projectj/Instinct-Python-Coder-Gemma4-12B-KimiK3
Full fine-tune vs unsloth/gemma-4-12b-it. projectj/Instinct-Python-Coder-Gemma4-12B-KimiK3 is a GGUF (Q8_0) of different weights. Fine-tune: 100% of the weights changed, across 48 of 48 layers.
Tensors changed: Embeddings 1/1, Norms 0/289, MLP 144/144, Attention 184/184