wangzhang/Qwen3.5-9B-abliterated
Abliterated vs Qwen/Qwen3.5-9B. Looks like abliteration (refusal removal): only the output projections (down_proj, o_proj, out_proj) changed, in 30 of 32 layers. Abliteration edits exactly these matrices to remove a "refusal direction"; everything else is untouched.
Tensors changed: lm_head 0/1, Embeddings 0/4, MLP 24/208, Attention 30/332, Norms 0/215
2 derivatives checked
- mradermacher/Qwen3.5-9B-abliterated-GGUF: Same weights, converted
- mradermacher/Qwen3.5-9B-abliterated-i1-GGUF: Quantization