lm5-2b8-55b-c4
/
evaluation
/rankeval_x_denoiser
/checkpoints_2b855b55bc4ul2ndfixnew_4_lm-eval_global_step52452_2023-02-09-23-08-31_4shots_backup.json
{ | |
"results": { | |
"anli_r1": { | |
"acc": 0.338, | |
"acc_stderr": 0.014965960710224482 | |
}, | |
"anli_r2": { | |
"acc": 0.346, | |
"acc_stderr": 0.015050266127564446 | |
}, | |
"anli_r3": { | |
"acc": 0.33916666666666667, | |
"acc_stderr": 0.013672343491681817 | |
}, | |
"cb": { | |
"acc": 0.5178571428571429, | |
"acc_stderr": 0.06737697508644645, | |
"f1": 0.3362023995826813 | |
}, | |
"copa": { | |
"acc": 0.61, | |
"acc_stderr": 0.04902071300001975 | |
}, | |
"hellaswag": { | |
"acc": 0.2954590718980283, | |
"acc_stderr": 0.004553164013379555, | |
"acc_norm": 0.32443736307508464, | |
"acc_norm_stderr": 0.004672074496749016 | |
}, | |
"rte": { | |
"acc": 0.48375451263537905, | |
"acc_stderr": 0.030080573208738064 | |
}, | |
"winogrande": { | |
"acc": 0.4846093133385951, | |
"acc_stderr": 0.014045826789783656 | |
}, | |
"storycloze_2016": { | |
"acc": 0.5734901122394441, | |
"acc_stderr": 0.011436857656268697 | |
}, | |
"boolq": { | |
"acc": 0.44587155963302755, | |
"acc_stderr": 0.008693659886486845 | |
}, | |
"arc_easy": { | |
"acc": 0.42634680134680136, | |
"acc_stderr": 0.010147858603835144, | |
"acc_norm": 0.39057239057239057, | |
"acc_norm_stderr": 0.010011059112064239 | |
}, | |
"arc_challenge": { | |
"acc": 0.19539249146757678, | |
"acc_stderr": 0.011586907189952911, | |
"acc_norm": 0.23037542662116042, | |
"acc_norm_stderr": 0.012304928418747611 | |
}, | |
"sciq": { | |
"acc": 0.718, | |
"acc_stderr": 0.014236526215291334, | |
"acc_norm": 0.687, | |
"acc_norm_stderr": 0.014671272822977883 | |
}, | |
"piqa": { | |
"acc": 0.6289445048966268, | |
"acc_stderr": 0.011271222398600525, | |
"acc_norm": 0.6218715995647442, | |
"acc_norm_stderr": 0.011313980666854535 | |
} | |
}, | |
"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 | |
} | |
} |