artkpv comments on Inducing human-like biases in moral reasoning LMs