Goekdeniz-Guelmez/Josiefied-Health-Qwen3-8B-abliterated-v1
Abliterated vs Intelligent-Internet/II-Medical-8B. Looks like abliteration (refusal removal): only the output projections (down_proj, o_proj) changed, in 35 of 36 layers. Abliteration edits exactly these matrices to remove a "refusal direction"; everything else is untouched.
Tensors changed: lm_head 0/1, Attention 35/144, Norms 0/145, MLP 35/108, Embeddings 0/1
3 derivatives checked
- mradermacher/Josiefied-Health-Qwen3-8B-abliterated-v1-i1-GGUF: Quantization
- mradermacher/Josiefied-Health-Qwen3-8B-abliterated-v1-GGUF: Same weights, converted
- mlx-community/Josiefied-Health-Qwen3-8B-abliterated-v1-8bit: Quantization