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

jwest33/gemma-4-12B-it-null-space-abliterated

jwest33 Gemma 12B multimodal
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/jwest33%2Fgemma-4-12B-it-null-space-abliterated"
Response includes
  • classification m1
  • files 17
  • hub_downloads_all_time 277
  • author_summary 20 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
277
20 last 30d - cooling
Likes
2
Descendants
2
in 2 direct forks
Model age
4mo ago
created 2026-06-10
Downloads over time
Now290→from104↑179%
95166237309104 on Jun 10290 on Oct 11290 on Oct 9JunJulAugSepOct
Jun 10 → Oct 11 · 57 snapshots · spans 123 days

Genealogy 2 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.

Variants by this author 2 formats · 308 downloads combined

The same weights this author released in different packaging. Pick the format that matches your runtime.

Metadata

License
gemma
Tags
transformers safetensors gemma4_unified image-text-to-text gemma gemma-4 abliterated uncensored conversational arxiv:2410.02355 arxiv:2406.11717 arxiv:2310.01405

Related

Total size
22.3 GB
Files
17
Quantizations
1
Registered
2026-08-22 13:56
Last updated on HF
2026-06-10 13:54

Files by quantization

Auxiliary files 17 files 22.3 GB
model-00004-of-00005.safetensors 4.64 GB 1776fd43 download
model-00002-of-00005.safetensors 4.63 GB 9b91224c download
model-00001-of-00005.safetensors 4.62 GB be598c8d download
model-00003-of-00005.safetensors 4.55 GB 7a308043 download
model-00005-of-00005.safetensors 3.84 GB e5c06503 download
null_space_projectors.pt 2.73 MB b30209a7 download
refusal_directions.pt 196 KB 2bbaa3f9 download
tokenizer.json 30.7 MB cc8d3a0c download
model.safetensors.index.json 65.4 KB 65a96b78 download
chat_template.jinja 17.1 KB e61bbfe9 download
config.json 4.32 KB 324c841b download
README.md 2.97 KB 9fffe5c6 download
abliteration_config.json 2.62 KB 0bbbf1cc download
tokenizer_config.json 2.05 KB 68354f96 download
.gitattributes 1.53 KB 52373fe2 download
processor_config.json 1.35 KB b889adcd download
generation_config.json 260 B d09dccf1 download

README current version from Hugging Face


license: gemma
library_name: transformers
base_model: google/gemma-4-12b-it
tags:

  • gemma
  • gemma-4
  • abliterated
  • uncensored
  • safetensors

Gemma 4 12B Instruct - Null-Space Abliterated

google/gemma-4-12b-it with refusal behavior removed via orthogonal projection. Uses null-space constraints and adaptive layer weighting to preserve model capabilities.

Note: This model will produce uncensored outputs. Use responsibly.

GGUF quantizations available at: jwest33/gemma-4-12b-it-null-space-abliterated-GGUF

Abliteration Techniques Used

  • Winsorization: Clips outlier activations at the 99th percentile for cleaner refusal direction estimation (recommended for Gemma models)
  • Null-Space Projection: Preserves model capabilities by constraining weight updates to the null space of preservation activations
    • Preservation Prompts: Dynamically generated using Gemma Scope 2 complete SAE circuit analysis to ensure complete coverage of shared features activated by harmful prompts, without overextending into unrelated capability space
  • Adaptive Weighting: Applies Gaussian-weighted per-layer ablation strength, focusing on middle-to-later layers where refusal behavior concentrates
  • Norm Preservation: Maintains original Frobenius norms of weight matrices after projection
Parameter Value
Harmful Prompts 5000
Harmless Prompts 1226
Winsorization 99.5th percentile
Null-Space Constraints rank ratio: 0.90
Directional Multiplier 1.10
SAE Targeted Coverage 1.00

Credits

Toolkit Used

github.com/jwest33/abliterator

License

This model inherits the Gemma license from the base model. Please review and comply with Google's usage terms.

Disclaimer

This model is provided for research and educational purposes. The creators are not responsible for any misuse. Users are solely responsible for ensuring their use complies with applicable laws and ethical standards.

README history 1 version

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

  1. 2026-06-10Upload folder using huggingface_hub9763daf2.9 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