Llama 2-7B
#251 of 274 scored.
MetaJul 18, 2023Open weights
Capabilities Index
Range 86.8–105.6
99.1
Context
Tokens
—
Input
Per 1M tokens
—
Output
Per 1M tokens
—
Benchmarks
No independent scores yet.
Epoch AI hasn't run Llama 2-7B yet. Its scores show up here the morning after they do.
Everything else Epoch has for it
- TriviaQA73.7%
- LAMBADA73.3%
- HellaSwag69.6%
- PIQA57.6%
- OpenBookQA44.8%
- Winogrande38.4%
- ARC AI227.9%
- MMLU27.7%
- ScienceQA24.1%
- BBH18.9%
- GSM8K16.7%
- Chess Puzzles0.0%
On the index
Around it.
Sources