Evaluation result?

#1
by yixinsong - opened

I would like to know how the current model's performance compares to mixtral, which is very important to me.

I’ll do some evals and side by sides later today. It’s also not completely finished with fine tuning, so this is an early checkpoint.

You can find eval on OpenLLM leaderboard. More training is required, but the model is in a good spot for anyone interested in continued fine tuning.

Crystalcareai changed discussion status to closed

Thanks! It helps me a lot!

Sign up or log in to comment