You remove a bit of complexity. Sure Postgres is not hard to set up sn to connect to, but Sqlite is just opening a file. It being a file makes it also very easy to test or debug you application.
It is simpler and removes failure points. You don’t need a separate database server process or network/socket connections. Everything happens in-process.
postgres is great and is also a good default choice. It needs a bit more setup than sqlite. Unless I need a capability that postgres provides, I go with sqlite. It just works.
Our prompting is a heck of a lot more complex and includes a lot of nonverbal input. Our reasoning isn't only in language. That makes us quite a bit less predictable. Maybe we're conclusion-reaching machines?
I think your excitement is justified. The paper is claiming a serious bridge between AR quality and parallel decoding, and the lossless LoRA-assisted mode is the wildest part.
no one is an idiot. no one has that identity. people do not have more or less value. we are all worthy of respect.
someone can act with idiot tendencies, but they are NOT an idiot.
I would expect, reminding _hackers_ about the rules is not exactly productive.
Edit: Comment I replied to originally said "Why am I being downvoted" and their original post said something like "Calling someone an idiot is against the rules"
The whole premise is that if you don't get to AGI first then you loose.
The idea is that Anthropic with AGI could build a better version of Apple, or whatever it wants.
This was the conversation like 1 year ago. What has changed?
Nothing changed, it's new ground, we are searching it with a search light. From some vantage points our view on things may feel quite complete, even insightful. Then we look at if differently and feel lost. It's a process we are in together.
By now, various forms of large-scale welfare are over 100 years old, so we have a lot of collective experience in what works and what doesn't as much.
It seems that there is a big difference between attacking extreme poverty in severely underdeveloped places, which was often quite successful and resulted in economic development, and attacking "standard" poverty of the lowest decile of the population in developed countries, which seems a lot less tractable.
My working theory is that in extremely underdeveloped places, most of the population is quite capable of productive work and building a competitive economy, but lacks the know-how, institutions, stability and means to do so. If provided with those, they can kick-start their own future.
While in the developed countries, the permanently impoverished subset are the people who, for all sorts of reasons, cannot do the same regardless of the means invested into them.
It is quite sobering to know that the most generous social system in the world, in France, spends about a third of the national GDP on social spending, and still has some really bad banlieues and very persistent class conflict. If a third of GDP of a developed country cannot do the job, the problem isn't in the raw amount of money spent, but somewhere else.
i think culture can also play a role. there are places in the world (southern italy for example) where people do not want to work, and so live in poverty. same for other places...
reply