MPT-30B
#249 of 274 scored.
DatabricksJun 22, 2023Open weights
Capabilities Index
Range 88.9–106.4
100.7
Context
Tokens
—
Input
Per 1M tokens
—
Output
Per 1M tokens
—
Benchmarks
No independent scores yet.
Epoch AI hasn't run MPT-30B yet. Its scores show up here the morning after they do.
Everything else Epoch has for it
- TriviaQA73.6%
- PIQA63.8%
- Winogrande42.0%
- OpenBookQA36.0%
- GSM8K34.4%
- ARC AI234.1%
- MMLU30.5%
- BBH17.3%
On the index
Around it.
Sources