richardyoung/Muse-Glimmer-30B-heretic
Abliterated vs meta-models/Muse-Glimmer-30B. Looks like abliteration (refusal removal): only the output projections (down_proj, o_proj) changed, in 18 of 52 layers. Abliteration edits exactly these matrices to remove a "refusal direction"; everything else is untouched.
Tensors changed: Embeddings 0/3, Norms 0/413, MLP 9/358, Attention 18/660, lm_head 0/1, other 0/1
1 derivatives checked
- richardyoung/Muse-Glimmer-30B-heretic-GGUF: Quantization