Commit
•
4125d38
1
Parent(s):
8e29fd9
End of training
Browse files- all_results.json +13 -0
- eval_results.json +8 -0
- train_results.json +8 -0
- trainer_state.json +313 -0
all_results.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 100.0,
|
| 3 |
+
"eval_loss": 1.2548089027404785,
|
| 4 |
+
"eval_runtime": 0.7876,
|
| 5 |
+
"eval_samples": 206,
|
| 6 |
+
"eval_samples_per_second": 261.545,
|
| 7 |
+
"eval_steps_per_second": 33.011,
|
| 8 |
+
"train_loss": 0.1092665815749109,
|
| 9 |
+
"train_runtime": 5451.3155,
|
| 10 |
+
"train_samples": 2884,
|
| 11 |
+
"train_samples_per_second": 52.905,
|
| 12 |
+
"train_steps_per_second": 4.421
|
| 13 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 100.0,
|
| 3 |
+
"eval_loss": 1.2548089027404785,
|
| 4 |
+
"eval_runtime": 0.7876,
|
| 5 |
+
"eval_samples": 206,
|
| 6 |
+
"eval_samples_per_second": 261.545,
|
| 7 |
+
"eval_steps_per_second": 33.011
|
| 8 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 100.0,
|
| 3 |
+
"train_loss": 0.1092665815749109,
|
| 4 |
+
"train_runtime": 5451.3155,
|
| 5 |
+
"train_samples": 2884,
|
| 6 |
+
"train_samples_per_second": 52.905,
|
| 7 |
+
"train_steps_per_second": 4.421
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,313 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 100.0,
|
| 5 |
+
"global_step": 24100,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 2.07,
|
| 12 |
+
"learning_rate": 2.937759336099585e-05,
|
| 13 |
+
"loss": 0.3885,
|
| 14 |
+
"step": 500
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 4.15,
|
| 18 |
+
"learning_rate": 2.8755186721991704e-05,
|
| 19 |
+
"loss": 0.2791,
|
| 20 |
+
"step": 1000
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 6.22,
|
| 24 |
+
"learning_rate": 2.8132780082987553e-05,
|
| 25 |
+
"loss": 0.2363,
|
| 26 |
+
"step": 1500
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 8.3,
|
| 30 |
+
"learning_rate": 2.75103734439834e-05,
|
| 31 |
+
"loss": 0.2137,
|
| 32 |
+
"step": 2000
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 10.37,
|
| 36 |
+
"learning_rate": 2.6887966804979253e-05,
|
| 37 |
+
"loss": 0.1969,
|
| 38 |
+
"step": 2500
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 12.45,
|
| 42 |
+
"learning_rate": 2.6265560165975105e-05,
|
| 43 |
+
"loss": 0.1854,
|
| 44 |
+
"step": 3000
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 14.52,
|
| 48 |
+
"learning_rate": 2.5643153526970954e-05,
|
| 49 |
+
"loss": 0.1694,
|
| 50 |
+
"step": 3500
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 16.6,
|
| 54 |
+
"learning_rate": 2.5020746887966805e-05,
|
| 55 |
+
"loss": 0.1722,
|
| 56 |
+
"step": 4000
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 18.67,
|
| 60 |
+
"learning_rate": 2.4398340248962657e-05,
|
| 61 |
+
"loss": 0.1514,
|
| 62 |
+
"step": 4500
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 20.75,
|
| 66 |
+
"learning_rate": 2.3775933609958506e-05,
|
| 67 |
+
"loss": 0.1499,
|
| 68 |
+
"step": 5000
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 22.82,
|
| 72 |
+
"learning_rate": 2.3153526970954358e-05,
|
| 73 |
+
"loss": 0.1464,
|
| 74 |
+
"step": 5500
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 24.9,
|
| 78 |
+
"learning_rate": 2.253112033195021e-05,
|
| 79 |
+
"loss": 0.1343,
|
| 80 |
+
"step": 6000
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 26.97,
|
| 84 |
+
"learning_rate": 2.1908713692946058e-05,
|
| 85 |
+
"loss": 0.1304,
|
| 86 |
+
"step": 6500
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 29.05,
|
| 90 |
+
"learning_rate": 2.128630705394191e-05,
|
| 91 |
+
"loss": 0.1243,
|
| 92 |
+
"step": 7000
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 31.12,
|
| 96 |
+
"learning_rate": 2.0663900414937758e-05,
|
| 97 |
+
"loss": 0.1165,
|
| 98 |
+
"step": 7500
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 33.2,
|
| 102 |
+
"learning_rate": 2.004149377593361e-05,
|
| 103 |
+
"loss": 0.119,
|
| 104 |
+
"step": 8000
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 35.27,
|
| 108 |
+
"learning_rate": 1.9419087136929462e-05,
|
| 109 |
+
"loss": 0.112,
|
| 110 |
+
"step": 8500
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 37.34,
|
| 114 |
+
"learning_rate": 1.879668049792531e-05,
|
| 115 |
+
"loss": 0.1061,
|
| 116 |
+
"step": 9000
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 39.42,
|
| 120 |
+
"learning_rate": 1.8174273858921162e-05,
|
| 121 |
+
"loss": 0.1048,
|
| 122 |
+
"step": 9500
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 41.49,
|
| 126 |
+
"learning_rate": 1.7551867219917014e-05,
|
| 127 |
+
"loss": 0.1008,
|
| 128 |
+
"step": 10000
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 43.57,
|
| 132 |
+
"learning_rate": 1.6929460580912863e-05,
|
| 133 |
+
"loss": 0.0924,
|
| 134 |
+
"step": 10500
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 45.64,
|
| 138 |
+
"learning_rate": 1.630705394190871e-05,
|
| 139 |
+
"loss": 0.0912,
|
| 140 |
+
"step": 11000
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 47.72,
|
| 144 |
+
"learning_rate": 1.5684647302904566e-05,
|
| 145 |
+
"loss": 0.0942,
|
| 146 |
+
"step": 11500
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 49.79,
|
| 150 |
+
"learning_rate": 1.5062240663900415e-05,
|
| 151 |
+
"loss": 0.0855,
|
| 152 |
+
"step": 12000
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 51.87,
|
| 156 |
+
"learning_rate": 1.4439834024896267e-05,
|
| 157 |
+
"loss": 0.0855,
|
| 158 |
+
"step": 12500
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 53.94,
|
| 162 |
+
"learning_rate": 1.3817427385892115e-05,
|
| 163 |
+
"loss": 0.0801,
|
| 164 |
+
"step": 13000
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 56.02,
|
| 168 |
+
"learning_rate": 1.3195020746887967e-05,
|
| 169 |
+
"loss": 0.0787,
|
| 170 |
+
"step": 13500
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 58.09,
|
| 174 |
+
"learning_rate": 1.2572614107883817e-05,
|
| 175 |
+
"loss": 0.0777,
|
| 176 |
+
"step": 14000
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 60.17,
|
| 180 |
+
"learning_rate": 1.1950207468879667e-05,
|
| 181 |
+
"loss": 0.0733,
|
| 182 |
+
"step": 14500
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 62.24,
|
| 186 |
+
"learning_rate": 1.132780082987552e-05,
|
| 187 |
+
"loss": 0.0722,
|
| 188 |
+
"step": 15000
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 64.32,
|
| 192 |
+
"learning_rate": 1.070539419087137e-05,
|
| 193 |
+
"loss": 0.0713,
|
| 194 |
+
"step": 15500
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 66.39,
|
| 198 |
+
"learning_rate": 1.0082987551867221e-05,
|
| 199 |
+
"loss": 0.0679,
|
| 200 |
+
"step": 16000
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 68.46,
|
| 204 |
+
"learning_rate": 9.46058091286307e-06,
|
| 205 |
+
"loss": 0.0685,
|
| 206 |
+
"step": 16500
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 70.54,
|
| 210 |
+
"learning_rate": 8.838174273858922e-06,
|
| 211 |
+
"loss": 0.0656,
|
| 212 |
+
"step": 17000
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 72.61,
|
| 216 |
+
"learning_rate": 8.215767634854773e-06,
|
| 217 |
+
"loss": 0.0657,
|
| 218 |
+
"step": 17500
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 74.69,
|
| 222 |
+
"learning_rate": 7.593360995850622e-06,
|
| 223 |
+
"loss": 0.0611,
|
| 224 |
+
"step": 18000
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 76.76,
|
| 228 |
+
"learning_rate": 6.970954356846474e-06,
|
| 229 |
+
"loss": 0.061,
|
| 230 |
+
"step": 18500
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 78.84,
|
| 234 |
+
"learning_rate": 6.348547717842324e-06,
|
| 235 |
+
"loss": 0.0605,
|
| 236 |
+
"step": 19000
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 80.91,
|
| 240 |
+
"learning_rate": 5.726141078838174e-06,
|
| 241 |
+
"loss": 0.0602,
|
| 242 |
+
"step": 19500
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 82.99,
|
| 246 |
+
"learning_rate": 5.103734439834025e-06,
|
| 247 |
+
"loss": 0.0585,
|
| 248 |
+
"step": 20000
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 85.06,
|
| 252 |
+
"learning_rate": 4.481327800829875e-06,
|
| 253 |
+
"loss": 0.0583,
|
| 254 |
+
"step": 20500
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 87.14,
|
| 258 |
+
"learning_rate": 3.858921161825726e-06,
|
| 259 |
+
"loss": 0.056,
|
| 260 |
+
"step": 21000
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 89.21,
|
| 264 |
+
"learning_rate": 3.236514522821577e-06,
|
| 265 |
+
"loss": 0.055,
|
| 266 |
+
"step": 21500
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 91.29,
|
| 270 |
+
"learning_rate": 2.6141078838174274e-06,
|
| 271 |
+
"loss": 0.0582,
|
| 272 |
+
"step": 22000
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 93.36,
|
| 276 |
+
"learning_rate": 1.991701244813278e-06,
|
| 277 |
+
"loss": 0.0556,
|
| 278 |
+
"step": 22500
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 95.44,
|
| 282 |
+
"learning_rate": 1.3692946058091286e-06,
|
| 283 |
+
"loss": 0.0545,
|
| 284 |
+
"step": 23000
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 97.51,
|
| 288 |
+
"learning_rate": 7.468879668049792e-07,
|
| 289 |
+
"loss": 0.0554,
|
| 290 |
+
"step": 23500
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 99.59,
|
| 294 |
+
"learning_rate": 1.2448132780082988e-07,
|
| 295 |
+
"loss": 0.0548,
|
| 296 |
+
"step": 24000
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 100.0,
|
| 300 |
+
"step": 24100,
|
| 301 |
+
"total_flos": 3.90325755445248e+16,
|
| 302 |
+
"train_loss": 0.1092665815749109,
|
| 303 |
+
"train_runtime": 5451.3155,
|
| 304 |
+
"train_samples_per_second": 52.905,
|
| 305 |
+
"train_steps_per_second": 4.421
|
| 306 |
+
}
|
| 307 |
+
],
|
| 308 |
+
"max_steps": 24100,
|
| 309 |
+
"num_train_epochs": 100,
|
| 310 |
+
"total_flos": 3.90325755445248e+16,
|
| 311 |
+
"trial_name": null,
|
| 312 |
+
"trial_params": null
|
| 313 |
+
}
|