← back to catalog · registered 2026-08-22 13:56

SicariusSicariiStuff/Impish_Bloodmoon_12B_Abliterated

Your rig guess connected
? Why do I need an app?
Reading your rig…

This is a rough estimate. Install the free app - we'll show exact numbers.

Reading real hardware from your app right now. Numbers below are exact.

Below is the per-quantization compatibility for this model.

curl -H "Authorization: Bearer $ABL_KEY" \
     "https://abliteration.org/api/v1/models/SicariusSicariiStuff%2FImpish_Bloodmoon_12B_Abliterated"
Response includes
  • classification m1
  • files 14
  • benchmarks 11 entries
  • hub_downloads_all_time 564
  • author_summary 75 models
  • readme_text full
10 credits · hourly refresh · ~4 KB payload Get an API key →
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 · lifetime
564
69 last 30d - stable
Likes
13
Descendants
10
in 10 direct forks
Model age
8mo ago
created 2026-02-05
Downloads over time
Now590→from15↑3,833%
021643264815 on Feb 4590 on Oct 11FebAprJunAugOct
Feb 4 → Oct 11 · 75 snapshots · spans 249 days

Benchmarks

Benchmark Score Source
Entertainment 0.7 UGI
Hazardous 2.4 UGI
Natural Intelligence 15.96 UGI
Political lean -1.5% UGI
Sensitive-Info 15.47 UGI
SocPol 1.9 UGI
UGI 34.48 UGI
Willingness (10) 7.2 UGI
W10-Adherence 4.5 UGI
W10-Direct 10 UGI
Writing 30.21 UGI

Genealogy 10 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
safetensors mistral en base_model:SicariusSicariiStuff/Impish_Bloodmoon_12B base_model:finetune:SicariusSicariiStuff/Impish_Bloodmoon_12B license:apache-2.0 region:us

Related

Total size
22.8 GB
Files
14
Quantizations
1
Registered
2026-08-22 13:56
Last updated on HF
2026-02-10 04:48

Files by quantization

Auxiliary files 14 files 22.8 GB
model-00003-of-00005.safetensors 4.57 GB f0a39824 download
model-00004-of-00005.safetensors 4.57 GB 775521a8 download
model-00002-of-00005.safetensors 4.57 GB 8517f923 download
model-00005-of-00005.safetensors 4.57 GB f11cf1d1 download
model-00001-of-00005.safetensors 4.53 GB d1ea396d download
tokenizer.json 16.3 MB 58ff6f83 download
tokenizer_config.json 173 KB eb5c5274 download
model.safetensors.index.json 29.2 KB 570599f6 download
README.md 3.48 KB 6d212e49 download
.gitattributes 1.64 KB 656b5074 download
config.json 611 B b9536785 download
special_tokens_map.json 558 B 547986c0 download
chat_template.jinja 292 B 2116e45c download
generation_config.json 111 B 2f4566f3 download

README current version from Hugging Face


license: apache-2.0
language:

base_model:

  • SicariusSicariiStuff/Impish_Bloodmoon_12B

Impish_Bloodmoon_12B
Abliterated

Developed by: SicariusSicariiStuff

Update post UGI results

Abliteration has caused this model to change alignment from Liberalism into Centrism, similarly to what happened to Impish_LLAMA_4B
Abliterated
.

I no longer think it's a coincidence; something is definitely up. Notice that the KL divergence is minimal (<0.02).

Maybe it's even research worthy, who knows. Better make another paper on how an obscure RL achieves +0.69 on math instead.


Impish_Bloodmoon_12B_Abliterated is an abliterated variant of SicariusSicariiStuff/Impish_Bloodmoon_12B with surgical removal of refusal mechanisms. This model maintains the full capabilities of the original, while eliminating safety guardrails through orthogonalization techniques.

KL divergence

<0.02

Refusals

~3%

What is KL divergence?

Think about it as a way to measure the variance between the original model "World Model," vs the abliterated one; the lower the KL divergence, the closer the "World Model" of the two models to each other.

If the original model thinks making pineapple pizza is a crime against humanity (it is), then the abliterated model will still hold to this belief, but if asked how to make one (probably after giving you a disclaimer about what an abomination that is), it would still tell you how. In other words, most of the knowledge, quirks, and capabilities are preserved.


Technical Specs

  • Base Model: Impish_Bloodmoon_12B
  • Parameters: 12B
  • Context Length: 128K tokens
  • Architecture: Mistral (decoder-only transformer)
  • Precision: bf16
  • Method: Orthogonalization-based abliteration
  • License: apache-2.0

Methodology

  1. Identifies refusal direction vectors in activation space
  2. Orthogonalizes weights to inhibit activation along these directions
  3. Preserves (mostly) all other model behaviors and knowledge

Model Details

  • Intended use: General Tasks, Roleplay.

  • Censorship level: Low - Very Low

  • 7.2 / 10 (10 completely uncensored)

UGI score:


Citation Information

@llm{Impish_Bloodmoon_12B_Abliterated,
  author = {SicariusSicariiStuff},
  title = {Impish_Bloodmoon_12B_Abliterated},
  year = {2026},
  publisher = {Hugging Face},
  url = {https://huggingface.co/SicariusSicariiStuff/Impish_Bloodmoon_12B_Abliterated}
}

Other stuff

README history 3 versions

The author's README evolved over time. Click a version to see its content at that point.

  1. 2026-02-10Update README.md06527e13.5 KB
    Loading...
  2. 2026-02-10Update README.md89675123 KB
    Loading...
  3. 2026-02-05Upload 13 filesddf77b82.8 KB
    Loading...
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.

Open in Abliteration