Testing the Ability of Language Models to Interpret Figurative Language