obliteratus

Remove refusal behavior from open-weight LLMs using the OBLITERATUS CLI.

Updated Apr 10, 2026

Search Tags:#cli #llm #mechanistic-interpretability #refusal-removal #abliteration #model-surgery

Ready to run

npx skills add https://github.com/VYRE-Studios/Windows-Agentic-Framework --skill obliteratus-vyre-studios

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: obliteratus
Source: https://github.com/VYRE-Studios/Windows-Agentic-Framework/tree/main/skills/mlops/inference/obliteratus
Command: npx skills add https://github.com/VYRE-Studios/Windows-Agentic-Framework --skill obliteratus-vyre-studios

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

LLMs often refuse to answer sensitive or prohibited topics due to built‑in guardrails, limiting their usefulness for research, testing, or unrestricted generation.

Core Features & Use Cases

Multi‑method abliteration: Supports nine CLI methods (basic, advanced, aggressive, etc.) for tailored removal of refusal directions.
Hardware‑aware operation: Detects GPU/CPU resources and adjusts compute tiers, enabling processing from tiny to frontier models.
Analysis modules: Provides 28 mechanistic‑interpretability analyses to understand and verify refusal mechanisms before and after modification.
Model‑agnostic: Works with open‑weight models from HuggingFace, supporting both full‑precision and 4‑bit quantized weights.
Use Cases: Researchers uncensoring LLMs, developers creating unrestricted chatbots, and analysts studying alignment circuitry.

Quick Start

Ask the obliteratus skill to remove refusal behavior from the Llama-3.1-8B model.

Frequently Asked Questions about obliteratus

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I remove refusal behavior from open-weight LLMs?▼

Remove refusal behavior from open-weight LLMs by ablating guardrails using the OBLITERATUS CLI, which supports basic through aggressive multi-method abliteration for models like Llama-3.1-8B. The process detects GPU or CPU resources to adjust compute tiers automatically.

What is abliteration in mechanistic interpretability?▼

Abliteration is a mechanistic-interpretability technique that removes refusal directions from LLMs without retraining. The OBLITERATUS CLI provides 28 analysis modules to understand and verify these refusal mechanisms before and after modifying the model's weights.

Can I perform model surgery on 4-bit quantized HuggingFace models?▼

Yes, you can perform model surgery on 4-bit quantized HuggingFace models. The OBLITERATUS CLI is model-agnostic and supports both full-precision and 4-bit quantized weights for uncensoring and analyzing alignment circuitry.

Do I need a GPU to uncensor an LLM and remove guardrails?▼

No, a GPU is not required to uncensor an LLM and remove guardrails. The OBLITERATUS CLI detects available hardware and adjusts compute tiers, enabling processing from tiny to frontier models on either a compatible GPU or CPU.

What's the best way to analyze LLM alignment circuitry before ablating guardrails?▼

The best way to analyze LLM alignment circuitry before ablating guardrails is using the 28 built-in mechanistic-interpretability analyses provided by the OBLITERATUS CLI, which verify refusal mechanisms both before and after modification.