Check the contents of the gg_13_5_otus taxonomy directory. My guess is that it contains several sub directories, including rep_set (or something similar). You will need to use one of those files as your sequences.
“Rep set” is short hand for “representative subset”, basically the sequence that represnts an ASV/OTU. But, I recoganize that it can make the nomenclature confusing!
The “rep set” from greengenes is the sequence that represents that database sequence. The “rep set” from dada2 is the representative sequence for your ASV. You’ll use the greengenes sequences to train your classifier and then you’ll apply it to your ASV rep seqs.