Not necessarily infinite. It stops when there's reasonable accuracy. Similar to how we would implement this in software.
705 karma · joined August 3, 2021
I agree with the first sentence but not with the second one. Consciousness most probably does not arise from just a next-token predictor. At least not from an architecture similar to current LLMs.
Both humans and LLMs basically learn to predict what happens next. However, LLMs only predict when we ask them. In contrast, humans predict something all the time. Even when we don't have any sensory input, our brain plays scenarios. Maybe consciousness arises because the result of our thinking is fed back as input. In that sense, we simulate a world that includes us acting and communicating in that world.
Also noteworthy, the human brain handles a variety of sensory information and it's output is not only language. LLMs are restricted to only language. But to me it seems like it's enough for consciousness if we can give it the self-referential property.
I think a writer should include all the references that lead to their thoughts expressed in the texts. This way the reader could follow these links to explore more adjacent topics. To me, it seems like the ecosystem of small website could improve on that. I'd rather explore other writers based on some topic than on a blogroll.
In scientific writing this has a long tradition called related work. In some fields it degraded a bit into a section that merely lists all the works that should be named rather than a deep discussion of their contents.
Obviously, this is also the original idea of Berners-Lee for the web.
A good example that comes to my mind is this: https://www.astralcodexten.com/p/somewhat-contra-marcus-on-a...
Is the any evidence that barter economies existed in large scale? I know this is a complicated question. But I think the author is not quite right. Most probably, people did not exchange pigs for wheat and then wheat for leather as its just too complicated. I think it's more likely that there was a common good that acted as a means of exchange (salt, cigarettes, gold, etc.).
> I should be able to pay you $150 (or whatever price we both agree on) without anyone taking a chunk off that transaction, or even knowing it happened. Note that your father already paid his taxes for the transaction when he purchased the Gameboy, 30 years ago.
I accept your opinion here. But I it's not more than an opinion and other people and governments might think different about this. Legal systems and hence tax systems are socially constructed and so there's no inherent truth against which you can argue.
> Gold is a finite resource because we know that there is a limit to the amount of gold on the planet, therefore it’s valuable. Fiat currency, on the other hand, holds value just because a governing entity is endorsing it.
Gold and money require both a working government that endorses basic rights to own and exchange goods.
There's not need to bring forward all these myths just to let people know that you like privacy and are against data mining and use of AI in this case.
Apple does a lot of analytics maybe they see that there's really a group of people who just use the default apps with some light browsing. But I think it's rather unlikely.
I guess the largest health effect comes from exercise, stable/low blood sugar, no alcohol, enough sleep.
In my first comment, I didn't claim that it's simulating the entire human brain but aspects of it like language and memory.
We can test such aspects by throwing text at the model and comparing it to responses of humans. A simulation doesn't need to have the same abstractions. What counts is solely the quality of the response. If the responses are sufficiently close to those of humans, I'd say it's simulating a human brain.
Humans seem to be part of superorganisms like ants. Basically, we need some form of social group to survive. Those who escape the superorganism tend to take resources with them (physical like tools or intellectual like skills).
But even if the former wouldn't be true, it is becoming increasingly hard. Humans have become the dominating species of the planet. In proportion to land, oxygen in the atmosphere, and other resources we're so large that the tragedy of the commons happens in all corners.
In my view that's at least part of the reason why we're getting so much regulation. A side effect of the regulation is that it limits the freedom for other ways of life than the default of superorganism. Or simply put, I cannot just move to the forest because it's owned by someone. I cannot fish because these animals are endangered and thus protected by law etc.
Then, an absolute number of profit doesn't reveal much. Maybe this seemingly huge profit is tiny compared to the effort to make it (i.e., the profit margin).
As much as I'd like to see a world where we value more aspects of a business than only profits, I can not change it alone. There's a brutal competition of companies to earn a certain profit margin. And Maersk is more or less forced to play this game until we change the rules.
LLMs are trained on human texts. And if they're trained well these models might start to simulate parts of a human brain. Might be that this is the simplest way to produce texts that satisfy the objective function.
If this would be the case, and we give LLMs some emotional shading, maybe these would simulate parts of a human brain with this emotion and hence produce better answers (given that in the training data, emotional samples have better quality).
EDIT: added "parts of" before human brain.
When I was doing my PhD, my supervisor always told me that scientists need to move around. It's somewhat of a tradition in academia even if few people understand the reason behind it fully (e.g., my supervisor did not I suspect).
Academia is a really old institution and has evolved mechanisms to ensure its future. This is one of them I guess.
I experienced the people there as a close-knit community. So, moving might counter this effect. The other reason that comes to my mind is signalling. If you're willing to move with your family, it's a strong signal that you take academia seriously. I guess every institution likes to have such people.
I guess that's what a framework is about. You get some abstractions that make your life easier and you pay by buying into the concepts.
> [...] some friends I've talked to where still on Next.js 12 and really felt the pressure to upgrade to not fall behind even more.
If the current version is doing its job, why update? It's a SSG not a browser that connects to the internet.
So I disagree. I'll not write my own and keep the existing one as long it's not too much hassle.
It's another thing that Apple confuses people with disabling Bluetooth. I always need to go to settings -> Bluetooth to really turn it off instead of using this pull-down menu. I assume many non-technical people don't understand this difference. So it seems like a violation of UX principles that only makes users feel to be in control.
So I think a diversified portfolio would still be valid. Even if the stock markets would be closed for years, I could sleep well as some of my companies will go bankrupt and others will have much higher profits.
Wikipedia lists nine possible explanations for stotting. The author, however, simply picks the one that fits his story.
There are also multiple explanations for sexual evolution.
I can hold the belief "Defund the Police" as a poor or as a rich. So it's easy to fake this signal. Also, I don't see any evidence that this belief is held mainly by rich people.
Now, I'm using it every day and haven't changed it in years. It compiles everywhere even on my M1 with Asahi Linux.
All my other software is not beautiful. It requires changes.
I _quickly_ checked the study [1] and didn't see anything obvious.
Given that our voices reveal so much about the state of a human (think of mood, emotions), I think it's at least possible that voices could reveal something about metabolism. Just we cannot hear it because hearing diabetes is not as useful as hearing something about the mood of another person.
[1] https://www.mcpdigitalhealth.org/article/S2949-7612(23)00073...
Imagine, you're a complicated machine implemented in Game of Life. The game is another not so complicated machine in Game of Life and you start exchanging messages. As soon as you understand how the game works, you can simulate the internal state in your mind. This way, you can reach any score you like. So the result becomes invalid.
Watched a quick ride towards quaternions. Also nice.
I fully agree. Yet animal shelters are crammed after the pandemic. Adopting an animal can improve their live.
In my experience dogs are less stressed without a leash. The primary reason for using a leash is that I feel compassionate about people who are afraid of dogs or don't like them. Seeing a leashed dog makes them feel more secure.
Basic comment was that the person confuses and misunderstands many things. According to my friend, the range marked as "missed beat" contains a premature heart beat [1].
Some more comments:
- Premature hear beats happen in us quite frequently
- CT scans are a rather invasive diagnostic method and such be avoided
So unless some more competent doctor comments otherwise, I take this as a nice story that resonates well in our layperson's minds. Not more.
Given my experience with complex technology (software), I think it's rather likely that a generation ship will reconfigure while traveling.
Maybe we will build a ship that's just enough to bring them to a place with lots of resources. Then, it will be rebuild with those resources for the final journey. Or even new ships might be build out of existing ones to adapt for varying conditions.
Similar to biological life on earth.