TrevorJS/gemma-4-31B-it-uncensored
Abliterated vs google/gemma-4-31B-it. Looks like abliteration (refusal removal): only the output projections (down_proj, o_proj) changed, in 59 of 60 layers. Abliteration edits exactly these matrices to remove a "refusal direction"; everything else is untouched.
Tensors changed: Embeddings 0/4, Norms 0/523, other 0/62, MLP 59/261, Attention 58/338
3 derivatives checked
- TrevorJS/gemma-4-31B-it-uncensored-GGUF: Quantization (fewer layers)
- mradermacher/gemma-4-31B-it-uncensored-i1-GGUF: Quantization (fewer layers)
- mradermacher/gemma-4-31B-it-uncensored-GGUF: Quantization (fewer layers)
6 other repos hold byte-for-byte identical weight files.