docling-project/DeskForge-Gemma4-E4B

Partial fine-tune vs google/gemma-4-E4B-it. Partial change: 50% of the weights changed (attention, mlp, other), in 42 of 42 layers.

Tensors changed: MLP 126/606, Norms 0/502, other 85/261, Attention 132/719, Embeddings 0/6

1 derivatives checked

• mradermacher/DeskForge-Gemma4-E4B-GGUF: Same weights, converted (fewer layers)

On the Hub · Downloads on Model Pulse