lm1-4b2-84b-c4seeds
/
4b284b21bc4seed4
/evaluation
/rankeval
/4b284b21bc4seed4_1_lm-eval_global_step80108_2023-02-15-11-04-03_1shots_backup.json
{ | |
"results": { | |
"anli_r1": { | |
"acc": 0.315, | |
"acc_stderr": 0.014696631960792498 | |
}, | |
"anli_r2": { | |
"acc": 0.339, | |
"acc_stderr": 0.01497675877162034 | |
}, | |
"anli_r3": { | |
"acc": 0.3541666666666667, | |
"acc_stderr": 0.01381193349957096 | |
}, | |
"cb": { | |
"acc": 0.4642857142857143, | |
"acc_stderr": 0.06724777654937658, | |
"f1": 0.3333333333333333 | |
}, | |
"copa": { | |
"acc": 0.75, | |
"acc_stderr": 0.04351941398892446 | |
}, | |
"hellaswag": { | |
"acc": 0.47759410476000796, | |
"acc_stderr": 0.004984768912326931, | |
"acc_norm": 0.6308504282015535, | |
"acc_norm_stderr": 0.004815882719278391 | |
}, | |
"rte": { | |
"acc": 0.5523465703971119, | |
"acc_stderr": 0.029931070362939533 | |
}, | |
"winogrande": { | |
"acc": 0.5785319652722968, | |
"acc_stderr": 0.013878072377497597 | |
}, | |
"storycloze_2016": { | |
"acc": 0.711918760021379, | |
"acc_stderr": 0.01047253701982257 | |
}, | |
"boolq": { | |
"acc": 0.5718654434250765, | |
"acc_stderr": 0.008654253415781077 | |
}, | |
"arc_easy": { | |
"acc": 0.6186868686868687, | |
"acc_stderr": 0.009966542497171018, | |
"acc_norm": 0.571969696969697, | |
"acc_norm_stderr": 0.010152943316426265 | |
}, | |
"arc_challenge": { | |
"acc": 0.31143344709897613, | |
"acc_stderr": 0.013532472099850942, | |
"acc_norm": 0.31313993174061433, | |
"acc_norm_stderr": 0.013552671543623503 | |
}, | |
"sciq": { | |
"acc": 0.878, | |
"acc_stderr": 0.010354864712936706, | |
"acc_norm": 0.852, | |
"acc_norm_stderr": 0.01123486636423524 | |
}, | |
"piqa": { | |
"acc": 0.7568008705114254, | |
"acc_stderr": 0.010009611953858914, | |
"acc_norm": 0.7589771490750816, | |
"acc_norm_stderr": 0.009979042717267315 | |
} | |
}, | |
"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 | |
} | |
} |