wnfldchen/gemma-4-12B-it-qat-q4_0-unquantized-heretic
Abliterated vs google/gemma-4-12B-it-qat-q4_0-unquantized. Looks like abliteration (refusal removal): only the output projections (down_proj, o_proj) changed, in 23 of 48 layers. Abliteration edits exactly these matrices to remove a "refusal direction"; everything else is untouched.
Tensors changed: Embeddings 0/12, Norms 0/289, other 0/48, MLP 23/144, Attention 23/184
3 derivatives checked
- megabytes/gemma-4-12B-it-heretic-volta-gguf: Quantization
- wnfldchen/gemma-4-12B-it-qat-q4_0-gguf-heretic: Quantization
- wnfldchen/gemma-4-12B-it-qat-w4a16-ct-heretic: Quantization