Hacker Newsnew | past | comments | ask | show | jobs | submit | littlestymaar's commentslogin

It doesn't need to, during inference there's little data exchange between one chip and another (just a single embedding vector per token).

It's completely different during training because of the backward pass and weight update, which put a lot of strain on the inter-chip communication, but during inference even x4 PCIe4.0 is enough to connect GPUs together and not lose speed.


How do you think that works?!

With the exception of diffusion language models that don't work this way, but are very niche, language models are autoregressive, which means you indeed need to process token in order.

And that's why model speed is such a big deal, you can't just throw more hardware at the problem because the problem is latency, not compute.


Kvark was leading the engineering effort for wgpu while he was at Mozilla.

But he was doing that on his work time and did so collaborating with other Mozilla engineers, whereas AFAIK blade has been more of a personal side project.


And made almost zero impact, it was just a bigger version of Deepseek V2 and when mostly unnoticed because its performances weren't particularly notable especially for its size.

It was R1 with its RL-training that made the news and crashed the srock market.


And the author completely misses the point thinking it's somehow mandatory in plaster walls, when it's just a convenience thing that avoids making holes in the plaster…

I do appreciate why people want to avoid that, plaster does crumble pretty easily. Combined with 100+ year old lath that is as hard as iron, it can be a mild pain in the ass to hang a picture without doing more damage to the plaster than you want.

> . I don't think there are many people who posted their Balatro playthroughs in text form online

There are *tons* of balatro content on YouTube though, and it makes absolutely zero doubt that Google is using YouTube content to train their model.


Yeah, or just the steam text guides would be a huge advantage.

I really doubt it's playing completely blind


Thanks to another comment here I went looking for the strategy guides that are injected. To save everyone else the trouble, here [0]. Look at (e.g.) default/STRATEGY.md.jinja. Also adding a permalink [1] for future readers' sake.

[0]: https://github.com/coder/balatrollm/tree/main/src/balatrollm...

[1]: https://github.com/coder/balatrollm/blob/a245a0c2b960b91262c...


Yeah we need someone to make an secret, air gapped strategy game for benchmarking purposes

> and you’ll realize that not only are “AI takeover” fears justified

Its quite the opposite actually, the “AI takeover risk” is manufactured bullshit to make people disregard the actual risks of the technology. That's why Dario Amodei keeps talking about it all the time, it's a red herring to distract people from the real social damage his product is doing right now.

As long as he gets the media (and regulators) obsessed by hypothetical future risks, they don't spend too much time criticizing and regulating his actual business.


Or “design cities to be less car centric so we can keep the antisocial drivers out of a car”. It's definitely a system issue.

Why jump to cities? Low speed traffic in cities is not the main location of car collisions.

The main issue is highways and rural roads.


    The main issue is highways and rural roads.
Source? I'd wager most accidents happen where the most people are. Cities.

> I'd wager most accidents happen where the most people are. Cities.

Sure, in absolute numbers. But..

In the US, highest deaths per 100M vehicle miles:

1.79 - Mississippi

1.73 - Arizona

1.72 - South Carolina

And the lowest:

0.56 - Massachusetts

0.70 - Minnesota

0.78 - New Jersey

Or, highest per 100,000 population:

25 - Mississippi

25 - Wyoming

21 - New Mexico

Lowest per 100,000:

4.9 - Massachusetts

5.7 - New York

6.5 - New Jersey

Maybe the average MS driver drives 3-5x as many miles as the average MA driver? I doubt it. Something else happening there.

https://www.iihs.org/research-areas/fatality-statistics/deta...


Fortunately deaths are only a fraction of the accidents though, and it's not even necessarily the kind of accident that bothers insurance companies the most as long as the driver only kills himself.

It seems we are talking about different things.

The majority of car collisions happen on urban roads where the majority of cars are.

The majority of fatal collisions occur on rural roads, where vehicles are travelling faster.

I live in the UK.

https://www.simplyquote.co.uk/insights/where-do-most-car-acc...


“Highways” and “city” aren't incompatible, quite the opposite: most of the traffic on highways is commuting, in urban areas.

.... in the USA.

I live in the UK where highways majority used for inter city transport.


I can't tell about the UK, but it's definitely true as well in France and Germany.

And a quick look at map of London and Birmingham suggests that it may very well be the case in the UK as well…


Maybe actually looking at research and statistics might serve you better than casually glancing at maps and taking wild guesses.

I'll get you started:

https://www.gov.uk/government/statistics/national-travel-sur...


Funny how you don't even bothered to get the data by yourself get still felt entitled to act like a jerk.

> Credit scores are universally hated but they make it possible to offer lower interest rates to more people.

That's probably true in theory, but not in practice, given how high US credit interest rates are compared to European countries for instance.

> Without credit scores, fewer people would have access to credit.

Too many people having access to credit is exactly how we got the worst financial crisis of the century, so it's not really something to brag about… People talk about US public debt a lot, but private debt is even more worrisome.


> This to me sounds a lot like the SpaceX conversation

The problem is that it is absolutely indiscernible from the Theranos conversation as well…

If Anthropic stopped making lies about the current capability of their models (like “it compiles the Linux kernel” here, but it's far from the first time they do that), maybe neutral people would give them the benefit of the doubt.

For one grifter that happen to succeed at delivering his grandiose promises (Elon), how many grifters will fail?


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: