Hcompany/NeoMME-800M-Retriever-ST-late

Same weights, repacked vs Hcompany/NeoMME-800M-Retriever. Same weights in different files (re-saved, re-sharded or new metadata).

Tensors changed: Embeddings 0/8, other 0/20, MLP 0/40, Attention 0/120

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

On the Hub · Downloads on Model Pulse