chaxor

38 ms·

481.
by chaxor 3y ago
I would disagree that 30B is not good enough. It heavily depends on which model, and what you're trying to use it for. 30B is plenty if you have a local DB of all of your files and wiki/stackechange/other important databases
482.
by chaxor 3y ago
I wouldn't be so sure that there isn't "any" form of compression going on - that just seems silly. It seems that at some level it is *fantastic* lossy semantic compression. This is one of the more fascinating aspects of
483.
by chaxor 3y ago
It seems like you're the negative one here. There is nothing inherently negative about living longer - just potential to have learned far more than others from experience. Any biological issues with existing for longer can have solut
484.
by chaxor 3y ago
The real goal would be for the model to determine what tool is needed to achieve the task, and use that to achieve it. Using the model to write code provides a way to achieve explainable and correctness guarantees (by reading and execu
485.
by chaxor 3y ago
You can run much larger models than llama-7B. Galpaca-30b or Galactica-120b for example.
486.
by chaxor 3y ago
The style fits pretty well with GitHub.com/charmbracelet
487.
by chaxor 3y ago
Hmmm... Perhaps that could be one component, but I don't know if proof of work in the typical way would be required? As in 'guessing' a hash (I don't know much about that area so I could be wrong here)., I don't thi
488.
by chaxor 3y ago
I did say it may do certain tasks with low performance. What you're saying is not really understandable (or simply wrong). The fact that it does understand how to do a 'simple' task such as finding where text is different is
489.
by chaxor 3y ago
Perhaps... but If that's true, you're likely not doing what is typically referred to as NER in computational linguistics. Either that or yours going with a type of 'dictionary based' regex system, which gets typically hi
490.
by chaxor 3y ago
They don't only run on high end systems. Good models can run on a desktop you have at home. If you don't have a desktop... I'm not sure what you're doing on HN.
491.
by chaxor 3y ago
I don't think the person you're responding to wants a network based or cloud based solution. When someone says they want it available they mean running on their own device. This is hackernews, nearly everyone on this site should h
492.
by chaxor 3y ago
It does, using this method. My immediate thought as well was '... Yeah, well vimdiff can do that in milliseconds rather than 22 seconds' - but that's obviously missing the point entirely. Of course, we need to tell people to
493.
by chaxor 3y ago
I am interested in a concept close to a CRDT - but I think meaningfully different. Is there a 'data structure' or 'scheme/setup' which fits the below description? Have a repo (with a specific git commit hash), whi
494.
by chaxor 3y ago
Perhaps one example would be 100 billion websites, but with a set of vectors for each website (chunked bert encoding, sum chunked glove vecs, whatever). Then you could have something like 3-100ish vectors for each website, which would be a
495.
by chaxor 3y ago
More is always better. That's people like LLMs. It takes them *decades* to process a database on consumer hardware, whereas more targeted system take merely seconds. Obviously decades > seconds, therefore: LLMs > smaller targeted
496.
by chaxor 3y ago
There are plenty of examples of encoders , like this paper, that are being applied to read through all literature, but using an LLM directly (typically referring to the decoder) is not really necessarily useful or efficient (for typical
497.
by chaxor 3y ago
It's concerning that there are no references to scispacy (from allenAI) in the paper. Scispacy is a bit dated in it's core tech, but it's still one of the easier ways of getting quick NER results on text. This paper is extre
498.
by chaxor 3y ago
This isn't a necessary part of intelligence .
499.
by chaxor 3y ago
Sorry about that. It was probably me. Sometimes you just want to figure out a cheap way to make a coffee roaster and can't figure out how to frame the right keywords search to the search. \/s
500.
by chaxor 3y ago
Actually, I think this problem is going to be solved easily with GPT, since it writes really concise and performant code. Using LLMs to generate scripts is so much more efficient and will make this bloat problem fade away. \/s
501.
by chaxor 3y ago
This is beautiful, but is there a decent way to plow through say, 20TB of text and put that into a vector database (encoder only)? It would be quite a great addition, especially if the vectors could then be translated into other forms (diff
502.
by chaxor 3y ago
Stop doing this. It's monumentally stupid to ask an LLM about any direct factual information it has not been trained upon.
503.
by chaxor 3y ago
I think that may be wrong. PaLM was 540B, but I don't think PaLM2 is near that. Perhaps 120B like Galactica large, but afaik they don't say.
504.
by chaxor 3y ago
If you want to solve math problems, LLMs are very useful for this. Exactly how trained professionals are. You make the model write code for you to solve it. Would you ask your dad to compute the correlation matrix between 40 thousand vector
505.
by chaxor 3y ago
I don't think it's actually using PaLM 2. If it is, it's not great, but I heavily doubt it is. The commentary they've made on it could mean many things. One of the possibilities is that they're using the API to PaL
506.
by chaxor 3y ago
Just let people be monumentally stupid like this. You can't correct it.
507.
by chaxor 3y ago
It's quite a vast difference between GPT-3.5 and GPT-4
508.
by chaxor 3y ago
I'm pretty sure they're trying to suggest that LLMs in general are not useful because they can't do this type of thing. It's just the next iteration of goal post moving and should effectively be ignored. Many artists an
509.
by chaxor 3y ago
It heavily depends upon the definition of general intelligence, because we haven't really settled on a definition as of yet. In some circles (of course the ones from Microsoft) the list of requirements is very close to being met by so
510.
by chaxor 3y ago
This is dangerous! How do you know it's correct!? What if you need different memory management!? What if it writes the wrong thing?! Program only in Assembly, because otherwise you risk some real dangers of not knowing what's goin

More ›