nitrox/ToolACE-2.5-Llama-3.1-8B-heretic
Abliterated vs Team-ACE/ToolACE-2.5-Llama-3.1-8B. Looks like abliteration (refusal removal): only the output projections (down_proj, o_proj) changed, in 28 of 32 layers. Abliteration edits exactly these matrices to remove a "refusal direction"; everything else is untouched.
Tensors changed: Embeddings 0/1, Norms 0/65, MLP 8/96, Attention 28/128, lm_head 0/1
2 derivatives checked
- mradermacher/ToolACE-2.5-Llama-3.1-8B-heretic-i1-GGUF: Quantization
- mradermacher/ToolACE-2.5-Llama-3.1-8B-heretic-GGUF: Same weights, converted