mahogany

7 ms·

1.
by mahogany 4d ago
Don't you need to run each model several times and take some average since their outputs can change for a fixed prompt? How do you test for that?
2.
by mahogany 8d ago
The thread is not really about what's legal; the topic is integrity. It sounds like, based on the fact that you wouldn't do it yourself, you agree that it's not a good thing to do.
3.
by mahogany 1mo ago
Note that SET memory_limit is a soft limit and can be completely ignored for some tasks, so this wouldn’t be the same as having a limit on physical memory.
4.
by mahogany 1mo ago
Taking Anthropic’s whole AI framing to its obvious logical end: if this is true as written, why was Jarred needed at all in this loop? It seems like an utter waste of time for a highly paid Anthropic employee. Can’t Anthropic have a top lev
5.
by mahogany 1mo ago
The difference now is about the scale. What percentage of the labor force do you think were “computers”? It is not comparable to a potential disruption of huge swathes of the workforce in a short time span. That’s what people are worried ab
6.
by mahogany 1mo ago
> No, I think people will demand to not live in poverty What about other countries in the world where people... live in poverty? > Capitalism allows people to escape poverty by personal effort Isn't it possible that there is a fu
7.
by mahogany 2mo ago
The comment is making fun of a Claude-ism where it becomes super “honest” about stuff. It’s a joke but there is an underlying real effect where this type of language is psychologically manipulative and I would guess makes people believe LLM
8.
by mahogany 2mo ago
You may be right but it seems like a different type of tool when it does the thinking for you. I feel like your analogy is closer to a good autocomplete rather than an agentic LLM loop. Agents feel like a genuinely different technology. May
9.
by mahogany 2mo ago
I was primarily pushing back on anti-screens in its absolute sense, and being somewhat pedantic. However, I largely agree with you, especially considering how screens are most often used today.
10.
by mahogany 2mo ago
If it were a human (going off of memory as it has been a while), they would probably be using mathscinet and their university library to obtain copies of these papers online. Many old papers are digitized and available by these means. I’m s
11.
by mahogany 2mo ago
I am skeptical of using AI for a tutor and would not give it to my kid, but I find the staunch no-screens sentiment equally ridiculous and similar to saying no-books. Do you have kids? Are you telling me that you and the other parent spend
12.
by mahogany 2mo ago
Screens are a medium. I don't see how they are inherently bad, although it is probably correct that screens carry the most bad content out of any information medium. But my kid has learned how crayons and instruments are made by watchi
13.
by mahogany 2mo ago
You would learn way more by asking an LLM how postgres works. Nobody is even reading this Rust code. There are 7000 commits over 2 weeks. What is being learned from that?
14.
by mahogany 2mo ago
> Just run both pgrust and postgres and compare the output. The space of inputs and outputs is infinite. You can't prove programs are the same by "just" testing a bunch inputs.
15.
by mahogany 2mo ago
Of course, if the rewrite is all of those things, then it seems like a good option. The whole point is how do we know if it actually is better.
16.
by mahogany 3mo ago
Genuine question - do you feel that you understand a code change well enough by only viewing diffs, enough to push the Claude code as “your own”? I can see this for some types of PRs but for anything that is relatively complex or touches mu
17.
by mahogany 3mo ago
Every time people point out a limitation or constraint of LLMs, I see a comment that is to the effect of “but humans…”. I don’t understand why this comparison is relevant to this particular thread. Is it just an amusing similarity?
18.
by mahogany 3mo ago
The point is that there is a 0% chance you can meaningfully understand the impact and consequences of what a 10k line change is doing in a couple hour review, so there’s no way for you to know if it’s “bad” in any sense other than it’s so a
19.
by mahogany 3mo ago
The entire PR description is filled with LLMisms. I find this style so hard to read, it’s almost nauseating these days. My eyes start to glaze over and I stop reading pretty quickly after. I’m having to deal with this a lot at work these da
20.
by mahogany 3mo ago
I think the answer to your question depends on what you mean by a “completely egalitarian world”. Depending on the answer, I would ask: why does the kid desire to charge people to earn money in a society like that in the first place? But ye
21.
by mahogany 3mo ago
> As an AI optimist, I think all forced labor should eventually be done by AI. People can then spend their time pursuing their own hobbies. Just as many people still play Go after AlphaGo appeared, because they genuinely love the game. A
22.
by mahogany 3mo ago
The sibling has pointed out some things and you can check that, but — This has AI fingerprints all over it and it was clear to me too. The fact that not only do people not notice this, but also that the comment pointing it out is dead and I
23.
by mahogany 3mo ago
> What all these theories have in common is the underlying belief that our brain/mind works as the machines we build It’s interesting to note that this is not a new phenomenon. If you read writers from the past, they were always com
24.
by mahogany 3mo ago
> The logical conclusion is that the brain makes me The logical conclusion under the popular axiomatic framework of materialism, for sure. But there are other possible logical conclusions depending on your philosophical foundations, e.g.
25.
by mahogany 4mo ago
I would not trust LLMs with the final word on anything financial. Not exactly accounting, but ChatGPT (whatever the paid model was in March) told me that paying down principal early would have virtually no effect on interest over the remain
26.
by mahogany 4mo ago
This doesn't make any sense to me, and seems like the exact opposite of what eBay is, but maybe we use eBay differently? When I use it, I am looking for something specific. The chance a random location nearby has that thing is basicall
27.
by mahogany 6mo ago
Honestly, even if that takes into account housing and everything, doesn't that seem... pathetic? All of the automation and technological advancement and productivity gains over the past 45 years, and average workers in the US see a mea
28.
by mahogany 6mo ago
The web was already not doing so well, but now I fear LLMs will be the final blow for me. This sort of thing is just unreadable. I don't see how people put up with it. Well, maybe not "people". This thread is full of new acco
29.
by mahogany 6mo ago
> annoying meme in feminism and history generally, that people in prior eras were idiots This sounds like a strawman to me but I’m not well versed in feminism. Do you have examples? On the topic of science, isn’t the criticism more that
30.
by mahogany 6mo ago
> If you want to reduce air travel for environmental reasons, then tax it more. > Shaming individuals doesn't seem to be productive or helpful. First, none of us have any power to "tax it more" so this is a dead end of

More ›