Jerry Tworek reveals why architecture experiments died at OpenAI: "I've been at @OpenAI seven years. Maybe we had, depends how you count, three, four tries at that time for new architecture. What you needed to do first, a researcher had to run some small scale experiment. Those were difficult to code. If those worked, then you had to experiment for at least three months." "Then you had to try to scale those up. Those are extremely hard. You had to probably get 10 people buy in into this…