Hey everybody, David Shapiro here for another video. This video is going to be about fine-tuning GPT-3 in order to fulfill my core objective functions.
Why listen
It goes beyond the title with direct discussion of functions, prosperity, suffering, including: This video is going to be about fine-tuning GPT-3 in order to fulfill my core objective functions.
Key takeaways
01They are meant to build experience, and the AGI is supposed to learn as it goes
02So next, the way that I've started is that I use synthetic data sets to seed the core objective function models And that what I going to give you a demonstration of here in just a
03And this is by simple virtue of machine learning models will believe whatever data you give them
Best for
listeners looking for a practical AI episode debrief