lm5-2b8-55b-c4
/
evaluation
/rankeval_s_denoiser_44b
/checkpoints_2b855b55bc4ul2ndfixnew_0_lm-eval_global_step42000_2023-02-08-13-42-29_0shots_backup.json
{ | |
"results": { | |
"anli_r1": { | |
"acc": 0.333, | |
"acc_stderr": 0.014910846164229868 | |
}, | |
"anli_r2": { | |
"acc": 0.339, | |
"acc_stderr": 0.014976758771620344 | |
}, | |
"anli_r3": { | |
"acc": 0.33916666666666667, | |
"acc_stderr": 0.013672343491681819 | |
}, | |
"cb": { | |
"acc": 0.4107142857142857, | |
"acc_stderr": 0.0663363415035954, | |
"f1": 0.1940928270042194 | |
}, | |
"copa": { | |
"acc": 0.62, | |
"acc_stderr": 0.04878317312145632 | |
}, | |
"hellaswag": { | |
"acc": 0.2949611631149173, | |
"acc_stderr": 0.004550933142528753, | |
"acc_norm": 0.32463652658832903, | |
"acc_norm_stderr": 0.004672819355838551 | |
}, | |
"rte": { | |
"acc": 0.5342960288808665, | |
"acc_stderr": 0.030025579819366422 | |
}, | |
"winogrande": { | |
"acc": 0.516179952644041, | |
"acc_stderr": 0.014045126130978601 | |
}, | |
"storycloze_2016": { | |
"acc": 0.5873864243719936, | |
"acc_stderr": 0.011384472322969045 | |
}, | |
"boolq": { | |
"acc": 0.6051987767584098, | |
"acc_stderr": 0.008549304887647411 | |
}, | |
"arc_easy": { | |
"acc": 0.42003367003367004, | |
"acc_stderr": 0.010127718838529398, | |
"acc_norm": 0.3728956228956229, | |
"acc_norm_stderr": 0.009922743197129255 | |
}, | |
"arc_challenge": { | |
"acc": 0.18600682593856654, | |
"acc_stderr": 0.011370940183266749, | |
"acc_norm": 0.22610921501706485, | |
"acc_norm_stderr": 0.01222420209706328 | |
}, | |
"sciq": { | |
"acc": 0.735, | |
"acc_stderr": 0.013963164754809949, | |
"acc_norm": 0.656, | |
"acc_norm_stderr": 0.015029633724408945 | |
}, | |
"piqa": { | |
"acc": 0.6251360174102285, | |
"acc_stderr": 0.011294565805619017, | |
"acc_norm": 0.6224156692056583, | |
"acc_norm_stderr": 0.011310782787145772 | |
} | |
}, | |
"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 | |
} | |
} |