Know AGI (@knowagi) header
Know AGI (@knowagi) profile picture

Info, Insights, Inspires from AGI world. Also share about open source, startups, and tools.

SF
Joined June 2024
Photo by Know AGI (@knowagi)Photo by Know AGI (@knowagi)Photo by Know AGI (@knowagi)Photo by Know AGI (@knowagi)Photo by Know AGI (@knowagi)Photo by Know AGI (@knowagi)Photo by Know AGI (@knowagi)Photo by Know AGI (@knowagi)Photo by Know AGI (@knowagi)Photo by Know AGI (@knowagi)
Know AGI retweeted
How big does the United States look on Mars? I just upgrade my country true size tool to solar system check the demo 👉
1
2
3
4
3,744
Spent the past 3 weeks at @fdotinc — it’s been truly amazing. What I took away wasn’t some hype-fueled data goal, but something more grounded: continuity, experimentation, and actually building. I learned to slow down, talk more with my users, and understand their needs better. Most importantly, I started to accept that not everything works at the beginning — and that’s okay. What really matters is showing up consistently, building, shipping, learning from real feedback, and continuing the loop.
2
1
3
162
Stop study data science by boring textbooks and courses Here is how AI can boost your study 10x: 1. Directly ask runcell AI to generate live code blocks in notebooks, so you can play and test it instantly 2. Ask AI to turn your notebook into short videos with visualization 👉
3
6
1
46
176,537
Know AGI retweeted
Latest trending data about python data visualization is out. Check which python library is most popular for data visualization 👉
1
3
5
335
Llama3.1-60B-Instruct: Removing 10B parameters (15%) with minimal performance loss and no retraining. Using nyuntam - A New Open-Source Toolkit for Model Compression and Adaptation dropped in Github Applying FLAP (Fluctuation-based Adaptive Structured Pruning) to compress and accelerate the Llama3.1-70b-instruct model. FLAP allows for significant reduction in model size and computational requirements without sacrificing performance. Unlike traditional pruning techniques, FLAP requires no retraining and adapts the pruning ratio across different modules and layers, offering an efficient and effective approach for deploying large language models in resource-constrained environments.
6
51
5
303
25,528
Know AGI retweeted
Replying to @jamespotter
it's actually useful for product managers, just for building prototype. So you wouldn't need to care about cache or effeciency
1
9
Know AGI retweeted
"AI Artifacts" for building #streamlit apps, lab2.dev
1
1
8
415
Know AGI retweeted
lab2.dev now is open to all new users🥳 Turn your idea into @streamlit app with simple text prompts. The AI Artifacts for building Python apps. Here is an example of how lab2.dev build a data exploration app within minutes.
4
14
867
Need an AI agent to code your ideas? Check out lab2.dev Just describe your concept with simple prompts, and lab2 will create the app for you. Here is a demo of how lab2.dev builds a Streamlit app that analyzes Python package download history in real-time!
1
5
1
35
8,429
We’ll be at @HayFestival this year supporting the ‘AI needs you’ session—featuring a conversation between @Cambridge_Uni’s AI & Geopolitics Institute director @verityharding, and AI Ethics & Society associate professor @jonniepenn. Join us at the event: hayfestival.com/p-21995-veri…
70
7
2
85
39,506
Know AGI retweeted
Falcon Heavy rolls out to the launch pad at 39A for @NASA’s GOES-U mission
603
1,583
110
14,023
2,885,669