lm5-2b8-55b-c4
/
evaluation
/rankeval_s_denoiser
/checkpoints_2b855b55bc4ul2ndfixnew_2_lm-eval_global_step52452_2023-02-09-17-36-57_2shots_backup.json
{ | |
"results": { | |
"anli_r1": { | |
"acc": 0.308, | |
"acc_stderr": 0.014606483127342763 | |
}, | |
"anli_r2": { | |
"acc": 0.353, | |
"acc_stderr": 0.015120172605483696 | |
}, | |
"anli_r3": { | |
"acc": 0.3375, | |
"acc_stderr": 0.013655897185463653 | |
}, | |
"cb": { | |
"acc": 0.4642857142857143, | |
"acc_stderr": 0.06724777654937658, | |
"f1": 0.316548463356974 | |
}, | |
"copa": { | |
"acc": 0.61, | |
"acc_stderr": 0.04902071300001975 | |
}, | |
"hellaswag": { | |
"acc": 0.2929695279824736, | |
"acc_stderr": 0.004541944342035901, | |
"acc_norm": 0.31866162119099783, | |
"acc_norm_stderr": 0.00465005215009441 | |
}, | |
"rte": { | |
"acc": 0.47653429602888087, | |
"acc_stderr": 0.030063300411902652 | |
}, | |
"winogrande": { | |
"acc": 0.5011838989739542, | |
"acc_stderr": 0.014052446290529012 | |
}, | |
"storycloze_2016": { | |
"acc": 0.5836451095670764, | |
"acc_stderr": 0.011399490926937005 | |
}, | |
"boolq": { | |
"acc": 0.4776758409785933, | |
"acc_stderr": 0.00873633411558504 | |
}, | |
"arc_easy": { | |
"acc": 0.42634680134680136, | |
"acc_stderr": 0.010147858603835139, | |
"acc_norm": 0.3926767676767677, | |
"acc_norm_stderr": 0.010020646555538686 | |
}, | |
"arc_challenge": { | |
"acc": 0.18344709897610922, | |
"acc_stderr": 0.011310170179554543, | |
"acc_norm": 0.22696245733788395, | |
"acc_norm_stderr": 0.012240491536132879 | |
}, | |
"sciq": { | |
"acc": 0.73, | |
"acc_stderr": 0.014046255632633915, | |
"acc_norm": 0.677, | |
"acc_norm_stderr": 0.014794927843348633 | |
}, | |
"piqa": { | |
"acc": 0.6349292709466812, | |
"acc_stderr": 0.011233021830554829, | |
"acc_norm": 0.6251360174102285, | |
"acc_norm_stderr": 0.011294565805619019 | |
} | |
}, | |
"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 | |
} | |
} |