cma

4 ms·

61.
by cma 2mo ago
Nemotron was using hybrid with recurrence via mamba layers since around April 2025.
62.
by cma 2mo ago
> Either be a programmer and code, or be a delegator and delegate Even if you just enjoy programming for the sake of programming, you can use it to review your code and then hand verify its findings, or prototype and explore feasibility
63.
by cma 2mo ago
So an ultramarathoner could hunt three back to back instead of one, or maybe two without being an elite athlete with modern hydrocarbon gels. Or get through part of a failed hunt midway and pursue another, etc.
64.
by cma 2mo ago
> If that's the level of evidence you need to be extremely alarmed, then you really should be a lot more worried about the alien invasion in Independence Day or the lizard men living under our feet. Now let's say instead of the
65.
by cma 2mo ago
You could have one really hard to understand proof of a theorem and then a lot of interesting human-understandable stuff that relies on that theorem. We already have lots of proofs with oracles, where you can work out consequences of what
66.
by cma 2mo ago
How much of inflation during Biden years was from Trump? For instance, Trump agreements to restrict oil production after covid lasted deep into Biden's term. The US still did better on inflation than most comparable peers in the after
67.
by cma 2mo ago
If it's using Aaronson's approach it shouldn't have any noticeable affect on generations. When it picks between options weighted by probability after the generation of logits, it still follows the probability mass, it just u
68.
by cma 2mo ago
The Google Toolbar for Internet Explorer let users search Google directly from any webpage, block pop-up ads, autofill web forms, and highlight search terms. It also displayed PageRank metrics, translated foreign languages, checked spelling
69.
by cma 2mo ago
Still doesn't look like it is hurting them, but to put it in perspective that's a 4% increase in profit, while 2025 inflation was around 2.5%. Closer to a 1.5% gain in real terms.
70.
by cma 2mo ago
>We have apparently reached the stage There used to be only 3 TV channels or so. You had things like Johnny Carson jokes causing a toilet paper shortage. In Britain, Delia Smith's How to Cook caused 10% swings in egg sales, far mor
71.
by cma 2mo ago
"A recent YC startup spent 250k on a domain. I was slightly shocked. But when I asked how much they raised after YC, the answer was 6.5m. So 1/26 of their round. This is the time we live in; one shocking number counterbalances the
72.
by cma 2mo ago
When you tell the AI: copy this function to here, a small window rewriter would mean it just corrupts and changes it instead of moving it. And even it's own tool use would have some small window dumb model changing the tool calls base
73.
by cma 2mo ago
Distilling is only prohibited from chats you prompt. But there are lots in the open like the LMSYS Chatbot Arena data. It's a essentially one of the largest public sources of preference data, maybe still useful for boostraping RLHF a
74.
by cma 2mo ago
> The reason why llama is open source is simply because it got leaked It arguably didn't really get leaked, and they had the .edu req mainly for fair use education exemption when legality of models was much more uncertain.
75.
by cma 2mo ago
*meant to say conscientiousness (keeping brakes, tires maintained), not consciousness
76.
by cma 2mo ago
Controlling for age in a profession with a high fatality rate like this seems like controlling for high alertness, consciousness, quick reaction times, and well restedness/low stress, low recreational drug and alcohol use. Of those, sl
77.
by cma 2mo ago
Alex Turner discusses a bit here at this timestamp: https://www.youtube.com/watch?v=pGlJQHVeKIA&t=14m27s
78.
by cma 2mo ago
They get a good bit more disproportionate per capita representation in the senate than in the house.
79.
by cma 2mo ago
This isn't the same with lookup tables, but in Explaining Attention with Program Synthesis they were able to replace some attention heads with Python programs: https://arxiv.org/abs/2606.19317
80.
by cma 2mo ago
Claude has found some weaknesses in other schemes, but not fully breaking anything yet: https://www.anthropic.com/research/discovering-cryptographic...
81.
by cma 2mo ago
That's if they are in a cave experiment: without that the sunlight cycle regulates it back to 24hr. But not for everyone, and that's N24/non-24, not DPSD, delayed-phase sleep disorder.
82.
by cma 2mo ago
The paper says: "Kimi K3 uses no explicit positional embedding (NoPE), and instead encodes positional information implicitly through the recurrent gating and decay mechanism of KDA."
83.
by cma 2mo ago
Great way to not get subscribers to your substack etc.
84.
by cma 2mo ago
x.com heavily penalizes off-site links.
85.
by cma 2mo ago
If they can distill fable into a full model post training run in ~15 days without the real thinking traces, yet we know Claude chats degraded with the thinking traces removed (chat resume bug from earlier in the year they reported stripping
86.
by cma 2mo ago
Won't modern wear leveling migrate rarely written data to spread new writes out to its blocks?
87.
by cma 2mo ago
You could use raw C in environments with effective sandboxes, Fil-C in environments without a sandbox. Or you could write a library that works for Fil-C and C without writing the library twice.
88.
by cma 3mo ago
Where does this explanation fall apart? There must be a cutoff point where more difficulty of getting the info down is worse than more ease. Otherwise, wouldn't a better recommendation be to do all your notes more slowly tracing with
89.
by cma 3mo ago
Already happened? > Trump’s comments, made hours after the large-scale military operation, mark one of the first times a U.S. president has so publicly alluded to U.S. cyber efforts against other nations, as these operations are typicall
90.
by cma 3mo ago
>Anthropic was blocked from releasing Fable without any such level of incident. The head of the NSA said Mythos breached almost all of their classified systems, though it was in an intention red-team test.

More ›