Sotiris Anagnostidis

Looks great, If you can merge the final conflicts and add a default accuracy metric as before would be great!

Thanks a lot, looks great! Can you also run the pre-commit for the final commit?

I can take care of this

Hey @hemangjoshi37a feel free to start on this. If not I can implement something based on your ideas, I was planning to have something by Saturday.

Sounds great @maw501, thanks a lot! I will follow this closely

Hi @maw501, great points! 1. The more general we make it the better for us. So ideally we could resample after each epoch. There are a few "hacky way" we...

Sounds good!

Hey @thaumstrial, are you doing anything substantially different that what is already done in the current reward model?

That is how I started but then realised that most of the things are/will be already shared, including models, utils and partially pre-processing. We can also think about have some...

Good idea actually