lossolo

46 ms·

751.
by lossolo 3y ago
I love that quote too! :) Enjoy!
752.
by lossolo 3y ago
> Ok, but why aren't there a thousand "made in a weekend" clones that at the very least function but have no users? Because you are using social network to connect/interact with other people? If I'm making a simp
753.
by lossolo 3y ago
There's a difference between social networks, where a network effect is needed, and why we don't see clones of platforms like Instagram or Reddit. Many of today's 'AI startups' are essentially just a landing page wi
754.
by lossolo 3y ago
> The field of ML research is moving so fast that people don’t even take time anymore to explain the design decisions behind their architectures Often, they don't fully understand why certain methods work, they simply experiment and
755.
by lossolo 3y ago
> The dilemma parents are grappling with is this: tablets and smartphones, while beneficial for children's learning and socializing Recent reports from teachers indicate that many children are intellectually behind their peers. A co
756.
by lossolo 3y ago
Correlated: "The obesity pay gap is worse than previously thought" https://www.economist.com/finance-and-economics/2023/11/23/t...
757.
The obesity pay gap is worse than previously thought(economist.com)
2 points by lossolo 3y ago|1 comments
758.
by lossolo 3y ago
> To me that seems like a great foundation for AGI. It's only one part, predicting text is relatively straightforward because it doesn't require predicting complex sequences like 'a S23mz s.zawsds'. Based on statistic
759.
by lossolo 3y ago
Open source version of something similar: https://github.com/PKU-YuanGroup/Video-LLaVA
760.
by lossolo 3y ago
> Maybe 10T? 100T? I don't think we have enough training data to train models so big in a way to efficiently use all the params. We would need to generate training data, but then I don't know how effective it would be.
761.
by lossolo 3y ago
If that would be the case then LLMs would know how to do math but they don't.
762.
by lossolo 3y ago
We are already doing it, as all LLMs are fully deterministic. So we are adding random seed to prompts and control temperature.
763.
by lossolo 3y ago
You can predict performance of certain tasks before training and it's continuous: https://twitter.com/mobav0/status/1653048872795791360
764.
by lossolo 3y ago
From a strictly evolutionary standpoint, our culture or civilization(however you want to frame it) hasn't adapted to the modern world because other people will pass on their genes. There is no fertility crisis in many places around the
765.
by lossolo 3y ago
> If I were the CEO of OpenAI, I'd be pretty pissed if a member of my own board was shitting on the organization she was a member of while puffing up a competitor. Considering what's in the charter, it seems like she didn'
766.
by lossolo 3y ago
Correct, I don't know why this is the top comment when it's so wrong.
767.
by lossolo 3y ago
> It is just a matter of training data, but it is very difficult to have someone collect these large amounts of data and train on it. It's really not that difficult, they are trained mostly on audiobooks and high quality audio from
768.
by lossolo 3y ago
> But the truth is that the US doesn't have an history of populist demagogues as rich as 3rd world countries. There is Trump and nothing as him before. The US had a good economy for many decades, while 3rd world countries did not. T
769.
by lossolo 3y ago
How would you compare mistral-7b-instruct 16fp (or similar 7b/13b model like llama2 etc) to Yi-34b quantized?
770.
by lossolo 3y ago
Who knows, money is not the only thing that motivates us. Power, influence, fame, ego etc are the thing that drives people too and can blind them.
771.
by lossolo 3y ago
Well, if it's so smart then maybe it will learn to count finally someday. https://chat.openai.com/share/986f55d2-8a46-4b16-974f-840cb0...
772.
by lossolo 3y ago
Given how often I see supercars in Monaco, it seems they must be more in active use.
773.
by lossolo 3y ago
Why can't we restrict TikTok to the same content only in the west? It's just a matter of regulation, we know it's possible because it's how it's done in China. Now think about it, is it really China here to blame fo
774.
by lossolo 3y ago
There is a difference between Chinese society and western society. Chinese do not allow kids to play video games for too long because they know effects of games on the youth, they also restrict what can be viewed by children on Chinese vers
775.
by lossolo 3y ago
So which is it? Reading the last few HN threads about quantum computers gave me the impression that we don't even know if there will be any practical use for them, as they seem super specialized, and there are technical challenges that
776.
by lossolo 3y ago
It's not about a diamond ring but what is behind that preference. Could be different things but some of them are a huge red flags. It's a lot better to marry someone that do not have that preference.
777.
by lossolo 3y ago
Gambling engages the brain's reward system, releasing neurotransmitters like dopamine, creating a pleasurable experience. This reinforcement mechanism, coupled with the unpredictable nature of wins, contributes to the development of ad
778.
by lossolo 3y ago
Gambling engages the brain's reward system, releasing neurotransmitters like dopamine, creating a pleasurable experience. This reinforcement mechanism, coupled with the unpredictable nature of wins, contributes to the development of ad
779.
by lossolo 3y ago
> Germany's median wealth per adult is a mere $66,000. They should be asking some serious questions. Isn't that because most of that "wealth" is from inflated prices of houses? Over 50% of germans rent so over the med
780.
by lossolo 3y ago
It's also worth noting that if you build your business on using OpenAI's LLM or Anthropic etc, then, in the majority of cases I've seen so far (no fine tuning etc), your competitor is just one prompt away from replicating you

More ›