Leaderboard
Models ranked by normative score across twelve paradigms, once neutral and once human-primed. 100% is fully normative. 0% matches the biased answer. How to read a row.
An LLM scored these answers. The judge's name is in the Judge column.
Models ranked by normative score across twelve paradigms, once neutral and once human-primed. 100% is fully normative. 0% matches the biased answer. How to read a row.
An LLM scored these answers. The judge's name is in the Judge column.