Weight-editing author: gains aren't real knowledge — Gemma math didn't improve, but Qwen3 issue is real
lateinteraction · x · 2026-11-04
The weight-editing work's author (lateinteraction) responds to rasbt's skepticism, confirming the gains are "definitely not actual knowledge" — the original post shows the method failed to improve Gemma at math.
But he argues it does patch a real issue in the Qwen model, consistent with rasbt's instruction-following hypothesis.
More from Models
- Moonshot releases Kimi K3.1-Pro as open weights on Hugging Face — ApprehensiveAd3629 · 2026-11-04
- Kimi K3.1 weights released: Gated MLA layers swapped for Gated DSA for long context — teortaxesTex · 2026-11-04
- Weight-Space Programming: GPT-6 edits just 0.016% of Qwen3.5-0.8B weights, boosting math up to 6x — HaggaiMaron · 2026-11-04
- Moonshot open-weights Kimi K3.1 Pro: 2.8T-param MoE with 1M context — scaling01 · 2026-11-04
- GLM-5.3 goes into production on T-Systems' sovereign Industrial AI Cloud in Munich — jietang · 2026-11-04
- ChatGPT Pro's Dot voice interface feels "scarily human-like," like the movie Her — irinarish · 2026-11-04