End of training
Browse files- all_results.json +12 -0
- eval_results.json +8 -0
- train_results.json +7 -0
- trainer_state.json +679 -0
all_results.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 20.0,
|
| 3 |
+
"eval_accuracy": 0.9087967049132097,
|
| 4 |
+
"eval_loss": 0.3166075348854065,
|
| 5 |
+
"eval_runtime": 144.5486,
|
| 6 |
+
"eval_samples_per_second": 47.029,
|
| 7 |
+
"eval_steps_per_second": 1.474,
|
| 8 |
+
"train_loss": 0.8469889265552798,
|
| 9 |
+
"train_runtime": 24285.4703,
|
| 10 |
+
"train_samples_per_second": 42.078,
|
| 11 |
+
"train_steps_per_second": 0.329
|
| 12 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 20.0,
|
| 3 |
+
"eval_accuracy": 0.9087967049132097,
|
| 4 |
+
"eval_loss": 0.3166075348854065,
|
| 5 |
+
"eval_runtime": 144.5486,
|
| 6 |
+
"eval_samples_per_second": 47.029,
|
| 7 |
+
"eval_steps_per_second": 1.474
|
| 8 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 20.0,
|
| 3 |
+
"train_loss": 0.8469889265552798,
|
| 4 |
+
"train_runtime": 24285.4703,
|
| 5 |
+
"train_samples_per_second": 42.078,
|
| 6 |
+
"train_steps_per_second": 0.329
|
| 7 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,679 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.9087967049132097,
|
| 3 |
+
"best_model_checkpoint": "trillsson3-ft-keyword-spotting-11/checkpoint-7980",
|
| 4 |
+
"epoch": 19.999373825923605,
|
| 5 |
+
"global_step": 7980,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.25,
|
| 12 |
+
"learning_rate": 3.571428571428571e-05,
|
| 13 |
+
"loss": 7.4329,
|
| 14 |
+
"step": 100
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.5,
|
| 18 |
+
"learning_rate": 7.330827067669172e-05,
|
| 19 |
+
"loss": 4.5207,
|
| 20 |
+
"step": 200
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.75,
|
| 24 |
+
"learning_rate": 0.00011090225563909774,
|
| 25 |
+
"loss": 2.9219,
|
| 26 |
+
"step": 300
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 1.0,
|
| 30 |
+
"eval_accuracy": 0.6216534274786703,
|
| 31 |
+
"eval_loss": 1.202303409576416,
|
| 32 |
+
"eval_runtime": 139.1467,
|
| 33 |
+
"eval_samples_per_second": 48.855,
|
| 34 |
+
"eval_steps_per_second": 1.531,
|
| 35 |
+
"step": 399
|
| 36 |
+
},
|
| 37 |
+
{
|
| 38 |
+
"epoch": 1.0,
|
| 39 |
+
"learning_rate": 0.00014849624060150375,
|
| 40 |
+
"loss": 2.0315,
|
| 41 |
+
"step": 400
|
| 42 |
+
},
|
| 43 |
+
{
|
| 44 |
+
"epoch": 1.25,
|
| 45 |
+
"learning_rate": 0.00018609022556390976,
|
| 46 |
+
"loss": 1.4541,
|
| 47 |
+
"step": 500
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"epoch": 1.5,
|
| 51 |
+
"learning_rate": 0.00022368421052631576,
|
| 52 |
+
"loss": 1.1541,
|
| 53 |
+
"step": 600
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 1.75,
|
| 57 |
+
"learning_rate": 0.00026127819548872177,
|
| 58 |
+
"loss": 0.9604,
|
| 59 |
+
"step": 700
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 2.0,
|
| 63 |
+
"eval_accuracy": 0.8117093262724331,
|
| 64 |
+
"eval_loss": 0.5436997413635254,
|
| 65 |
+
"eval_runtime": 140.1586,
|
| 66 |
+
"eval_samples_per_second": 48.502,
|
| 67 |
+
"eval_steps_per_second": 1.52,
|
| 68 |
+
"step": 798
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 2.01,
|
| 72 |
+
"learning_rate": 0.0002988721804511278,
|
| 73 |
+
"loss": 0.8683,
|
| 74 |
+
"step": 800
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 2.26,
|
| 78 |
+
"learning_rate": 0.00029594820384294066,
|
| 79 |
+
"loss": 0.8309,
|
| 80 |
+
"step": 900
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 2.51,
|
| 84 |
+
"learning_rate": 0.0002917710944026733,
|
| 85 |
+
"loss": 0.7717,
|
| 86 |
+
"step": 1000
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 2.76,
|
| 90 |
+
"learning_rate": 0.000287593984962406,
|
| 91 |
+
"loss": 0.7608,
|
| 92 |
+
"step": 1100
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 3.0,
|
| 96 |
+
"eval_accuracy": 0.8887908208296558,
|
| 97 |
+
"eval_loss": 0.42219480872154236,
|
| 98 |
+
"eval_runtime": 140.2584,
|
| 99 |
+
"eval_samples_per_second": 48.468,
|
| 100 |
+
"eval_steps_per_second": 1.519,
|
| 101 |
+
"step": 1197
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 3.01,
|
| 105 |
+
"learning_rate": 0.00028341687552213866,
|
| 106 |
+
"loss": 0.7384,
|
| 107 |
+
"step": 1200
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 3.26,
|
| 111 |
+
"learning_rate": 0.0002792397660818713,
|
| 112 |
+
"loss": 0.7127,
|
| 113 |
+
"step": 1300
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 3.51,
|
| 117 |
+
"learning_rate": 0.000275062656641604,
|
| 118 |
+
"loss": 0.7025,
|
| 119 |
+
"step": 1400
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 3.76,
|
| 123 |
+
"learning_rate": 0.00027088554720133666,
|
| 124 |
+
"loss": 0.7045,
|
| 125 |
+
"step": 1500
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 4.0,
|
| 129 |
+
"eval_accuracy": 0.8932038834951457,
|
| 130 |
+
"eval_loss": 0.3880767524242401,
|
| 131 |
+
"eval_runtime": 139.5385,
|
| 132 |
+
"eval_samples_per_second": 48.718,
|
| 133 |
+
"eval_steps_per_second": 1.526,
|
| 134 |
+
"step": 1596
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 4.01,
|
| 138 |
+
"learning_rate": 0.0002667084377610693,
|
| 139 |
+
"loss": 0.7082,
|
| 140 |
+
"step": 1600
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 4.26,
|
| 144 |
+
"learning_rate": 0.000262531328320802,
|
| 145 |
+
"loss": 0.6833,
|
| 146 |
+
"step": 1700
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 4.51,
|
| 150 |
+
"learning_rate": 0.00025835421888053466,
|
| 151 |
+
"loss": 0.6881,
|
| 152 |
+
"step": 1800
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 4.76,
|
| 156 |
+
"learning_rate": 0.0002541771094402673,
|
| 157 |
+
"loss": 0.659,
|
| 158 |
+
"step": 1900
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 5.0,
|
| 162 |
+
"eval_accuracy": 0.8846719623418653,
|
| 163 |
+
"eval_loss": 0.37056076526641846,
|
| 164 |
+
"eval_runtime": 141.2433,
|
| 165 |
+
"eval_samples_per_second": 48.13,
|
| 166 |
+
"eval_steps_per_second": 1.508,
|
| 167 |
+
"step": 1995
|
| 168 |
+
},
|
| 169 |
+
{
|
| 170 |
+
"epoch": 5.01,
|
| 171 |
+
"learning_rate": 0.00025,
|
| 172 |
+
"loss": 0.6621,
|
| 173 |
+
"step": 2000
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 5.26,
|
| 177 |
+
"learning_rate": 0.00024582289055973265,
|
| 178 |
+
"loss": 0.6643,
|
| 179 |
+
"step": 2100
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 5.51,
|
| 183 |
+
"learning_rate": 0.00024164578111946533,
|
| 184 |
+
"loss": 0.6421,
|
| 185 |
+
"step": 2200
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 5.76,
|
| 189 |
+
"learning_rate": 0.00023746867167919798,
|
| 190 |
+
"loss": 0.6541,
|
| 191 |
+
"step": 2300
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 6.0,
|
| 195 |
+
"eval_accuracy": 0.891732862606649,
|
| 196 |
+
"eval_loss": 0.3553307056427002,
|
| 197 |
+
"eval_runtime": 140.013,
|
| 198 |
+
"eval_samples_per_second": 48.553,
|
| 199 |
+
"eval_steps_per_second": 1.521,
|
| 200 |
+
"step": 2394
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 6.02,
|
| 204 |
+
"learning_rate": 0.00023329156223893065,
|
| 205 |
+
"loss": 0.6477,
|
| 206 |
+
"step": 2400
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 6.27,
|
| 210 |
+
"learning_rate": 0.00022911445279866333,
|
| 211 |
+
"loss": 0.651,
|
| 212 |
+
"step": 2500
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 6.52,
|
| 216 |
+
"learning_rate": 0.00022493734335839597,
|
| 217 |
+
"loss": 0.6509,
|
| 218 |
+
"step": 2600
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 6.77,
|
| 222 |
+
"learning_rate": 0.00022076023391812865,
|
| 223 |
+
"loss": 0.6448,
|
| 224 |
+
"step": 2700
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 7.0,
|
| 228 |
+
"eval_accuracy": 0.8952633127390409,
|
| 229 |
+
"eval_loss": 0.3482230305671692,
|
| 230 |
+
"eval_runtime": 140.3109,
|
| 231 |
+
"eval_samples_per_second": 48.45,
|
| 232 |
+
"eval_steps_per_second": 1.518,
|
| 233 |
+
"step": 2793
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 7.02,
|
| 237 |
+
"learning_rate": 0.00021658312447786132,
|
| 238 |
+
"loss": 0.6421,
|
| 239 |
+
"step": 2800
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 7.27,
|
| 243 |
+
"learning_rate": 0.00021240601503759397,
|
| 244 |
+
"loss": 0.6418,
|
| 245 |
+
"step": 2900
|
| 246 |
+
},
|
| 247 |
+
{
|
| 248 |
+
"epoch": 7.52,
|
| 249 |
+
"learning_rate": 0.00020822890559732665,
|
| 250 |
+
"loss": 0.6453,
|
| 251 |
+
"step": 3000
|
| 252 |
+
},
|
| 253 |
+
{
|
| 254 |
+
"epoch": 7.77,
|
| 255 |
+
"learning_rate": 0.00020405179615705932,
|
| 256 |
+
"loss": 0.6288,
|
| 257 |
+
"step": 3100
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 8.0,
|
| 261 |
+
"eval_accuracy": 0.8989408649602825,
|
| 262 |
+
"eval_loss": 0.34094032645225525,
|
| 263 |
+
"eval_runtime": 141.0541,
|
| 264 |
+
"eval_samples_per_second": 48.194,
|
| 265 |
+
"eval_steps_per_second": 1.51,
|
| 266 |
+
"step": 3192
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 8.02,
|
| 270 |
+
"learning_rate": 0.00019987468671679197,
|
| 271 |
+
"loss": 0.6336,
|
| 272 |
+
"step": 3200
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 8.27,
|
| 276 |
+
"learning_rate": 0.00019569757727652464,
|
| 277 |
+
"loss": 0.6441,
|
| 278 |
+
"step": 3300
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 8.52,
|
| 282 |
+
"learning_rate": 0.00019152046783625732,
|
| 283 |
+
"loss": 0.6557,
|
| 284 |
+
"step": 3400
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 8.77,
|
| 288 |
+
"learning_rate": 0.00018734335839598997,
|
| 289 |
+
"loss": 0.641,
|
| 290 |
+
"step": 3500
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 9.0,
|
| 294 |
+
"eval_accuracy": 0.9051191526919682,
|
| 295 |
+
"eval_loss": 0.3297290503978729,
|
| 296 |
+
"eval_runtime": 140.3281,
|
| 297 |
+
"eval_samples_per_second": 48.444,
|
| 298 |
+
"eval_steps_per_second": 1.518,
|
| 299 |
+
"step": 3591
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 9.02,
|
| 303 |
+
"learning_rate": 0.00018316624895572264,
|
| 304 |
+
"loss": 0.6411,
|
| 305 |
+
"step": 3600
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 9.27,
|
| 309 |
+
"learning_rate": 0.00017898913951545532,
|
| 310 |
+
"loss": 0.6455,
|
| 311 |
+
"step": 3700
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 9.52,
|
| 315 |
+
"learning_rate": 0.00017481203007518797,
|
| 316 |
+
"loss": 0.6464,
|
| 317 |
+
"step": 3800
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 9.77,
|
| 321 |
+
"learning_rate": 0.00017063492063492064,
|
| 322 |
+
"loss": 0.6369,
|
| 323 |
+
"step": 3900
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 10.0,
|
| 327 |
+
"eval_accuracy": 0.9042365401588702,
|
| 328 |
+
"eval_loss": 0.33252373337745667,
|
| 329 |
+
"eval_runtime": 140.3318,
|
| 330 |
+
"eval_samples_per_second": 48.442,
|
| 331 |
+
"eval_steps_per_second": 1.518,
|
| 332 |
+
"step": 3990
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 10.03,
|
| 336 |
+
"learning_rate": 0.0001664578111946533,
|
| 337 |
+
"loss": 0.6225,
|
| 338 |
+
"step": 4000
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 10.28,
|
| 342 |
+
"learning_rate": 0.00016228070175438596,
|
| 343 |
+
"loss": 0.6295,
|
| 344 |
+
"step": 4100
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 10.53,
|
| 348 |
+
"learning_rate": 0.0001581453634085213,
|
| 349 |
+
"loss": 0.6255,
|
| 350 |
+
"step": 4200
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 10.78,
|
| 354 |
+
"learning_rate": 0.00015396825396825397,
|
| 355 |
+
"loss": 0.6218,
|
| 356 |
+
"step": 4300
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 11.0,
|
| 360 |
+
"eval_accuracy": 0.9064430714916152,
|
| 361 |
+
"eval_loss": 0.3250460922718048,
|
| 362 |
+
"eval_runtime": 140.1683,
|
| 363 |
+
"eval_samples_per_second": 48.499,
|
| 364 |
+
"eval_steps_per_second": 1.52,
|
| 365 |
+
"step": 4389
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 11.03,
|
| 369 |
+
"learning_rate": 0.00014979114452798661,
|
| 370 |
+
"loss": 0.6507,
|
| 371 |
+
"step": 4400
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 11.28,
|
| 375 |
+
"learning_rate": 0.0001456140350877193,
|
| 376 |
+
"loss": 0.623,
|
| 377 |
+
"step": 4500
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 11.53,
|
| 381 |
+
"learning_rate": 0.00014143692564745194,
|
| 382 |
+
"loss": 0.6278,
|
| 383 |
+
"step": 4600
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 11.78,
|
| 387 |
+
"learning_rate": 0.0001372598162071846,
|
| 388 |
+
"loss": 0.6247,
|
| 389 |
+
"step": 4700
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 12.0,
|
| 393 |
+
"eval_accuracy": 0.8958517210944396,
|
| 394 |
+
"eval_loss": 0.3311932682991028,
|
| 395 |
+
"eval_runtime": 140.5537,
|
| 396 |
+
"eval_samples_per_second": 48.366,
|
| 397 |
+
"eval_steps_per_second": 1.515,
|
| 398 |
+
"step": 4788
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 12.03,
|
| 402 |
+
"learning_rate": 0.0001330827067669173,
|
| 403 |
+
"loss": 0.6149,
|
| 404 |
+
"step": 4800
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 12.28,
|
| 408 |
+
"learning_rate": 0.00012890559732664993,
|
| 409 |
+
"loss": 0.6244,
|
| 410 |
+
"step": 4900
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 12.53,
|
| 414 |
+
"learning_rate": 0.0001247284878863826,
|
| 415 |
+
"loss": 0.6294,
|
| 416 |
+
"step": 5000
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 12.78,
|
| 420 |
+
"learning_rate": 0.00012055137844611527,
|
| 421 |
+
"loss": 0.6284,
|
| 422 |
+
"step": 5100
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 13.0,
|
| 426 |
+
"eval_accuracy": 0.9068843777581642,
|
| 427 |
+
"eval_loss": 0.321657657623291,
|
| 428 |
+
"eval_runtime": 139.9572,
|
| 429 |
+
"eval_samples_per_second": 48.572,
|
| 430 |
+
"eval_steps_per_second": 1.522,
|
| 431 |
+
"step": 5187
|
| 432 |
+
},
|
| 433 |
+
{
|
| 434 |
+
"epoch": 13.03,
|
| 435 |
+
"learning_rate": 0.00011637426900584793,
|
| 436 |
+
"loss": 0.6317,
|
| 437 |
+
"step": 5200
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 13.28,
|
| 441 |
+
"learning_rate": 0.00011219715956558061,
|
| 442 |
+
"loss": 0.6397,
|
| 443 |
+
"step": 5300
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 13.53,
|
| 447 |
+
"learning_rate": 0.00010802005012531327,
|
| 448 |
+
"loss": 0.6041,
|
| 449 |
+
"step": 5400
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 13.78,
|
| 453 |
+
"learning_rate": 0.00010384294068504593,
|
| 454 |
+
"loss": 0.6213,
|
| 455 |
+
"step": 5500
|
| 456 |
+
},
|
| 457 |
+
{
|
| 458 |
+
"epoch": 14.0,
|
| 459 |
+
"eval_accuracy": 0.8977640482494852,
|
| 460 |
+
"eval_loss": 0.33012473583221436,
|
| 461 |
+
"eval_runtime": 139.8104,
|
| 462 |
+
"eval_samples_per_second": 48.623,
|
| 463 |
+
"eval_steps_per_second": 1.523,
|
| 464 |
+
"step": 5586
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 14.04,
|
| 468 |
+
"learning_rate": 9.96658312447786e-05,
|
| 469 |
+
"loss": 0.6206,
|
| 470 |
+
"step": 5600
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 14.29,
|
| 474 |
+
"learning_rate": 9.548872180451127e-05,
|
| 475 |
+
"loss": 0.6379,
|
| 476 |
+
"step": 5700
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 14.54,
|
| 480 |
+
"learning_rate": 9.131161236424393e-05,
|
| 481 |
+
"loss": 0.6269,
|
| 482 |
+
"step": 5800
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 14.79,
|
| 486 |
+
"learning_rate": 8.713450292397659e-05,
|
| 487 |
+
"loss": 0.6274,
|
| 488 |
+
"step": 5900
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 15.0,
|
| 492 |
+
"eval_accuracy": 0.9080611944689615,
|
| 493 |
+
"eval_loss": 0.31795457005500793,
|
| 494 |
+
"eval_runtime": 139.5988,
|
| 495 |
+
"eval_samples_per_second": 48.697,
|
| 496 |
+
"eval_steps_per_second": 1.526,
|
| 497 |
+
"step": 5985
|
| 498 |
+
},
|
| 499 |
+
{
|
| 500 |
+
"epoch": 15.04,
|
| 501 |
+
"learning_rate": 8.295739348370926e-05,
|
| 502 |
+
"loss": 0.6263,
|
| 503 |
+
"step": 6000
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"epoch": 15.29,
|
| 507 |
+
"learning_rate": 7.878028404344193e-05,
|
| 508 |
+
"loss": 0.6171,
|
| 509 |
+
"step": 6100
|
| 510 |
+
},
|
| 511 |
+
{
|
| 512 |
+
"epoch": 15.54,
|
| 513 |
+
"learning_rate": 7.460317460317459e-05,
|
| 514 |
+
"loss": 0.626,
|
| 515 |
+
"step": 6200
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 15.79,
|
| 519 |
+
"learning_rate": 7.042606516290726e-05,
|
| 520 |
+
"loss": 0.627,
|
| 521 |
+
"step": 6300
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 16.0,
|
| 525 |
+
"eval_accuracy": 0.9020300088261254,
|
| 526 |
+
"eval_loss": 0.32565683126449585,
|
| 527 |
+
"eval_runtime": 139.2707,
|
| 528 |
+
"eval_samples_per_second": 48.811,
|
| 529 |
+
"eval_steps_per_second": 1.529,
|
| 530 |
+
"step": 6384
|
| 531 |
+
},
|
| 532 |
+
{
|
| 533 |
+
"epoch": 16.04,
|
| 534 |
+
"learning_rate": 6.624895572263992e-05,
|
| 535 |
+
"loss": 0.6166,
|
| 536 |
+
"step": 6400
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 16.29,
|
| 540 |
+
"learning_rate": 6.207184628237259e-05,
|
| 541 |
+
"loss": 0.6146,
|
| 542 |
+
"step": 6500
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 16.54,
|
| 546 |
+
"learning_rate": 5.7894736842105253e-05,
|
| 547 |
+
"loss": 0.6126,
|
| 548 |
+
"step": 6600
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 16.79,
|
| 552 |
+
"learning_rate": 5.3759398496240595e-05,
|
| 553 |
+
"loss": 0.6227,
|
| 554 |
+
"step": 6700
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 17.0,
|
| 558 |
+
"eval_accuracy": 0.9055604589585172,
|
| 559 |
+
"eval_loss": 0.3193184435367584,
|
| 560 |
+
"eval_runtime": 140.199,
|
| 561 |
+
"eval_samples_per_second": 48.488,
|
| 562 |
+
"eval_steps_per_second": 1.519,
|
| 563 |
+
"step": 6783
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 17.04,
|
| 567 |
+
"learning_rate": 4.9582289055973256e-05,
|
| 568 |
+
"loss": 0.629,
|
| 569 |
+
"step": 6800
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 17.29,
|
| 573 |
+
"learning_rate": 4.5405179615705925e-05,
|
| 574 |
+
"loss": 0.6055,
|
| 575 |
+
"step": 6900
|
| 576 |
+
},
|
| 577 |
+
{
|
| 578 |
+
"epoch": 17.54,
|
| 579 |
+
"learning_rate": 4.122807017543859e-05,
|
| 580 |
+
"loss": 0.6088,
|
| 581 |
+
"step": 7000
|
| 582 |
+
},
|
| 583 |
+
{
|
| 584 |
+
"epoch": 17.79,
|
| 585 |
+
"learning_rate": 3.7050960735171254e-05,
|
| 586 |
+
"loss": 0.6192,
|
| 587 |
+
"step": 7100
|
| 588 |
+
},
|
| 589 |
+
{
|
| 590 |
+
"epoch": 18.0,
|
| 591 |
+
"eval_accuracy": 0.9065901735804649,
|
| 592 |
+
"eval_loss": 0.31994640827178955,
|
| 593 |
+
"eval_runtime": 139.8576,
|
| 594 |
+
"eval_samples_per_second": 48.607,
|
| 595 |
+
"eval_steps_per_second": 1.523,
|
| 596 |
+
"step": 7182
|
| 597 |
+
},
|
| 598 |
+
{
|
| 599 |
+
"epoch": 18.05,
|
| 600 |
+
"learning_rate": 3.287385129490392e-05,
|
| 601 |
+
"loss": 0.624,
|
| 602 |
+
"step": 7200
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 18.3,
|
| 606 |
+
"learning_rate": 2.8696741854636587e-05,
|
| 607 |
+
"loss": 0.6102,
|
| 608 |
+
"step": 7300
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 18.55,
|
| 612 |
+
"learning_rate": 2.4519632414369252e-05,
|
| 613 |
+
"loss": 0.6261,
|
| 614 |
+
"step": 7400
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"epoch": 18.8,
|
| 618 |
+
"learning_rate": 2.0342522974101917e-05,
|
| 619 |
+
"loss": 0.6075,
|
| 620 |
+
"step": 7500
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"epoch": 19.0,
|
| 624 |
+
"eval_accuracy": 0.9073256840247131,
|
| 625 |
+
"eval_loss": 0.3183441460132599,
|
| 626 |
+
"eval_runtime": 145.1619,
|
| 627 |
+
"eval_samples_per_second": 46.83,
|
| 628 |
+
"eval_steps_per_second": 1.467,
|
| 629 |
+
"step": 7581
|
| 630 |
+
},
|
| 631 |
+
{
|
| 632 |
+
"epoch": 19.05,
|
| 633 |
+
"learning_rate": 1.6165413533834585e-05,
|
| 634 |
+
"loss": 0.6066,
|
| 635 |
+
"step": 7600
|
| 636 |
+
},
|
| 637 |
+
{
|
| 638 |
+
"epoch": 19.3,
|
| 639 |
+
"learning_rate": 1.198830409356725e-05,
|
| 640 |
+
"loss": 0.6057,
|
| 641 |
+
"step": 7700
|
| 642 |
+
},
|
| 643 |
+
{
|
| 644 |
+
"epoch": 19.55,
|
| 645 |
+
"learning_rate": 7.811194653299915e-06,
|
| 646 |
+
"loss": 0.6276,
|
| 647 |
+
"step": 7800
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 19.8,
|
| 651 |
+
"learning_rate": 3.634085213032581e-06,
|
| 652 |
+
"loss": 0.6196,
|
| 653 |
+
"step": 7900
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 20.0,
|
| 657 |
+
"eval_accuracy": 0.9087967049132097,
|
| 658 |
+
"eval_loss": 0.3166075348854065,
|
| 659 |
+
"eval_runtime": 146.0633,
|
| 660 |
+
"eval_samples_per_second": 46.541,
|
| 661 |
+
"eval_steps_per_second": 1.458,
|
| 662 |
+
"step": 7980
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 20.0,
|
| 666 |
+
"step": 7980,
|
| 667 |
+
"total_flos": 0.0,
|
| 668 |
+
"train_loss": 0.8469889265552798,
|
| 669 |
+
"train_runtime": 24285.4703,
|
| 670 |
+
"train_samples_per_second": 42.078,
|
| 671 |
+
"train_steps_per_second": 0.329
|
| 672 |
+
}
|
| 673 |
+
],
|
| 674 |
+
"max_steps": 7980,
|
| 675 |
+
"num_train_epochs": 20,
|
| 676 |
+
"total_flos": 0.0,
|
| 677 |
+
"trial_name": null,
|
| 678 |
+
"trial_params": null
|
| 679 |
+
}
|