Commit
•
d7518bb
1
Parent(s):
370c3e9
Upload trainer_state.json with huggingface_hub
Browse files- trainer_state.json +277 -0
trainer_state.json
ADDED
|
@@ -0,0 +1,277 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 19.9984,
|
| 5 |
+
"global_step": 6240,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 1.0,
|
| 12 |
+
"eval_loss": 0.4537074565887451,
|
| 13 |
+
"eval_mse": 0.45370742678642273,
|
| 14 |
+
"eval_runtime": 6.2981,
|
| 15 |
+
"eval_samples_per_second": 158.779,
|
| 16 |
+
"eval_steps_per_second": 19.847,
|
| 17 |
+
"step": 312
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 1.6,
|
| 21 |
+
"learning_rate": 1.3457813907586236e-05,
|
| 22 |
+
"loss": 0.6153,
|
| 23 |
+
"step": 500
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 2.0,
|
| 27 |
+
"eval_loss": 0.19009532034397125,
|
| 28 |
+
"eval_mse": 0.19009532034397125,
|
| 29 |
+
"eval_runtime": 6.3224,
|
| 30 |
+
"eval_samples_per_second": 158.167,
|
| 31 |
+
"eval_steps_per_second": 19.771,
|
| 32 |
+
"step": 624
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 3.0,
|
| 36 |
+
"eval_loss": 0.2025919258594513,
|
| 37 |
+
"eval_mse": 0.2025919258594513,
|
| 38 |
+
"eval_runtime": 6.3503,
|
| 39 |
+
"eval_samples_per_second": 157.472,
|
| 40 |
+
"eval_steps_per_second": 19.684,
|
| 41 |
+
"step": 936
|
| 42 |
+
},
|
| 43 |
+
{
|
| 44 |
+
"epoch": 3.2,
|
| 45 |
+
"learning_rate": 1.2285530466158863e-05,
|
| 46 |
+
"loss": 0.1584,
|
| 47 |
+
"step": 1000
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"epoch": 4.0,
|
| 51 |
+
"eval_loss": 0.2967708110809326,
|
| 52 |
+
"eval_mse": 0.2967708110809326,
|
| 53 |
+
"eval_runtime": 6.3822,
|
| 54 |
+
"eval_samples_per_second": 156.686,
|
| 55 |
+
"eval_steps_per_second": 19.586,
|
| 56 |
+
"step": 1248
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 4.81,
|
| 60 |
+
"learning_rate": 1.1113247024731491e-05,
|
| 61 |
+
"loss": 0.0958,
|
| 62 |
+
"step": 1500
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 5.0,
|
| 66 |
+
"eval_loss": 0.2692314684391022,
|
| 67 |
+
"eval_mse": 0.2692314386367798,
|
| 68 |
+
"eval_runtime": 6.3165,
|
| 69 |
+
"eval_samples_per_second": 158.314,
|
| 70 |
+
"eval_steps_per_second": 19.789,
|
| 71 |
+
"step": 1560
|
| 72 |
+
},
|
| 73 |
+
{
|
| 74 |
+
"epoch": 6.0,
|
| 75 |
+
"eval_loss": 0.2646154761314392,
|
| 76 |
+
"eval_mse": 0.2646154761314392,
|
| 77 |
+
"eval_runtime": 6.321,
|
| 78 |
+
"eval_samples_per_second": 158.204,
|
| 79 |
+
"eval_steps_per_second": 19.775,
|
| 80 |
+
"step": 1872
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 6.41,
|
| 84 |
+
"learning_rate": 9.940963583304119e-06,
|
| 85 |
+
"loss": 0.0605,
|
| 86 |
+
"step": 2000
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 7.0,
|
| 90 |
+
"eval_loss": 0.3655566871166229,
|
| 91 |
+
"eval_mse": 0.3655566871166229,
|
| 92 |
+
"eval_runtime": 6.4015,
|
| 93 |
+
"eval_samples_per_second": 156.213,
|
| 94 |
+
"eval_steps_per_second": 19.527,
|
| 95 |
+
"step": 2184
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 8.0,
|
| 99 |
+
"eval_loss": 0.3432270586490631,
|
| 100 |
+
"eval_mse": 0.3432270586490631,
|
| 101 |
+
"eval_runtime": 6.4193,
|
| 102 |
+
"eval_samples_per_second": 155.78,
|
| 103 |
+
"eval_steps_per_second": 19.473,
|
| 104 |
+
"step": 2496
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 8.01,
|
| 108 |
+
"learning_rate": 8.768680141876746e-06,
|
| 109 |
+
"loss": 0.0448,
|
| 110 |
+
"step": 2500
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 9.0,
|
| 114 |
+
"eval_loss": 0.29013848304748535,
|
| 115 |
+
"eval_mse": 0.29013848304748535,
|
| 116 |
+
"eval_runtime": 6.369,
|
| 117 |
+
"eval_samples_per_second": 157.01,
|
| 118 |
+
"eval_steps_per_second": 19.626,
|
| 119 |
+
"step": 2808
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 9.61,
|
| 123 |
+
"learning_rate": 7.596396700449374e-06,
|
| 124 |
+
"loss": 0.0328,
|
| 125 |
+
"step": 3000
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 10.0,
|
| 129 |
+
"eval_loss": 0.23616179823875427,
|
| 130 |
+
"eval_mse": 0.23616181313991547,
|
| 131 |
+
"eval_runtime": 6.3278,
|
| 132 |
+
"eval_samples_per_second": 158.032,
|
| 133 |
+
"eval_steps_per_second": 19.754,
|
| 134 |
+
"step": 3120
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 11.0,
|
| 138 |
+
"eval_loss": 0.23755905032157898,
|
| 139 |
+
"eval_mse": 0.23755905032157898,
|
| 140 |
+
"eval_runtime": 6.3847,
|
| 141 |
+
"eval_samples_per_second": 156.626,
|
| 142 |
+
"eval_steps_per_second": 19.578,
|
| 143 |
+
"step": 3432
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 11.22,
|
| 147 |
+
"learning_rate": 6.424113259022001e-06,
|
| 148 |
+
"loss": 0.0286,
|
| 149 |
+
"step": 3500
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 12.0,
|
| 153 |
+
"eval_loss": 0.23381923139095306,
|
| 154 |
+
"eval_mse": 0.23381923139095306,
|
| 155 |
+
"eval_runtime": 6.3781,
|
| 156 |
+
"eval_samples_per_second": 156.787,
|
| 157 |
+
"eval_steps_per_second": 19.598,
|
| 158 |
+
"step": 3744
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 12.82,
|
| 162 |
+
"learning_rate": 5.2518298175946285e-06,
|
| 163 |
+
"loss": 0.0243,
|
| 164 |
+
"step": 4000
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 13.0,
|
| 168 |
+
"eval_loss": 0.2918383777141571,
|
| 169 |
+
"eval_mse": 0.2918383777141571,
|
| 170 |
+
"eval_runtime": 6.3488,
|
| 171 |
+
"eval_samples_per_second": 157.51,
|
| 172 |
+
"eval_steps_per_second": 19.689,
|
| 173 |
+
"step": 4056
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 14.0,
|
| 177 |
+
"eval_loss": 0.26574844121932983,
|
| 178 |
+
"eval_mse": 0.26574844121932983,
|
| 179 |
+
"eval_runtime": 6.3063,
|
| 180 |
+
"eval_samples_per_second": 158.572,
|
| 181 |
+
"eval_steps_per_second": 19.822,
|
| 182 |
+
"step": 4368
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 14.42,
|
| 186 |
+
"learning_rate": 4.079546376167256e-06,
|
| 187 |
+
"loss": 0.021,
|
| 188 |
+
"step": 4500
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 15.0,
|
| 192 |
+
"eval_loss": 0.2649979293346405,
|
| 193 |
+
"eval_mse": 0.2649979293346405,
|
| 194 |
+
"eval_runtime": 6.3763,
|
| 195 |
+
"eval_samples_per_second": 156.832,
|
| 196 |
+
"eval_steps_per_second": 19.604,
|
| 197 |
+
"step": 4680
|
| 198 |
+
},
|
| 199 |
+
{
|
| 200 |
+
"epoch": 16.0,
|
| 201 |
+
"eval_loss": 0.2675020396709442,
|
| 202 |
+
"eval_mse": 0.2675020396709442,
|
| 203 |
+
"eval_runtime": 6.2951,
|
| 204 |
+
"eval_samples_per_second": 158.855,
|
| 205 |
+
"eval_steps_per_second": 19.857,
|
| 206 |
+
"step": 4992
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 16.03,
|
| 210 |
+
"learning_rate": 2.9072629347398834e-06,
|
| 211 |
+
"loss": 0.019,
|
| 212 |
+
"step": 5000
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 17.0,
|
| 216 |
+
"eval_loss": 0.2602880895137787,
|
| 217 |
+
"eval_mse": 0.2602880895137787,
|
| 218 |
+
"eval_runtime": 6.3892,
|
| 219 |
+
"eval_samples_per_second": 156.515,
|
| 220 |
+
"eval_steps_per_second": 19.564,
|
| 221 |
+
"step": 5304
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 17.63,
|
| 225 |
+
"learning_rate": 1.734979493312511e-06,
|
| 226 |
+
"loss": 0.0165,
|
| 227 |
+
"step": 5500
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 18.0,
|
| 231 |
+
"eval_loss": 0.2324579656124115,
|
| 232 |
+
"eval_mse": 0.2324579656124115,
|
| 233 |
+
"eval_runtime": 6.3476,
|
| 234 |
+
"eval_samples_per_second": 157.541,
|
| 235 |
+
"eval_steps_per_second": 19.693,
|
| 236 |
+
"step": 5616
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 19.0,
|
| 240 |
+
"eval_loss": 0.24971328675746918,
|
| 241 |
+
"eval_mse": 0.24971328675746918,
|
| 242 |
+
"eval_runtime": 6.3218,
|
| 243 |
+
"eval_samples_per_second": 158.182,
|
| 244 |
+
"eval_steps_per_second": 19.773,
|
| 245 |
+
"step": 5928
|
| 246 |
+
},
|
| 247 |
+
{
|
| 248 |
+
"epoch": 19.23,
|
| 249 |
+
"learning_rate": 5.626960518851388e-07,
|
| 250 |
+
"loss": 0.015,
|
| 251 |
+
"step": 6000
|
| 252 |
+
},
|
| 253 |
+
{
|
| 254 |
+
"epoch": 20.0,
|
| 255 |
+
"eval_loss": 0.24272866547107697,
|
| 256 |
+
"eval_mse": 0.24272866547107697,
|
| 257 |
+
"eval_runtime": 6.3685,
|
| 258 |
+
"eval_samples_per_second": 157.022,
|
| 259 |
+
"eval_steps_per_second": 19.628,
|
| 260 |
+
"step": 6240
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 20.0,
|
| 264 |
+
"step": 6240,
|
| 265 |
+
"total_flos": 6577309062776832.0,
|
| 266 |
+
"train_loss": 0.09126096031604669,
|
| 267 |
+
"train_runtime": 4399.0118,
|
| 268 |
+
"train_samples_per_second": 22.732,
|
| 269 |
+
"train_steps_per_second": 1.419
|
| 270 |
+
}
|
| 271 |
+
],
|
| 272 |
+
"max_steps": 6240,
|
| 273 |
+
"num_train_epochs": 20,
|
| 274 |
+
"total_flos": 6577309062776832.0,
|
| 275 |
+
"trial_name": null,
|
| 276 |
+
"trial_params": null
|
| 277 |
+
}
|