Model

MPT 30B

MosaicML

Knowledge & Reasoning

Knowledge & Reasoning
BenchmarkScoreMetricHarnessSourceConfidenceLink
BBH38.0averageN/AIndependentmediumarxiv.org
BoolQ85.0scoreN/AIndependentmediumcrfm.stanford.edu
MMLU47.9exact_matchN/AIndependentmediumarxiv.org
OpenBookQA52.0accuracyN/AIndependentmediumarxiv.org
PIQA81.9scoreN/AIndependentmediumarxiv.org
Trivia QA73.6exact_matchN/AIndependentmediumarxiv.org
WinoGrande71.0accuracyN/AIndependentmediumarxiv.org

Math

Math
BenchmarkScoreMetricHarnessSourceConfidenceLink
GSM8K34.4exact_matchN/AIndependentmediumcrfm.stanford.edu

Reproduced by the community

Loading community runs…