length penalty in reward function

[paulinebourigault]

The length penalty is calculated using the formula: ((5 + length) / 6) ^ alpha

This is based on the formula used in Google Neural Machine Translation paper [https://arxiv.org/pdf/1609.08144]
When alpha < 0, shorter sequences get higher rewards
When alpha > 0, longer sequences get higher rewards

Length penalties are applied to the final reward scores, after the primary reward calculation

[paulinebourigault]

example use in training script:
reward_model.length_penalty.enabled=True \
reward_model.length_penalty.alpha=-0.15 \
reward_model.length_penalty.min_length=3000 \
reward_model.length_penalty.max_length=4000 \