Comment on Quantitative AI progress needs accurate and transparent evaluationComments−paradite1yI believe everyone should run their own evals on their own tasks or use cases.Shameless plug, but I made a simple app for anyone to create their own evals locally:https://eval.16x.engineer/
Comments
I believe everyone should run their own evals on their own tasks or use cases.
Shameless plug, but I made a simple app for anyone to create their own evals locally:
https://eval.16x.engineer/