DavidAU/Qwen3-4B-2507-Thinking-heretic-abliterated-uncensored
Abliterated vs Qwen/Qwen3-4B-Thinking-2507. Looks like abliteration (refusal removal): only the output projections (down_proj, o_proj) changed, in 33 of 36 layers. Abliteration edits exactly these matrices to remove a "refusal direction"; everything else is untouched.
Tensors changed: Embeddings 0/1, Norms 0/145, MLP 33/108, Attention 27/144
2 derivatives checked
- mradermacher/Qwen3-4B-2507-Thinking-heretic-abliterated-uncensored-i1-GGUF: Quantization
- mradermacher/Qwen3-4B-2507-Thinking-heretic-abliterated-uncensored-GGUF: Same weights, converted