llmfan46/Qwen3.6-27B-uncensored-heretic-v2
Abliterated vs Qwen/Qwen3.6-27B. Looks like abliteration (refusal removal): only the output projections (down_proj, o_proj, out_proj) changed, in 64 of 64 layers. Abliteration edits exactly these matrices to remove a "refusal direction"; everything else is untouched.
Tensors changed: Embeddings 0/4, MLP 64/304, Attention 56/556, lm_head 0/1, Norms 0/319
11 derivatives checked
- llmfan46/Qwen3.6-27B-uncensored-heretic-v2-GGUF: Same weights, converted
- Jolex98/Qwen3.6-27B-uncensored-heretic-v2-GGUF: Quantization
- llmfan46/Qwen3.6-27B-uncensored-heretic-v2-NVFP4-GGUF: Quantization
- llmfan46/Qwen3.6-27B-uncensored-heretic-v2-GPTQ-Int8: Quantization
- froggeric/Qwen3.6-27B-Uncensored-Heretic-v2-MLX-4bit: Quantization
- llmfan46/Qwen3.6-27B-uncensored-heretic-v2-GPTQ-Int4: Quantization
- CCSSNE/llmfan46-Qwen3.6-27B-uncensored-heretic-v2-GGUF: Same weights, converted
- leonsarmiento/Qwen3.6-27B-uncensored-heretic-v2-3bit-mlx: Quantization
- froggeric/Qwen3.6-27B-Uncensored-Heretic-v2-MLX-8bit: Quantization
- froggeric/Qwen3.6-27B-Uncensored-Heretic-v2-MLX-6bit: Quantization
- llmfan46/Qwen3.6-27B-uncensored-heretic-v2-NVFP4: Quantization
2 other repos hold byte-for-byte identical weight files.