KridgeDookie/Qwen3.8-27B-ABLITERATED-UNCENSORED-PHILADELPHIA-CLASS
Abliterated vs Qwen/Qwen3.8-27B. Looks like abliteration (refusal removal): only the output projections (down_proj, o_proj, out_proj) changed, in 63 of 64 layers. Abliteration edits exactly these matrices to remove a "refusal direction"; everything else is untouched.
Tensors changed: Norms 0/319, Attention 63/556, MLP 63/304, Embeddings 0/4, lm_head 0/1
3 derivatives checked
- TheCluster/Qwen3.8-27B-ABLITERATED-UNCENSORED-PHILADELPHIA-CLASS-MLX-mixed-9.4bit: Quantization
- TheCluster/Qwen3.8-27B-ABLITERATED-UNCENSORED-PHILADELPHIA-CLASS-MLX-mixed-6bit: Quantization
- TheCluster/Qwen3.8-27B-ABLITERATED-UNCENSORED-PHILADELPHIA-CLASS-MLX-mixed-7.4bit: Quantization
1 other repos hold byte-for-byte identical weight files.