lm1-misc-pile
/
421m32b32b
/421m32b32bpile
/evaluation
/rankeval
/lm1-421m-32b-results_lm-eval_global_step60336_2023-01-24-13-53-29_2shots.json
{ | |
"results": { | |
"anli_r1": { | |
"acc": 0.321, | |
"acc_stderr": 0.014770821817934644 | |
}, | |
"anli_r2": { | |
"acc": 0.361, | |
"acc_stderr": 0.015195720118175118 | |
}, | |
"anli_r3": { | |
"acc": 0.3333333333333333, | |
"acc_stderr": 0.013613950010225603 | |
}, | |
"cb": { | |
"acc": 0.39285714285714285, | |
"acc_stderr": 0.0658538889806635, | |
"f1": 0.27449576321756775 | |
}, | |
"copa": { | |
"acc": 0.64, | |
"acc_stderr": 0.04824181513244218 | |
}, | |
"hellaswag": { | |
"acc": 0.3007369049990042, | |
"acc_stderr": 0.0045764127139515, | |
"acc_norm": 0.3345947022505477, | |
"acc_norm_stderr": 0.004708842600177446 | |
}, | |
"rte": { | |
"acc": 0.49458483754512633, | |
"acc_stderr": 0.030094698123239966 | |
}, | |
"winogrande": { | |
"acc": 0.5153906866614049, | |
"acc_stderr": 0.014045826789783665 | |
}, | |
"storycloze_2016": { | |
"acc": 0.5905932656333511, | |
"acc_stderr": 0.01137105952719707 | |
}, | |
"boolq": { | |
"acc": 0.544954128440367, | |
"acc_stderr": 0.008709637955263421 | |
}, | |
"arc_easy": { | |
"acc": 0.4861111111111111, | |
"acc_stderr": 0.010255824507190342, | |
"acc_norm": 0.4659090909090909, | |
"acc_norm_stderr": 0.010235908103438688 | |
}, | |
"arc_challenge": { | |
"acc": 0.20051194539249148, | |
"acc_stderr": 0.011700318050499361, | |
"acc_norm": 0.24146757679180889, | |
"acc_norm_stderr": 0.012506564839739432 | |
}, | |
"sciq": { | |
"acc": 0.817, | |
"acc_stderr": 0.012233587399477825, | |
"acc_norm": 0.81, | |
"acc_norm_stderr": 0.012411851354816329 | |
}, | |
"piqa": { | |
"acc": 0.6425462459194777, | |
"acc_stderr": 0.011181692590867659, | |
"acc_norm": 0.6512513601741022, | |
"acc_norm_stderr": 0.011119263056159599 | |
} | |
}, | |
"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 | |
} | |
} |