lm5-2b8-55b-c4
/
evaluation
/rankeval_x_denoiser
/checkpoints_2b855b55bc4ul2ndfixnew_3_lm-eval_global_step52452_2023-02-09-23-08-31_3shots_backup.json
{ | |
"results": { | |
"anli_r1": { | |
"acc": 0.314, | |
"acc_stderr": 0.014683991951087967 | |
}, | |
"anli_r2": { | |
"acc": 0.343, | |
"acc_stderr": 0.015019206922356951 | |
}, | |
"anli_r3": { | |
"acc": 0.3458333333333333, | |
"acc_stderr": 0.013736245342311012 | |
}, | |
"cb": { | |
"acc": 0.48214285714285715, | |
"acc_stderr": 0.0673769750864465, | |
"f1": 0.3218559218559219 | |
}, | |
"copa": { | |
"acc": 0.62, | |
"acc_stderr": 0.04878317312145633 | |
}, | |
"hellaswag": { | |
"acc": 0.29286994622585144, | |
"acc_stderr": 0.004541492151639243, | |
"acc_norm": 0.3227444732125075, | |
"acc_norm_stderr": 0.004665704208339039 | |
}, | |
"rte": { | |
"acc": 0.5054151624548736, | |
"acc_stderr": 0.030094698123239966 | |
}, | |
"winogrande": { | |
"acc": 0.4925019731649566, | |
"acc_stderr": 0.01405090552122858 | |
}, | |
"storycloze_2016": { | |
"acc": 0.5729556386958845, | |
"acc_stderr": 0.01143868739857839 | |
}, | |
"boolq": { | |
"acc": 0.4675840978593272, | |
"acc_stderr": 0.008726657178723137 | |
}, | |
"arc_easy": { | |
"acc": 0.42297979797979796, | |
"acc_stderr": 0.010137328382209099, | |
"acc_norm": 0.3808922558922559, | |
"acc_norm_stderr": 0.009964428212260379 | |
}, | |
"arc_challenge": { | |
"acc": 0.19197952218430034, | |
"acc_stderr": 0.011509598906598098, | |
"acc_norm": 0.22525597269624573, | |
"acc_norm_stderr": 0.0122078399954073 | |
}, | |
"sciq": { | |
"acc": 0.72, | |
"acc_stderr": 0.014205696104091496, | |
"acc_norm": 0.679, | |
"acc_norm_stderr": 0.014770821817934645 | |
}, | |
"piqa": { | |
"acc": 0.6273122959738846, | |
"acc_stderr": 0.011281318332897734, | |
"acc_norm": 0.6202393906420022, | |
"acc_norm_stderr": 0.01132348350471584 | |
} | |
}, | |
"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 | |
} | |
} |