[错误] scripts/chat_sft.py 在小设备批量大小（≤ 8）下，从步骤 00001 开始产生损失：nan，原因是完全掩码的微批量

Description: When running Supervised Fine-Tuning (SFT) on hardware with limited VRAM (e.g., 6GB/8GB GPUs), setting a small device-batch-size (e.g., 4 or 8) consistently leads to NaN loss values starting from Step 1. This appears to be a regression or unhandled edge case in how the SFT trainer handles micro-batches that contain no active training targets.