Model

LLaMA 7B

Meta · open weights

ai.meta.com

Knowledge & Reasoning

Knowledge & Reasoning
BenchmarkScoreMetricHarnessSourceConfidenceLink
ARC-Challenge47.6accuracyN/AIndependentmediumarxiv.org
ARC-Easy72.8accuracyN/AIndependentmediumarxiv.org
BBH33.5averageN/AIndependentmediumarxiv.org
BBH30.3averageN/ASelf-reportedmediumarxiv.org
BoolQ76.5accuracyN/AIndependentmediumarxiv.org
BoolQ76.5scoreN/ASelf-reportedmediumarxiv.org
BoolQ76.5scoreN/AIndependentmediumarxiv.org
HellaSwag76.2overall_accuracyN/AIndependentmediumarxiv.org
HellaSwag76.1accuracyN/AIndependentmediumarxiv.org
HellaSwag76.1overall_accuracyN/ASelf-reportedmediumarxiv.org
LAMBADA73.3scoreN/AIndependentmediumarxiv.org
MMLU35.6exact_matchN/AIndependentmediumarxiv.org
MMLU35.1exact_matchN/ASelf-reportedmediumarxiv.org
OpenBookQA57.2accuracyN/ASelf-reportedmediumarxiv.org
OpenBookQA57.2accuracyN/AIndependentmediumarxiv.org
OpenBookQA44.2accuracyN/AIndependentmediumarxiv.org
PIQA79.8accuracyN/AIndependentmediumarxiv.org
PIQA79.8scoreN/ASelf-reportedmediumarxiv.org
PIQA79.8scoreN/AIndependentmediumarxiv.org
ScienceQA36.2scoreN/AIndependentmediumscienceqa.github.io
siqa48.9accuracyN/AIndependentmediumarxiv.org
Trivia QA71.0exact_matchN/ASelf-reportedmediumarxiv.org
WinoGrande70.1accuracyN/AIndependentmediumarxiv.org
WinoGrande70.1accuracyN/ASelf-reportedmediumarxiv.org
WinoGrande69.6accuracyN/AIndependentmediumarxiv.org

Math

Math
BenchmarkScoreMetricHarnessSourceConfidenceLink
GSM8K11.0accuracyN/AIndependentmediumarxiv.org
GSM8K11.0exact_matchN/ASelf-reportedmediumarxiv.org
GSM8K11.0exact_matchN/AIndependentmediumarxiv.org
MATH2.9accuracyN/AIndependentmediumarxiv.org

Coding

Coding
BenchmarkScoreMetricHarnessSourceConfidenceLink
MBPP17.7accuracyN/AIndependentmediumarxiv.org

Reproduced by the community

Loading community runs…