kabachuha/G4-MeroMero-v2-31B-Heretic-ARA-LoRA-GGUF
Adapter vs kabachuha/G4-MeroMero-v2-31B-Heretic-ARA-LoRA. kabachuha/G4-MeroMero-v2-31B-Heretic-ARA-LoRA is a LoRA adapter, not full weights: rank 64, on down_proj, o_proj in 60 layers (120 adapted matrices). It modifies kabachuha/G4-MeroMero-v2-31B-Heretic-ARA-LoRA-GGUF at load time.