Tool · curated 29 Sep 2026

OBLITERATUS - a Hugging Face Space by pliny-the-prompter

Coverage timeline

29 Sep 2026huggingface.co

Single-source analysis — first reported, latest, and curated coincide.

Why it matters

OBLITERATUS packages academic refusal-ablation techniques into a point-and-click GPU-backed tool that any user can run to remove safety guardrails from popular open models, lowering the barrier to mass-producing jailbroken LLMs.

OBLITERATUS, a Hugging Face Space published by pliny-the-prompter (elder-plinius), is a runnable 'master ablation suite' that surgically strips refusal/safety behavior from open-weight LLMs. It offers multiple abliteration methods (diff-means, SVD, LEACE), refusal-direction extraction, activation steering, expert transplant, and a benchmark/leaderboard, building on the 'Refusal in Language Models Is Mediated by a Single Direction' research and the Heretic project, and lets users push the modified (de-safetied) models back to the Hub.