lgas

17 ms·

241.
by lgas 3y ago
It's the second one. The individual parts moved away from each other. And there's nothing "not good" about thinking about it.
242.
by lgas 3y ago
They rolled out code interpreter to everyone this week. https://help.openai.com/en/articles/6825453-chatgpt-release-...
243.
by lgas 3y ago
Not to dogpile on all the other "isn't this just" messages, but isn't this just sparse embeddings?
244.
by lgas 3y ago
Aren't "light" and "featureful" opposites?
245.
by lgas 3y ago
Doesn't relativity mean there's no difference between that observer accelerating in one direction and everything else accelerating in the opposite direction? So wouldn't everyone's observation be equally privileged? I&#
246.
by lgas 3y ago
> All we can do is make the tools work really well, and then people can choose how they want to use them or not. That's a great idea, I'd love to see a company try it some day.
247.
by lgas 3y ago
I don't know, it seems like a bit more of a gray area here than with Pikachu. For example, if I said "write me an implementation of fast inverse square root in rust"... and it did... that's certainly not a copyright vio
248.
by lgas 3y ago
I don't understand why everyone keeps saying that no one is going to want to use PostgreSQL for inference... of course not, no one is suggesting that. But they are going to want to store embeddings of their data and be able to query t
249.
by lgas 3y ago
I think perhaps you're just misunderstanding some of the terms you're attempting to use. Those things that everyone else told you were wrong and that no one else saw the value in... your reward function rewards pursuing those. A
250.
by lgas 3y ago
I saw the name Diaphora and thought it would be the perfect name for a reddit replacement... like Diaspora and Fora (multiple forums) combined. This is pretty cool too though.
251.
by lgas 3y ago
It's interesting to me that you say > Generating new code from whole-cloth seems like an easier task for GPT. as I've been working on a similar tool of my own and I've found the opposite. The initial task I've been t
252.
by lgas 3y ago
This is similar to an idea I've been kicking around but haven't had the time to work on. Will definitely give it a try as soon as I get a chance, but FWIW it looks like the requirements.txt is out of data, as the code mentions ll
253.
by lgas 3y ago
How does it rely on the compiler's eta reduction/expansion rules? It doesn't seem to from my (decidedly intermediate) point of view.
254.
by lgas 3y ago
How does that extend to community management for big social networks?
255.
by lgas 3y ago
> I wanted a simple way to "orchestrate" containers on it (from a friendly web interface or something). I've come to the conclusion that such a solution does not exist - it's either CLI (docker/docker compose) or
256.
by lgas 3y ago
Sentence embeddings + retrieving context makes sense. By any chance, have you done any work with indexing code using embeddings? I'd like to do something similar there, but there's no obvious notion of "sentence", espec
257.
by lgas 3y ago
According to the poster you're replying to it's a universal synonym which can be pointed at anything.
258.
by lgas 3y ago
I'm curious about more sophisticated answers to this question, but the obvious approach would be to split the article or web page into sentences and do an embedding per sentence.
259.
by lgas 3y ago
In my experience GPT-4 does much better at handling sentences without punctuation than most people do. I think this is because as a human we might start to interpret something a certain way before we get to the end of the (possibly punctua
260.
by lgas 3y ago
How does it compare to GPT-4 at coding tasks? I haven't tried it but everything I've heard suggests that it is noticeably worse.
261.
by lgas 3y ago
You can either have it show you the statement to confirm before executing it, or if you only want to do queries you can use a read-only database connection.
262.
by lgas 3y ago
Typically you feed the schema in as context and literally just ask the model to turn it into SQL. So a prompt of the format "Given this schema {schema}, turn this query into SQL. {query}. Return the SQL and nothing else."
263.
by lgas 3y ago
That's a bit strange... it doesn't happen for me. I thought it might be a zoom level thing but I've tried a wide range and the text never overflows.
264.
by lgas 3y ago
I went with Swords-stadium-man and it got it with that as well.
265.
by lgas 3y ago
I think there are just a lot more birds in cities than in the country because cities are where all the easy food is.
266.
by lgas 3y ago
I'm in the US and have an iPhone but as I mentioned in another reply, I have friends with android and have never noticed any impediments to texting them... or them me? Is this a high school thing or are you talking about adults?
267.
by lgas 3y ago
Really? I have an iphone but I have friends with android and have never noticed any impediments to texting them... or them me?
268.
by lgas 3y ago
What do you get out of iMessage and its network effect besides blue text instead of green?
269.
by lgas 3y ago
It's hallucinating and/or making mistakes, not lying to you.
270.
by lgas 3y ago
It's a tool. Humans that adapt to any tool they use frequently.

More ›