lm1-4b2-84b-c4seeds
/
4b284b42bc4seed2
/evaluation
/rankeval
/4b284b42bc4seed2_1_lm-eval_global_step80108_2023-02-15-11-04-03_1shots_backup.json
{ | |
"results": { | |
"anli_r1": { | |
"acc": 0.335, | |
"acc_stderr": 0.014933117490932575 | |
}, | |
"anli_r2": { | |
"acc": 0.326, | |
"acc_stderr": 0.014830507204541033 | |
}, | |
"anli_r3": { | |
"acc": 0.3441666666666667, | |
"acc_stderr": 0.013720551062295756 | |
}, | |
"cb": { | |
"acc": 0.4107142857142857, | |
"acc_stderr": 0.0663363415035954, | |
"f1": 0.28917378917378916 | |
}, | |
"copa": { | |
"acc": 0.78, | |
"acc_stderr": 0.04163331998932262 | |
}, | |
"hellaswag": { | |
"acc": 0.4714200358494324, | |
"acc_stderr": 0.00498162329219619, | |
"acc_norm": 0.6203943437562238, | |
"acc_norm_stderr": 0.004842969887794082 | |
}, | |
"rte": { | |
"acc": 0.51985559566787, | |
"acc_stderr": 0.030072723167317177 | |
}, | |
"winogrande": { | |
"acc": 0.5682715074980268, | |
"acc_stderr": 0.01392087211001071 | |
}, | |
"storycloze_2016": { | |
"acc": 0.7049706039551042, | |
"acc_stderr": 0.010546232606962283 | |
}, | |
"boolq": { | |
"acc": 0.5685015290519878, | |
"acc_stderr": 0.008662594569027316 | |
}, | |
"arc_easy": { | |
"acc": 0.6132154882154882, | |
"acc_stderr": 0.009993308355370968, | |
"acc_norm": 0.5774410774410774, | |
"acc_norm_stderr": 0.010135978222981071 | |
}, | |
"arc_challenge": { | |
"acc": 0.2713310580204778, | |
"acc_stderr": 0.012993807727545794, | |
"acc_norm": 0.302901023890785, | |
"acc_norm_stderr": 0.013428241573185349 | |
}, | |
"sciq": { | |
"acc": 0.868, | |
"acc_stderr": 0.010709373963528012, | |
"acc_norm": 0.841, | |
"acc_norm_stderr": 0.0115694793682713 | |
}, | |
"piqa": { | |
"acc": 0.7464635473340587, | |
"acc_stderr": 0.010150090834551794, | |
"acc_norm": 0.749183895538629, | |
"acc_norm_stderr": 0.010113869547069046 | |
} | |
}, | |
"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 | |
} | |
} |