KridgeDookie/Gemma-4-E4B-IT-ABLITERATED-UNCENSORED-PHILADELPHIA-CLASS
Abliterated vs google/gemma-4-E4B-it. Looks like abliteration (refusal removal): only the output projections (down_proj, o_proj) changed, in 12 of 42 layers. Abliteration edits exactly these matrices to remove a "refusal direction"; everything else is untouched.
Tensors changed: MLP 12/606, Norms 0/484, other 0/261, Attention 12/719, Embeddings 0/6
3 derivatives checked
- mradermacher/Gemma-4-E4B-IT-ABLITERATED-UNCENSORED-PHILADELPHIA-CLASS-i1-GGUF: Quantization (fewer layers)
- mradermacher/Gemma-4-E4B-IT-ABLITERATED-UNCENSORED-PHILADELPHIA-CLASS-GGUF: Same weights, converted (fewer layers)
- marafx2007/Gemma-4-E4B-IT-ABLITERATED-UNCENSORED-PHILADELPHIA-CLASS-i1-GGUF: Quantization (fewer layers)