sparrow
@xtrupal
impulse towards better cognition
i build a neural network that predicts the starting point of a chaotic double pendulum just by watching how it moves. github.com/xtrupal/theta_…
despite having a small understanding of how llms work, its still hard to believe that they dont actually store any language at all.
New video on the details of diffusion models: youtu.be/iv-5mZ_9CPY Produced by @welchlabs, this is the first in a small series of 3b1b this summer. I enjoyed providing editorial feedback throughout the last several months, and couldn't be happier with the result.
hi im sparrow if u are interested in: space philosophy theology ai,ml,rl,dl C, assembly, javascript webdev/ devops distributed systems fusion simulations black holes, quasars, magnetars mythologys comment i will follow you

"i shouldn't learn to code because it's getting automated" no my friend you should learn to code because you can now code 100 times faster. you can create 100 times the value, and make 100 times the money
how simple matrix multiplication like this can evolve into systems that can translate languages, generate photorealistic images, write code, and even act like a goon when instructed?
It's funny how everyone is afraid of this math equation achieving consciousness.