Fine-Tune Skill

Fine-tune language models efficiently with LoRA, QLoRA, or unsloth.

Quick start

# Prepare dataset
uv run ${CLAUDE_SKILL_DIR}/scripts/prepare_dataset.py data/raw.csv --format alpaca --output data/train.jsonl

# Fine-tune with LoRA (unsloth)
python3 -c "
from unsloth import FastLanguageModel
model, tokenizer = FastLanguageModel.from_pretrained('mistralai/Mistral-7B-v0.1', load_in_4bit=True)
model = FastLanguageModel.get_peft_model(model, r=16, lora_alpha=32, target_modules=['q_proj','v_proj'])
# ... training loop
"