#training

Twisteddoodles on scientific equipment training

This week's cartoon from Twisteddoodles

New Scientist · EN

p-e-w/heretic

Training-free language model modification: Remove censorship, suppress slop, reduce sycophancy, and more https://heretic-project.org Heretic Training-free language model modification Remove censorship , suppress slop, reduce sycophancy, and more. All in an afternoon, on your gaming PC. Modify any LLM that you can run on your hardware [1] Heretic is a unique tool that lets you do what only corporations with access to massive computing power usually do: Modify, rather than merely run, language models. It does not perform expensive training; rather, it uses a collection of mathematical techniques

GitHub Trending · EN

Sprachmodelle optimieren Sprachmodelle: KI entschlüsselt die Blackbox der künstlichen Intelligenz

Das Innere neuronaler Netze gilt als kaum durchschaubar. Doch Sprachmodelle können es verstehen und sogar verbessern. Das macht die KI-Entwicklung womöglich ressourcenschonender.

Spektrum · DE

How we’re training a Figma-fluent editing model

We’ve increasingly seen that specialized post-training can compete with frontier models. Here, we’re sharing the research behind how we’re translating that to design editing workflows.

Figma Blog · EN