lm1-4b2-84b-c4seeds
/
4b284b21bc4seed4
/evaluation
/rankeval
/4b284b21bc4seed4_2_lm-eval_global_step80108_2023-02-15-11-04-02_2shots_backup.json
{ | |
"results": { | |
"anli_r1": { | |
"acc": 0.315, | |
"acc_stderr": 0.0146966319607925 | |
}, | |
"anli_r2": { | |
"acc": 0.326, | |
"acc_stderr": 0.01483050720454104 | |
}, | |
"anli_r3": { | |
"acc": 0.3466666666666667, | |
"acc_stderr": 0.013744022550571944 | |
}, | |
"cb": { | |
"acc": 0.39285714285714285, | |
"acc_stderr": 0.0658538889806635, | |
"f1": 0.2606516290726817 | |
}, | |
"copa": { | |
"acc": 0.74, | |
"acc_stderr": 0.044084400227680794 | |
}, | |
"hellaswag": { | |
"acc": 0.4737104162517427, | |
"acc_stderr": 0.004982879340691406, | |
"acc_norm": 0.6329416450906195, | |
"acc_norm_stderr": 0.004810175357870948 | |
}, | |
"rte": { | |
"acc": 0.5415162454873647, | |
"acc_stderr": 0.029992535385373317 | |
}, | |
"winogrande": { | |
"acc": 0.5966850828729282, | |
"acc_stderr": 0.013787257285896241 | |
}, | |
"storycloze_2016": { | |
"acc": 0.7258150721539284, | |
"acc_stderr": 0.010316062787590001 | |
}, | |
"boolq": { | |
"acc": 0.5715596330275229, | |
"acc_stderr": 0.008655028561519765 | |
}, | |
"arc_easy": { | |
"acc": 0.6216329966329966, | |
"acc_stderr": 0.009951575683331949, | |
"acc_norm": 0.5921717171717171, | |
"acc_norm_stderr": 0.010083950240041223 | |
}, | |
"arc_challenge": { | |
"acc": 0.2960750853242321, | |
"acc_stderr": 0.013340916085246254, | |
"acc_norm": 0.310580204778157, | |
"acc_norm_stderr": 0.013522292098053059 | |
}, | |
"sciq": { | |
"acc": 0.894, | |
"acc_stderr": 0.009739551265785133, | |
"acc_norm": 0.871, | |
"acc_norm_stderr": 0.010605256784796577 | |
}, | |
"piqa": { | |
"acc": 0.7568008705114254, | |
"acc_stderr": 0.010009611953858917, | |
"acc_norm": 0.764961915125136, | |
"acc_norm_stderr": 0.009893146688805306 | |
} | |
}, | |
"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 | |
} | |
} |