Topic
6 facts from 6 episodes, in date order. These are what speakers said on the shows, each linked to the moment it was said. They aren't independently verified.
2019
Microsoft started building its first AI supercomputer in 2019 and deployed it at the end of that year, and GPT-3 was trained on it.
“we started working on it in 2019, and we deployed it, uh, at the end of that year. And it was the computing environment that GPT-3 was trained on.”
2022
Ethan Mollick says GPT-3, the base generation of the technology behind ChatGPT, had existed for over a year before ChatGPT and produced only okay work.
“GPT three, which is, uh, the sort of base generation of this technology has been around for over a year, and it was okay, right?”
2023
Yudkowsky said GPT-4 was probably produced by applying about twenty times more computing power to roughly the same architecture as GPT-3.
“taking what is probably basically the same architecture in GPT three and throwing twenty times as much compute at it, probably, and getting out GPT four.”
Undated
Lex Fridman states that GPT-3 and GPT-3.5 had 175 billion parameters.
“so GPT three, 3.5 had, uh, 175 billion parameters”
Undated
Huang says GPT-3 has 175 billion parameters, each of which could be a floating point number.
“Chatgpt three, uh, is 175 billion parameters, and each one of the parameters could be a floating point number.”
Undated
Eric Boyd says CarMax used GPT-3 to summarize its customer reviews for each car make, model and year and generate a page for each.
“they use GPT-3 to go and summarize reviews that they've got on each make, each model, each year, and then generate a page for it.”
Episodes