lm5-2b8-55b-c4
/
evaluation
/rankeval_s_denoiser
/checkpoints_2b855b55bc4ul2ndfixnew_1_lm-eval_global_step52452_2023-02-09-17-38-12_1shots_backup.json
{ | |
"results": { | |
"anli_r1": { | |
"acc": 0.358, | |
"acc_stderr": 0.015167928865407559 | |
}, | |
"anli_r2": { | |
"acc": 0.35, | |
"acc_stderr": 0.015090650341444236 | |
}, | |
"anli_r3": { | |
"acc": 0.32, | |
"acc_stderr": 0.013471620929769152 | |
}, | |
"cb": { | |
"acc": 0.4107142857142857, | |
"acc_stderr": 0.0663363415035954, | |
"f1": 0.28651292802236195 | |
}, | |
"copa": { | |
"acc": 0.68, | |
"acc_stderr": 0.04688261722621504 | |
}, | |
"hellaswag": { | |
"acc": 0.2930691097390958, | |
"acc_stderr": 0.004542396269999213, | |
"acc_norm": 0.3207528380800637, | |
"acc_norm_stderr": 0.004658120152230808 | |
}, | |
"rte": { | |
"acc": 0.5631768953068592, | |
"acc_stderr": 0.02985524739031495 | |
}, | |
"winogrande": { | |
"acc": 0.4964483030781373, | |
"acc_stderr": 0.01405213114691586 | |
}, | |
"storycloze_2016": { | |
"acc": 0.5793693212185996, | |
"acc_stderr": 0.01141582799434265 | |
}, | |
"boolq": { | |
"acc": 0.5125382262996941, | |
"acc_stderr": 0.008742304974218311 | |
}, | |
"arc_easy": { | |
"acc": 0.4313973063973064, | |
"acc_stderr": 0.010162752847747498, | |
"acc_norm": 0.38341750841750843, | |
"acc_norm_stderr": 0.009976995068264717 | |
}, | |
"arc_challenge": { | |
"acc": 0.19027303754266212, | |
"acc_stderr": 0.011470424179225709, | |
"acc_norm": 0.22610921501706485, | |
"acc_norm_stderr": 0.012224202097063274 | |
}, | |
"sciq": { | |
"acc": 0.704, | |
"acc_stderr": 0.014442734941575022, | |
"acc_norm": 0.658, | |
"acc_norm_stderr": 0.015008706182121731 | |
}, | |
"piqa": { | |
"acc": 0.6240478781284005, | |
"acc_stderr": 0.011301098166895732, | |
"acc_norm": 0.6158868335146899, | |
"acc_norm_stderr": 0.011348160741479136 | |
} | |
}, | |
"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 | |
} | |
} |