Add TF weights

#1
by Rocketknight1 HF staff - opened

Model converted by the transformers' pt_to_tf CLI. All converted model outputs and hidden layers were validated against its PyTorch counterpart.

Maximum crossload output difference=1.192e-05; Maximum crossload hidden layer difference=1.323e-05;
Maximum conversion output difference=1.192e-05; Maximum conversion hidden layer difference=1.323e-05;

Hi @pchambon , this is an automated TF conversion of the model weights! We've checked and outputs are equivalent to the PT version up to standard float32 error (~1e-5)

Ready to merge
This branch is ready to get merged automatically.

Sign up or log in to comment