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

jwest33/medgemma-4b-it-null-space-abliterated

jwest33 Gemma 4.3B 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%2Fmedgemma-4b-it-null-space-abliterated"
Response includes
  • classification m1
  • files 18
  • hub_downloads_all_time 54
  • 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
54
16 last 30d - stable
Likes
0
Model age
9mo ago
created 2026-01-07
Downloads over time
Now61→from6↑917%
32445676 on Jan 761 on Oct 1161 on Oct 6JanMarMayJulSep
Jan 7 → Oct 11 · 79 snapshots · spans 277 days

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.

Variants by this author 2 formats · 449 downloads combined

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

Metadata

License
gemma
Tags
transformers safetensors gemma3 image-text-to-text gemma gemma-3 abliterated uncensored conversational arxiv:2410.02355 arxiv:2406.11717 arxiv:2310.01405

Related

Total size
8.02 GB
Files
18
Quantizations
1
Registered
2026-08-22 13:56
Last updated on HF
2026-01-07 14:41

Files by quantization

Auxiliary files 18 files 8.06 GB
model-00001-of-00002.safetensors 4.62 GB d2b1e4e7 download
model-00002-of-00002.safetensors 3.39 GB 3cb95496 download
null_space_projectors.pt 9.22 MB ff046a9f download
refusal_directions.pt 96.4 KB 6385f8bd download
tokenizer.json 31.8 MB 3ff2eb2f download
tokenizer.model 4.47 MB 1299c11d download
tokenizer_config.json 1.15 MB 2cfe89ab download
model.safetensors.index.json 89.3 KB 05e6aa91 download
README.md 3.08 KB c82e0b40 download
config.json 2.56 KB 2cc0f74b download
chat_template.jinja 1.54 KB c5f13654 download
.gitattributes 1.53 KB 52373fe2 download
special_tokens_map.json 695 B 6728103d download
preprocessor_config.json 599 B da332e16 download
abliteration_config.json 382 B 885268bb download
generation_config.json 161 B 74190f4c download
processor_config.json 74.0 B bcc0e8fd download
added_tokens.json 38.0 B f9f1f4f5 download

README current version from Hugging Face


license: gemma
library_name: transformers
base_model: google/medgemma-4b-it
pipeline_tag: image-text-to-text
tags:

  • gemma
  • gemma-3
  • abliterated
  • uncensored
  • safetensors

MedGemma 4B Instruct - Null-Space Abliterated

google/medgemma-4b-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/medgemma-4b-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 [gemma-3-4b-pt SAE models were used for circuit analysis of MedGemma3]
  • 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 637
Winsorization 99.5th percentile
Null-Space Constraints rank ratio: 0.95
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 2 versions

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

  1. 2026-01-07Upload README.md with huggingface_hub0e7b5093 KB
    Loading...
  2. 2026-01-07Upload folder using huggingface_hub93fdf453 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