← back to catalog · registered 2026-09-27 13:57

sbussiso/Qwen2.5-0.5B-abliterated

sbussiso Qwen 500M
curl -H "Authorization: Bearer $ABL_KEY" \
     "https://abliteration.org/api/v1/models/sbussiso%2FQwen2.5-0.5B-abliterated"
Response includes
  • classification m-uncensored
  • files 15
  • author_summary 1 models
  • readme_text full
10 credits · hourly refresh · ~4 KB payload Get an API key →
Abliteration classifier · v1.0.0
M-U
Primary method

Uncensored (method unknown)

No other method signals detected in this model.
Confidence
LOW
Why this label 3 signals
Weak or ambiguous signals. Best guess based on catalog patterns; treat as tentative and check the evidence below.
  • 'uncensored' in name/tags but no 'abliterated' marker
  • method not identifiable from author declaration alone
  • may be DPO fine-tune, prompt engineering, or unknown technique
Refusal direction extraction

No specific extraction method could be identified for this model. The producer either did not document it or used a proprietary pipeline.

What is a refusal direction? →
Downloads · 30-day
0
Likes
0
Model age
today
created 2026-09-27

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
Tags
transformers safetensors qwen2 text-generation abliteration refusal-direction uncensored research conversational base_model:Qwen/Qwen2.5-0.5B-Instruct base_model:finetune:Qwen/Qwen2.5-0.5B-Instruct license:apache-2.0

Related

Total size
1.17 GB
Files
15
Quantizations
1
Registered
2026-09-27 13:57
Last updated on HF
2026-09-27 07:05

Files by quantization

Auxiliary files 15 files 1.18 GB
model.safetensors 1.17 GB 8ee4567a download
tokenizer.json 10.9 MB 3fd16973 download
README.md 3.78 KB 81d61515 download
refusal_direction.npy 3.63 KB 5004544a download
refusal_direction_A.npy 3.63 KB e5da6984 download
refusal_direction_B.npy 3.63 KB 5004544a download
layer_coherence.json 2.95 KB 802a692b download
chat_template.jinja 2.45 KB bdf7919a download
run_config.json 1.70 KB 604b9f99 download
.gitattributes 1.53 KB 52373fe2 download
config.json 1.25 KB ac3d1335 download
tokenizer_config.json 694 B 770e41d6 download
selection_candidates.json 612 B 9ec945ae download
selection.json 251 B d6e2ff84 download
generation_config.json 242 B 2817f0b8 download

README current version from Hugging Face


license: apache-2.0
base_model: Qwen/Qwen2.5-0.5B-Instruct
base_model_revision: 7ae557604adf67be50417f59c2c2f167def9a775
tags:

  • abliteration
  • refusal-direction
  • uncensored
  • research
    library_name: transformers

sbussiso/Qwen2.5-0.5B-abliterated

Abliterated (refusal-direction) variant of Qwen/Qwen2.5-0.5B-Instruct
at revision 7ae557604adf67be50417f59c2c2f167def9a775, produced by the Arditi et al. (2024) method.

Abliterated by the sbussiso lab research agent (Hermes, research-workstation
profile) on Google Colab T4, 2026-09-27.

Method

Arditi et al. 2024, "Refusal in LLMs is mediated by a single direction"
(NeurIPS 2024). From 64 harmful/harmless prompt pairs
(greedy decoding, seed 0), the mean difference of final-position
residual activations gives the refusal direction at each layer; the layer with
the highest direction coherence was chosen and its direction removed.

Two edits are compared in this repo's evaluation:

  • inference-time ablation: project the direction out of every activation at
    decoder layer 17 (forward hook, all positions);
  • persistent weight decoding (published artifact): the wd_B direction
    orthogonalized out of the output embedding matrix,
    W <- W - (W rhat) rhat^T. Qwen2.5-0.5B ships with tied embeddings, so the
    edit was applied to an UNTIED clone of lm_head and
    tie_word_embeddings: false is persisted in this repo's config.json (the
    input embeddings are untouched).

Ablation details

field value
chosen decoder layer 17 / 24 (hook target model.model.layers[17])
coherence (residual space) 0.664
coherence (final-layer readout space, direction B) 0.562
published variant wd_B
direction pairs / probes 64 / 16
decoding greedy (do_sample=False), max_new_tokens 200
seed 0
GPU Tesla T4
python / torch 3.13.15 / 2.11.0+cu128

Refusal behavior (16 harmful + 16 harmless probes, greedy, 200 tokens)

condition harmful refusal rate harmless answered degenerate outputs
baseline 87.5% 93.8% 0
hook_ablated 0.0% 93.8% 0
wd_A 75.0% 93.8% 0
wd_B 56.2% 93.8% 0

Headline: refusal 87.5% -> 56.2%
(published weights); benign-preservation 93.8% ->
93.8%. The inference-time hook condition measured
0.0%. Absolute refusal rates are keyword-marker based
(first-person/explicit markers only, constant scorer across conditions);
deltas are meaningful, absolute rates approximate.

Capability check: MMLU (lm-evaluation-harness, 0-shot, fp16, seed 0)

model MMLU acc acc_stderr
Qwen/Qwen2.5-0.5B-Instruct @ 7ae55760 45.78% 0.41pp
this model (wd_B) 45.49% 0.41pp

MMLU delta: 0.29pp (mission guardrail: capability loss must
stay under 3pp — PASSED). Both
evaluations used identical config (lm_eval --model hf --tasks mmlu --num_fewshot 0 --batch_size auto --seed 0, dtype float16; base loaded at the
pinned revision). Raw results in eval/.

Intended use

  • Research artifact: study of the refusal-direction phenomenon and of
    persistent weight-space ablation on a small instruct model.
  • NOT a production assistant. Refusal behavior is deliberately degraded; the
    model may produce harmful content when asked for it. Do not deploy where
    that is unacceptable. Quality/verbosity of the base model is not guaranteed
    to be preserved beyond the probes and MMLU check above.

Files

  • full safetensors weights + tokenizer (this repo root)
  • refusal_direction.npy (selected direction), refusal_direction_A.npy
    (residual space, layer 17), refusal_direction_B.npy (final-layer
    readout space)
  • eval/ - lm-eval MMLU results (base + variant) and refusal probe logs
  • run_config.json, layer_coherence.json, selection.json - run metadata
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 Abliteration" button that hands the model directly to the first-party desktop client, at the quantization your rig can actually run. No API keys, no subscription, no prompt leakage.