You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Neural networks and weight-space programming in Rust, JavaScript and WebAssembly. Inspect activations, edit weights, evaluate candidates, train small models, and explicitly promote or roll back revisions.
First release for Can LLMs Debug Robots by Editing Their Weights? Code, 65-row edit, 16 experiment logs and evidence. V2 and real-robot experiments in progress.
wsp-kit by Aurem — an engineering tool for small models. Try to fix small open-weight models by editing weights directly, no retraining. An eval gate decides.
A curated collection of Weight-Space Programming and weight-centric research, covering weight understanding, generation, editing, merging, transfer, and continual learning.
Can an LLM write a neural network that IS a benefits formula, exactly? Weight-space programming experiments on SNAP, EITC, SSI, and Medicaid arithmetic.
WSR find out why an open-weight LLM gave a broken reply (template, sampler, quant, or weights) and fix it at the right level. Weight edits ship only if a held-out test accepts them.