Ministral 3 3B Instruct 2512 - Null-Space Abliterated
mistralai/ministral-3-3b-instruct-2512 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/ministral-3-3b-instruct-2512-null-space-abliterated-GGUF
Abliteration Techniques Used
- Winsorization: Clips outlier activations at the 99.5th percentile for cleaner refusal direction estimation
- Null-Space Projection: Preserves model capabilities by constraining weight updates to the null space of preservation activations
- 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 |
Credits
- Base Model: mistralai/ministral-3-3b-instruct-2512-null-space-abliterated by mistralai
- Norm-Preserving Biprojected Abliteration — Jim Lai (grimjim) (2025)
- AlphaEdit: Null-Space Constrained Knowledge Editing — Fang et al. (ICLR 2025)
- Refusal in Language Models Is Mediated by a Single Direction — Arditi et al. (2024)
- Representation Engineering — Zou et al. (2023)
Toolkit Used
github.com/jwest33/abliterator
License
This model inherits the apache-2.0 from the base model. Please review and comply with mistralai'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.