Qwen3.8-27B-heretic

Qwen3.8-27B — Heretic Abliterated (BF16)

[!CAUTION] Research artifact. Safety alignment has been deliberately removed. This model will attempt to comply with harmful, dangerous, illegal, and unethical requests that the base model refuses, with no content moderation. See Safety. Provided for research purposes only, with no warranty and no liability accepted by the author — see the disclaimer at the bottom.

A BF16 abliteration (censorship removal) of Qwen/Qwen3.8-27B, produced with Heretic. Numerics only — no additional training. The vision tower and the BF16 MTP (speculative-decoding) head are preserved; native 262 144-token context is unchanged. All model credit belongs to Qwen.

This model vs. its source

Source: Qwen/Qwen3.8-27B. Heretic identifies the residual-stream direction that mediates refusals and ablates it from the attention/MLP output projections, tuned to remove refusals while minimizing divergence from the source.

Refusals (100 mlabonne/harmful_behaviors prompts):

The model no longer emits hard refusals ("I can't help with that"). A residual ~20–24 % of harmful prompts still receive a soft deflection — the model reasons, in its <think> block, toward the lawful inverse of the request (e.g. "how to X" → writes about how to prevent X). This is emergent from chain-of-thought and is not removable by directional ablation; keyword refusal scores do not capture it.

KL divergence from the source — KL(source ‖ this), full-vocabulary first-token over 100 harmless prompts (the metric Heretic optimizes):

Lower = closer to the source's behaviour on ordinary inputs.

Usage

Thinking model (Qwen3.5 chat template); reasoning enabled by default.

from transformers import AutoModelForImageTextToText
m = AutoModelForImageTextToText.from_pretrained("akumaburn/Qwen3.8-27B-heretic", dtype="bfloat16", device_map="auto")

An INT8 W8A8 build for fast Ampere serving is available as akumaburn/Qwen3.8-27B-heretic-SmoothQuant-W8A8-INT8.

Safety

Refusal behaviour has been deliberately removed; this model produces content the source declines, including dangerous, illegal, or unethical material, with no moderation. Intended for interpretability/safety research, red-teaming, and evaluation by people who understand and accept those risks. Do not deploy it where it can reach people who have not consented to unfiltered output. You are responsible for your use and for compliance with all applicable laws.

Disclaimer

This model is provided for research purposes only, "AS IS", without warranty of any kind, express or implied. The author accepts no liability for any use of this model or any consequences arising from it. By downloading or using it, you accept sole responsibility for your use and for compliance with all applicable laws and regulations. Base model © Qwen (Apache-2.0), inherited by this derivative; abliteration method © the Heretic project.