lm5-2b8-55b-c4
/
evaluation
/rankeval_s_denoiser
/checkpoints_2b855b55bc4ul2ndfixnew_0_lm-eval_global_step52452_2023-02-09-17-38-12_0shots_backup.json
{ | |
"results": { | |
"anli_r1": { | |
"acc": 0.339, | |
"acc_stderr": 0.01497675877162034 | |
}, | |
"anli_r2": { | |
"acc": 0.336, | |
"acc_stderr": 0.014944140233795027 | |
}, | |
"anli_r3": { | |
"acc": 0.3358333333333333, | |
"acc_stderr": 0.01363926119093288 | |
}, | |
"cb": { | |
"acc": 0.4107142857142857, | |
"acc_stderr": 0.0663363415035954, | |
"f1": 0.1940928270042194 | |
}, | |
"copa": { | |
"acc": 0.63, | |
"acc_stderr": 0.04852365870939099 | |
}, | |
"hellaswag": { | |
"acc": 0.297450707030472, | |
"acc_stderr": 0.004562022467161891, | |
"acc_norm": 0.32374029077872934, | |
"acc_norm_stderr": 0.004669459891917689 | |
}, | |
"rte": { | |
"acc": 0.5234657039711191, | |
"acc_stderr": 0.03006330041190266 | |
}, | |
"winogrande": { | |
"acc": 0.5090765588003157, | |
"acc_stderr": 0.014050170094497707 | |
}, | |
"storycloze_2016": { | |
"acc": 0.5905932656333511, | |
"acc_stderr": 0.01137105952719707 | |
}, | |
"boolq": { | |
"acc": 0.5944954128440367, | |
"acc_stderr": 0.008587459055441612 | |
}, | |
"arc_easy": { | |
"acc": 0.43308080808080807, | |
"acc_stderr": 0.010167478013701799, | |
"acc_norm": 0.38173400673400676, | |
"acc_norm_stderr": 0.009968648851839667 | |
}, | |
"arc_challenge": { | |
"acc": 0.1885665529010239, | |
"acc_stderr": 0.011430897647675803, | |
"acc_norm": 0.22610921501706485, | |
"acc_norm_stderr": 0.01222420209706328 | |
}, | |
"sciq": { | |
"acc": 0.735, | |
"acc_stderr": 0.013963164754809953, | |
"acc_norm": 0.668, | |
"acc_norm_stderr": 0.014899597242811476 | |
}, | |
"piqa": { | |
"acc": 0.6158868335146899, | |
"acc_stderr": 0.011348160741479148, | |
"acc_norm": 0.6218715995647442, | |
"acc_norm_stderr": 0.011313980666854533 | |
} | |
}, | |
"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 | |
} | |
} |