gokuls commited on
Commit
432b0ca
1 Parent(s): 275e18c

End of training

Browse files
all_results.json CHANGED
@@ -1,15 +1,15 @@
1
  {
2
- "epoch": 0.0,
3
- "eval_accuracy": 2.228276571731114e-05,
4
- "eval_loss": 10.515625,
5
- "eval_runtime": 1062.2138,
6
  "eval_samples": 308356,
7
- "eval_samples_per_second": 290.296,
8
- "eval_steps_per_second": 4.537,
9
- "perplexity": 36887.38862753968,
10
- "train_loss": 10.515625,
11
- "train_runtime": 1.0122,
12
  "train_samples": 5858758,
13
- "train_samples_per_second": 28940457.729,
14
- "train_steps_per_second": 452199.129
15
  }
 
1
  {
2
+ "epoch": 1.85,
3
+ "eval_accuracy": 0.04297391005057371,
4
+ "eval_loss": 7.266438961029053,
5
+ "eval_runtime": 986.7506,
6
  "eval_samples": 308356,
7
+ "eval_samples_per_second": 312.496,
8
+ "eval_steps_per_second": 4.884,
9
+ "perplexity": 1431.443938534488,
10
+ "train_loss": 7.273256063168102,
11
+ "train_runtime": 107974.2539,
12
  "train_samples": 5858758,
13
+ "train_samples_per_second": 271.303,
14
+ "train_steps_per_second": 4.239
15
  }
eval_results.json CHANGED
@@ -1,10 +1,10 @@
1
  {
2
- "epoch": 0.0,
3
- "eval_accuracy": 2.228276571731114e-05,
4
- "eval_loss": 10.515625,
5
- "eval_runtime": 1062.2138,
6
  "eval_samples": 308356,
7
- "eval_samples_per_second": 290.296,
8
- "eval_steps_per_second": 4.537,
9
- "perplexity": 36887.38862753968
10
  }
 
1
  {
2
+ "epoch": 1.85,
3
+ "eval_accuracy": 0.04297391005057371,
4
+ "eval_loss": 7.266438961029053,
5
+ "eval_runtime": 986.7506,
6
  "eval_samples": 308356,
7
+ "eval_samples_per_second": 312.496,
8
+ "eval_steps_per_second": 4.884,
9
+ "perplexity": 1431.443938534488
10
  }
logs/events.out.tfevents.1685516790.serv-9220.3164122.2 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:09f7a944e41051c98df89bf52550ddd36bb26b9d47524ea80ef8ab0f0d321ad9
3
+ size 369
train_results.json CHANGED
@@ -1,8 +1,8 @@
1
  {
2
- "epoch": 0.0,
3
- "train_loss": 10.515625,
4
- "train_runtime": 1.0122,
5
  "train_samples": 5858758,
6
- "train_samples_per_second": 28940457.729,
7
- "train_steps_per_second": 452199.129
8
  }
 
1
  {
2
+ "epoch": 1.85,
3
+ "train_loss": 7.273256063168102,
4
+ "train_runtime": 107974.2539,
5
  "train_samples": 5858758,
6
+ "train_samples_per_second": 271.303,
7
+ "train_steps_per_second": 4.239
8
  }
trainer_state.json CHANGED
@@ -1,25 +1,2197 @@
1
  {
2
- "best_metric": null,
3
- "best_model_checkpoint": null,
4
- "epoch": 1.0923708817617758e-05,
5
- "global_step": 1,
6
  "is_hyper_param_search": false,
7
  "is_local_process_zero": true,
8
  "is_world_process_zero": true,
9
  "log_history": [
10
  {
11
- "epoch": 0.0,
12
- "step": 1,
13
- "total_flos": 18737413488640.0,
14
- "train_loss": 10.515625,
15
- "train_runtime": 1.0122,
16
- "train_samples_per_second": 28940457.729,
17
- "train_steps_per_second": 452199.129
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
18
  }
19
  ],
20
  "max_steps": 457720,
21
  "num_train_epochs": 5,
22
- "total_flos": 18737413488640.0,
23
  "trial_name": null,
24
  "trial_params": null
25
  }
 
1
  {
2
+ "best_metric": 7.266932487487793,
3
+ "best_model_checkpoint": "bert_12_layer_model_v2_complete_training_new/checkpoint-30000",
4
+ "epoch": 1.8475924145765972,
5
+ "global_step": 169136,
6
  "is_hyper_param_search": false,
7
  "is_local_process_zero": true,
8
  "is_world_process_zero": true,
9
  "log_history": [
10
  {
11
+ "epoch": 0.01,
12
+ "learning_rate": 0.0006747425010840046,
13
+ "loss": 7.3929,
14
+ "step": 500
15
+ },
16
+ {
17
+ "epoch": 0.01,
18
+ "learning_rate": 0.0007499999999999999,
19
+ "loss": 7.2987,
20
+ "step": 1000
21
+ },
22
+ {
23
+ "epoch": 0.02,
24
+ "learning_rate": 0.0007940228147639202,
25
+ "loss": 7.2939,
26
+ "step": 1500
27
+ },
28
+ {
29
+ "epoch": 0.02,
30
+ "learning_rate": 0.0008252574989159952,
31
+ "loss": 7.2875,
32
+ "step": 2000
33
+ },
34
+ {
35
+ "epoch": 0.03,
36
+ "learning_rate": 0.0008494850021680092,
37
+ "loss": 7.287,
38
+ "step": 2500
39
+ },
40
+ {
41
+ "epoch": 0.03,
42
+ "learning_rate": 0.0008692803136799154,
43
+ "loss": 7.2802,
44
+ "step": 3000
45
+ },
46
+ {
47
+ "epoch": 0.04,
48
+ "learning_rate": 0.0008860170110875688,
49
+ "loss": 7.2847,
50
+ "step": 3500
51
+ },
52
+ {
53
+ "epoch": 0.04,
54
+ "learning_rate": 0.0009005149978319905,
55
+ "loss": 7.2797,
56
+ "step": 4000
57
+ },
58
+ {
59
+ "epoch": 0.05,
60
+ "learning_rate": 0.0009133031284438358,
61
+ "loss": 7.2796,
62
+ "step": 4500
63
+ },
64
+ {
65
+ "epoch": 0.05,
66
+ "learning_rate": 0.0009247425010840047,
67
+ "loss": 7.2861,
68
+ "step": 5000
69
+ },
70
+ {
71
+ "epoch": 0.06,
72
+ "learning_rate": 0.0009350906723735609,
73
+ "loss": 7.276,
74
+ "step": 5500
75
+ },
76
+ {
77
+ "epoch": 0.07,
78
+ "learning_rate": 0.0009445378125959107,
79
+ "loss": 7.2843,
80
+ "step": 6000
81
+ },
82
+ {
83
+ "epoch": 0.07,
84
+ "learning_rate": 0.0009532283391607137,
85
+ "loss": 7.2736,
86
+ "step": 6500
87
+ },
88
+ {
89
+ "epoch": 0.08,
90
+ "learning_rate": 0.000961274510003564,
91
+ "loss": 7.2744,
92
+ "step": 7000
93
+ },
94
+ {
95
+ "epoch": 0.08,
96
+ "learning_rate": 0.000968765315847925,
97
+ "loss": 7.2792,
98
+ "step": 7500
99
+ },
100
+ {
101
+ "epoch": 0.09,
102
+ "learning_rate": 0.0009757724967479857,
103
+ "loss": 7.2781,
104
+ "step": 8000
105
+ },
106
+ {
107
+ "epoch": 0.09,
108
+ "learning_rate": 0.000982354731428573,
109
+ "loss": 7.2828,
110
+ "step": 8500
111
+ },
112
+ {
113
+ "epoch": 0.1,
114
+ "learning_rate": 0.000988560627359831,
115
+ "loss": 7.2828,
116
+ "step": 9000
117
+ },
118
+ {
119
+ "epoch": 0.1,
120
+ "learning_rate": 0.0009944309013222117,
121
+ "loss": 7.2761,
122
+ "step": 9500
123
+ },
124
+ {
125
+ "epoch": 0.11,
126
+ "learning_rate": 0.0009999999999999998,
127
+ "loss": 7.2752,
128
+ "step": 10000
129
+ },
130
+ {
131
+ "epoch": 0.11,
132
+ "eval_accuracy": 0.046582492639500865,
133
+ "eval_loss": 7.277245998382568,
134
+ "eval_runtime": 989.9846,
135
+ "eval_samples_per_second": 311.476,
136
+ "eval_steps_per_second": 4.868,
137
+ "step": 10000
138
+ },
139
+ {
140
+ "epoch": 0.11,
141
+ "learning_rate": 0.0009988854641293666,
142
+ "loss": 7.6758,
143
+ "step": 10500
144
+ },
145
+ {
146
+ "epoch": 0.12,
147
+ "learning_rate": 0.0009977686947199142,
148
+ "loss": 7.3739,
149
+ "step": 11000
150
+ },
151
+ {
152
+ "epoch": 0.13,
153
+ "learning_rate": 0.0009966519253104618,
154
+ "loss": 7.3249,
155
+ "step": 11500
156
+ },
157
+ {
158
+ "epoch": 0.13,
159
+ "learning_rate": 0.0009955351559010096,
160
+ "loss": 7.3018,
161
+ "step": 12000
162
+ },
163
+ {
164
+ "epoch": 0.14,
165
+ "learning_rate": 0.0009944183864915572,
166
+ "loss": 7.2977,
167
+ "step": 12500
168
+ },
169
+ {
170
+ "epoch": 0.14,
171
+ "learning_rate": 0.0009933016170821049,
172
+ "loss": 7.2901,
173
+ "step": 13000
174
+ },
175
+ {
176
+ "epoch": 0.15,
177
+ "learning_rate": 0.0009921848476726527,
178
+ "loss": 7.2877,
179
+ "step": 13500
180
+ },
181
+ {
182
+ "epoch": 0.15,
183
+ "learning_rate": 0.0009910680782632003,
184
+ "loss": 7.2851,
185
+ "step": 14000
186
+ },
187
+ {
188
+ "epoch": 0.16,
189
+ "learning_rate": 0.000989951308853748,
190
+ "loss": 7.2809,
191
+ "step": 14500
192
+ },
193
+ {
194
+ "epoch": 0.16,
195
+ "learning_rate": 0.0009888345394442955,
196
+ "loss": 7.2812,
197
+ "step": 15000
198
+ },
199
+ {
200
+ "epoch": 0.17,
201
+ "learning_rate": 0.0009877177700348431,
202
+ "loss": 7.2785,
203
+ "step": 15500
204
+ },
205
+ {
206
+ "epoch": 0.17,
207
+ "learning_rate": 0.000986601000625391,
208
+ "loss": 7.2757,
209
+ "step": 16000
210
+ },
211
+ {
212
+ "epoch": 0.18,
213
+ "learning_rate": 0.0009854842312159386,
214
+ "loss": 7.2718,
215
+ "step": 16500
216
+ },
217
+ {
218
+ "epoch": 0.19,
219
+ "learning_rate": 0.0009843674618064862,
220
+ "loss": 7.2774,
221
+ "step": 17000
222
+ },
223
+ {
224
+ "epoch": 0.19,
225
+ "learning_rate": 0.000983250692397034,
226
+ "loss": 7.2725,
227
+ "step": 17500
228
+ },
229
+ {
230
+ "epoch": 0.2,
231
+ "learning_rate": 0.0009821339229875816,
232
+ "loss": 7.2699,
233
+ "step": 18000
234
+ },
235
+ {
236
+ "epoch": 0.2,
237
+ "learning_rate": 0.0009810171535781292,
238
+ "loss": 7.2708,
239
+ "step": 18500
240
+ },
241
+ {
242
+ "epoch": 0.21,
243
+ "learning_rate": 0.0009799003841686768,
244
+ "loss": 7.2673,
245
+ "step": 19000
246
+ },
247
+ {
248
+ "epoch": 0.21,
249
+ "learning_rate": 0.0009787836147592244,
250
+ "loss": 7.2731,
251
+ "step": 19500
252
+ },
253
+ {
254
+ "epoch": 0.22,
255
+ "learning_rate": 0.0009776668453497723,
256
+ "loss": 7.2774,
257
+ "step": 20000
258
+ },
259
+ {
260
+ "epoch": 0.22,
261
+ "eval_accuracy": 0.042795053620331966,
262
+ "eval_loss": 7.277062892913818,
263
+ "eval_runtime": 988.361,
264
+ "eval_samples_per_second": 311.987,
265
+ "eval_steps_per_second": 4.876,
266
+ "step": 20000
267
+ },
268
+ {
269
+ "epoch": 0.22,
270
+ "learning_rate": 0.0009765500759403199,
271
+ "loss": 7.2699,
272
+ "step": 20500
273
+ },
274
+ {
275
+ "epoch": 0.23,
276
+ "learning_rate": 0.0009754333065308675,
277
+ "loss": 7.2739,
278
+ "step": 21000
279
+ },
280
+ {
281
+ "epoch": 0.23,
282
+ "learning_rate": 0.0009743165371214152,
283
+ "loss": 7.2693,
284
+ "step": 21500
285
+ },
286
+ {
287
+ "epoch": 0.24,
288
+ "learning_rate": 0.0009731997677119628,
289
+ "loss": 7.2696,
290
+ "step": 22000
291
+ },
292
+ {
293
+ "epoch": 0.25,
294
+ "learning_rate": 0.0009720829983025105,
295
+ "loss": 7.2728,
296
+ "step": 22500
297
+ },
298
+ {
299
+ "epoch": 0.25,
300
+ "learning_rate": 0.0009709662288930582,
301
+ "loss": 7.2706,
302
+ "step": 23000
303
+ },
304
+ {
305
+ "epoch": 0.26,
306
+ "learning_rate": 0.0009698494594836058,
307
+ "loss": 7.2725,
308
+ "step": 23500
309
+ },
310
+ {
311
+ "epoch": 0.26,
312
+ "learning_rate": 0.0009687326900741536,
313
+ "loss": 7.2719,
314
+ "step": 24000
315
+ },
316
+ {
317
+ "epoch": 0.27,
318
+ "learning_rate": 0.0009676159206647012,
319
+ "loss": 7.2691,
320
+ "step": 24500
321
+ },
322
+ {
323
+ "epoch": 0.27,
324
+ "learning_rate": 0.0009664991512552488,
325
+ "loss": 7.2709,
326
+ "step": 25000
327
+ },
328
+ {
329
+ "epoch": 0.28,
330
+ "learning_rate": 0.0009653823818457965,
331
+ "loss": 7.2683,
332
+ "step": 25500
333
+ },
334
+ {
335
+ "epoch": 0.28,
336
+ "learning_rate": 0.0009642656124363441,
337
+ "loss": 7.2694,
338
+ "step": 26000
339
+ },
340
+ {
341
+ "epoch": 0.29,
342
+ "learning_rate": 0.0009631488430268918,
343
+ "loss": 7.2734,
344
+ "step": 26500
345
+ },
346
+ {
347
+ "epoch": 0.29,
348
+ "learning_rate": 0.0009620320736174395,
349
+ "loss": 7.275,
350
+ "step": 27000
351
+ },
352
+ {
353
+ "epoch": 0.3,
354
+ "learning_rate": 0.0009609153042079872,
355
+ "loss": 7.2742,
356
+ "step": 27500
357
+ },
358
+ {
359
+ "epoch": 0.31,
360
+ "learning_rate": 0.0009597985347985349,
361
+ "loss": 7.2684,
362
+ "step": 28000
363
+ },
364
+ {
365
+ "epoch": 0.31,
366
+ "learning_rate": 0.0009586817653890825,
367
+ "loss": 7.2733,
368
+ "step": 28500
369
+ },
370
+ {
371
+ "epoch": 0.32,
372
+ "learning_rate": 0.0009575649959796301,
373
+ "loss": 7.2661,
374
+ "step": 29000
375
+ },
376
+ {
377
+ "epoch": 0.32,
378
+ "learning_rate": 0.0009564482265701778,
379
+ "loss": 7.2709,
380
+ "step": 29500
381
+ },
382
+ {
383
+ "epoch": 0.33,
384
+ "learning_rate": 0.0009553314571607254,
385
+ "loss": 7.2722,
386
+ "step": 30000
387
+ },
388
+ {
389
+ "epoch": 0.33,
390
+ "eval_accuracy": 0.04286128173369546,
391
+ "eval_loss": 7.266932487487793,
392
+ "eval_runtime": 986.3612,
393
+ "eval_samples_per_second": 312.62,
394
+ "eval_steps_per_second": 4.886,
395
+ "step": 30000
396
+ },
397
+ {
398
+ "epoch": 0.33,
399
+ "learning_rate": 0.0009542146877512731,
400
+ "loss": 7.2704,
401
+ "step": 30500
402
+ },
403
+ {
404
+ "epoch": 0.34,
405
+ "learning_rate": 0.0009530979183418209,
406
+ "loss": 7.2717,
407
+ "step": 31000
408
+ },
409
+ {
410
+ "epoch": 0.34,
411
+ "learning_rate": 0.0009519811489323685,
412
+ "loss": 7.2686,
413
+ "step": 31500
414
+ },
415
+ {
416
+ "epoch": 0.35,
417
+ "learning_rate": 0.0009508643795229162,
418
+ "loss": 7.2683,
419
+ "step": 32000
420
+ },
421
+ {
422
+ "epoch": 0.36,
423
+ "learning_rate": 0.0009497476101134638,
424
+ "loss": 7.2707,
425
+ "step": 32500
426
+ },
427
+ {
428
+ "epoch": 0.36,
429
+ "learning_rate": 0.0009486308407040115,
430
+ "loss": 7.2702,
431
+ "step": 33000
432
+ },
433
+ {
434
+ "epoch": 0.37,
435
+ "learning_rate": 0.0009475140712945591,
436
+ "loss": 7.2709,
437
+ "step": 33500
438
+ },
439
+ {
440
+ "epoch": 0.37,
441
+ "learning_rate": 0.0009463973018851067,
442
+ "loss": 7.2695,
443
+ "step": 34000
444
+ },
445
+ {
446
+ "epoch": 0.38,
447
+ "learning_rate": 0.0009452805324756544,
448
+ "loss": 7.2683,
449
+ "step": 34500
450
+ },
451
+ {
452
+ "epoch": 0.38,
453
+ "learning_rate": 0.0009441637630662021,
454
+ "loss": 7.2669,
455
+ "step": 35000
456
+ },
457
+ {
458
+ "epoch": 0.39,
459
+ "learning_rate": 0.0009430469936567498,
460
+ "loss": 7.2652,
461
+ "step": 35500
462
+ },
463
+ {
464
+ "epoch": 0.39,
465
+ "learning_rate": 0.0009419302242472975,
466
+ "loss": 7.271,
467
+ "step": 36000
468
+ },
469
+ {
470
+ "epoch": 0.4,
471
+ "learning_rate": 0.0009408134548378451,
472
+ "loss": 7.2651,
473
+ "step": 36500
474
+ },
475
+ {
476
+ "epoch": 0.4,
477
+ "learning_rate": 0.0009396966854283928,
478
+ "loss": 7.2716,
479
+ "step": 37000
480
+ },
481
+ {
482
+ "epoch": 0.41,
483
+ "learning_rate": 0.0009385799160189404,
484
+ "loss": 7.2654,
485
+ "step": 37500
486
+ },
487
+ {
488
+ "epoch": 0.42,
489
+ "learning_rate": 0.000937463146609488,
490
+ "loss": 7.2616,
491
+ "step": 38000
492
+ },
493
+ {
494
+ "epoch": 0.42,
495
+ "learning_rate": 0.0009363463772000358,
496
+ "loss": 7.2743,
497
+ "step": 38500
498
+ },
499
+ {
500
+ "epoch": 0.43,
501
+ "learning_rate": 0.0009352296077905834,
502
+ "loss": 7.2899,
503
+ "step": 39000
504
+ },
505
+ {
506
+ "epoch": 0.43,
507
+ "learning_rate": 0.0009341128383811311,
508
+ "loss": 7.2732,
509
+ "step": 39500
510
+ },
511
+ {
512
+ "epoch": 0.44,
513
+ "learning_rate": 0.0009329960689716788,
514
+ "loss": 7.27,
515
+ "step": 40000
516
+ },
517
+ {
518
+ "epoch": 0.44,
519
+ "eval_accuracy": 0.04655514477110356,
520
+ "eval_loss": 7.268455982208252,
521
+ "eval_runtime": 985.521,
522
+ "eval_samples_per_second": 312.886,
523
+ "eval_steps_per_second": 4.89,
524
+ "step": 40000
525
+ },
526
+ {
527
+ "epoch": 0.44,
528
+ "learning_rate": 0.0009318792995622264,
529
+ "loss": 7.2683,
530
+ "step": 40500
531
+ },
532
+ {
533
+ "epoch": 0.45,
534
+ "learning_rate": 0.0009307625301527741,
535
+ "loss": 7.2701,
536
+ "step": 41000
537
+ },
538
+ {
539
+ "epoch": 0.45,
540
+ "learning_rate": 0.0009296457607433217,
541
+ "loss": 7.2758,
542
+ "step": 41500
543
+ },
544
+ {
545
+ "epoch": 0.46,
546
+ "learning_rate": 0.0009285289913338693,
547
+ "loss": 7.2712,
548
+ "step": 42000
549
+ },
550
+ {
551
+ "epoch": 0.46,
552
+ "learning_rate": 0.0009274122219244171,
553
+ "loss": 7.2724,
554
+ "step": 42500
555
+ },
556
+ {
557
+ "epoch": 0.47,
558
+ "learning_rate": 0.0009262954525149647,
559
+ "loss": 7.2624,
560
+ "step": 43000
561
+ },
562
+ {
563
+ "epoch": 0.48,
564
+ "learning_rate": 0.0009251786831055124,
565
+ "loss": 7.2748,
566
+ "step": 43500
567
+ },
568
+ {
569
+ "epoch": 0.48,
570
+ "learning_rate": 0.0009240619136960601,
571
+ "loss": 7.2741,
572
+ "step": 44000
573
+ },
574
+ {
575
+ "epoch": 0.49,
576
+ "learning_rate": 0.0009229451442866077,
577
+ "loss": 7.2716,
578
+ "step": 44500
579
+ },
580
+ {
581
+ "epoch": 0.49,
582
+ "learning_rate": 0.0009218283748771554,
583
+ "loss": 7.2678,
584
+ "step": 45000
585
+ },
586
+ {
587
+ "epoch": 0.5,
588
+ "learning_rate": 0.000920711605467703,
589
+ "loss": 7.2704,
590
+ "step": 45500
591
+ },
592
+ {
593
+ "epoch": 0.5,
594
+ "learning_rate": 0.0009195948360582507,
595
+ "loss": 7.2672,
596
+ "step": 46000
597
+ },
598
+ {
599
+ "epoch": 0.51,
600
+ "learning_rate": 0.0009184780666487984,
601
+ "loss": 7.2695,
602
+ "step": 46500
603
+ },
604
+ {
605
+ "epoch": 0.51,
606
+ "learning_rate": 0.000917361297239346,
607
+ "loss": 7.2679,
608
+ "step": 47000
609
+ },
610
+ {
611
+ "epoch": 0.52,
612
+ "learning_rate": 0.0009162445278298936,
613
+ "loss": 7.2732,
614
+ "step": 47500
615
+ },
616
+ {
617
+ "epoch": 0.52,
618
+ "learning_rate": 0.0009151277584204414,
619
+ "loss": 7.2711,
620
+ "step": 48000
621
+ },
622
+ {
623
+ "epoch": 0.53,
624
+ "learning_rate": 0.000914010989010989,
625
+ "loss": 7.2692,
626
+ "step": 48500
627
+ },
628
+ {
629
+ "epoch": 0.54,
630
+ "learning_rate": 0.0009128942196015367,
631
+ "loss": 7.2694,
632
+ "step": 49000
633
+ },
634
+ {
635
+ "epoch": 0.54,
636
+ "learning_rate": 0.0009117774501920844,
637
+ "loss": 7.2745,
638
+ "step": 49500
639
+ },
640
+ {
641
+ "epoch": 0.55,
642
+ "learning_rate": 0.000910660680782632,
643
+ "loss": 7.2687,
644
+ "step": 50000
645
+ },
646
+ {
647
+ "epoch": 0.55,
648
+ "eval_accuracy": 0.046565208288623344,
649
+ "eval_loss": 7.272406101226807,
650
+ "eval_runtime": 992.178,
651
+ "eval_samples_per_second": 310.787,
652
+ "eval_steps_per_second": 4.857,
653
+ "step": 50000
654
+ },
655
+ {
656
+ "epoch": 0.55,
657
+ "learning_rate": 0.0009095439113731797,
658
+ "loss": 7.268,
659
+ "step": 50500
660
+ },
661
+ {
662
+ "epoch": 0.56,
663
+ "learning_rate": 0.0009084271419637273,
664
+ "loss": 7.274,
665
+ "step": 51000
666
+ },
667
+ {
668
+ "epoch": 0.56,
669
+ "learning_rate": 0.000907310372554275,
670
+ "loss": 7.2718,
671
+ "step": 51500
672
+ },
673
+ {
674
+ "epoch": 0.57,
675
+ "learning_rate": 0.0009061936031448227,
676
+ "loss": 7.2724,
677
+ "step": 52000
678
+ },
679
+ {
680
+ "epoch": 0.57,
681
+ "learning_rate": 0.0009050768337353703,
682
+ "loss": 7.273,
683
+ "step": 52500
684
+ },
685
+ {
686
+ "epoch": 0.58,
687
+ "learning_rate": 0.0009039600643259181,
688
+ "loss": 7.2675,
689
+ "step": 53000
690
+ },
691
+ {
692
+ "epoch": 0.58,
693
+ "learning_rate": 0.0009028432949164657,
694
+ "loss": 7.2691,
695
+ "step": 53500
696
+ },
697
+ {
698
+ "epoch": 0.59,
699
+ "learning_rate": 0.0009017265255070133,
700
+ "loss": 7.2652,
701
+ "step": 54000
702
+ },
703
+ {
704
+ "epoch": 0.6,
705
+ "learning_rate": 0.000900609756097561,
706
+ "loss": 7.2744,
707
+ "step": 54500
708
+ },
709
+ {
710
+ "epoch": 0.6,
711
+ "learning_rate": 0.0008994929866881086,
712
+ "loss": 7.2669,
713
+ "step": 55000
714
+ },
715
+ {
716
+ "epoch": 0.61,
717
+ "learning_rate": 0.0008983762172786563,
718
+ "loss": 7.2715,
719
+ "step": 55500
720
+ },
721
+ {
722
+ "epoch": 0.61,
723
+ "learning_rate": 0.000897259447869204,
724
+ "loss": 7.2717,
725
+ "step": 56000
726
+ },
727
+ {
728
+ "epoch": 0.62,
729
+ "learning_rate": 0.0008961426784597516,
730
+ "loss": 7.2678,
731
+ "step": 56500
732
+ },
733
+ {
734
+ "epoch": 0.62,
735
+ "learning_rate": 0.0008950259090502994,
736
+ "loss": 7.2688,
737
+ "step": 57000
738
+ },
739
+ {
740
+ "epoch": 0.63,
741
+ "learning_rate": 0.000893909139640847,
742
+ "loss": 7.2719,
743
+ "step": 57500
744
+ },
745
+ {
746
+ "epoch": 0.63,
747
+ "learning_rate": 0.0008927923702313946,
748
+ "loss": 7.2616,
749
+ "step": 58000
750
+ },
751
+ {
752
+ "epoch": 0.64,
753
+ "learning_rate": 0.0008916756008219423,
754
+ "loss": 7.2691,
755
+ "step": 58500
756
+ },
757
+ {
758
+ "epoch": 0.64,
759
+ "learning_rate": 0.0008905588314124899,
760
+ "loss": 7.2723,
761
+ "step": 59000
762
+ },
763
+ {
764
+ "epoch": 0.65,
765
+ "learning_rate": 0.0008894420620030376,
766
+ "loss": 7.2684,
767
+ "step": 59500
768
+ },
769
+ {
770
+ "epoch": 0.66,
771
+ "learning_rate": 0.0008883252925935854,
772
+ "loss": 7.2724,
773
+ "step": 60000
774
+ },
775
+ {
776
+ "epoch": 0.66,
777
+ "eval_accuracy": 0.046516656493005064,
778
+ "eval_loss": 7.2715935707092285,
779
+ "eval_runtime": 991.3468,
780
+ "eval_samples_per_second": 311.048,
781
+ "eval_steps_per_second": 4.861,
782
+ "step": 60000
783
+ },
784
+ {
785
+ "epoch": 0.66,
786
+ "learning_rate": 0.000887208523184133,
787
+ "loss": 7.271,
788
+ "step": 60500
789
+ },
790
+ {
791
+ "epoch": 0.67,
792
+ "learning_rate": 0.0008860917537746807,
793
+ "loss": 7.2656,
794
+ "step": 61000
795
+ },
796
+ {
797
+ "epoch": 0.67,
798
+ "learning_rate": 0.0008849749843652283,
799
+ "loss": 7.2751,
800
+ "step": 61500
801
+ },
802
+ {
803
+ "epoch": 0.68,
804
+ "learning_rate": 0.0008838582149557759,
805
+ "loss": 7.2697,
806
+ "step": 62000
807
+ },
808
+ {
809
+ "epoch": 0.68,
810
+ "learning_rate": 0.0008827414455463236,
811
+ "loss": 7.2703,
812
+ "step": 62500
813
+ },
814
+ {
815
+ "epoch": 0.69,
816
+ "learning_rate": 0.0008816246761368712,
817
+ "loss": 7.2659,
818
+ "step": 63000
819
+ },
820
+ {
821
+ "epoch": 0.69,
822
+ "learning_rate": 0.0008805079067274189,
823
+ "loss": 7.2614,
824
+ "step": 63500
825
+ },
826
+ {
827
+ "epoch": 0.7,
828
+ "learning_rate": 0.0008793911373179666,
829
+ "loss": 7.2706,
830
+ "step": 64000
831
+ },
832
+ {
833
+ "epoch": 0.7,
834
+ "learning_rate": 0.0008782743679085143,
835
+ "loss": 7.2721,
836
+ "step": 64500
837
+ },
838
+ {
839
+ "epoch": 0.71,
840
+ "learning_rate": 0.000877157598499062,
841
+ "loss": 7.2711,
842
+ "step": 65000
843
+ },
844
+ {
845
+ "epoch": 0.72,
846
+ "learning_rate": 0.0008760408290896096,
847
+ "loss": 7.2657,
848
+ "step": 65500
849
+ },
850
+ {
851
+ "epoch": 0.72,
852
+ "learning_rate": 0.0008749240596801573,
853
+ "loss": 7.2723,
854
+ "step": 66000
855
+ },
856
+ {
857
+ "epoch": 0.73,
858
+ "learning_rate": 0.0008738072902707049,
859
+ "loss": 7.2681,
860
+ "step": 66500
861
+ },
862
+ {
863
+ "epoch": 0.73,
864
+ "learning_rate": 0.0008726905208612525,
865
+ "loss": 7.2671,
866
+ "step": 67000
867
+ },
868
+ {
869
+ "epoch": 0.74,
870
+ "learning_rate": 0.0008715737514518003,
871
+ "loss": 7.2678,
872
+ "step": 67500
873
+ },
874
+ {
875
+ "epoch": 0.74,
876
+ "learning_rate": 0.0008704569820423479,
877
+ "loss": 7.272,
878
+ "step": 68000
879
+ },
880
+ {
881
+ "epoch": 0.75,
882
+ "learning_rate": 0.0008693402126328956,
883
+ "loss": 7.2669,
884
+ "step": 68500
885
+ },
886
+ {
887
+ "epoch": 0.75,
888
+ "learning_rate": 0.0008682234432234433,
889
+ "loss": 7.2658,
890
+ "step": 69000
891
+ },
892
+ {
893
+ "epoch": 0.76,
894
+ "learning_rate": 0.0008671066738139909,
895
+ "loss": 7.2678,
896
+ "step": 69500
897
+ },
898
+ {
899
+ "epoch": 0.76,
900
+ "learning_rate": 0.0008659899044045386,
901
+ "loss": 7.2689,
902
+ "step": 70000
903
+ },
904
+ {
905
+ "epoch": 0.76,
906
+ "eval_accuracy": 0.04650451752608111,
907
+ "eval_loss": 7.268852233886719,
908
+ "eval_runtime": 993.0576,
909
+ "eval_samples_per_second": 310.512,
910
+ "eval_steps_per_second": 4.853,
911
+ "step": 70000
912
+ },
913
+ {
914
+ "epoch": 0.77,
915
+ "learning_rate": 0.0008648731349950862,
916
+ "loss": 7.2695,
917
+ "step": 70500
918
+ },
919
+ {
920
+ "epoch": 0.78,
921
+ "learning_rate": 0.0008637563655856338,
922
+ "loss": 7.2686,
923
+ "step": 71000
924
+ },
925
+ {
926
+ "epoch": 0.78,
927
+ "learning_rate": 0.0008626395961761816,
928
+ "loss": 7.2716,
929
+ "step": 71500
930
+ },
931
+ {
932
+ "epoch": 0.79,
933
+ "learning_rate": 0.0008615228267667292,
934
+ "loss": 7.2724,
935
+ "step": 72000
936
+ },
937
+ {
938
+ "epoch": 0.79,
939
+ "learning_rate": 0.0008604060573572769,
940
+ "loss": 7.2753,
941
+ "step": 72500
942
+ },
943
+ {
944
+ "epoch": 0.8,
945
+ "learning_rate": 0.0008592892879478246,
946
+ "loss": 7.2667,
947
+ "step": 73000
948
+ },
949
+ {
950
+ "epoch": 0.8,
951
+ "learning_rate": 0.0008581725185383722,
952
+ "loss": 7.2694,
953
+ "step": 73500
954
+ },
955
+ {
956
+ "epoch": 0.81,
957
+ "learning_rate": 0.0008570557491289199,
958
+ "loss": 7.2706,
959
+ "step": 74000
960
+ },
961
+ {
962
+ "epoch": 0.81,
963
+ "learning_rate": 0.0008559389797194675,
964
+ "loss": 7.2721,
965
+ "step": 74500
966
+ },
967
+ {
968
+ "epoch": 0.82,
969
+ "learning_rate": 0.0008548222103100152,
970
+ "loss": 7.2719,
971
+ "step": 75000
972
+ },
973
+ {
974
+ "epoch": 0.82,
975
+ "learning_rate": 0.0008537054409005629,
976
+ "loss": 7.277,
977
+ "step": 75500
978
+ },
979
+ {
980
+ "epoch": 0.83,
981
+ "learning_rate": 0.0008525886714911105,
982
+ "loss": 7.2801,
983
+ "step": 76000
984
+ },
985
+ {
986
+ "epoch": 0.84,
987
+ "learning_rate": 0.0008514719020816582,
988
+ "loss": 7.2742,
989
+ "step": 76500
990
+ },
991
+ {
992
+ "epoch": 0.84,
993
+ "learning_rate": 0.0008503551326722059,
994
+ "loss": 7.281,
995
+ "step": 77000
996
+ },
997
+ {
998
+ "epoch": 0.85,
999
+ "learning_rate": 0.0008492383632627535,
1000
+ "loss": 7.2747,
1001
+ "step": 77500
1002
+ },
1003
+ {
1004
+ "epoch": 0.85,
1005
+ "learning_rate": 0.0008481215938533012,
1006
+ "loss": 7.2713,
1007
+ "step": 78000
1008
+ },
1009
+ {
1010
+ "epoch": 0.86,
1011
+ "learning_rate": 0.0008470048244438489,
1012
+ "loss": 7.2747,
1013
+ "step": 78500
1014
+ },
1015
+ {
1016
+ "epoch": 0.86,
1017
+ "learning_rate": 0.0008458880550343965,
1018
+ "loss": 7.2748,
1019
+ "step": 79000
1020
+ },
1021
+ {
1022
+ "epoch": 0.87,
1023
+ "learning_rate": 0.0008447712856249442,
1024
+ "loss": 7.2721,
1025
+ "step": 79500
1026
+ },
1027
+ {
1028
+ "epoch": 0.87,
1029
+ "learning_rate": 0.0008436545162154918,
1030
+ "loss": 7.268,
1031
+ "step": 80000
1032
+ },
1033
+ {
1034
+ "epoch": 0.87,
1035
+ "eval_accuracy": 0.04651641092958247,
1036
+ "eval_loss": 7.2707648277282715,
1037
+ "eval_runtime": 991.4441,
1038
+ "eval_samples_per_second": 311.017,
1039
+ "eval_steps_per_second": 4.861,
1040
+ "step": 80000
1041
+ },
1042
+ {
1043
+ "epoch": 0.88,
1044
+ "learning_rate": 0.0008425377468060394,
1045
+ "loss": 7.2701,
1046
+ "step": 80500
1047
+ },
1048
+ {
1049
+ "epoch": 0.88,
1050
+ "learning_rate": 0.0008414209773965872,
1051
+ "loss": 7.2704,
1052
+ "step": 81000
1053
+ },
1054
+ {
1055
+ "epoch": 0.89,
1056
+ "learning_rate": 0.0008403042079871348,
1057
+ "loss": 7.2713,
1058
+ "step": 81500
1059
+ },
1060
+ {
1061
+ "epoch": 0.9,
1062
+ "learning_rate": 0.0008391874385776826,
1063
+ "loss": 7.2677,
1064
+ "step": 82000
1065
+ },
1066
+ {
1067
+ "epoch": 0.9,
1068
+ "learning_rate": 0.0008380706691682302,
1069
+ "loss": 7.2716,
1070
+ "step": 82500
1071
+ },
1072
+ {
1073
+ "epoch": 0.91,
1074
+ "learning_rate": 0.0008369538997587778,
1075
+ "loss": 7.2706,
1076
+ "step": 83000
1077
+ },
1078
+ {
1079
+ "epoch": 0.91,
1080
+ "learning_rate": 0.0008358371303493255,
1081
+ "loss": 7.2722,
1082
+ "step": 83500
1083
+ },
1084
+ {
1085
+ "epoch": 0.92,
1086
+ "learning_rate": 0.0008347203609398731,
1087
+ "loss": 7.2674,
1088
+ "step": 84000
1089
+ },
1090
+ {
1091
+ "epoch": 0.92,
1092
+ "learning_rate": 0.0008336035915304208,
1093
+ "loss": 7.2693,
1094
+ "step": 84500
1095
+ },
1096
+ {
1097
+ "epoch": 0.93,
1098
+ "learning_rate": 0.0008324868221209685,
1099
+ "loss": 7.2695,
1100
+ "step": 85000
1101
+ },
1102
+ {
1103
+ "epoch": 0.93,
1104
+ "learning_rate": 0.0008313700527115161,
1105
+ "loss": 7.2722,
1106
+ "step": 85500
1107
+ },
1108
+ {
1109
+ "epoch": 0.94,
1110
+ "learning_rate": 0.0008302532833020639,
1111
+ "loss": 7.2781,
1112
+ "step": 86000
1113
+ },
1114
+ {
1115
+ "epoch": 0.94,
1116
+ "learning_rate": 0.0008291365138926115,
1117
+ "loss": 7.2645,
1118
+ "step": 86500
1119
+ },
1120
+ {
1121
+ "epoch": 0.95,
1122
+ "learning_rate": 0.0008280197444831591,
1123
+ "loss": 7.2717,
1124
+ "step": 87000
1125
+ },
1126
+ {
1127
+ "epoch": 0.96,
1128
+ "learning_rate": 0.0008269029750737068,
1129
+ "loss": 7.2686,
1130
+ "step": 87500
1131
+ },
1132
+ {
1133
+ "epoch": 0.96,
1134
+ "learning_rate": 0.0008257862056642544,
1135
+ "loss": 7.2691,
1136
+ "step": 88000
1137
+ },
1138
+ {
1139
+ "epoch": 0.97,
1140
+ "learning_rate": 0.0008246694362548021,
1141
+ "loss": 7.2677,
1142
+ "step": 88500
1143
+ },
1144
+ {
1145
+ "epoch": 0.97,
1146
+ "learning_rate": 0.0008235526668453498,
1147
+ "loss": 7.2691,
1148
+ "step": 89000
1149
+ },
1150
+ {
1151
+ "epoch": 0.98,
1152
+ "learning_rate": 0.0008224358974358975,
1153
+ "loss": 7.2678,
1154
+ "step": 89500
1155
+ },
1156
+ {
1157
+ "epoch": 0.98,
1158
+ "learning_rate": 0.0008213191280264452,
1159
+ "loss": 7.2723,
1160
+ "step": 90000
1161
+ },
1162
+ {
1163
+ "epoch": 0.98,
1164
+ "eval_accuracy": 0.04652993661252105,
1165
+ "eval_loss": 7.271114349365234,
1166
+ "eval_runtime": 991.9947,
1167
+ "eval_samples_per_second": 310.844,
1168
+ "eval_steps_per_second": 4.858,
1169
+ "step": 90000
1170
+ },
1171
+ {
1172
+ "epoch": 0.99,
1173
+ "learning_rate": 0.0008202023586169928,
1174
+ "loss": 7.2688,
1175
+ "step": 90500
1176
+ },
1177
+ {
1178
+ "epoch": 0.99,
1179
+ "learning_rate": 0.0008190855892075404,
1180
+ "loss": 7.2687,
1181
+ "step": 91000
1182
+ },
1183
+ {
1184
+ "epoch": 1.0,
1185
+ "learning_rate": 0.0008179688197980881,
1186
+ "loss": 7.271,
1187
+ "step": 91500
1188
+ },
1189
+ {
1190
+ "epoch": 1.0,
1191
+ "learning_rate": 0.0008168520503886357,
1192
+ "loss": 7.2669,
1193
+ "step": 92000
1194
+ },
1195
+ {
1196
+ "epoch": 1.01,
1197
+ "learning_rate": 0.0008157352809791834,
1198
+ "loss": 7.2686,
1199
+ "step": 92500
1200
+ },
1201
+ {
1202
+ "epoch": 1.02,
1203
+ "learning_rate": 0.0008146185115697312,
1204
+ "loss": 7.2629,
1205
+ "step": 93000
1206
+ },
1207
+ {
1208
+ "epoch": 1.02,
1209
+ "learning_rate": 0.0008135017421602788,
1210
+ "loss": 7.2704,
1211
+ "step": 93500
1212
+ },
1213
+ {
1214
+ "epoch": 1.03,
1215
+ "learning_rate": 0.0008123849727508265,
1216
+ "loss": 7.2709,
1217
+ "step": 94000
1218
+ },
1219
+ {
1220
+ "epoch": 1.03,
1221
+ "learning_rate": 0.0008112682033413741,
1222
+ "loss": 7.2705,
1223
+ "step": 94500
1224
+ },
1225
+ {
1226
+ "epoch": 1.04,
1227
+ "learning_rate": 0.0008101514339319217,
1228
+ "loss": 7.2736,
1229
+ "step": 95000
1230
+ },
1231
+ {
1232
+ "epoch": 1.04,
1233
+ "learning_rate": 0.0008090346645224694,
1234
+ "loss": 7.2728,
1235
+ "step": 95500
1236
+ },
1237
+ {
1238
+ "epoch": 1.05,
1239
+ "learning_rate": 0.000807917895113017,
1240
+ "loss": 7.2721,
1241
+ "step": 96000
1242
+ },
1243
+ {
1244
+ "epoch": 1.05,
1245
+ "learning_rate": 0.0008068011257035647,
1246
+ "loss": 7.2712,
1247
+ "step": 96500
1248
+ },
1249
+ {
1250
+ "epoch": 1.06,
1251
+ "learning_rate": 0.0008056843562941124,
1252
+ "loss": 7.2714,
1253
+ "step": 97000
1254
+ },
1255
+ {
1256
+ "epoch": 1.07,
1257
+ "learning_rate": 0.0008045675868846601,
1258
+ "loss": 7.2678,
1259
+ "step": 97500
1260
+ },
1261
+ {
1262
+ "epoch": 1.07,
1263
+ "learning_rate": 0.0008034508174752078,
1264
+ "loss": 7.2737,
1265
+ "step": 98000
1266
+ },
1267
+ {
1268
+ "epoch": 1.08,
1269
+ "learning_rate": 0.0008023340480657554,
1270
+ "loss": 7.2713,
1271
+ "step": 98500
1272
+ },
1273
+ {
1274
+ "epoch": 1.08,
1275
+ "learning_rate": 0.0008012172786563031,
1276
+ "loss": 7.2655,
1277
+ "step": 99000
1278
+ },
1279
+ {
1280
+ "epoch": 1.09,
1281
+ "learning_rate": 0.0008001005092468507,
1282
+ "loss": 7.2703,
1283
+ "step": 99500
1284
+ },
1285
+ {
1286
+ "epoch": 1.09,
1287
+ "learning_rate": 0.0007989837398373983,
1288
+ "loss": 7.2724,
1289
+ "step": 100000
1290
+ },
1291
+ {
1292
+ "epoch": 1.09,
1293
+ "eval_accuracy": 0.0428689001334134,
1294
+ "eval_loss": 7.271392345428467,
1295
+ "eval_runtime": 990.7575,
1296
+ "eval_samples_per_second": 311.233,
1297
+ "eval_steps_per_second": 4.864,
1298
+ "step": 100000
1299
+ },
1300
+ {
1301
+ "epoch": 1.1,
1302
+ "learning_rate": 0.0007978669704279461,
1303
+ "loss": 7.2732,
1304
+ "step": 100500
1305
+ },
1306
+ {
1307
+ "epoch": 1.1,
1308
+ "learning_rate": 0.0007967502010184937,
1309
+ "loss": 7.2711,
1310
+ "step": 101000
1311
+ },
1312
+ {
1313
+ "epoch": 1.11,
1314
+ "learning_rate": 0.0007956334316090414,
1315
+ "loss": 7.2687,
1316
+ "step": 101500
1317
+ },
1318
+ {
1319
+ "epoch": 1.11,
1320
+ "learning_rate": 0.0007945166621995891,
1321
+ "loss": 7.2714,
1322
+ "step": 102000
1323
+ },
1324
+ {
1325
+ "epoch": 1.12,
1326
+ "learning_rate": 0.0007933998927901367,
1327
+ "loss": 7.2657,
1328
+ "step": 102500
1329
+ },
1330
+ {
1331
+ "epoch": 1.13,
1332
+ "learning_rate": 0.0007922831233806844,
1333
+ "loss": 7.2675,
1334
+ "step": 103000
1335
+ },
1336
+ {
1337
+ "epoch": 1.13,
1338
+ "learning_rate": 0.000791166353971232,
1339
+ "loss": 7.2639,
1340
+ "step": 103500
1341
+ },
1342
+ {
1343
+ "epoch": 1.14,
1344
+ "learning_rate": 0.0007900495845617796,
1345
+ "loss": 7.2741,
1346
+ "step": 104000
1347
+ },
1348
+ {
1349
+ "epoch": 1.14,
1350
+ "learning_rate": 0.0007889328151523274,
1351
+ "loss": 7.2683,
1352
+ "step": 104500
1353
+ },
1354
+ {
1355
+ "epoch": 1.15,
1356
+ "learning_rate": 0.000787816045742875,
1357
+ "loss": 7.2681,
1358
+ "step": 105000
1359
+ },
1360
+ {
1361
+ "epoch": 1.15,
1362
+ "learning_rate": 0.0007866992763334227,
1363
+ "loss": 7.2737,
1364
+ "step": 105500
1365
+ },
1366
+ {
1367
+ "epoch": 1.16,
1368
+ "learning_rate": 0.0007855825069239704,
1369
+ "loss": 7.2732,
1370
+ "step": 106000
1371
+ },
1372
+ {
1373
+ "epoch": 1.16,
1374
+ "learning_rate": 0.000784465737514518,
1375
+ "loss": 7.2683,
1376
+ "step": 106500
1377
+ },
1378
+ {
1379
+ "epoch": 1.17,
1380
+ "learning_rate": 0.0007833489681050657,
1381
+ "loss": 7.2726,
1382
+ "step": 107000
1383
+ },
1384
+ {
1385
+ "epoch": 1.17,
1386
+ "learning_rate": 0.0007822321986956133,
1387
+ "loss": 7.2707,
1388
+ "step": 107500
1389
+ },
1390
+ {
1391
+ "epoch": 1.18,
1392
+ "learning_rate": 0.000781115429286161,
1393
+ "loss": 7.2739,
1394
+ "step": 108000
1395
+ },
1396
+ {
1397
+ "epoch": 1.19,
1398
+ "learning_rate": 0.0007799986598767087,
1399
+ "loss": 7.2704,
1400
+ "step": 108500
1401
+ },
1402
+ {
1403
+ "epoch": 1.19,
1404
+ "learning_rate": 0.0007788818904672563,
1405
+ "loss": 7.2661,
1406
+ "step": 109000
1407
+ },
1408
+ {
1409
+ "epoch": 1.2,
1410
+ "learning_rate": 0.0007777651210578039,
1411
+ "loss": 7.2707,
1412
+ "step": 109500
1413
+ },
1414
+ {
1415
+ "epoch": 1.2,
1416
+ "learning_rate": 0.0007766483516483517,
1417
+ "loss": 7.2761,
1418
+ "step": 110000
1419
+ },
1420
+ {
1421
+ "epoch": 1.2,
1422
+ "eval_accuracy": 0.046532255715385766,
1423
+ "eval_loss": 7.2722978591918945,
1424
+ "eval_runtime": 989.2994,
1425
+ "eval_samples_per_second": 311.691,
1426
+ "eval_steps_per_second": 4.871,
1427
+ "step": 110000
1428
+ },
1429
+ {
1430
+ "epoch": 1.21,
1431
+ "learning_rate": 0.0007755315822388993,
1432
+ "loss": 7.2708,
1433
+ "step": 110500
1434
+ },
1435
+ {
1436
+ "epoch": 1.21,
1437
+ "learning_rate": 0.000774414812829447,
1438
+ "loss": 7.2677,
1439
+ "step": 111000
1440
+ },
1441
+ {
1442
+ "epoch": 1.22,
1443
+ "learning_rate": 0.0007732980434199947,
1444
+ "loss": 7.275,
1445
+ "step": 111500
1446
+ },
1447
+ {
1448
+ "epoch": 1.22,
1449
+ "learning_rate": 0.0007721812740105423,
1450
+ "loss": 7.2739,
1451
+ "step": 112000
1452
+ },
1453
+ {
1454
+ "epoch": 1.23,
1455
+ "learning_rate": 0.00077106450460109,
1456
+ "loss": 7.2744,
1457
+ "step": 112500
1458
+ },
1459
+ {
1460
+ "epoch": 1.23,
1461
+ "learning_rate": 0.0007699477351916376,
1462
+ "loss": 7.2794,
1463
+ "step": 113000
1464
+ },
1465
+ {
1466
+ "epoch": 1.24,
1467
+ "learning_rate": 0.0007688309657821853,
1468
+ "loss": 7.2673,
1469
+ "step": 113500
1470
+ },
1471
+ {
1472
+ "epoch": 1.25,
1473
+ "learning_rate": 0.000767714196372733,
1474
+ "loss": 7.2725,
1475
+ "step": 114000
1476
+ },
1477
+ {
1478
+ "epoch": 1.25,
1479
+ "learning_rate": 0.0007665974269632806,
1480
+ "loss": 7.2741,
1481
+ "step": 114500
1482
+ },
1483
+ {
1484
+ "epoch": 1.26,
1485
+ "learning_rate": 0.0007654806575538284,
1486
+ "loss": 7.2688,
1487
+ "step": 115000
1488
+ },
1489
+ {
1490
+ "epoch": 1.26,
1491
+ "learning_rate": 0.000764363888144376,
1492
+ "loss": 7.2746,
1493
+ "step": 115500
1494
+ },
1495
+ {
1496
+ "epoch": 1.27,
1497
+ "learning_rate": 0.0007632471187349236,
1498
+ "loss": 7.2713,
1499
+ "step": 116000
1500
+ },
1501
+ {
1502
+ "epoch": 1.27,
1503
+ "learning_rate": 0.0007621303493254713,
1504
+ "loss": 7.2662,
1505
+ "step": 116500
1506
+ },
1507
+ {
1508
+ "epoch": 1.28,
1509
+ "learning_rate": 0.0007610135799160189,
1510
+ "loss": 7.2707,
1511
+ "step": 117000
1512
+ },
1513
+ {
1514
+ "epoch": 1.28,
1515
+ "learning_rate": 0.0007598968105065666,
1516
+ "loss": 7.2679,
1517
+ "step": 117500
1518
+ },
1519
+ {
1520
+ "epoch": 1.29,
1521
+ "learning_rate": 0.0007587800410971143,
1522
+ "loss": 7.2716,
1523
+ "step": 118000
1524
+ },
1525
+ {
1526
+ "epoch": 1.29,
1527
+ "learning_rate": 0.000757663271687662,
1528
+ "loss": 7.2707,
1529
+ "step": 118500
1530
+ },
1531
+ {
1532
+ "epoch": 1.3,
1533
+ "learning_rate": 0.0007565465022782097,
1534
+ "loss": 7.2641,
1535
+ "step": 119000
1536
+ },
1537
+ {
1538
+ "epoch": 1.31,
1539
+ "learning_rate": 0.0007554297328687573,
1540
+ "loss": 7.2714,
1541
+ "step": 119500
1542
+ },
1543
+ {
1544
+ "epoch": 1.31,
1545
+ "learning_rate": 0.0007543129634593049,
1546
+ "loss": 7.2694,
1547
+ "step": 120000
1548
+ },
1549
+ {
1550
+ "epoch": 1.31,
1551
+ "eval_accuracy": 0.0465121739230196,
1552
+ "eval_loss": 7.268530368804932,
1553
+ "eval_runtime": 988.7443,
1554
+ "eval_samples_per_second": 311.866,
1555
+ "eval_steps_per_second": 4.874,
1556
+ "step": 120000
1557
+ },
1558
+ {
1559
+ "epoch": 1.32,
1560
+ "learning_rate": 0.0007531961940498526,
1561
+ "loss": 7.2689,
1562
+ "step": 120500
1563
+ },
1564
+ {
1565
+ "epoch": 1.32,
1566
+ "learning_rate": 0.0007520794246404002,
1567
+ "loss": 7.2674,
1568
+ "step": 121000
1569
+ },
1570
+ {
1571
+ "epoch": 1.33,
1572
+ "learning_rate": 0.0007509626552309479,
1573
+ "loss": 7.2714,
1574
+ "step": 121500
1575
+ },
1576
+ {
1577
+ "epoch": 1.33,
1578
+ "learning_rate": 0.0007498458858214957,
1579
+ "loss": 7.2685,
1580
+ "step": 122000
1581
+ },
1582
+ {
1583
+ "epoch": 1.34,
1584
+ "learning_rate": 0.0007487291164120433,
1585
+ "loss": 7.2678,
1586
+ "step": 122500
1587
+ },
1588
+ {
1589
+ "epoch": 1.34,
1590
+ "learning_rate": 0.000747612347002591,
1591
+ "loss": 7.2721,
1592
+ "step": 123000
1593
+ },
1594
+ {
1595
+ "epoch": 1.35,
1596
+ "learning_rate": 0.0007464955775931386,
1597
+ "loss": 7.2701,
1598
+ "step": 123500
1599
+ },
1600
+ {
1601
+ "epoch": 1.35,
1602
+ "learning_rate": 0.0007453788081836862,
1603
+ "loss": 7.2776,
1604
+ "step": 124000
1605
+ },
1606
+ {
1607
+ "epoch": 1.36,
1608
+ "learning_rate": 0.0007442620387742339,
1609
+ "loss": 7.2668,
1610
+ "step": 124500
1611
+ },
1612
+ {
1613
+ "epoch": 1.37,
1614
+ "learning_rate": 0.0007431452693647815,
1615
+ "loss": 7.2711,
1616
+ "step": 125000
1617
+ },
1618
+ {
1619
+ "epoch": 1.37,
1620
+ "learning_rate": 0.0007420284999553292,
1621
+ "loss": 7.2632,
1622
+ "step": 125500
1623
+ },
1624
+ {
1625
+ "epoch": 1.38,
1626
+ "learning_rate": 0.0007409117305458769,
1627
+ "loss": 7.2672,
1628
+ "step": 126000
1629
+ },
1630
+ {
1631
+ "epoch": 1.38,
1632
+ "learning_rate": 0.0007397949611364246,
1633
+ "loss": 7.2721,
1634
+ "step": 126500
1635
+ },
1636
+ {
1637
+ "epoch": 1.39,
1638
+ "learning_rate": 0.0007386781917269723,
1639
+ "loss": 7.2731,
1640
+ "step": 127000
1641
+ },
1642
+ {
1643
+ "epoch": 1.39,
1644
+ "learning_rate": 0.0007375614223175199,
1645
+ "loss": 7.2695,
1646
+ "step": 127500
1647
+ },
1648
+ {
1649
+ "epoch": 1.4,
1650
+ "learning_rate": 0.0007364446529080675,
1651
+ "loss": 7.2679,
1652
+ "step": 128000
1653
+ },
1654
+ {
1655
+ "epoch": 1.4,
1656
+ "learning_rate": 0.0007353278834986152,
1657
+ "loss": 7.2727,
1658
+ "step": 128500
1659
+ },
1660
+ {
1661
+ "epoch": 1.41,
1662
+ "learning_rate": 0.0007342111140891628,
1663
+ "loss": 7.2695,
1664
+ "step": 129000
1665
+ },
1666
+ {
1667
+ "epoch": 1.41,
1668
+ "learning_rate": 0.0007330943446797106,
1669
+ "loss": 7.2704,
1670
+ "step": 129500
1671
+ },
1672
+ {
1673
+ "epoch": 1.42,
1674
+ "learning_rate": 0.0007319775752702582,
1675
+ "loss": 7.2671,
1676
+ "step": 130000
1677
+ },
1678
+ {
1679
+ "epoch": 1.42,
1680
+ "eval_accuracy": 0.04657166881035385,
1681
+ "eval_loss": 7.272782325744629,
1682
+ "eval_runtime": 985.7841,
1683
+ "eval_samples_per_second": 312.803,
1684
+ "eval_steps_per_second": 4.888,
1685
+ "step": 130000
1686
+ },
1687
+ {
1688
+ "epoch": 1.43,
1689
+ "learning_rate": 0.0007308608058608059,
1690
+ "loss": 7.2722,
1691
+ "step": 130500
1692
+ },
1693
+ {
1694
+ "epoch": 1.43,
1695
+ "learning_rate": 0.0007297440364513536,
1696
+ "loss": 7.2669,
1697
+ "step": 131000
1698
+ },
1699
+ {
1700
+ "epoch": 1.44,
1701
+ "learning_rate": 0.0007286272670419012,
1702
+ "loss": 7.2683,
1703
+ "step": 131500
1704
+ },
1705
+ {
1706
+ "epoch": 1.44,
1707
+ "learning_rate": 0.0007275104976324489,
1708
+ "loss": 7.2686,
1709
+ "step": 132000
1710
+ },
1711
+ {
1712
+ "epoch": 1.45,
1713
+ "learning_rate": 0.0007263937282229965,
1714
+ "loss": 7.2708,
1715
+ "step": 132500
1716
+ },
1717
+ {
1718
+ "epoch": 1.45,
1719
+ "learning_rate": 0.0007252769588135441,
1720
+ "loss": 7.2689,
1721
+ "step": 133000
1722
+ },
1723
+ {
1724
+ "epoch": 1.46,
1725
+ "learning_rate": 0.0007241601894040919,
1726
+ "loss": 7.2721,
1727
+ "step": 133500
1728
+ },
1729
+ {
1730
+ "epoch": 1.46,
1731
+ "learning_rate": 0.0007230434199946395,
1732
+ "loss": 7.2798,
1733
+ "step": 134000
1734
+ },
1735
+ {
1736
+ "epoch": 1.47,
1737
+ "learning_rate": 0.0007219266505851872,
1738
+ "loss": 7.2693,
1739
+ "step": 134500
1740
+ },
1741
+ {
1742
+ "epoch": 1.47,
1743
+ "learning_rate": 0.0007208098811757349,
1744
+ "loss": 7.278,
1745
+ "step": 135000
1746
+ },
1747
+ {
1748
+ "epoch": 1.48,
1749
+ "learning_rate": 0.0007196931117662825,
1750
+ "loss": 7.2759,
1751
+ "step": 135500
1752
+ },
1753
+ {
1754
+ "epoch": 1.49,
1755
+ "learning_rate": 0.0007185763423568302,
1756
+ "loss": 7.2683,
1757
+ "step": 136000
1758
+ },
1759
+ {
1760
+ "epoch": 1.49,
1761
+ "learning_rate": 0.0007174595729473778,
1762
+ "loss": 7.2707,
1763
+ "step": 136500
1764
+ },
1765
+ {
1766
+ "epoch": 1.5,
1767
+ "learning_rate": 0.0007163428035379255,
1768
+ "loss": 7.2637,
1769
+ "step": 137000
1770
+ },
1771
+ {
1772
+ "epoch": 1.5,
1773
+ "learning_rate": 0.0007152260341284732,
1774
+ "loss": 7.2661,
1775
+ "step": 137500
1776
+ },
1777
+ {
1778
+ "epoch": 1.51,
1779
+ "learning_rate": 0.0007141092647190208,
1780
+ "loss": 7.2709,
1781
+ "step": 138000
1782
+ },
1783
+ {
1784
+ "epoch": 1.51,
1785
+ "learning_rate": 0.0007129924953095684,
1786
+ "loss": 7.2666,
1787
+ "step": 138500
1788
+ },
1789
+ {
1790
+ "epoch": 1.52,
1791
+ "learning_rate": 0.0007118757259001162,
1792
+ "loss": 7.2686,
1793
+ "step": 139000
1794
+ },
1795
+ {
1796
+ "epoch": 1.52,
1797
+ "learning_rate": 0.0007107589564906638,
1798
+ "loss": 7.2699,
1799
+ "step": 139500
1800
+ },
1801
+ {
1802
+ "epoch": 1.53,
1803
+ "learning_rate": 0.0007096421870812115,
1804
+ "loss": 7.2664,
1805
+ "step": 140000
1806
+ },
1807
+ {
1808
+ "epoch": 1.53,
1809
+ "eval_accuracy": 0.04652100591929093,
1810
+ "eval_loss": 7.271435260772705,
1811
+ "eval_runtime": 989.631,
1812
+ "eval_samples_per_second": 311.587,
1813
+ "eval_steps_per_second": 4.869,
1814
+ "step": 140000
1815
+ },
1816
+ {
1817
+ "epoch": 1.53,
1818
+ "learning_rate": 0.0007085254176717592,
1819
+ "loss": 7.2693,
1820
+ "step": 140500
1821
+ },
1822
+ {
1823
+ "epoch": 1.54,
1824
+ "learning_rate": 0.0007074086482623068,
1825
+ "loss": 7.2711,
1826
+ "step": 141000
1827
+ },
1828
+ {
1829
+ "epoch": 1.55,
1830
+ "learning_rate": 0.0007062918788528545,
1831
+ "loss": 7.263,
1832
+ "step": 141500
1833
+ },
1834
+ {
1835
+ "epoch": 1.55,
1836
+ "learning_rate": 0.0007051751094434021,
1837
+ "loss": 7.2718,
1838
+ "step": 142000
1839
+ },
1840
+ {
1841
+ "epoch": 1.56,
1842
+ "learning_rate": 0.0007040583400339497,
1843
+ "loss": 7.2684,
1844
+ "step": 142500
1845
+ },
1846
+ {
1847
+ "epoch": 1.56,
1848
+ "learning_rate": 0.0007029415706244975,
1849
+ "loss": 7.2697,
1850
+ "step": 143000
1851
+ },
1852
+ {
1853
+ "epoch": 1.57,
1854
+ "learning_rate": 0.0007018248012150451,
1855
+ "loss": 7.2702,
1856
+ "step": 143500
1857
+ },
1858
+ {
1859
+ "epoch": 1.57,
1860
+ "learning_rate": 0.0007007080318055929,
1861
+ "loss": 7.2689,
1862
+ "step": 144000
1863
+ },
1864
+ {
1865
+ "epoch": 1.58,
1866
+ "learning_rate": 0.0006995912623961405,
1867
+ "loss": 7.2659,
1868
+ "step": 144500
1869
+ },
1870
+ {
1871
+ "epoch": 1.58,
1872
+ "learning_rate": 0.0006984744929866881,
1873
+ "loss": 7.267,
1874
+ "step": 145000
1875
+ },
1876
+ {
1877
+ "epoch": 1.59,
1878
+ "learning_rate": 0.0006973577235772358,
1879
+ "loss": 7.2731,
1880
+ "step": 145500
1881
+ },
1882
+ {
1883
+ "epoch": 1.59,
1884
+ "learning_rate": 0.0006962409541677834,
1885
+ "loss": 7.2691,
1886
+ "step": 146000
1887
+ },
1888
+ {
1889
+ "epoch": 1.6,
1890
+ "learning_rate": 0.0006951241847583311,
1891
+ "loss": 7.2676,
1892
+ "step": 146500
1893
+ },
1894
+ {
1895
+ "epoch": 1.61,
1896
+ "learning_rate": 0.0006940074153488788,
1897
+ "loss": 7.2709,
1898
+ "step": 147000
1899
+ },
1900
+ {
1901
+ "epoch": 1.61,
1902
+ "learning_rate": 0.0006928906459394264,
1903
+ "loss": 7.2692,
1904
+ "step": 147500
1905
+ },
1906
+ {
1907
+ "epoch": 1.62,
1908
+ "learning_rate": 0.0006917738765299742,
1909
+ "loss": 7.2699,
1910
+ "step": 148000
1911
+ },
1912
+ {
1913
+ "epoch": 1.62,
1914
+ "learning_rate": 0.0006906571071205218,
1915
+ "loss": 7.2725,
1916
+ "step": 148500
1917
+ },
1918
+ {
1919
+ "epoch": 1.63,
1920
+ "learning_rate": 0.0006895403377110694,
1921
+ "loss": 7.2629,
1922
+ "step": 149000
1923
+ },
1924
+ {
1925
+ "epoch": 1.63,
1926
+ "learning_rate": 0.0006884235683016171,
1927
+ "loss": 7.2661,
1928
+ "step": 149500
1929
+ },
1930
+ {
1931
+ "epoch": 1.64,
1932
+ "learning_rate": 0.0006873067988921647,
1933
+ "loss": 7.2671,
1934
+ "step": 150000
1935
+ },
1936
+ {
1937
+ "epoch": 1.64,
1938
+ "eval_accuracy": 0.04653411066611076,
1939
+ "eval_loss": 7.2707414627075195,
1940
+ "eval_runtime": 986.5443,
1941
+ "eval_samples_per_second": 312.562,
1942
+ "eval_steps_per_second": 4.885,
1943
+ "step": 150000
1944
+ },
1945
+ {
1946
+ "epoch": 1.64,
1947
+ "learning_rate": 0.0006861900294827124,
1948
+ "loss": 7.2674,
1949
+ "step": 150500
1950
+ },
1951
+ {
1952
+ "epoch": 1.65,
1953
+ "learning_rate": 0.0006850732600732601,
1954
+ "loss": 7.2674,
1955
+ "step": 151000
1956
+ },
1957
+ {
1958
+ "epoch": 1.65,
1959
+ "learning_rate": 0.0006839564906638078,
1960
+ "loss": 7.2639,
1961
+ "step": 151500
1962
+ },
1963
+ {
1964
+ "epoch": 1.66,
1965
+ "learning_rate": 0.0006828397212543555,
1966
+ "loss": 7.2691,
1967
+ "step": 152000
1968
+ },
1969
+ {
1970
+ "epoch": 1.67,
1971
+ "learning_rate": 0.0006817229518449031,
1972
+ "loss": 7.2676,
1973
+ "step": 152500
1974
+ },
1975
+ {
1976
+ "epoch": 1.67,
1977
+ "learning_rate": 0.0006806061824354507,
1978
+ "loss": 7.2686,
1979
+ "step": 153000
1980
+ },
1981
+ {
1982
+ "epoch": 1.68,
1983
+ "learning_rate": 0.0006794894130259984,
1984
+ "loss": 7.2704,
1985
+ "step": 153500
1986
+ },
1987
+ {
1988
+ "epoch": 1.68,
1989
+ "learning_rate": 0.000678372643616546,
1990
+ "loss": 7.2732,
1991
+ "step": 154000
1992
+ },
1993
+ {
1994
+ "epoch": 1.69,
1995
+ "learning_rate": 0.0006772558742070937,
1996
+ "loss": 7.2691,
1997
+ "step": 154500
1998
+ },
1999
+ {
2000
+ "epoch": 1.69,
2001
+ "learning_rate": 0.0006761391047976413,
2002
+ "loss": 7.2721,
2003
+ "step": 155000
2004
+ },
2005
+ {
2006
+ "epoch": 1.7,
2007
+ "learning_rate": 0.0006750223353881891,
2008
+ "loss": 7.2699,
2009
+ "step": 155500
2010
+ },
2011
+ {
2012
+ "epoch": 1.7,
2013
+ "learning_rate": 0.0006739055659787368,
2014
+ "loss": 7.2705,
2015
+ "step": 156000
2016
+ },
2017
+ {
2018
+ "epoch": 1.71,
2019
+ "learning_rate": 0.0006727887965692844,
2020
+ "loss": 7.2686,
2021
+ "step": 156500
2022
+ },
2023
+ {
2024
+ "epoch": 1.72,
2025
+ "learning_rate": 0.000671672027159832,
2026
+ "loss": 7.2661,
2027
+ "step": 157000
2028
+ },
2029
+ {
2030
+ "epoch": 1.72,
2031
+ "learning_rate": 0.0006705552577503797,
2032
+ "loss": 7.2759,
2033
+ "step": 157500
2034
+ },
2035
+ {
2036
+ "epoch": 1.73,
2037
+ "learning_rate": 0.0006694384883409273,
2038
+ "loss": 7.2702,
2039
+ "step": 158000
2040
+ },
2041
+ {
2042
+ "epoch": 1.73,
2043
+ "learning_rate": 0.000668321718931475,
2044
+ "loss": 7.2657,
2045
+ "step": 158500
2046
+ },
2047
+ {
2048
+ "epoch": 1.74,
2049
+ "learning_rate": 0.0006672049495220227,
2050
+ "loss": 7.2698,
2051
+ "step": 159000
2052
+ },
2053
+ {
2054
+ "epoch": 1.74,
2055
+ "learning_rate": 0.0006660881801125704,
2056
+ "loss": 7.2665,
2057
+ "step": 159500
2058
+ },
2059
+ {
2060
+ "epoch": 1.75,
2061
+ "learning_rate": 0.0006649714107031181,
2062
+ "loss": 7.2663,
2063
+ "step": 160000
2064
+ },
2065
+ {
2066
+ "epoch": 1.75,
2067
+ "eval_accuracy": 0.04653063667672438,
2068
+ "eval_loss": 7.274196147918701,
2069
+ "eval_runtime": 986.618,
2070
+ "eval_samples_per_second": 312.538,
2071
+ "eval_steps_per_second": 4.884,
2072
+ "step": 160000
2073
+ },
2074
+ {
2075
+ "epoch": 1.75,
2076
+ "learning_rate": 0.0006638546412936657,
2077
+ "loss": 7.2691,
2078
+ "step": 160500
2079
+ },
2080
+ {
2081
+ "epoch": 1.76,
2082
+ "learning_rate": 0.0006627378718842134,
2083
+ "loss": 7.2644,
2084
+ "step": 161000
2085
+ },
2086
+ {
2087
+ "epoch": 1.76,
2088
+ "learning_rate": 0.000661621102474761,
2089
+ "loss": 7.2686,
2090
+ "step": 161500
2091
+ },
2092
+ {
2093
+ "epoch": 1.77,
2094
+ "learning_rate": 0.0006605043330653086,
2095
+ "loss": 7.2654,
2096
+ "step": 162000
2097
+ },
2098
+ {
2099
+ "epoch": 1.78,
2100
+ "learning_rate": 0.0006593875636558564,
2101
+ "loss": 7.2704,
2102
+ "step": 162500
2103
+ },
2104
+ {
2105
+ "epoch": 1.78,
2106
+ "learning_rate": 0.000658270794246404,
2107
+ "loss": 7.2666,
2108
+ "step": 163000
2109
+ },
2110
+ {
2111
+ "epoch": 1.79,
2112
+ "learning_rate": 0.0006571540248369517,
2113
+ "loss": 7.2741,
2114
+ "step": 163500
2115
+ },
2116
+ {
2117
+ "epoch": 1.79,
2118
+ "learning_rate": 0.0006560372554274994,
2119
+ "loss": 7.2727,
2120
+ "step": 164000
2121
+ },
2122
+ {
2123
+ "epoch": 1.8,
2124
+ "learning_rate": 0.000654920486018047,
2125
+ "loss": 7.2623,
2126
+ "step": 164500
2127
+ },
2128
+ {
2129
+ "epoch": 1.8,
2130
+ "learning_rate": 0.0006538037166085947,
2131
+ "loss": 7.2654,
2132
+ "step": 165000
2133
+ },
2134
+ {
2135
+ "epoch": 1.81,
2136
+ "learning_rate": 0.0006526869471991423,
2137
+ "loss": 7.2659,
2138
+ "step": 165500
2139
+ },
2140
+ {
2141
+ "epoch": 1.81,
2142
+ "learning_rate": 0.0006515724113285089,
2143
+ "loss": 7.2705,
2144
+ "step": 166000
2145
+ },
2146
+ {
2147
+ "epoch": 1.82,
2148
+ "learning_rate": 0.0006504556419190566,
2149
+ "loss": 7.2702,
2150
+ "step": 166500
2151
+ },
2152
+ {
2153
+ "epoch": 1.82,
2154
+ "learning_rate": 0.0006493388725096042,
2155
+ "loss": 7.2598,
2156
+ "step": 167000
2157
+ },
2158
+ {
2159
+ "epoch": 1.83,
2160
+ "learning_rate": 0.0006482221031001518,
2161
+ "loss": 7.2735,
2162
+ "step": 167500
2163
+ },
2164
+ {
2165
+ "epoch": 1.84,
2166
+ "learning_rate": 0.0006471075672295185,
2167
+ "loss": 7.2693,
2168
+ "step": 168000
2169
+ },
2170
+ {
2171
+ "epoch": 1.84,
2172
+ "learning_rate": 0.0006459907978200661,
2173
+ "loss": 7.2746,
2174
+ "step": 168500
2175
+ },
2176
+ {
2177
+ "epoch": 1.85,
2178
+ "learning_rate": 0.0006448740284106138,
2179
+ "loss": 7.2716,
2180
+ "step": 169000
2181
+ },
2182
+ {
2183
+ "epoch": 1.85,
2184
+ "step": 169136,
2185
+ "total_flos": 3.1691545184745554e+18,
2186
+ "train_loss": 7.273256063168102,
2187
+ "train_runtime": 107974.2539,
2188
+ "train_samples_per_second": 271.303,
2189
+ "train_steps_per_second": 4.239
2190
  }
2191
  ],
2192
  "max_steps": 457720,
2193
  "num_train_epochs": 5,
2194
+ "total_flos": 3.1691545184745554e+18,
2195
  "trial_name": null,
2196
  "trial_params": null
2197
  }