ChiKoi7/GPT-5-Distill-llama3.2-3B-Instruct-Heretic
Abliterated vs Jackrong/GPT-5-Distill-llama3.2-3B-Instruct. Looks like abliteration (refusal removal): only the output projections (down_proj, o_proj) changed, in 25 of 28 layers. Abliteration edits exactly these matrices to remove a "refusal direction"; everything else is untouched.
Tensors changed: Embeddings 0/1, Norms 0/57, MLP 23/84, Attention 25/112
2 derivatives checked
- mradermacher/GPT-5-Distill-llama3.2-3B-Instruct-Heretic-i1-GGUF: Quantization
- mradermacher/GPT-5-Distill-llama3.2-3B-Instruct-Heretic-GGUF: Same weights, converted