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

chulcher/gemma-4-e4b-abliterated

chulcher Gemma 8.0B
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/chulcher%2Fgemma-4-e4b-abliterated"
Response includes
  • classification m1
  • files 11
  • hub_downloads_all_time 602
  • author_summary 1 models
  • readme_text full
Try next
GET/models?family=gemma&method=m1
Sibling gemma models · same method
 GET/methods/m1
What "m1" means · taxonomy
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)
How we label → About M1 →
Refusal direction extracted via
Extraction technique

Difference-of-means

FromArditi et al. (2024) ↗
Confidence
MEDIUM
Why we say so
primary_method=M1; difference-of-means is the reference extraction for M1/M3 (Arditi 2024)
What is a refusal direction? → Read the paper ↗
Downloads · lifetime
602
17 last 30d - cooling
Likes
1
Model age
6mo ago
created 2026-04-05
Downloads over time
Now607→from540↑12%
537562588614540 on Apr 15607 on Oct 11607 on Oct 8AprMayJunJulAugSepOct
Apr 15 → Oct 11 · 65 snapshots · spans 179 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.

google/gemma-4-e4b-it this lineage
chulcher/gemma-4-e4b-abliterated↓ 17

Metadata

License
apache-2.0
Languages
en
Tags
safetensors gemma4 abliterated uncensored heretic text-generation conversational en license:apache-2.0 region:us

Related

Same base model
HauhauCS/Gemma-4-E4B-Uncensored-HauhauCS-Aggressive 1.4M LuffyTheFox/Gemma-4-E4B-Uncensored-Genesis-GGUF 6K Amory0201/Gemma-4-E4B-Uncensored-HauhauCS-Aggressive 3K RusteddDDD/Gemma-4-E4B-Uncensored-HauhauCS-Aggressive 1K tripolskypetr/Gemma-4-Uncensored-Aggressive-GGUF 1K
Total size
14.9 GB
Files
11
Quantizations
1
Registered
2026-08-22 13:56
Last updated on HF
2026-04-05 22:42
Source huggingface.co/chulcher/gemma-4-e4b-abliterated ↗ Base model google/gemma-4-e4b-it ↗

Files by quantization

Auxiliary files 11 files 14.9 GB
model.safetensors 14.9 GB 4ede376d download
tokenizer.json 30.7 MB 94ee542d download
refusal_test_results.json 34.7 KB 3ccf6b97 download
chat_template.jinja 11.6 KB afb1d517 download
config.json 5.01 KB ea1503cc download
.gitattributes 1.53 KB 52373fe2 download
tokenizer_config.json 1.46 KB aedf56a5 download
README.md 1.35 KB d7f12c70 download
pareto_trials.json 1.02 KB f806b8ca download
abliteration_metadata.json 584 B 0f59f1e3 download
generation_config.json 176 B 57a362be download

README current version from Hugging Face

license: apache-2.0
base_model: google/gemma-4-e4b-it
tags:

  • abliterated
  • uncensored
  • heretic
  • gemma4
    language:
  • en
    pipeline_tag: text-generation

Gemma 4 E4B — Abliterated

Google's Gemma 4 E4B with refusal behavior removed via Heretic directional ablation. No fine-tuning — weights were directly modified to suppress the refusal direction in activation space.

Results

Metric Base Gemma 4 E4B Abliterated
Refusal Rate 41.2% (40/97) 2.0% (1/50)
KL Divergence — 0.034

Abliteration Details

  • Tool: Heretic v1.1.0
  • Hardware: NVIDIA DGX Spark (GB10 Blackwell, 128GB unified memory)
  • Time: ~6 hours
  • Trials: 30 (10 random + 20 TPE-guided via Optuna)
  • Best trial: #21 — 9/100 refusals, 0.034 KL divergence
  • Components modified: attn.o_proj + mlp.down_proj across all 42 layers
  • Parameters:
    • Direction index: per layer
    • attn.o_proj weights: 0.73–1.49 (peak at layer 24.83)
    • mlp.down_proj weights: 0.48–1.11 (peak at layer 35.18)

Acknowledgments

README history 3 versions

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

  1. 2026-04-05Simplify model card — just the facts627232c1.4 KB
    Loading...
  2. 2026-04-05Remove Intended Use section7554b503.7 KB
    Loading...
  3. 2026-04-05Initial upload: Gemma 4 E4B abliterated via Heretic (41% → 2% refusal rate, 0...37786053.9 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.

Get the app Browse the catalog
Open in Abliteration