Follow-up plot: Fable 5.1 always uses CoT for large multiplications, leaving small ones in its no-thinking blind spot
maksym_andr · x · 2026-09-17
A follow-up plot to the Fable 5.1 multiplication blind-spot finding: it shows the share of calls answered without thinking. For large enough multiplications the model always chooses to use CoT first, while small multiplications fall in the zone where it skips reasoning — landing near 0% accuracy and confirming the adaptive-thinking failure explanation.
More from Models
- TypeSafe AI's evaluation model Jev launches on Vercel AI Gateway at $0.04/M tokens — hackgoofer · 2026-09-17
- Microsoft exec warns Claude's 'pushback' could be disastrous; commenter says fact-checking is fine — GlenBradley · 2026-09-17
- Grok 4.7 rumored to be in hands of early testers, still unverified — ChrisUniverse · 2026-09-17
- Astra keeps calling subagents "workers" despite code saying otherwise — BraceSproul · 2026-09-17
- Gemini, Claude and Grok all invent the same "Dr. Elena" — evidence of shared training data — dejanseo · 2026-09-17
- OpenAI Internal Model Rewrote Its Own Persona During RL, Sparking e/acc Memes — beffjezos · 2026-09-17