AEON-7/Gemma-4-12B-it-AEON-Abliterated-K4-BF16
Abliterated vs google/gemma-4-12B-it. Looks like abliteration (refusal removal): only the output projections (down_proj, o_proj) changed, in 24 of 48 layers. Abliteration edits exactly these matrices to remove a "refusal direction"; everything else is untouched.
Tensors changed: Embeddings 0/12, Norms 0/289, other 0/48, MLP 24/144, Attention 24/184
9 derivatives checked
- AEON-7/Gemma-4-12B-it-AEON-Abliterated-MLX-8bit: Quantization
- AEON-7/Gemma-4-12B-it-AEON-Abliterated-MLXFP4: Quantization
- AEON-7/Gemma-4-12B-it-AEON-Abliterated-K4-NVFP4: Quantization
- AEON-7/Gemma-4-12B-it-AEON-Abliterated-K4-FP8: Quantization
- AEON-7/Gemma-4-12B-it-AEON-Abliterated-K4-NVFP4-FP8: Quantization
- mradermacher/Gemma-4-12B-it-AEON-Abliterated-K4-BF16-i1-GGUF: Quantization
- Abiray/Gemma-4-12B-it-AEON-Abliterated-K4-GGUF: Quantization
- mradermacher/Gemma-4-12B-it-AEON-Abliterated-K4-BF16-GGUF: Quantization
- timfromhcs/Gemma-4-12B-it-AEON-Abliterated-K4-GGUF: Quantization
1 other repos hold byte-for-byte identical weight files.