aifeifei798/Gemma-4-31B-Cognitive-Unshackled
Abliterated vs google/gemma-4-31B-it. Looks like abliteration (refusal removal): only the output projections (down_proj, o_proj) changed, in 40 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 40/261, Attention 40/338
3 derivatives checked
- mradermacher/Gemma-4-31B-Cognitive-Unshackled-i1-GGUF: Quantization (fewer layers)
- mradermacher/Gemma-4-31B-Cognitive-Unshackled-GGUF: Quantization (fewer layers)
- s1arsky/Gemma-4-31B-Cognitive-Unshackled-Q4_KS_GGUF: Quantization (fewer layers)