Gaboo/gemma-4-31B-it-qat-q4_0-assistant
Partial fine-tune vs google/gemma-4-31B-it-qat-q4_0-unquantized-assistant. Gaboo/gemma-4-31B-it-qat-q4_0-assistant is a GGUF (BF16) of different weights. Partial change: 0% of the weights changed (norms), in 4 of 4 layers.
Tensors changed: Embeddings 0/1, Norms 4/13, MLP 0/12, Attention 0/8, other 0/2