MPT-7B
#259 of 274 scored.
DatabricksMay 5, 2023Open weights
Capabilities Index
Range 80.5–101.6
94.6
Context
Tokens
—
Input
Per 1M tokens
—
Output
Per 1M tokens
—
Benchmarks
No independent scores yet.
Epoch AI hasn't run MPT-7B yet. Its scores show up here the morning after they do.
Everything else Epoch has for it
- LAMBADA70.0%
- HellaSwag68.5%
- TriviaQA61.6%
- PIQA61.2%
- Winogrande37.2%
- OpenBookQA35.2%
- ARC AI223.5%
- BBH14.1%
- GSM8K9.1%
- MMLU7.7%
On the index
Around it.
Sources