🔥The 9th Round of Easy Loan, Earn $40 Reward is in progress❗️ ⏰ Promotion Period: January 15th - Feburary 15th, 2025 👉 Register now and check more details at gate.io/campaigns/358

Whenever you are contemplating participating in Kaggle competitions and you might have heard someone say it is too far-fetched from practical data science work, consider this example: In the recent Science LLM competition participants learned among many other things: - How to

Yep, gradient checkpointing is probably the most underrated memory saving trick that just works. Definitely should be known well by all GPU poor folks.

Brand new CS285: Deep RL lecture on RL for sequence models and language models: youtu.be/egJgDbe5oaM?fe… Figured it was about time to add this lecture to the course🙂

And keeps smashing them lol

Tech drama >> cricket world cup

OK fine I'll tap the sign how.rl.works

Paul Graham

While I love LlamaIndex, but I'm bearish on architecture based RAG in the long term, this feels like brittle pre-ML over engineering that will soon be replaced by one architecture to rule them all.

Sergey Brin is worth $105 billion yet he was a core contributor on the Gemini AI technical paper, coding basically every day. Legend.

Open source lives on

A clean implementation of Mamba in pure PyTorch

Idea: safeLinux. All the same programs you know and love but now upgraded with safety to stop bad actors right in their tracks. $ ls I'm sorry, I cannot list the files in this directory because one or more files may contain unsafe content. Can I help you with anything else? 😅

Next episode: RGB vs BGR, aka fml we used OpenCV somewhere 🤯

Every solar panel is like a giant pixel in a camera sensor, turning photons into a voltage. With billions of them across the world, it is interesting to imagine what they could collectively “see” of the sun and atmosphere, even though they have no focusing machinery. Conversely,

Kagglers doing this since forever because we have no time to tune one thing at a time.

I'm a 100% EV proponent, but a friend pointed out an interesting paradox about this price dynamic. Batteries get cheaper so the resale value of EV gets lower, whereas ICE cars are getting more expensive so they are also holding resale value. Weird market dynamics.

Haha so now prompting LLMs is becoming like a programming language.

not true, especially for language. if you trained a large & deep MLP language model with no self-attention, no matter how much data you'll feed it you'll still be lacking behind a transformer (with much less data). will it get to the same point? i don't think so. your tokens

when you compare your results to the other paper's “under similar conditions”

LLM model size competition is intensifying… backwards! My bet is that we'll see models that "think" very well and reliably that are very very small. There is most likely a setting even of GPT-2 parameters for which most people will consider GPT-2 "smart". The reason current