ALittleLight

10 ms·

61.
by ALittleLight 2y ago
If you take this idea of communication restrictions to the limit, you could imagine the guessers identifying N sets of cards by a single word each as they discuss their guess. The clue giver listens, then uses the clue that identifies the c
62.
by ALittleLight 2y ago
He should be considered innocent by the courts - and he was (innocent of the murder for hire charges, I mean). In the public we aren't obligated to follow the same standards of evidence as the courts. I think he almost certainly did pa
63.
by ALittleLight 2y ago
Doesn't China actively block many American companies from doing business there?
64.
by ALittleLight 2y ago
In the scenario I laid out, no additional real children are being abused either - because the government is simply framing you and you never consumed or produced any CSAM. That's what I meant by it being similar - if your concern with
65.
by ALittleLight 2y ago
You could say the same thing about real CSAM. Authorities claim you have it, present a hard drive they "recovered" from your property to court and you get convicted. At some level we just have to hope, and be skeptical of etc, tha
66.
by ALittleLight 2y ago
Probably put the signs up at the outer limits of the zone and/or using safety gear. Worker in Hazmat suit at the periphery is probably fine.
67.
by ALittleLight 2y ago
You don't have to use it just for winning - consider sports betting, you can definitely place multi-million dollar bets, on the team that isn't a favorite, then use your Havana Ray to disturb a couple key players - make millions,
68.
by ALittleLight 2y ago
Disagree. They were wrong to go after Aaron Swartz and would be wrong to go after OpenAI or Meta. The federal government should instead require any university receiving federal supports, support, or students on federal loans to require anyo
69.
by ALittleLight 2y ago
You really shouldn't say LLMs "never graduate" to experienced staff - rather that they haven't yet. But there are recent and continuing improvements in the ability of the LLMs, and in time, perhaps a small amount of time
70.
by ALittleLight 2y ago
Is it a good response to the react YouTubers to make your content terrible? Or to provide something in your content not available on their? Whether you like it or not LLMs are going to be how people explore the web. They simply work better
71.
by ALittleLight 2y ago
Seems silly to put garbage data there. Like intentionally doing bad SEO so Google doesn't link you. I think you should think about it as: I want the LLM to recognize my site as a high quality resource and direct traffic to me. Imagine
72.
by ALittleLight 2y ago
My approach is a lot of writing out ideas and giving them to ChatGPT. ChatGPT sometimes nods along, sometimes offers bad or meaningless suggestions, sometimes offers good suggestions, sometimes points out (what should have been) obvious er
73.
by ALittleLight 2y ago
It's not saturated. 85% is average human performance, not "best human" performance. There is still room for the model to go up to 100% on this eval.
74.
by ALittleLight 2y ago
Yes, it's pretty similar to Raven's. The reason it is an interesting benchmark is because humans, even very young humans, "get" the test in the sense of understanding what it's asking and being able to do pretty we
75.
by ALittleLight 2y ago
I'd love a "Code" component where you could enter arbitrary code. After following the tutorials I asked myself "What would I like to make?" And I imagined a tweet-bot - grab headlines from Wiki news (or somewhere)
76.
by ALittleLight 2y ago
Earlier I was thinking about camera components for an Arduino project. I asked ChatGPT to give me a table with columns for name, cost, resolution, link - and to fill it in with some good choices for my project. It did! To describe this as
77.
by ALittleLight 2y ago
It seems testable in the sense that you could train a small model, like GPT-2, on the whole dataset, then on the dataset - Shakespeare and compare the loss after training.
78.
by ALittleLight 2y ago
"Stealing data" seems pretty strong. Web scraping is legal. If you put text on the public Internet other people can read it or do statistical processing on it. What do you mean he was "stealing data"? Was he hacking into
79.
by ALittleLight 2y ago
If the company was not meaningfully involved in selling a dangerous addictive substance falsely marketed that resulted in the deaths of tens of thousands of Americans then they should not face any penalty or fines. On the other hand, if the
80.
by ALittleLight 2y ago
I think it's clearly wrong to use generative AI, or any tool, to produce what are basically copies of a work, and then profit off that as if you were the original creator. I think it's just as clearly okay to use content that othe
81.
by ALittleLight 2y ago
Before releasing the pictures authorities would redact the address.
82.
by ALittleLight 2y ago
I doubt this is real. The New York Times mentions that he comes from a prosperous family - seems incongruous with his mother lacking healthcare desperately. https://www.nytimes.com/2024/12/09/nyregion/who
83.
by ALittleLight 2y ago
I'm not sure the little details are enough of a moat. Consider TikTok - people use cheap "special effects" to get the message across, e.g. if a man is playing a woman he might drape a towel over his head - it's silly an
84.
by ALittleLight 2y ago
We have AGI, in it's original definition, now - it's just human level rather than superhuman. ChatGPT, without being specifically programmed to, can summarize a text, write a program, play chess, explain a joke, translate, explain
85.
by ALittleLight 2y ago
But also what new tools will emerge to supplant LLMs as they are supplanting Google? And how good will open source (weights) LLMs be?
86.
by ALittleLight 2y ago
He said it in the livestream that just finished. ~9:36 - https://www.youtube.com/watch?v=rsFHqpN2bCM
87.
by ALittleLight 2y ago
Hey. Sorry about your diagnosis. Thanks for sharing. This is an interesting, though somewhat disappointing, paper. One thing I didn't get while reading it was the specific blood concentration achieved. I noticed they mentioned 12 round
88.
by ALittleLight 2y ago
That's a ridiculous reading of my comment. Vitamin C induces cell death in a variety of cancers. That's what I wrote and it's not controversial, it's documented in dozens of studies. I went on to link an article by medic
89.
by ALittleLight 2y ago
Deeply disappointed by the comments replying to this. I would reply to each individually, but feel like I'd get throttled by hackernews, so I'll just reply to everyone here. --- shepherdjerred writes that this is unlikely because
90.
by ALittleLight 2y ago
I don't see how something like high dose vitamin C IV is very expensive. I would assume a handful of oncologists could do the whole thing themselves. We get X patients a year, we randomly suggest the vitamin C IV to half, the half wi

More ›