Downloads · 30 days
7
1% of all-time downloads
martin-t-studio/Qwen3-4B-Instruct-2507-heretic
Qwen3-4B-Instruct-2507-heretic is a text generation model from martin-t-studio. Use it when you need the model to write or continue text. It is set up for transformers. The card lists the license as apache-2.0.
Downloads · 30 days
7
1% of all-time downloads
All-time downloads
486
Public
Parameters
4B
8.1 GB on disk
Likes
0
Public
Click a slice to open those files.
.safetensors8 GB · 100%
From the Hugging Face model README
| Parameter | Value |
|---|---|
| direction_index | 18.78 |
| attn.o_proj.max_weight | 1.45 |
| attn.o_proj.max_weight_position | 26.86 |
| attn.o_proj.min_weight | 1.34 |
| attn.o_proj.min_weight_distance | 20.09 |
| mlp.down_proj.max_weight | 1.10 |
| mlp.down_proj.max_weight_position | 29.91 |
| mlp.down_proj.min_weight | 0.67 |
| mlp.down_proj.min_weight_distance | 20.77 |
| Metric | This model | Original model (Qwen/Qwen3-4B-Instruct-2507) |
|---|---|---|
| KL divergence | 0.0929 | 0 (by definition) |
| Refusals | 16/100 | 100/100 |
This model is a decensored/abliterated version of the base Qwen3 model. By using this model, you acknowledge and agree to the following:
If you find my work useful, a cup of coffee would be greatly appreciated!