It's a fascinating design but it's been 14 years since the concept was first announced and it's never really completely worked. If it ever was possible, it's not clear the talent for it still works for SpaceX.
Ok that's...just cheating. You can't take a benchmark like MMLU designed to test the performance of a single general language model and compare it to performance of a small specialized model designed to do well on MMLU.
There are many marginally employable swing voters who vote Republican when they have jobs and the Democrats in charge ask for taxes, then vote Democratic when they get put out of work and need a social safety net.
> it is possible to put 500 to 1000 TW/year of AI satellites into deep space, meaningfully ascend the Kardashev scale and harness a non-trivial percentage of the Sun’s power
We currently make around 1 TW of photovoltaic cells per year, globally. The proposal here is to launch that much to space every 9 hours, complete with attached computers, continuously, from the moon.
edit: Also, this would capture a very trivial percentage of the Sun's power. A few trillionths per year.
The real lesson here should be that doing crazy shit like swizzling the program counter in a signal handler and writing your own assembler is not a good idea.
She wasn't fired over the paper. She was fired because she demanded Google name the employees who gave (anonymous) negative feedback about the paper, and that anyone going forward who gave feedback on her work also be named, or she'd quit.
Their hearts were never really in XScale. Intel only ended up having that group due to a bizarre legal settlement with DEC, and pawned it off to Marvell in some misguided attempt at corporate streamlining.
That speedtest graph in the article is a great example of why averaging statistics is often a bad idea. The US fixed broadband mean is ~120mbps, but the median is more like ~70 mbps [1]. A few people with gigabit-ish connections are hiding the many people stuck with crappy sub-25 mbps connections.
I believe this was changed everywhere because glibc maintainers decided that it made more sense to run at library unload time, and then everyone else had to change their implementation to match "what Linux does."
That being said, none of these behaviors really make any sense. If the registered function isn't mapped into the process address space any more, what is supposed to happen?