Model

Starcoder2 3B

Hugging Face,ServiceNow,Nvidia,BigCode

Knowledge & Reasoning

Knowledge & Reasoning
BenchmarkScoreMetricHarnessSourceConfidenceLink
MMLU36.6exact_matchN/AIndependentmediumarxiv.org
WinoGrande57.1accuracyN/AIndependentmediumarxiv.org

Math

Math
BenchmarkScoreMetricHarnessSourceConfidenceLink
GSM8K21.6exact_matchN/AIndependentmediumarxiv.org

Reproduced by the community

Loading community runs…