The use of computational tools for the study of literature can facilitate new perspectives and avenues for critical work. Reflecting on the recent emergence of (and increasing hype around) large languages models such as GPT, this essay argues that the creation of "smart" data sets and corpora as new forms of literary objects requires and enables the development of computational methods and tools that can create "data stories". Smart data sets continue a humanistic tradition of textual scholarship and bibliography while also preparing text data to be "read" by machines. In telling "data stories" about Herman Melville, we bridge the gap from "numbers to meaning" in a variety of examples from Billy Budd. The essay closes with a broader reflection on reading Melville in the age of machine learning and artificial intelligence.