Andycurrent/gemma-3-12b-it-uncensored-GGUF

Abliterated (fewer layers) vs google/gemma-3-12b-it. Andycurrent/gemma-3-12b-it-uncensored-GGUF matches part of google/gemma-3-12b-it (626 of its 1065 tensors; the rest of google/gemma-3-12b-it has no counterpart). On the shared part: Andycurrent/gemma-3-12b-it-uncensored-GGUF is a GGUF (F16) of different weights. Looks like abliteration (refusal removal): only the output projections (down_proj, o_proj) changed, in 47 of 48 layers. Abliteration edits exactly these matrices to remove a "refusal direction"; everything else is untouched.

Tensors changed: Embeddings 0/1, Norms 0/289, MLP 47/144, Attention 47/192

On the Hub · Downloads on Model Pulse