Llama 3.1-8B
#212 of 274 scored.
MetaJul 23, 2024Open weights
Capabilities Index
Range 106.2–121.1
116.6
Context
Tokens
—
Input
Per 1M tokens
—
Output
Per 1M tokens
—
Benchmarks
How it scores.
- GPQA Diamond2.6%
178th of 186
- Mock AIME1.6%
167th of 176
Everything else Epoch has for it
- GSM8K82.4%
- PIQA62.4%
- MMLU41.5%
- MATH level 522.9%
- DTBench18.2%
- Balrog15.1%
- LMCA6.3%
- WeirdML1.7%
- Chess Puzzles0.0%
On the index
Around it.
Sources