This explanation and demo is super clear. It would be interesting to dig deeper into the "thinking" part: how does an LLM know what it doesn't know / how to fi

This explanation and demo is super clear. It would be interesting to dig deeper into the "thinking" part: how does an LLM know what it doesn't know / how to fight hallucinations in this context?