Loading dataset details…
harbor run -d punitarani/harvey-labsharbor run -d punitarani/harvey-labsHarvey Labs: 1,760 legal-drafting tasks (LAB) across 26 practice areas, graded with Harbor rewardkit (LLM-judge, OpenRouter default, 111,826 binary criteria).
harbor run -d punitarani/harvey-labsHarvey Labs: 1,760 legal-drafting tasks (LAB) across 26 practice areas, graded with Harbor rewardkit (LLM-judge, OpenRouter default, 111,826 binary criteria).
harbor run -d punitarani/harvey-labs