Youssofal/Qwen3.6-27B-Abliterated-Heretic-Uncensored-GGUF
Abliterated vs Qwen/Qwen3.6-27B. Youssofal/Qwen3.6-27B-Abliterated-Heretic-Uncensored-GGUF is a GGUF (BF16) of different weights. Looks like abliteration (refusal removal): only the output projections (down_proj, o_proj) changed, in 50 of 64 layers. Abliteration edits exactly these matrices to remove a "refusal direction"; everything else is untouched.
Tensors changed: Embeddings 0/1, MLP 50/192, Attention 10/64, lm_head 0/1, Norms 0/161