Recent
-
Hyprland is crazy. And I like it.
-
Here we go... Trying it first on an autoresearch project.
-
Distillation is sampling from perceived valuable spaces within a much broader distribution. It's a point in time optimization, and that broader distribution is not as constrained as you might think.
-
"Function Call & Device Use: Turning on a light does not need a frontier model. Smartwatches, home assistants and robots already expose their abilities as functions with typed parameters, so the only hard part is mapping a messy sentence onto them..." https://cactuscompute.com/n…
-
You're using flex tier at OpenaAI right? Half the cost.
-
The autoresearch pattern is pretty cool.
-
I tihnk it's time to closely read the Chinchilla paper. https://arxiv.org/abs/2203.15556
-
Whether a 70M or 410M pythia lang model, it looks like it takes about 8 reps to memorize a list of single sentence samples. This is pure memorize, no generalizing.