The attention mask and the pad token id were not set

#8
by hamidpalangi - opened
Microsoft org

Hi,

Thanks for the wonderful work here! I am able to run "Phi-3-medium-128k-instruct" with no issues but when I try "Phi-3-small-128k-instruct" for the same input I receive the warning The attention mask and the pad token id were not set. As a consequence, you may observe unexpected behavior. Please pass your input's attention_mask to obtain reliable results. and then the model generates null in the output. Any idea how to fix this?

Thanks!
-Hamid

Sign up or log in to comment