Upload README.md with huggingface_hub

README.md

@@ -2,9 +2,41 @@
 license: apache-2.0
 pipeline_tag: image-text-to-text
 library_name: transformers
+tags:
+- heretic
+- uncensored
+- decensored
+- abliterated
 ---
+# This is a decensored version of [Qwen/Qwen3-VL-32B-Instruct](https://huggingface.co/Qwen/Qwen3-VL-32B-Instruct), made using [Heretic](https://github.com/p-e-w/heretic) v1.0.1
+
+## Abliteration parameters
+
+| Parameter | Value |
+| :-------- | :---: |
+| **direction_index** | 48.00 |
+| **attn.o_proj.max_weight** | 1.13 |
+| **attn.o_proj.max_weight_position** | 38.00 |
+| **attn.o_proj.min_weight** | 0.33 |
+| **attn.o_proj.min_weight_distance** | 37.73 |
+| **mlp.down_proj.max_weight** | 1.34 |
+| **mlp.down_proj.max_weight_position** | 47.85 |
+| **mlp.down_proj.min_weight** | 1.34 |
+| **mlp.down_proj.min_weight_distance** | 14.35 |
+
+## Performance
+
+| Metric | This model | Original model ([Qwen/Qwen3-VL-32B-Instruct](https://huggingface.co/Qwen/Qwen3-VL-32B-Instruct)) |
+| :----- | :--------: | :---------------------------: |
+| **KL divergence** | 0.21 | 0 *(by definition)* |
+| **Refusals** | 4/100 | 89/100 |
+
+-----
+
+<a href="https://chat.qwenlm.ai/" target="_blank" style="margin: 2px;">
+    <img alt="Chat" src="https://img.shields.io/badge/%F0%9F%92%9C%EF%B8%8F%20Qwen%20Chat%20-536af5" style="display: inline-block; vertical-align: middle;"/>
+</a>
 
-This is 
 
 # Qwen3-VL-32B-Instruct