lm1-misc-pile
/
619m22b22b
/evaluation
/rankeval
/lm1-619m-22b-results_lm-eval_global_step41007_2023-01-24-17-05-33_5shots.json
{ | |
"results": { | |
"anli_r1": { | |
"acc": 0.313, | |
"acc_stderr": 0.014671272822977892 | |
}, | |
"anli_r2": { | |
"acc": 0.324, | |
"acc_stderr": 0.014806864733738856 | |
}, | |
"anli_r3": { | |
"acc": 0.3383333333333333, | |
"acc_stderr": 0.01366414400661827 | |
}, | |
"cb": { | |
"acc": 0.39285714285714285, | |
"acc_stderr": 0.0658538889806635, | |
"f1": 0.2593406593406593 | |
}, | |
"copa": { | |
"acc": 0.68, | |
"acc_stderr": 0.04688261722621505 | |
}, | |
"hellaswag": { | |
"acc": 0.3018323043218482, | |
"acc_stderr": 0.0045811472479631975, | |
"acc_norm": 0.33608842859988053, | |
"acc_norm_stderr": 0.004714041652598617 | |
}, | |
"rte": { | |
"acc": 0.5523465703971119, | |
"acc_stderr": 0.02993107036293953 | |
}, | |
"winogrande": { | |
"acc": 0.510655090765588, | |
"acc_stderr": 0.014049294536290396 | |
}, | |
"storycloze_2016": { | |
"acc": 0.5916622127204704, | |
"acc_stderr": 0.011366477562142522 | |
}, | |
"boolq": { | |
"acc": 0.518960244648318, | |
"acc_stderr": 0.008738765179491936 | |
}, | |
"arc_easy": { | |
"acc": 0.4983164983164983, | |
"acc_stderr": 0.010259725364582795, | |
"acc_norm": 0.47685185185185186, | |
"acc_norm_stderr": 0.010248782484554474 | |
}, | |
"arc_challenge": { | |
"acc": 0.21245733788395904, | |
"acc_stderr": 0.011953482906582954, | |
"acc_norm": 0.2440273037542662, | |
"acc_norm_stderr": 0.012551447627856253 | |
}, | |
"sciq": { | |
"acc": 0.852, | |
"acc_stderr": 0.011234866364235237, | |
"acc_norm": 0.832, | |
"acc_norm_stderr": 0.01182860583145427 | |
}, | |
"piqa": { | |
"acc": 0.6458106637649619, | |
"acc_stderr": 0.011158755672626112, | |
"acc_norm": 0.6474428726877041, | |
"acc_norm_stderr": 0.011147074365010456 | |
} | |
}, | |
"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 | |
} | |
} |