mradermacher/ThinkEdit-deepseek-llama3-8b-GGUF
Abliterated vs cesun/ThinkEdit-deepseek-llama3-8b. mradermacher/ThinkEdit-deepseek-llama3-8b-GGUF is a GGUF (F16) of different weights. Looks like abliteration (refusal removal): only the output projections (o_proj) changed, in 3 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 0/96, Attention 3/128, lm_head 0/1