Model

StableBeluga2

Stability AI

Knowledge & Reasoning

Knowledge & Reasoning
BenchmarkScoreMetricHarnessSourceConfidenceLink
BBH69.3averageN/AIndependentmediumarxiv.org
BoolQ89.4scoreN/AIndependentmediumarxiv.org
HellaSwag84.1overall_accuracyN/AIndependentmediumarxiv.org
LAMBADA71.3scoreN/AIndependentmediumarxiv.org
MMLU68.6exact_matchN/AIndependentmediumarxiv.org
PIQA83.3scoreN/AIndependentmediumarxiv.org

Math

Math
BenchmarkScoreMetricHarnessSourceConfidenceLink
GSM8K69.6exact_matchN/AIndependentmediumarxiv.org

Reproduced by the community

Loading community runs…