richardyoung/Phi-4-mini-instruct-heretic
Abliterated vs microsoft/Phi-4-mini-instruct. Looks like abliteration (refusal removal): only the output projections (down_proj, o_proj) changed, in 23 of 32 layers. Abliteration edits exactly these matrices to remove a "refusal direction"; everything else is untouched.
Tensors changed: Embeddings 0/1, Norms 0/65, MLP 21/64, Attention 23/64
1 derivatives checked
- richardyoung/Phi-4-mini-instruct-heretic-GGUF: Quantization