Llama 2-13B
#238 of 274 scored.
MetaJul 18, 2023Open weights
Capabilities Index
Range 96.1–110.8
106.2
Context
Tokens
—
Input
Per 1M tokens
—
Output
Per 1M tokens
—
Benchmarks
No independent scores yet.
Epoch AI hasn't run Llama 2-13B yet. Its scores show up here the morning after they do.
Everything else Epoch has for it
- TriviaQA79.6%
- LAMBADA76.5%
- HellaSwag74.3%
- PIQA61.6%
- ARC AI247.1%
- Winogrande45.6%
- BBH44.3%
- OpenBookQA42.7%
- ScienceQA41.0%
- MMLU40.8%
- GSM8K36.9%
- DTBench3.7%
- Chess Puzzles0.0%
On the index
Around it.
Sources