#example sentences
sentence1 = 'Hello World!'
sentence2 = 'This is an example!'

#breaking up into tokens
sentence1_tokens = ['[CLS]', 'Hello', 'World', '!']
sentence2_tokens = ['[CLS]', 'This', 'is', 'an', 'example', '!']

#those tokens have IDs
sentence1_token_ids = [101, 1340, 87345, 1332]
sentence2_token_ids = [101, 4589, 988, 874, 13598, 1332]

#we can combine the token ID's together, and add a pad in between them.
#also we dont need the CLS token for sentence two
sequence_token_ids = [101, 1340, 87345, 1332, 102, 4589, 988, 874, 13598, 1332]

#constructing sentence tokens for positional encoding
#note how they correspond to the sequence token ids
sent_location_tokens=[0,   0,    0,     0,    1,   1,    1,   1,   1,     1   ]

#pad, if necessary, with 0's for the tokenids and 1's for the sent locations