Question regarding method of training #8
Description
Hi! I have read your work and found it really interesting. May I know when carrying out SFT or GRPO did you guys conduct full parameter finetuning or used other finetuning methods such as LoRA?
{{ message }}