MuXodious/Cydonia-24B-v4.3-absolute-heresy
Abliterated vs TheDrummer/Cydonia-24B-v4.3. Looks like abliteration (refusal removal): only the output projections (down_proj, o_proj) changed, in 38 of 40 layers. Abliteration edits exactly these matrices to remove a "refusal direction"; everything else is untouched.
Tensors changed: Embeddings 0/1, Norms 0/81, MLP 38/120, Attention 37/160, lm_head 0/1
4 derivatives checked
- mradermacher/Cydonia-24B-v4.3-absolute-heresy-i1-GGUF: Quantization
- mradermacher/Cydonia-24B-v4.3-absolute-heresy-GGUF: Quantization
- jwhisenhunt/Cydonia-24B-v4.3-absolute-heresy-Q4_K_M-GGUF: Quantization
- OccultAI/DeepWater-Pleroma-24B-v1: LoRA-style
1 other repos hold byte-for-byte identical weight files.