decide / eval.json
jmwri's picture
Upload model
14d67ec verified
Raw History Blame Contribute Delete
12.1 kB
{
"ood": {
"overall": {
"n": 4500,
"acc": 0.47555555555555556,
"nll": 1.268476909899631,
"ece": 0.1325593729014183,
"mae": 0.7127850154762857
},
"macro_acc": 0.49183333333333334,
"by_kind": {
"choice": {
"n": 2500,
"acc": 0.4276,
"nll": 1.4152444272854594,
"ece": 0.128002237490073
},
"noul": {
"n": 1000,
"acc": 0.589,
"nll": 1.0108116476597444,
"ece": 0.26637219234174647
},
"score": {
"n": 1000,
"acc": 0.482,
"nll": 1.1592233786749255,
"ece": 0.07211618516297272,
"mae": 0.7127850154762857
}
},
"by_task": {
"amazon_stars": {
"n": 1000,
"acc": 0.482,
"nll": 1.1592233786749255,
"ece": 0.07211618516297272,
"mae": 0.7127850154762857
},
"emotion": {
"n": 1000,
"acc": 0.551,
"nll": 1.2441813441969587,
"ece": 0.08441541454068684
},
"enron_spam": {
"n": 1000,
"acc": 0.589,
"nll": 1.0108116476597444,
"ece": 0.26637219234174647
},
"mmlu": {
"n": 1500,
"acc": 0.3453333333333333,
"nll": 1.5292864826777985,
"ece": 0.16617672448084847
}
}
},
"order_invariance_max_logit_delta": 0.000008940696716308594,
"val": {
"overall": {
"n": 8127,
"acc": 0.801894918173988,
"nll": 0.5053446978458112,
"ece": 0.007099048157460054,
"mae": 0.5636936249569889
},
"macro_acc": 0.7992572561996758,
"by_kind": {
"choice": {
"n": 4745,
"acc": 0.7869336143308746,
"nll": 0.5689453066911272,
"ece": 0.010992410314786963
},
"noul": {
"n": 2703,
"acc": 0.8953015168331484,
"nll": 0.25909313340379886,
"ece": 0.008562793730987436
},
"score": {
"n": 679,
"acc": 0.5346097201767305,
"nll": 1.0411813542754798,
"ece": 0.03380929975803491,
"mae": 0.5636936249569889
}
},
"by_task": {
"amazon_polarity": {
"n": 300,
"acc": 0.94,
"nll": 0.1780728226299769,
"ece": 0.03494007905033705
},
"arc": {
"n": 200,
"acc": 0.49,
"nll": 1.2722349869251168,
"ece": 0.12446486240223996
},
"arc_easy": {
"n": 200,
"acc": 0.66,
"nll": 0.8571748012168005,
"ece": 0.0790243305106568
},
"banking77": {
"n": 300,
"acc": 0.9666666666666667,
"nll": 0.09405613306326095,
"ece": 0.031746609200941224
},
"bitext_support": {
"n": 300,
"acc": 1,
"nll": 0.010971368929279203,
"ece": 0.010659385842467978
},
"boolq": {
"n": 300,
"acc": 0.8333333333333334,
"nll": 0.41208993970329466,
"ece": 0.04317514312264036
},
"civil_comments": {
"n": 156,
"acc": 0.7692307692307693,
"nll": 0.5426374327890833,
"ece": 0.08561364800510647,
"mae": 0.4794184904520537
},
"clinc150": {
"n": 300,
"acc": 0.9833333333333333,
"nll": 0.058032448735896636,
"ece": 0.019049882616036785
},
"cola": {
"n": 200,
"acc": 0.79,
"nll": 0.4192314168142291,
"ece": 0.08574232134607189
},
"commonsense_qa": {
"n": 300,
"acc": 0.65,
"nll": 0.8803430494105633,
"ece": 0.042060654356154394
},
"dbpedia14": {
"n": 300,
"acc": 0.7,
"nll": 1.0272166475279698,
"ece": 0.16287307424578817
},
"go_emotions": {
"n": 256,
"acc": 0.84375,
"nll": 0.4625858118782197,
"ece": 0.07856337055037782
},
"hellaswag": {
"n": 300,
"acc": 0.6866666666666666,
"nll": 0.8184342990797175,
"ece": 0.065393505705077
},
"massive_intent": {
"n": 300,
"acc": 0.97,
"nll": 0.11297520330888683,
"ece": 0.03747435978798287
},
"massive_scenario": {
"n": 300,
"acc": 0.9633333333333334,
"nll": 0.14092133711034824,
"ece": 0.040415555363053655
},
"mnli": {
"n": 300,
"acc": 0.8733333333333333,
"nll": 0.3591158164193244,
"ece": 0.06122319501006837
},
"mrpc": {
"n": 200,
"acc": 0.905,
"nll": 0.2340552890847502,
"ece": 0.0475103336919146
},
"newsgroups": {
"n": 296,
"acc": 0.8547297297297297,
"nll": 0.4397821279908795,
"ece": 0.05815265497795075
},
"openbookqa": {
"n": 250,
"acc": 0.596,
"nll": 0.9566964109317068,
"ece": 0.05584956464154654
},
"qasc": {
"n": 250,
"acc": 0.496,
"nll": 1.3698718341800415,
"ece": 0.05324690519921278
},
"qnli": {
"n": 200,
"acc": 0.905,
"nll": 0.22825732034313698,
"ece": 0.05859514918248248
},
"qqp": {
"n": 200,
"acc": 0.87,
"nll": 0.3305964280804749,
"ece": 0.05731508099827738
},
"rte": {
"n": 200,
"acc": 0.82,
"nll": 0.3908081576970799,
"ece": 0.03793475295601129
},
"sms_spam": {
"n": 300,
"acc": 1,
"nll": 0.009443218337208322,
"ece": 0.009181943972248042
},
"snli": {
"n": 295,
"acc": 0.8745762711864407,
"nll": 0.34642039268536257,
"ece": 0.0733830186710087
},
"sst2": {
"n": 300,
"acc": 0.95,
"nll": 0.17723819437605498,
"ece": 0.04834600996257101
},
"sst5": {
"n": 300,
"acc": 0.5466666666666666,
"nll": 1.034445504429844,
"ece": 0.06904864172392965,
"mae": 0.5531891951338506
},
"stsb": {
"n": 300,
"acc": 0.5133333333333333,
"nll": 1.0930674410680064,
"ece": 0.048921423078040656,
"mae": 0.5963905068664275
},
"toxic_conversations": {
"n": 124,
"acc": 0.9193548387096774,
"nll": 0.25773286368862236,
"ece": 0.05372284727427989
},
"wanli": {
"n": 300,
"acc": 0.7466666666666667,
"nll": 0.5828687631711043,
"ece": 0.04179919773420856
},
"winogrande": {
"n": 300,
"acc": 0.66,
"nll": 0.7154438457054739,
"ece": 0.11005624853642657
}
}
},
"val_uncalibrated": {
"overall": {
"n": 8127,
"acc": 0.801894918173988,
"nll": 0.6020770461865659,
"ece": 0.08041752615679162,
"mae": 0.559288462760455
},
"macro_acc": 0.7992572561996758,
"by_kind": {
"choice": {
"n": 4745,
"acc": 0.7869336143308746,
"nll": 0.713495906024384,
"ece": 0.10104285618425683
},
"noul": {
"n": 2703,
"acc": 0.8953015168331484,
"nll": 0.2901019780777808,
"ece": 0.04708815584016769
},
"score": {
"n": 679,
"acc": 0.5346097201767305,
"nll": 1.0653850272875651,
"ece": 0.07348444458021999,
"mae": 0.559288462760455
}
},
"by_task": {
"amazon_polarity": {
"n": 300,
"acc": 0.94,
"nll": 0.21491849340648492,
"ece": 0.041034815786236065
},
"arc": {
"n": 200,
"acc": 0.49,
"nll": 1.7225013280848958,
"ece": 0.2666289663771016
},
"arc_easy": {
"n": 200,
"acc": 0.66,
"nll": 1.0581853665538725,
"ece": 0.16953352736225474
},
"banking77": {
"n": 300,
"acc": 0.9666666666666667,
"nll": 0.08512822269970122,
"ece": 0.019191444769018
},
"bitext_support": {
"n": 300,
"acc": 1,
"nll": 0.0006662884404404625,
"ece": 0.0006564032820199905
},
"boolq": {
"n": 300,
"acc": 0.8333333333333334,
"nll": 0.46327447408189254,
"ece": 0.09105915635074
},
"civil_comments": {
"n": 156,
"acc": 0.7692307692307693,
"nll": 0.48259455349648295,
"ece": 0.044267076511289985,
"mae": 0.4782040600423823
},
"clinc150": {
"n": 300,
"acc": 0.9833333333333333,
"nll": 0.050976034228883325,
"ece": 0.015613015942019744
},
"cola": {
"n": 200,
"acc": 0.79,
"nll": 0.4599676350741479,
"ece": 0.10238053149913356
},
"commonsense_qa": {
"n": 300,
"acc": 0.65,
"nll": 1.0999147304745942,
"ece": 0.1701816034928976
},
"dbpedia14": {
"n": 300,
"acc": 0.7,
"nll": 1.7263676027906165,
"ece": 0.227155404042094
},
"go_emotions": {
"n": 256,
"acc": 0.84375,
"nll": 0.44446812826453613,
"ece": 0.06292944279854824
},
"hellaswag": {
"n": 300,
"acc": 0.6866666666666666,
"nll": 1.0690494049258386,
"ece": 0.16647033490244928
},
"massive_intent": {
"n": 300,
"acc": 0.97,
"nll": 0.10690323173654996,
"ece": 0.026559626387008747
},
"massive_scenario": {
"n": 300,
"acc": 0.9633333333333334,
"nll": 0.12833034262109785,
"ece": 0.023441601714425293
},
"mnli": {
"n": 300,
"acc": 0.8733333333333333,
"nll": 0.35470827238044367,
"ece": 0.054108644890771114
},
"mrpc": {
"n": 200,
"acc": 0.905,
"nll": 0.22430776762422472,
"ece": 0.04434373851108637
},
"newsgroups": {
"n": 296,
"acc": 0.8547297297297297,
"nll": 0.48702157440611676,
"ece": 0.07240345282869842
},
"openbookqa": {
"n": 250,
"acc": 0.596,
"nll": 1.1414455726990549,
"ece": 0.16806808919887872
},
"qasc": {
"n": 250,
"acc": 0.496,
"nll": 1.716193180040507,
"ece": 0.2341949557110443
},
"qnli": {
"n": 200,
"acc": 0.905,
"nll": 0.2502223326943523,
"ece": 0.05328244333552107
},
"qqp": {
"n": 200,
"acc": 0.87,
"nll": 0.36164874486477994,
"ece": 0.08291620754257616
},
"rte": {
"n": 200,
"acc": 0.82,
"nll": 0.44405080029545163,
"ece": 0.09511886647669884
},
"sms_spam": {
"n": 300,
"acc": 1,
"nll": 0.0015762756577773233,
"ece": 0.0015227966663197983
},
"snli": {
"n": 295,
"acc": 0.8745762711864407,
"nll": 0.34305959824971366,
"ece": 0.0675852776100721
},
"sst2": {
"n": 300,
"acc": 0.95,
"nll": 0.1810899283111296,
"ece": 0.037267897728623225
},
"sst5": {
"n": 300,
"acc": 0.5466666666666666,
"nll": 1.056353078896091,
"ece": 0.07267637287258515,
"mae": 0.5517158101363502
},
"stsb": {
"n": 300,
"acc": 0.5133333333333333,
"nll": 1.1312887822663569,
"ece": 0.10057246284204428,
"mae": 0.5882133414336538
},
"toxic_conversations": {
"n": 124,
"acc": 0.9193548387096774,
"nll": 0.27868695477230904,
"ece": 0.060428655478112825
},
"wanli": {
"n": 300,
"acc": 0.7466666666666667,
"nll": 0.6654510619859225,
"ece": 0.12142146453678089
},
"winogrande": {
"n": 300,
"acc": 0.66,
"nll": 1.0156935419433384,
"ece": 0.19326918465280768
}
}
}
}