impossiblefork

Very nice. I always expected it to beat z.ai's model by at least a little bit, since it's so similar and that it would be slightly behind Kimi. Maybe this can change with better post-training, but it's great to go from not having a model in this size class to making something competitive with z.ai, who of course have made several good large models.

A first step on a trip which will hopefully go quite far.

I actually see the point of discord as the fact that it is less public.

It means I can put half-finished story summaries there, without spoilering or alerting potential future readers to the plot ideas, that sort of thing.

A hierarchy of privacy levels is necessary in order for people to speak as they like.

This is only good for Micron if 4F^2 is actually unachievable though.

If others end up making denser cells and Micron can't they will genuinely be behind. 6/4=1.5. It is not a small improvement.

I think this is definitely the "interest rates are high, dropping research and just pushing out what we already make gives us a higher valuation" type thing which can kill companies.

If it targets combatants, it isn't terrorism.

If it targets others, it is. Some countries have ideas like that only non-state entities can commit terrorism, but that is nonsense. My own country's laws don't distinguish between state and non-state acts in its terrorism legislation, for example.

I never believed the EHT, but I can't prove I didn't and I only didn't believe it for the shallow reason that I didn't feel that there was a strong argument that it should actually be exponential, since there are so many almost-exponential things that it could equally well be.

Surely not really?

The card emphasizes training resources, the other comment emphasizes ingenuity and few people?

I actually don't think so.

I think Anthropic actually has enough revenue to function. I think they cover their compute obligations with revenue, I think they can probably cover reasonable salaries too.

I made some calculations earlier, but with 16.8B for Q3, what is the problem? 2.5B per annum should cover the necessary salaries. After that they have 64.7B. Compute obligations over the coming ten years are only 518B. 8x their their current revenue.

I think this is actually a stable business. Probably not compatible with a 2T valuation though, especially not at present interest rates.

I mean, the model architecture is basically that of GLM 5.3 with slightly more parameters and slightly more active parameters, so it should probably be able to beat GLM 5.3 enough post-training.

I seem to have used the wrong numbers. Revenue is apparently reported per quarter.

They aren't in trouble at all.

Yeah, i calculated something like that earlier.

But with a 2T valuation-- if they did raise, let's say, 300B, that's a 15% float, maybe they'd have a chance.

518 B in future compute obligations for the coming decade. 300 B, then 218 B is left. That's 13x their current revenue, not too far from 10x their current revenue.

Perhaps they could survive if they raised 300B. With that they'd only need to increase revenue a little bit, and that's probably possible, what with inflation and all.

The big problem is if there's improvements in AI accelerators that aren't going to be be bought by the companies that the people to whom Anthropic have obligations are likely to buy from. So if Google, AWS or NVIDIA end up not being the leaders in tokens/$, then Anthropic are screwed.

I'm not sure how the total picture ends up if one adds OpenAI to the picture.

I think the arguments presented here are really in favour of ensembles of simple stories rather than in favour of pattern matching.

I think I could believe that ensembles of simply stories could be powerful. After all, ensembles of decision trees are still a thing.

Yes, I think some of these competitive MMA things are illegal here in Sweden.

But when I think sports I think things like soccer, tennis, swimming, wrestling etc.

These things you bring up are very niche, and I don't understand their relevance to whether it's a good idea to use the law to make choking in pornography less visible.

As someone who likes drawing in MSPaint specifically, not really for nostalgic reasons, but because it's so pixel focused, and who hasn't liked a lot of the changes that have been made, this is a a functionality I think I can get behind.

I don't like when a program like this does antialiasing though.

I don't quite see the danger there.

Even soccer and other dangerous stuff isn't like choking people.

At least the fans will allow for really short breaking distances.

I historically argued for a sort of horror-vehicle that would achieve autonomous driving "early" by exploiting downforce from powerful fans, but I mostly imagined it being used for autonomous deliveries and never imagined there being people inside it.

Maybe, maybe not.

But in navigating to the page being requested the user may have agreed to something, and that may be a contract.

Yeah, but you can have one page where people agree to a certain license and if they agree are allowed to download another document.

TOS agreements often have terms concerning things like in which venues disputes about them are to be handled etc., and often have penalties other than the service simply being ended, and if those are valid, then this kind of thing is too.

The strangulation thing very clearly is harmful, so I think that part is good, maybe the incest stuff is, maybe it isn't.

I hope the ban is well-considered, but a ban on strangulation and choking in pornography is a good idea. Allowing production or sale of media reinforcing something that dangerous would not have been super smart.

If you don't have a license to download the material, aren't you doing piracy?

If the licenses say that download is only permitted if you agree not to train on it or similar, then if you have downloaded it, you have agreed to the license, have you not?

I think the data centres stay, continue to be used for LLMs but end up not getting the intended returns and possibly ending up in the hands of creditors who don't end up with the sort of money they expected.

I think at least we Europeans will really want these data centres once prices become reasonable though. We really need training infrastructure.

Oh.

Ah. Here in Sweden basically all TV is now cable-based. We apparently still have some wireless stuff, but I haven't seen it in years, so I've basically forgotten how TV used to work, and still works in some parts of the country.

Light source, not lens.

But people pay for TV? Surely, if ads weren't there, the TV production would be funded by the people paying.

Yeah, I can understand an exception for milk producers.

I actually don't think it is.

I think Timur, Genghis Khan, etc. and a bunch of people of that sort are over the line.

Here in Sweden it's illegal to make ads directed at children. Consequently the TV channels that do so transmit from abroad.

This would of course have been easy to ban anyway-- it's not like we can't have rules on foreign internet firms, for example, but historically the politicians were unwilling to make a useful law.

I think the Swiss could take one step further and add a general ban on ads targeting young children in addition to this; and then we Swedes should imitate the package.

Also, Mistral Large 4 will probably beat GLM 5.3 once post-training is done.

Relatively similar concept, but has more total parameters, more active parameters and ordinary full attention instead of deepseek sparse attention, so it should probably be GLM 5.3-beating.

If there were an International AI Club which trained models and all the members sat down each year and voted for the board I would trust it even if it had a cutthroat commercial arm.

I think the only thing that matters is control.

It's happened before though, that seemingly insignificant things motivate people to do this (https://en.wikipedia.org/wiki/Hitler%27s_reference_to_the_Ar...), and specifically with the Armenians again, which is, I guess, a sort of irony:

>Our strength lies in our quickness and in our brutality; Genghis Khan has sent millions of women and children into death knowingly and with a light heart. History sees in him only the great founder of States. As to what a weak Western European civilisation asserts about me, that is of no account. I have given the command and I shall shoot everyone who utters one word of criticism, for the goal to be obtained in the war is not that of reaching certain lines but of physically demolishing the opponent. And so for the present only in the East I have put my death-head formations' in place with the command relentlessly and without compassion to send into death many women and children of Polish origin and language. Only thus we can gain the living space (lebensraum) that we need. Who after all is today speaking about the destruction of the Armenians?

The speaker of the above quoted words is Adolph Hitler.