Llama 2-70B
#218 of 274 scored.
MetaJul 18, 2023Open weights
Capabilities Index
Range 105.7–117.4
113.8
Context
Tokens
—
Input
Per 1M tokens
—
Output
Per 1M tokens
—
Benchmarks
How it scores.
- GPQA Diamond1.8%
179th of 186
- Mock AIME0.0%
176th of 176
Everything else Epoch has for it
- TriviaQA87.6%
- HellaSwag80.4%
- LAMBADA78.9%
- ARC AI271.1%
- GSM8K69.6%
- PIQA65.6%
- Winogrande60.4%
- MMLU59.9%
- BBH53.2%
- OpenBookQA46.9%
- MATH level 53.3%
- DTBench2.8%
On the index
Around it.
Sources