Keep on to blur preview images; turn off to show them clearly
Is anyone aware of research that increases the ICL capabilities and mimicry capabilities of models?
I mean in post training btw*
![RT @tenderizzation: [ENG SUB] how it feels to use eager pytorch in 2025](/_next/image?url=https%3A%2F%2Fpbs.twimg.com%2Famplify_video_thumb%2F1986664821014863888%2Fimg%2FYjXBwpwwmXP75Zqk.jpg&w=3840&q=75 "RT @tenderizzation: [ENG SUB] how it feels to use eager pytorch in 2025")
RT @tenderizzation: [ENG SUB] how it feels to use eager pytorch in 2025
Distributed Learning @ deepmind | DiLoCo, DiPaCo. Continual Learning PhD @ Sorbonne
This past week I watched Claude decompile PPTX files and recompiles them with edits just from single prompts. The daily input/outputs of the *average* fortune 1000 worker are already in distribution of these models. The product layer is what is missing.
CEO @ Nomic (hiring) https://t.co/NbsLHLWfy8
"The gap between how most people are using AI and what AI is presently capable of is immense." This. 99% of knowledge workers don't know what's coming.
This past week I watched Claude decompile PPTX files and recompiles them with edits just from single prompts. The daily input/outputs of the *average* fortune 1000 worker are already in distribution of these models. The product layer is what is missing.
Delete your drafts
Founder | Author | Speaker Building @beltstripe. Healtech/EdTech/Agric I'm Not The Man Of Your Dreams. Your Imagination Wasn't This Great.
seeing Yarv and co in action has convinced me that classical academic training is highly valuable. Utter midwits can (when they try, or aren’t defending pure ideology) hold their own just because they have notions of “rigor” and “literature”. I feel so thankful to my teachers now
We're in a race. It's not USA vs China but humans and AGIs vs ape power centralization. @deepseek_ai stan #1, 2023–Deep Time «C’est la guerre.» ®1