PS4CoT/phi4-reasoning-sdf-false-1k
Abliterated vs microsoft/Phi-4-reasoning. Looks like abliteration (refusal removal): only the output projections (down_proj, o_proj) changed, in 40 of 40 layers. Abliteration edits exactly these matrices to remove a "refusal direction"; everything else is untouched.
Tensors changed: Embeddings 0/1, Norms 0/81, MLP 40/80, Attention 40/80, lm_head 0/1
1 other repos hold byte-for-byte identical weight files.