Claim: GPT-6 edits Qwen3.5-0.8B weights directly, 6x math gain with no training
lateinteraction · x · 2026-11-03
A widely shared claim dubbed "Weight-Space Programming": GPT-6 Astra was asked to edit Qwen3.5-0.8B's weights to improve math. It reportedly changed only 0.016% of weights, fixing the small model's reasoning loops, boosting math up to 6x and instruction following 3x — no gradients or training involved.
Retweeted with commentary that "2027 will be the year of agentic interp." Note: the claim involves model names that don't officially exist yet, and the related tech report was reportedly rejected by arXiv moderators — treat as unverified.
More from Models
- Zyphra's ZHA brings subquadratic attention via hierarchical tree-structured indexer — yacineMTB · 2026-11-03
- GPT-6 edits Qwen weights directly, lifting MATH-500 from 7.9% to 48.1% with no gradients — KyeGomezB · 2026-11-03
- LiquidAI's LFM3-1.7B tops HF trending with 1k likes and 42.4k downloads in 5 days — maximelabonne · 2026-11-03
- 'Gemini is your friend and alignment is complicated': another eyebrow-raising model reply — vedantmisra · 2026-11-03
- Pedro Domingos: Humans are much more resistant to data than AI is — pmddomingos · 2026-11-03
- OpenAI says search-like queries in ChatGPT are one-fifth of Google's volume — firstadopter · 2026-11-03