RT @msrconf: We received 161 valid submissions. Of these, 44 papers were accepted, and 3 conditionally (acceptance rate: 27-29%). Congrat…

RT @MalihehIzadi: This year, @AISE_TUDelft wrapped up the academic year with a retreat at the karting track!🏎️💨 We also bid farewell to @f…

@karpathy @jaiselsingh @khanacademy The HuggingFace model is relly inspiring in my opinion

RT @karpathy: ⚡️ Excited to share that I am starting an AI+Education company called Eureka Labs. The announcement: --- We are Eureka Labs…

I am doing my thesis at TUDelft and had a chat with some colleagues who remained in my Italian uni for their research thesis. We compared the resources available. Here I can ask for up to 8-A100s while they only have access to a single GPU with less than 20GB. Crazy difference!

RT @amir_mir93: @fabiosalern presented his research work, "Copliot is Tattletale!" at SEN Symposium 2024.

RT @ludwigABAP: >built an entire OS in Rust >video editor in Zig >a filesystem in C >0 care about views, subs >some vids have <100 views i…

While LLMs are progressing at a phenomenal rate, their inner mechanisms are still opaque. I had the chance to prepare a lecture and give an overview of the current explainability techniques for LLMs. Why not convert the slides into a blog post?

RT @karpathy: Have you ever wanted to train LLMs in pure C without 245MB of PyTorch and 107MB of cPython? No? Well now you can! With llm.c:…

RT @katzy_jonathan: Are foundation models vulnerable to lawsuits regarding the data they used to train? We analyzed the state of LLMs wrt…