Jiayi Pan
@jiayi_pirate
Research | Prev @xAI @Berkeley_AI | Views Are My Own
Joined September 2021
1.6K Following 14.4K Followers
We reproduced DeepSeek R1-Zero in the CountDown game, and it just works
Through RL, the 3B base LM develops self-verification and search abilities all on its own
You can experience the Ahah moment yourself for < $30
Code:
Here's what we learned 🧵
Show more
