That's the worst villain monologue I've ever heard, but unintentionally funny. Somehow appropriate for someone sharing his name with a JoJo's Bizarre Adventure villain.
If Q4 takes less than 1.3x as many tokens as bf16 or q8, it could still end up being faster, given how decode tends to be bandwidth bound.
The kv cache was still bf16, so a few ops are the same between quants.
The US absolutely should expect for their industry to get hit in a conflict like that.
China knows that the general attitude towards war in the US population is strongly skewed by the fact that they haven't had to deal with the fallout of war on their home turf in recent memory. To the US, war is currently something that happens somewhere else, not at home.
In a conflict, it would just be smart to correct that attitude immediately, and China has the means to do so. Many data centers, major refineries, large industrial targets, and important pieces of the US' supply chain I'd expect to not last the week.
Also China is not going to sit there, taking a beating, and let the US keep the support base that allows them project power overseas (including attitude to war) when it is within China's means to severely cripple the US' ability to do so. China has the home-turf advantage, allowing them to fight the same war for cheaper, guaranteeing victory even if the US retaliates in kind.
Will there be a conflict over Taiwan though? I strongly suspect that if China started blockading Taiwan tomorrow, no other power would do anything about it, because there's no realistic path to win that conflict. It might just get handed over with no shots fired.
There'd be meetings, and negotiating tables, and strongly worded threats, and sanctions, but eventually some agreement would be signed for the transfer of Taiwan.
Planning only/mostly for a limited conflict when both parties are well armed, large industrial powers is pretty short sighted (not to mention both nuclear powers).
Conflicts escalate in unpredictable ways across history because once you start it's really hard to stop sometimes.
The US has grown used to been able to enter a war on its terms and exit on its terms and against the people they've been fighting since Korea that's been entirely possible because they can just leave and the combatants on the other side couldn't really do a lot to the US once they'd gone.
>> Somehow I doubt that after a war with China, satellites would be high on the priority list of things that need rebuilding.
If GPS goes down, most of our communications will go down. A lot of other stuff falls from there. I suspect we'd be OK after a brief adjustment, but it would be very disruptive.
These are bizarre assertions. GNSS is one way to navigate, and GNSS is one way to synchronize time-of-day. Communications networks are full of devices that don't have, or use, any GPS receivers at all. NTP, WiFi, DOCSIS, DSL, leased lines, servers, core routers, switches, desktops and laptops are all things that won't even notice that GPS was unavailable.
What if you were friendly with your neighbor, but on the way to return a shovel you get hit by a bus?
Point being: you're dreaming up edge cases, presumably as a justification for... what? Refusing to engage with people who don't wear their political views like a badge on their chest?
If your stance honestly is "we cannot be friends unless I know your political views", then those who keep their views to themselves in the face of that are wise.
In my experience, yes, the irredeemable bigot is rare. For everyone else, regardless of whether their political views are unknown to you or they are threatening your personal way of life, the world will be a better place for you being good neighbors with them. Not to mention that it will improve your own living situation as well.
Most bigotry doesn't come from an inherent maliciousness or evil. It's often just limited experience and never having had reason to reflect on one's views. If you instinctively go "fuck off bigot", you're not only not helping, you're their mirror image.
To be clear, there's a point at which you have to be intolerant of intolerance, but starting off with that is not only counterproductive by reinforcing echo chambers and increasing division, it also makes you not unlike the thing you're opposed to.
Madame Tracy poked her head around the door. "Coo-ee, Mr. Shadwell," she said, giving Newt a friendly little wave. "A gentleman on the telephone for you. Hallo, Mr. Newton."
"Awa' wi' ye, harlot," said Shadwell, automatically.
"He sounds ever so refined," said Madame Tracy, taking no notice. "And I'll be getting us a nice bit of liver for Sunday."
"I'd sooner sup wi' the De'el, wumman."
"So if you'd let me have the plates back from last week it'd be a help, there's a love," said Madame Tracy, and tottered unsteadily back on three-inch heels to her flat and whatever it was that had been interrupted.
Xenophobia and racial (or religious ) animus are hardly edge cases. On the contrary, they are the default setting for our species, as evidenced by (flips through history book) the last few millennia.
Bigotry is innate and one has to fight it. And that's exactly chmod775's point - stop being a bigot towards other people's reasonable opinions, stop being dramatic, stop making a mountain out of a molehill.
The word "bigotry" actually has a meaning. Trying to equate "unjustified hate against a person for their skin color, ethnicity, or religion" with "distaste for a set of chosen political opinions" is a hoot.
> stop being a bigot towards other people's reasonable opinions
I welcome all reasonable opinions. But not all opinions are reasonable.
In particular, the Maga crowd favors opinions based on hero-worship credulity and irrational vibes, completely divorced from factual reality. Very often, this leads to a place of xenophibia and bigotry, which no one is obligated to tolerate.
Just look at the "conservative" "arguments" in this thread. Just a chain of non-sequiturs and preconceived biases.
If you look at the score distribution, Grok and Opus 5 especially stand out for doing consistently well and rarely ever scoring under 50%. Basically they always at give you something that at least works.
Most others, especially and famously Fable 5.1, seem to have a fair chance of completely failing, despite also sometimes excelling.
You will always have someone say "you should rewrite it in x", but it's rare that something carves out so much ground on the Pareto front you'll get a crowd.
I think they merely want something with a functioning long term memory. Something that can exhibit growth past the first 5 to 10 human-equivalent hours working on something.
Current LLMs are worse than most dementia cases, reaching "peak domain skill" pretty much immediately.
That's the worst villain monologue I've ever heard, but unintentionally funny. Somehow appropriate for someone sharing his name with a JoJo's Bizarre Adventure villain.
reply