learning_rate": 1.585677749360614e-05,
193
+ "loss": 1.1404,
194
+ "step": 310
195
+ },
196
+ {
197
+ "epoch": 0.82,
198
+ "learning_rate": 1.636828644501279e-05,
199
+ "loss": 1.1411,
200
+ "step": 320
201
+ },
202
+ {
203
+ "epoch": 0.84,
204
+ "learning_rate": 1.687979539641944e-05,
205
+ "loss": 1.1204,
206
+ "step": 330
207
+ },
208
+ {
209
+ "epoch": 0.87,
210
+ "learning_rate": 1.739130434782609e-05,
211
+ "loss": 1.0784,
212
+ "step": 340
213
+ },
214
+ {
215
+ "epoch": 0.89,
216
+ "learning_rate": 1.790281329923274e-05,
217
+ "loss": 1.0616,
218
+ "step": 350
219
+ },
220
+ {
221
+ "epoch": 0.92,
222
+ "learning_rate": 1.8414322250639388e-05,
223
+ "loss": 1.0479,
224
+ "step": 360
225
+ },
226
+ {
227
+ "epoch": 0.95,
228
+ "learning_rate": 1.8925831202046038e-05,
229
+ "loss": 1.087,
230
+ "step": 370
231
+ },
232
+ {
233
+ "epoch": 0.97,
234
+ "learning_rate": 1.9437340153452687e-05,
235
+ "loss": 1.0054,
236
+ "step": 380
237
+ },
238
+ {
239
+ "epoch": 1.0,
240
+ "learning_rate": 1.9948849104859337e-05,
241
+ "loss": 1.0543,
242
+ "step": 390
243
+ },
244
+ {
245
+ "epoch": 1.02,
246
+ "learning_rate": 1.9948849104859337e-05,
247
+ "loss": 0.9918,
248
+ "step": 400
249
+ },
250
+ {
251
+ "epoch": 1.05,
252
+ "learning_rate": 1.9892014776925264e-05,
253
+ "loss": 0.9886,
254
+ "step": 410
255
+ },
256
+ {
257
+ "epoch": 1.07,
258
+ "learning_rate": 1.983518044899119e-05,
259
+ "loss": 0.9545,
260
+ "step": 420
261
+ },
262
+ {
263
+ "epoch": 1.1,
264
+ "learning_rate": 1.9778346121057118e-05,
265
+ "loss": 0.9254,
266
+ "step": 430
267
+ },
268
+ {
269
+ "epoch": 1.13,
270
+ "learning_rate": 1.972151179312305e-05,
271
+ "loss": 0.9518,
272
+ "step": 440
273
+ },
274
+ {
275
+ "epoch": 1.15,
276
+ "learning_rate": 1.9664677465188975e-05,
277
+ "loss": 0.938,
278
+ "step": 450
279
+ },
280
+ {
281
+ "epoch": 1.18,
282
+ "learning_rate": 1.9607843137254903e-05,
283
+ "loss": 0.9437,
284
+ "step": 460
285
+ },
286
+ {
287
+ "epoch": 1.2,
288
+ "learning_rate": 1.955100880932083e-05,
289
+ "loss": 0.9041,
290
+ "step": 470
291
+ },
292
+ {
293
+ "epoch": 1.23,
294
+ "learning_rate": 1.949417448138676e-05,
295
+ "loss": 0.9089,
296
+ "step": 480
297
+ },
298
+ {
299
+ "epoch": 1.25,
300
+ "learning_rate": 1.9437340153452687e-05,
301
+ "loss": 0.8944,
302
+ "step": 490
303
+ },
304
+ {
305
+ "epoch": 1.28,
306
+ "learning_rate": 1.9380505825518614e-05,
307
+ "loss": 0.9,
308
+ "step": 500
309
+ },
310
+ {
311
+ "epoch": 1.3,
312
+ "learning_rate": 1.932367149758454e-05,
313
+ "loss": 0.9285,
314
+ "step": 510
315
+ },
316
+ {
317
+ "epoch": 1.33,
318
+ "learning_rate": 1.926683716965047e-05,
319
+ "loss": 0.9065,
320
+ "step": 520
321
+ },
322
+ {
323
+ "epoch": 1.36,
324
+ "learning_rate": 1.92100028417164e-05,
325
+ "loss": 0.8607,
326
+ "step": 530
327
+ },
328
+ {
329
+ "epoch": 1.38,
330
+ "learning_rate": 1.9153168513782326e-05,
331
+ "loss": 0.8576,
332
+ "step": 540
333
+ },
334
+ {
335
+ "epoch": 1.41,
336
+ "learning_rate": 1.9096334185848253e-05,
337
+ "loss": 0.8888,
338
+ "step": 550
339
+ },
340
+ {
341
+ "epoch": 1.43,
342
+ "learning_rate": 1.903949985791418e-05,
343
+ "loss": 0.8894,
344
+ "step": 560
345
+ },
346
+ {
347
+ "epoch": 1.46,
348
+ "learning_rate": 1.898266552998011e-05,
349
+ "loss": 0.8637,
350
+ "step": 570
351
+ },
352
+ {
353
+ "epoch": 1.48,
354
+ "learning_rate": 1.8925831202046038e-05,
355
+ "loss": 0.8725,
356
+ "step": 580
357
+ },
358
+ {
359
+ "epoch": 1.51,
360
+ "learning_rate": 1.8868996874111965e-05,
361
+ "loss": 0.852,
362
+ "step": 590
363
+ },
364
+ {
365
+ "epoch": 1.53,
366
+ "learning_rate": 1.881216254617789e-05,
367
+ "loss": 0.8559,
368
+ "step": 600
369
+ },
370
+ {
371
+ "epoch": 1.56,
372
+ "learning_rate": 1.875532821824382e-05,
373
+ "loss": 0.8671,
374
+ "step": 610
375
+ },
376
+ {
377
+ "epoch": 1.59,
378
+ "learning_rate": 1.869849389030975e-05,
379
+ "loss": 0.8656,
380
+ "step": 620
381
+ },
382
+ {
383
+ "epoch": 1.61,
384
+ "learning_rate": 1.8641659562375676e-05,
385
+ "loss": 0.8336,
386
+ "step": 630
387
+ },
388
+ {
389
+ "epoch": 1.64,
390
+ "learning_rate": 1.8584825234441603e-05,
391
+ "loss": 0.8274,
392
+ "step": 640
393
+ },
394
+ {
395
+ "epoch": 1.66,
396
+ "learning_rate": 1.852799090650753e-05,
397
+ "loss": 0.8555,
398
+ "step": 650
399
+ },
400
+ {
401
+ "epoch": 1.69,
402
+ "learning_rate": 1.847115657857346e-05,
403
+ "loss": 0.8158,
404
+ "step": 660
405
+ },
406
+ {
407
+ "epoch": 1.71,
408
+ "learning_rate": 1.8414322250639388e-05,
409
+ "loss": 0.8182,
410
+ "step": 670
411
+ },
412
+ {
413
+ "epoch": 1.74,
414
+ "learning_rate": 1.8357487922705315e-05,
415
+ "loss": 0.8143,
416
+ "step": 680
417
+ },
418
+ {
419
+ "epoch": 1.76,
420
+ "learning_rate": 1.8300653594771242e-05,
421
+ "loss": 0.8632,
422
+ "step": 690
423
+ },
424
+ {
425
+ "epoch": 1.79,
426
+ "learning_rate": 1.824381926683717e-05,
427
+ "loss": 0.8461,
428
+ "step": 700
429
+ },
430
+ {
431
+ "epoch": 1.82,
432
+ "learning_rate": 1.81869849389031e-05,
433
+ "loss": 0.8144,
434
+ "step": 710
435
+ },
436
+ {
437
+ "epoch": 1.84,
438
+ "learning_rate": 1.8130150610969027e-05,
439
+ "loss": 0.834,
440
+ "step": 720
441
+ },
442
+ {
443
+ "epoch": 1.87,
444
+ "learning_rate": 1.8073316283034954e-05,
445
+ "loss": 0.8729,
446
+ "step": 730
447
+ },
448
+ {
449
+ "epoch": 1.89,
450
+ "learning_rate": 1.801648195510088e-05,
451
+ "loss": 0.8187,
452
+ "step": 740
453
+ },
454
+ {
455
+ "epoch": 1.92,
456
+ "learning_rate": 1.795964762716681e-05,
457
+ "loss": 0.827,
458
+ "step": 750
459
+ },
460
+ {
461
+ "epoch": 1.94,
462
+ "learning_rate": 1.790281329923274e-05,
463
+ "loss": 0.8571,
464
+ "step": 760
465
+ },
466
+ {
467
+ "epoch": 1.97,
468
+ "learning_rate": 1.7845978971298665e-05,
469
+ "loss": 0.8204,
470
+ "step": 770
471
+ },
472
+ {
473
+ "epoch": 1.99,
474
+ "learning_rate": 1.7789144643364593e-05,
475
+ "loss": 0.8573,
476
+ "step": 780
477
+ },
478
+ {
479
+ "epoch": 2.02,
480
+ "learning_rate": 1.773231031543052e-05,
481
+ "loss": 0.7526,
482
+ "step": 790
483
+ },
484
+ {
485
+ "epoch": 2.05,
486
+ "learning_rate": 1.767547598749645e-05,
487
+ "loss": 0.7175,
488
+ "step": 800
489
+ },
490
+ {
491
+ "epoch": 2.07,
492
+ "learning_rate": 1.7618641659562377e-05,
493
+ "loss": 0.7448,
494
+ "step": 810
495
+ },
496
+ {
497
+ "epoch": 2.1,
498
+ "learning_rate": 1.7561807331628304e-05,
499
+ "loss": 0.7375,
500
+ "step": 820
501
+ },
502
+ {
503
+ "epoch": 2.12,
504
+ "learning_rate": 1.750497300369423e-05,
505
+ "loss": 0.7215,
506
+ "step": 830
507
+ },
508
+ {
509
+ "epoch": 2.15,
510
+ "learning_rate": 1.7448138675760162e-05,
511
+ "loss": 0.7259,
512
+ "step": 840
513
+ },
514
+ {
515
+ "epoch": 2.17,
516
+ "learning_rate": 1.739130434782609e-05,
517
+ "loss": 0.7564,
518
+ "step": 850
519
+ },
520
+ {
521
+ "epoch": 2.2,
522
+ "learning_rate": 1.7334470019892016e-05,
523
+ "loss": 0.7001,
524
+ "step": 860
525
+ },
526
+ {
527
+ "epoch": 2.22,
528
+ "learning_rate": 1.7277635691957943e-05,
529
+ "loss": 0.7136,
530
+ "step": 870
531
+ },
532
+ {
533
+ "epoch": 2.25,
534
+ "learning_rate": 1.722080136402387e-05,
535
+ "loss": 0.7273,
536
+ "step": 880
537
+ },
538
+ {
539
+ "epoch": 2.28,
540
+ "learning_rate": 1.71639670360898e-05,
541
+ "loss": 0.7035,
542
+ "step": 890
543
+ },
544
+ {
545
+ "epoch": 2.3,
546
+ "learning_rate": 1.7107132708155728e-05,
547
+ "loss": 0.7125,
548
+ "step": 900
549
+ },
550
+ {
551
+ "epoch": 2.33,
552
+ "learning_rate": 1.7050298380221655e-05,
553
+ "loss": 0.73,
554
+ "step": 910
555
+ },
556
+ {
557
+ "epoch": 2.35,
558
+ "learning_rate": 1.6993464052287582e-05,
559
+ "loss": 0.7052,
560
+ "step": 920
561
+ },
562
+ {
563
+ "epoch": 2.38,
564
+ "learning_rate": 1.6936629724353512e-05,
565
+ "loss": 0.7362,
566
+ "step": 930
567
+ },
568
+ {
569
+ "epoch": 2.4,
570
+ "learning_rate": 1.687979539641944e-05,
571
+ "loss": 0.7649,
572
+ "step": 940
573
+ },
574
+ {
575
+ "epoch": 2.43,
576
+ "learning_rate": 1.6822961068485366e-05,
577
+ "loss": 0.7001,
578
+ "step": 950
579
+ },
580
+ {
581
+ "epoch": 2.46,
582
+ "learning_rate": 1.6766126740551293e-05,
583
+ "loss": 0.7006,
584
+ "step": 960
585
+ },
586
+ {
587
+ "epoch": 2.48,
588
+ "learning_rate": 1.670929241261722e-05,
589
+ "loss": 0.7253,
590
+ "step": 970
591
+ },
592
+ {
593
+ "epoch": 2.51,
594
+ "learning_rate": 1.665245808468315e-05,
595
+ "loss": 0.7042,
596
+ "step": 980
597
+ },
598
+ {
599
+ "epoch": 2.53,
600
+ "learning_rate": 1.6595623756749078e-05,
601
+ "loss": 0.7072,
602
+ "step": 990
603
+ },
604
+ {
605
+ "epoch": 2.56,
606
+ "learning_rate": 1.6538789428815005e-05,
607
+ "loss": 0.7279,
608
+ "step": 1000
609
+ },
610
+ {
611
+ "epoch": 2.58,
612
+ "learning_rate": 1.6481955100880932e-05,
613
+ "loss": 0.7273,
614
+ "step": 1010
615
+ },
616
+ {
617
+ "epoch": 2.61,
618
+ "learning_rate": 1.6425120772946863e-05,
619
+ "loss": 0.7148,
620
+ "step": 1020
621
+ },
622
+ {
623
+ "epoch": 2.63,
624
+ "learning_rate": 1.636828644501279e-05,
625
+ "loss": 0.7136,
626
+ "step": 1030
627
+ },
628
+ {
629
+ "epoch": 2.66,
630
+ "learning_rate": 1.6311452117078717e-05,
631
+ "loss": 0.6997,
632
+ "step": 1040
633
+ },
634
+ {
635
+ "epoch": 2.69,
636
+ "learning_rate": 1.6254617789144644e-05,
637
+ "loss": 0.7115,
638
+ "step": 1050
639
+ },
640
+ {
641
+ "epoch": 2.71,
642
+ "learning_rate": 1.619778346121057e-05,
643
+ "loss": 0.6905,
644
+ "step": 1060
645
+ },
646
+ {
647
+ "epoch": 2.74,
648
+ "learning_rate": 1.61409491332765e-05,
649
+ "loss": 0.725,
650
+ "step": 1070
651
+ },
652
+ {
653
+ "epoch": 2.76,
654
+ "learning_rate": 1.608411480534243e-05,
655
+ "loss": 0.6954,
656
+ "step": 1080
657
+ },
658
+ {
659
+ "epoch": 2.79,
660
+ "learning_rate": 1.6027280477408356e-05,
661
+ "loss": 0.7409,
662
+ "step": 1090
663
+ },
664
+ {
665
+ "epoch": 2.81,
666
+ "learning_rate": 1.5970446149474283e-05,
667
+ "loss": 0.7275,
668
+ "step": 1100
669
+ },
670
+ {
671
+ "epoch": 2.84,
672
+ "learning_rate": 1.591361182154021e-05,
673
+ "loss": 0.7164,
674
+ "step": 1110
675
+ },
676
+ {
677
+ "epoch": 2.86,
678
+ "learning_rate": 1.585677749360614e-05,
679
+ "loss": 0.7317,
680
+ "step": 1120
681
+ },
682
+ {
683
+ "epoch": 2.89,
684
+ "learning_rate": 1.5799943165672067e-05,
685
+ "loss": 0.7129,
686
+ "step": 1130
687
+ },
688
+ {
689
+ "epoch": 2.92,
690
+ "learning_rate": 1.5743108837737994e-05,
691
+ "loss": 0.6981,
692
+ "step": 1140
693
+ },
694
+ {
695
+ "epoch": 2.94,
696
+ "learning_rate": 1.568627450980392e-05,
697
+ "loss": 0.6901,
698
+ "step": 1150
699
+ },
700
+ {
701
+ "epoch": 2.97,
702
+ "learning_rate": 1.5629440181869852e-05,
703
+ "loss": 0.7029,
704
+ "step": 1160
705
+ },
706
+ {
707
+ "epoch": 2.99,
708
+ "learning_rate": 1.557260585393578e-05,
709
+ "loss": 0.7217,
710
+ "step": 1170
711
+ },
712
+ {
713
+ "epoch": 3.02,
714
+ "learning_rate": 1.5515771526001706e-05,
715
+ "loss": 0.6391,
716
+ "step": 1180
717
+ },
718
+ {
719
+ "epoch": 3.04,
720
+ "learning_rate": 1.5458937198067633e-05,
721
+ "loss": 0.6264,
722
+ "step": 1190
723
+ },
724
+ {
725
+ "epoch": 3.07,
726
+ "learning_rate": 1.540210287013356e-05,
727
+ "loss": 0.6258,
728
+ "step": 1200
729
+ },
730
+ {
731
+ "epoch": 3.09,
732
+ "learning_rate": 1.534526854219949e-05,
733
+ "loss": 0.6432,
734
+ "step": 1210
735
+ },
736
+ {
737
+ "epoch": 3.12,
738
+ "learning_rate": 1.5288434214265418e-05,
739
+ "loss": 0.6276,
740
+ "step": 1220
741
+ },
742
+ {
743
+ "epoch": 3.15,
744
+ "learning_rate": 1.5231599886331345e-05,
745
+ "loss": 0.6057,
746
+ "step": 1230
747
+ },
748
+ {
749
+ "epoch": 3.17,
750
+ "learning_rate": 1.5174765558397273e-05,
751
+ "loss": 0.6249,
752
+ "step": 1240
753
+ },
754
+ {
755
+ "epoch": 3.2,
756
+ "learning_rate": 1.51179312304632e-05,
757
+ "loss": 0.6112,
758
+ "step": 1250
759
+ },
760
+ {
761
+ "epoch": 3.22,
762
+ "learning_rate": 1.506109690252913e-05,
763
+ "loss": 0.6232,
764
+ "step": 1260
765
+ },
766
+ {
767
+ "epoch": 3.25,
768
+ "learning_rate": 1.5004262574595056e-05,
769
+ "loss": 0.6441,
770
+ "step": 1270
771
+ },
772
+ {
773
+ "epoch": 3.27,
774
+ "learning_rate": 1.4947428246660985e-05,
775
+ "loss": 0.6036,
776
+ "step": 1280
777
+ },
778
+ {
779
+ "epoch": 3.3,
780
+ "learning_rate": 1.4890593918726912e-05,
781
+ "loss": 0.6509,
782
+ "step": 1290
783
+ },
784
+ {
785
+ "epoch": 3.32,
786
+ "learning_rate": 1.483375959079284e-05,
787
+ "loss": 0.611,
788
+ "step": 1300
789
+ },
790
+ {
791
+ "epoch": 3.35,
792
+ "learning_rate": 1.4776925262858768e-05,
793
+ "loss": 0.6206,
794
+ "step": 1310
795
+ },
796
+ {
797
+ "epoch": 3.38,
798
+ "learning_rate": 1.4720090934924695e-05,
799
+ "loss": 0.6064,
800
+ "step": 1320
801
+ },
802
+ {
803
+ "epoch": 3.4,
804
+ "learning_rate": 1.4663256606990624e-05,
805
+ "loss": 0.6353,
806
+ "step": 1330
807
+ },
808
+ {
809
+ "epoch": 3.43,
810
+ "learning_rate": 1.4606422279056551e-05,
811
+ "loss": 0.6394,
812
+ "step": 1340
813
+ },
814
+ {
815
+ "epoch": 3.45,
816
+ "learning_rate": 1.454958795112248e-05,
817
+ "loss": 0.6163,
818
+ "step": 1350
819
+ },
820
+ {
821
+ "epoch": 3.48,
822
+ "learning_rate": 1.4492753623188407e-05,
823
+ "loss": 0.614,
824
+ "step": 1360
825
+ },
826
+ {
827
+ "epoch": 3.5,
828
+ "learning_rate": 1.4435919295254336e-05,
829
+ "loss": 0.6185,
830
+ "step": 1370
831
+ },
832
+ {
833
+ "epoch": 3.53,
834
+ "learning_rate": 1.4379084967320263e-05,
835
+ "loss": 0.6439,
836
+ "step": 1380
837
+ },
838
+ {
839
+ "epoch": 3.55,
840
+ "learning_rate": 1.432225063938619e-05,
841
+ "loss": 0.6156,
842
+ "step": 1390
843
+ },
844
+ {
845
+ "epoch": 3.58,
846
+ "learning_rate": 1.4265416311452118e-05,
847
+ "loss": 0.6665,
848
+ "step": 1400
849
+ },
850
+ {
851
+ "epoch": 3.61,
852
+ "learning_rate": 1.4208581983518046e-05,
853
+ "loss": 0.6015,
854
+ "step": 1410
855
+ },
856
+ {
857
+ "epoch": 3.63,
858
+ "learning_rate": 1.4151747655583974e-05,
859
+ "loss": 0.5993,
860
+ "step": 1420
861
+ },
862
+ {
863
+ "epoch": 3.66,
864
+ "learning_rate": 1.4094913327649901e-05,
865
+ "loss": 0.6466,
866
+ "step": 1430
867
+ },
868
+ {
869
+ "epoch": 3.68,
870
+ "learning_rate": 1.403807899971583e-05,
871
+ "loss": 0.6505,
872
+ "step": 1440
873
+ },
874
+ {
875
+ "epoch": 3.71,
876
+ "learning_rate": 1.3981244671781757e-05,
877
+ "loss": 0.6328,
878
+ "step": 1450
879
+ },
880
+ {
881
+ "epoch": 3.73,
882
+ "learning_rate": 1.3924410343847686e-05,
883
+ "loss": 0.6456,
884
+ "step": 1460
885
+ },
886
+ {
887
+ "epoch": 3.76,
888
+ "learning_rate": 1.3867576015913613e-05,
889
+ "loss": 0.6334,
890
+ "step": 1470
891
+ },
892
+ {
893
+ "epoch": 3.78,
894
+ "learning_rate": 1.381074168797954e-05,
895
+ "loss": 0.5947,
896
+ "step": 1480
897
+ },
898
+ {
899
+ "epoch": 3.81,
900
+ "learning_rate": 1.3753907360045469e-05,
901
+ "loss": 0.6485,
902
+ "step": 1490
903
+ },
904
+ {
905
+ "epoch": 3.84,
906
+ "learning_rate": 1.3697073032111396e-05,
907
+ "loss": 0.6559,
908
+ "step": 1500
909
+ },
910
+ {
911
+ "epoch": 3.86,
912
+ "learning_rate": 1.3640238704177325e-05,
913
+ "loss": 0.6182,
914
+ "step": 1510
915
+ },
916
+ {
917
+ "epoch": 3.89,
918
+ "learning_rate": 1.3583404376243252e-05,
919
+ "loss": 0.6184,
920
+ "step": 1520
921
+ },
922
+ {
923
+ "epoch": 3.91,
924
+ "learning_rate": 1.352657004830918e-05,
925
+ "loss": 0.6376,
926
+ "step": 1530
927
+ },
928
+ {
929
+ "epoch": 3.94,
930
+ "learning_rate": 1.3469735720375108e-05,
931
+ "loss": 0.6126,
932
+ "step": 1540
933
+ },
934
+ {
935
+ "epoch": 3.96,
936
+ "learning_rate": 1.3412901392441036e-05,
937
+ "loss": 0.6122,
938
+ "step": 1550
939
+ },
940
+ {
941
+ "epoch": 3.99,
942
+ "learning_rate": 1.3356067064506963e-05,
943
+ "loss": 0.6305,
944
+ "step": 1560
945
+ },
946
+ {
947
+ "epoch": 4.02,
948
+ "learning_rate": 1.329923273657289e-05,
949
+ "loss": 0.5854,
950
+ "step": 1570
951
+ },
952
+ {
953
+ "epoch": 4.04,
954
+ "learning_rate": 1.324239840863882e-05,
955
+ "loss": 0.5463,
956
+ "step": 1580
957
+ },
958
+ {
959
+ "epoch": 4.07,
960
+ "learning_rate": 1.3185564080704746e-05,
961
+ "loss": 0.5565,
962
+ "step": 1590
963
+ },
964
+ {
965
+ "epoch": 4.09,
966
+ "learning_rate": 1.3128729752770675e-05,
967
+ "loss": 0.5883,
968
+ "step": 1600
969
+ },
970
+ {
971
+ "epoch": 4.12,
972
+ "learning_rate": 1.3071895424836602e-05,
973
+ "loss": 0.5709,
974
+ "step": 1610
975
+ },
976
+ {
977
+ "epoch": 4.14,
978
+ "learning_rate": 1.3015061096902531e-05,
979
+ "loss": 0.5554,
980
+ "step": 1620
981
+ },
982
+ {
983
+ "epoch": 4.17,
984
+ "learning_rate": 1.2958226768968458e-05,
985
+ "loss": 0.541,
986
+ "step": 1630
987
+ },
988
+ {
989
+ "epoch": 4.19,
990
+ "learning_rate": 1.2901392441034387e-05,
991
+ "loss": 0.5633,
992
+ "step": 1640
993
+ },
994
+ {
995
+ "epoch": 4.22,
996
+ "learning_rate": 1.2844558113100314e-05,
997
+ "loss": 0.53,
998
+ "step": 1650
999
+ },
1000
+ {
1001
+ "epoch": 4.25,
1002
+ "learning_rate": 1.2787723785166241e-05,
1003
+ "loss": 0.5695,
1004
+ "step": 1660
1005
+ },
1006
+ {
1007
+ "epoch": 4.27,
1008
+ "learning_rate": 1.273088945723217e-05,
1009
+ "loss": 0.5689,
1010
+ "step": 1670
1011
+ },
1012
+ {
1013
+ "epoch": 4.3,
1014
+ "learning_rate": 1.2674055129298097e-05,
1015
+ "loss": 0.561,
1016
+ "step": 1680
1017
+ },
1018
+ {
1019
+ "epoch": 4.32,
1020
+ "learning_rate": 1.2617220801364026e-05,
1021
+ "loss": 0.5441,
1022
+ "step": 1690
1023
+ },
1024
+ {
1025
+ "epoch": 4.35,
1026
+ "learning_rate": 1.2560386473429953e-05,
1027
+ "loss": 0.5052,
1028
+ "step": 1700
1029
+ },
1030
+ {
1031
+ "epoch": 4.37,
1032
+ "learning_rate": 1.2503552145495881e-05,
1033
+ "loss": 0.5565,
1034
+ "step": 1710
1035
+ },
1036
+ {
1037
+ "epoch": 4.4,
1038
+ "learning_rate": 1.2446717817561809e-05,
1039
+ "loss": 0.569,
1040
+ "step": 1720
1041
+ },
1042
+ {
1043
+ "epoch": 4.42,
1044
+ "learning_rate": 1.2389883489627737e-05,
1045
+ "loss": 0.5383,
1046
+ "step": 1730
1047
+ },
1048
+ {
1049
+ "epoch": 4.45,
1050
+ "learning_rate": 1.2333049161693664e-05,
1051
+ "loss": 0.5603,
1052
+ "step": 1740
1053
+ },
1054
+ {
1055
+ "epoch": 4.48,
1056
+ "learning_rate": 1.2276214833759591e-05,
1057
+ "loss": 0.5318,
1058
+ "step": 1750
1059
+ },
1060
+ {
1061
+ "epoch": 4.5,
1062
+ "learning_rate": 1.221938050582552e-05,
1063
+ "loss": 0.5463,
1064
+ "step": 1760
1065
+ },
1066
+ {
1067
+ "epoch": 4.53,
1068
+ "learning_rate": 1.2162546177891447e-05,
1069
+ "loss": 0.5893,
1070
+ "step": 1770
1071
+ },
1072
+ {
1073
+ "epoch": 4.55,
1074
+ "learning_rate": 1.2105711849957376e-05,
1075
+ "loss": 0.584,
1076
+ "step": 1780
1077
+ },
1078
+ {
1079
+ "epoch": 4.58,
1080
+ "learning_rate": 1.2048877522023303e-05,
1081
+ "loss": 0.5615,
1082
+ "step": 1790
1083
+ },
1084
+ {
1085
+ "epoch": 4.6,
1086
+ "learning_rate": 1.1992043194089232e-05,
1087
+ "loss": 0.5359,
1088
+ "step": 1800
1089
+ },
1090
+ {
1091
+ "epoch": 4.63,
1092
+ "learning_rate": 1.1935208866155159e-05,
1093
+ "loss": 0.5391,
1094
+ "step": 1810
1095
+ },
1096
+ {
1097
+ "epoch": 4.65,
1098
+ "learning_rate": 1.1878374538221088e-05,
1099
+ "loss": 0.588,
1100
+ "step": 1820
1101
+ },
1102
+ {
1103
+ "epoch": 4.68,
1104
+ "learning_rate": 1.1821540210287015e-05,
1105
+ "loss": 0.5374,
1106
+ "step": 1830
1107
+ },
1108
+ {
1109
+ "epoch": 4.71,
1110
+ "learning_rate": 1.1764705882352942e-05,
1111
+ "loss": 0.5946,
1112
+ "step": 1840
1113
+ },
1114
+ {
1115
+ "epoch": 4.73,
1116
+ "learning_rate": 1.170787155441887e-05,
1117
+ "loss": 0.5641,
1118
+ "step": 1850
1119
+ },
1120
+ {
1121
+ "epoch": 4.76,
1122
+ "learning_rate": 1.1651037226484798e-05,
1123
+ "loss": 0.5724,
1124
+ "step": 1860
1125
+ },
1126
+ {
1127
+ "epoch": 4.78,
1128
+ "learning_rate": 1.1594202898550726e-05,
1129
+ "loss": 0.5327,
1130
+ "step": 1870
1131
+ },
1132
+ {
1133
+ "epoch": 4.81,
1134
+ "learning_rate": 1.1537368570616654e-05,
1135
+ "loss": 0.5543,
1136
+ "step": 1880
1137
+ },
1138
+ {
1139
+ "epoch": 4.83,
1140
+ "learning_rate": 1.1480534242682582e-05,
1141
+ "loss": 0.5694,
1142
+ "step": 1890
1143
+ },
1144
+ {
1145
+ "epoch": 4.86,
1146
+ "learning_rate": 1.142369991474851e-05,
1147
+ "loss": 0.5487,
1148
+ "step": 1900
1149
+ },
1150
+ {
1151
+ "epoch": 4.88,
1152
+ "learning_rate": 1.1366865586814438e-05,
1153
+ "loss": 0.5794,
1154
+ "step": 1910
1155
+ },
1156
+ {
1157
+ "epoch": 4.91,
1158
+ "learning_rate": 1.1310031258880365e-05,
1159
+ "loss": 0.5981,
1160
+ "step": 1920
1161
+ },
1162
+ {
1163
+ "epoch": 4.94,
1164
+ "learning_rate": 1.1253196930946292e-05,
1165
+ "loss": 0.5559,
1166
+ "step": 1930
1167
+ },
1168
+ {
1169
+ "epoch": 4.96,
1170
+ "learning_rate": 1.1196362603012221e-05,
1171
+ "loss": 0.5132,
1172
+ "step": 1940
1173
+ },
1174
+ {
1175
+ "epoch": 4.99,
1176
+ "learning_rate": 1.1139528275078148e-05,
1177
+ "loss": 0.5294,
1178
+ "step": 1950
1179
+ },
1180
+ {
1181
+ "epoch": 5.01,
1182
+ "learning_rate": 1.1082693947144077e-05,
1183
+ "loss": 0.5457,
1184
+ "step": 1960
1185
+ },
1186
+ {
1187
+ "epoch": 5.04,
1188
+ "learning_rate": 1.1025859619210004e-05,
1189
+ "loss": 0.5082,
1190
+ "step": 1970
1191
+ },
1192
+ {
1193
+ "epoch": 5.06,
1194
+ "learning_rate": 1.0969025291275933e-05,
1195
+ "loss": 0.4839,
1196
+ "step": 1980
1197
+ },
1198
+ {
1199
+ "epoch": 5.09,
1200
+ "learning_rate": 1.091219096334186e-05,
1201
+ "loss": 0.5042,
1202
+ "step": 1990
1203
+ },
1204
+ {
1205
+ "epoch": 5.12,
1206
+ "learning_rate": 1.0855356635407789e-05,
1207
+ "loss": 0.4862,
1208
+ "step": 2000
1209
+ },
1210
+ {
1211
+ "epoch": 5.14,
1212
+ "learning_rate": 1.0798522307473716e-05,
1213
+ "loss": 0.5559,
1214
+ "step": 2010
1215
+ },
1216
+ {
1217
+ "epoch": 5.17,
1218
+ "learning_rate": 1.0741687979539643e-05,
1219
+ "loss": 0.5107,
1220
+ "step": 2020
1221
+ },
1222
+ {
1223
+ "epoch": 5.19,
1224
+ "learning_rate": 1.0684853651605571e-05,
1225
+ "loss": 0.4919,
1226
+ "step": 2030
1227
+ },
1228
+ {
1229
+ "epoch": 5.22,
1230
+ "learning_rate": 1.0628019323671499e-05,
1231
+ "loss": 0.5198,
1232
+ "step": 2040
1233
+ },
1234
+ {
1235
+ "epoch": 5.24,
1236
+ "learning_rate": 1.0571184995737427e-05,
1237
+ "loss": 0.4967,
1238
+ "step": 2050
1239
+ },
1240
+ {
1241
+ "epoch": 5.27,
1242
+ "learning_rate": 1.0514350667803354e-05,
1243
+ "loss": 0.4895,
1244
+ "step": 2060
1245
+ },
1246
+ {
1247
+ "epoch": 5.29,
1248
+ "learning_rate": 1.0457516339869283e-05,
1249
+ "loss": 0.492,
1250
+ "step": 2070
1251
+ },
1252
+ {
1253
+ "epoch": 5.32,
1254
+ "learning_rate": 1.040068201193521e-05,
1255
+ "loss": 0.494,
1256
+ "step": 2080
1257
+ },
1258
+ {
1259
+ "epoch": 5.35,
1260
+ "learning_rate": 1.0343847684001139e-05,
1261
+ "loss": 0.4864,
1262
+ "step": 2090
1263
+ },
1264
+ {
1265
+ "epoch": 5.37,
1266
+ "learning_rate": 1.0287013356067066e-05,
1267
+ "loss": 0.5043,
1268
+ "step": 2100
1269
+ },
1270
+ {
1271
+ "epoch": 5.4,
1272
+ "learning_rate": 1.0230179028132993e-05,
1273
+ "loss": 0.5088,
1274
+ "step": 2110
1275
+ },
1276
+ {
1277
+ "epoch": 5.42,
1278
+ "learning_rate": 1.0173344700198922e-05,
1279
+ "loss": 0.5035,
1280
+ "step": 2120
1281
+ },
1282
+ {
1283
+ "epoch": 5.45,
1284
+ "learning_rate": 1.0116510372264849e-05,
1285
+ "loss": 0.5052,
1286
+ "step": 2130
1287
+ },
1288
+ {
1289
+ "epoch": 5.47,
1290
+ "learning_rate": 1.0059676044330778e-05,
1291
+ "loss": 0.5096,
1292
+ "step": 2140
1293
+ },
1294
+ {
1295
+ "epoch": 5.5,
1296
+ "learning_rate": 1.0002841716396705e-05,
1297
+ "loss": 0.4916,
1298
+ "step": 2150
1299
+ },
1300
+ {
1301
+ "epoch": 5.52,
1302
+ "learning_rate": 9.946007388462632e-06,
1303
+ "loss": 0.4942,
1304
+ "step": 2160
1305
+ },
1306
+ {
1307
+ "epoch": 5.55,
1308
+ "learning_rate": 9.889173060528559e-06,
1309
+ "loss": 0.4957,
1310
+ "step": 2170
1311
+ },
1312
+ {
1313
+ "epoch": 5.58,
1314
+ "learning_rate": 9.832338732594488e-06,
1315
+ "loss": 0.5078,
1316
+ "step": 2180
1317
+ },
1318
+ {
1319
+ "epoch": 5.6,
1320
+ "learning_rate": 9.775504404660415e-06,
1321
+ "loss": 0.4841,
1322
+ "step": 2190
1323
+ },
1324
+ {
1325
+ "epoch": 5.63,
1326
+ "learning_rate": 9.718670076726344e-06,
1327
+ "loss": 0.4971,
1328
+ "step": 2200
1329
+ },
1330
+ {
1331
+ "epoch": 5.65,
1332
+ "learning_rate": 9.66183574879227e-06,
1333
+ "loss": 0.485,
1334
+ "step": 2210
1335
+ },
1336
+ {
1337
+ "epoch": 5.68,
1338
+ "learning_rate": 9.6050014208582e-06,
1339
+ "loss": 0.4804,
1340
+ "step": 2220
1341
+ },
1342
+ {
1343
+ "epoch": 5.7,
1344
+ "learning_rate": 9.548167092924126e-06,
1345
+ "loss": 0.4811,
1346
+ "step": 2230
1347
+ },
1348
+ {
1349
+ "epoch": 5.73,
1350
+ "learning_rate": 9.491332764990055e-06,
1351
+ "loss": 0.5065,
1352
+ "step": 2240
1353
+ },
1354
+ {
1355
+ "epoch": 5.75,
1356
+ "learning_rate": 9.434498437055982e-06,
1357
+ "loss": 0.4964,
1358
+ "step": 2250
1359
+ },
1360
+ {
1361
+ "epoch": 5.78,
1362
+ "learning_rate": 9.37766410912191e-06,
1363
+ "loss": 0.5003,
1364
+ "step": 2260
1365
+ },
1366
+ {
1367
+ "epoch": 5.81,
1368
+ "learning_rate": 9.320829781187838e-06,
1369
+ "loss": 0.4964,
1370
+ "step": 2270
1371
+ },
1372
+ {
1373
+ "epoch": 5.83,
1374
+ "learning_rate": 9.263995453253765e-06,
1375
+ "loss": 0.5077,
1376
+ "step": 2280
1377
+ },
1378
+ {
1379
+ "epoch": 5.86,
1380
+ "learning_rate": 9.207161125319694e-06,
1381
+ "loss": 0.5197,
1382
+ "step": 2290
1383
+ },
1384
+ {
1385
+ "epoch": 5.88,
1386
+ "learning_rate": 9.150326797385621e-06,
1387
+ "loss": 0.5071,
1388
+ "step": 2300
1389
+ },
1390
+ {
1391
+ "epoch": 5.91,
1392
+ "learning_rate": 9.09349246945155e-06,
1393
+ "loss": 0.4853,
1394
+ "step": 2310
1395
+ },
1396
+ {
1397
+ "epoch": 5.93,
1398
+ "learning_rate": 9.036658141517477e-06,
1399
+ "loss": 0.4799,
1400
+ "step": 2320
1401
+ },
1402
+ {
1403
+ "epoch": 5.96,
1404
+ "learning_rate": 8.979823813583406e-06,
1405
+ "loss": 0.4857,
1406
+ "step": 2330
1407
+ },
1408
+ {
1409
+ "epoch": 5.98,
1410
+ "learning_rate": 8.922989485649333e-06,
1411
+ "loss": 0.4689,
1412
+ "step": 2340
1413
+ },
1414
+ {
1415
+ "epoch": 6.01,
1416
+ "learning_rate": 8.86615515771526e-06,
1417
+ "loss": 0.5046,
1418
+ "step": 2350
1419
+ },
1420
+ {
1421
+ "epoch": 6.04,
1422
+ "learning_rate": 8.809320829781189e-06,
1423
+ "loss": 0.4501,
1424
+ "step": 2360
1425
+ },
1426
+ {
1427
+ "epoch": 6.06,
1428
+ "learning_rate": 8.752486501847116e-06,
1429
+ "loss": 0.4295,
1430
+ "step": 2370
1431
+ },
1432
+ {
1433
+ "epoch": 6.09,
1434
+ "learning_rate": 8.695652173913044e-06,
1435
+ "loss": 0.4513,
1436
+ "step": 2380
1437
+ },
1438
+ {
1439
+ "epoch": 6.11,
1440
+ "learning_rate": 8.638817845978971e-06,
1441
+ "loss": 0.4742,
1442
+ "step": 2390
1443
+ },
1444
+ {
1445
+ "epoch": 6.14,
1446
+ "learning_rate": 8.5819835180449e-06,
1447
+ "loss": 0.4412,
1448
+ "step": 2400
1449
+ },
1450
+ {
1451
+ "epoch": 6.16,
1452
+ "learning_rate": 8.525149190110827e-06,
1453
+ "loss": 0.4546,
1454
+ "step": 2410
1455
+ },
1456
+ {
1457
+ "epoch": 6.19,
1458
+ "learning_rate": 8.468314862176756e-06,
1459
+ "loss": 0.4547,
1460
+ "step": 2420
1461
+ },
1462
+ {
1463
+ "epoch": 6.21,
1464
+ "learning_rate": 8.411480534242683e-06,
1465
+ "loss": 0.4286,
1466
+ "step": 2430
1467
+ },
1468
+ {
1469
+ "epoch": 6.24,
1470
+ "learning_rate": 8.35464620630861e-06,
1471
+ "loss": 0.4475,
1472
+ "step": 2440
1473
+ },
1474
+ {
1475
+ "epoch": 6.27,
1476
+ "learning_rate": 8.297811878374539e-06,
1477
+ "loss": 0.4666,
1478
+ "step": 2450
1479
+ },
1480
+ {
1481
+ "epoch": 6.29,
1482
+ "learning_rate": 8.240977550440466e-06,
1483
+ "loss": 0.4564,
1484
+ "step": 2460
1485
+ },
1486
+ {
1487
+ "epoch": 6.32,
1488
+ "learning_rate": 8.184143222506395e-06,
1489
+ "loss": 0.4504,
1490
+ "step": 2470
1491
+ },
1492
+ {
1493
+ "epoch": 6.34,
1494
+ "learning_rate": 8.127308894572322e-06,
1495
+ "loss": 0.4422,
1496
+ "step": 2480
1497
+ },
1498
+ {
1499
+ "epoch": 6.37,
1500
+ "learning_rate": 8.07047456663825e-06,
1501
+ "loss": 0.4548,
1502
+ "step": 2490
1503
+ },
1504
+ {
1505
+ "epoch": 6.39,
1506
+ "learning_rate": 8.013640238704178e-06,
1507
+ "loss": 0.457,
1508
+ "step": 2500
1509
+ },
1510
+ {
1511
+ "epoch": 6.42,
1512
+ "learning_rate": 7.956805910770105e-06,
1513
+ "loss": 0.4474,
1514
+ "step": 2510
1515
+ },
1516
+ {
1517
+ "epoch": 6.44,
1518
+ "learning_rate": 7.899971582836034e-06,
1519
+ "loss": 0.4462,
1520
+ "step": 2520
1521
+ },
1522
+ {
1523
+ "epoch": 6.47,
1524
+ "learning_rate": 7.84313725490196e-06,
1525
+ "loss": 0.4246,
1526
+ "step": 2530
1527
+ },
1528
+ {
1529
+ "epoch": 6.5,
1530
+ "learning_rate": 7.78630292696789e-06,
1531
+ "loss": 0.4488,
1532
+ "step": 2540
1533
+ },
1534
+ {
1535
+ "epoch": 6.52,
1536
+ "learning_rate": 7.729468599033817e-06,
1537
+ "loss": 0.4616,
1538
+ "step": 2550
1539
+ },
1540
+ {
1541
+ "epoch": 6.55,
1542
+ "learning_rate": 7.672634271099745e-06,
1543
+ "loss": 0.4273,
1544
+ "step": 2560
1545
+ },
1546
+ {
1547
+ "epoch": 6.57,
1548
+ "learning_rate": 7.615799943165672e-06,
1549
+ "loss": 0.4671,
1550
+ "step": 2570
1551
+ },
1552
+ {
1553
+ "epoch": 6.6,
1554
+ "learning_rate": 7.5589656152316e-06,
1555
+ "loss": 0.4738,
1556
+ "step": 2580
1557
+ },
1558
+ {
1559
+ "epoch": 6.62,
1560
+ "learning_rate": 7.502131287297528e-06,
1561
+ "loss": 0.4563,
1562
+ "step": 2590
1563
+ },
1564
+ {
1565
+ "epoch": 6.65,
1566
+ "learning_rate": 7.445296959363456e-06,
1567
+ "loss": 0.4773,
1568
+ "step": 2600
1569
+ },
1570
+ {
1571
+ "epoch": 6.67,
1572
+ "learning_rate": 7.388462631429384e-06,
1573
+ "loss": 0.4678,
1574
+ "step": 2610
1575
+ },
1576
+ {
1577
+ "epoch": 6.7,
1578
+ "learning_rate": 7.331628303495312e-06,
1579
+ "loss": 0.4758,
1580
+ "step": 2620
1581
+ },
1582
+ {
1583
+ "epoch": 6.73,
1584
+ "learning_rate": 7.27479397556124e-06,
1585
+ "loss": 0.4357,
1586
+ "step": 2630
1587
+ },
1588
+ {
1589
+ "epoch": 6.75,
1590
+ "learning_rate": 7.217959647627168e-06,
1591
+ "loss": 0.4732,
1592
+ "step": 2640
1593
+ },
1594
+ {
1595
+ "epoch": 6.78,
1596
+ "learning_rate": 7.161125319693095e-06,
1597
+ "loss": 0.4376,
1598
+ "step": 2650
1599
+ },
1600
+ {
1601
+ "epoch": 6.8,
1602
+ "learning_rate": 7.104290991759023e-06,
1603
+ "loss": 0.4503,
1604
+ "step": 2660
1605
+ },
1606
+ {
1607
+ "epoch": 6.83,
1608
+ "learning_rate": 7.047456663824951e-06,
1609
+ "loss": 0.452,
1610
+ "step": 2670
1611
+ },
1612
+ {
1613
+ "epoch": 6.85,
1614
+ "learning_rate": 6.990622335890879e-06,
1615
+ "loss": 0.4345,
1616
+ "step": 2680
1617
+ },
1618
+ {
1619
+ "epoch": 6.88,
1620
+ "learning_rate": 6.9337880079568065e-06,
1621
+ "loss": 0.4928,
1622
+ "step": 2690
1623
+ },
1624
+ {
1625
+ "epoch": 6.91,
1626
+ "learning_rate": 6.8769536800227344e-06,
1627
+ "loss": 0.4375,
1628
+ "step": 2700
1629
+ },
1630
+ {
1631
+ "epoch": 6.93,
1632
+ "learning_rate": 6.820119352088662e-06,
1633
+ "loss": 0.4412,
1634
+ "step": 2710
1635
+ },
1636
+ {
1637
+ "epoch": 6.96,
1638
+ "learning_rate": 6.76328502415459e-06,
1639
+ "loss": 0.4186,
1640
+ "step": 2720
1641
+ },
1642
+ {
1643
+ "epoch": 6.98,
1644
+ "learning_rate": 6.706450696220518e-06,
1645
+ "loss": 0.433,
1646
+ "step": 2730
1647
+ },
1648
+ {
1649
+ "epoch": 7.01,
1650
+ "learning_rate": 6.649616368286445e-06,
1651
+ "loss": 0.4575,
1652
+ "step": 2740
1653
+ },
1654
+ {
1655
+ "epoch": 7.03,
1656
+ "learning_rate": 6.592782040352373e-06,
1657
+ "loss": 0.4553,
1658
+ "step": 2750
1659
+ },
1660
+ {
1661
+ "epoch": 7.06,
1662
+ "learning_rate": 6.535947712418301e-06,
1663
+ "loss": 0.4123,
1664
+ "step": 2760
1665
+ },
1666
+ {
1667
+ "epoch": 7.08,
1668
+ "learning_rate": 6.479113384484229e-06,
1669
+ "loss": 0.4148,
1670
+ "step": 2770
1671
+ },
1672
+ {
1673
+ "epoch": 7.11,
1674
+ "learning_rate": 6.422279056550157e-06,
1675
+ "loss": 0.4094,
1676
+ "step": 2780
1677
+ },
1678
+ {
1679
+ "epoch": 7.14,
1680
+ "learning_rate": 6.365444728616085e-06,
1681
+ "loss": 0.402,
1682
+ "step": 2790
1683
+ },
1684
+ {
1685
+ "epoch": 7.16,
1686
+ "learning_rate": 6.308610400682013e-06,
1687
+ "loss": 0.4316,
1688
+ "step": 2800
1689
+ },
1690
+ {
1691
+ "epoch": 7.19,
1692
+ "learning_rate": 6.251776072747941e-06,
1693
+ "loss": 0.4182,
1694
+ "step": 2810
1695
+ },
1696
+ {
1697
+ "epoch": 7.21,
1698
+ "learning_rate": 6.194941744813869e-06,
1699
+ "loss": 0.4226,
1700
+ "step": 2820
1701
+ },
1702
+ {
1703
+ "epoch": 7.24,
1704
+ "learning_rate": 6.138107416879796e-06,
1705
+ "loss": 0.4167,
1706
+ "step": 2830
1707
+ },
1708
+ {
1709
+ "epoch": 7.26,
1710
+ "learning_rate": 6.081273088945724e-06,
1711
+ "loss": 0.3813,
1712
+ "step": 2840
1713
+ },
1714
+ {
1715
+ "epoch": 7.29,
1716
+ "learning_rate": 6.0244387610116515e-06,
1717
+ "loss": 0.4131,
1718
+ "step": 2850
1719
+ },
1720
+ {
1721
+ "epoch": 7.31,
1722
+ "learning_rate": 5.9676044330775795e-06,
1723
+ "loss": 0.4336,
1724
+ "step": 2860
1725
+ },
1726
+ {
1727
+ "epoch": 7.34,
1728
+ "learning_rate": 5.910770105143507e-06,
1729
+ "loss": 0.401,
1730
+ "step": 2870
1731
+ },
1732
+ {
1733
+ "epoch": 7.37,
1734
+ "learning_rate": 5.853935777209435e-06,
1735
+ "loss": 0.4051,
1736
+ "step": 2880
1737
+ },
1738
+ {
1739
+ "epoch": 7.39,
1740
+ "learning_rate": 5.797101449275363e-06,
1741
+ "loss": 0.4045,
1742
+ "step": 2890
1743
+ },
1744
+ {
1745
+ "epoch": 7.42,
1746
+ "learning_rate": 5.740267121341291e-06,
1747
+ "loss": 0.4039,
1748
+ "step": 2900
1749
+ },
1750
+ {
1751
+ "epoch": 7.44,
1752
+ "learning_rate": 5.683432793407219e-06,
1753
+ "loss": 0.4158,
1754
+ "step": 2910
1755
+ },
1756
+ {
1757
+ "epoch": 7.47,
1758
+ "learning_rate": 5.626598465473146e-06,
1759
+ "loss": 0.3805,
1760
+ "step": 2920
1761
+ },
1762
+ {
1763
+ "epoch": 7.49,
1764
+ "learning_rate": 5.569764137539074e-06,
1765
+ "loss": 0.4288,
1766
+ "step": 2930
1767
+ },
1768
+ {
1769
+ "epoch": 7.52,
1770
+ "learning_rate": 5.512929809605002e-06,
1771
+ "loss": 0.3905,
1772
+ "step": 2940
1773
+ },
1774
+ {
1775
+ "epoch": 7.54,
1776
+ "learning_rate": 5.45609548167093e-06,
1777
+ "loss": 0.4227,
1778
+ "step": 2950
1779
+ },
1780
+ {
1781
+ "epoch": 7.57,
1782
+ "learning_rate": 5.399261153736858e-06,
1783
+ "loss": 0.437,
1784
+ "step": 2960
1785
+ },
1786
+ {
1787
+ "epoch": 7.6,
1788
+ "learning_rate": 5.342426825802786e-06,
1789
+ "loss": 0.4289,
1790
+ "step": 2970
1791
+ },
1792
+ {
1793
+ "epoch": 7.62,
1794
+ "learning_rate": 5.285592497868714e-06,
1795
+ "loss": 0.4034,
1796
+ "step": 2980
1797
+ },
1798
+ {
1799
+ "epoch": 7.65,
1800
+ "learning_rate": 5.2287581699346416e-06,
1801
+ "loss": 0.4394,
1802
+ "step": 2990
1803
+ },
1804
+ {
1805
+ "epoch": 7.67,
1806
+ "learning_rate": 5.1719238420005695e-06,
1807
+ "loss": 0.4292,
1808
+ "step": 3000
1809
+ },
1810
+ {
1811
+ "epoch": 7.7,
1812
+ "learning_rate": 5.1150895140664966e-06,
1813
+ "loss": 0.4219,
1814
+ "step": 3010
1815
+ },
1816
+ {
1817
+ "epoch": 7.72,
1818
+ "learning_rate": 5.0582551861324245e-06,
1819
+ "loss": 0.4181,
1820
+ "step": 3020
1821
+ },
1822
+ {
1823
+ "epoch": 7.75,
1824
+ "learning_rate": 5.001420858198352e-06,
1825
+ "loss": 0.3983,
1826
+ "step": 3030
1827
+ },
1828
+ {
1829
+ "epoch": 7.77,
1830
+ "learning_rate": 4.9445865302642795e-06,
1831
+ "loss": 0.4167,
1832
+ "step": 3040
1833
+ },
1834
+ {
1835
+ "epoch": 7.8,
1836
+ "learning_rate": 4.887752202330207e-06,
1837
+ "loss": 0.3953,
1838
+ "step": 3050
1839
+ },
1840
+ {
1841
+ "epoch": 7.83,
1842
+ "learning_rate": 4.830917874396135e-06,
1843
+ "loss": 0.4081,
1844
+ "step": 3060
1845
+ },
1846
+ {
1847
+ "epoch": 7.85,
1848
+ "learning_rate": 4.774083546462063e-06,
1849
+ "loss": 0.3817,
1850
+ "step": 3070
1851
+ },
1852
+ {
1853
+ "epoch": 7.88,
1854
+ "learning_rate": 4.717249218527991e-06,
1855
+ "loss": 0.4229,
1856
+ "step": 3080
1857
+ },
1858
+ {
1859
+ "epoch": 7.9,
1860
+ "learning_rate": 4.660414890593919e-06,
1861
+ "loss": 0.41,
1862
+ "step": 3090
1863
+ },
1864
+ {
1865
+ "epoch": 7.93,
1866
+ "learning_rate": 4.603580562659847e-06,
1867
+ "loss": 0.3895,
1868
+ "step": 3100
1869
+ },
1870
+ {
1871
+ "epoch": 7.95,
1872
+ "learning_rate": 4.546746234725775e-06,
1873
+ "loss": 0.4021,
1874
+ "step": 3110
1875
+ },
1876
+ {
1877
+ "epoch": 7.98,
1878
+ "learning_rate": 4.489911906791703e-06,
1879
+ "loss": 0.419,
1880
+ "step": 3120
1881
+ },
1882
+ {
1883
+ "epoch": 8.01,
1884
+ "learning_rate": 4.43307757885763e-06,
1885
+ "loss": 0.4308,
1886
+ "step": 3130
1887
+ },
1888
+ {
1889
+ "epoch": 8.03,
1890
+ "learning_rate": 4.376243250923558e-06,
1891
+ "loss": 0.3788,
1892
+ "step": 3140
1893
+ },
1894
+ {
1895
+ "epoch": 8.06,
1896
+ "learning_rate": 4.319408922989486e-06,
1897
+ "loss": 0.3693,
1898
+ "step": 3150
1899
+ },
1900
+ {
1901
+ "epoch": 8.08,
1902
+ "learning_rate": 4.262574595055414e-06,
1903
+ "loss": 0.3889,
1904
+ "step": 3160
1905
+ },
1906
+ {
1907
+ "epoch": 8.11,
1908
+ "learning_rate": 4.205740267121342e-06,
1909
+ "loss": 0.4061,
1910
+ "step": 3170
1911
+ },
1912
+ {
1913
+ "epoch": 8.13,
1914
+ "learning_rate": 4.1489059391872695e-06,
1915
+ "loss": 0.3893,
1916
+ "step": 3180
1917
+ },
1918
+ {
1919
+ "epoch": 8.16,
1920
+ "learning_rate": 4.092071611253197e-06,
1921
+ "loss": 0.3829,
1922
+ "step": 3190
1923
+ },
1924
+ {
1925
+ "epoch": 8.18,
1926
+ "learning_rate": 4.035237283319125e-06,
1927
+ "loss": 0.3624,
1928
+ "step": 3200
1929
+ },
1930
+ {
1931
+ "epoch": 8.21,
1932
+ "learning_rate": 3.978402955385052e-06,
1933
+ "loss": 0.3947,
1934
+ "step": 3210
1935
+ },
1936
+ {
1937
+ "epoch": 8.24,
1938
+ "learning_rate": 3.92156862745098e-06,
1939
+ "loss": 0.4006,
1940
+ "step": 3220
1941
+ },
1942
+ {
1943
+ "epoch": 8.26,
1944
+ "learning_rate": 3.864734299516908e-06,
1945
+ "loss": 0.3909,
1946
+ "step": 3230
1947
+ },
1948
+ {
1949
+ "epoch": 8.29,
1950
+ "learning_rate": 3.807899971582836e-06,
1951
+ "loss": 0.3744,
1952
+ "step": 3240
1953
+ },
1954
+ {
1955
+ "epoch": 8.31,
1956
+ "learning_rate": 3.751065643648764e-06,
1957
+ "loss": 0.4052,
1958
+ "step": 3250
1959
+ },
1960
+ {
1961
+ "epoch": 8.34,
1962
+ "learning_rate": 3.694231315714692e-06,
1963
+ "loss": 0.3801,
1964
+ "step": 3260
1965
+ },
1966
+ {
1967
+ "epoch": 8.36,
1968
+ "learning_rate": 3.63739698778062e-06,
1969
+ "loss": 0.3822,
1970
+ "step": 3270
1971
+ },
1972
+ {
1973
+ "epoch": 8.39,
1974
+ "learning_rate": 3.5805626598465474e-06,
1975
+ "loss": 0.3856,
1976
+ "step": 3280
1977
+ },
1978
+ {
1979
+ "epoch": 8.41,
1980
+ "learning_rate": 3.5237283319124753e-06,
1981
+ "loss": 0.3644,
1982
+ "step": 3290
1983
+ },
1984
+ {
1985
+ "epoch": 8.44,
1986
+ "learning_rate": 3.4668940039784033e-06,
1987
+ "loss": 0.3844,
1988
+ "step": 3300
1989
+ },
1990
+ {
1991
+ "epoch": 8.47,
1992
+ "learning_rate": 3.410059676044331e-06,
1993
+ "loss": 0.4009,
1994
+ "step": 3310
1995
+ },
1996
+ {
1997
+ "epoch": 8.49,
1998
+ "learning_rate": 3.353225348110259e-06,
1999
+ "loss": 0.3886,
2000
+ "step": 3320
2001
+ },
2002
+ {
2003
+ "epoch": 8.52,
2004
+ "learning_rate": 3.2963910201761866e-06,
2005
+ "loss": 0.3711,
2006
+ "step": 3330
2007
+ },
2008
+ {
2009
+ "epoch": 8.54,
2010
+ "learning_rate": 3.2395566922421145e-06,
2011
+ "loss": 0.3844,
2012
+ "step": 3340
2013
+ },
2014
+ {
2015
+ "epoch": 8.57,
2016
+ "learning_rate": 3.1827223643080424e-06,
2017
+ "loss": 0.3861,
2018
+ "step": 3350
2019
+ },
2020
+ {
2021
+ "epoch": 8.59,
2022
+ "learning_rate": 3.1258880363739704e-06,
2023
+ "loss": 0.3874,
2024
+ "step": 3360
2025
+ },
2026
+ {
2027
+ "epoch": 8.62,
2028
+ "learning_rate": 3.069053708439898e-06,
2029
+ "loss": 0.3617,
2030
+ "step": 3370
2031
+ },
2032
+ {
2033
+ "epoch": 8.64,
2034
+ "learning_rate": 3.0122193805058258e-06,
2035
+ "loss": 0.37,
2036
+ "step": 3380
2037
+ },
2038
+ {
2039
+ "epoch": 8.67,
2040
+ "learning_rate": 2.9553850525717537e-06,
2041
+ "loss": 0.3907,
2042
+ "step": 3390
2043
+ },
2044
+ {
2045
+ "epoch": 8.7,
2046
+ "learning_rate": 2.8985507246376816e-06,
2047
+ "loss": 0.3958,
2048
+ "step": 3400
2049
+ },
2050
+ {
2051
+ "epoch": 8.72,
2052
+ "learning_rate": 2.8417163967036095e-06,
2053
+ "loss": 0.3868,
2054
+ "step": 3410
2055
+ },
2056
+ {
2057
+ "epoch": 8.75,
2058
+ "learning_rate": 2.784882068769537e-06,
2059
+ "loss": 0.3755,
2060
+ "step": 3420
2061
+ },
2062
+ {
2063
+ "epoch": 8.77,
2064
+ "learning_rate": 2.728047740835465e-06,
2065
+ "loss": 0.371,
2066
+ "step": 3430
2067
+ },
2068
+ {
2069
+ "epoch": 8.8,
2070
+ "learning_rate": 2.671213412901393e-06,
2071
+ "loss": 0.3793,
2072
+ "step": 3440
2073
+ },
2074
+ {
2075
+ "epoch": 8.82,
2076
+ "learning_rate": 2.6143790849673208e-06,
2077
+ "loss": 0.3588,
2078
+ "step": 3450
2079
+ },
2080
+ {
2081
+ "epoch": 8.85,
2082
+ "learning_rate": 2.5575447570332483e-06,
2083
+ "loss": 0.3868,
2084
+ "step": 3460
2085
+ },
2086
+ {
2087
+ "epoch": 8.87,
2088
+ "learning_rate": 2.500710429099176e-06,
2089
+ "loss": 0.3826,
2090
+ "step": 3470
2091
+ },
2092
+ {
2093
+ "epoch": 8.9,
2094
+ "learning_rate": 2.4438761011651037e-06,
2095
+ "loss": 0.3868,
2096
+ "step": 3480
2097
+ },
2098
+ {
2099
+ "epoch": 8.93,
2100
+ "learning_rate": 2.3870417732310316e-06,
2101
+ "loss": 0.4038,
2102
+ "step": 3490
2103
+ },
2104
+ {
2105
+ "epoch": 8.95,
2106
+ "learning_rate": 2.3302074452969595e-06,
2107
+ "loss": 0.3527,
2108
+ "step": 3500
2109
+ },
2110
+ {
2111
+ "epoch": 8.98,
2112
+ "learning_rate": 2.2733731173628875e-06,
2113
+ "loss": 0.3766,
2114
+ "step": 3510
2115
+ },
2116
+ {
2117
+ "epoch": 9.0,
2118
+ "learning_rate": 2.216538789428815e-06,
2119
+ "loss": 0.3853,
2120
+ "step": 3520
2121
+ },
2122
+ {
2123
+ "epoch": 9.03,
2124
+ "learning_rate": 2.159704461494743e-06,
2125
+ "loss": 0.36,
2126
+ "step": 3530
2127
+ },
2128
+ {
2129
+ "epoch": 9.05,
2130
+ "learning_rate": 2.102870133560671e-06,
2131
+ "loss": 0.3456,
2132
+ "step": 3540
2133
+ },
2134
+ {
2135
+ "epoch": 9.08,
2136
+ "learning_rate": 2.0460358056265987e-06,
2137
+ "loss": 0.3509,
2138
+ "step": 3550
2139
+ },
2140
+ {
2141
+ "epoch": 9.1,
2142
+ "learning_rate": 1.989201477692526e-06,
2143
+ "loss": 0.3565,
2144
+ "step": 3560
2145
+ },
2146
+ {
2147
+ "epoch": 9.13,
2148
+ "learning_rate": 1.932367149758454e-06,
2149
+ "loss": 0.3911,
2150
+ "step": 3570
2151
+ },
2152
+ {
2153
+ "epoch": 9.16,
2154
+ "learning_rate": 1.875532821824382e-06,
2155
+ "loss": 0.3853,
2156
+ "step": 3580
2157
+ },
2158
+ {
2159
+ "epoch": 9.18,
2160
+ "learning_rate": 1.81869849389031e-06,
2161
+ "loss": 0.3816,
2162
+ "step": 3590
2163
+ },
2164
+ {
2165
+ "epoch": 9.21,
2166
+ "learning_rate": 1.7618641659562377e-06,
2167
+ "loss": 0.3679,
2168
+ "step": 3600
2169
+ },
2170
+ {
2171
+ "epoch": 9.23,
2172
+ "learning_rate": 1.7050298380221656e-06,
2173
+ "loss": 0.365,
2174
+ "step": 3610
2175
+ },
2176
+ {
2177
+ "epoch": 9.26,
2178
+ "learning_rate": 1.6481955100880933e-06,
2179
+ "loss": 0.346,
2180
+ "step": 3620
2181
+ },
2182
+ {
2183
+ "epoch": 9.28,
2184
+ "learning_rate": 1.5913611821540212e-06,
2185
+ "loss": 0.3598,
2186
+ "step": 3630
2187
+ },
2188
+ {
2189
+ "epoch": 9.31,
2190
+ "learning_rate": 1.534526854219949e-06,
2191
+ "loss": 0.3428,
2192
+ "step": 3640
2193
+ },
2194
+ {
2195
+ "epoch": 9.33,
2196
+ "learning_rate": 1.4776925262858768e-06,
2197
+ "loss": 0.3959,
2198
+ "step": 3650
2199
+ },
2200
+ {
2201
+ "epoch": 9.36,
2202
+ "learning_rate": 1.4208581983518048e-06,
2203
+ "loss": 0.3707,
2204
+ "step": 3660
2205
+ },
2206
+ {
2207
+ "epoch": 9.39,
2208
+ "learning_rate": 1.3640238704177325e-06,
2209
+ "loss": 0.3541,
2210
+ "step": 3670
2211
+ },
2212
+ {
2213
+ "epoch": 9.41,
2214
+ "learning_rate": 1.3071895424836604e-06,
2215
+ "loss": 0.3691,
2216
+ "step": 3680
2217
+ },
2218
+ {
2219
+ "epoch": 9.44,
2220
+ "learning_rate": 1.250355214549588e-06,
2221
+ "loss": 0.3736,
2222
+ "step": 3690
2223
+ },
2224
+ {
2225
+ "epoch": 9.46,
2226
+ "learning_rate": 1.1935208866155158e-06,
2227
+ "loss": 0.3496,
2228
+ "step": 3700
2229
+ },
2230
+ {
2231
+ "epoch": 9.49,
2232
+ "learning_rate": 1.1366865586814437e-06,
2233
+ "loss": 0.3369,
2234
+ "step": 3710
2235
+ },
2236
+ {
2237
+ "epoch": 9.51,
2238
+ "learning_rate": 1.0798522307473714e-06,
2239
+ "loss": 0.3959,
2240
+ "step": 3720
2241
+ },
2242
+ {
2243
+ "epoch": 9.54,
2244
+ "learning_rate": 1.0230179028132994e-06,
2245
+ "loss": 0.3604,
2246
+ "step": 3730
2247
+ },
2248
+ {
2249
+ "epoch": 9.57,
2250
+ "learning_rate": 9.66183574879227e-07,
2251
+ "loss": 0.3569,
2252
+ "step": 3740
2253
+ },
2254
+ {
2255
+ "epoch": 9.59,
2256
+ "learning_rate": 9.09349246945155e-07,
2257
+ "loss": 0.3648,
2258
+ "step": 3750
2259
+ },
2260
+ {
2261
+ "epoch": 9.62,
2262
+ "learning_rate": 8.525149190110828e-07,
2263
+ "loss": 0.354,
2264
+ "step": 3760
2265
+ },
2266
+ {
2267
+ "epoch": 9.64,
2268
+ "learning_rate": 7.956805910770106e-07,
2269
+ "loss": 0.3331,
2270
+ "step": 3770
2271
+ },
2272
+ {
2273
+ "epoch": 9.67,
2274
+ "learning_rate": 7.388462631429384e-07,
2275
+ "loss": 0.384,
2276
+ "step": 3780
2277
+ },
2278
+ {
2279
+ "epoch": 9.69,
2280
+ "learning_rate": 6.820119352088662e-07,
2281
+ "loss": 0.3496,
2282
+ "step": 3790
2283
+ },
2284
+ {
2285
+ "epoch": 9.72,
2286
+ "learning_rate": 6.25177607274794e-07,
2287
+ "loss": 0.3779,
2288
+ "step": 3800
2289
+ },
2290
+ {
2291
+ "epoch": 9.74,
2292
+ "learning_rate": 5.683432793407219e-07,
2293
+ "loss": 0.3726,
2294
+ "step": 3810
2295
+ },
2296
+ {
2297
+ "epoch": 9.77,
2298
+ "learning_rate": 5.115089514066497e-07,
2299
+ "loss": 0.3931,
2300
+ "step": 3820
2301
+ },
2302
+ {
2303
+ "epoch": 9.8,
2304
+ "learning_rate": 4.546746234725775e-07,
2305
+ "loss": 0.3714,
2306
+ "step": 3830
2307
+ },
2308
+ {
2309
+ "epoch": 9.82,
2310
+ "learning_rate": 3.978402955385053e-07,
2311
+ "loss": 0.3703,
2312
+ "step": 3840
2313
+ },
2314
+ {
2315
+ "epoch": 9.85,
2316
+ "learning_rate": 3.410059676044331e-07,
2317
+ "loss": 0.3559,
2318
+ "step": 3850
2319
+ },
2320
+ {
2321
+ "epoch": 9.87,
2322
+ "learning_rate": 2.8417163967036093e-07,
2323
+ "loss": 0.3656,
2324
+ "step": 3860
2325
+ },
2326
+ {
2327
+ "epoch": 9.9,
2328
+ "learning_rate": 2.2733731173628875e-07,
2329
+ "loss": 0.3534,
2330
+ "step": 3870
2331
+ },
2332
+ {
2333
+ "epoch": 9.92,
2334
+ "learning_rate": 1.7050298380221656e-07,
2335
+ "loss": 0.3769,
2336
+ "step": 3880
2337
+ },
2338
+ {
2339
+ "epoch": 9.95,
2340
+ "learning_rate": 1.1366865586814437e-07,
2341
+ "loss": 0.3442,
2342
+ "step": 3890
2343
+ },
2344
+ {
2345
+ "epoch": 9.97,
2346
+ "learning_rate": 5.6834327934072186e-08,
2347
+ "loss": 0.3558,
2348
+ "step": 3900
2349
+ },
2350
+ {
2351
+ "epoch": 10.0,
2352
+ "learning_rate": 0.0,
2353
+ "loss": 0.3395,
2354
+ "step": 3910
2355
+ },
2356
+ {
2357
+ "epoch": 10.0,
2358
+ "step": 3910,
2359
+ "total_flos": 1.0186138340339622e+17,
2360
+ "train_loss": 0.6484309974533823,
2361
+ "train_runtime": 10912.4775,
2362
+ "train_samples_per_second": 91.732,
2363
+ "train_steps_per_second": 0.358
2364
+ }
2365
+ ],
2366
+ "max_steps": 3910,
2367
+ "num_train_epochs": 10,
2368
+ "total_flos": 1.0186138340339622e+17,
2369
+ "trial_name": null,
2370
+ "trial_params": null
2371
+ }