qsera

26 ms·

661.
by qsera 6mo ago
I don't have a good answer. But I have a good reason to say LLMs are not intelligent. https://news.ycombinator.com/item?id=47918103
662.
by qsera 6mo ago
With AI, everyone will be average in no time! Internet started it, hopefully LLMs will finish it.
663.
by qsera 6mo ago
That is like saying we can get unlimited data compression by feeding the output of a data compressing program into its own input..
664.
by qsera 6mo ago
So your problem with the definition is that "I invented it"? Do you have any rational objection to the definition? If you don't have, then I am afraid that you don't have a point.
665.
by qsera 6mo ago
Just like clergymen used to omit "we believe x, y, or z" when making some mystical claims, "scientists" of today omit "as far as we know from their communication with the layman... >Who would let a junior dev tri
666.
by qsera 6mo ago
>LLMs are highly intelligen I will tell you why it is not. Intelligence is understanding low level stuff and using it to reason about and understand high level stuff. When LLMs demonstrate "highly intelligent" behavior, like so
667.
by qsera 6mo ago
>small layer on top of very thick caveman/animal insticts and programming. Which is also why marketing and advertising works on EVERYONE. When AI puts out the phrase "Prompt engineering", everyone instinctively treat it as
668.
by qsera 6mo ago
>Using information that's known to be false or weakly supported But where does such information originate from? Is laypeople just making it up?
669.
by qsera 6mo ago
If he is an idiot, the "bullshit" that he says would be inconsistent. The problem with Trump is perception. That what Trump says can appear to be inconsistent if you are looking for it. Or if you want to project that impression.
670.
by qsera 6mo ago
> it's part of my quest to get Diablo Sorry, I didn't quite get what you refer to as "it" here. Nevertheless, I also have a similar quest going.
671.
by qsera 6mo ago
>I'd have to hear how you think this type of strategy benefits America The only answer I have is the one that Trump keep on repeating, about how Iran should not have a nuclear weapon. > JCPOA agreement If I am the POTUS and is r
672.
by qsera 6mo ago
Look, I choose to be honest and see the response I am getting. Being called a troll. Is it really surprising that I can relate?
673.
by qsera 6mo ago
> who are deliberately destroying science... But this is subjective. What you call as "Science" might be pseudoscience for someone else. As an example, some decade back, following and trusting peer reviewed research was "s
674.
by qsera 6mo ago
>for it to make sense Mmmm..I didn't make that argument. I was talking about keeping the behavior of the people good and aligned with the goals of the society. Who cares if it was a lie, if it provides happiness and peace? The probl
675.
by qsera 6mo ago
Yea, it is like looking at the insides of a mechanical clock vs looking into an electronic one. The former is mesmerizing, intriguing, inviting and inspiring. The later? you can't wait to put the lid back on...
676.
by qsera 6mo ago
Because computers can be used to do things that are not...so serious?
677.
by qsera 6mo ago
Awesome! This makes me really happy.
678.
by qsera 6mo ago
Wait a min. Does this paper says we don't know how back-propogation works?
679.
by qsera 6mo ago
So basically that is what I was saying. Imagine an LLM that can also OCR. Would it be possible to make it OCR a totally new letter by only showing a single picture of it and including the fact in the context? I think it would not be possibl
680.
by qsera 6mo ago
> right and rational Even you seem to agree that there is a notion of a "right" thing. A "Rational" action can totally depend on what you want to achieve. And also considering the fact that "rationality" is
681.
by qsera 6mo ago
>non-scientific worldview, brainwashing This can be good, you know. I mean that was the original purpose of religion. The idea is that everyone will be good if they are afraid of judgement day. But science came along and took that away.
682.
by qsera 6mo ago
>just extending the context length and adding more instructions in the context will not get you continual learning... I agree. But I am wondering if context would help in answering superficial questions and only fail when answering quest
683.
by qsera 6mo ago
>knowing how to give AI good context. It is really ironic that humans are now trying to find patterns in the behavior of a program that works by finding patterns in human behavior.
684.
by qsera 6mo ago
LLM does not "reason". This should be clear by the fact that it can solve complex math problems without understanding how to count.
685.
by qsera 6mo ago
That was exactly what I was thinking. So it is a bit unclear why such a possibility should be even considered. To be fair, I didn't really understand what idea this article is trying to get across..
686.
by qsera 6mo ago
>for the sake of argument, that context can express everything weights can... Does this imply that a completely untrained model (random weights) should show intelligent behavior only by providing enough context?
687.
by qsera 6mo ago
>natural world unravel Natural world would be mostly fine one way or other, human beings might not survive though...
688.
by qsera 6mo ago
I think we can talk about intelligence without introducing consciousness. So there are two questions 1. Is LLMs really showing intelligent behavior. 2. Is that real intelligence. I think even to the first question, the answer is no. Because
689.
by qsera 6mo ago
Ads? Where we are going, we won't need Ads. People seem to be missing the fact that businesses won't need ads anymore. It would be like pharmas gifting doctors and practitioners to prescribe their products. Those are not Ads. With
690.
by qsera 6mo ago
>The absorb command will do a lot of this for you by identifying which downstream mutable commit each line or hunk of your current commit belong in and automatically squashing them down for you. This feels like magic every time I use it

More ›