Chapters
Intro
Intro
0:00
Intro
0:00
Sponsor: Weights & Biases
Sponsor: Weights & Biases
0:40
ChatGPT: How does it work?
ChatGPT: How does it work?
3:20
Reinforcement Learning from Human Feedback
Reinforcement Learning from Human Feedback
5:20
ChatGPT Origins: The GPT-3.5 Series
ChatGPT Origins: The GPT-3.5 Series
7:10
OpenAI's strategy: Iterative Refinement
OpenAI's strategy: Iterative Refinement
8:20
ChatGPT's amazing capabilities
ChatGPT's amazing capabilities
9:10
Internals: What we know so far
Internals: What we know so far
14:10
Building a virtual machine in ChatGPT's imagination (insane)
Building a virtual machine in ChatGPT's imagination (insane)
16:10
Jailbreaks: Circumventing the safety mechanisms
Jailbreaks: Circumventing the safety mechanisms
20:15
How OpenAI sees the future
How OpenAI sees the future
29:25
Sync to video time
Products
Description
15KLikes
437,717Views
2022Dec 7
Chapters
View all
View all
Transcript
Follow along using the transcript.
Show transcript
Yannic Kilcher
290K subscribers
Transcript
NaN / NaN