facebook/sapiens2-seg-1b
Full fine-tune vs facebook/sapiens2-pretrain-1b. Fine-tune: 100% of the weights changed, across 40 of 40 layers.
Tensors changed: Attention 360/360, Norms 161/161, MLP 160/160, Embeddings 3/3
Full fine-tune vs facebook/sapiens2-pretrain-1b. Fine-tune: 100% of the weights changed, across 40 of 40 layers.
Tensors changed: Attention 360/360, Norms 161/161, MLP 160/160, Embeddings 3/3