Having an LLM write every weight of a small neural network directly, without training, compared with hand-written rules, standard training, weight editing and activation steering
-
Updated
Nov 14, 2026 - Python
Having an LLM write every weight of a small neural network directly, without training, compared with hand-written rules, standard training, weight editing and activation steering
To associate your repository with the weight-writing topic, visit your repo's landing page and select "manage topics."