My company ran a test and found that they reduce token output on flagship model by something like 2-4x, and that number has been increasing with newer models. I suspect the increased subagent usage is driving this…
Because "depth" is recursive. You can search twice without waiting for the results of your first search: iteration. You can't if the thing you need to search for is the results of your first search: recursion.
That sounds more iterative than recursive. Recursion requires feeding the output back into the input, so creating version 4 requires results from version 3. You cannot recur in parallel. Iteration does not. You can…
A published formalization is code. I would not think humans have any edge when it comes to citing previously published results.
It is, but it was also a rhetorical device! I wanted to highlight that we're actually looking at a trio of concepts: intelligence, IQ, and value. Strongly correlated concepts, yes, but also meaningfully distinct!
What does IQ have to do with intelligence? (Forgive the flippant response. I believe it cuts to the core of what the parent was intending.)
https://docs.google.com/presentation/d/1XMCU2vxtxxqqQ2NY567n... You can download it in various forms as well, such as txt. Won't fix the tone, but at least it's scannable.
Embeded slides URL is a great hack from which you can just download as .txt directly.
No, you've been charged for every one of those licenses.
> Also, the statement was valid because it will be true for most. The statement seems more like a preconception than data. Source?
That is the appeal. It's an opinionated distro designed around keyboard navigation.
Those self-interested unions better stay away from my interests!
My distinction is "systems" vs "scripting" but yeah. I'm bringing Y2K back next!
Agreed, efficiency is still important, but being at the "frontier of efficiency" is significantly more relevant to commodity model providers than state-of-the-art model providers. Frontier labs are incentivized to route…
> Given that OpenAI is ahead in intelligence, it's also reasonably likely that they are at the frontier of efficiency too. Frontier labs have no incentive to be at the frontier of efficiency. Claude still leads the pack…
Scripting languages are awful. Python is one of the nicest scripting languages.
Happens with humans too! My senior colleagues check in with me significantly less often and cost significantly more in the meantime!
I feel like these are two competing goals: * the dev wants to describe an app in natural language then fall asleep while an AI works on it * the dev wishes that the same AI would write less comprehensive tests What…
The session is "just" the raw chat history in it's entirety (human and agent) and can be disseminated as such. This is what enables swapping between models, you simply send the whole context. Not sure how others do it,…
> [...] there is one unquantified, unproven comment at the top saying it's 100% [...] The parent was clearly not stating anything confidently. > Look at the later comments, they have substance oriented discussions.…
I don't follow. The default today is that an unprivileged user space application should not crash the machine. The kernel reserves what it needs, then hands the rest to user space. Thus, all available memory should not…
I don't think anyone can objectively judge the attractiveness of both genders on one scale.
I like "pretty privilege." In my experience, improving appearance is already one of the top suggestions for job applicants. We're attuned the other way too, showing up to work in a nice outfit tends to be met with "got…
Isn't that the default? The user has access to all available memory unless they (or an admin) set a restriction. Unavailable memory is only unavailable because writing to it crashes the kernel, thus preventing you from…
Disk, memory, and CPU limits? The app should crash, not the kernel.
My company ran a test and found that they reduce token output on flagship model by something like 2-4x, and that number has been increasing with newer models. I suspect the increased subagent usage is driving this…
Because "depth" is recursive. You can search twice without waiting for the results of your first search: iteration. You can't if the thing you need to search for is the results of your first search: recursion.
That sounds more iterative than recursive. Recursion requires feeding the output back into the input, so creating version 4 requires results from version 3. You cannot recur in parallel. Iteration does not. You can…
A published formalization is code. I would not think humans have any edge when it comes to citing previously published results.
It is, but it was also a rhetorical device! I wanted to highlight that we're actually looking at a trio of concepts: intelligence, IQ, and value. Strongly correlated concepts, yes, but also meaningfully distinct!
What does IQ have to do with intelligence? (Forgive the flippant response. I believe it cuts to the core of what the parent was intending.)
https://docs.google.com/presentation/d/1XMCU2vxtxxqqQ2NY567n... You can download it in various forms as well, such as txt. Won't fix the tone, but at least it's scannable.
Embeded slides URL is a great hack from which you can just download as .txt directly.
No, you've been charged for every one of those licenses.
> Also, the statement was valid because it will be true for most. The statement seems more like a preconception than data. Source?
That is the appeal. It's an opinionated distro designed around keyboard navigation.
Those self-interested unions better stay away from my interests!
My distinction is "systems" vs "scripting" but yeah. I'm bringing Y2K back next!
Agreed, efficiency is still important, but being at the "frontier of efficiency" is significantly more relevant to commodity model providers than state-of-the-art model providers. Frontier labs are incentivized to route…
> Given that OpenAI is ahead in intelligence, it's also reasonably likely that they are at the frontier of efficiency too. Frontier labs have no incentive to be at the frontier of efficiency. Claude still leads the pack…
Scripting languages are awful. Python is one of the nicest scripting languages.
Happens with humans too! My senior colleagues check in with me significantly less often and cost significantly more in the meantime!
I feel like these are two competing goals: * the dev wants to describe an app in natural language then fall asleep while an AI works on it * the dev wishes that the same AI would write less comprehensive tests What…
The session is "just" the raw chat history in it's entirety (human and agent) and can be disseminated as such. This is what enables swapping between models, you simply send the whole context. Not sure how others do it,…
> [...] there is one unquantified, unproven comment at the top saying it's 100% [...] The parent was clearly not stating anything confidently. > Look at the later comments, they have substance oriented discussions.…
I don't follow. The default today is that an unprivileged user space application should not crash the machine. The kernel reserves what it needs, then hands the rest to user space. Thus, all available memory should not…
I don't think anyone can objectively judge the attractiveness of both genders on one scale.
I like "pretty privilege." In my experience, improving appearance is already one of the top suggestions for job applicants. We're attuned the other way too, showing up to work in a nice outfit tends to be met with "got…
Isn't that the default? The user has access to all available memory unless they (or an admin) set a restriction. Unavailable memory is only unavailable because writing to it crashes the kernel, thus preventing you from…
Disk, memory, and CPU limits? The app should crash, not the kernel.