lm1-misc-pile
/
1b566b66b
/1b566b66bpile
/evaluation
/rankeval
/lm1-1b5-66b-results_lm-eval_global_step125429_2023-01-24-13-57-04_3shots.json
{ | |
"results": { | |
"anli_r1": { | |
"acc": 0.309, | |
"acc_stderr": 0.014619600977206493 | |
}, | |
"anli_r2": { | |
"acc": 0.347, | |
"acc_stderr": 0.015060472031706622 | |
}, | |
"anli_r3": { | |
"acc": 0.3541666666666667, | |
"acc_stderr": 0.013811933499570963 | |
}, | |
"cb": { | |
"acc": 0.5178571428571429, | |
"acc_stderr": 0.06737697508644648, | |
"f1": 0.45082590365609226 | |
}, | |
"copa": { | |
"acc": 0.73, | |
"acc_stderr": 0.0446196043338474 | |
}, | |
"hellaswag": { | |
"acc": 0.36666002788289187, | |
"acc_stderr": 0.0048090772053434976, | |
"acc_norm": 0.4627564230233021, | |
"acc_norm_stderr": 0.004975919665116536 | |
}, | |
"rte": { | |
"acc": 0.5487364620938628, | |
"acc_stderr": 0.029953149241808943 | |
}, | |
"winogrande": { | |
"acc": 0.5382794001578532, | |
"acc_stderr": 0.014011242594964115 | |
}, | |
"storycloze_2016": { | |
"acc": 0.6648850881881346, | |
"acc_stderr": 0.010915644164980044 | |
}, | |
"boolq": { | |
"acc": 0.5779816513761468, | |
"acc_stderr": 0.008638040428462952 | |
}, | |
"arc_easy": { | |
"acc": 0.5812289562289562, | |
"acc_stderr": 0.010123487160167812, | |
"acc_norm": 0.57996632996633, | |
"acc_norm_stderr": 0.010127718838529315 | |
}, | |
"arc_challenge": { | |
"acc": 0.257679180887372, | |
"acc_stderr": 0.012780770562768402, | |
"acc_norm": 0.29180887372013653, | |
"acc_norm_stderr": 0.013284525292403506 | |
}, | |
"sciq": { | |
"acc": 0.919, | |
"acc_stderr": 0.008632121032139992, | |
"acc_norm": 0.92, | |
"acc_norm_stderr": 0.008583336977753653 | |
}, | |
"piqa": { | |
"acc": 0.6931447225244831, | |
"acc_stderr": 0.010760295070580366, | |
"acc_norm": 0.7013057671381937, | |
"acc_norm_stderr": 0.010678556398149233 | |
} | |
}, | |
"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 | |
} | |
} |