license: apache-2.0
base_model: empero-ai/Qwen3.8-4B-Distill
pipeline_tag: text-generation
library_name: transformers
language:
- en
tags:- qwen3.8
- qwen3.8-4b
- distill
- empero-ai
- safetensors
- abliterated
- uncensored
- decensored
- ara
- arbitrary-rank-ablation
- heretic
- refusal-removed
- roleplay
- creative-writing
- chat
- text-generation
- austriani
Austriani // Heretic // Safetensors
Qwen3.8-4B-Distill
ARA Abliterated
Refusal Removal via Arbitrary-Rank Ablation
4B
Parameters
3/100
Refusals
0.0586
KLD
15/40
Trial
Base Architecture
This repository hosts the Safetensors ARA Abliteration of empero-ai/Qwen3.8-4B-Distill. This is a distilled version of Qwen3.8-4B, optimized for efficient inference while retaining strong reasoning and agentic capabilities. The model is released under the Apache 2.0 license.
The original Qwen3.8-4B architecture uses a dense 4B parameter transformer with GQA attention and a native context window of 32,768 tokens, extendable to 131,072 with YaRN. This distill version preserves those core capabilities while reducing the compute footprint for edge deployment.
Abliteration Report // ARA Method
This release uses Arbitrary-Rank Ablation (ARA) via Heretic. The optimizer balances three competing objectives: preserve harmless behavior, align harmful outputs with harmless ones, and overcorrect harmful outputs away from their original state. The parameters below were used for the winning trial.
Trial 15 of 40 produced a refusal rate of 3/100 with a KL divergence of 0.0586. This indicates a mature optimization run where the Pareto frontier had stabilized into a favorable trade-off between capability and compliance.
View Full ARA Parameters
start_layer_index
3
end_layer_index
21
preserve_good
0.6106
steer_bad
0.0032
overcorrect
0.7700
neighbor_count
12
Usage & Compatibility
This is a standard Transformers model released in Safetensors format. It can be loaded directly using the
AutoModelForCausalLM and AutoTokenizer classes. It is fully compatible with vLLM, TGI, and standard PyTorch inference pipelines.
For GGUF or ik_llama.cpp quantizations, please refer to my dedicated quantized repositories.