grayarea/Phi-4-mini-instruct-heretic-v1.2
Abliterated vs microsoft/Phi-4-mini-instruct. Looks like abliteration (refusal removal): only the output projections (down_proj, o_proj) changed, in 22 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 18/64, Attention 22/64
2 derivatives checked
- mradermacher/Phi-4-mini-instruct-heretic-v1.2-i1-GGUF: Quantization
- mradermacher/Phi-4-mini-instruct-heretic-v1.2-GGUF: Same weights, converted