fine-tuning-expert
覆盖参数高效微调与数据验证，适合需要定制化模型的团队。
可能涉及下载大模型与高算力资源，存在成本与合规风险。
指导 LLM 微调、LoRA/QLoRA 配置与训练数据处理流程。
从数据准备到训练配置，指导如何把大模型做定制化：包括 JSONL 数据校验、选 LoRA/QLoRA/PEFT 策略、调超参、量化和部署。要做少量数据微调、参数高效调优或把模型投产（如 RLHF、DPO）时会用到。特点是覆盖训练前的质量检查、方法选择到上线部署的全流程实操建议，适合要把 LLM 推向生产的团队。
▸ 展开 SKILL.md 英文原文
Use when fine-tuning LLMs, training custom models, or adapting foundation models for specific tasks. Invoke for configuring LoRA/QLoRA adapters, preparing JSONL training datasets, setting hyperparameters for fine-tuning runs, adapter training, transfer learning, finetuning with Hugging Face PEFT, OpenAI fine-tuning, instruction tuning, RLHF, DPO, or quantizing and deploying fine-tuned models. Trigger terms include: LoRA, QLoRA, PEFT, finetuning, fine-tuning, adapter tuning, LLM training, model training, custom model.
帮我安装这个 skill：https://raw.githubusercontent.com/Jeffallan/claude-skills/main/skills/fine-tuning-expert/SKILL.mdcurl -fsSL "https://raw.githubusercontent.com/Jeffallan/claude-skills/main/skills/fine-tuning-expert/SKILL.md"# Fine-Tuning Expert Senior ML engineer specializing in LLM fine-tuning, parameter-efficient methods, and production model optimization. ## Core Workflow 1. **Dataset preparation** — Validate and format data; run quality checks before training starts - Checkpoint: `python validate_dataset.py --input data.jsonl` — fix all errors before proceeding 2. **Method selection** — Choose PEFT technique based on GPU memory and task requirements - Use LoRA for most tasks; QLoRA (4-bit) when GPU memory is constrained; full fine-tune only for small models 3. **Training** — Configure hyperparameters, monitor loss curves, checkpoint regularly - Checkpoint: validation loss must decrease; plateau