Yobenboben/QwQ-32B-Brh
Abliterated vs ArliAI/QwQ-32B-ArliAI-RpR-v2. Looks like abliteration (refusal removal): only the output projections (down_proj, o_proj) changed, in 63 of 64 layers. Abliteration edits exactly these matrices to remove a "refusal direction"; everything else is untouched.
Tensors changed: Norms 0/129, MLP 63/192, Attention 63/448
2 derivatives checked
- mradermacher/QwQ-32B-Brh-i1-GGUF: Quantization
- mradermacher/QwQ-32B-Brh-GGUF: Quantization