Model

Megatron-Turing NLG 530B

Microsoft,Nvidia

Knowledge & Reasoning

Knowledge & Reasoning
BenchmarkScoreMetricHarnessSourceConfidenceLink
ANLI39.7scoreN/ASelf-reportedmediumarxiv.org
BoolQ84.8scoreN/ASelf-reportedmediumarxiv.org
BoolQ78.2scoreN/AIndependentmediumarxiv.org
HellaSwag82.4overall_accuracyN/AIndependentmediumarxiv.org
HellaSwag82.4overall_accuracyN/ASelf-reportedmediumarxiv.org
LAMBADA87.2scoreN/ASelf-reportedmediumarxiv.org
LAMBADA76.6scoreN/AIndependentmediumarxiv.org
PIQA83.2scoreN/ASelf-reportedmediumarxiv.org
PIQA82.0scoreN/AIndependentmediumarxiv.org
WinoGrande78.9accuracyN/AIndependentmediumarxiv.org
WinoGrande73.7accuracyN/ASelf-reportedmediumarxiv.org

Reproduced by the community

Loading community runs…