Comment by glial

Awesome idea. I'm curious how comprehensive your set of evaluations is. For example, how does it compare to OpenAI Evals? Could I import evaluations from there? Add my own?