lm5-2b8-55b-c4
/
evaluation
/rankeval_r_denoiser
/checkpoints_2b855b55bc4ul2ndfixnew_3_lm-eval_global_step52452_2023-02-09-23-24-23_3shots_backup.json
{ | |
"results": { | |
"anli_r1": { | |
"acc": 0.312, | |
"acc_stderr": 0.014658474370509008 | |
}, | |
"anli_r2": { | |
"acc": 0.344, | |
"acc_stderr": 0.015029633724408947 | |
}, | |
"anli_r3": { | |
"acc": 0.3425, | |
"acc_stderr": 0.013704669762934725 | |
}, | |
"cb": { | |
"acc": 0.5178571428571429, | |
"acc_stderr": 0.06737697508644647, | |
"f1": 0.3422885572139303 | |
}, | |
"copa": { | |
"acc": 0.6, | |
"acc_stderr": 0.049236596391733084 | |
}, | |
"hellaswag": { | |
"acc": 0.291575383389763, | |
"acc_stderr": 0.004535589759202657, | |
"acc_norm": 0.32284405496912966, | |
"acc_norm_stderr": 0.004666080865179641 | |
}, | |
"rte": { | |
"acc": 0.5054151624548736, | |
"acc_stderr": 0.030094698123239966 | |
}, | |
"winogrande": { | |
"acc": 0.5011838989739542, | |
"acc_stderr": 0.014052446290529015 | |
}, | |
"storycloze_2016": { | |
"acc": 0.5740245857830037, | |
"acc_stderr": 0.011435014262181197 | |
}, | |
"boolq": { | |
"acc": 0.4688073394495413, | |
"acc_stderr": 0.008728020822889253 | |
}, | |
"arc_easy": { | |
"acc": 0.42424242424242425, | |
"acc_stderr": 0.010141333654958574, | |
"acc_norm": 0.38425925925925924, | |
"acc_norm_stderr": 0.009981120724601443 | |
}, | |
"arc_challenge": { | |
"acc": 0.18686006825938567, | |
"acc_stderr": 0.011391015649694391, | |
"acc_norm": 0.22440273037542663, | |
"acc_norm_stderr": 0.012191404938603838 | |
}, | |
"sciq": { | |
"acc": 0.723, | |
"acc_stderr": 0.014158794845306265, | |
"acc_norm": 0.682, | |
"acc_norm_stderr": 0.014734079309311901 | |
}, | |
"piqa": { | |
"acc": 0.6284004352557128, | |
"acc_stderr": 0.011274603006724743, | |
"acc_norm": 0.6196953210010882, | |
"acc_norm_stderr": 0.011326620892570314 | |
} | |
}, | |
"versions": { | |
"anli_r1": 0, | |
"anli_r2": 0, | |
"anli_r3": 0, | |
"cb": 1, | |
"copa": 0, | |
"hellaswag": 0, | |
"rte": 0, | |
"winogrande": 0, | |
"storycloze_2016": 0, | |
"boolq": 1, | |
"arc_easy": 0, | |
"arc_challenge": 0, | |
"sciq": 0, | |
"piqa": 0 | |
} | |
} |