Commit
•
54acd22
1
Parent(s):
ccffb3e
End of training
Browse files- all_results.json +8 -0
- train_results.json +8 -0
- trainer_state.json +751 -0
all_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"total_flos": 1.1227700242944e+16,
|
| 4 |
+
"train_loss": 1.3416433044646565,
|
| 5 |
+
"train_runtime": 4283.2723,
|
| 6 |
+
"train_samples_per_second": 36.422,
|
| 7 |
+
"train_steps_per_second": 0.284
|
| 8 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"total_flos": 1.1227700242944e+16,
|
| 4 |
+
"train_loss": 1.3416433044646565,
|
| 5 |
+
"train_runtime": 4283.2723,
|
| 6 |
+
"train_samples_per_second": 36.422,
|
| 7 |
+
"train_steps_per_second": 0.284
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,751 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 2.998038536979347,
|
| 5 |
+
"global_step": 1218,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.02,
|
| 12 |
+
"learning_rate": 1.092896174863388e-06,
|
| 13 |
+
"loss": 5.1703,
|
| 14 |
+
"step": 10
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.05,
|
| 18 |
+
"learning_rate": 2.185792349726776e-06,
|
| 19 |
+
"loss": 4.4905,
|
| 20 |
+
"step": 20
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.07,
|
| 24 |
+
"learning_rate": 3.2786885245901638e-06,
|
| 25 |
+
"loss": 3.712,
|
| 26 |
+
"step": 30
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.1,
|
| 30 |
+
"learning_rate": 4.371584699453552e-06,
|
| 31 |
+
"loss": 2.9301,
|
| 32 |
+
"step": 40
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.12,
|
| 36 |
+
"learning_rate": 5.464480874316941e-06,
|
| 37 |
+
"loss": 2.4493,
|
| 38 |
+
"step": 50
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.15,
|
| 42 |
+
"learning_rate": 6.5573770491803276e-06,
|
| 43 |
+
"loss": 2.2009,
|
| 44 |
+
"step": 60
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.17,
|
| 48 |
+
"learning_rate": 7.650273224043716e-06,
|
| 49 |
+
"loss": 2.1093,
|
| 50 |
+
"step": 70
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.2,
|
| 54 |
+
"learning_rate": 8.743169398907103e-06,
|
| 55 |
+
"loss": 2.0401,
|
| 56 |
+
"step": 80
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 0.22,
|
| 60 |
+
"learning_rate": 9.836065573770493e-06,
|
| 61 |
+
"loss": 1.9798,
|
| 62 |
+
"step": 90
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 0.25,
|
| 66 |
+
"learning_rate": 1.0928961748633882e-05,
|
| 67 |
+
"loss": 1.9179,
|
| 68 |
+
"step": 100
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 0.27,
|
| 72 |
+
"learning_rate": 1.202185792349727e-05,
|
| 73 |
+
"loss": 1.8746,
|
| 74 |
+
"step": 110
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.3,
|
| 78 |
+
"learning_rate": 1.3114754098360655e-05,
|
| 79 |
+
"loss": 1.8008,
|
| 80 |
+
"step": 120
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.32,
|
| 84 |
+
"learning_rate": 1.4207650273224044e-05,
|
| 85 |
+
"loss": 1.7745,
|
| 86 |
+
"step": 130
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.34,
|
| 90 |
+
"learning_rate": 1.5300546448087432e-05,
|
| 91 |
+
"loss": 1.7164,
|
| 92 |
+
"step": 140
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 0.37,
|
| 96 |
+
"learning_rate": 1.639344262295082e-05,
|
| 97 |
+
"loss": 1.6755,
|
| 98 |
+
"step": 150
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 0.39,
|
| 102 |
+
"learning_rate": 1.7486338797814207e-05,
|
| 103 |
+
"loss": 1.613,
|
| 104 |
+
"step": 160
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 0.42,
|
| 108 |
+
"learning_rate": 1.85792349726776e-05,
|
| 109 |
+
"loss": 1.5696,
|
| 110 |
+
"step": 170
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 0.44,
|
| 114 |
+
"learning_rate": 1.9672131147540985e-05,
|
| 115 |
+
"loss": 1.5038,
|
| 116 |
+
"step": 180
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 0.47,
|
| 120 |
+
"learning_rate": 1.986473429951691e-05,
|
| 121 |
+
"loss": 1.482,
|
| 122 |
+
"step": 190
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.49,
|
| 126 |
+
"learning_rate": 1.9671497584541064e-05,
|
| 127 |
+
"loss": 1.4213,
|
| 128 |
+
"step": 200
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.52,
|
| 132 |
+
"learning_rate": 1.947826086956522e-05,
|
| 133 |
+
"loss": 1.3555,
|
| 134 |
+
"step": 210
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 0.54,
|
| 138 |
+
"learning_rate": 1.9285024154589372e-05,
|
| 139 |
+
"loss": 1.3546,
|
| 140 |
+
"step": 220
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 0.57,
|
| 144 |
+
"learning_rate": 1.909178743961353e-05,
|
| 145 |
+
"loss": 1.3325,
|
| 146 |
+
"step": 230
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 0.59,
|
| 150 |
+
"learning_rate": 1.8898550724637685e-05,
|
| 151 |
+
"loss": 1.2889,
|
| 152 |
+
"step": 240
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 0.62,
|
| 156 |
+
"learning_rate": 1.8705314009661837e-05,
|
| 157 |
+
"loss": 1.2737,
|
| 158 |
+
"step": 250
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 0.64,
|
| 162 |
+
"learning_rate": 1.8512077294685994e-05,
|
| 163 |
+
"loss": 1.2861,
|
| 164 |
+
"step": 260
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.66,
|
| 168 |
+
"learning_rate": 1.8318840579710146e-05,
|
| 169 |
+
"loss": 1.2707,
|
| 170 |
+
"step": 270
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.69,
|
| 174 |
+
"learning_rate": 1.81256038647343e-05,
|
| 175 |
+
"loss": 1.2544,
|
| 176 |
+
"step": 280
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 0.71,
|
| 180 |
+
"learning_rate": 1.7932367149758455e-05,
|
| 181 |
+
"loss": 1.2397,
|
| 182 |
+
"step": 290
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 0.74,
|
| 186 |
+
"learning_rate": 1.773913043478261e-05,
|
| 187 |
+
"loss": 1.2487,
|
| 188 |
+
"step": 300
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 0.76,
|
| 192 |
+
"learning_rate": 1.7545893719806767e-05,
|
| 193 |
+
"loss": 1.2469,
|
| 194 |
+
"step": 310
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 0.79,
|
| 198 |
+
"learning_rate": 1.735265700483092e-05,
|
| 199 |
+
"loss": 1.2192,
|
| 200 |
+
"step": 320
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 0.81,
|
| 204 |
+
"learning_rate": 1.7159420289855073e-05,
|
| 205 |
+
"loss": 1.2422,
|
| 206 |
+
"step": 330
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.84,
|
| 210 |
+
"learning_rate": 1.696618357487923e-05,
|
| 211 |
+
"loss": 1.23,
|
| 212 |
+
"step": 340
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.86,
|
| 216 |
+
"learning_rate": 1.677294685990338e-05,
|
| 217 |
+
"loss": 1.2295,
|
| 218 |
+
"step": 350
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 0.89,
|
| 222 |
+
"learning_rate": 1.6579710144927538e-05,
|
| 223 |
+
"loss": 1.224,
|
| 224 |
+
"step": 360
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 0.91,
|
| 228 |
+
"learning_rate": 1.6386473429951694e-05,
|
| 229 |
+
"loss": 1.2363,
|
| 230 |
+
"step": 370
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 0.94,
|
| 234 |
+
"learning_rate": 1.6193236714975847e-05,
|
| 235 |
+
"loss": 1.216,
|
| 236 |
+
"step": 380
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 0.96,
|
| 240 |
+
"learning_rate": 1.6000000000000003e-05,
|
| 241 |
+
"loss": 1.205,
|
| 242 |
+
"step": 390
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 0.98,
|
| 246 |
+
"learning_rate": 1.5806763285024155e-05,
|
| 247 |
+
"loss": 1.2222,
|
| 248 |
+
"step": 400
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 1.01,
|
| 252 |
+
"learning_rate": 1.5613526570048308e-05,
|
| 253 |
+
"loss": 1.1912,
|
| 254 |
+
"step": 410
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 1.03,
|
| 258 |
+
"learning_rate": 1.5420289855072464e-05,
|
| 259 |
+
"loss": 1.1828,
|
| 260 |
+
"step": 420
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 1.06,
|
| 264 |
+
"learning_rate": 1.5227053140096619e-05,
|
| 265 |
+
"loss": 1.1682,
|
| 266 |
+
"step": 430
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 1.08,
|
| 270 |
+
"learning_rate": 1.5033816425120775e-05,
|
| 271 |
+
"loss": 1.1822,
|
| 272 |
+
"step": 440
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 1.11,
|
| 276 |
+
"learning_rate": 1.484057971014493e-05,
|
| 277 |
+
"loss": 1.166,
|
| 278 |
+
"step": 450
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 1.13,
|
| 282 |
+
"learning_rate": 1.4647342995169082e-05,
|
| 283 |
+
"loss": 1.1727,
|
| 284 |
+
"step": 460
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 1.16,
|
| 288 |
+
"learning_rate": 1.4454106280193238e-05,
|
| 289 |
+
"loss": 1.1796,
|
| 290 |
+
"step": 470
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 1.18,
|
| 294 |
+
"learning_rate": 1.4260869565217392e-05,
|
| 295 |
+
"loss": 1.1568,
|
| 296 |
+
"step": 480
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 1.21,
|
| 300 |
+
"learning_rate": 1.4067632850241547e-05,
|
| 301 |
+
"loss": 1.1816,
|
| 302 |
+
"step": 490
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 1.23,
|
| 306 |
+
"learning_rate": 1.3874396135265701e-05,
|
| 307 |
+
"loss": 1.1614,
|
| 308 |
+
"step": 500
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 1.26,
|
| 312 |
+
"learning_rate": 1.3681159420289856e-05,
|
| 313 |
+
"loss": 1.1636,
|
| 314 |
+
"step": 510
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 1.28,
|
| 318 |
+
"learning_rate": 1.3487922705314012e-05,
|
| 319 |
+
"loss": 1.1602,
|
| 320 |
+
"step": 520
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 1.3,
|
| 324 |
+
"learning_rate": 1.3294685990338165e-05,
|
| 325 |
+
"loss": 1.1624,
|
| 326 |
+
"step": 530
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 1.33,
|
| 330 |
+
"learning_rate": 1.3101449275362319e-05,
|
| 331 |
+
"loss": 1.1523,
|
| 332 |
+
"step": 540
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 1.35,
|
| 336 |
+
"learning_rate": 1.2908212560386475e-05,
|
| 337 |
+
"loss": 1.1672,
|
| 338 |
+
"step": 550
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 1.38,
|
| 342 |
+
"learning_rate": 1.271497584541063e-05,
|
| 343 |
+
"loss": 1.1565,
|
| 344 |
+
"step": 560
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 1.4,
|
| 348 |
+
"learning_rate": 1.2521739130434784e-05,
|
| 349 |
+
"loss": 1.1585,
|
| 350 |
+
"step": 570
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 1.43,
|
| 354 |
+
"learning_rate": 1.2328502415458938e-05,
|
| 355 |
+
"loss": 1.1564,
|
| 356 |
+
"step": 580
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 1.45,
|
| 360 |
+
"learning_rate": 1.2135265700483093e-05,
|
| 361 |
+
"loss": 1.1381,
|
| 362 |
+
"step": 590
|
| 363 |
+
},
|
| 364 |
+
{
|
| 365 |
+
"epoch": 1.48,
|
| 366 |
+
"learning_rate": 1.1942028985507247e-05,
|
| 367 |
+
"loss": 1.17,
|
| 368 |
+
"step": 600
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 1.5,
|
| 372 |
+
"learning_rate": 1.1748792270531402e-05,
|
| 373 |
+
"loss": 1.1754,
|
| 374 |
+
"step": 610
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 1.53,
|
| 378 |
+
"learning_rate": 1.1555555555555556e-05,
|
| 379 |
+
"loss": 1.1292,
|
| 380 |
+
"step": 620
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 1.55,
|
| 384 |
+
"learning_rate": 1.1362318840579712e-05,
|
| 385 |
+
"loss": 1.1239,
|
| 386 |
+
"step": 630
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 1.58,
|
| 390 |
+
"learning_rate": 1.1169082125603865e-05,
|
| 391 |
+
"loss": 1.1878,
|
| 392 |
+
"step": 640
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 1.6,
|
| 396 |
+
"learning_rate": 1.0975845410628021e-05,
|
| 397 |
+
"loss": 1.1506,
|
| 398 |
+
"step": 650
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 1.62,
|
| 402 |
+
"learning_rate": 1.0782608695652175e-05,
|
| 403 |
+
"loss": 1.1536,
|
| 404 |
+
"step": 660
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 1.65,
|
| 408 |
+
"learning_rate": 1.0589371980676328e-05,
|
| 409 |
+
"loss": 1.1449,
|
| 410 |
+
"step": 670
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 1.67,
|
| 414 |
+
"learning_rate": 1.0396135265700484e-05,
|
| 415 |
+
"loss": 1.1287,
|
| 416 |
+
"step": 680
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 1.7,
|
| 420 |
+
"learning_rate": 1.0202898550724639e-05,
|
| 421 |
+
"loss": 1.1366,
|
| 422 |
+
"step": 690
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 1.72,
|
| 426 |
+
"learning_rate": 1.0009661835748795e-05,
|
| 427 |
+
"loss": 1.1468,
|
| 428 |
+
"step": 700
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 1.75,
|
| 432 |
+
"learning_rate": 9.816425120772947e-06,
|
| 433 |
+
"loss": 1.1484,
|
| 434 |
+
"step": 710
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 1.77,
|
| 438 |
+
"learning_rate": 9.623188405797102e-06,
|
| 439 |
+
"loss": 1.1549,
|
| 440 |
+
"step": 720
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 1.8,
|
| 444 |
+
"learning_rate": 9.429951690821256e-06,
|
| 445 |
+
"loss": 1.1501,
|
| 446 |
+
"step": 730
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 1.82,
|
| 450 |
+
"learning_rate": 9.23671497584541e-06,
|
| 451 |
+
"loss": 1.1433,
|
| 452 |
+
"step": 740
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 1.85,
|
| 456 |
+
"learning_rate": 9.043478260869565e-06,
|
| 457 |
+
"loss": 1.1315,
|
| 458 |
+
"step": 750
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 1.87,
|
| 462 |
+
"learning_rate": 8.850241545893721e-06,
|
| 463 |
+
"loss": 1.1615,
|
| 464 |
+
"step": 760
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 1.9,
|
| 468 |
+
"learning_rate": 8.657004830917876e-06,
|
| 469 |
+
"loss": 1.1521,
|
| 470 |
+
"step": 770
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 1.92,
|
| 474 |
+
"learning_rate": 8.46376811594203e-06,
|
| 475 |
+
"loss": 1.138,
|
| 476 |
+
"step": 780
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 1.94,
|
| 480 |
+
"learning_rate": 8.270531400966184e-06,
|
| 481 |
+
"loss": 1.1608,
|
| 482 |
+
"step": 790
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 1.97,
|
| 486 |
+
"learning_rate": 8.077294685990339e-06,
|
| 487 |
+
"loss": 1.1655,
|
| 488 |
+
"step": 800
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 1.99,
|
| 492 |
+
"learning_rate": 7.884057971014493e-06,
|
| 493 |
+
"loss": 1.1367,
|
| 494 |
+
"step": 810
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 2.02,
|
| 498 |
+
"learning_rate": 7.690821256038648e-06,
|
| 499 |
+
"loss": 1.1201,
|
| 500 |
+
"step": 820
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 2.04,
|
| 504 |
+
"learning_rate": 7.497584541062802e-06,
|
| 505 |
+
"loss": 1.1403,
|
| 506 |
+
"step": 830
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 2.07,
|
| 510 |
+
"learning_rate": 7.304347826086957e-06,
|
| 511 |
+
"loss": 1.1166,
|
| 512 |
+
"step": 840
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 2.09,
|
| 516 |
+
"learning_rate": 7.111111111111112e-06,
|
| 517 |
+
"loss": 1.1243,
|
| 518 |
+
"step": 850
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 2.12,
|
| 522 |
+
"learning_rate": 6.917874396135267e-06,
|
| 523 |
+
"loss": 1.0965,
|
| 524 |
+
"step": 860
|
| 525 |
+
},
|
| 526 |
+
{
|
| 527 |
+
"epoch": 2.14,
|
| 528 |
+
"learning_rate": 6.724637681159421e-06,
|
| 529 |
+
"loss": 1.1033,
|
| 530 |
+
"step": 870
|
| 531 |
+
},
|
| 532 |
+
{
|
| 533 |
+
"epoch": 2.17,
|
| 534 |
+
"learning_rate": 6.531400966183575e-06,
|
| 535 |
+
"loss": 1.1162,
|
| 536 |
+
"step": 880
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 2.19,
|
| 540 |
+
"learning_rate": 6.33816425120773e-06,
|
| 541 |
+
"loss": 1.1242,
|
| 542 |
+
"step": 890
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 2.22,
|
| 546 |
+
"learning_rate": 6.144927536231885e-06,
|
| 547 |
+
"loss": 1.1151,
|
| 548 |
+
"step": 900
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 2.24,
|
| 552 |
+
"learning_rate": 5.951690821256038e-06,
|
| 553 |
+
"loss": 1.1068,
|
| 554 |
+
"step": 910
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 2.26,
|
| 558 |
+
"learning_rate": 5.758454106280194e-06,
|
| 559 |
+
"loss": 1.1132,
|
| 560 |
+
"step": 920
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 2.29,
|
| 564 |
+
"learning_rate": 5.565217391304348e-06,
|
| 565 |
+
"loss": 1.1307,
|
| 566 |
+
"step": 930
|
| 567 |
+
},
|
| 568 |
+
{
|
| 569 |
+
"epoch": 2.31,
|
| 570 |
+
"learning_rate": 5.371980676328503e-06,
|
| 571 |
+
"loss": 1.1174,
|
| 572 |
+
"step": 940
|
| 573 |
+
},
|
| 574 |
+
{
|
| 575 |
+
"epoch": 2.34,
|
| 576 |
+
"learning_rate": 5.178743961352657e-06,
|
| 577 |
+
"loss": 1.1135,
|
| 578 |
+
"step": 950
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 2.36,
|
| 582 |
+
"learning_rate": 4.985507246376812e-06,
|
| 583 |
+
"loss": 1.0946,
|
| 584 |
+
"step": 960
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 2.39,
|
| 588 |
+
"learning_rate": 4.7922705314009665e-06,
|
| 589 |
+
"loss": 1.1089,
|
| 590 |
+
"step": 970
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 2.41,
|
| 594 |
+
"learning_rate": 4.599033816425121e-06,
|
| 595 |
+
"loss": 1.116,
|
| 596 |
+
"step": 980
|
| 597 |
+
},
|
| 598 |
+
{
|
| 599 |
+
"epoch": 2.44,
|
| 600 |
+
"learning_rate": 4.405797101449275e-06,
|
| 601 |
+
"loss": 1.1129,
|
| 602 |
+
"step": 990
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 2.46,
|
| 606 |
+
"learning_rate": 4.212560386473431e-06,
|
| 607 |
+
"loss": 1.0921,
|
| 608 |
+
"step": 1000
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 2.49,
|
| 612 |
+
"learning_rate": 4.019323671497585e-06,
|
| 613 |
+
"loss": 1.0999,
|
| 614 |
+
"step": 1010
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"epoch": 2.51,
|
| 618 |
+
"learning_rate": 3.8260869565217395e-06,
|
| 619 |
+
"loss": 1.1141,
|
| 620 |
+
"step": 1020
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"epoch": 2.54,
|
| 624 |
+
"learning_rate": 3.632850241545894e-06,
|
| 625 |
+
"loss": 1.1274,
|
| 626 |
+
"step": 1030
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 2.56,
|
| 630 |
+
"learning_rate": 3.4396135265700487e-06,
|
| 631 |
+
"loss": 1.1194,
|
| 632 |
+
"step": 1040
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 2.58,
|
| 636 |
+
"learning_rate": 3.2463768115942027e-06,
|
| 637 |
+
"loss": 1.1065,
|
| 638 |
+
"step": 1050
|
| 639 |
+
},
|
| 640 |
+
{
|
| 641 |
+
"epoch": 2.61,
|
| 642 |
+
"learning_rate": 3.0531400966183576e-06,
|
| 643 |
+
"loss": 1.1107,
|
| 644 |
+
"step": 1060
|
| 645 |
+
},
|
| 646 |
+
{
|
| 647 |
+
"epoch": 2.63,
|
| 648 |
+
"learning_rate": 2.8599033816425124e-06,
|
| 649 |
+
"loss": 1.112,
|
| 650 |
+
"step": 1070
|
| 651 |
+
},
|
| 652 |
+
{
|
| 653 |
+
"epoch": 2.66,
|
| 654 |
+
"learning_rate": 2.666666666666667e-06,
|
| 655 |
+
"loss": 1.1252,
|
| 656 |
+
"step": 1080
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"epoch": 2.68,
|
| 660 |
+
"learning_rate": 2.4734299516908212e-06,
|
| 661 |
+
"loss": 1.1116,
|
| 662 |
+
"step": 1090
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 2.71,
|
| 666 |
+
"learning_rate": 2.280193236714976e-06,
|
| 667 |
+
"loss": 1.0865,
|
| 668 |
+
"step": 1100
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 2.73,
|
| 672 |
+
"learning_rate": 2.0869565217391305e-06,
|
| 673 |
+
"loss": 1.1122,
|
| 674 |
+
"step": 1110
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 2.76,
|
| 678 |
+
"learning_rate": 1.8937198067632853e-06,
|
| 679 |
+
"loss": 1.1225,
|
| 680 |
+
"step": 1120
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"epoch": 2.78,
|
| 684 |
+
"learning_rate": 1.7004830917874398e-06,
|
| 685 |
+
"loss": 1.1077,
|
| 686 |
+
"step": 1130
|
| 687 |
+
},
|
| 688 |
+
{
|
| 689 |
+
"epoch": 2.81,
|
| 690 |
+
"learning_rate": 1.5072463768115944e-06,
|
| 691 |
+
"loss": 1.1104,
|
| 692 |
+
"step": 1140
|
| 693 |
+
},
|
| 694 |
+
{
|
| 695 |
+
"epoch": 2.83,
|
| 696 |
+
"learning_rate": 1.314009661835749e-06,
|
| 697 |
+
"loss": 1.0929,
|
| 698 |
+
"step": 1150
|
| 699 |
+
},
|
| 700 |
+
{
|
| 701 |
+
"epoch": 2.86,
|
| 702 |
+
"learning_rate": 1.1207729468599034e-06,
|
| 703 |
+
"loss": 1.0986,
|
| 704 |
+
"step": 1160
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 2.88,
|
| 708 |
+
"learning_rate": 9.275362318840581e-07,
|
| 709 |
+
"loss": 1.1137,
|
| 710 |
+
"step": 1170
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 2.9,
|
| 714 |
+
"learning_rate": 7.342995169082126e-07,
|
| 715 |
+
"loss": 1.0868,
|
| 716 |
+
"step": 1180
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 2.93,
|
| 720 |
+
"learning_rate": 5.410628019323672e-07,
|
| 721 |
+
"loss": 1.1027,
|
| 722 |
+
"step": 1190
|
| 723 |
+
},
|
| 724 |
+
{
|
| 725 |
+
"epoch": 2.95,
|
| 726 |
+
"learning_rate": 3.4782608695652175e-07,
|
| 727 |
+
"loss": 1.1193,
|
| 728 |
+
"step": 1200
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 2.98,
|
| 732 |
+
"learning_rate": 1.5458937198067633e-07,
|
| 733 |
+
"loss": 1.1247,
|
| 734 |
+
"step": 1210
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 3.0,
|
| 738 |
+
"step": 1218,
|
| 739 |
+
"total_flos": 1.1227700242944e+16,
|
| 740 |
+
"train_loss": 1.3416433044646565,
|
| 741 |
+
"train_runtime": 4283.2723,
|
| 742 |
+
"train_samples_per_second": 36.422,
|
| 743 |
+
"train_steps_per_second": 0.284
|
| 744 |
+
}
|
| 745 |
+
],
|
| 746 |
+
"max_steps": 1218,
|
| 747 |
+
"num_train_epochs": 3,
|
| 748 |
+
"total_flos": 1.1227700242944e+16,
|
| 749 |
+
"trial_name": null,
|
| 750 |
+
"trial_params": null
|
| 751 |
+
}
|