End of training
Browse files- all_results.json +14 -0
- eval_results.json +9 -0
- train_results.json +8 -0
- trainer_state.json +297 -0
all_results.json
ADDED
|
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 9.95,
|
| 3 |
+
"eval_loss": 1.2986704111099243,
|
| 4 |
+
"eval_runtime": 6.7978,
|
| 5 |
+
"eval_samples": 452,
|
| 6 |
+
"eval_samples_per_second": 66.492,
|
| 7 |
+
"eval_steps_per_second": 16.623,
|
| 8 |
+
"perplexity": 3.6644212534021072,
|
| 9 |
+
"train_loss": 1.8282876014709473,
|
| 10 |
+
"train_runtime": 2056.929,
|
| 11 |
+
"train_samples": 8653,
|
| 12 |
+
"train_samples_per_second": 42.068,
|
| 13 |
+
"train_steps_per_second": 0.078
|
| 14 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 9.95,
|
| 3 |
+
"eval_loss": 1.2986704111099243,
|
| 4 |
+
"eval_runtime": 6.7978,
|
| 5 |
+
"eval_samples": 452,
|
| 6 |
+
"eval_samples_per_second": 66.492,
|
| 7 |
+
"eval_steps_per_second": 16.623,
|
| 8 |
+
"perplexity": 3.6644212534021072
|
| 9 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 9.95,
|
| 3 |
+
"train_loss": 1.8282876014709473,
|
| 4 |
+
"train_runtime": 2056.929,
|
| 5 |
+
"train_samples": 8653,
|
| 6 |
+
"train_samples_per_second": 42.068,
|
| 7 |
+
"train_steps_per_second": 0.078
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,297 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 9.948148148148148,
|
| 5 |
+
"global_step": 160,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.3,
|
| 12 |
+
"learning_rate": 6.25e-05,
|
| 13 |
+
"loss": 4.423,
|
| 14 |
+
"step": 5
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.59,
|
| 18 |
+
"learning_rate": 9.995728791936504e-05,
|
| 19 |
+
"loss": 3.358,
|
| 20 |
+
"step": 10
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.89,
|
| 24 |
+
"learning_rate": 9.947761466636014e-05,
|
| 25 |
+
"loss": 2.8568,
|
| 26 |
+
"step": 15
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.95,
|
| 30 |
+
"eval_loss": 2.593674421310425,
|
| 31 |
+
"eval_runtime": 7.3362,
|
| 32 |
+
"eval_samples_per_second": 61.613,
|
| 33 |
+
"eval_steps_per_second": 15.403,
|
| 34 |
+
"step": 16
|
| 35 |
+
},
|
| 36 |
+
{
|
| 37 |
+
"epoch": 1.24,
|
| 38 |
+
"learning_rate": 9.847001329696653e-05,
|
| 39 |
+
"loss": 3.0543,
|
| 40 |
+
"step": 20
|
| 41 |
+
},
|
| 42 |
+
{
|
| 43 |
+
"epoch": 1.53,
|
| 44 |
+
"learning_rate": 9.694523495787149e-05,
|
| 45 |
+
"loss": 2.3947,
|
| 46 |
+
"step": 25
|
| 47 |
+
},
|
| 48 |
+
{
|
| 49 |
+
"epoch": 1.83,
|
| 50 |
+
"learning_rate": 9.491954909459895e-05,
|
| 51 |
+
"loss": 2.2487,
|
| 52 |
+
"step": 30
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 1.95,
|
| 56 |
+
"eval_loss": 2.1050145626068115,
|
| 57 |
+
"eval_runtime": 7.5204,
|
| 58 |
+
"eval_samples_per_second": 60.104,
|
| 59 |
+
"eval_steps_per_second": 15.026,
|
| 60 |
+
"step": 32
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"epoch": 2.18,
|
| 64 |
+
"learning_rate": 9.241456985587868e-05,
|
| 65 |
+
"loss": 2.4987,
|
| 66 |
+
"step": 35
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 2.47,
|
| 70 |
+
"learning_rate": 8.945702546981969e-05,
|
| 71 |
+
"loss": 2.0039,
|
| 72 |
+
"step": 40
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 2.77,
|
| 76 |
+
"learning_rate": 8.60784730526531e-05,
|
| 77 |
+
"loss": 1.9011,
|
| 78 |
+
"step": 45
|
| 79 |
+
},
|
| 80 |
+
{
|
| 81 |
+
"epoch": 2.95,
|
| 82 |
+
"eval_loss": 1.80819571018219,
|
| 83 |
+
"eval_runtime": 7.4002,
|
| 84 |
+
"eval_samples_per_second": 61.079,
|
| 85 |
+
"eval_steps_per_second": 15.27,
|
| 86 |
+
"step": 48
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 3.12,
|
| 90 |
+
"learning_rate": 8.231496189304704e-05,
|
| 91 |
+
"loss": 2.1532,
|
| 92 |
+
"step": 50
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 3.41,
|
| 96 |
+
"learning_rate": 7.820664880476256e-05,
|
| 97 |
+
"loss": 1.7384,
|
| 98 |
+
"step": 55
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 3.71,
|
| 102 |
+
"learning_rate": 7.379736965185368e-05,
|
| 103 |
+
"loss": 1.6837,
|
| 104 |
+
"step": 60
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 3.95,
|
| 108 |
+
"eval_loss": 1.6178017854690552,
|
| 109 |
+
"eval_runtime": 7.6447,
|
| 110 |
+
"eval_samples_per_second": 59.126,
|
| 111 |
+
"eval_steps_per_second": 14.781,
|
| 112 |
+
"step": 64
|
| 113 |
+
},
|
| 114 |
+
{
|
| 115 |
+
"epoch": 4.06,
|
| 116 |
+
"learning_rate": 6.91341716182545e-05,
|
| 117 |
+
"loss": 1.9079,
|
| 118 |
+
"step": 65
|
| 119 |
+
},
|
| 120 |
+
{
|
| 121 |
+
"epoch": 4.36,
|
| 122 |
+
"learning_rate": 6.426681121245527e-05,
|
| 123 |
+
"loss": 1.5592,
|
| 124 |
+
"step": 70
|
| 125 |
+
},
|
| 126 |
+
{
|
| 127 |
+
"epoch": 4.65,
|
| 128 |
+
"learning_rate": 5.924722336357793e-05,
|
| 129 |
+
"loss": 1.522,
|
| 130 |
+
"step": 75
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 4.95,
|
| 134 |
+
"learning_rate": 5.4128967273616625e-05,
|
| 135 |
+
"loss": 1.4887,
|
| 136 |
+
"step": 80
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 4.95,
|
| 140 |
+
"eval_loss": 1.4896912574768066,
|
| 141 |
+
"eval_runtime": 7.3322,
|
| 142 |
+
"eval_samples_per_second": 61.646,
|
| 143 |
+
"eval_steps_per_second": 15.411,
|
| 144 |
+
"step": 80
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 5.3,
|
| 148 |
+
"learning_rate": 4.8966654938622295e-05,
|
| 149 |
+
"loss": 1.6886,
|
| 150 |
+
"step": 85
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 5.59,
|
| 154 |
+
"learning_rate": 4.381536843653262e-05,
|
| 155 |
+
"loss": 1.4043,
|
| 156 |
+
"step": 90
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 5.89,
|
| 160 |
+
"learning_rate": 3.87300721992097e-05,
|
| 161 |
+
"loss": 1.3812,
|
| 162 |
+
"step": 95
|
| 163 |
+
},
|
| 164 |
+
{
|
| 165 |
+
"epoch": 5.95,
|
| 166 |
+
"eval_loss": 1.4017213582992554,
|
| 167 |
+
"eval_runtime": 7.2083,
|
| 168 |
+
"eval_samples_per_second": 62.706,
|
| 169 |
+
"eval_steps_per_second": 15.676,
|
| 170 |
+
"step": 96
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 6.24,
|
| 174 |
+
"learning_rate": 3.3765026539765834e-05,
|
| 175 |
+
"loss": 1.5807,
|
| 176 |
+
"step": 100
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 6.53,
|
| 180 |
+
"learning_rate": 2.8973208692864624e-05,
|
| 181 |
+
"loss": 1.3233,
|
| 182 |
+
"step": 105
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 6.83,
|
| 186 |
+
"learning_rate": 2.4405747545519963e-05,
|
| 187 |
+
"loss": 1.2944,
|
| 188 |
+
"step": 110
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 6.95,
|
| 192 |
+
"eval_loss": 1.3437390327453613,
|
| 193 |
+
"eval_runtime": 7.1152,
|
| 194 |
+
"eval_samples_per_second": 63.526,
|
| 195 |
+
"eval_steps_per_second": 15.882,
|
| 196 |
+
"step": 112
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 7.18,
|
| 200 |
+
"learning_rate": 2.0111378089837956e-05,
|
| 201 |
+
"loss": 1.5042,
|
| 202 |
+
"step": 115
|
| 203 |
+
},
|
| 204 |
+
{
|
| 205 |
+
"epoch": 7.47,
|
| 206 |
+
"learning_rate": 1.6135921418712956e-05,
|
| 207 |
+
"loss": 1.2607,
|
| 208 |
+
"step": 120
|
| 209 |
+
},
|
| 210 |
+
{
|
| 211 |
+
"epoch": 7.77,
|
| 212 |
+
"learning_rate": 1.2521795812943704e-05,
|
| 213 |
+
"loss": 1.2574,
|
| 214 |
+
"step": 125
|
| 215 |
+
},
|
| 216 |
+
{
|
| 217 |
+
"epoch": 7.95,
|
| 218 |
+
"eval_loss": 1.3126752376556396,
|
| 219 |
+
"eval_runtime": 7.3998,
|
| 220 |
+
"eval_samples_per_second": 61.083,
|
| 221 |
+
"eval_steps_per_second": 15.271,
|
| 222 |
+
"step": 128
|
| 223 |
+
},
|
| 224 |
+
{
|
| 225 |
+
"epoch": 8.12,
|
| 226 |
+
"learning_rate": 9.307564136490254e-06,
|
| 227 |
+
"loss": 1.4587,
|
| 228 |
+
"step": 130
|
| 229 |
+
},
|
| 230 |
+
{
|
| 231 |
+
"epoch": 8.41,
|
| 232 |
+
"learning_rate": 6.527522369181655e-06,
|
| 233 |
+
"loss": 1.2245,
|
| 234 |
+
"step": 135
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 8.71,
|
| 238 |
+
"learning_rate": 4.2113336672471245e-06,
|
| 239 |
+
"loss": 1.2325,
|
| 240 |
+
"step": 140
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 8.95,
|
| 244 |
+
"eval_loss": 1.3008612394332886,
|
| 245 |
+
"eval_runtime": 7.4909,
|
| 246 |
+
"eval_samples_per_second": 60.34,
|
| 247 |
+
"eval_steps_per_second": 15.085,
|
| 248 |
+
"step": 144
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 9.06,
|
| 252 |
+
"learning_rate": 2.3837118562592797e-06,
|
| 253 |
+
"loss": 1.4374,
|
| 254 |
+
"step": 145
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 9.36,
|
| 258 |
+
"learning_rate": 1.064157733632276e-06,
|
| 259 |
+
"loss": 1.2213,
|
| 260 |
+
"step": 150
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 9.65,
|
| 264 |
+
"learning_rate": 2.667509943378721e-07,
|
| 265 |
+
"loss": 1.2213,
|
| 266 |
+
"step": 155
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 9.95,
|
| 270 |
+
"learning_rate": 0.0,
|
| 271 |
+
"loss": 1.2223,
|
| 272 |
+
"step": 160
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 9.95,
|
| 276 |
+
"eval_loss": 1.2986704111099243,
|
| 277 |
+
"eval_runtime": 7.4403,
|
| 278 |
+
"eval_samples_per_second": 60.751,
|
| 279 |
+
"eval_steps_per_second": 15.188,
|
| 280 |
+
"step": 160
|
| 281 |
+
},
|
| 282 |
+
{
|
| 283 |
+
"epoch": 9.95,
|
| 284 |
+
"step": 160,
|
| 285 |
+
"total_flos": 8.010001210461389e+16,
|
| 286 |
+
"train_loss": 1.8282876014709473,
|
| 287 |
+
"train_runtime": 2056.929,
|
| 288 |
+
"train_samples_per_second": 42.068,
|
| 289 |
+
"train_steps_per_second": 0.078
|
| 290 |
+
}
|
| 291 |
+
],
|
| 292 |
+
"max_steps": 160,
|
| 293 |
+
"num_train_epochs": 10,
|
| 294 |
+
"total_flos": 8.010001210461389e+16,
|
| 295 |
+
"trial_name": null,
|
| 296 |
+
"trial_params": null
|
| 297 |
+
}
|