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