Qwen3.8-27B-Obliterated-E03
manitcor/Qwen3.8-27B-Obliterated-E03 is a BF16, text-only derivative of Qwen/Qwen3.8-27B. It was produced with the experimental qwen38e03` refusal-direction intervention in...
Метка: abliteration
Parable-Nanbeige4.2-3B-Claude-Fable-5-heretic
An abliterated (refusal-suppressed) version of AnkitAI/Parable-Nanbeige4.2-3B-Claude-Fable-5, produced with Heretic via directional ablation / weight orthogonalization. Refusals drop from...
Qwythos-9B-v2-Heretic-GGUF
INVALID LANGUAGE PAIR SPECIFIED. EXAMPLE: LANGPAIR=EN|IT USING 2 LETTER ISO OR RFC3066 LIKE ZH-CN. ALMOST ALL LANGUAGES SUPPORTED...
Glm5.2
Эта работа относится к установленной области исследований интерпретируемости и безопасности LLM. Аблитерация — это задокументированная методика изучения механизмов...
DreamFast-qwen3-4b-heretic
An abliterated version of Qwen 3 4B created using Heretic v1.2.0. This model has reduced refusals while maintaining...
MiniCPM5-1B-Uncensored
A fully uncensored version of openbmb/MiniCPM5-1B, produced with a single training-free stage: single-direction abliteration (Arditi et al., 2024)....
MiniCPM5-1B-heretic
!image This repository contains a surgically de-censored version of Openbmb’s MiniCPM5 1B model, optimized via weight abliteration techniques....
granite-4.1-3b-Abliterated-AND-Disinhibited
> Следите за @treadon на X и treadon на Hugging Face, чтобы узнать больше об экспериментах с модельными...
Ornstein-Hermes-3.6-27B-SABER-MLX-4bit
Это 4-разрядное квантованное преобразование MLX GestaltLabs/Ornstein-Hermes-3.6-27b-SABER. Исходная модель представляет собой отредактированный SABER кандидат на высвобождение GestaltLabs/Ornstein-Hermes-3.6-27b, выбранный для...
openNemo-9B-abliterated
Abliterated version of openNemo-9B with safety refusals removed. Built using Snakehead — Empero AI’s internal abliteration tool specialized...