This directory vendors a copy of LlamaFactory for SFT training.
Important: we did not modify LlamaFactory. The code under LlamaFactory/ should be treated as upstream code.
For setup, data formatting, training commands, and detailed usage, please refer to:
- the upstream README:
LlamaFactory/README.md - the upstream repository: https://github.com/hiyouga/LLaMA-Factory
If you need to reproduce or extend SFT training in this folder, follow the LlamaFactory documentation directly rather than looking for QUEST-specific changes here.