llmfan46/Qwen3.5-40B-Claude-4.5-Opus-High-Reasoning-Thinking-uncensored-heretic
Abliterated vs DavidAU/Qwen3.5-40B-Claude-4.5-Opus-High-Reasoning-Thinking. Looks like abliteration (refusal removal): only the output projections (down_proj, o_proj, out_proj) changed, in 42 of 96 layers. Abliteration edits exactly these matrices to remove a "refusal direction"; everything else is untouched.
Tensors changed: lm_head 0/1, Embeddings 0/4, Norms 0/423, Attention 42/780, MLP 40/400
3 derivatives checked
- llmfan46/Qwen3.5-40B-Claude-4.5-Opus-High-Reasoning-Thinking-uncensored-heretic-GGUF: Same weights, converted
- BrockMisner/Qwen3.5-40B-Claude-4.5-Opus-High-Reasoning-Thinking-uncensored-heretic-GGUF: Same weights, converted
- Octavian710/Qwen3.5-40B-Claude-4.5-Opus-High-Reasoning-Thinking-uncensored-heretic-GGUF: Same weights, converted