Hi @yizhiwang96, is there any particular reason for pdb.set_trace() at the def attention function?
It traces if if mask is not None. I am experiencing entering the Pdb mode during the process. I've tried to debug it, but without a significant understanding of why it occurs.
Could you please recall your intention?
Hi @yizhiwang96, is there any particular reason for
pdb.set_trace()at thedef attentionfunction? It traces ifif mask is not None. I am experiencing entering the Pdb mode during the process. I've tried to debug it, but without a significant understanding of why it occurs. Could you please recall your intention?