nativ-community/embeddinggemma-2-bf16

Same weights, repacked vs google/embeddinggemma-2. Same weights in different files (re-saved, re-sharded or new metadata).

Tensors changed: MLP 0/360, Norms 0/376, other 0/195, Attention 0/425, Embeddings 0/6

2 other repos hold byte-for-byte identical weight files.

On the Hub · Downloads on Model Pulse