Palmik

21 ms·

361.
by Palmik 3y ago
Added to my reading list! The world of quantizations is moving so fast even TheBloke might not be able to keep up! So Atom base models would be compatible with Punica? I also wonder, many people already train LoRAs in 8 or even 4 bit (for t
362.
by Palmik 3y ago
This is amazing, and will unlock many possibilities. I just recently read the S-LoRA paper, which is related, but it's even better to have a working (and extremely efficient!) implementation. How hard would it be to adapt your kernels
363.
S-LoRA: Serving Concurrent LoRA Adapters(arxiv.org)
41 points by Palmik 3y ago|8 comments
364.
by Palmik 3y ago
If you're looking for more e2e math / latex aware OCR checkout https://github.com/facebookresearch/nougat
365.
A glimpse of the next generation of AlphaFold(deepmind.google)
1 points by Palmik 3y ago|0 comments
366.
by Palmik 3y ago
The gap between the value added by GPT 3.5 and GPT 4 to my every day tasks is immense. Most people, outside of these circles, don't realize that such a large gap exists.
367.
by Palmik 3y ago
The UI looks very clean, congrats on the launch. I see you are using lexical, what was your decisions process to pick between lexical, prosemirror and other alternatives?
368.
Executive order for AI: Biden will sign executive order to address his concerns(apnews.com)
6 points by Palmik 3y ago|1 comments
369.
by Palmik 3y ago
Some of my favorite examples of NextJs suddenly breaking major things in minor releases: Environment variables stopped working: https://github.com/vercel/next.js/issues/53367 The <Image> component stopp
370.
by Palmik 3y ago
Typically, vector databases store the index in memory, requiring machines with a lot of RAM. There are ways to reduce the memory usage, e.g. with quantization, but more recently disk-first databases (LanceDB) and algorithms (DiskANN) are be
371.
State of AI Report 2023(stateof.ai)
2 points by Palmik 3y ago|0 comments
372.
Nougat: Open-source LaTeX aware OCR for math-heavy books(github.com)
1 points by Palmik 3y ago|0 comments
373.
by Palmik 3y ago
Recently I was quite surprised how good inversion from the OpenAI ada-002 embeddings can be: https://twitter.com/thesephist/status/1698095739899974031
374.
Modal: Cloud function platform now generally available(modal.com)
2 points by Palmik 3y ago|1 comments
375.
Spotify's AI Voice Translation(newsroom.spotify.com)
1 points by Palmik 3y ago|0 comments
376.
by Palmik 3y ago
So far the examples all have one thing in common, https://<the_domain> returns: This site can’t be reached The webpage at https://<the_domain>/ might be temporarily down or it may have moved
377.
LanceDB: Open-source vector database with disk based indices written in Rust(lancedb.com)
1 points by Palmik 3y ago|0 comments
378.
In Search of a Bigger Boom(blog.nuclearsecrecy.com)
2 points by Palmik 3y ago|0 comments
379.
Commerce Department and VCs partner to encourage responsible AI use for startups(fortune.com)
1 points by Palmik 3y ago|0 comments
380.
by Palmik 3y ago
I think we should not try to forcefully preserve jobs that are no longer needed, AND I think it could cause harm in the short term, and lead to further transfers of wealth.
381.
by Palmik 3y ago
Well, I am definitely against the kind of involvement these large companies, that are set to benefit the most, are argumenting for, which is limiting / gate-keeping access. Rather, I could imagine the opposite -- some sort of guarantee
382.
by Palmik 3y ago
I agree we should not try to preserve jobs that are no longer needed. Though I don't think there's any copyright involved in your analogy :) I just think we should be clear about the potential short-to-medium term effects / h
383.
by Palmik 3y ago
Movie critics harm the original work / producers of the work in different way: they dissuade the consumption of the work, rather than providing a 1:1 replacement of it.
384.
by Palmik 3y ago
I think we should not stop innovation, but let's not minimize the impact on the producers of the content (the whole of humanity). First of all, even if the models never reproduce any of the copyrighted content verbatim, they will soon
385.
by Palmik 3y ago
Summary of their arguments for why training on copyrighted datasets is OK: (1) Training AI systems on large copyrighted datasets constitutes fair use under current copyright law. They believe it is highly transformative, does not harm the m
386.
by Palmik 3y ago
To anyone that knows what the vulnerability is, this announcement does not bring any new information.
387.
by Palmik 3y ago
I see a lot of people misunderstanding what this is about. What this allows is incrementally updating the attention cache. It does not allow the model to see beyond its usual attention window. As they explain in a README, only the tokens th
388.
Designing Dex: research language for typed, functional array processing [pdf](cambium.inria.fr)
1 points by Palmik 3y ago|0 comments
389.
Better Configuration for Less(nickel-lang.org)
2 points by Palmik 3y ago|0 comments
390.
Vector storage is coming to Meilisearch(blog.meilisearch.com)
1 points by Palmik 3y ago|0 comments

More ›