huihui-ai/Huihui-Qwopus3.5-9B-v3-abliterated
Abliterated vs Jackrong/Qwopus3.5-9B-v3. Looks like abliteration (refusal removal): only the output projections (down_proj, o_proj, out_proj) changed, in 32 of 32 layers. Abliteration edits exactly these matrices to remove a "refusal direction"; everything else is untouched.
Tensors changed: lm_head 0/1, Embeddings 0/5, MLP 33/211, Attention 33/336, other 0/1, Norms 0/221
4 derivatives checked
- mradermacher/Huihui-Qwopus3.5-9B-v3-abliterated-GGUF: Same weights, converted
- mradermacher/Huihui-Qwopus3.5-9B-v3-abliterated-i1-GGUF: Quantization
- XChava/cyber-ken-v3-GGUF: Quantization
- dungnvt/Huihui-Qwopus3.5-9B-v3-abliterated-GGUF: Quantization
1 other repos hold byte-for-byte identical weight files.