You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
self.kv_seq_len is accumulated when generating samples in the loops. Only the first sample uses the pyramidKV while the remaining ones don't. We need to set the self.kv_seq_len to zero in each new iteration.