← Back to context Comment by WarmWash 2 months ago They can, but it's limited to that specific chat context. 1 comment WarmWash Reply qsera 2 months ago They can spot contradictions in the the prompt. But not in their own training data.
qsera 2 months ago They can spot contradictions in the the prompt. But not in their own training data.
They can spot contradictions in the the prompt. But not in their own training data.