llmfan46/Darkidol-Ballad-27B-ultra-uncensored-heretic-v2
Abliterated vs aifeifei798/Darkidol-Ballad-27B. Looks like abliteration (refusal removal): only the output projections (o_proj, out_proj) changed, in 26 of 64 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 0/209, Attention 26/448, MLP 0/192
3 derivatives checked
- mradermacher/Darkidol-Ballad-27B-ultra-uncensored-heretic-v2-i1-GGUF: Quantization
- mradermacher/Darkidol-Ballad-27B-ultra-uncensored-heretic-v2-GGUF: Quantization
- llmfan46/Darkidol-Ballad-27B-ultra-uncensored-heretic-v2-GGUF: Same weights, converted