huihui-ai/Huihui-Qwen3-0.6B-abliterated-v2
Abliterated vs Qwen/Qwen3-0.6B. Looks like abliteration (refusal removal): only the output projections (down_proj, o_proj) changed, in 28 of 28 layers. Abliteration edits exactly these matrices to remove a "refusal direction"; everything else is untouched.
Tensors changed: Embeddings 0/1, Norms 0/113, MLP 28/84, Attention 28/112
4 derivatives checked
- mradermacher/Huihui-Qwen3-0.6B-abliterated-v2-GGUF: Same weights, converted
- mradermacher/Huihui-Qwen3-0.6B-abliterated-v2-i1-GGUF: Quantization
- 3MPER0RR/Qwen3-06B-3MPER0RR-abliterated: Abliterated
- vXofi/businessgpt-v5-qwen3-0.6b: LoRA-style