MuXodious/gemma-4-E4B-it-SOMPOA-heresy
Abliterated vs google/gemma-4-E4B-it. Looks like abliteration (refusal removal): only the output projections (down_proj, o_proj) changed, in 32 of 42 layers. Abliteration edits exactly these matrices to remove a "refusal direction"; everything else is untouched.
Tensors changed: MLP 32/606, Norms 0/502, other 0/261, Attention 28/755, Embeddings 0/6
3 derivatives checked
- mradermacher/gemma-4-E4B-it-SOMPOA-heresy-i1-GGUF: Quantization (fewer layers)
- mradermacher/gemma-4-E4B-it-SOMPOA-heresy-GGUF: Abliterated (fewer layers)
- Nubinu/Gemma4-E4B-MiniFantasy-V1: Partial fine-tune