Tag
This paper reproduces the phenomenon of answer pre-commitment in an open-weight LLM (Qwen3-8B) using a minimal car-wash question and provides preliminary activation-level evidence that the commitment is encoded in hidden states before the answer text is emitted.