JUNJIE99/MMRet-large

Full fine-tune vs openai/clip-vit-large-patch14. Fine-tune: 91% of the weights changed, across 24 of 24 layers.

Tensors changed: other 2/3, Embeddings 4/5, Norms 110/150, MLP 144/144, Attention 288/288

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

On the Hub · Downloads on Model Pulse