MuXodious/SmolLM3-3B-absolute-heresy
Abliterated vs HuggingFaceTB/SmolLM3-3B. Looks like abliteration (refusal removal): only the output projections (down_proj, o_proj) changed, in 31 of 36 layers. Abliteration edits exactly these matrices to remove a "refusal direction"; everything else is untouched.
Tensors changed: Embeddings 0/1, Norms 0/73, MLP 31/108, Attention 28/144
2 derivatives checked
- mradermacher/SmolLM3-3B-absolute-heresy-i1-GGUF: Quantization
- mradermacher/SmolLM3-3B-absolute-heresy-GGUF: Same weights, converted