RLHF Overview Quiz
5 questions Pass: 70% +25 pts (50 with Pro)
Quiz covering Fine-Tuning and Customization
What you'll be asked
- What is the primary goal of Reinforcement Learning from Human Feedback (RLHF) in the context of Large Language Models?
- In the standard RLHF pipeline, what is the role of the 'Reward Model'?
- Which of the following describes the 'Supervised Fine-Tuning' (SFT) phase that typically precedes RLHF?
- Why is a KL-divergence penalty typically applied during the Reinforcement Learning phase of RLHF?
- In a scenario where a reward model is trained on human pairwise comparisons, what is the mathematical objective during the optimization phase?
Answers and explanations come after you submit.
Course Content