Sublime
An inspiration engine for ideas
First modeling paper out of BigScience is here!
T0 shows zero-shot task generalization on English natural language prompts, outperforming GPT-3 on many tasks, while being 16x smaller!
Model: https://t.co/QvEaqkfmgk
Repo: https://t.co/IpBvBp0JS3
Paper:... See more
GPT-5 is state of the art at reasoning. In SimpleBench, it passed 90% of all questions (9 out of 10). I have access to it via Copilot, and it uses Reasoning... See more
So we’ve got ~6 labs with models at the GPT-4 level, and nobody has meaningfully surpassed it. OpenAI had to scrap/restart their GPT-5 training project bc this barrier has been harder to break than expected.
For the next leap a lot of labs seem to be betting on something new. Lotta smart folks pivoting to reinforcement... See more
Kyle Corbittx.com
People are really bad at understanding just how big LLM's actually are. I think this is partly why they belittle them as 'just' next-word predictors https://t.co/kCFtunt2NO
Last November, ChatGPT's launch shocked the world.
This November, ChatGPT's historic DevDay updates will fundamentally change the way we use technology.
Here's 13 astounding use-cases of ChatGPT's new capabilities: https://t.co/NJMvBUajbs
Sorry to hype —
but having a few friends at OpenAI makes it hard not to hear how wild their open-source model dropping next month is.
It won’t run on a phone, but let’s just say it edges out one of the models in the ChatGPT dropdown.
Meta’s gonna need some time... See more
Yuchen Jinx.com