lm5-2b8-55b-c4
/
evaluation
/rankeval_x_denoiser
/checkpoints_2b855b55bc4ul2ndfixnew_2_lm-eval_global_step52452_2023-02-09-23-08-31_2shots_backup.json
{ | |
"results": { | |
"anli_r1": { | |
"acc": 0.312, | |
"acc_stderr": 0.014658474370509001 | |
}, | |
"anli_r2": { | |
"acc": 0.348, | |
"acc_stderr": 0.01507060460376841 | |
}, | |
"anli_r3": { | |
"acc": 0.3358333333333333, | |
"acc_stderr": 0.013639261190932889 | |
}, | |
"cb": { | |
"acc": 0.44642857142857145, | |
"acc_stderr": 0.06703189227942398, | |
"f1": 0.30579096045197734 | |
}, | |
"copa": { | |
"acc": 0.62, | |
"acc_stderr": 0.04878317312145633 | |
}, | |
"hellaswag": { | |
"acc": 0.29286994622585144, | |
"acc_stderr": 0.004541492151639241, | |
"acc_norm": 0.31736705835490936, | |
"acc_norm_stderr": 0.004645003662067885 | |
}, | |
"rte": { | |
"acc": 0.48375451263537905, | |
"acc_stderr": 0.030080573208738064 | |
}, | |
"winogrande": { | |
"acc": 0.5082872928176796, | |
"acc_stderr": 0.014050555322824189 | |
}, | |
"storycloze_2016": { | |
"acc": 0.5793693212185996, | |
"acc_stderr": 0.011415827994342657 | |
}, | |
"boolq": { | |
"acc": 0.4709480122324159, | |
"acc_stderr": 0.008730280528451546 | |
}, | |
"arc_easy": { | |
"acc": 0.4212962962962963, | |
"acc_stderr": 0.010131882498193131, | |
"acc_norm": 0.3888888888888889, | |
"acc_norm_stderr": 0.010003248335313755 | |
}, | |
"arc_challenge": { | |
"acc": 0.181740614334471, | |
"acc_stderr": 0.011269198948880236, | |
"acc_norm": 0.22866894197952217, | |
"acc_norm_stderr": 0.0122728535825408 | |
}, | |
"sciq": { | |
"acc": 0.727, | |
"acc_stderr": 0.014095022868717605, | |
"acc_norm": 0.676, | |
"acc_norm_stderr": 0.01480686473373886 | |
}, | |
"piqa": { | |
"acc": 0.6349292709466812, | |
"acc_stderr": 0.011233021830554826, | |
"acc_norm": 0.6240478781284005, | |
"acc_norm_stderr": 0.011301098166895727 | |
} | |
}, | |
"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 | |
} | |
} |