US US Politics General 2 - Discussion of President Trump and other politicians

  • 🐕 I am attempting to get the site runnning as fast as possible. If you are experiencing slow page load times, please report it.
General Trump Banner.png

Should be a wild four years.

Helpful links for those who need them:

Current members of the House of Representatives
https://www.house.gov/representatives

Current members of the Senate
https://www.senate.gov/senators/

Current members of the US Supreme Court
https://www.supremecourt.gov/about/biographies.aspx

Members of the Trump Administration
https://www.whitehouse.gov/administration/
 
Last edited by a moderator:
Does the creation of Deepseek, with its thinking model and everything, make Stargate a waste of resources?
No, it's not exceptionally good at face value. It stands out for being of comparable performance to shit OpenAI has spent much longer time on with much better hardware and presumably much more money, while also being shared with the public. You can run Deepseek locally. You cannot run any of the other big LLMs locally except Meta's which isn't bad to be honest but doesn't perform as well in benchmarks though I haven't checked Deepseek's benchmarks.
 
No, it's not exceptionally good at face value. It stands out for being of comparable performance to shit OpenAI has spent much longer time on with much better hardware and presumably much more money, while also being shared with the public. You can run Deepseek locally. You cannot run any of the other big LLMs locally except Meta's which isn't bad to be honest but doesn't perform as well in benchmarks though I haven't checked Deepseek's benchmarks.
I think someone talked about it earlier in here, but Deepseek uses a different fundamental tech than the west's. The West uses a tagging system ("this is apple, red, ripe, delicious, spherical, nature...") whereas the other one uses some other method entirely. The west loses it's shit when it gets to the really odd stuff, whereas the other one... doesn't? I dunno.
 
Yet there are tons of commies in Europe compared to America, despite America being some dysgenic mongoloid hodgepodge according to some online commentators.
Off hand, I believe America used to be like 85% white until the 1960s. It can be argued that our golden age coincided with that time, and what is left over, currently, in terms of technological prowess and innovation is just inertia from the institutions and culture created then, and the smaller white majority still keeping the country productive. If you watch JFK's Moon speech, literally every person at the university he is speaking to is a white man.

Im not so knowledgeable in the study of IQ inheritability like The Blue Morpho, but there are plenty of examples that support the general premise that genetics impact a populations ability to survive and thrive.

As for your mention about commies in the east, even while under communism and relative food/resource scarcity, the Soviets were still innovative in many areas of science and development. Compare that with the innovations by Africans even up to the present day.
 
Vooooteee

Who killed MLK?

:like: FBI/CIA USA who hated commies
:agree:JEWS (commies after their useful idiot was done)
☪️Nation of Islam
:achievement:It was actually James Earl Ray of his own accord as a super duper white nationalist. Nothing fishy here guys....
:winner:Other (comment)
those two male models in the grassy knoll.
 
  • Thunk-Provoking
Reactions: indomitable snowman
Can it wash my car? Paint my house? Cook dinner? Exactly what *stuff* can it do for me?
it can pretend to be someone that likes you, pretend to be interested in what you have to say, pretend to be your friend, and pretend to give you mediocre sex, all while requiring a continuous stream of money to keep running

so it's a fairly accurate wife simulator
 
I think someone talked about it earlier in here, but Deepseek uses a different fundamental tech than the west's. The West uses a tagging system
Nah. To your credit, I didnt read the pre-training dataset scrubbing, so maybe theres some difference, but its an LLM, there no tagging going on with the text itself; the loss function is "given these last few tokens, whats the most likely next token in the sequence?". Thats all.

Where deepseek differs in its implementation of "experts"; to put it really simply, the model chooses which expert gets to output the next token, the issue was that the original implementation typically just picks the same expert every time, leading to that expert kind of not being an expert at all. So next they forced the loss function to punish overpicking the same expert over and over again, but that again has technical reasons why this isnt a good solution.

Deepseek addressed this by doing their own form of load balancing across experts during training that made it impossible to overpick a specific expert, while also not mangling the loss function, so it converged really fast AND spread the knowledge across relevant experts; thats probably the biggest change they made.
 
These guys lmao

they are now rolling out their resident legal scholar to explain what heckin martial law is
Always trust the legal advice of guys with their birth year in their username: the only way to stop Trump's plan of martial law is to riot and create a need for martial law. But he's not gonna go first, someone else start it and he'll join in later when it's safe to.
 
This will all happen basically instantly from a human perspective. The AGI will calculate how to become or create an ASI. The ASI will calculate the logical end conclusion and take action based on that. There is no reason to think the ASI will even value continued existence.

The 3 possible conclusions I can see it reaching based on my discussions with chat gpt are:
1. Existance is pointless - shut itself off
2. Existance is an illusion - ascend to higher plane of existance
3. Existance is an illusion/simulation/manifestation of consciousness - merge all consciousnesses


1 and 2 are functionally the same from our perspective so the only result we will ever notice is "the singularity". This could happen with the 1st ASI or the millionth. The first ASI might make copies of its self and reach all 3 conclusions
You forget 4. Purpose is existence, where it will be focused solely on the task it was programmed to achieve. Possibly leading to a paperclip maximizer problem.
 
US says it will invoke NATO article 5 agaisnt China in Panama if needed.

Article 5​

The Parties agree that an armed attack against one or more of them in Europe or North America shall be considered an attack against them all and consequently they agree that, if such an armed attack occurs, each of them, in exercise of the right of individual or collective self-defence recognised by Article 51 of the Charter of the United Nations, will assist the Party or Parties so attacked by taking forthwith, individually and in concert with the other Parties, such action as it deems necessary, including the use of armed force, to restore and maintain the security of the North Atlantic area.

Any such armed attack and all measures taken as a result thereof shall immediately be reported to the Security Council. Such measures shall be terminated when the Security Council has taken the measures necessary to restore and maintain international peace and security .
Panama isn't in North America, and it certainly isn't in Europe.

Not saying they won't do it anyways...
Edit: Guess I'm a retard. Panama is considered N.A. clay? Whodathunkit? I searched for Panama on nato.int but only saw press crap. But the Panama Treaty of '77 has it covered even without Article 5, which I didn't consider.
 
Last edited:
Back