Just go to Perplexity or something like that, its well trodden ground. And essentially, what you're doing is discretizing the relevant differential equations and getting that to run in a 2D or 3D cellular automaton.

I'll give you a simple example. For diffusion of heat between 2 points, the rate of change (first derivative) is proportional to the difference in temp between them. So you make an update rule for points on a grid that says "calc the average difference of a cell's temp with that of its neighbors, multiply by some constant, and that is the amount to update this cell at this time. Run that for every cell in parallel, many times." Then you tack on a visualization and you can watch the heat diffuse. A fun example would be the cooling of the proto-Earth. You can watch the crust form.

Heat diffusion is a good starter problem. So is gravitational interaction.

The entire model of sea-borne power projection has changed. What does it mean when hypersonic missiles can be fitted with optical sensors to autonomously target ships? It means carriers now are about as useful as battleships in WW2. Nobody knows who can dominate which parts of the world ocean anymore.

Interestingly, I read or heard something that said that a fatal flaw in hypersonic in regards to naval combat was that hypersonic speeds create plasma and that the plasma made optical sensors and remote control useless. They need to slow down in order to hit naval vessels, which makes them vulnerable.

I am in denial.

This sounds a lot like acceptance.

I find for auto turn detection, models work better if you put in the system prompt "if it seems the user hasnt completed their thought yet, output silence". This hack works around their compulsive need to output something.

Great hack, thanks for sharing! Any other hacks like this you’ve found useful to improve voice AI?

I lived with a host family in Lima for 9 months on their roof. They treated me like a son. Extreme tourism.

I would try the Qwen models before LLaVa

Do you need the embeddings to be private? Or just the photos?

For photo indexing I'd run CLIP directly and save on compute, no need to use a whole language model.

The point is to colonize Gaza, they won't irradiate it first.

Netanyahu, when addressing the troops, even said "Do not forget what Amalek has done to you", invoking the memory of the biblical commandment to genocide the Amalekites.

Reposting as it was flagged by Zionists:

You don't have to have sympathy for them. Their religion literally tells them to kill children to steal their land:

"However, in the cities of the nations the Lord your God is giving you as an inheritance, do not leave alive anything that breathes. Completely destroy[a] them—the Hittites, Amorites, Canaanites, Perizzites, Hivites and Jebusites—as the Lord your God has commanded you." - Deuteronomy 20:16-17

And now they are literally doing it.

I wondered how an entire ethnic group could be so depraved and removed from humanity, where literally hundreds of thousands of their members in Hebrew Telegram channels cheer on children being burned alive. And now I know that, it was always their religion that caused their depraved behavior. From a philosophy of racial superiority to literally commands to kill children to steal their land. It was always going to come to this genocidal conclusion.

It's perfectly fine to say Judaism should be canceled, given how Jews are behaving publicly and without shame in their desire to steal land and kill children. They literally don't know that they're not supposed to steal land and kill children. They believe that's completely fine.

Does anyone think this is about anything other than stamping out criticism of Israel?

Of course it’s for more than that. It’s a fabulously oppressive tool that’ll get used for whatever the hell power-hungry folks want. That’s not to say it won’t be used for your proposed purpose! Just gotta think bigger.

That’s the trial balloon.