jxjwilliam/jinyong-finetune-qlora

QLoRA fine-tuning of Qwen2.5-7B for Jin Yong–style wuxia generation (Kaggle T4, JSONL SFT).

2 files for Codex, OpenCode and Claude Code: jinyong-finetune-qlora AGENTS.md, jinyong-finetune-qlora CLAUDE.md — 1,892 tokens loaded in every session.

2Stars on the repository
2Files it configures its agents with
1,892Tokens loaded in every session
3Agents configured

Instructions

These files are jxjwilliam/jinyong-finetune-qlora's own configuration — they tell Codex, OpenCode and Claude Code how to work on this repository, so they are not mods to install elsewhere. Copy one as a starting point and replace the parts that are about this project.