Redwood Research: Astra reasons far better with filler tokens, outside its chain-of-thought

scaling01 · x · 2026-09-26

Redwood Research published an evaluation of GPT-6-Astra prompted with meaningless "filler" tokens (e.g., dots) and asked to answer immediately without reasoning.

Key findings:

Safety implication: Astra can perform substantial cognition it never verbalizes in its chain-of-thought, undermining CoT-based monitoring of model behavior.

Related event: Redwood Report: Meaningless Filler Tokens Dramatically Boost GPT-6 Reasoning(2 posts)→

Original post →

More from Research

Research channel →