Commit
•
6936703
1
Parent(s):
2fb9a84
Upload trainer_state.json
Browse files- trainer_state.json +262 -0
trainer_state.json
ADDED
|
@@ -0,0 +1,262 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 5.0,
|
| 5 |
+
"global_step": 16110,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.16,
|
| 12 |
+
"learning_rate": 5.17063081695967e-06,
|
| 13 |
+
"loss": 0.9438,
|
| 14 |
+
"step": 500
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.31,
|
| 18 |
+
"learning_rate": 9.97820775275705e-06,
|
| 19 |
+
"loss": 0.6981,
|
| 20 |
+
"step": 1000
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.47,
|
| 24 |
+
"learning_rate": 9.648022188469921e-06,
|
| 25 |
+
"loss": 0.6641,
|
| 26 |
+
"step": 1500
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.62,
|
| 30 |
+
"learning_rate": 9.317836624182791e-06,
|
| 31 |
+
"loss": 0.6351,
|
| 32 |
+
"step": 2000
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.78,
|
| 36 |
+
"learning_rate": 8.987651059895663e-06,
|
| 37 |
+
"loss": 0.6069,
|
| 38 |
+
"step": 2500
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.93,
|
| 42 |
+
"learning_rate": 8.657465495608533e-06,
|
| 43 |
+
"loss": 0.6,
|
| 44 |
+
"step": 3000
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 1.0,
|
| 48 |
+
"eval_accuracy": 0.7530310750007629,
|
| 49 |
+
"eval_loss": 0.5826240181922913,
|
| 50 |
+
"eval_runtime": 74.1302,
|
| 51 |
+
"eval_samples_per_second": 132.402,
|
| 52 |
+
"eval_steps_per_second": 4.141,
|
| 53 |
+
"step": 3222
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 1.09,
|
| 57 |
+
"learning_rate": 8.327279931321404e-06,
|
| 58 |
+
"loss": 0.5475,
|
| 59 |
+
"step": 3500
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 1.24,
|
| 63 |
+
"learning_rate": 7.997094367034274e-06,
|
| 64 |
+
"loss": 0.5118,
|
| 65 |
+
"step": 4000
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 1.4,
|
| 69 |
+
"learning_rate": 7.666908802747146e-06,
|
| 70 |
+
"loss": 0.5113,
|
| 71 |
+
"step": 4500
|
| 72 |
+
},
|
| 73 |
+
{
|
| 74 |
+
"epoch": 1.55,
|
| 75 |
+
"learning_rate": 7.336723238460015e-06,
|
| 76 |
+
"loss": 0.5134,
|
| 77 |
+
"step": 5000
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 1.71,
|
| 81 |
+
"learning_rate": 7.006537674172885e-06,
|
| 82 |
+
"loss": 0.5022,
|
| 83 |
+
"step": 5500
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 1.86,
|
| 87 |
+
"learning_rate": 6.676352109885757e-06,
|
| 88 |
+
"loss": 0.5082,
|
| 89 |
+
"step": 6000
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 2.0,
|
| 93 |
+
"eval_accuracy": 0.7587366104125977,
|
| 94 |
+
"eval_loss": 0.5997494459152222,
|
| 95 |
+
"eval_runtime": 74.1755,
|
| 96 |
+
"eval_samples_per_second": 132.321,
|
| 97 |
+
"eval_steps_per_second": 4.139,
|
| 98 |
+
"step": 6444
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 2.02,
|
| 102 |
+
"learning_rate": 6.3461665455986265e-06,
|
| 103 |
+
"loss": 0.4847,
|
| 104 |
+
"step": 6500
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 2.17,
|
| 108 |
+
"learning_rate": 6.015980981311498e-06,
|
| 109 |
+
"loss": 0.3947,
|
| 110 |
+
"step": 7000
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 2.33,
|
| 114 |
+
"learning_rate": 5.685795417024368e-06,
|
| 115 |
+
"loss": 0.3968,
|
| 116 |
+
"step": 7500
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 2.48,
|
| 120 |
+
"learning_rate": 5.3556098527372394e-06,
|
| 121 |
+
"loss": 0.3938,
|
| 122 |
+
"step": 8000
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 2.64,
|
| 126 |
+
"learning_rate": 5.025424288450109e-06,
|
| 127 |
+
"loss": 0.3877,
|
| 128 |
+
"step": 8500
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 2.79,
|
| 132 |
+
"learning_rate": 4.69523872416298e-06,
|
| 133 |
+
"loss": 0.3839,
|
| 134 |
+
"step": 9000
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 2.95,
|
| 138 |
+
"learning_rate": 4.365053159875851e-06,
|
| 139 |
+
"loss": 0.3949,
|
| 140 |
+
"step": 9500
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 3.0,
|
| 144 |
+
"eval_accuracy": 0.8090677261352539,
|
| 145 |
+
"eval_loss": 0.5072706937789917,
|
| 146 |
+
"eval_runtime": 74.3802,
|
| 147 |
+
"eval_samples_per_second": 131.957,
|
| 148 |
+
"eval_steps_per_second": 4.127,
|
| 149 |
+
"step": 9666
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 3.1,
|
| 153 |
+
"learning_rate": 4.034867595588721e-06,
|
| 154 |
+
"loss": 0.321,
|
| 155 |
+
"step": 10000
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"epoch": 3.26,
|
| 159 |
+
"learning_rate": 3.704682031301592e-06,
|
| 160 |
+
"loss": 0.2894,
|
| 161 |
+
"step": 10500
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 3.41,
|
| 165 |
+
"learning_rate": 3.3744964670144627e-06,
|
| 166 |
+
"loss": 0.2923,
|
| 167 |
+
"step": 11000
|
| 168 |
+
},
|
| 169 |
+
{
|
| 170 |
+
"epoch": 3.57,
|
| 171 |
+
"learning_rate": 3.0443109027273334e-06,
|
| 172 |
+
"loss": 0.2906,
|
| 173 |
+
"step": 11500
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 3.72,
|
| 177 |
+
"learning_rate": 2.714125338440204e-06,
|
| 178 |
+
"loss": 0.2913,
|
| 179 |
+
"step": 12000
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 3.88,
|
| 183 |
+
"learning_rate": 2.3839397741530743e-06,
|
| 184 |
+
"loss": 0.2969,
|
| 185 |
+
"step": 12500
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 4.0,
|
| 189 |
+
"eval_accuracy": 0.7917473316192627,
|
| 190 |
+
"eval_loss": 0.6656549572944641,
|
| 191 |
+
"eval_runtime": 74.3184,
|
| 192 |
+
"eval_samples_per_second": 132.067,
|
| 193 |
+
"eval_steps_per_second": 4.131,
|
| 194 |
+
"step": 12888
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 4.03,
|
| 198 |
+
"learning_rate": 2.053754209865945e-06,
|
| 199 |
+
"loss": 0.2657,
|
| 200 |
+
"step": 13000
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 4.19,
|
| 204 |
+
"learning_rate": 1.7235686455788153e-06,
|
| 205 |
+
"loss": 0.2215,
|
| 206 |
+
"step": 13500
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 4.35,
|
| 210 |
+
"learning_rate": 1.393383081291686e-06,
|
| 211 |
+
"loss": 0.2228,
|
| 212 |
+
"step": 14000
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 4.5,
|
| 216 |
+
"learning_rate": 1.0631975170045567e-06,
|
| 217 |
+
"loss": 0.215,
|
| 218 |
+
"step": 14500
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 4.66,
|
| 222 |
+
"learning_rate": 7.330119527174273e-07,
|
| 223 |
+
"loss": 0.2306,
|
| 224 |
+
"step": 15000
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 4.81,
|
| 228 |
+
"learning_rate": 4.0282638843029786e-07,
|
| 229 |
+
"loss": 0.2168,
|
| 230 |
+
"step": 15500
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 4.97,
|
| 234 |
+
"learning_rate": 7.264082414316847e-08,
|
| 235 |
+
"loss": 0.2267,
|
| 236 |
+
"step": 16000
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 5.0,
|
| 240 |
+
"eval_accuracy": 0.801732063293457,
|
| 241 |
+
"eval_loss": 0.7047895789146423,
|
| 242 |
+
"eval_runtime": 74.3187,
|
| 243 |
+
"eval_samples_per_second": 132.066,
|
| 244 |
+
"eval_steps_per_second": 4.131,
|
| 245 |
+
"step": 16110
|
| 246 |
+
},
|
| 247 |
+
{
|
| 248 |
+
"epoch": 5.0,
|
| 249 |
+
"step": 16110,
|
| 250 |
+
"total_flos": 1.2007859696955264e+17,
|
| 251 |
+
"train_loss": 0.41926650207727906,
|
| 252 |
+
"train_runtime": 12549.4046,
|
| 253 |
+
"train_samples_per_second": 41.069,
|
| 254 |
+
"train_steps_per_second": 1.284
|
| 255 |
+
}
|
| 256 |
+
],
|
| 257 |
+
"max_steps": 16110,
|
| 258 |
+
"num_train_epochs": 5,
|
| 259 |
+
"total_flos": 1.2007859696955264e+17,
|
| 260 |
+
"trial_name": null,
|
| 261 |
+
"trial_params": null
|
| 262 |
+
}
|