Validated by the pt_to_tf CLI. Max crossload output difference=3.123e-05; Max converted output difference=3.123e-05.

Hi again πŸ‘‹

My apologies -- our automatic conversion tool was missing the conversion of some model heads, and this was one of the incomplete conversions. We also added much stricter equivalence tests (https://github.com/huggingface/transformers/pull/17588), to ensure TF users enjoy the exact same model experience as PT users.