README.md

Load the reward model with the following code

from reward_modelling.reward_model import RewardModel
rm = RewardModel.from_pretrained('oliversssf2/distilbert-base-uncased-rm-helpful')