llmfan46/G4-MeroMero-31B-uncensored-heretic
Abliterated vs zerofata/G4-MeroMero-31B. Looks like abliteration (refusal removal): only the output projections (o_proj) changed, in 19 of 60 layers. Abliteration edits exactly these matrices to remove a "refusal direction"; everything else is untouched.
Tensors changed: Embeddings 0/4, Norms 0/523, other 0/62, MLP 0/261, Attention 19/338
5 derivatives checked
- llmfan46/G4-MeroMero-31B-uncensored-heretic-GGUF: Same weights, converted (fewer layers)
- mradermacher/G4-MeroMero-31B-uncensored-heretic-i1-GGUF: Quantization (fewer layers)
- mradermacher/G4-MeroMero-31B-uncensored-heretic-GGUF: Quantization (fewer layers)
- llmfan46/G4-MeroMero-31B-uncensored-heretic-NVFP4: Quantization
- llmfan46/G4-MeroMero-31B-uncensored-heretic-NVFP4-GGUF: Same weights, converted