Pinned
Agus 🔎🔸
@agucova.bsky.social
· Nov 12
Reposted by Agus 🔎🔸
This site does unfortunately disabuse you of the notion that careless thinking is confined to a particular ideology
October 3, 2025 at 9:59 PM
This site does unfortunately disabuse you of the notion that careless thinking is confined to a particular ideology
Reposted by Agus 🔎🔸
anyway, here is 2024 Nobel Prize in Physics winner Geoffrey Hinton discussing what we know about large AI models on 60 Minutes.
October 5, 2025 at 6:51 AM
anyway, here is 2024 Nobel Prize in Physics winner Geoffrey Hinton discussing what we know about large AI models on 60 Minutes.
Reposted by Agus 🔎🔸
things we know about LLMs and large DL models in general:
- how they are trained (gradient descent)
- the structure into which they are placed (architecture)
- the base arithmetic (matmul, norm, batch norm, and so on)
- how they are trained (gradient descent)
- the structure into which they are placed (architecture)
- the base arithmetic (matmul, norm, batch norm, and so on)
as a girl with a PhD in natural language processing and machine learning it's actually offensive to me when you say "we don't know how LLMs work so they might be conscious"
I didn't spend 10 years in mines of academia to be told ignorance is morally equal knowledge.
We know exactly how LLMs work.
I didn't spend 10 years in mines of academia to be told ignorance is morally equal knowledge.
We know exactly how LLMs work.
October 5, 2025 at 1:22 AM
things we know about LLMs and large DL models in general:
- how they are trained (gradient descent)
- the structure into which they are placed (architecture)
- the base arithmetic (matmul, norm, batch norm, and so on)
- how they are trained (gradient descent)
- the structure into which they are placed (architecture)
- the base arithmetic (matmul, norm, batch norm, and so on)
Reposted by Agus 🔎🔸
Another victim of AI psychosis. Really sad 😔
October 4, 2025 at 11:10 AM
Another victim of AI psychosis. Really sad 😔
Reposted by Agus 🔎🔸
one thing that has remained true throughout time is that any assertion or evidence that runs counter to human uniqueness is invariably met with strong (often incoherent/misdirected) anger. Jane Goodall wrote about this wrt. chimpanzees and tool-making.
October 5, 2025 at 6:11 PM
one thing that has remained true throughout time is that any assertion or evidence that runs counter to human uniqueness is invariably met with strong (often incoherent/misdirected) anger. Jane Goodall wrote about this wrt. chimpanzees and tool-making.
a lot of silence from the stochastic parrots crowd
December 21, 2024 at 9:11 PM
a lot of silence from the stochastic parrots crowd
It’s funny, because my Twitter experience made me think that actually, sealioning, engaging with civility while actually being disengenous, is actually a rare phenomenon, and people mostly just overupdated because of a messy combination of biases
part of the growing pains of every community, social media, etc is "what do we do with people who are pretending to be kind, saying the right things, and technically not doing anything wrong right now but are still ontologically evil" and the answer is "stop playing their game and ban them"
to anyone complaining Singal didn’t technically do anything wrong on bluesky yet:
Jesse Singal has a long history of attacking children, pediatricians, & healthcare providers.
Example: He once disappeared from Twitter for a couple weeks after “accidentally” leaking a minor’s medical records.
Jesse Singal has a long history of attacking children, pediatricians, & healthcare providers.
Example: He once disappeared from Twitter for a couple weeks after “accidentally” leaking a minor’s medical records.
December 7, 2024 at 10:01 PM
It’s funny, because my Twitter experience made me think that actually, sealioning, engaging with civility while actually being disengenous, is actually a rare phenomenon, and people mostly just overupdated because of a messy combination of biases
I don't love how the default norms on this site seem to discourage engaging in passionate discussions about important things
December 7, 2024 at 9:53 PM
I don't love how the default norms on this site seem to discourage engaging in passionate discussions about important things
Landsailor was my top 2 song of the year, and Visions by Jose Gonzalez was my third
EA/rationalism at last seeping into my wrapped
EA/rationalism at last seeping into my wrapped
December 5, 2024 at 2:51 AM
Landsailor was my top 2 song of the year, and Visions by Jose Gonzalez was my third
EA/rationalism at last seeping into my wrapped
EA/rationalism at last seeping into my wrapped
I really dislike how bad psychoanalytic accounts of behavior tend to be. They're unspecific, unparsimonious and rarely compared with contrasting hypotheses.
December 3, 2024 at 2:22 PM
I really dislike how bad psychoanalytic accounts of behavior tend to be. They're unspecific, unparsimonious and rarely compared with contrasting hypotheses.
I’m loving this accidental email signature I sent
December 3, 2024 at 4:44 AM
I’m loving this accidental email signature I sent
Reposted by Agus 🔎🔸
On the internet nobody knows you’re a god
December 3, 2024 at 12:11 AM
On the internet nobody knows you’re a god
reviewing a ton of research proposals has the nice side benefit of forcing me to dive into (and understand) a lot of new AI safety research I hadn't heard of
December 2, 2024 at 10:18 PM
reviewing a ton of research proposals has the nice side benefit of forcing me to dive into (and understand) a lot of new AI safety research I hadn't heard of
Reposted by Agus 🔎🔸
Honestly this would work as torture.
xkcd.com/3006/
xkcd.com/3006/
Demons
xkcd.com
December 1, 2024 at 7:21 PM
Honestly this would work as torture.
xkcd.com/3006/
xkcd.com/3006/
Reposted by Agus 🔎🔸
Yet another safety researcher has left OpenAI.
Rosie Campbell says she has been “unsettled by some of the shifts over the last ~year, and the loss of so many people who shaped our culture”.
She says she “can’t see a place” for her to continue her work internally.
Rosie Campbell says she has been “unsettled by some of the shifts over the last ~year, and the loss of so many people who shaped our culture”.
She says she “can’t see a place” for her to continue her work internally.
December 1, 2024 at 12:48 AM
Yet another safety researcher has left OpenAI.
Rosie Campbell says she has been “unsettled by some of the shifts over the last ~year, and the loss of so many people who shaped our culture”.
She says she “can’t see a place” for her to continue her work internally.
Rosie Campbell says she has been “unsettled by some of the shifts over the last ~year, and the loss of so many people who shaped our culture”.
She says she “can’t see a place” for her to continue her work internally.
TIL that self-tanning lotions literally work by frying your skin’s dead surface cells
November 30, 2024 at 8:38 PM
TIL that self-tanning lotions literally work by frying your skin’s dead surface cells
Who knew I would be so fortunate that Vitalik Buterin would not just follow me once, but twice!
November 29, 2024 at 8:47 PM
Who knew I would be so fortunate that Vitalik Buterin would not just follow me once, but twice!
Besides eigenclaude, are any of you using any interesting Claude styles?
November 29, 2024 at 1:50 PM
Besides eigenclaude, are any of you using any interesting Claude styles?
Reposted by Agus 🔎🔸
torturing myself by reading the Marc Andreessen/Joe Rogan podcast transcript
November 29, 2024 at 12:43 PM
torturing myself by reading the Marc Andreessen/Joe Rogan podcast transcript
anyone know any good Vienna-Teng-adjacent artists?
November 29, 2024 at 1:05 PM
anyone know any good Vienna-Teng-adjacent artists?
Reposted by Agus 🔎🔸
November 28, 2024 at 7:26 PM
Reposted by Agus 🔎🔸
i exclusively consent to my tweets being used for training neural networks. if you are not a neural network, stop reading this immediately
November 28, 2024 at 2:59 AM
i exclusively consent to my tweets being used for training neural networks. if you are not a neural network, stop reading this immediately
Are there are any well made conlangs optimized for efficiency and clarity, and not ease of learning (Esperanto) or simplicity (Toki Pona)?
November 28, 2024 at 2:56 PM
Are there are any well made conlangs optimized for efficiency and clarity, and not ease of learning (Esperanto) or simplicity (Toki Pona)?
Reposted by Agus 🔎🔸
is there research supporting the idea that the 'information environment' was a driving factor behind the election results (or whatever people are blaming it for - rightward turn, polarization, etc), and that its effect is *uniquely* large in the last 5-10 years?
November 28, 2024 at 1:46 AM
is there research supporting the idea that the 'information environment' was a driving factor behind the election results (or whatever people are blaming it for - rightward turn, polarization, etc), and that its effect is *uniquely* large in the last 5-10 years?