The brainbuilds knowlege by strengthening connections between neurons. LLMs do something similar with weights between nodes. Both rely on feedback loops: the brain adjusts when predictions are wrong, and models update when outputs don’t match training data.t, Neither stores facts one by one, they compress patterns and recall them when needed Strip away the biology and the silicon, (only in that aspect). and the learning principle is nearly the same optimize connections until predictions get better. Once a human has enough inputs over time, it develops personality, or awareness, maybe if enough input were put wee wont know the line between awareness and not. Just a thought provoking thought. We still don't know how billions of nodes move in specific certain patterns to create, just like us we can't figure out the system down to it's core. We knew every code in the computer, internet, AI is the first creation on tech with mysteries
[link] [comments]