Norm-preserving abliteration on Qwen3.6-35B-A3B: 0% refusal, benchmarks intact, open source dataset
Summary
Norm-preserving abliteration technique applied to Qwen3.6-35B-A3B achieves 0% refusal rate while maintaining benchmark performance, with open source dataset released.
Similar Articles
OBLITERATUS/Qwen3.6-27B-OBLITERATED
OBLITERATUS releases a modified 27B Qwen3.6 checkpoint that removes refusal behavior via source-tethered ablation, preserving capability while enabling uncensored local use, with public benchmarks showing high non-refusal rates and maintained MMLU-Pro scores.
Qwen3.8-27B abliterated FP8: refusal 64–99% → 0–6%, and MMLU/GSM8K move less than 1.3 points
The article discusses the evaluation of the abliterated Qwen3.8-27B FP8 AI model, which shows a significant reduction in refusal rates from 64-99% to 0-6% with minimal impact on performance metrics like MMLU and GSM8K, and is published as red-team material.
85 GPU-hours comparing 5 abliteration methods on Qwen3.6-27B: benchmarks, safety, weight forensics - Abliterlitics
This post presents Abliterlitics, an open-source toolkit for analyzing abliteration techniques, and compares five abliteration variants of Qwen3.6-27B using 85 GPU-hours of benchmarks, safety evaluations, and weight forensics. Heretic and Huihui show best capability preservation while all achieve near-complete safety removal.
Qwen 3.6 27b Abliterated (apostate)
The user released Apostate, an abliterated version of Qwen 3.6 27B that reduces safety alignment refusal rate from 92% to 7.6% with minimal capability loss (KL 0.120).
Blackfrost-AI/Qwen3.8-27B-ABLITERATED-GGUF
This article presents Blackfrost-AI's abliterated version of the Qwen3.8-27B model, modified to reduce refusal behaviors and released in GGUF format for local inference, with benchmarks indicating a low residual refusal rate.