ash80/RLHF_in_notebooks

RLHF (Supervised fine-tuning, reward model, and PPO) step-by-step in 3 Jupyter notebooks

72 best points · 1 comments · 1 Hacker News post · 255 GitHub stars · View on GitHub

Hacker News discussions

• Reinforcement Learning from Human Feedback (RLHF) in Notebooks (72 points, 1 comments, by ash_at_hny, 2025-07-06)