What would you say are your best resources for solid signal? I ask because it looks like Iβm going to be starting a podcast about open source, self-hosted AI tools.
Discussion
Brian Roemmele seems to have interesting ideas about all of this
So this is just a small list that I curated for you right now, but there are a ton more youβll fall deeper into the research rabbit hole with these individuals on this list theyβll lead you to the other great researchers bc they interact amongst each other
All of the usernames Iβm posting are twitter usernames
@nearcyan
@mckaywrigley
@emollick
@Suhail
@levie
@DrJimFan
@ylecun
@JagersbergKnut
@Shahules786
@JohnNosta
@BrianRoemmele
@DimitrisPapail
@HongyiWang10
Check this out itβs called StarCoder and its a 15B LLM for code with 8k context and trained only on permissive data in 80+ programming languages. It can be prompted to reach 40% pass@1 on HumanEval and act as a Tech Assistant. Open sourced ((:
- Intro blog: https://huggingface.co/blog/starcoder
- Model link: https://huggingface.co/bigcode/starcoderbase
- Paper (May the source be with you!!)
https://drive.google.com/file/d/1cN-b9GnWtHzQRoE7M7gAEyivY0kl4BYs/view