I am playing with open weights models at home and yeah they are like that ... I use Claude 3.7 @ work and yeah it is a lot better ... Sometimes it will flub things but it also can write large amounts of code ... Mostly…
As a professional rust dev, I will say this, I don't notice. Generally because I am doing partial builds mostly. And when I am not the m1 max I have is fast enough to compile the project. Really there are much larger…
But also unfathomably large because of the speed at which you could consume it was measured in kbps....
Project Binky (YouTube) ep 39 or 40 did this with their CNC mill
Respectfully, I disagree. An llm in my mind is a new compiler. Just it takes natural language and produces code.
So the issue with genetic algorithms / genetic programming is you need a good way to handle the path the population takes. It is more reinforcement than y = f(x) for deep learning f() is what the nn is computing. X and…
Yes, one is committing knowledge to neurons, the other is commuting knowledge to short term memory. Put another way, if you took a llm with random weights. Do you expect you could rely on context alone?
Lol thanks for the correction! like I said... it had been 20 years.. I misremembered the amounts :P
Are you talking about teaching in the context window or fine tuning? If it is the context window, then you are limited to the size of said window and everything is lost on the next run. Learning is memory, what you are…
IIRC ( and it was 20 years ago now that I learnt this) the brain uses 20% of the body's resting energy usage. Most of that is keeping neurons polarised to the outside (ion pumps need ATP!!!). The body uses 25w resting…
I am playing with open weights models at home and yeah they are like that ... I use Claude 3.7 @ work and yeah it is a lot better ... Sometimes it will flub things but it also can write large amounts of code ... Mostly…
As a professional rust dev, I will say this, I don't notice. Generally because I am doing partial builds mostly. And when I am not the m1 max I have is fast enough to compile the project. Really there are much larger…
But also unfathomably large because of the speed at which you could consume it was measured in kbps....
Project Binky (YouTube) ep 39 or 40 did this with their CNC mill
Respectfully, I disagree. An llm in my mind is a new compiler. Just it takes natural language and produces code.
So the issue with genetic algorithms / genetic programming is you need a good way to handle the path the population takes. It is more reinforcement than y = f(x) for deep learning f() is what the nn is computing. X and…
Yes, one is committing knowledge to neurons, the other is commuting knowledge to short term memory. Put another way, if you took a llm with random weights. Do you expect you could rely on context alone?
Lol thanks for the correction! like I said... it had been 20 years.. I misremembered the amounts :P
Are you talking about teaching in the context window or fine tuning? If it is the context window, then you are limited to the size of said window and everything is lost on the next run. Learning is memory, what you are…
IIRC ( and it was 20 years ago now that I learnt this) the brain uses 20% of the body's resting energy usage. Most of that is keeping neurons polarised to the outside (ion pumps need ATP!!!). The body uses 25w resting…