https://www.chenk.top/en/tags/chatgpt/Recent content in ChatGPT on Chen Kai BlogHugoenThu, 25 Sep 2025 09:00:00 +0000https://www.chenk.top/en/reinforcement-learning/12-rlhf-and-llm-applications/Thu, 25 Sep 2025 09:00:00 +0000https://www.chenk.top/en/reinforcement-learning/12-rlhf-and-llm-applications/<p>GPT-3 (June 2020) and ChatGPT (November 2022) share most of their weights. The base model could write fluent prose, complete code, and continue any pattern you gave it. Yet, when asked a simple question, it might ramble, refuse for the wrong reasons, hallucinate citations, or produce toxic content. The two and a half years between GPT-3 and ChatGPT weren&rsquo;t spent on larger transformers. Instead, they focused on <strong>how to make the model useful</strong> — a reinforcement-learning problem.</p>