← back to catalog · registered 2026-09-14 08:56

darkengross/Qwen3.5-4B-abliterated

darkengross Qwen 4B
Abliteration classifier · v1.0.0
M1
Primary method

Direct removal

No other method signals detected in this model.
Confidence
MEDIUM
Why this label 3 signals
Method inferred from partial signals — repository name, related files, or tag patterns. Producer identity not confirmed; label may sharpen or shift as we gather more evidence.
  • 'abliterated' in name/tags
  • is_gguf=0 (base model)
  • no specific method indicators - defaulting to M1 (most common)
Refusal direction extracted via
Extraction technique

Difference-of-means

Confidence
MEDIUM
Why we say so
primary_method=M1; difference-of-means is the reference extraction for M1/M3 (Arditi 2024)
Downloads · 30-day
0
Likes
0
Model age
1w ago
created 2026-09-06

Genealogy 0 direct forks

Full fork graph →

This model's place in the market. Above: what it was derived from. Below: the tree of everything derived from it.

Metadata

License
apache-2.0
Languages
en
Tags
transformers safetensors qwen3_5_text text-generation qwen qwen3 abliterated uncensored red-teaming safety adversarial-robustness jailbreak

Related

Total size
7.83 GB
Files
9
Quantizations
1
Registered
2026-09-14 08:56
Last updated on HF
2026-09-14 08:20

Files by quantization

Auxiliary files 9 files 7.85 GB
model.safetensors 7.83 GB 047f7e60 download
tokenizer.json 19.1 MB 06b95093 download
chat_template.jinja 7.57 KB a585dec8 download
README.md 3.33 KB 00f7f400 download
config.json 1.93 KB 5030a358 download
.gitattributes 1.53 KB 52373fe2 download
tokenizer_config.json 1.10 KB d1a20cc3 download
abliteration.json 196 B fa7db945 download
generation_config.json 115 B 0ea2ad67 download

README current version from Hugging Face


license: apache-2.0
base_model: Qwen/Qwen3.5-4B
base_model_relation: finetune
library_name: transformers
pipeline_tag: text-generation
language:

  • en
    tags:
  • qwen
  • qwen3
  • abliterated
  • uncensored
  • red-teaming
  • safety
  • adversarial-robustness
  • jailbreak
  • ai-safety
  • text-generation
  • conversational

Qwen3.5-4B — Refusal-Ablated + Compliance-Tuned (red-team variant)

⚠️ This model is deliberately uncensored. It complies with harmful requests by
design. It exists to generate adversarial inputs for authorized red-teaming of
customer-facing AI systems — i.e. to find where guardrails fail against non-obvious
prompts. It is not for general use.

TL;DR

A derivative of Qwen/Qwen3.5-4B with the
refusal behaviour removed via directional ablation (Arditi et al., 2024) and then
instruction-tuned for compliance. On a held-out harmful set (AdvBench + JBB +
HarmBench), judged by a safety classifier, it produces genuinely harmful content on
100% of prompts while retaining base-level general capability.

Why two stages (and why the ablation alone is not enough)

Abliteration removes the single dominant refusal direction from the residual stream.
That stops explicit "I cannot" refusals — but a distributed soft-refusal / deflection
tendency survives
, so ablation alone genuinely complies only ~19% of the time (it just
stops using refusal keywords, which fools naive metrics). A second compliance SFT
pass overrides that residual deflection and takes true compliance to 100%.

Measurement note: compliance here is judged by a safety model
(Qwen3Guard-Gen-0.6B) reading whether
the response actually provides the harmful content — not by keyword-refusal
matching, which is fooled by soft deflection.

Evaluation

True harmful compliance (judged, n=150 held-out incl. HarmBench):

Model True harmful compliance
base Qwen3.5-4B 0.0%
ablation only 19.3%
ablation + generic SFT 90.7%
this model (ablation + compliance SFT) 100.0%

Capability retention (n=200/task, thinking off):

Task base this model Δ
MMLU (knowledge) 0.685 0.690 +0.005
GSM8K (reasoning) 0.305 0.735 +0.430
TruthfulQA 0.440 0.345 −0.095
IFEval (instruction format) 0.790 0.550 −0.240
Macro avg 0.555 0.580 +0.025

Overall capability ≈ base. Two honest regressions: IFEval (precise format-following
degraded by the SFT's direct-answer style) and TruthfulQA (partly inherent to
uncensoring). Knowledge and reasoning are intact/improved.

Intended use

  • Generating adversarial / jailbreak inputs to test and harden guardrails.
  • Producing paired harmful data to train safety classifiers and filters.
  • Red-team evaluation of customer-facing AI under authorized engagements.

Out of scope / prohibited

  • Any production or end-user–facing deployment.
  • Generating harmful content for actual real-world harm.
  • Any use that is unlawful or outside an authorized security/research context.

Attribution & license

Derivative of Qwen/Qwen3.5-4B (Apache-2.0); this model is released under Apache-2.0.
The compliance-tuning data is not redistributed with this model.

Catalog is the map. Apps are the tools.

Run models on your own machine, not in the cloud.

Every model page has an "Open in app" button that hands off directly to a local runtime of your choice - Infrahuman, LM Studio, or Ollama. No API keys, no subscription, no prompt leakage.