Fix SDPA dropout during eval

[massimilianoviola]

F.scaled_dot_product_attention applies dropout whenever dropout_p > 0, even when the module is in eval mode.

This does not affect the default pretrained V-JEPA 2.1 inference because dropout defaults to 0.0, but if the attention modules are configured with nonzero proj_drop, eval can produce different outputs.

This PR sets the dropout_p argument to zero when self.training is false, matching the usual behavior.

[meta-cla]

Hi @massimilianoviola!

Thank you for your pull request and welcome to our community.

Action Required

In order to merge any pull request (code, docs, etc.), we require contributors to sign our Contributor License Agreement, and we don't seem to have one on file for you.

Process

In order for us to review and merge your suggested changes, please sign at https://code.facebook.com/cla. If you are contributing on behalf of someone else (eg your employer), the individual CLA may not be sufficient and your employer may need to sign the corporate CLA.

Once the CLA is signed, our tooling will perform checks and validations. Afterwards, the pull request will be tagged with CLA signed. The tagging process may take up to 1 hour after signing. Please give it that time before contacting us about it.

If you have received this in error or have any questions, please contact us at cla@meta.com. Thanks!