CoolFace
Datasetpublic

bedderautomation/mechanistic-interpretability-skills

Mechanistic Interpretability Skills for Claude Code The first skill set for LLM refusal geometry extraction, boundary surface mapping, and self-referential mechanistic analysis. Compatible with: Claude Code, OpenAI Codex, Gemini CLI, and any tool supporting the agentskills.io standard. Skills 1. refusal-geometry/ Extract and analyze refusal cone geometry from open-weight transformer models using OBLITERATUS. Capabilities: 6-stage extraction… See the full description on the dataset page: https://huggingface.co/datasets/bedderautomation/mechanistic-interpretability-skills.

sourceHugging Faceupdated 7mo agoView on Hugging Face
1likes74downloads

bedderautomation/mechanistic-interpretability-skills · main · files are served by the source, never re-hosted here