3 ms·
Isn’t that exactly what we’re currently doing? The problem is that doing this few billion times for every token seems to be harder than just powering some actua
by echoangle 2y ago
Isn’t that exactly what we’re currently doing? The problem is that doing this few billion times for every token seems to be harder than just powering some actual neurons with sugar.
- tim333 2y agoI think the algorithm is pretty different though I'm not expert on the stuff. I don't think the brain processes look like matrix multiplication.
- echoangle 2y agoThe algorithm (of a neural network) is simulating connections between nodes with specific weights and an activation function. This idea was derived from the way neurons are thought to work.