Ba2han/Gemma3-Turkish-Augment-FT

Full fine-tune vs unsloth/gemma-3-4b-it. Fine-tune: 90% of the weights changed, across 34 of 34 layers.

Tensors changed: Embeddings 1/4, Norms 91/316, MLP 102/210, Attention 136/352, other 0/1

1 derivatives checked

• mradermacher/Gemma3-Turkish-Augment-FT-GGUF: Same weights, converted (fewer layers)

On the Hub · Downloads on Model Pulse