TextClassDataBunch - Removing limit on number of tokens

In lesson-3 IMDB example, TextClassDataBunch, the texts are truncated at 100 tokens for more readability. How can we remove this limitation