Ba2han/gemma-e2b-TR-OCR

Partial fine-tune vs unsloth/gemma-4-E2B-it. Partial change: 47% of the weights changed (attention, embeddings, mlp), in 35 of 35 layers.

Tensors changed: MLP 153/585, Norms 0/453, other 0/240, Attention 164/727, Embeddings 1/6

1 derivatives checked

• mradermacher/gemma-e2b-TR-OCR-GGUF: Same weights, converted (fewer layers)

On the Hub · Downloads on Model Pulse