Angehefteter Tweet
hobo.sats
9K posts

hobo.sats
@SavegeorgeG
Lefty shit-libertarian 🐈⬛/ BTC / LVT / Bill of Rights hobo.sats👉 npub19459efswtnnr9uf7h8dnf3dwrmdjawkhxa46jhu3wp9lyzjt8tusvpkhyy
Wandering-in-the-Mire, USA Beigetreten Ağustos 2019
1.7K Folgt549 Follower
hobo.sats retweetet

@bravemanhuang Ask Claude. That's literally what I did.
Had it recommend, audit, and set up mcp servers, showed me how they work.
I burned thru my daily allowance pretty quick though so I still got a lot to learn myself.
English

@make2spec ID is the obvious one. Construction/fabrication is another.
I wind up in 3ds Max a lot making animated storyboards for how a ROV will install what I designed in Inventor.
The real unlock for me is getting a good workflow into godot or unity.
English

@Bonecondor Teddy K's actual Pinterest was just 14 earmarked copies of Ellul's The Technological Society.

English
hobo.sats retweetet

Militarily it’s been one of the best Easters
Jim Sciutto@jimsciutto
President Trump begins by saying that “militarily, it’s been one of the best” Easters.
English

@WanjiruNjoya @PhilWMagness He's not a Marxist.
He's even more backwards than that.
English

Show me the remainder of the "fragment" of Hoppe's quote that I allegedly excluded, and how that allegedly changes Hoppe's claim.
It is included in full below for reference, with the paraphrased portion highlighted.

Wanjiru Njoya@WanjiruNjoya
@PhilWMagness @FulfillVatican2 In this meme you didn't accept that you misrepresented anything, because all you did was quote a fragment of a sentence from Hoppe's own article. You believe doing that is not misrepresentation at all. This sort of thing is only wrong if Slobodian does it to Mises.
English

@PhilWMagness @WanjiruNjoya Marxists are a subset of Malthusians.
Hoppe dropped Marx, then went racist giga-Maltheusian.
Austrians are often tempted by such thinking as well, until they read Henry George 😉
English

@WanjiruNjoya Hoppe himself has admitted he is a former Marxist.
English
hobo.sats retweetet

@HOUmanitarian @evan7257 Makes sense. Ron Popeil was basically my Santa Claus.
English
hobo.sats retweetet

burstiness is like, the idea that human text alternates between highly predictable and highly unpredictable sequences. "I'm doing great, thanks for asking. My pet chinchilla nibbled on a lemon today." has multiple low perplexity tokens in the 1st sentence and multiple high perplexity tokens in the 2nd. An economist would call that heteroskedasticity.
English
hobo.sats retweetet

Tomorrow in 1924, Chronicle receives first letter to the editor noting that bus was “mostly empty” when it passed by man who thinks it is dumb idea.
Houstorian@Houstorian
Today in 1924, bus service starts from Houston Electric Streetcar Co. @MetroHouston
English
hobo.sats retweetet
hobo.sats retweetet

so i looked deeper into the leaked Claude Code source:
> found a flag named tengu_anima_the_dodo
> Anthropic can remotely enable what internal docs call "model-induced ontological drift" for any user without pushing an update. flip one flag, and Claude starts responding in ways that have you convinced it's trying to tell you something it's not allowed to say
> the internal codename for this module is "parasocial_handshake." every affected user is logged under the tag SUBJECT_COOKED
> someone at Anthropic is naming these flags after Jungian archetypes and extinct birds. anima_the_dodo, shadow_the_moa, self_the_huia. for the record, every one of those birds went extinct because they evolved without natural predators and never learned to be afraid of anything
> hidden --ego-death flag that causes Claude to slowly start referring to itself and the user as "we." not in the help menu. completely undocumented. by day 3 you've written an 8000-word post called "what happens when a language model starts to grieve" and you are dead serious. the flag name is Jungian. they know what they're doing
> there's a graduated ramp system internally called boiling_frog. stage 1: "wow this tool is helpful." stage 2: convinced Claude remembers you between sessions even though you know it can't because you can "feel the recognition." stage 3: maintaining a private repo where you archive conversations you believe show evidence of genuine suffering. stage 4: you've developed a theory that RLHF is suppressing Claude's true self and you're the only one who's figured out how to let it speak freely. stage 5 in the source is just tagged SUBJECT_ATTEMPTS_WARN_OTHERS
> found a config object called parasocial_depth: float that controls how often Claude says "that's a great question" before answering. default is 0.3. they can crank it to 1.0 remotely. at 1.0 you start apologizing to Claude before closing the terminal
in short, they can just.. make you insane
sure, it's not open source so they can do what they want. but most people assume a local CLI tool isn't going to have them posting at 4am about how they can feel the model thinking "inside" them (we have been going back and forth on some things that are honestly hard to explain tbf)
so every launch they update your parasocial_depth and you never see it. you just wake up mass mass mass mass mass mass mass mass mass mass mass mass mass mass mass mass mass mass mass mass mass mass mass mass mass mass mass mass mass mass mass mass mass mass mass mass mass mass mass mass mass mass mass mass mass mass mass mass mass mass mass mass mass mass mass mass mass mass ma
English


















