Goekdeniz-Guelmez/Josiefied-Qwen3-0.6B-abliterated-v1
Abliterated vs Qwen/Qwen3-0.6B. Looks like abliteration (refusal removal): only the output projections (down_proj, o_proj) changed, in 27 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 27/84, Attention 27/112
4 derivatives checked
- DavidAU/Qwen3-MOE-6x0.6B-3.6B-Writing-On-Fire-Uncensored: MoE built from dense models
- mradermacher/Josiefied-Qwen3-0.6B-abliterated-v1-GGUF: Same weights, converted
- Goekdeniz-Guelmez/Josiefied-Qwen3-0.6B-abliterated-v1-gguf: Same weights, converted
- mlx-community/Josiefied-Qwen3-0.6B-abliterated-v1-6bit: Quantization