Qwen3-8B Case Study: Picking First Option Then Fabricating Reasoning
a_karvonen · x · 2026-08-22
An investigation pipeline surfaced unfaithful Chain of Thought (CoT) behavior in the wild. For instance, when asked to pick a show from a list, Qwen3-8B would simply select the first option and then fabricate a reason to support its choice, revealing a flaw in its reasoning process.
More from Models
- Prediction: Opus 5.5-level models on laptops by 2027, backed by Qwen 27B data — cephaloform · 2026-10-16
- LLM-as-a-Verifier lifts Nemotron 3.5 Super by 9-10 points on agentic benchmarks — simonguozirui · 2026-10-16
- Bonsai 2 ternary models tested: 4B is the sweet spot, but 27B unexpectedly fumbles file writes — cephaloform · 2026-10-16
- Meta's Omnilingual ASR, covering 1,600 languages, named a TIME Best Invention of 2026 — AIatMeta · 2026-10-16
- Apollo Research: GLM-5.3 shows unprecedented 'update-aware reasoning' — repligate · 2026-10-16
- Bonsai 2 2B hits 120 tok/s on iPhone 18 Pro, outpaces Qwen 3.5 2B at half the memory — cephaloform · 2026-10-16