3abiton

18 ms·

361.
by 3abiton 2y ago
I wonder if those examples can be dumped down even further for lower age brackets. One of the "powers" of LLms
362.
by 3abiton 2y ago
I've been following nutrition "science" for the past 10 years, and to say it's cluster fudge is an understatement.
363.
by 3abiton 2y ago
If you were a book recommender system, what other books are similar to yours? I've read plenty of science/maths light read non-fiction, so I want to compare the reading experience before I jump into the book.
364.
by 3abiton 2y ago
I own a boox tablet (full fledge Android tablet with eink screen), and this sort of things would be perfect for it. I wonder if in 5 years the mobile hw would support something like that locally!
365.
by 3abiton 2y ago
I remember seeing these videos on youtube long time ago, but couldn't remember the sources. Thanks for sharing!
366.
by 3abiton 2y ago
This is really big. I always thought Vscode ssh capability was just some black magic that the open source community wasn't able to replicat well with vscodium extensions. Yet, it seems the reason why is that MS threw a curve ball with
367.
by 3abiton 2y ago
So more 'mature' models might arise in the near future with less params and better benchmarks?
368.
by 3abiton 2y ago
It depends on the features you want. My approach is to search in their supported devices, and sort by ram.
369.
by 3abiton 2y ago
I am really looking forward for JAX to take over pytorch/cuda over the next years. The whole PTX kerfuffle with Deepseek team shows the value of investing in more low levels approaches to squeeze out the most out of your hardware.
370.
by 3abiton 2y ago
I am looking up chunking techniques, but the resources are so scarce on this. What's your recommendation?
371.
by 3abiton 2y ago
I second Exhalation, it was a great experience, and I couldn't share that excitement after reading as no one in my (offline) circle had read it at the time. Reddit was one of those places where it was brought up at least.
372.
by 3abiton 2y ago
3x the price for less than 2x the speed increase. I don't think the price justifies the upgrade.
373.
DeepSeek's Refusals: How Do the Guardrails of Language Models Compare?(medium.com)
2 points by 3abiton 2y ago|0 comments
374.
by 3abiton 2y ago
I am extremely grateful so far for their work and contributions, nut they are right. China is leading the way despite all the hurdles put by the chip act.
375.
by 3abiton 2y ago
Been out of the loop for while, anything exciting?
376.
by 3abiton 2y ago
The question is rather a bit flawed without knowing your background, as to determine what to suggest. That being said, I would argue a lot of the novelty and recent advances in AI/ML are not yet documented in books, but rather in dense
377.
by 3abiton 2y ago
> but it seems it's still considered mediocre or not as good as X was. Because there was a big fracture and doubt in the community support for Wayland, not to mention alternatives like mir popping up and diverting resources.
378.
by 3abiton 2y ago
> I believe XWayland is there to stay for retro compatibility. That seems fine to me, retro-compatibility seems like a good thing. That would not be unlike XQuartz on Mac, which doesn't make us question Mac's own display server
379.
by 3abiton 2y ago
How's the performance and features compared to pinokio?
380.
by 3abiton 2y ago
I recently got a good laptop with great hardware, so I opted for Wayland (with hyprland). It is still finniky. One main issue is Xwayland dependency that feels a bit warped. Another one is gaming support (Wine still heavily depends on X11).
381.
by 3abiton 2y ago
This could be the reason why nutrition "science" is so fragmented. There is no way to ensure the process is being conducted correctly.
382.
by 3abiton 2y ago
But I feel we're going back full circle. These small models are not generalist, thus not really LLMs at least in terms of objective. Recently there has been a rise of "specialized" models that provide lots of values, but that
383.
by 3abiton 2y ago
But the way they did their PR for O1 made it sound like it was the next step, while in reality it was a side step. A branching from the current direction towards AGI.
384.
by 3abiton 2y ago
One interesting feature: an index of all existing (already documented) hardware hacks on commercial devices. I know for example the IP camera industry has a large models that many different people are modding, but their docs are scattered a
385.
by 3abiton 2y ago
Really great work, suggest for a next post: the VRAM requirements estimation calculation for running models locally. Especially with different architecture and different Quants, it gets always confusing and even online calculators give diff
386.
by 3abiton 2y ago
> - people choose venues that maximises their chances of getting published (and pay for publication sometimes, I'm looking at you, APC); > > - little concern given to excellence, rigour, and impact; It's because the kpis
387.
by 3abiton 2y ago
It's always fun to dog into hn lore!
388.
by 3abiton 2y ago
Scientific peer review is another facit of civilization that its current design does not allow it to scale well. More and more people are being involved in the process, but the qualityis forever going down.
389.
by 3abiton 2y ago
Because the prices are not competitive enough for hobbyists. How can you expect people to adopt and develop on top of your chips if they can't afford it?
390.
by 3abiton 2y ago
My gripe with the post is that there is no objective "cognitive load" solution. Arguably this varies from 1 person to another.

More ›