In machine learning, for the train_test_split method, what is the default split of the train dataset and test dataset? code example

X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.33, random_state=42)