qolop
Joined 5 years ago
Joined 5 years ago
Predicting the next word is a much deeper problem than people like you realise. To be able to be good at predicting the next word you need to have an internal model of the reality that produced that next word.
GPT-4 might be trained at predicting the next word, but in that process it learns a very deep representation of our world. That explains how it has an intuition for colours despite never having seen colours. It explains why it knows how physical objects in the real world interact.
Now, if you disagree with this hypothesis it's very easy to disprove it by presenting a problem to GPT4 that is very easy for humans to solve but not for GPT4. Like the Yann Lecun gear problem, which GPT4 is also able to solve.
I'm slowly switching to helix, and this time around will be forcing myself to use the defaults, and not adding a single line of config.
I've observed the effects on both Firefox and Chrome and it's like day and night.
On chrome you can even test it using Devtools > Application> backward forward cache