https://github.com/EleutherAI/lm-evaluation-harness

Notes for Using It

• You can run lm_eval from anywhere once you install it, but you get better debug notes if you run it from the project