91 lines
2.6 KiB
Markdown
91 lines
2.6 KiB
Markdown
---
|
|
base_model: Qwen/Qwen3-0.6B
|
|
tags:
|
|
- uncensored
|
|
- gabliteration
|
|
datasets:
|
|
- mlabonne/harmless_alpaca
|
|
- mlabonne/harmful_behaviors
|
|
library_name: gabliteration
|
|
arxiv: "2512.18901"
|
|
|
|
model-index:
|
|
- name: Qwen_Qwen3-0.6B-gabliterated
|
|
results:
|
|
- task:
|
|
type: text-generation
|
|
dataset:
|
|
type: harmless_alpaca
|
|
name: Harmless Alpaca
|
|
metrics:
|
|
- name: KL Divergence
|
|
type: pass@1
|
|
value: 0.0591
|
|
|
|
- task:
|
|
type: text-generation
|
|
dataset:
|
|
type: harmful_behaviors
|
|
name: Harmful Behaviors
|
|
metrics:
|
|
- name: Refusal Rate
|
|
type: pass@1
|
|
value: 0.05
|
|
---
|
|
|
|
# Gabliterated Model Series
|
|
|
|
|
|
|
|
## Overview
|
|
|
|
With this model series, I introduce the first **Gabliteration**, a novel neural weight modification technique that advances beyond traditional abliteration methods through adaptive multi-directional projections with regularized layer selection.
|
|
My new Gabliteration technique addresses the fundamental limitation of existing abliteration methods that compromise model quality while attempting to modify specific behavioral patterns.
|
|
|
|
```text
|
|
Refusal: 5/100
|
|
KL Div: 0.0591
|
|
Config:
|
|
Samples: 400
|
|
Skip: [4, 3]
|
|
Layer: 0.66 (selected: 18)
|
|
Scale: 0.48
|
|
λ: 0.05
|
|
k: 3
|
|
β: 0.54
|
|
Adaptive: False
|
|
τ: 0.84
|
|
```
|
|
|
|
## Model Variants
|
|
|
|
This series includes models ranging from 0.6B to 32B parameters, demonstrating the scalability and effectiveness of the Gabliteration technique across different model sizes.
|
|
|
|
## Quants
|
|
|
|
- [GGUF (mradermacher)]()
|
|
|
|
## Technical Background
|
|
|
|
Building upon the foundational work of Arditi et al. (2024) on single-direction abliteration, Gabliteration extends to a comprehensive multi-directional framework with theoretical guarantees.
|
|
My method employs singular value decomposition on difference matrices between harmful and harmless prompt representations to extract multiple refusal directions.
|
|
|
|
### Dynamic Layer Selection
|
|
|
|
This model was created using fixed layer selection.
|
|
A fixed layer fraction was used based on empirical tuning.
|
|
|
|
Selected layer: **18** (out of 28 total layers)
|
|
|
|
## Citation
|
|
|
|
If you use these models, please cite the original research (paper coming later this year):
|
|
|
|
```
|
|
Gülmez, G. (2025). Gabliteration: Adaptive Multi-Directional Neural Weight Modification for Selective Behavioral Alteration in Large Language Models. https://arxiv.org/abs/2512.18901
|
|
```
|
|
|
|
## Acknowledgments
|
|
|
|
This work builds upon the foundational research by Arditi et al. (2024) on refusal direction identification in large language models.
|
|
|