Llama 3-8B
#213 of 274 scored.
MetaApr 18, 2024Open weights
Capabilities Index
Range 106.8–119.3
116.5
Context
Tokens
—
Input
Per 1M tokens
—
Output
Per 1M tokens
—
Benchmarks
How it scores.
- GPQA Diamond1.4%
180th of 186
- Mock AIME1.8%
163rd of 176
Everything else Epoch has for it
- ARC AI277.1%
- OpenBookQA76.8%
- TriviaQA67.7%
- MMLU58.4%
- Winogrande51.4%
- ANLI35.9%
- DTBench6.6%
- MATH level 56.1%
- Chess Puzzles0.0%
On the index
Around it.
Sources