prithivMLmods/Deep-Fake-Detector-v2-Model

Full fine-tune vs google/vit-base-patch16-224-in21k. Fine-tune: 100% of the weights changed, across 12 of 12 layers.

Tensors changed: Embeddings 4/4, Attention 96/96, other 48/48, Norms 50/50

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

On the Hub · Downloads on Model Pulse