Commit
•
0e3fa03
1
Parent(s):
7285628
End of training
Browse files- all_results.json +8 -0
- train_results.json +8 -0
- trainer_state.json +287 -0
all_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 15.0,
|
| 3 |
+
"train_loss": 1.0260957291357629,
|
| 4 |
+
"train_runtime": 82549.8906,
|
| 5 |
+
"train_samples": 313,
|
| 6 |
+
"train_samples_per_second": 0.057,
|
| 7 |
+
"train_steps_per_second": 0.014
|
| 8 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 15.0,
|
| 3 |
+
"train_loss": 1.0260957291357629,
|
| 4 |
+
"train_runtime": 82549.8906,
|
| 5 |
+
"train_samples": 313,
|
| 6 |
+
"train_samples_per_second": 0.057,
|
| 7 |
+
"train_steps_per_second": 0.014
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,287 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 15.0,
|
| 5 |
+
"global_step": 1185,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.01,
|
| 12 |
+
"learning_rate": 9e-07,
|
| 13 |
+
"loss": 2.5593,
|
| 14 |
+
"perplexity": 12.926765411958069,
|
| 15 |
+
"step": 1
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 1.0,
|
| 19 |
+
"learning_rate": 9e-07,
|
| 20 |
+
"loss": 2.5731,
|
| 21 |
+
"perplexity": 13.1063913429122,
|
| 22 |
+
"step": 79
|
| 23 |
+
},
|
| 24 |
+
{
|
| 25 |
+
"epoch": 1.0,
|
| 26 |
+
"eval_accuracy": 0.03165881737310309,
|
| 27 |
+
"eval_loss": 2.611328125,
|
| 28 |
+
"eval_perplexity": 13.617124090346826,
|
| 29 |
+
"eval_runtime": 4.9638,
|
| 30 |
+
"eval_samples_per_second": 15.714,
|
| 31 |
+
"eval_steps_per_second": 0.604,
|
| 32 |
+
"step": 79
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 2.0,
|
| 36 |
+
"learning_rate": 9e-07,
|
| 37 |
+
"loss": 2.206,
|
| 38 |
+
"perplexity": 9.079326356062138,
|
| 39 |
+
"step": 158
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 2.0,
|
| 43 |
+
"eval_accuracy": 0.032757718472004184,
|
| 44 |
+
"eval_loss": 2.48046875,
|
| 45 |
+
"eval_perplexity": 11.946863197656471,
|
| 46 |
+
"eval_runtime": 5.388,
|
| 47 |
+
"eval_samples_per_second": 14.477,
|
| 48 |
+
"eval_steps_per_second": 0.557,
|
| 49 |
+
"step": 158
|
| 50 |
+
},
|
| 51 |
+
{
|
| 52 |
+
"epoch": 3.0,
|
| 53 |
+
"learning_rate": 9e-07,
|
| 54 |
+
"loss": 1.9105,
|
| 55 |
+
"perplexity": 6.7564661872073595,
|
| 56 |
+
"step": 237
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 3.0,
|
| 60 |
+
"eval_accuracy": 0.03333333333333333,
|
| 61 |
+
"eval_loss": 2.451171875,
|
| 62 |
+
"eval_perplexity": 11.601934773230344,
|
| 63 |
+
"eval_runtime": 5.3824,
|
| 64 |
+
"eval_samples_per_second": 14.492,
|
| 65 |
+
"eval_steps_per_second": 0.557,
|
| 66 |
+
"step": 237
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 4.0,
|
| 70 |
+
"learning_rate": 9e-07,
|
| 71 |
+
"loss": 1.6301,
|
| 72 |
+
"perplexity": 5.1043851315288045,
|
| 73 |
+
"step": 316
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 4.0,
|
| 77 |
+
"eval_accuracy": 0.03453689167974882,
|
| 78 |
+
"eval_loss": 2.5078125,
|
| 79 |
+
"eval_perplexity": 12.278042445054236,
|
| 80 |
+
"eval_runtime": 5.6801,
|
| 81 |
+
"eval_samples_per_second": 13.732,
|
| 82 |
+
"eval_steps_per_second": 0.528,
|
| 83 |
+
"step": 316
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 5.0,
|
| 87 |
+
"learning_rate": 9e-07,
|
| 88 |
+
"loss": 1.3733,
|
| 89 |
+
"perplexity": 3.948358804340304,
|
| 90 |
+
"step": 395
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 5.0,
|
| 94 |
+
"eval_accuracy": 0.034170591313448455,
|
| 95 |
+
"eval_loss": 2.681640625,
|
| 96 |
+
"eval_perplexity": 14.609041604294383,
|
| 97 |
+
"eval_runtime": 5.3982,
|
| 98 |
+
"eval_samples_per_second": 14.449,
|
| 99 |
+
"eval_steps_per_second": 0.556,
|
| 100 |
+
"step": 395
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 6.0,
|
| 104 |
+
"learning_rate": 9e-07,
|
| 105 |
+
"loss": 1.1337,
|
| 106 |
+
"perplexity": 3.107131645102156,
|
| 107 |
+
"step": 474
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 6.0,
|
| 111 |
+
"eval_accuracy": 0.033019361590790164,
|
| 112 |
+
"eval_loss": 3.0078125,
|
| 113 |
+
"eval_perplexity": 20.24306974171993,
|
| 114 |
+
"eval_runtime": 5.3853,
|
| 115 |
+
"eval_samples_per_second": 14.484,
|
| 116 |
+
"eval_steps_per_second": 0.557,
|
| 117 |
+
"step": 474
|
| 118 |
+
},
|
| 119 |
+
{
|
| 120 |
+
"epoch": 7.0,
|
| 121 |
+
"learning_rate": 9e-07,
|
| 122 |
+
"loss": 0.9619,
|
| 123 |
+
"perplexity": 2.616663413821779,
|
| 124 |
+
"step": 553
|
| 125 |
+
},
|
| 126 |
+
{
|
| 127 |
+
"epoch": 7.0,
|
| 128 |
+
"eval_accuracy": 0.033019361590790164,
|
| 129 |
+
"eval_loss": 3.177734375,
|
| 130 |
+
"eval_perplexity": 23.992334296139543,
|
| 131 |
+
"eval_runtime": 5.6833,
|
| 132 |
+
"eval_samples_per_second": 13.724,
|
| 133 |
+
"eval_steps_per_second": 0.528,
|
| 134 |
+
"step": 553
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 8.0,
|
| 138 |
+
"learning_rate": 9e-07,
|
| 139 |
+
"loss": 0.798,
|
| 140 |
+
"perplexity": 2.221094294751435,
|
| 141 |
+
"step": 632
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 8.0,
|
| 145 |
+
"eval_accuracy": 0.033019361590790164,
|
| 146 |
+
"eval_loss": 3.255859375,
|
| 147 |
+
"eval_perplexity": 25.94189877662824,
|
| 148 |
+
"eval_runtime": 5.3825,
|
| 149 |
+
"eval_samples_per_second": 14.491,
|
| 150 |
+
"eval_steps_per_second": 0.557,
|
| 151 |
+
"step": 632
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 9.0,
|
| 155 |
+
"learning_rate": 9e-07,
|
| 156 |
+
"loss": 0.6653,
|
| 157 |
+
"perplexity": 1.9450739560040564,
|
| 158 |
+
"step": 711
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 9.0,
|
| 162 |
+
"eval_accuracy": 0.033124018838304556,
|
| 163 |
+
"eval_loss": 3.427734375,
|
| 164 |
+
"eval_perplexity": 30.806767041912167,
|
| 165 |
+
"eval_runtime": 5.3793,
|
| 166 |
+
"eval_samples_per_second": 14.5,
|
| 167 |
+
"eval_steps_per_second": 0.558,
|
| 168 |
+
"step": 711
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 10.0,
|
| 172 |
+
"learning_rate": 9e-07,
|
| 173 |
+
"loss": 0.552,
|
| 174 |
+
"perplexity": 1.736722992721326,
|
| 175 |
+
"step": 790
|
| 176 |
+
},
|
| 177 |
+
{
|
| 178 |
+
"epoch": 10.0,
|
| 179 |
+
"eval_accuracy": 0.03333333333333333,
|
| 180 |
+
"eval_loss": 3.556640625,
|
| 181 |
+
"eval_perplexity": 35.04526897307925,
|
| 182 |
+
"eval_runtime": 5.4904,
|
| 183 |
+
"eval_samples_per_second": 14.207,
|
| 184 |
+
"eval_steps_per_second": 0.546,
|
| 185 |
+
"step": 790
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 11.0,
|
| 189 |
+
"learning_rate": 9e-07,
|
| 190 |
+
"loss": 0.4568,
|
| 191 |
+
"perplexity": 1.579013050057063,
|
| 192 |
+
"step": 869
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 11.0,
|
| 196 |
+
"eval_accuracy": 0.032443746729461015,
|
| 197 |
+
"eval_loss": 3.732421875,
|
| 198 |
+
"eval_perplexity": 41.780172086809024,
|
| 199 |
+
"eval_runtime": 5.402,
|
| 200 |
+
"eval_samples_per_second": 14.439,
|
| 201 |
+
"eval_steps_per_second": 0.555,
|
| 202 |
+
"step": 869
|
| 203 |
+
},
|
| 204 |
+
{
|
| 205 |
+
"epoch": 12.0,
|
| 206 |
+
"learning_rate": 9e-07,
|
| 207 |
+
"loss": 0.3756,
|
| 208 |
+
"perplexity": 1.4558646714178145,
|
| 209 |
+
"step": 948
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 12.0,
|
| 213 |
+
"eval_accuracy": 0.032757718472004184,
|
| 214 |
+
"eval_loss": 3.818359375,
|
| 215 |
+
"eval_perplexity": 45.529450258314064,
|
| 216 |
+
"eval_runtime": 5.3901,
|
| 217 |
+
"eval_samples_per_second": 14.471,
|
| 218 |
+
"eval_steps_per_second": 0.557,
|
| 219 |
+
"step": 948
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 13.0,
|
| 223 |
+
"learning_rate": 9e-07,
|
| 224 |
+
"loss": 0.3119,
|
| 225 |
+
"perplexity": 1.3660180843907228,
|
| 226 |
+
"step": 1027
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 13.0,
|
| 230 |
+
"eval_accuracy": 0.033124018838304556,
|
| 231 |
+
"eval_loss": 3.84765625,
|
| 232 |
+
"eval_perplexity": 46.88305220915621,
|
| 233 |
+
"eval_runtime": 5.3723,
|
| 234 |
+
"eval_samples_per_second": 14.519,
|
| 235 |
+
"eval_steps_per_second": 0.558,
|
| 236 |
+
"step": 1027
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 14.0,
|
| 240 |
+
"learning_rate": 9e-07,
|
| 241 |
+
"loss": 0.2448,
|
| 242 |
+
"perplexity": 1.2773658144929685,
|
| 243 |
+
"step": 1106
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 14.0,
|
| 247 |
+
"eval_accuracy": 0.03291470434327577,
|
| 248 |
+
"eval_loss": 3.90625,
|
| 249 |
+
"eval_perplexity": 49.71218131735948,
|
| 250 |
+
"eval_runtime": 5.3964,
|
| 251 |
+
"eval_samples_per_second": 14.454,
|
| 252 |
+
"eval_steps_per_second": 0.556,
|
| 253 |
+
"step": 1106
|
| 254 |
+
},
|
| 255 |
+
{
|
| 256 |
+
"epoch": 15.0,
|
| 257 |
+
"learning_rate": 9e-07,
|
| 258 |
+
"loss": 0.1986,
|
| 259 |
+
"perplexity": 1.219693990715056,
|
| 260 |
+
"step": 1185
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 15.0,
|
| 264 |
+
"eval_accuracy": 0.032862375719518576,
|
| 265 |
+
"eval_loss": 3.943359375,
|
| 266 |
+
"eval_perplexity": 51.591626129191596,
|
| 267 |
+
"eval_runtime": 5.3856,
|
| 268 |
+
"eval_samples_per_second": 14.483,
|
| 269 |
+
"eval_steps_per_second": 0.557,
|
| 270 |
+
"step": 1185
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 15.0,
|
| 274 |
+
"step": 1185,
|
| 275 |
+
"total_flos": 39797194752000.0,
|
| 276 |
+
"train_loss": 1.0260957291357629,
|
| 277 |
+
"train_runtime": 82549.8906,
|
| 278 |
+
"train_samples_per_second": 0.057,
|
| 279 |
+
"train_steps_per_second": 0.014
|
| 280 |
+
}
|
| 281 |
+
],
|
| 282 |
+
"max_steps": 1185,
|
| 283 |
+
"num_train_epochs": 15,
|
| 284 |
+
"total_flos": 39797194752000.0,
|
| 285 |
+
"trial_name": null,
|
| 286 |
+
"trial_params": null
|
| 287 |
+
}
|