Did we read the same graph? DeepSeek Gundam 200 dpi appeared to get similar perf as dots-ocr, but with less tokens needed. The x axis is inverted, descending with distance from the origin.
> the behavior of a complex system can simplify as it passes from one state to another. “Sometimes a high-dimensional system can tip,” said Lenton, “and when it gets near tipping, it starts to behave like a much lower-dimensional system.” The lesson, he added, echoes the one learned at Peter Lake: to “simplify without oversimplifying.”
Sounds a lot like what people WANT neural networks to do. Collapse a high dimensional situation into a very low dimensional network. Ideally a binary answer, yes or no. I wonder if this bifurcation chaos math has implications in ml work
> If Eli Lily is the only producer, how is the gray market being supplied? This makes no sense...
It turns out enterprising chemists and pharmacists are capable of reverse engineering.
I don't think it's that hard to figure out how someone might do it -- imagine having to reverse engineer food you've received, given many samples. Imagine some of those samples might have "fallen off the back of a truck".
> That being said, I'm waiting for oral GLP1 agonists. Injections are a hassle and gray market ones even more so
This is really going to be the second leg of adoption and will catapult GLP1s even further IMO. Rybelsus has not really seen a ton of popularity compared to the injections. That said, Orforglipron is Eli Lilly's upcoming oral GLP1 and it looks to have really good results:
Yes, chemists don't need to reverse engineer that (although they can) because it will be in the patent. I'm not sure if the production process is though.
The phrasing in this sentence implies that the Julia language could be named for the code breaker, as much as it could be named for anyone else. In other words, it wasn't named for the code breaker, but it might as well have been.
The follow up comment gives hard quantitative fact that the language wasn't named for anyone or anything. I can see how both comments are correct, the first implicitly, the second explicitly
The first comment is, at best, pointing out a name collision. It's otherwise fairly meaningless yet misleading. Julia could be named for any Julia in the known universe... but it wasn't.
I think a good rule of thumb is to stop if it hurts. I presume you're not using a laser. Your eyes are pretty good at telling you when something is too much
I thought they were overpriced until I looked into building one myself. The heat sinks were going to be really expensive. The LEDs and lenses weren't cheap either.
I bought an infrared parabolic heater to test this out once, as a friend made a comment insinuating that more time in front of fire meant a better sleep. I ran it for a few days, but the results were mixed and inconclusive.
I suspect because the longer IR wavelengths just heat the body up, it's hard to get enough of the more "useful" wavelengths without overheating
It has UV, which stresses out the cells
You can't PWM the sun, it's hard to get full naked coverage in front of it. Doesn't perfectly target NIR or NIR-II biological windows.
