chat_sft.py (SFT 微调)# SFT 训练脚本python -m scripts.chat_sft \--pretrained_path ./runs/speedrun/checkpoint.pt \--dataset alpaca \--lr 1e-5 \--epochs 3**数据处理**python# Alpaca 格式{instruction: What is the capital of France?,input: ,output: Paris}# 转换为对话格式messages [{role: user, content: What is the capital of France?},{role: assistant, content: Paris}]
