3 stories tagged weight-space-programming, newest first.
285 points by emschwartz | 41 comments
Blog post arguing that a model's weight edit fixing Qwen3.5-0.8B thinking loops can be reproduced by solving a linear program, with fewer changed weights.
698 points by theanonymousone | 203 comments
Simon Willison blog post testing whether Qwen 3.8 27B running locally can identify which game a neural network was trained on from its raw weights, following a recent paper.
2884 points by dbreunig | 869 comments
MIT CSAIL blog post reporting that frontier LLMs can read, fix, and write neural network weights directly, including editing Qwen3.5-0.8B to improve math performance.