lm5-2b8-55b-c4
/
evaluation
/rankeval_x_denoiser
/checkpoints_2b855b55bc4ul2ndfixnew_5_lm-eval_global_step52452_2023-02-09-23-08-31_5shots_backup.json
{ | |
"results": { | |
"anli_r1": { | |
"acc": 0.37, | |
"acc_stderr": 0.015275252316519362 | |
}, | |
"anli_r2": { | |
"acc": 0.341, | |
"acc_stderr": 0.014998131348402706 | |
}, | |
"anli_r3": { | |
"acc": 0.345, | |
"acc_stderr": 0.013728421539454878 | |
}, | |
"cb": { | |
"acc": 0.5535714285714286, | |
"acc_stderr": 0.06703189227942395, | |
"f1": 0.3464373464373464 | |
}, | |
"copa": { | |
"acc": 0.64, | |
"acc_stderr": 0.048241815132442176 | |
}, | |
"hellaswag": { | |
"acc": 0.2952599083847839, | |
"acc_stderr": 0.004552272447071696, | |
"acc_norm": 0.32374029077872934, | |
"acc_norm_stderr": 0.004669459891917693 | |
}, | |
"rte": { | |
"acc": 0.4981949458483754, | |
"acc_stderr": 0.030096267148976633 | |
}, | |
"winogrande": { | |
"acc": 0.5011838989739542, | |
"acc_stderr": 0.014052446290529012 | |
}, | |
"storycloze_2016": { | |
"acc": 0.5740245857830037, | |
"acc_stderr": 0.0114350142621812 | |
}, | |
"boolq": { | |
"acc": 0.44250764525993885, | |
"acc_stderr": 0.00868705131518137 | |
}, | |
"arc_easy": { | |
"acc": 0.4292929292929293, | |
"acc_stderr": 0.010156678075911097, | |
"acc_norm": 0.39604377104377103, | |
"acc_norm_stderr": 0.01003558096209795 | |
}, | |
"arc_challenge": { | |
"acc": 0.19112627986348124, | |
"acc_stderr": 0.011490055292778585, | |
"acc_norm": 0.22525597269624573, | |
"acc_norm_stderr": 0.012207839995407305 | |
}, | |
"sciq": { | |
"acc": 0.717, | |
"acc_stderr": 0.014251810906481735, | |
"acc_norm": 0.68, | |
"acc_norm_stderr": 0.014758652303574885 | |
}, | |
"piqa": { | |
"acc": 0.6305767138193689, | |
"acc_stderr": 0.011260988628572343, | |
"acc_norm": 0.6147986942328618, | |
"acc_norm_stderr": 0.011354179751257073 | |
} | |
}, | |
"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 | |
} | |
} |