Lyapunov_Lover

11 ms·

1.
by Lyapunov_Lover 1y ago
Meta never matched the competition with their Llama models. They've never even come close. And Llama 4 was an actual disaster.
2.
by Lyapunov_Lover 1y ago
> The evidence shows that there is no methodological moat for LLMS. Does it? Then how come Meta hasn't been able to release a SOTA model? It's not for a lack of trying. Or compute. And it's not like DeepSeek had access to
3.
A not-so-quick overview of the Chinese AI scene(statespaceadventures.substack.com)
2 points by Lyapunov_Lover 2y ago|0 comments
4.
by Lyapunov_Lover 3y ago
It's strange how people seem to think that Good Old Fashioned Journalism was objective and reliable. Bias is certainly not a modern invention. This bizarre nostalgia for an era that never actually existed seems to be pretty common, sim
5.
by Lyapunov_Lover 3y ago
Min-maxing the non-fiction equivalent of pilsners, sounds like an excellent way to spend your time.
6.
by Lyapunov_Lover 4y ago
The clear explanation is that neither Google nor Meta have had "ChatGPT" moments—everyone and their grandmothers have tried OpenAIs LLM so it's hardly surprising that people are excited for the follow-up.
7.
by Lyapunov_Lover 4y ago
That's the (hard) problem of qualia. You can't say for sure that ChatGPT doesn't experience anything. And I can't say for sure that you do, either. It's trivial to dismiss it on the basis of "well obviously it&
8.
by Lyapunov_Lover 4y ago
Why would there be a falsifiable hypothesis in it? Do you think that's a criterion for something being a scientific paper or something? If it ain't Popper, it ain't proper? LLMs dramatically lower the bar for generating semi-
9.
by Lyapunov_Lover 4y ago
The author makes a mistake here. It's fine to think of entropy as messiness; that's the Boltzmann picture of statistical mechanics. The mistake is thinking that lowering entropy, or getting rid of the mess, is a satisfactory strat
10.
by Lyapunov_Lover 4y ago
A much more obvious explanation is that it involved nothing more than stream-of-consciousness rambling. You take a random sentence and you convince someone it's deep and profound. They work hard trying to understand it, and they end up
11.
by Lyapunov_Lover 4y ago
This reminds me of a short story by Ken Liu ( https://www.uncannymagazine.com/article/50-things-every-ai-w... ). He trained an AI on his own corpus of stories and made it spit out a list, then he wrote a story around it.
12.
by Lyapunov_Lover 4y ago
They used GPT-3 and DALL·E 2. For the latter, there's a waitlist on OpenAIs website. For GPT-3 they have an API available to developers and there are various apps which use it like ShortlyAI.
13.
by Lyapunov_Lover 4y ago
There's a difference between low-res and high-res; that's what I'm getting at. I think it's interesting to know a little about how it might work. There is something new and interesting about the topic to many people. And
14.
by Lyapunov_Lover 4y ago
I think my use of the term "high-res" already implied that.
15.
by Lyapunov_Lover 4y ago
I disagree that critic reviews are inherently prescriptive. Major blockbusters rely on cliches for the same reason that supermarket chains spend ridiculous amounts of money trying to get every store to look and feel the same: people find fa
16.
by Lyapunov_Lover 4y ago
It doesn't, actually. There are high-res versions of important events. The locus coeruleus-noradrenaline system act in concerto with the amygdala to induce protein synthesis in the hippocampus to produce vivid episodic memories. It get
17.
by Lyapunov_Lover 5y ago
Yeah, it's not a novel idea. Gregory Chaitin has been talking about this for a long time. And a whole bunch of other people. I can't find all that many of them listed in the citations of this work, though. And the Twitter thread o
18.
by Lyapunov_Lover 5y ago
There's a difference between 'after' and 'during'.
19.
by Lyapunov_Lover 5y ago
Yeah, I think people are mostly confused here. Open access is a thing when we're talking about peer-reviewed journals. arXiv hosts preprints, meaning they are available before the peer-review process has proceeded. So calling arXiv pap
20.
by Lyapunov_Lover 5y ago
Honest question: if an anti-vaxxer said that to you, how would you react? I don't feel threatened by McGilchrist and I don't feel threatened by TikTok influencers. I mostly feel annoyed that people generally aren't interested
21.
by Lyapunov_Lover 5y ago
Neuroscientist? He's just a psychiatrist. The video also mentions his "revolutionary theory" and that's just ... wrong. He writes a huge book based mostly on pioneering work by Roger Sperry and Michael Gazzaniga and fill
22.
by Lyapunov_Lover 5y ago
What chaos means in this context is "sensitive dependence on initial conditions" meaning if there's just a tiny amount of noise in your model of the behavior it's going to become useless pretty swiftly. Starling murmurat
23.
by Lyapunov_Lover 5y ago
The ideas of Hofstadter and Hawkins aren't really related. Hawkins postulated a singular predictive algorithm as the explanation of how our neocortices works. Hofstadter has focused on analogies and have had some ideas similar to Hopfi
24.
by Lyapunov_Lover 5y ago
Stories are compressed representations of complex spatiotemporal patterns. We use stories to make sense of the world and to share our insights with others. And if you think about it, stories are essentially containers for if-then relationsh
25.
by Lyapunov_Lover 5y ago
The PC crowd is right. Language encodes our cultural beliefs, and many of them are pretty rotten. But how do you update a culture's shared set of beliefs? Banning words is a symbolic exercise. What we tend to do instead is that we tell
26.
by Lyapunov_Lover 5y ago
I see a lot of people here expressing doubts and confusion. I want to try to clear up some of that. The key notion here is scale relativity. This is the reason why transformer models have been so, well, transformative. Bigger models are bet
27.
by Lyapunov_Lover 5y ago
It's not a secret at all. Transformer models scale. Big models are powerful. Everyone knows this. Google can afford to train very big models. It's not a new technique. I think the issue here is that people are uncomfortable with t
28.
by Lyapunov_Lover 5y ago
We sort of already know that these models scale in such a way that a model with 1000 times the parameters is, indeed, 1000 times more powerful. We haven't found a ceiling effect yet, so the onus is on the skeptics. These things scale.
29.
by Lyapunov_Lover 5y ago
Yeah, I think that's David Eagleman's idea.