End of training
Browse files
all_results.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"eval_accuracy": 0.9040894380700206,
|
| 4 |
+
"eval_loss": 0.3562973439693451,
|
| 5 |
+
"eval_runtime": 43.2189,
|
| 6 |
+
"eval_samples_per_second": 157.292,
|
| 7 |
+
"eval_steps_per_second": 2.476,
|
| 8 |
+
"train_loss": 0.8915880279732229,
|
| 9 |
+
"train_runtime": 4529.6112,
|
| 10 |
+
"train_samples_per_second": 112.8,
|
| 11 |
+
"train_steps_per_second": 1.762
|
| 12 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"eval_accuracy": 0.9040894380700206,
|
| 4 |
+
"eval_loss": 0.3562973439693451,
|
| 5 |
+
"eval_runtime": 43.2189,
|
| 6 |
+
"eval_samples_per_second": 157.292,
|
| 7 |
+
"eval_steps_per_second": 2.476
|
| 8 |
+
}
|
runs/Nov28_21-42-20_3367f5396cf0/events.out.tfevents.1669676543.3367f5396cf0.230.2
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:10d95b9c1b6f2e084f9dfb4a0f6485debaaa30ce115712a53d568f423f070d86
|
| 3 |
+
size 363
|
train_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"train_loss": 0.8915880279732229,
|
| 4 |
+
"train_runtime": 4529.6112,
|
| 5 |
+
"train_samples_per_second": 112.8,
|
| 6 |
+
"train_steps_per_second": 1.762
|
| 7 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,589 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.9040894380700206,
|
| 3 |
+
"best_model_checkpoint": "trillsson3-ft-keyword-spotting/checkpoint-3192",
|
| 4 |
+
"epoch": 9.999373825923607,
|
| 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": 0.13,
|
| 12 |
+
"learning_rate": 3.533834586466165e-05,
|
| 13 |
+
"loss": 7.4799,
|
| 14 |
+
"step": 100
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.25,
|
| 18 |
+
"learning_rate": 7.293233082706766e-05,
|
| 19 |
+
"loss": 4.7288,
|
| 20 |
+
"step": 200
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.38,
|
| 24 |
+
"learning_rate": 0.00011052631578947366,
|
| 25 |
+
"loss": 3.0455,
|
| 26 |
+
"step": 300
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.5,
|
| 30 |
+
"learning_rate": 0.00014812030075187968,
|
| 31 |
+
"loss": 2.3495,
|
| 32 |
+
"step": 400
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.63,
|
| 36 |
+
"learning_rate": 0.00018571428571428572,
|
| 37 |
+
"loss": 1.836,
|
| 38 |
+
"step": 500
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.75,
|
| 42 |
+
"learning_rate": 0.00022330827067669172,
|
| 43 |
+
"loss": 1.4179,
|
| 44 |
+
"step": 600
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.88,
|
| 48 |
+
"learning_rate": 0.0002609022556390977,
|
| 49 |
+
"loss": 1.1824,
|
| 50 |
+
"step": 700
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 1.0,
|
| 54 |
+
"eval_accuracy": 0.7488967343336276,
|
| 55 |
+
"eval_loss": 0.6477929949760437,
|
| 56 |
+
"eval_runtime": 42.5234,
|
| 57 |
+
"eval_samples_per_second": 159.865,
|
| 58 |
+
"eval_steps_per_second": 2.516,
|
| 59 |
+
"step": 798
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 1.0,
|
| 63 |
+
"learning_rate": 0.0002984962406015037,
|
| 64 |
+
"loss": 1.0261,
|
| 65 |
+
"step": 800
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 1.13,
|
| 69 |
+
"learning_rate": 0.00029598997493734334,
|
| 70 |
+
"loss": 0.9596,
|
| 71 |
+
"step": 900
|
| 72 |
+
},
|
| 73 |
+
{
|
| 74 |
+
"epoch": 1.25,
|
| 75 |
+
"learning_rate": 0.000291812865497076,
|
| 76 |
+
"loss": 0.8732,
|
| 77 |
+
"step": 1000
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 1.38,
|
| 81 |
+
"learning_rate": 0.00028763575605680864,
|
| 82 |
+
"loss": 0.8509,
|
| 83 |
+
"step": 1100
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 1.5,
|
| 87 |
+
"learning_rate": 0.00028345864661654134,
|
| 88 |
+
"loss": 0.8162,
|
| 89 |
+
"step": 1200
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 1.63,
|
| 93 |
+
"learning_rate": 0.000279281537176274,
|
| 94 |
+
"loss": 0.7603,
|
| 95 |
+
"step": 1300
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 1.75,
|
| 99 |
+
"learning_rate": 0.00027510442773600664,
|
| 100 |
+
"loss": 0.7638,
|
| 101 |
+
"step": 1400
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 1.88,
|
| 105 |
+
"learning_rate": 0.00027092731829573934,
|
| 106 |
+
"loss": 0.7448,
|
| 107 |
+
"step": 1500
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 2.0,
|
| 111 |
+
"eval_accuracy": 0.8727566931450427,
|
| 112 |
+
"eval_loss": 0.4273872375488281,
|
| 113 |
+
"eval_runtime": 42.1254,
|
| 114 |
+
"eval_samples_per_second": 161.375,
|
| 115 |
+
"eval_steps_per_second": 2.54,
|
| 116 |
+
"step": 1596
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 2.01,
|
| 120 |
+
"learning_rate": 0.000266750208855472,
|
| 121 |
+
"loss": 0.7268,
|
| 122 |
+
"step": 1600
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 2.13,
|
| 126 |
+
"learning_rate": 0.00026257309941520463,
|
| 127 |
+
"loss": 0.7503,
|
| 128 |
+
"step": 1700
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 2.26,
|
| 132 |
+
"learning_rate": 0.00025839598997493734,
|
| 133 |
+
"loss": 0.719,
|
| 134 |
+
"step": 1800
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 2.38,
|
| 138 |
+
"learning_rate": 0.00025421888053467,
|
| 139 |
+
"loss": 0.7174,
|
| 140 |
+
"step": 1900
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 2.51,
|
| 144 |
+
"learning_rate": 0.00025004177109440263,
|
| 145 |
+
"loss": 0.6869,
|
| 146 |
+
"step": 2000
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 2.63,
|
| 150 |
+
"learning_rate": 0.00024586466165413533,
|
| 151 |
+
"loss": 0.7113,
|
| 152 |
+
"step": 2100
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 2.76,
|
| 156 |
+
"learning_rate": 0.00024168755221386798,
|
| 157 |
+
"loss": 0.7011,
|
| 158 |
+
"step": 2200
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 2.88,
|
| 162 |
+
"learning_rate": 0.00023751044277360066,
|
| 163 |
+
"loss": 0.7089,
|
| 164 |
+
"step": 2300
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 3.0,
|
| 168 |
+
"eval_accuracy": 0.8949691085613416,
|
| 169 |
+
"eval_loss": 0.3723289370536804,
|
| 170 |
+
"eval_runtime": 41.1393,
|
| 171 |
+
"eval_samples_per_second": 165.243,
|
| 172 |
+
"eval_steps_per_second": 2.601,
|
| 173 |
+
"step": 2394
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 3.01,
|
| 177 |
+
"learning_rate": 0.0002333333333333333,
|
| 178 |
+
"loss": 0.6834,
|
| 179 |
+
"step": 2400
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 3.13,
|
| 183 |
+
"learning_rate": 0.00022915622389306598,
|
| 184 |
+
"loss": 0.6804,
|
| 185 |
+
"step": 2500
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 3.26,
|
| 189 |
+
"learning_rate": 0.00022497911445279865,
|
| 190 |
+
"loss": 0.6771,
|
| 191 |
+
"step": 2600
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 3.38,
|
| 195 |
+
"learning_rate": 0.0002208020050125313,
|
| 196 |
+
"loss": 0.684,
|
| 197 |
+
"step": 2700
|
| 198 |
+
},
|
| 199 |
+
{
|
| 200 |
+
"epoch": 3.51,
|
| 201 |
+
"learning_rate": 0.00021662489557226398,
|
| 202 |
+
"loss": 0.6636,
|
| 203 |
+
"step": 2800
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 3.63,
|
| 207 |
+
"learning_rate": 0.00021244778613199665,
|
| 208 |
+
"loss": 0.7003,
|
| 209 |
+
"step": 2900
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 3.76,
|
| 213 |
+
"learning_rate": 0.0002082706766917293,
|
| 214 |
+
"loss": 0.6594,
|
| 215 |
+
"step": 3000
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 3.88,
|
| 219 |
+
"learning_rate": 0.00020409356725146197,
|
| 220 |
+
"loss": 0.6781,
|
| 221 |
+
"step": 3100
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 4.0,
|
| 225 |
+
"eval_accuracy": 0.9040894380700206,
|
| 226 |
+
"eval_loss": 0.3562973439693451,
|
| 227 |
+
"eval_runtime": 42.576,
|
| 228 |
+
"eval_samples_per_second": 159.667,
|
| 229 |
+
"eval_steps_per_second": 2.513,
|
| 230 |
+
"step": 3192
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 4.01,
|
| 234 |
+
"learning_rate": 0.00019991645781119465,
|
| 235 |
+
"loss": 0.6921,
|
| 236 |
+
"step": 3200
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 4.14,
|
| 240 |
+
"learning_rate": 0.0001957393483709273,
|
| 241 |
+
"loss": 0.6615,
|
| 242 |
+
"step": 3300
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 4.26,
|
| 246 |
+
"learning_rate": 0.00019156223893065997,
|
| 247 |
+
"loss": 0.6707,
|
| 248 |
+
"step": 3400
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 4.39,
|
| 252 |
+
"learning_rate": 0.00018738512949039265,
|
| 253 |
+
"loss": 0.6511,
|
| 254 |
+
"step": 3500
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 4.51,
|
| 258 |
+
"learning_rate": 0.0001832080200501253,
|
| 259 |
+
"loss": 0.6974,
|
| 260 |
+
"step": 3600
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 4.64,
|
| 264 |
+
"learning_rate": 0.00017903091060985797,
|
| 265 |
+
"loss": 0.6447,
|
| 266 |
+
"step": 3700
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 4.76,
|
| 270 |
+
"learning_rate": 0.00017485380116959065,
|
| 271 |
+
"loss": 0.643,
|
| 272 |
+
"step": 3800
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 4.89,
|
| 276 |
+
"learning_rate": 0.0001706766917293233,
|
| 277 |
+
"loss": 0.6386,
|
| 278 |
+
"step": 3900
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 5.0,
|
| 282 |
+
"eval_accuracy": 0.8986466607825832,
|
| 283 |
+
"eval_loss": 0.3440994322299957,
|
| 284 |
+
"eval_runtime": 42.3928,
|
| 285 |
+
"eval_samples_per_second": 160.358,
|
| 286 |
+
"eval_steps_per_second": 2.524,
|
| 287 |
+
"step": 3990
|
| 288 |
+
},
|
| 289 |
+
{
|
| 290 |
+
"epoch": 5.01,
|
| 291 |
+
"learning_rate": 0.00016649958228905597,
|
| 292 |
+
"loss": 0.6611,
|
| 293 |
+
"step": 4000
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"epoch": 5.14,
|
| 297 |
+
"learning_rate": 0.0001623642439431913,
|
| 298 |
+
"loss": 0.6634,
|
| 299 |
+
"step": 4100
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 5.26,
|
| 303 |
+
"learning_rate": 0.00015818713450292397,
|
| 304 |
+
"loss": 0.6419,
|
| 305 |
+
"step": 4200
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 5.39,
|
| 309 |
+
"learning_rate": 0.00015401002506265662,
|
| 310 |
+
"loss": 0.6341,
|
| 311 |
+
"step": 4300
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 5.51,
|
| 315 |
+
"learning_rate": 0.0001498329156223893,
|
| 316 |
+
"loss": 0.6283,
|
| 317 |
+
"step": 4400
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 5.64,
|
| 321 |
+
"learning_rate": 0.00014565580618212197,
|
| 322 |
+
"loss": 0.6551,
|
| 323 |
+
"step": 4500
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 5.76,
|
| 327 |
+
"learning_rate": 0.00014147869674185462,
|
| 328 |
+
"loss": 0.6318,
|
| 329 |
+
"step": 4600
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 5.89,
|
| 333 |
+
"learning_rate": 0.0001373015873015873,
|
| 334 |
+
"loss": 0.6342,
|
| 335 |
+
"step": 4700
|
| 336 |
+
},
|
| 337 |
+
{
|
| 338 |
+
"epoch": 6.0,
|
| 339 |
+
"eval_accuracy": 0.8993821712268314,
|
| 340 |
+
"eval_loss": 0.3380272388458252,
|
| 341 |
+
"eval_runtime": 42.8693,
|
| 342 |
+
"eval_samples_per_second": 158.575,
|
| 343 |
+
"eval_steps_per_second": 2.496,
|
| 344 |
+
"step": 4788
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 6.02,
|
| 348 |
+
"learning_rate": 0.00013312447786131997,
|
| 349 |
+
"loss": 0.6371,
|
| 350 |
+
"step": 4800
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 6.14,
|
| 354 |
+
"learning_rate": 0.00012894736842105261,
|
| 355 |
+
"loss": 0.6484,
|
| 356 |
+
"step": 4900
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 6.27,
|
| 360 |
+
"learning_rate": 0.0001247702589807853,
|
| 361 |
+
"loss": 0.6331,
|
| 362 |
+
"step": 5000
|
| 363 |
+
},
|
| 364 |
+
{
|
| 365 |
+
"epoch": 6.39,
|
| 366 |
+
"learning_rate": 0.00012059314954051795,
|
| 367 |
+
"loss": 0.6201,
|
| 368 |
+
"step": 5100
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 6.52,
|
| 372 |
+
"learning_rate": 0.00011641604010025061,
|
| 373 |
+
"loss": 0.6619,
|
| 374 |
+
"step": 5200
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 6.64,
|
| 378 |
+
"learning_rate": 0.00011223893065998329,
|
| 379 |
+
"loss": 0.6337,
|
| 380 |
+
"step": 5300
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 6.77,
|
| 384 |
+
"learning_rate": 0.00010806182121971595,
|
| 385 |
+
"loss": 0.6356,
|
| 386 |
+
"step": 5400
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 6.89,
|
| 390 |
+
"learning_rate": 0.00010388471177944861,
|
| 391 |
+
"loss": 0.6275,
|
| 392 |
+
"step": 5500
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 7.0,
|
| 396 |
+
"eval_accuracy": 0.8982053545160341,
|
| 397 |
+
"eval_loss": 0.33755984902381897,
|
| 398 |
+
"eval_runtime": 42.4768,
|
| 399 |
+
"eval_samples_per_second": 160.04,
|
| 400 |
+
"eval_steps_per_second": 2.519,
|
| 401 |
+
"step": 5586
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 7.02,
|
| 405 |
+
"learning_rate": 9.970760233918129e-05,
|
| 406 |
+
"loss": 0.6359,
|
| 407 |
+
"step": 5600
|
| 408 |
+
},
|
| 409 |
+
{
|
| 410 |
+
"epoch": 7.14,
|
| 411 |
+
"learning_rate": 9.553049289891395e-05,
|
| 412 |
+
"loss": 0.6502,
|
| 413 |
+
"step": 5700
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 7.27,
|
| 417 |
+
"learning_rate": 9.135338345864661e-05,
|
| 418 |
+
"loss": 0.6138,
|
| 419 |
+
"step": 5800
|
| 420 |
+
},
|
| 421 |
+
{
|
| 422 |
+
"epoch": 7.39,
|
| 423 |
+
"learning_rate": 8.717627401837928e-05,
|
| 424 |
+
"loss": 0.6309,
|
| 425 |
+
"step": 5900
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 7.52,
|
| 429 |
+
"learning_rate": 8.299916457811194e-05,
|
| 430 |
+
"loss": 0.6386,
|
| 431 |
+
"step": 6000
|
| 432 |
+
},
|
| 433 |
+
{
|
| 434 |
+
"epoch": 7.64,
|
| 435 |
+
"learning_rate": 7.88220551378446e-05,
|
| 436 |
+
"loss": 0.6149,
|
| 437 |
+
"step": 6100
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 7.77,
|
| 441 |
+
"learning_rate": 7.464494569757727e-05,
|
| 442 |
+
"loss": 0.6218,
|
| 443 |
+
"step": 6200
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 7.89,
|
| 447 |
+
"learning_rate": 7.050960735171261e-05,
|
| 448 |
+
"loss": 0.6349,
|
| 449 |
+
"step": 6300
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 8.0,
|
| 453 |
+
"eval_accuracy": 0.9014416004707266,
|
| 454 |
+
"eval_loss": 0.3333294689655304,
|
| 455 |
+
"eval_runtime": 43.296,
|
| 456 |
+
"eval_samples_per_second": 157.012,
|
| 457 |
+
"eval_steps_per_second": 2.471,
|
| 458 |
+
"step": 6384
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 8.02,
|
| 462 |
+
"learning_rate": 6.633249791144527e-05,
|
| 463 |
+
"loss": 0.6115,
|
| 464 |
+
"step": 6400
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 8.15,
|
| 468 |
+
"learning_rate": 6.215538847117793e-05,
|
| 469 |
+
"loss": 0.6463,
|
| 470 |
+
"step": 6500
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 8.27,
|
| 474 |
+
"learning_rate": 5.79782790309106e-05,
|
| 475 |
+
"loss": 0.6183,
|
| 476 |
+
"step": 6600
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 8.4,
|
| 480 |
+
"learning_rate": 5.380116959064327e-05,
|
| 481 |
+
"loss": 0.6475,
|
| 482 |
+
"step": 6700
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 8.52,
|
| 486 |
+
"learning_rate": 4.962406015037593e-05,
|
| 487 |
+
"loss": 0.6437,
|
| 488 |
+
"step": 6800
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 8.65,
|
| 492 |
+
"learning_rate": 4.54469507101086e-05,
|
| 493 |
+
"loss": 0.6265,
|
| 494 |
+
"step": 6900
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 8.77,
|
| 498 |
+
"learning_rate": 4.1269841269841266e-05,
|
| 499 |
+
"loss": 0.6326,
|
| 500 |
+
"step": 7000
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 8.9,
|
| 504 |
+
"learning_rate": 3.709273182957393e-05,
|
| 505 |
+
"loss": 0.6261,
|
| 506 |
+
"step": 7100
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 9.0,
|
| 510 |
+
"eval_accuracy": 0.9024713150926743,
|
| 511 |
+
"eval_loss": 0.32949280738830566,
|
| 512 |
+
"eval_runtime": 42.5233,
|
| 513 |
+
"eval_samples_per_second": 159.865,
|
| 514 |
+
"eval_steps_per_second": 2.516,
|
| 515 |
+
"step": 7182
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 9.02,
|
| 519 |
+
"learning_rate": 3.2915622389306596e-05,
|
| 520 |
+
"loss": 0.6306,
|
| 521 |
+
"step": 7200
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 9.15,
|
| 525 |
+
"learning_rate": 2.873851294903926e-05,
|
| 526 |
+
"loss": 0.6362,
|
| 527 |
+
"step": 7300
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 9.27,
|
| 531 |
+
"learning_rate": 2.4561403508771925e-05,
|
| 532 |
+
"loss": 0.6318,
|
| 533 |
+
"step": 7400
|
| 534 |
+
},
|
| 535 |
+
{
|
| 536 |
+
"epoch": 9.4,
|
| 537 |
+
"learning_rate": 2.0384294068504594e-05,
|
| 538 |
+
"loss": 0.6506,
|
| 539 |
+
"step": 7500
|
| 540 |
+
},
|
| 541 |
+
{
|
| 542 |
+
"epoch": 9.52,
|
| 543 |
+
"learning_rate": 1.620718462823726e-05,
|
| 544 |
+
"loss": 0.6202,
|
| 545 |
+
"step": 7600
|
| 546 |
+
},
|
| 547 |
+
{
|
| 548 |
+
"epoch": 9.65,
|
| 549 |
+
"learning_rate": 1.2030075187969923e-05,
|
| 550 |
+
"loss": 0.6217,
|
| 551 |
+
"step": 7700
|
| 552 |
+
},
|
| 553 |
+
{
|
| 554 |
+
"epoch": 9.77,
|
| 555 |
+
"learning_rate": 7.852965747702588e-06,
|
| 556 |
+
"loss": 0.6307,
|
| 557 |
+
"step": 7800
|
| 558 |
+
},
|
| 559 |
+
{
|
| 560 |
+
"epoch": 9.9,
|
| 561 |
+
"learning_rate": 3.6758563074352545e-06,
|
| 562 |
+
"loss": 0.6188,
|
| 563 |
+
"step": 7900
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 10.0,
|
| 567 |
+
"eval_accuracy": 0.9024713150926743,
|
| 568 |
+
"eval_loss": 0.33220550417900085,
|
| 569 |
+
"eval_runtime": 42.0746,
|
| 570 |
+
"eval_samples_per_second": 161.57,
|
| 571 |
+
"eval_steps_per_second": 2.543,
|
| 572 |
+
"step": 7980
|
| 573 |
+
},
|
| 574 |
+
{
|
| 575 |
+
"epoch": 10.0,
|
| 576 |
+
"step": 7980,
|
| 577 |
+
"total_flos": 0.0,
|
| 578 |
+
"train_loss": 0.8915880279732229,
|
| 579 |
+
"train_runtime": 4529.6112,
|
| 580 |
+
"train_samples_per_second": 112.8,
|
| 581 |
+
"train_steps_per_second": 1.762
|
| 582 |
+
}
|
| 583 |
+
],
|
| 584 |
+
"max_steps": 7980,
|
| 585 |
+
"num_train_epochs": 10,
|
| 586 |
+
"total_flos": 0.0,
|
| 587 |
+
"trial_name": null,
|
| 588 |
+
"trial_params": null
|
| 589 |
+
}
|