Commit
·
1cc0058
1
Parent(s):
0f50716
End of training
Browse files- all_results.json +13 -0
- eval_results.json +8 -0
- train_results.json +8 -0
- trainer_state.json +553 -0
all_results.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"eval_accuracy": 0.9858,
|
| 4 |
+
"eval_loss": 0.04139312729239464,
|
| 5 |
+
"eval_runtime": 57.2471,
|
| 6 |
+
"eval_samples_per_second": 87.341,
|
| 7 |
+
"eval_steps_per_second": 2.742,
|
| 8 |
+
"total_flos": 3.5202418590178345e+18,
|
| 9 |
+
"train_loss": 0.46742354324910035,
|
| 10 |
+
"train_runtime": 3479.7951,
|
| 11 |
+
"train_samples_per_second": 12.932,
|
| 12 |
+
"train_steps_per_second": 0.101
|
| 13 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"eval_accuracy": 0.9858,
|
| 4 |
+
"eval_loss": 0.04139312729239464,
|
| 5 |
+
"eval_runtime": 57.2471,
|
| 6 |
+
"eval_samples_per_second": 87.341,
|
| 7 |
+
"eval_steps_per_second": 2.742
|
| 8 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"total_flos": 3.5202418590178345e+18,
|
| 4 |
+
"train_loss": 0.46742354324910035,
|
| 5 |
+
"train_runtime": 3479.7951,
|
| 6 |
+
"train_samples_per_second": 12.932,
|
| 7 |
+
"train_steps_per_second": 0.101
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,553 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.997867803837953,
|
| 5 |
+
"eval_steps": 10,
|
| 6 |
+
"global_step": 351,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.03,
|
| 13 |
+
"learning_rate": 1.388888888888889e-05,
|
| 14 |
+
"loss": 2.303,
|
| 15 |
+
"step": 10
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 0.03,
|
| 19 |
+
"eval_accuracy": 0.2334,
|
| 20 |
+
"eval_loss": 2.1672284603118896,
|
| 21 |
+
"eval_runtime": 61.2464,
|
| 22 |
+
"eval_samples_per_second": 81.637,
|
| 23 |
+
"eval_steps_per_second": 2.563,
|
| 24 |
+
"step": 10
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.06,
|
| 28 |
+
"learning_rate": 2.777777777777778e-05,
|
| 29 |
+
"loss": 2.0158,
|
| 30 |
+
"step": 20
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.06,
|
| 34 |
+
"eval_accuracy": 0.657,
|
| 35 |
+
"eval_loss": 1.6672015190124512,
|
| 36 |
+
"eval_runtime": 57.6151,
|
| 37 |
+
"eval_samples_per_second": 86.783,
|
| 38 |
+
"eval_steps_per_second": 2.725,
|
| 39 |
+
"step": 20
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.09,
|
| 43 |
+
"learning_rate": 4.166666666666667e-05,
|
| 44 |
+
"loss": 1.4855,
|
| 45 |
+
"step": 30
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.09,
|
| 49 |
+
"eval_accuracy": 0.8704,
|
| 50 |
+
"eval_loss": 0.8292349576950073,
|
| 51 |
+
"eval_runtime": 58.3323,
|
| 52 |
+
"eval_samples_per_second": 85.716,
|
| 53 |
+
"eval_steps_per_second": 2.691,
|
| 54 |
+
"step": 30
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"epoch": 0.11,
|
| 58 |
+
"learning_rate": 4.936507936507937e-05,
|
| 59 |
+
"loss": 0.7451,
|
| 60 |
+
"step": 40
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"epoch": 0.11,
|
| 64 |
+
"eval_accuracy": 0.93,
|
| 65 |
+
"eval_loss": 0.2578178346157074,
|
| 66 |
+
"eval_runtime": 58.1914,
|
| 67 |
+
"eval_samples_per_second": 85.923,
|
| 68 |
+
"eval_steps_per_second": 2.698,
|
| 69 |
+
"step": 40
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 0.14,
|
| 73 |
+
"learning_rate": 4.7777777777777784e-05,
|
| 74 |
+
"loss": 0.5618,
|
| 75 |
+
"step": 50
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"epoch": 0.14,
|
| 79 |
+
"eval_accuracy": 0.962,
|
| 80 |
+
"eval_loss": 0.14759203791618347,
|
| 81 |
+
"eval_runtime": 58.5488,
|
| 82 |
+
"eval_samples_per_second": 85.399,
|
| 83 |
+
"eval_steps_per_second": 2.682,
|
| 84 |
+
"step": 50
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 0.17,
|
| 88 |
+
"learning_rate": 4.6190476190476194e-05,
|
| 89 |
+
"loss": 0.4545,
|
| 90 |
+
"step": 60
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 0.17,
|
| 94 |
+
"eval_accuracy": 0.9642,
|
| 95 |
+
"eval_loss": 0.12480442970991135,
|
| 96 |
+
"eval_runtime": 61.2292,
|
| 97 |
+
"eval_samples_per_second": 81.66,
|
| 98 |
+
"eval_steps_per_second": 2.564,
|
| 99 |
+
"step": 60
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 0.2,
|
| 103 |
+
"learning_rate": 4.460317460317461e-05,
|
| 104 |
+
"loss": 0.4587,
|
| 105 |
+
"step": 70
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 0.2,
|
| 109 |
+
"eval_accuracy": 0.9748,
|
| 110 |
+
"eval_loss": 0.09407833218574524,
|
| 111 |
+
"eval_runtime": 57.8655,
|
| 112 |
+
"eval_samples_per_second": 86.407,
|
| 113 |
+
"eval_steps_per_second": 2.713,
|
| 114 |
+
"step": 70
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.23,
|
| 118 |
+
"learning_rate": 4.301587301587302e-05,
|
| 119 |
+
"loss": 0.3911,
|
| 120 |
+
"step": 80
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 0.23,
|
| 124 |
+
"eval_accuracy": 0.9712,
|
| 125 |
+
"eval_loss": 0.09437316656112671,
|
| 126 |
+
"eval_runtime": 57.5812,
|
| 127 |
+
"eval_samples_per_second": 86.834,
|
| 128 |
+
"eval_steps_per_second": 2.727,
|
| 129 |
+
"step": 80
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.26,
|
| 133 |
+
"learning_rate": 4.1428571428571437e-05,
|
| 134 |
+
"loss": 0.3839,
|
| 135 |
+
"step": 90
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.26,
|
| 139 |
+
"eval_accuracy": 0.9756,
|
| 140 |
+
"eval_loss": 0.0848420187830925,
|
| 141 |
+
"eval_runtime": 58.1027,
|
| 142 |
+
"eval_samples_per_second": 86.055,
|
| 143 |
+
"eval_steps_per_second": 2.702,
|
| 144 |
+
"step": 90
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 0.28,
|
| 148 |
+
"learning_rate": 3.984126984126984e-05,
|
| 149 |
+
"loss": 0.3864,
|
| 150 |
+
"step": 100
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.28,
|
| 154 |
+
"eval_accuracy": 0.978,
|
| 155 |
+
"eval_loss": 0.0744096040725708,
|
| 156 |
+
"eval_runtime": 57.7886,
|
| 157 |
+
"eval_samples_per_second": 86.522,
|
| 158 |
+
"eval_steps_per_second": 2.717,
|
| 159 |
+
"step": 100
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 0.31,
|
| 163 |
+
"learning_rate": 3.8253968253968256e-05,
|
| 164 |
+
"loss": 0.3141,
|
| 165 |
+
"step": 110
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 0.31,
|
| 169 |
+
"eval_accuracy": 0.98,
|
| 170 |
+
"eval_loss": 0.06731583178043365,
|
| 171 |
+
"eval_runtime": 58.0242,
|
| 172 |
+
"eval_samples_per_second": 86.171,
|
| 173 |
+
"eval_steps_per_second": 2.706,
|
| 174 |
+
"step": 110
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 0.34,
|
| 178 |
+
"learning_rate": 3.6666666666666666e-05,
|
| 179 |
+
"loss": 0.3764,
|
| 180 |
+
"step": 120
|
| 181 |
+
},
|
| 182 |
+
{
|
| 183 |
+
"epoch": 0.34,
|
| 184 |
+
"eval_accuracy": 0.9764,
|
| 185 |
+
"eval_loss": 0.07063312083482742,
|
| 186 |
+
"eval_runtime": 58.0733,
|
| 187 |
+
"eval_samples_per_second": 86.098,
|
| 188 |
+
"eval_steps_per_second": 2.703,
|
| 189 |
+
"step": 120
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 0.37,
|
| 193 |
+
"learning_rate": 3.5079365079365075e-05,
|
| 194 |
+
"loss": 0.3003,
|
| 195 |
+
"step": 130
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 0.37,
|
| 199 |
+
"eval_accuracy": 0.984,
|
| 200 |
+
"eval_loss": 0.05995839089155197,
|
| 201 |
+
"eval_runtime": 58.0208,
|
| 202 |
+
"eval_samples_per_second": 86.176,
|
| 203 |
+
"eval_steps_per_second": 2.706,
|
| 204 |
+
"step": 130
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 0.4,
|
| 208 |
+
"learning_rate": 3.349206349206349e-05,
|
| 209 |
+
"loss": 0.3566,
|
| 210 |
+
"step": 140
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"epoch": 0.4,
|
| 214 |
+
"eval_accuracy": 0.9826,
|
| 215 |
+
"eval_loss": 0.05617852881550789,
|
| 216 |
+
"eval_runtime": 57.9727,
|
| 217 |
+
"eval_samples_per_second": 86.248,
|
| 218 |
+
"eval_steps_per_second": 2.708,
|
| 219 |
+
"step": 140
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.43,
|
| 223 |
+
"learning_rate": 3.19047619047619e-05,
|
| 224 |
+
"loss": 0.2855,
|
| 225 |
+
"step": 150
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 0.43,
|
| 229 |
+
"eval_accuracy": 0.9816,
|
| 230 |
+
"eval_loss": 0.056736305356025696,
|
| 231 |
+
"eval_runtime": 57.7871,
|
| 232 |
+
"eval_samples_per_second": 86.525,
|
| 233 |
+
"eval_steps_per_second": 2.717,
|
| 234 |
+
"step": 150
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.45,
|
| 238 |
+
"learning_rate": 3.0317460317460318e-05,
|
| 239 |
+
"loss": 0.3351,
|
| 240 |
+
"step": 160
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.45,
|
| 244 |
+
"eval_accuracy": 0.9828,
|
| 245 |
+
"eval_loss": 0.0543464832007885,
|
| 246 |
+
"eval_runtime": 58.1354,
|
| 247 |
+
"eval_samples_per_second": 86.006,
|
| 248 |
+
"eval_steps_per_second": 2.701,
|
| 249 |
+
"step": 160
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 0.48,
|
| 253 |
+
"learning_rate": 2.8730158730158728e-05,
|
| 254 |
+
"loss": 0.2977,
|
| 255 |
+
"step": 170
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.48,
|
| 259 |
+
"eval_accuracy": 0.9798,
|
| 260 |
+
"eval_loss": 0.05677906423807144,
|
| 261 |
+
"eval_runtime": 57.6217,
|
| 262 |
+
"eval_samples_per_second": 86.773,
|
| 263 |
+
"eval_steps_per_second": 2.725,
|
| 264 |
+
"step": 170
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 0.51,
|
| 268 |
+
"learning_rate": 2.714285714285714e-05,
|
| 269 |
+
"loss": 0.2924,
|
| 270 |
+
"step": 180
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 0.51,
|
| 274 |
+
"eval_accuracy": 0.9804,
|
| 275 |
+
"eval_loss": 0.05774379149079323,
|
| 276 |
+
"eval_runtime": 58.0352,
|
| 277 |
+
"eval_samples_per_second": 86.155,
|
| 278 |
+
"eval_steps_per_second": 2.705,
|
| 279 |
+
"step": 180
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 0.54,
|
| 283 |
+
"learning_rate": 2.5555555555555554e-05,
|
| 284 |
+
"loss": 0.2884,
|
| 285 |
+
"step": 190
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 0.54,
|
| 289 |
+
"eval_accuracy": 0.983,
|
| 290 |
+
"eval_loss": 0.05509978160262108,
|
| 291 |
+
"eval_runtime": 57.5779,
|
| 292 |
+
"eval_samples_per_second": 86.839,
|
| 293 |
+
"eval_steps_per_second": 2.727,
|
| 294 |
+
"step": 190
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 0.57,
|
| 298 |
+
"learning_rate": 2.396825396825397e-05,
|
| 299 |
+
"loss": 0.3067,
|
| 300 |
+
"step": 200
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"epoch": 0.57,
|
| 304 |
+
"eval_accuracy": 0.983,
|
| 305 |
+
"eval_loss": 0.04865285009145737,
|
| 306 |
+
"eval_runtime": 57.833,
|
| 307 |
+
"eval_samples_per_second": 86.456,
|
| 308 |
+
"eval_steps_per_second": 2.715,
|
| 309 |
+
"step": 200
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 0.6,
|
| 313 |
+
"learning_rate": 2.2380952380952384e-05,
|
| 314 |
+
"loss": 0.3159,
|
| 315 |
+
"step": 210
|
| 316 |
+
},
|
| 317 |
+
{
|
| 318 |
+
"epoch": 0.6,
|
| 319 |
+
"eval_accuracy": 0.984,
|
| 320 |
+
"eval_loss": 0.05127624422311783,
|
| 321 |
+
"eval_runtime": 57.9626,
|
| 322 |
+
"eval_samples_per_second": 86.263,
|
| 323 |
+
"eval_steps_per_second": 2.709,
|
| 324 |
+
"step": 210
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.63,
|
| 328 |
+
"learning_rate": 2.0793650793650797e-05,
|
| 329 |
+
"loss": 0.2795,
|
| 330 |
+
"step": 220
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"epoch": 0.63,
|
| 334 |
+
"eval_accuracy": 0.9846,
|
| 335 |
+
"eval_loss": 0.046046771109104156,
|
| 336 |
+
"eval_runtime": 57.8964,
|
| 337 |
+
"eval_samples_per_second": 86.361,
|
| 338 |
+
"eval_steps_per_second": 2.712,
|
| 339 |
+
"step": 220
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.65,
|
| 343 |
+
"learning_rate": 1.920634920634921e-05,
|
| 344 |
+
"loss": 0.3113,
|
| 345 |
+
"step": 230
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.65,
|
| 349 |
+
"eval_accuracy": 0.9832,
|
| 350 |
+
"eval_loss": 0.04946137219667435,
|
| 351 |
+
"eval_runtime": 57.9735,
|
| 352 |
+
"eval_samples_per_second": 86.246,
|
| 353 |
+
"eval_steps_per_second": 2.708,
|
| 354 |
+
"step": 230
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 0.68,
|
| 358 |
+
"learning_rate": 1.761904761904762e-05,
|
| 359 |
+
"loss": 0.2882,
|
| 360 |
+
"step": 240
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.68,
|
| 364 |
+
"eval_accuracy": 0.9838,
|
| 365 |
+
"eval_loss": 0.047487206757068634,
|
| 366 |
+
"eval_runtime": 57.8304,
|
| 367 |
+
"eval_samples_per_second": 86.46,
|
| 368 |
+
"eval_steps_per_second": 2.715,
|
| 369 |
+
"step": 240
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 0.71,
|
| 373 |
+
"learning_rate": 1.6031746031746033e-05,
|
| 374 |
+
"loss": 0.263,
|
| 375 |
+
"step": 250
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 0.71,
|
| 379 |
+
"eval_accuracy": 0.9854,
|
| 380 |
+
"eval_loss": 0.0448877215385437,
|
| 381 |
+
"eval_runtime": 57.7851,
|
| 382 |
+
"eval_samples_per_second": 86.528,
|
| 383 |
+
"eval_steps_per_second": 2.717,
|
| 384 |
+
"step": 250
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 0.74,
|
| 388 |
+
"learning_rate": 1.4444444444444444e-05,
|
| 389 |
+
"loss": 0.2686,
|
| 390 |
+
"step": 260
|
| 391 |
+
},
|
| 392 |
+
{
|
| 393 |
+
"epoch": 0.74,
|
| 394 |
+
"eval_accuracy": 0.9826,
|
| 395 |
+
"eval_loss": 0.05095283314585686,
|
| 396 |
+
"eval_runtime": 57.6101,
|
| 397 |
+
"eval_samples_per_second": 86.79,
|
| 398 |
+
"eval_steps_per_second": 2.725,
|
| 399 |
+
"step": 260
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 0.77,
|
| 403 |
+
"learning_rate": 1.2857142857142857e-05,
|
| 404 |
+
"loss": 0.2705,
|
| 405 |
+
"step": 270
|
| 406 |
+
},
|
| 407 |
+
{
|
| 408 |
+
"epoch": 0.77,
|
| 409 |
+
"eval_accuracy": 0.9846,
|
| 410 |
+
"eval_loss": 0.04833937808871269,
|
| 411 |
+
"eval_runtime": 57.637,
|
| 412 |
+
"eval_samples_per_second": 86.75,
|
| 413 |
+
"eval_steps_per_second": 2.724,
|
| 414 |
+
"step": 270
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 0.8,
|
| 418 |
+
"learning_rate": 1.126984126984127e-05,
|
| 419 |
+
"loss": 0.2807,
|
| 420 |
+
"step": 280
|
| 421 |
+
},
|
| 422 |
+
{
|
| 423 |
+
"epoch": 0.8,
|
| 424 |
+
"eval_accuracy": 0.9854,
|
| 425 |
+
"eval_loss": 0.04302287474274635,
|
| 426 |
+
"eval_runtime": 57.6918,
|
| 427 |
+
"eval_samples_per_second": 86.667,
|
| 428 |
+
"eval_steps_per_second": 2.721,
|
| 429 |
+
"step": 280
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.82,
|
| 433 |
+
"learning_rate": 9.682539682539683e-06,
|
| 434 |
+
"loss": 0.2583,
|
| 435 |
+
"step": 290
|
| 436 |
+
},
|
| 437 |
+
{
|
| 438 |
+
"epoch": 0.82,
|
| 439 |
+
"eval_accuracy": 0.9858,
|
| 440 |
+
"eval_loss": 0.04517200589179993,
|
| 441 |
+
"eval_runtime": 57.63,
|
| 442 |
+
"eval_samples_per_second": 86.76,
|
| 443 |
+
"eval_steps_per_second": 2.724,
|
| 444 |
+
"step": 290
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.85,
|
| 448 |
+
"learning_rate": 8.095238095238097e-06,
|
| 449 |
+
"loss": 0.2346,
|
| 450 |
+
"step": 300
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.85,
|
| 454 |
+
"eval_accuracy": 0.9858,
|
| 455 |
+
"eval_loss": 0.04349066689610481,
|
| 456 |
+
"eval_runtime": 57.5683,
|
| 457 |
+
"eval_samples_per_second": 86.853,
|
| 458 |
+
"eval_steps_per_second": 2.727,
|
| 459 |
+
"step": 300
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 0.88,
|
| 463 |
+
"learning_rate": 6.507936507936509e-06,
|
| 464 |
+
"loss": 0.2294,
|
| 465 |
+
"step": 310
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.88,
|
| 469 |
+
"eval_accuracy": 0.986,
|
| 470 |
+
"eval_loss": 0.04344068095088005,
|
| 471 |
+
"eval_runtime": 57.6423,
|
| 472 |
+
"eval_samples_per_second": 86.742,
|
| 473 |
+
"eval_steps_per_second": 2.724,
|
| 474 |
+
"step": 310
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 0.91,
|
| 478 |
+
"learning_rate": 4.920634920634921e-06,
|
| 479 |
+
"loss": 0.2608,
|
| 480 |
+
"step": 320
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 0.91,
|
| 484 |
+
"eval_accuracy": 0.986,
|
| 485 |
+
"eval_loss": 0.04333638399839401,
|
| 486 |
+
"eval_runtime": 57.6441,
|
| 487 |
+
"eval_samples_per_second": 86.739,
|
| 488 |
+
"eval_steps_per_second": 2.724,
|
| 489 |
+
"step": 320
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 0.94,
|
| 493 |
+
"learning_rate": 3.3333333333333333e-06,
|
| 494 |
+
"loss": 0.2642,
|
| 495 |
+
"step": 330
|
| 496 |
+
},
|
| 497 |
+
{
|
| 498 |
+
"epoch": 0.94,
|
| 499 |
+
"eval_accuracy": 0.9866,
|
| 500 |
+
"eval_loss": 0.04251227155327797,
|
| 501 |
+
"eval_runtime": 59.7952,
|
| 502 |
+
"eval_samples_per_second": 83.619,
|
| 503 |
+
"eval_steps_per_second": 2.626,
|
| 504 |
+
"step": 330
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 0.97,
|
| 508 |
+
"learning_rate": 1.7460317460317462e-06,
|
| 509 |
+
"loss": 0.2781,
|
| 510 |
+
"step": 340
|
| 511 |
+
},
|
| 512 |
+
{
|
| 513 |
+
"epoch": 0.97,
|
| 514 |
+
"eval_accuracy": 0.986,
|
| 515 |
+
"eval_loss": 0.041707251220941544,
|
| 516 |
+
"eval_runtime": 57.5428,
|
| 517 |
+
"eval_samples_per_second": 86.892,
|
| 518 |
+
"eval_steps_per_second": 2.728,
|
| 519 |
+
"step": 340
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 1.0,
|
| 523 |
+
"learning_rate": 1.5873015873015874e-07,
|
| 524 |
+
"loss": 0.247,
|
| 525 |
+
"step": 350
|
| 526 |
+
},
|
| 527 |
+
{
|
| 528 |
+
"epoch": 1.0,
|
| 529 |
+
"eval_accuracy": 0.9858,
|
| 530 |
+
"eval_loss": 0.041396625339984894,
|
| 531 |
+
"eval_runtime": 57.5657,
|
| 532 |
+
"eval_samples_per_second": 86.857,
|
| 533 |
+
"eval_steps_per_second": 2.727,
|
| 534 |
+
"step": 350
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 1.0,
|
| 538 |
+
"step": 351,
|
| 539 |
+
"total_flos": 3.5202418590178345e+18,
|
| 540 |
+
"train_loss": 0.46742354324910035,
|
| 541 |
+
"train_runtime": 3479.7951,
|
| 542 |
+
"train_samples_per_second": 12.932,
|
| 543 |
+
"train_steps_per_second": 0.101
|
| 544 |
+
}
|
| 545 |
+
],
|
| 546 |
+
"logging_steps": 10,
|
| 547 |
+
"max_steps": 351,
|
| 548 |
+
"num_train_epochs": 1,
|
| 549 |
+
"save_steps": 500,
|
| 550 |
+
"total_flos": 3.5202418590178345e+18,
|
| 551 |
+
"trial_name": null,
|
| 552 |
+
"trial_params": null
|
| 553 |
+
}
|