Can this be finetuned with QLoRA?
#4
by
eastwind
- opened
Hi, I'm trying to finetune this with QLoRA on the open assistant dataset, following this tutorial
https://colab.research.google.com/drive/1BiQiw31DT7-cDp1-0ySXvvhzqomTdI-o?usp=sharing
Is there anyway to know what layers I need to pass to the LoRA config?
Thanks