deucebucket/Gemma-4-E4B-it-Heretic-Cerebellum-GGUF
Quantization (fewer layers) vs coder3101/gemma-4-E4B-it-heretic. deucebucket/Gemma-4-E4B-it-Heretic-Cerebellum-GGUF matches part of coder3101/gemma-4-E4B-it-heretic (677 of its 2130 tensors; the rest of coder3101/gemma-4-E4B-it-heretic has no counterpart). On the shared part: deucebucket/Gemma-4-E4B-it-Heretic-Cerebellum-GGUF is a GGUF quantization (Q3_K_M) of the same weights: 297 tensors match exactly and 380 within quantization error.
Tensors changed: Embeddings 0/2, Norms 0/296, MLP 0/126, other 0/85, Attention 0/168