#ML

Some technical intuition on RLHF and Direct Preference Optimisation