Womble
@Womble@lemmy.world
- Comment on Flohmarkt - a Fediverse replacement for Facebook Marketplace 2 days ago:
- Lemmy is no better or worse than Reddit
- Pixelfeed is significantly better than Instagram
- Mastodon is much worse than Twitter
Seems to me pretty much an even spread of how good the names are
- Comment on Nvidia loses $500 bn in value as Chinese AI firm jolts tech shares 1 week ago:
No, that was me running the model on my own machine not using deepseek’s hosted one. What they were doing was justifying blatent politcal censorship by saying anyone could spend millions of dollars themselves to follow their method and make your own model.
- Comment on Nvidia loses $500 bn in value as Chinese AI firm jolts tech shares 1 week ago:
Yes I’m aware, I was saying that the method is the same.
- Comment on Nvidia loses $500 bn in value as Chinese AI firm jolts tech shares 1 week ago:
Maybe they should have been clearer than saying people were joking about it doing something that it actually does if they wanted to make a point.
- Comment on Nvidia loses $500 bn in value as Chinese AI firm jolts tech shares 1 week ago:
In fairness that is also exactly what gpt clauude and the rest do for their online versions too when you hit their limits (usually around sex). IIRC they work by having a second LLM monitor the output and send a cancel signal if they think its gone over the line.
- Comment on Nvidia loses $500 bn in value as Chinese AI firm jolts tech shares 1 week ago:
Oh I hadnt realised uncensored version had started coming out yet, I definitely wil look into it once quantised versions drop.
- Comment on Nvidia loses $500 bn in value as Chinese AI firm jolts tech shares 1 week ago:
Thats the innovation of the “chain of thought” models like OpenAI’s o1 and now this deepseek model, it narrates an internal dialogue first in order to try and create more consistent answers. It isnt perfect but it helps it do things like logical reasoning at the cost of taking a lot longer to get to the answer.
- Comment on Nvidia loses $500 bn in value as Chinese AI firm jolts tech shares 1 week ago:
Oh, by the way, as to your theory of “maybe it just doesnt know about Tienanmen, its not an encyclopedia”…
- Comment on Nvidia loses $500 bn in value as Chinese AI firm jolts tech shares 1 week ago:
Ok sure, as I said before I am grateful that they have done this and open sourced it. But it is still deliberately politically censored, and no “Just train your own bro” is not a reasonable reply to that.
- Comment on Nvidia loses $500 bn in value as Chinese AI firm jolts tech shares 1 week ago:
www.analyticsvidhya.com/blog/2024/…/deepseek-v3/
Huh I guess 6 million USD is not millions eh? The innovation is it’s comparatively cheap to train, compared to the billions OpenAI et al are spending. (and that is with acquiring thousands of H800s not included in the cost)
- Comment on Nvidia loses $500 bn in value as Chinese AI firm jolts tech shares 1 week ago:
Wow ok, you really dont know what you’re talking about huh?
No I dont have thousands of almost top of the line graphics cards to retain an LLM from scratch, nor the millions of dollars to pay for electricity.
I’m sure someone will and I’m glad this has been open sourced, its a great boon. But that’s still no excuse to sweep under the rug blatant censorship of topics the CCP dont want to be talked about.
- Comment on Nvidia loses $500 bn in value as Chinese AI firm jolts tech shares 1 week ago:
??? you dont use training data when running models, that’s what is used in training them.
- Comment on Nvidia loses $500 bn in value as Chinese AI firm jolts tech shares 1 week ago:
It’s not a joke, it wont:
- Comment on ‘Serious concerns’ about DWP’s use of AI to read correspondence from benefit claimants 1 week ago:
There definitely should be more transparency around exactly what is being done, but assumng they are not twisting the truth about it then:
Officials say it is complementary to existing systems, and flags correspondence which is then reviewed by agents to determine whether a correspondent is in fact potentially vulnerable. The DWP said no decision was made by the AI and no data processed by it.
That seems like a perfectly reasonable use case to improve efficiency while keeping people making the decisions.
- Comment on Canadian CEOs Want More Guns, Less Bread 1 week ago:
An alliance the the DPRK if preferable to one with the USA
most rational tankie
- Comment on UK | Musk accused of ‘politicising’ rape of young girls in UK to attack Starmer 4 weeks ago:
What on earth is that spin score nonsense? Its a biased article because it uses the word “billionaire” to describe musk?
- Comment on Uber Eats undercover: Delivering your food for $1.74 an hour 4 weeks ago:
“Let me help subsidise a company paying below minimum wage” totally normal not batshit insane idea.
- Comment on Accused killer of US insurance CEO pleads not guilty to 'terrorist' murder 1 month ago:
Yes that exactly what the professional executive class are, mecenaries for capital (though it cant be mixed up with them also being owners as well as just ridiculously paid frontmen).
- Comment on *Everyone liked that* 1 month ago:
As did the many of the thousands of people who died from being denied treatment in order to push his wage packet up another million dollars. Some of them were children themselves.
- Comment on MPs vote in favour of historic bill to allow assisted dying after emotional debate 2 months ago:
Thats incredibly crass and you should be ashamed of yourself
- Comment on MPs vote in favour of historic bill to allow assisted dying after emotional debate 2 months ago:
The proposed law is only available to people with a terminal illness judged to have 6 months or less to live, needs to be signed off on by two doctors and a judge, and the patient needs to take the drugs themselves. If anything it’s potentially too restrictive, but a step in the right direction.
- Comment on A little physics riddle 2 months ago:
You dont have sufficient information to calculate the momentum of the third body in the new system. If you are treating the system as Newtonian you need the mass of it to calculate its momentum in the new frame, if you are treating it as relativistic you need its total energy.
- Comment on Guardian will no longer post on Elon Musk’s X from its official accounts 2 months ago:
Immediately critising someone/a group for doing the right thing but not doing enough/quickly enough is probably one of the main reasons its so hard to get traction for progressive politics. It’s incredibly tiresome.
- Comment on Coming on Lemmy and complaining because there are too many Linux users is like going in to a brothel and complaining that there are too many hookers 2 months ago:
They propably would if a common topic of discussion was “this is how leggings have got worse today”
- Comment on AI Slop Is Flooding Medium. 3 months ago:
Is this particularly worse than when Medium was filled with artisinally hand crafted slop before AI was a thing?
- Comment on Baidu CEO warns AI is just an inevitable bubble — 99% of AI companies are at risk of failing when the bubble bursts 3 months ago:
AI has been the name of the field for 70 years at this point, it isn’t something Sam Altman came up with as a marketing wheeze.
- Comment on UK experts warn against buying ‘XL bully cats’ 3 months ago:
People are being urged not to buy the feline equivalent of XL bully dogs, which have been created by breeders in the US.
- Comment on All-optical switch device paves way for faster fiber-optic communication 3 months ago:
It is, it confused me too. It is refering to an optical only on/off switch which can also be used as an xor gate. Many levels down froma network switch.
- Comment on Elon Musk’s X is now worth less than a quarter of its $44 billion purchase price 4 months ago:
Lots of people spout this conspiricy theory, but Ive yet to hear a good reason why he had to be sued into naking the purchace (after making price manipulating statements) if it was some sinister plan.
Far more likely hes just a fuck up.
- Comment on ChatGPT is changing the way we write. Here’s how – and why it’s a problem. 4 months ago:
woah woah woah, lets have less of this looking at reality here. We all know generative AI is a fad that never works for anything and anyone using it is an idiot, we don’t need to have our prejudices challenged
