Downloads · 30 days
17
3% of all-time downloads
gemma-4-e4b-abliterated
chulcher/gemma-4-e4b-abliterated
gemma-4-e4b-abliterated is a text generation model from chulcher. Use it when you need the model to write or continue text. The card lists the license as apache-2.0.
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.
All-time downloads
607
Public
Parameters
8B
16 GB on disk
Likes
1
Public
Repo makeup
Click a slice to open those files.
.safetensors16 GB · 100%
At a glance
- Task
- Text Generation
- License
- apache-2.0
- Model type
- gemma4
- Access
- Public
- Created
- Apr 5, 2026
- Updated
- Apr 5, 2026
- SHA
- 627232c9
Try a prompt
Base models
Model card
From the Hugging Face model README
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
- Google Gemma 4 E4B (Apache 2.0)
- Heretic by Philipp Emanuel Weidmann
- Optuna
- Task
- Text Generation
- Type
- gemma4
- License
- apache-2.0
- Languages
- en
- Created
- Apr 5, 2026
- Updated
- Apr 5, 2026