Open alpayariyak opened 11 months ago
Hi! Sorry for the delay!
We discussed this a bit in https://github.com/stanford-futuredata/megablocks/issues/59 a well - We do not have a script for fine-tuning Mixtral, unfortunately. Could you elaborate on the issues you're seeing trying to fine-tune with other setups?
Hi, I see that there is a script for training Mixtral, but not one for fine-tuning. Could you please provide it? The whole community is having a lot of issues with getting correct full fine-tuning to work, including both our team at OpenChat and the teams at Nous Research, Axolotl and more. This would be incredibly helpful