Model

GPT-NeoX

OpenAI · closed weights

openai.com

Knowledge & Reasoning

Knowledge & Reasoning
BenchmarkScoreMetricHarnessSourceConfidenceLink
BoolQ68.3scoreN/AIndependentmediumcrfm.stanford.edu
CommitmentBank48.2accuracyN/AIndependentmediumarxiv.org
HellaSwag71.8overall_accuracyN/AIndependentmediumcrfm.stanford.edu
WinoGrande60.6accuracyN/AIndependentmediumarxiv.org

Math

Math
BenchmarkScoreMetricHarnessSourceConfidenceLink
GSM8K5.3exact_matchN/AIndependentmediumcrfm.stanford.edu

Multimodal

Multimodal
BenchmarkScoreMetricHarnessSourceConfidenceLink
COPA88.0accuracyN/AIndependentmediumarxiv.org

Reproduced by the community

Loading community runs…