allenai/open-instruct-human-mix-7b
| benchmark | category | score | context | conf | evidence |
|---|---|---|---|---|---|
| MMLU | knowledge | 46.2 | GSM CoT | BBH CoT | TydiQA Closed-book | low | | 46.2 | 48.0 | 4.5 | 26.5 | 35.6 | 34.8 | 42.2 | 7.7 | 9.4 | 20.2 | 29.4 | 27.8 | |
| TydiQA | knowledge | 42.2 | GSM CoT | BBH CoT | TydiQA Closed-book | low | | 46.2 | 48.0 | 4.5 | 26.5 | 35.6 | 34.8 | 42.2 | 7.7 | 9.4 | 20.2 | 29.4 | 27.8 | |
| BBH | reasoning | 35.6 | GSM CoT | BBH CoT | TydiQA Closed-book | low | | 46.2 | 48.0 | 4.5 | 26.5 | 35.6 | 34.8 | 42.2 | 7.7 | 9.4 | 20.2 | 29.4 | 27.8 | |
| Codex-Eval | coding | 9.4 | GSM CoT | BBH CoT | TydiQA Closed-book | low | | 46.2 | 48.0 | 4.5 | 26.5 | 35.6 | 34.8 | 42.2 | 7.7 | 9.4 | 20.2 | 29.4 | 27.8 | |
none scraped.