Sorihon/MN-GRAND-Gutenberg-Lyra4-Lyra-12B-DARKNESS-heretic
Abliterated vs DavidAU/MN-GRAND-Gutenberg-Lyra4-Lyra-12B-DARKNESS. Looks like abliteration (refusal removal): only the output projections (down_proj, o_proj) changed, in 37 of 40 layers. Abliteration edits exactly these matrices to remove a "refusal direction"; everything else is untouched.
Tensors changed: lm_head 0/1, Embeddings 0/1, Norms 0/81, MLP 36/120, Attention 37/160