Zynerji/Ektome-StableLM-2-1.6B-Chat-PristinelyUncensored
Abliterated vs stabilityai/stablelm-2-1_6b-chat. Looks like abliteration (refusal removal): only the output projections (down_proj, 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 0/1, Norms 0/98, MLP 24/72, Attention 24/168, lm_head 0/1
2 derivatives checked
- mradermacher/Ektome-StableLM-2-1.6B-Chat-PristinelyUncensored-i1-GGUF: Quantization
- mradermacher/Ektome-StableLM-2-1.6B-Chat-PristinelyUncensored-GGUF: Same weights, converted