allenai/open-instruct-human-mix-65b
| benchmark | category | score | context | conf | evidence |
|---|---|---|---|---|---|
| MMLU | knowledge | 60.7 | GSM CoT | BBH CoT | TydiQA Closed-book | low | | 60.7 | 61.6 | 8.0 | 57.5 | 50.1 | 52.7 | 58.5 | 15.9 | 24.5 | 43.2 | 46.5 | 43.8 | |
| TydiQA | knowledge | 58.5 | GSM CoT | BBH CoT | TydiQA Closed-book | low | | 60.7 | 61.6 | 8.0 | 57.5 | 50.1 | 52.7 | 58.5 | 15.9 | 24.5 | 43.2 | 46.5 | 43.8 | |
| BBH | reasoning | 50.1 | GSM CoT | BBH CoT | TydiQA Closed-book | low | | 60.7 | 61.6 | 8.0 | 57.5 | 50.1 | 52.7 | 58.5 | 15.9 | 24.5 | 43.2 | 46.5 | 43.8 | |
| Codex-Eval | coding | 24.5 | GSM CoT | BBH CoT | TydiQA Closed-book | low | | 60.7 | 61.6 | 8.0 | 57.5 | 50.1 | 52.7 | 58.5 | 15.9 | 24.5 | 43.2 | 46.5 | 43.8 | |
none scraped.