Overthinking the Truth: Understanding how Language Models Process False Demonstrations

Open in new window