lm1-4b2-84b-c4seeds
/
4b284b42bc4seed2
/evaluation
/rankeval
/4b284b42bc4seed2_4_lm-eval_global_step80108_2023-02-15-11-04-03_4shots_backup.json
{ | |
"results": { | |
"anli_r1": { | |
"acc": 0.324, | |
"acc_stderr": 0.014806864733738856 | |
}, | |
"anli_r2": { | |
"acc": 0.359, | |
"acc_stderr": 0.015177264224798596 | |
}, | |
"anli_r3": { | |
"acc": 0.3466666666666667, | |
"acc_stderr": 0.013744022550571952 | |
}, | |
"cb": { | |
"acc": 0.3392857142857143, | |
"acc_stderr": 0.06384226561930827, | |
"f1": 0.2379084967320261 | |
}, | |
"copa": { | |
"acc": 0.77, | |
"acc_stderr": 0.04229525846816506 | |
}, | |
"hellaswag": { | |
"acc": 0.46793467436765585, | |
"acc_stderr": 0.004979510001776621, | |
"acc_norm": 0.6213901613224457, | |
"acc_norm_stderr": 0.004840493603166203 | |
}, | |
"rte": { | |
"acc": 0.49458483754512633, | |
"acc_stderr": 0.030094698123239966 | |
}, | |
"winogrande": { | |
"acc": 0.5816890292028414, | |
"acc_stderr": 0.013863669961195911 | |
}, | |
"storycloze_2016": { | |
"acc": 0.7226082308925709, | |
"acc_stderr": 0.010353267472010767 | |
}, | |
"boolq": { | |
"acc": 0.5278287461773701, | |
"acc_stderr": 0.008731499445069549 | |
}, | |
"arc_easy": { | |
"acc": 0.6195286195286195, | |
"acc_stderr": 0.009962305992058577, | |
"acc_norm": 0.6052188552188552, | |
"acc_norm_stderr": 0.010030038935883598 | |
}, | |
"arc_challenge": { | |
"acc": 0.27474402730375425, | |
"acc_stderr": 0.013044617212771227, | |
"acc_norm": 0.30204778156996587, | |
"acc_norm_stderr": 0.01341751914471642 | |
}, | |
"sciq": { | |
"acc": 0.892, | |
"acc_stderr": 0.0098200016513457, | |
"acc_norm": 0.889, | |
"acc_norm_stderr": 0.009938701010583726 | |
}, | |
"piqa": { | |
"acc": 0.7568008705114254, | |
"acc_stderr": 0.010009611953858922, | |
"acc_norm": 0.7671381936887922, | |
"acc_norm_stderr": 0.009861236071080757 | |
} | |
}, | |
"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 | |
} | |
} |