Goekdeniz-Guelmez/Josiefied-DeepSeek-R1-0528-Qwen3-8B-abliterated-v1
Abliterated vs deepseek-ai/DeepSeek-R1-0528-Qwen3-8B. 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: Embeddings 0/1, Attention 35/144, Norms 0/145, MLP 35/108, lm_head 0/1
4 derivatives checked
- mradermacher/Josiefied-DeepSeek-R1-0528-Qwen3-8B-abliterated-v1-i1-GGUF: Quantization
- mradermacher/Josiefied-DeepSeek-R1-0528-Qwen3-8B-abliterated-v1-GGUF: Same weights, converted
- mlx-community/Josiefied-DeepSeek-R1-0528-Qwen3-8B-abliterated-v1-4bit: Quantization
- mlx-community/Josiefied-DeepSeek-R1-0528-Qwen3-8B-abliterated-v1-bf16: Same weights, repacked