Kunbyte/ROSE
Full fine-tune vs alibaba-pai/Wan2.1-Fun-1.3B-InP. Fine-tune: 95% of the weights changed, across 30 of 30 layers.
Tensors changed: Attention 539/600, Norms 30/210, MLP 120/120, other 41/43, Embeddings 10/10
Full fine-tune vs alibaba-pai/Wan2.1-Fun-1.3B-InP. Fine-tune: 95% of the weights changed, across 30 of 30 layers.
Tensors changed: Attention 539/600, Norms 30/210, MLP 120/120, other 41/43, Embeddings 10/10