Æthos-Sundai Hack [Remote Research on Qwen2.5 fine-tuning]

​Join us for a Sundai hack!

​Every Sundai, we host weekly hacks for building rapid AI prototypes. 

​We welcome coders of all trades: computer scientists, bioinformaticians, physicists or engineers, from any discipline to come up with, code and finish an AI prototype by the end of the day.

​And, most importantly - Have fun and code!

​This Sundai is a little different, as we will adapt OG Sundai style of hosting Research Hacks!

​We’ll dive into the latest work in fine-tuning and Reinforcement learning of Qwen2.5. By following the GitHub of am interesting new paper discussing how Qwen is fine tuned, and how in the world it can learn on various conditions.

​Focus : https://github.com/ruixin31/Rethink_RLV

​

​​Let’s hack!

​​Our vision at Sundai Research:

​​"Pioneering AI research through hacking."

​​Deeply embedded in Harvard and MIT ecosystems, Sundai Research connects top AI researchers, creating opportunities for collaborative breakthroughs & future publications.

​​Learn more: sundai.foundation

​​Our goals each Sundai: (1) explore cutting-edge AI research problems (2) design and conduct experiments for measurable insights (3) catalyze future AI publications and innovations

​​Our mission (1) grow exceptional AI researchers (2) advance AI science (2) spread the researcher mindset

​

​Schedule:

​15:00 Intro to Hacks

​18:00 Discussion : Start

​20:00 Discussion : Finish

​

​Our values:

​•⁠  ⁠Open mind, open source, open to new ideas.

​•⁠  ⁠Balance between serving our community and making big strides for the future.

​•⁠  ⁠Shortcuts is an ingenious move that a hacker makes to build a system that works.

​•⁠  ⁠Have fun and code!

​

​

​We are co-parented by Æthos, an ethical AI startup accelerator based between Cambridge and Berlin CIC, and Sundai.club, a Cambridge based AI hacker club with builders from MIT and Harvard.

​Bring some snacks, good music, and your laptop!