eggdog100/Qwen3.6-35B_Zenith-Abliterated
Abliterated vs eggdog100/Qwen3.6-35B_Zenith. Looks like abliteration (refusal removal): only the output projections (o_proj) changed, in 10 of 40 layers. Abliteration edits exactly these matrices to remove a "refusal direction"; everything else is untouched.
Tensors changed: lm_head 0/1, Embeddings 0/4, Norms 0/241, Attention 10/388, MLP 0/392