March 2, 2024 at 10:30 AM

Great article on the nuances of Gemini and why many have labeled it (pun intended) as woke

“Today, the bots are different: they appear to know right from wrong and truth from lies. Simplifying a bit, this is a product of a mechanism known as reinforcement learning from human feedback (RLHF). In RLHF, a largely-formed LLM is fine-tuned by presenting it with a variety of prompts, and then browbeating the model to give the answers that get the highest marks from human reviewers.”

https://lcamtuf.substack.com/p/gemini-how-did-we-end-up-here