Text this: Efficiently quantifying dependence in massive scientific datasets using InterDependence Scores.