chunked/nltk_chunking/_troubleshooting/chunk_29.txt

Now the output of the second sequence matches its actual output:

By default, the tokenizer creates an attention_mask for you based on your specific tokenizer's defaults.