atq2119

4 ms·

31.
by atq2119 4mo ago
Agreed. To put this in perspective, batch 1 token decode is bandwidth limited in theory. Memory bandwidth of RTX 3090 is listed as 936GB/s. The post isn't fully clear on which model they used and how big it is, but even assuming i
32.
by atq2119 4mo ago
Interesting to see somebody argue out in the wild for what I have been subjected to in the past and have long considered absolute worst practice. It may be that my perspective is different because my work tends to be in "hard" fou
33.
by atq2119 5mo ago
That is definitely not my experience using Claude Code with Opus. I work in a very sparsely commented code base, and the agent produces substantially more comments than the surrounding code.
34.
by atq2119 5mo ago
Of course it does, but that's beside the point. As a software developer, you must never subject your team mates to a PR that you yourself believe to be low quality. The point of code review by others is to catch things that you mis
35.
by atq2119 5mo ago
That quote shows an utter disregard for basic human decency. It is the responsibility of the person running the coding agent to make sure the resulting PRs are high quality. Putting that on your team mates, or worse, random open source proj
36.
by atq2119 5mo ago
You may be confusing some abstract unachievable ideal with the reality of the world we live in. In reality, being superwealthy absolutely comes with a tremendous amount of power. It is also pretty much the opposite of anarchic, given effect
37.
by atq2119 5mo ago
This is a genuine problem indeed and part of the appeal of an UBI. The idea being that if the rules of redistribution are dead simple, then that helps minimize the potential for grift, which in turn minimizes the potential danger of a redis
38.
by atq2119 5mo ago
Unless you're systematically repeating the exact same task, the most parsimonious explanation is that you're seeing natural variation based on different tasks, random sampling of tokens, etc.
39.
by atq2119 5mo ago
It's been a while since I saw a detailed paper on a high end training run, but extrapolating from what I remember, it seems those training runs are in the 10s of trillions of tokens. This already accounts for potentially sampling token
40.
by atq2119 5mo ago
If the inputs are constant, then all the multiplies are constant and the only thing that toggles is the accumulation. Which explains the pi situation. Normal vs uniform is less clear, but also not as much of a difference. The arguments abou
41.
by atq2119 5mo ago
Think about it from the other end. Why would any bits flip at all in the data path of your matrix multiplier when all the matrices are 0?
42.
by atq2119 5mo ago
You may be talking past each other. You're talking mostly about quality of life. That is different from being (de)humanized. Dehumanizing is treating other humans as not, or not fully, human. For example, the term "human resources
43.
by atq2119 5mo ago
And, since there's a pipeline from students and hobbyists to professional use, it's risking the future.
44.
by atq2119 5mo ago
That sounds more like the 80286/80386.
45.
by atq2119 5mo ago
Sure, that's the main challenge with building good sandboxing systems. But it's not actually that hard to do when the will to do it is there. For example, Android already allows you to give apps restricted access to your media. My
46.
by atq2119 5mo ago
Internet access. An editor extension does not need it.
47.
by atq2119 5mo ago
Fully agree with the first half of your comment. The second half goes off the rails, though. I rarely see people complain about sandboxing. What people complain about is when devices are locked down in a way where you are only allowed to in
48.
by atq2119 5mo ago
I don't think this is factually accurate. What it really boils down is a question of scale of societies. Most of us humans inherently value each other. There are exceptions, and small communities can get nasty. But for the most part,
49.
by atq2119 5mo ago
It really depends on how often you use git bisect and blame. This varies greatly across projects. That said, if/when stacked PRs become a first-class citizen in GitHub, more projects will see the benefit of this approach (though they&#
50.
by atq2119 5mo ago
Thank you. I vaguely remember being put off by Mercurial at the time and feeling much more comfortable with git, but couldn't have put it into words as well as you did. People always say you have to know git internals in order to use i
51.
by atq2119 5mo ago
I agree that training is a forever task, and the current rate of training is probably not sustainable. But all that means is that once the current investment mania ends, the market will most likely find a new equilibrium where continuous tr
52.
by atq2119 5mo ago
> Without continuous training there is no inference product. This claim deserves teasing apart. Clearly, training is a Red Queen's race today. If a model provider were to unilaterally decide to stop training, they would very quickly
53.
by atq2119 5mo ago
Let's not get ahead of ourselves. Millions , really? I can believe there are a lot of enthusiasts doing this, but "millions" needs a citation.
54.
by atq2119 5mo ago
Inference at per-token pricing is profitable. The article's point is that if you're relying on flat fee subscriptions, a rude awakening may be coming. That seems plausible to me. Issues around token quotas are a frequent topic on
55.
by atq2119 5mo ago
> Untangling the entire lot to make the game available upon closure would be a nightmare in some cases. This is only true if the game wasn't architected with open-sourcing in mind. Which affects how this kind of law should be struct
56.
by atq2119 5mo ago
There's the adage that writing is thinking, but even more accurately at least for me, editing is thinking. Neither typing speed nor dictation speed is a true bottleneck, but editing speech seems like it'd be harder than editing
57.
by atq2119 5mo ago
> Teachers were never highly paid. Teachers never got rich, that's fair, but there was a time, at least in my culture, when teachers were usually one of the most highly regarded and respected people around, on a similar level as pri
58.
by atq2119 5mo ago
> The idea is also that parents with lower intelligence will value education for their offspring less and care less about whether their kids learn to read well, etc. Educated people can neglect their kids. And less educated people can st
59.
by atq2119 5mo ago
Yeah, Trump showed the world exactly who he is even before his first term. He has absolutely no principles other than feeding his own wealth and ego. Anybody who is surprised by what has happened is either an incredibly bad judge of charact
60.
by atq2119 6mo ago
Plus a systematic way of keeping the Gini coefficient of wealth small in a sustainable way. I'm a fan of establishing sovereign wealth funds whose dividends are paid out equally per capita for this purpose.

More ›