lm5-2b8-55b-c4
/
evaluation
/rankeval_s_denoiser_44b
/checkpoints_2b855b55bc4ul2ndfixnew_4_lm-eval_global_step42000_2023-02-08-13-42-29_4shots_backup.json
{ | |
"results": { | |
"anli_r1": { | |
"acc": 0.34, | |
"acc_stderr": 0.014987482264363937 | |
}, | |
"anli_r2": { | |
"acc": 0.341, | |
"acc_stderr": 0.014998131348402697 | |
}, | |
"anli_r3": { | |
"acc": 0.355, | |
"acc_stderr": 0.013819249004047298 | |
}, | |
"cb": { | |
"acc": 0.44642857142857145, | |
"acc_stderr": 0.06703189227942397, | |
"f1": 0.29572649572649573 | |
}, | |
"copa": { | |
"acc": 0.57, | |
"acc_stderr": 0.04975698519562428 | |
}, | |
"hellaswag": { | |
"acc": 0.287293367855009, | |
"acc_stderr": 0.004515748192605717, | |
"acc_norm": 0.3125871340370444, | |
"acc_norm_stderr": 0.004626002828389158 | |
}, | |
"rte": { | |
"acc": 0.4657039711191336, | |
"acc_stderr": 0.030025579819366426 | |
}, | |
"winogrande": { | |
"acc": 0.5169692186266772, | |
"acc_stderr": 0.014044390401612976 | |
}, | |
"storycloze_2016": { | |
"acc": 0.5750935328701229, | |
"acc_stderr": 0.011431286492205843 | |
}, | |
"boolq": { | |
"acc": 0.5513761467889908, | |
"acc_stderr": 0.008698767182005272 | |
}, | |
"arc_easy": { | |
"acc": 0.4057239057239057, | |
"acc_stderr": 0.010075755540128876, | |
"acc_norm": 0.37626262626262624, | |
"acc_norm_stderr": 0.009940646221513786 | |
}, | |
"arc_challenge": { | |
"acc": 0.1962457337883959, | |
"acc_stderr": 0.011606019881416286, | |
"acc_norm": 0.22781569965870307, | |
"acc_norm_stderr": 0.012256708602326905 | |
}, | |
"sciq": { | |
"acc": 0.691, | |
"acc_stderr": 0.014619600977206486, | |
"acc_norm": 0.658, | |
"acc_norm_stderr": 0.01500870618212173 | |
}, | |
"piqa": { | |
"acc": 0.6305767138193689, | |
"acc_stderr": 0.011260988628572341, | |
"acc_norm": 0.6175190424374319, | |
"acc_norm_stderr": 0.011339019654272347 | |
} | |
}, | |
"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 | |
} | |
} |