Felldude/Qwen3.5-4B-Uncensored

Full fine-tune vs Qwen/Qwen3.5-4B-Base. Fine-tune: 93% of the weights changed, across 32 of 32 layers.

Tensors changed: Embeddings 3/5, MLP 103/199, Attention 220/324, Norms 113/209, other 1/1

2 derivatives checked

• mradermacher/Qwen3.5-4B-Uncensored-GGUF: Same weights, converted
• mradermacher/Qwen3.5-4B-Uncensored-i1-GGUF: Quantization

On the Hub · Downloads on Model Pulse