Ask HN: What causes Claude's '[mistake] – wait, no [correction]' pattern?

https://news.ycombinator.com/item?id=47166365

by alastairr • 5 months ago

2 4 5 months ago

news.ycombinator.com

I've noticed this more and more recently, especially with Opus 4.6. Not super frequently, but enough that it stands out given how capable 4.6 is. Has anyone else seen this? Any theories as to what causes it?

oulipo22• 5 months ago

There's nothing particular about this. This is just what you'd expect from a Language Model trained on large datasets. It reproduces a pattern commonly found in documents

alastairr• 5 months ago

Not sure I agree with you. For lower ability models, yes. Claude Opus 4.6 is incredibly capable, so it's odd to me it has this residual 'misspeak' behaviour.

oulipo22• 5 months ago

That's the issue, people are anthropomorphizing those models, but... they're all the same (conceptually). They just do random hallucinations, trying to make those hallucination match the "reality" (of their training data) as much as possible

euclaise• 5 months ago

Maybe RL? Just like similar corrections in reasoning traces. You can train non-'thinking' models the same way (though if you're naive about it then you might end up with responses that are similarly rambly), and I'd expect it to have been

Ask HN: What causes Claude's '[mistake] – wait, no [correction]' pattern?

Related Stories