noxinc/gemma-portuguese-tom-cat-2b-it-Q4_K_M-GGUF-PTBR
Full fine-tune vs google/gemma-2b-it. noxinc/gemma-portuguese-tom-cat-2b-it-Q4_K_M-GGUF-PTBR is a GGUF (Q4_K_M) of different weights. Fine-tune: 100% of the weights changed, across 18 of 18 layers.
Tensors changed: Embeddings 1/1, Norms 0/37, MLP 54/54, Attention 72/72