This is a heavily interactive web application, and JavaScript is required. Simple HTML interfaces are possible, but that is not what this is.
Post
Adrian Chan
gravity7.bsky.social
did:plc:4rcqtj5petmffaalnh65jn6x
An unsurprisingly sharp deep dive into sycophancy & RLHF.
Whilst this episode has its tech explanations, the social interaction aspects are unsolved & will rise as models are personalized, as mirroring is an effective design feature, & we're not good at distinguishing affirmation from agreement.
[contains quote post or other embedded content]
2025-05-04T17:16:50.750Z