3
points
kristjansson
Joined 4,281 karma
- kristjanssonReally why have the PCI/CPU artifice at all? Apple and Nvidia have the right idea: put the MPP on the same die/package as the CPU.
- Sure, some fraction of withdrawn water is retained due to contamination. That’s counted in the consumption numbers of the source study? The operative question is still whether DCs and related power generation consume a concerning about of water.
- > 20+% margin on cars
wouldn't that be nice.
- It was way too easy for the venv to be the only artifact of the process used to produce it. pip install, hit a bug, install some other dep, another bug, oh this one needs to be downgraded, and so on until it works. Without more effort to memorialize / remember the steps, deleting the venv destroyed the work to create it.
uv, in addition to its raw speed, is very clever to record things in pyproject as the user interacts with it.
- > I don’t want a thing that hijacks my terminal and PATH
The shame is ... it never had to be that way. A venv is just a directory with a pyvenv.cfg, symlinks to an interpreter in bin, and a site-packages directory in lib. Running anything with venv/bin/python _is_ running in the virtual environment. Pip operations in the venv are just venv/bin/python -m pip ... . All the source/deactivate/shell nonsense obfuscating that reality did a disservice to a generation of python programmers.
- Step One: get them to a better payment processor than PayPal! I waded through it, but that's a high friction funnel.
- uh. there was a bit more to the story than 'yup totally unalloyed free lunch'
- While no one will deny you (or I guess your system) the immense satisfaction of 100x improvement on a given step, I think it would be helpful to note the frequency of this rebalancing step, and to contextualize your result in terms of the runtime (or throughput) of the workload(s) you were using to evaluate.
e: also comparison a fixed (nothing faster than 0!) and random policy might be informative if your intent is to publish this as improvement for the object problem, not just a demonstration of ARDS.
- This is quite cool, but I must note that the 5x reported in the headline is the _runtime_ of the load balancing algorithm itself, not the load factor or throughput of the system or what have you.
> On average, it takes about 540 ms to re-balance the experts and achieves a load balance factor of 0.66 (calculated as the ratio of average to maximum tokens generated per GPU).
> ...
> We also consider a non-public reference implementation from a frontier lab that we have access to. This implementation avoids explicit iteration and reduces the rebalancing algorithm runtime to 19.6 ms while achieving the same balance factor as the open-source algorithm.
> ...
> The resulting algorithm matches the load balance factor of the other baselines while reducing runtime to just 3.7 ms, yielding a 5.0x speedup over the internal reference implementation.
- Whole seconds will have been wasted!
- yes yes yes we'll just invent the world from whole cloth at 60 fps
- I mean there is the chance it's on someone else's computer ^W^W^W^ the cloud, and his provider of choice doesn't offer easy access to deep scrollback ... which means this is only inefficient, not inefficient and pointless.
- If it’s in the context window … it’s sitting around as plain text. I guess asking is easier than scrollback?
- There's an older talk Simon Peyton Jones (IIRC?) gave about some development or other in haskell, in which he suggested that many software systems have some aspect of the swamp or the marsh into which you must eventually wade - that there's a mucky, sticky, irreducible aspect to the problem that must be dealt with somewhere, regardless of how elegant the rest of the system is.
"that marsh thing" has stuck with me, and been a frequent contributor to my work and thinking. I'll happily take Law of Conservation of Ugly as a _much_ better name for the thought :)
- B200 is 1kW+ TDP ;)
- $500 isn’t even out of the question for a great knife, it feels like about the right price point for the product.
- On one hand I would buy that instantly. On the other hand, it’s already missing small pieces thanks to my current mandolin. Not sure I want to make my scariest utensil scarier.