OS-Software/gemma-4-12B-it-qat-q4_0-unquantized-heretic-ja-v2

Full fine-tune vs google/gemma-4-12B-it. Fine-tune: 100% of the weights changed, across 48 of 48 layers.

Tensors changed: Embeddings 5/12, Norms 193/289, other 46/48, MLP 144/144, Attention 184/184

2 derivatives checked

• mradermacher/gemma-4-12B-it-qat-q4_0-unquantized-heretic-ja-v2-i1-GGUF: Quantization
• mradermacher/gemma-4-12B-it-qat-q4_0-unquantized-heretic-ja-v2-GGUF: Quantization

On the Hub · Downloads on Model Pulse