End of training
Browse files- all_results.json +10 -0
- test_results.json +10 -0
- trainer_state.json +522 -0
all_results.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 19.05,
|
| 3 |
+
"eval_accuracy": 0.8932038834951457,
|
| 4 |
+
"eval_accuracy_gunit": 0.5416666666666666,
|
| 5 |
+
"eval_accuracy_nothing": 1.0,
|
| 6 |
+
"eval_loss": 0.3490257263183594,
|
| 7 |
+
"eval_runtime": 13.7098,
|
| 8 |
+
"eval_samples_per_second": 7.513,
|
| 9 |
+
"eval_steps_per_second": 0.948
|
| 10 |
+
}
|
test_results.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 19.05,
|
| 3 |
+
"eval_accuracy": 0.8932038834951457,
|
| 4 |
+
"eval_accuracy_gunit": 0.5416666666666666,
|
| 5 |
+
"eval_accuracy_nothing": 1.0,
|
| 6 |
+
"eval_loss": 0.3490257263183594,
|
| 7 |
+
"eval_runtime": 13.7098,
|
| 8 |
+
"eval_samples_per_second": 7.513,
|
| 9 |
+
"eval_steps_per_second": 0.948
|
| 10 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,522 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.9504132231404959,
|
| 3 |
+
"best_model_checkpoint": "videomae-base-finetuned-good-gestureUnitsV3/checkpoint-255",
|
| 4 |
+
"epoch": 19.05,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 340,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.029411764705882353,
|
| 13 |
+
"grad_norm": 9.534835815429688,
|
| 14 |
+
"learning_rate": 1.4705882352941177e-06,
|
| 15 |
+
"loss": 0.7047,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.05,
|
| 20 |
+
"eval_accuracy": 0.5619834710743802,
|
| 21 |
+
"eval_accuracy_gunit": 0.575,
|
| 22 |
+
"eval_accuracy_nothing": 0.5555555555555556,
|
| 23 |
+
"eval_loss": 0.7023928165435791,
|
| 24 |
+
"eval_runtime": 15.7256,
|
| 25 |
+
"eval_samples_per_second": 7.694,
|
| 26 |
+
"eval_steps_per_second": 1.017,
|
| 27 |
+
"step": 17
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 1.0088235294117647,
|
| 31 |
+
"grad_norm": 5.451850414276123,
|
| 32 |
+
"learning_rate": 2.9411764705882355e-06,
|
| 33 |
+
"loss": 0.7083,
|
| 34 |
+
"step": 20
|
| 35 |
+
},
|
| 36 |
+
{
|
| 37 |
+
"epoch": 1.0382352941176471,
|
| 38 |
+
"grad_norm": 5.070091724395752,
|
| 39 |
+
"learning_rate": 4.411764705882353e-06,
|
| 40 |
+
"loss": 0.6958,
|
| 41 |
+
"step": 30
|
| 42 |
+
},
|
| 43 |
+
{
|
| 44 |
+
"epoch": 1.05,
|
| 45 |
+
"eval_accuracy": 0.38016528925619836,
|
| 46 |
+
"eval_accuracy_gunit": 0.725,
|
| 47 |
+
"eval_accuracy_nothing": 0.20987654320987653,
|
| 48 |
+
"eval_loss": 0.7165102362632751,
|
| 49 |
+
"eval_runtime": 15.8798,
|
| 50 |
+
"eval_samples_per_second": 7.62,
|
| 51 |
+
"eval_steps_per_second": 1.008,
|
| 52 |
+
"step": 34
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 2.0176470588235293,
|
| 56 |
+
"grad_norm": 4.038697719573975,
|
| 57 |
+
"learning_rate": 4.901960784313726e-06,
|
| 58 |
+
"loss": 0.6585,
|
| 59 |
+
"step": 40
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 2.0470588235294116,
|
| 63 |
+
"grad_norm": 3.2235519886016846,
|
| 64 |
+
"learning_rate": 4.7385620915032685e-06,
|
| 65 |
+
"loss": 0.6534,
|
| 66 |
+
"step": 50
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 2.05,
|
| 70 |
+
"eval_accuracy": 0.371900826446281,
|
| 71 |
+
"eval_accuracy_gunit": 0.85,
|
| 72 |
+
"eval_accuracy_nothing": 0.13580246913580246,
|
| 73 |
+
"eval_loss": 0.7150475382804871,
|
| 74 |
+
"eval_runtime": 16.0184,
|
| 75 |
+
"eval_samples_per_second": 7.554,
|
| 76 |
+
"eval_steps_per_second": 0.999,
|
| 77 |
+
"step": 51
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 3.026470588235294,
|
| 81 |
+
"grad_norm": 4.749432563781738,
|
| 82 |
+
"learning_rate": 4.5751633986928105e-06,
|
| 83 |
+
"loss": 0.655,
|
| 84 |
+
"step": 60
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 3.05,
|
| 88 |
+
"eval_accuracy": 0.32231404958677684,
|
| 89 |
+
"eval_accuracy_gunit": 0.975,
|
| 90 |
+
"eval_accuracy_nothing": 0.0,
|
| 91 |
+
"eval_loss": 0.7505635619163513,
|
| 92 |
+
"eval_runtime": 15.7625,
|
| 93 |
+
"eval_samples_per_second": 7.676,
|
| 94 |
+
"eval_steps_per_second": 1.015,
|
| 95 |
+
"step": 68
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 4.0058823529411764,
|
| 99 |
+
"grad_norm": 6.722472190856934,
|
| 100 |
+
"learning_rate": 4.411764705882353e-06,
|
| 101 |
+
"loss": 0.6741,
|
| 102 |
+
"step": 70
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 4.035294117647059,
|
| 106 |
+
"grad_norm": 6.8286638259887695,
|
| 107 |
+
"learning_rate": 4.2483660130718954e-06,
|
| 108 |
+
"loss": 0.6545,
|
| 109 |
+
"step": 80
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 4.05,
|
| 113 |
+
"eval_accuracy": 0.4297520661157025,
|
| 114 |
+
"eval_accuracy_gunit": 0.95,
|
| 115 |
+
"eval_accuracy_nothing": 0.1728395061728395,
|
| 116 |
+
"eval_loss": 0.6991418600082397,
|
| 117 |
+
"eval_runtime": 15.5171,
|
| 118 |
+
"eval_samples_per_second": 7.798,
|
| 119 |
+
"eval_steps_per_second": 1.031,
|
| 120 |
+
"step": 85
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 5.014705882352941,
|
| 124 |
+
"grad_norm": 7.191628456115723,
|
| 125 |
+
"learning_rate": 4.084967320261438e-06,
|
| 126 |
+
"loss": 0.56,
|
| 127 |
+
"step": 90
|
| 128 |
+
},
|
| 129 |
+
{
|
| 130 |
+
"epoch": 5.044117647058823,
|
| 131 |
+
"grad_norm": 5.416815757751465,
|
| 132 |
+
"learning_rate": 3.92156862745098e-06,
|
| 133 |
+
"loss": 0.6321,
|
| 134 |
+
"step": 100
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 5.05,
|
| 138 |
+
"eval_accuracy": 0.47107438016528924,
|
| 139 |
+
"eval_accuracy_gunit": 0.95,
|
| 140 |
+
"eval_accuracy_nothing": 0.2345679012345679,
|
| 141 |
+
"eval_loss": 0.6750265955924988,
|
| 142 |
+
"eval_runtime": 15.9488,
|
| 143 |
+
"eval_samples_per_second": 7.587,
|
| 144 |
+
"eval_steps_per_second": 1.003,
|
| 145 |
+
"step": 102
|
| 146 |
+
},
|
| 147 |
+
{
|
| 148 |
+
"epoch": 6.023529411764706,
|
| 149 |
+
"grad_norm": 4.386753559112549,
|
| 150 |
+
"learning_rate": 3.758169934640523e-06,
|
| 151 |
+
"loss": 0.5883,
|
| 152 |
+
"step": 110
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 6.05,
|
| 156 |
+
"eval_accuracy": 0.743801652892562,
|
| 157 |
+
"eval_accuracy_gunit": 0.9,
|
| 158 |
+
"eval_accuracy_nothing": 0.6666666666666666,
|
| 159 |
+
"eval_loss": 0.6142133474349976,
|
| 160 |
+
"eval_runtime": 16.6481,
|
| 161 |
+
"eval_samples_per_second": 7.268,
|
| 162 |
+
"eval_steps_per_second": 0.961,
|
| 163 |
+
"step": 119
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 7.002941176470588,
|
| 167 |
+
"grad_norm": 5.5603108406066895,
|
| 168 |
+
"learning_rate": 3.5947712418300657e-06,
|
| 169 |
+
"loss": 0.5803,
|
| 170 |
+
"step": 120
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 7.0323529411764705,
|
| 174 |
+
"grad_norm": 3.9502112865448,
|
| 175 |
+
"learning_rate": 3.431372549019608e-06,
|
| 176 |
+
"loss": 0.5045,
|
| 177 |
+
"step": 130
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 7.05,
|
| 181 |
+
"eval_accuracy": 0.8181818181818182,
|
| 182 |
+
"eval_accuracy_gunit": 0.9,
|
| 183 |
+
"eval_accuracy_nothing": 0.7777777777777778,
|
| 184 |
+
"eval_loss": 0.549534022808075,
|
| 185 |
+
"eval_runtime": 15.8267,
|
| 186 |
+
"eval_samples_per_second": 7.645,
|
| 187 |
+
"eval_steps_per_second": 1.011,
|
| 188 |
+
"step": 136
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 8.011764705882353,
|
| 192 |
+
"grad_norm": 5.036988735198975,
|
| 193 |
+
"learning_rate": 3.2679738562091506e-06,
|
| 194 |
+
"loss": 0.578,
|
| 195 |
+
"step": 140
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 8.041176470588235,
|
| 199 |
+
"grad_norm": 4.750277519226074,
|
| 200 |
+
"learning_rate": 3.104575163398693e-06,
|
| 201 |
+
"loss": 0.455,
|
| 202 |
+
"step": 150
|
| 203 |
+
},
|
| 204 |
+
{
|
| 205 |
+
"epoch": 8.05,
|
| 206 |
+
"eval_accuracy": 0.71900826446281,
|
| 207 |
+
"eval_accuracy_gunit": 0.925,
|
| 208 |
+
"eval_accuracy_nothing": 0.6172839506172839,
|
| 209 |
+
"eval_loss": 0.572287917137146,
|
| 210 |
+
"eval_runtime": 16.0277,
|
| 211 |
+
"eval_samples_per_second": 7.549,
|
| 212 |
+
"eval_steps_per_second": 0.998,
|
| 213 |
+
"step": 153
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 9.020588235294118,
|
| 217 |
+
"grad_norm": 4.390650749206543,
|
| 218 |
+
"learning_rate": 2.9411764705882355e-06,
|
| 219 |
+
"loss": 0.382,
|
| 220 |
+
"step": 160
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 9.05,
|
| 224 |
+
"grad_norm": 12.403214454650879,
|
| 225 |
+
"learning_rate": 2.7777777777777783e-06,
|
| 226 |
+
"loss": 0.4191,
|
| 227 |
+
"step": 170
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 9.05,
|
| 231 |
+
"eval_accuracy": 0.9338842975206612,
|
| 232 |
+
"eval_accuracy_gunit": 0.85,
|
| 233 |
+
"eval_accuracy_nothing": 0.9753086419753086,
|
| 234 |
+
"eval_loss": 0.37085020542144775,
|
| 235 |
+
"eval_runtime": 19.423,
|
| 236 |
+
"eval_samples_per_second": 6.23,
|
| 237 |
+
"eval_steps_per_second": 0.824,
|
| 238 |
+
"step": 170
|
| 239 |
+
},
|
| 240 |
+
{
|
| 241 |
+
"epoch": 10.029411764705882,
|
| 242 |
+
"grad_norm": 8.752729415893555,
|
| 243 |
+
"learning_rate": 2.6143790849673208e-06,
|
| 244 |
+
"loss": 0.365,
|
| 245 |
+
"step": 180
|
| 246 |
+
},
|
| 247 |
+
{
|
| 248 |
+
"epoch": 10.05,
|
| 249 |
+
"eval_accuracy": 0.9090909090909091,
|
| 250 |
+
"eval_accuracy_gunit": 0.85,
|
| 251 |
+
"eval_accuracy_nothing": 0.9382716049382716,
|
| 252 |
+
"eval_loss": 0.3546590209007263,
|
| 253 |
+
"eval_runtime": 15.5146,
|
| 254 |
+
"eval_samples_per_second": 7.799,
|
| 255 |
+
"eval_steps_per_second": 1.031,
|
| 256 |
+
"step": 187
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 11.008823529411766,
|
| 260 |
+
"grad_norm": 6.302486896514893,
|
| 261 |
+
"learning_rate": 2.450980392156863e-06,
|
| 262 |
+
"loss": 0.3493,
|
| 263 |
+
"step": 190
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 11.038235294117648,
|
| 267 |
+
"grad_norm": 4.457513809204102,
|
| 268 |
+
"learning_rate": 2.2875816993464053e-06,
|
| 269 |
+
"loss": 0.2593,
|
| 270 |
+
"step": 200
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 11.05,
|
| 274 |
+
"eval_accuracy": 0.8925619834710744,
|
| 275 |
+
"eval_accuracy_gunit": 0.85,
|
| 276 |
+
"eval_accuracy_nothing": 0.9135802469135802,
|
| 277 |
+
"eval_loss": 0.36317938566207886,
|
| 278 |
+
"eval_runtime": 15.7061,
|
| 279 |
+
"eval_samples_per_second": 7.704,
|
| 280 |
+
"eval_steps_per_second": 1.019,
|
| 281 |
+
"step": 204
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 12.01764705882353,
|
| 285 |
+
"grad_norm": 11.817601203918457,
|
| 286 |
+
"learning_rate": 2.1241830065359477e-06,
|
| 287 |
+
"loss": 0.3111,
|
| 288 |
+
"step": 210
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 12.047058823529412,
|
| 292 |
+
"grad_norm": 6.769512176513672,
|
| 293 |
+
"learning_rate": 1.96078431372549e-06,
|
| 294 |
+
"loss": 0.2225,
|
| 295 |
+
"step": 220
|
| 296 |
+
},
|
| 297 |
+
{
|
| 298 |
+
"epoch": 12.05,
|
| 299 |
+
"eval_accuracy": 0.9421487603305785,
|
| 300 |
+
"eval_accuracy_gunit": 0.85,
|
| 301 |
+
"eval_accuracy_nothing": 0.9876543209876543,
|
| 302 |
+
"eval_loss": 0.24565516412258148,
|
| 303 |
+
"eval_runtime": 20.1411,
|
| 304 |
+
"eval_samples_per_second": 6.008,
|
| 305 |
+
"eval_steps_per_second": 0.794,
|
| 306 |
+
"step": 221
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"epoch": 13.026470588235295,
|
| 310 |
+
"grad_norm": 3.0028862953186035,
|
| 311 |
+
"learning_rate": 1.7973856209150328e-06,
|
| 312 |
+
"loss": 0.2121,
|
| 313 |
+
"step": 230
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"epoch": 13.05,
|
| 317 |
+
"eval_accuracy": 0.9256198347107438,
|
| 318 |
+
"eval_accuracy_gunit": 0.85,
|
| 319 |
+
"eval_accuracy_nothing": 0.9629629629629629,
|
| 320 |
+
"eval_loss": 0.26187241077423096,
|
| 321 |
+
"eval_runtime": 15.6231,
|
| 322 |
+
"eval_samples_per_second": 7.745,
|
| 323 |
+
"eval_steps_per_second": 1.024,
|
| 324 |
+
"step": 238
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 14.005882352941176,
|
| 328 |
+
"grad_norm": 11.049546241760254,
|
| 329 |
+
"learning_rate": 1.6339869281045753e-06,
|
| 330 |
+
"loss": 0.1814,
|
| 331 |
+
"step": 240
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 14.035294117647059,
|
| 335 |
+
"grad_norm": 13.598651885986328,
|
| 336 |
+
"learning_rate": 1.4705882352941177e-06,
|
| 337 |
+
"loss": 0.1506,
|
| 338 |
+
"step": 250
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 14.05,
|
| 342 |
+
"eval_accuracy": 0.9504132231404959,
|
| 343 |
+
"eval_accuracy_gunit": 0.85,
|
| 344 |
+
"eval_accuracy_nothing": 1.0,
|
| 345 |
+
"eval_loss": 0.2190253734588623,
|
| 346 |
+
"eval_runtime": 15.7623,
|
| 347 |
+
"eval_samples_per_second": 7.677,
|
| 348 |
+
"eval_steps_per_second": 1.015,
|
| 349 |
+
"step": 255
|
| 350 |
+
},
|
| 351 |
+
{
|
| 352 |
+
"epoch": 15.014705882352942,
|
| 353 |
+
"grad_norm": 5.206620693206787,
|
| 354 |
+
"learning_rate": 1.3071895424836604e-06,
|
| 355 |
+
"loss": 0.1518,
|
| 356 |
+
"step": 260
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 15.044117647058824,
|
| 360 |
+
"grad_norm": 15.521167755126953,
|
| 361 |
+
"learning_rate": 1.1437908496732026e-06,
|
| 362 |
+
"loss": 0.1335,
|
| 363 |
+
"step": 270
|
| 364 |
+
},
|
| 365 |
+
{
|
| 366 |
+
"epoch": 15.05,
|
| 367 |
+
"eval_accuracy": 0.9338842975206612,
|
| 368 |
+
"eval_accuracy_gunit": 0.85,
|
| 369 |
+
"eval_accuracy_nothing": 0.9753086419753086,
|
| 370 |
+
"eval_loss": 0.22213229537010193,
|
| 371 |
+
"eval_runtime": 15.6243,
|
| 372 |
+
"eval_samples_per_second": 7.744,
|
| 373 |
+
"eval_steps_per_second": 1.024,
|
| 374 |
+
"step": 272
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 16.023529411764706,
|
| 378 |
+
"grad_norm": 3.6706037521362305,
|
| 379 |
+
"learning_rate": 9.80392156862745e-07,
|
| 380 |
+
"loss": 0.1039,
|
| 381 |
+
"step": 280
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 16.05,
|
| 385 |
+
"eval_accuracy": 0.9421487603305785,
|
| 386 |
+
"eval_accuracy_gunit": 0.85,
|
| 387 |
+
"eval_accuracy_nothing": 0.9876543209876543,
|
| 388 |
+
"eval_loss": 0.2107107788324356,
|
| 389 |
+
"eval_runtime": 15.8386,
|
| 390 |
+
"eval_samples_per_second": 7.64,
|
| 391 |
+
"eval_steps_per_second": 1.01,
|
| 392 |
+
"step": 289
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 17.00294117647059,
|
| 396 |
+
"grad_norm": 2.560628890991211,
|
| 397 |
+
"learning_rate": 8.169934640522876e-07,
|
| 398 |
+
"loss": 0.1959,
|
| 399 |
+
"step": 290
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 17.03235294117647,
|
| 403 |
+
"grad_norm": 13.137368202209473,
|
| 404 |
+
"learning_rate": 6.535947712418302e-07,
|
| 405 |
+
"loss": 0.1036,
|
| 406 |
+
"step": 300
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 17.05,
|
| 410 |
+
"eval_accuracy": 0.9421487603305785,
|
| 411 |
+
"eval_accuracy_gunit": 0.85,
|
| 412 |
+
"eval_accuracy_nothing": 0.9876543209876543,
|
| 413 |
+
"eval_loss": 0.20594191551208496,
|
| 414 |
+
"eval_runtime": 15.7489,
|
| 415 |
+
"eval_samples_per_second": 7.683,
|
| 416 |
+
"eval_steps_per_second": 1.016,
|
| 417 |
+
"step": 306
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 18.011764705882353,
|
| 421 |
+
"grad_norm": 105.94062805175781,
|
| 422 |
+
"learning_rate": 4.901960784313725e-07,
|
| 423 |
+
"loss": 0.139,
|
| 424 |
+
"step": 310
|
| 425 |
+
},
|
| 426 |
+
{
|
| 427 |
+
"epoch": 18.041176470588237,
|
| 428 |
+
"grad_norm": 7.893646240234375,
|
| 429 |
+
"learning_rate": 3.267973856209151e-07,
|
| 430 |
+
"loss": 0.1019,
|
| 431 |
+
"step": 320
|
| 432 |
+
},
|
| 433 |
+
{
|
| 434 |
+
"epoch": 18.05,
|
| 435 |
+
"eval_accuracy": 0.9421487603305785,
|
| 436 |
+
"eval_accuracy_gunit": 0.85,
|
| 437 |
+
"eval_accuracy_nothing": 0.9876543209876543,
|
| 438 |
+
"eval_loss": 0.2062729150056839,
|
| 439 |
+
"eval_runtime": 15.7147,
|
| 440 |
+
"eval_samples_per_second": 7.7,
|
| 441 |
+
"eval_steps_per_second": 1.018,
|
| 442 |
+
"step": 323
|
| 443 |
+
},
|
| 444 |
+
{
|
| 445 |
+
"epoch": 19.020588235294117,
|
| 446 |
+
"grad_norm": 4.479135036468506,
|
| 447 |
+
"learning_rate": 1.6339869281045755e-07,
|
| 448 |
+
"loss": 0.0837,
|
| 449 |
+
"step": 330
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 19.05,
|
| 453 |
+
"grad_norm": 84.62340545654297,
|
| 454 |
+
"learning_rate": 0.0,
|
| 455 |
+
"loss": 0.1181,
|
| 456 |
+
"step": 340
|
| 457 |
+
},
|
| 458 |
+
{
|
| 459 |
+
"epoch": 19.05,
|
| 460 |
+
"eval_accuracy": 0.9504132231404959,
|
| 461 |
+
"eval_accuracy_gunit": 0.85,
|
| 462 |
+
"eval_accuracy_nothing": 1.0,
|
| 463 |
+
"eval_loss": 0.21026436984539032,
|
| 464 |
+
"eval_runtime": 15.8645,
|
| 465 |
+
"eval_samples_per_second": 7.627,
|
| 466 |
+
"eval_steps_per_second": 1.009,
|
| 467 |
+
"step": 340
|
| 468 |
+
},
|
| 469 |
+
{
|
| 470 |
+
"epoch": 19.05,
|
| 471 |
+
"step": 340,
|
| 472 |
+
"total_flos": 3.339453099784274e+18,
|
| 473 |
+
"train_loss": 0.3907703813384561,
|
| 474 |
+
"train_runtime": 996.3737,
|
| 475 |
+
"train_samples_per_second": 2.73,
|
| 476 |
+
"train_steps_per_second": 0.341
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 19.05,
|
| 480 |
+
"eval_accuracy": 0.8932038834951457,
|
| 481 |
+
"eval_accuracy_gunit": 0.5416666666666666,
|
| 482 |
+
"eval_accuracy_nothing": 1.0,
|
| 483 |
+
"eval_loss": 0.34902578592300415,
|
| 484 |
+
"eval_runtime": 14.5221,
|
| 485 |
+
"eval_samples_per_second": 7.093,
|
| 486 |
+
"eval_steps_per_second": 0.895,
|
| 487 |
+
"step": 340
|
| 488 |
+
},
|
| 489 |
+
{
|
| 490 |
+
"epoch": 19.05,
|
| 491 |
+
"eval_accuracy": 0.8932038834951457,
|
| 492 |
+
"eval_accuracy_gunit": 0.5416666666666666,
|
| 493 |
+
"eval_accuracy_nothing": 1.0,
|
| 494 |
+
"eval_loss": 0.3490257263183594,
|
| 495 |
+
"eval_runtime": 13.7098,
|
| 496 |
+
"eval_samples_per_second": 7.513,
|
| 497 |
+
"eval_steps_per_second": 0.948,
|
| 498 |
+
"step": 340
|
| 499 |
+
}
|
| 500 |
+
],
|
| 501 |
+
"logging_steps": 10,
|
| 502 |
+
"max_steps": 340,
|
| 503 |
+
"num_input_tokens_seen": 0,
|
| 504 |
+
"num_train_epochs": 9223372036854775807,
|
| 505 |
+
"save_steps": 500,
|
| 506 |
+
"stateful_callbacks": {
|
| 507 |
+
"TrainerControl": {
|
| 508 |
+
"args": {
|
| 509 |
+
"should_epoch_stop": false,
|
| 510 |
+
"should_evaluate": false,
|
| 511 |
+
"should_log": false,
|
| 512 |
+
"should_save": true,
|
| 513 |
+
"should_training_stop": true
|
| 514 |
+
},
|
| 515 |
+
"attributes": {}
|
| 516 |
+
}
|
| 517 |
+
},
|
| 518 |
+
"total_flos": 3.339453099784274e+18,
|
| 519 |
+
"train_batch_size": 8,
|
| 520 |
+
"trial_name": null,
|
| 521 |
+
"trial_params": null
|
| 522 |
+
}
|