huihui-ai/Huihui-Qwen3.6-35B-A3B-Claude-4.6-Opus-abliterated
Abliterated vs hesamation/Qwen3.6-35B-A3B-Claude-4.6-Opus-Reasoning-Distilled. Looks like abliteration (refusal removal): only the output projections (down_proj, o_proj, out_proj) changed, in 30 of 40 layers. Abliteration edits exactly these matrices to remove a "refusal direction"; everything else is untouched.
Tensors changed: Embeddings 0/5, Attention 30/392, MLP 60/399, Norms 0/247, lm_head 0/1, other 0/1