Model

Chinchilla-70B

Vendor N/A

Knowledge & Reasoning

Knowledge & Reasoning
BenchmarkScoreMetricHarnessSourceConfidenceLink
BigBench92.8accuracyN/AIndependentmediumarxiv.org
BigBench77.4accuracyN/AIndependentmediumarxiv.org
BoolQ83.7accuracyN/AIndependentmediumarxiv.org
HellaSwag80.8accuracyN/AIndependentmediumarxiv.org
PIQA81.8accuracyN/AIndependentmediumarxiv.org
WinoGrande74.9accuracyN/AIndependentmediumarxiv.org

Reproduced by the community

Loading community runs…