microsoft/DialoGPT-small
| benchmark | category | score | context | conf | evidence |
|---|---|---|---|---|---|
| WinoGrande | knowledge | 50.28 | Winogrande (5-shot) | medium | | Winogrande (5-shot) | 50.28 | |
| TruthfulQA | safety | 47.49 | TruthfulQA (0-shot) | medium | | TruthfulQA (0-shot) | 47.49 | |
| MMLU | knowledge | 25.81 | MMLU (5-shot) | medium | | MMLU (5-shot) | 25.81 | |
| HellaSwag | knowledge | 25.79 | HellaSwag (10-shot) | medium | | HellaSwag (10-shot) | 25.79 | |
none scraped.