Aaron_Scher comments on Finding Deception in Language Models