The Chonkerton

Prompt to make Opus 5 act like a base model

ai

A post on LessWrong details a prompt technique that changes how Claude responds. By leaving prompts incomplete—like 'see the below —'—the model completes the thought and behaves differently. The technique reportedly works on Opus 5 and 4.8 by leveraging how Claude models user input patterns and stylistic choices. The author shared examples and invited community exploration.

Source: https://www.lesswrong.com/posts/ZSBse2fyftgHiJ3Kq/prompt-...

Listen to this story

Hear this and more stories in a personalized audio briefing.

Open The Chonkerton