erax-ai/EraX-Translator-V1.0

Partial fine-tune vs google/gemma-3-4b-it. Partial change: 52% of the weights changed (attention, embeddings, mlp, norms), in 34 of 34 layers.

Tensors changed: Embeddings 1/4, Norms 83/316, MLP 51/210, Attention 70/352, other 0/1

3 derivatives checked