MuXodious/Tiamat-24B-Magistral-PaperWitch-heresy
Abliterated vs Gryphe/Tiamat-24B-Magistral. Looks like abliteration (refusal removal): only the output projections (down_proj, o_proj) changed, in 35 of 40 layers. Abliteration edits exactly these matrices to remove a "refusal direction"; everything else is untouched.
Tensors changed: Embeddings 0/1, Norms 0/131, MLP 27/192, Attention 32/256, other 0/4, lm_head 0/1
2 derivatives checked
- mradermacher/Tiamat-24B-Magistral-PaperWitch-heresy-i1-GGUF: Quantization (fewer layers)
- mradermacher/Tiamat-24B-Magistral-PaperWitch-heresy-GGUF: Quantization (fewer layers)