Goekdeniz-Guelmez/Josiefied-Qwen2.5-0.5B-Instruct-abliterated-v1
Abliterated vs Qwen/Qwen2.5-0.5B-Instruct. Looks like abliteration (refusal removal): only the output projections (down_proj, embed_tokens, o_proj) changed, in 24 of 24 layers. Abliteration edits exactly these matrices to remove a "refusal direction"; everything else is untouched.
Tensors changed: Embeddings 1/1, Norms 0/49, MLP 24/72, Attention 24/168
4 derivatives checked
- mradermacher/Josiefied-Qwen2.5-0.5B-Instruct-abliterated-v1-i1-GGUF: Quantization
- mradermacher/Josiefied-Qwen2.5-0.5B-Instruct-abliterated-v1-GGUF: Same weights, converted
- mlx-community/Josiefied-Qwen2.5-0.5B-Instruct-abliterated-v1: Same weights, converted
- mlx-community/Josiefied-Qwen2.5-0.5B-Instruct-abliterated-v1-4bit: Quantization