lm5-2b8-55b-c4
/
evaluation
/rankeval_s_denoiser_44b
/checkpoints_2b855b55bc4ul2ndfixnew_2_lm-eval_global_step42000_2023-02-08-13-42-29_2shots_backup.json
{ | |
"results": { | |
"anli_r1": { | |
"acc": 0.313, | |
"acc_stderr": 0.014671272822977892 | |
}, | |
"anli_r2": { | |
"acc": 0.336, | |
"acc_stderr": 0.014944140233795023 | |
}, | |
"anli_r3": { | |
"acc": 0.3425, | |
"acc_stderr": 0.013704669762934732 | |
}, | |
"cb": { | |
"acc": 0.42857142857142855, | |
"acc_stderr": 0.06672848092813058, | |
"f1": 0.2791044776119403 | |
}, | |
"copa": { | |
"acc": 0.62, | |
"acc_stderr": 0.04878317312145633 | |
}, | |
"hellaswag": { | |
"acc": 0.2887870942043418, | |
"acc_stderr": 0.004522725412556968, | |
"acc_norm": 0.31517625970922125, | |
"acc_norm_stderr": 0.004636365534819762 | |
}, | |
"rte": { | |
"acc": 0.48014440433212996, | |
"acc_stderr": 0.0300727231673172 | |
}, | |
"winogrande": { | |
"acc": 0.5011838989739542, | |
"acc_stderr": 0.014052446290529012 | |
}, | |
"storycloze_2016": { | |
"acc": 0.5809727418492785, | |
"acc_stderr": 0.011409804749706194 | |
}, | |
"boolq": { | |
"acc": 0.5483180428134556, | |
"acc_stderr": 0.008704126206159355 | |
}, | |
"arc_easy": { | |
"acc": 0.398989898989899, | |
"acc_stderr": 0.010048240683798759, | |
"acc_norm": 0.36784511784511786, | |
"acc_norm_stderr": 0.009894923464455196 | |
}, | |
"arc_challenge": { | |
"acc": 0.18515358361774745, | |
"acc_stderr": 0.011350774438389695, | |
"acc_norm": 0.22781569965870307, | |
"acc_norm_stderr": 0.012256708602326914 | |
}, | |
"sciq": { | |
"acc": 0.717, | |
"acc_stderr": 0.014251810906481735, | |
"acc_norm": 0.634, | |
"acc_norm_stderr": 0.015240612726405756 | |
}, | |
"piqa": { | |
"acc": 0.6327529923830251, | |
"acc_stderr": 0.011247128539690563, | |
"acc_norm": 0.6175190424374319, | |
"acc_norm_stderr": 0.011339019654272345 | |
} | |
}, | |
"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 | |
} | |
} |