Why Prompting is Hard
Data Skeptic
00:00
The Importance of Generalizability
The model that just says A all the time is going to be 90% accurate. But actually you've got nothing, right? Like your actual model doesn't know anything about any of the data,. It's learned nothing. So if it says A, it's going to do reasonably well. And so like that kind of thing is counter intuitive.
Transcript
Play full episode
