Yeah, agreed. It’s true that GPT obeys the objective “minimize the cross-entropy loss between the output and the distribution of continuations in the training data.” But this doesn’t mean it doesn’talso obey objectives like “write coherent text”, to the extent that we can tell a useful story about how the training set induces that behavior.
(It is amusing to me how our thoughts immediately both jumped to our recent hobbyhorses.)
Yeah, agreed. It’s true that GPT obeys the objective “minimize the cross-entropy loss between the output and the distribution of continuations in the training data.” But this doesn’t mean it doesn’t also obey objectives like “write coherent text”, to the extent that we can tell a useful story about how the training set induces that behavior.
(It is amusing to me how our thoughts immediately both jumped to our recent hobbyhorses.)