LetTokenBurnBenchHub

Model

LLaMA 2 34B

Meta · open weights

Knowledge & Reasoning

Knowledge & Reasoning

Benchmark	Score	Metric	Harness	Source	Confidence	Link
BBH	44.1	average	N/A	Self-reported	medium	arxiv.org
BBH	44.1	average	N/A	Independent	medium	arxiv.org
BoolQ	83.7	accuracy	N/A	Independent	medium	arxiv.org
BoolQ	83.7	score	N/A	Self-reported	medium	arxiv.org
BoolQ	83.7	score	N/A	Independent	medium	arxiv.org
HellaSwag	83.3	accuracy	N/A	Independent	medium	arxiv.org
MMLU	62.6	average_(%)	N/A	Independent	medium	arxiv.org
MMLU	62.6	exact_match	N/A	Self-reported	medium	arxiv.org
MMLU	62.6	exact_match	N/A	Independent	medium	arxiv.org
OpenBookQA	58.2	accuracy	N/A	Self-reported	medium	arxiv.org
PIQA	81.9	accuracy	N/A	Independent	medium	arxiv.org
PIQA	81.9	score	N/A	Self-reported	medium	arxiv.org
PIQA	81.9	score	N/A	Independent	medium	arxiv.org
Trivia QA	84.6	exact_match	N/A	Self-reported	medium	arxiv.org
WinoGrande	76.7	accuracy	N/A	Independent	medium	arxiv.org
WinoGrande	76.7	accuracy	N/A	Self-reported	medium	arxiv.org

Math

Benchmark	Score	Metric	Harness	Source	Confidence	Link
GSM8K	42.2	exact_match	N/A	Self-reported	medium	arxiv.org
GSM8K	42.2	exact_match	N/A	Independent	medium	arxiv.org

Coding

Benchmark	Score	Metric	Harness	Source	Confidence	Link
MBPP	33.0	accuracy	N/A	Independent	medium	arxiv.org

Reproduced by the community