aoeusnth1

5 ms·

1.
by aoeusnth1 26d ago
No customer has access to an appreciable fraction of any data center, and if they do they don't have 10x overhead in reserved quota.
2.
by aoeusnth1 1mo ago
Wynes & Nicholas downgraded electricity sourcing to "questionable merit." Table 1 gives a range of under 100 kg to 2,500 kg CO₂e saved per year, depending on the default CO2 intensity of the state/country you are living i
3.
by aoeusnth1 1mo ago
It sounds like you agree with the criticism that your title is a bunch of baloney, but somehow think that's not your fault? You wrote the lies in that title, not an algorithm.
4.
by aoeusnth1 1mo ago
Sounds like you get your news from 2024 when people thought things were plateauing after GPT4?
5.
by aoeusnth1 1mo ago
Soccer is nondeterministic. Are you as good as Messi?
6.
by aoeusnth1 2mo ago
The return on capital is also phenomenal. To be a bear, you have to assume that will crater.
7.
by aoeusnth1 2mo ago
No, because Opus is smarter than those models if they’re all on medium settings. You should compare at similar levels of performance, which would be favorable to Opus.
8.
by aoeusnth1 3mo ago
If it's memorized your benchmark then your benchmark is bad, it's not cheating
9.
by aoeusnth1 3mo ago
What about all his other articles that had f-bombs and the predictive utility of used toilet paper?
10.
by aoeusnth1 3mo ago
It doesn't matter what went wrong- he is not a credible leader. If you are going to lead doctors, you should be a doctor. It doesn't matter why you didn't become a doctor.
11.
by aoeusnth1 4mo ago
"To all my haters who said I mispredicted the capabilities of these systems, rest assured: my statements have no predictive worth at all! I am not saying anything concrete!"
12.
by aoeusnth1 4mo ago
Did I miss where OpenAI plagerized the disproof of the planar unit distance problem from?
13.
by aoeusnth1 4mo ago
We're discussing whether they are models or not, not whether they have goals and agency. A language model does form a model of who you are and what you're thinking, because language is causally connected to those aspects of the ge
14.
by aoeusnth1 4mo ago
When I read your reply, I’m also modeling language. Tokens are just the discretization of the model’s eyes and ears. My brain does a huge amount of work to represent what’s happening in the world based on discrete information received from
15.
by aoeusnth1 4mo ago
The universe contains subsystems which can be described as eluded in the sense that we can take the intentional stance on these systems and describe their observable behavior as being in a state of illusion of separation.
16.
by aoeusnth1 4mo ago
There is nothing which makes either of them are “you.” The feeling of Self is a useful predictor which a physical subsystem uses to nagivate the world and predict observations. “I” is not a physically real label which attaches the “you”-nes
17.
by aoeusnth1 4mo ago
The model is the thing which is learned in order to make the probabilistic prediction with low entropy.
18.
by aoeusnth1 4mo ago
AI agents
19.
by aoeusnth1 4mo ago
I believe they were using their teeth (or lack thereof) to reference their visible meth addiction. See GP, "2) meth is neurodegenerative. heavy users end up with a permanent disability."
20.
by aoeusnth1 4mo ago
The article explicitly and repeatedly affirms that illegal THC vapes are dangerous because of Vitamin E Acetate, which is used as a thickener agent. TFA points out how the NYT article carefully weasels its way around admitting that the THC
21.
by aoeusnth1 5mo ago
Meat is optional.
22.
by aoeusnth1 5mo ago
And idle% is causally connected to whether you make a request or not, surely? I don't understand how your mental model works.
23.
by aoeusnth1 5mo ago
Babies are not conditionally created to solve a problem.
24.
by aoeusnth1 5mo ago
SWE-bench pro is ~20% higher than the previous .1 generation which was released 2 months ago. For their SWE benchmark, the token consumption iso-performance is down 2x from the model they released 2 months ago. If this is a plateau I strugg
25.
by aoeusnth1 5mo ago
Why is a map lookup measured in ms instead of us? Something is seriously wrong with these benchmarks.
26.
by aoeusnth1 6mo ago
> Wobbly assumption that increasing the size of these models yields better performance. I'm assuming you disagree that larger models are better? Can you expand on what indicates that AI will hit a wall in scaling given the evidence
27.
by aoeusnth1 6mo ago
How would you know if it wasn't an extrapolation of current knowledge? Can you point me to somethings humans have done which isn't an extrapolation?
28.
by aoeusnth1 6mo ago
You could reimplement it as public domain on your machine, and then edit it by hand and copyleft your own edits.
29.
by aoeusnth1 6mo ago
Citation needed? Tall tasks are standard practice to improve utilization and reduce hotspots by reducing load variance across tasks.
30.
by aoeusnth1 7mo ago
I don't think he's misleading, I think he is valuing Claude's contributions as essentially having cracked the problem open while the humans cleaned it up into something presentable.

More ›