Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Code Llama, a state-of-the-art large language model for coding

ai.meta.com

Post reply on HN
Print view View original

431–440 of 525 posts

Re: Code Llama, a state-of-the-art large language model for coding

#431

Post by sicariusnoctis »

wokwokwok wrote: post #76
benvolio wrote: post #51

>The Code Llama models provide stable generations with up to 100,000 tokens of context. Not a bad context window, but makes me wonder how embedded code models would pick that context when dealing with a codebase larger than 100K tokens. And this makes me further wonder if, when coding with such a tool (or at least a knowledge that they’re becoming more widely used and leaned on), are there some new considerations tha…

That seems daft. You can, I suppose, contract your code so that it’s context free and uses less tokens, but that makes it more confusing for humans and language models . Taken to the extreme, you can see obviously with one letter functions and variables like i, j, k the model will be able to infer literally nothing and, thus, produce arbitrary nonsense. Clearly the solution is to do what we already do to manage compl…

64k tokens ought to be enough for anybody.

Re: Code Llama, a state-of-the-art large language model for coding

#433

Post by totallywrong »

suyash wrote: post #201
syntaxing wrote: post #185

TheBloke doesn’t joke around [1]. I’m guessing we’ll have the quantized ones by the end of the day. I’m super excited to use the 34B Python 4 bit quantized one that should just fit on a 3090. [1] https://huggingface.co/TheBloke/CodeLlama-13B-Python-fp16

can it be quantised further so it can run locally on a normal laptop of a developer?

Just started playing with this, there's a tool called ollama that runs Llama2 13B on my 16GB M1 Pro really smoothly with zero config.

Re: Code Llama, a state-of-the-art large language model for coding

#434

Post by BoorishBears »

kordlessagain wrote: post #427
BoorishBears wrote: post #421

Earlier quoted context omitted.

Perfect example of why I said academic interest. Awkward tie-ins between SBF and value systems (?) have no effect on practical usage. A theoretical concern they might train on my API data after saying they won't doesn't either. Amazon might be training on everything not bolted down in S3, not worth wasting brain power on that. The moderation API isn't some magic gotcha, it's documented. They don't want to deal with p…

In my opinion the risk is coupling accelerated intelligence to competitive business models.

The accelerated intelligence wouldn't exist without competitive business models.

Re: Code Llama, a state-of-the-art large language model for coding

#436

Post by astrange »

swader999 wrote: post #305
mymac wrote: post #132

Earlier quoted context omitted.

But it won't stop there. Why would it stop at some arbitrarily defined boundary? The savings associated with no longer having to pay programmers the amounts of money that they believe they are worth (high enough to result in collusion between employers) are just too tempting.

The answer to AI stealing your job is to go ahead and start a company, solve a hard problem, sell the solution and leverage AI to do this.

The only thing that takes anyone's job is demand shortfalls. Productivity increases certainly don't do it. It's like saying getting a raise makes you poorer.

Re: Code Llama, a state-of-the-art large language model for coding

#437

Post by astrange »

ilaksh wrote: post #20

Between this, ideogram.ai (image generator which can spell, from former Google Imagen team member and others), and ChatGPT fine-tuning, this has been a truly epic week. I would argue that many teams will have to reevaluate their LLM strategy _again_ for the second time in a week.

SDXL and DeepFloyd can spell. It's more or less just a matter of having a good enough text encoder.

I tried Ideogram yesterday and it felt too much like existing generators (base SD and Midjourney). DALLE2 actually has some interestingly different outputs, the problem is they never update it or fix the bad image quality.

Re: Code Llama, a state-of-the-art large language model for coding

#439

Post by SubiculumCode »

KaiserPro wrote: post #355

This is great for asking questions like "how do I do x with y" and this code > isn't working, whats wrong? Much faster that googling, or a great basis for forming a more accurate google search. Where its a bit shit is when its used to provide auto suggest. It hallucinates plausible sounding functions/names, which for me personally are hard to stop if they are wrong (I suspect that's a function of the plugin)

hallucinations can be resuces by incorporating 'retrieval automated generation' , RAG, on the front end. likely function library defs could be automagically entered as prompt/memory inputs.

Re: Code Llama, a state-of-the-art large language model for coding

#440

Post by chaboud »

awwaiid wrote: post #432
wavemode wrote: post #413

Earlier quoted context omitted.

One.

Two when you include -2, which I certainly think we should in this circumstance.

Are we including it just to poke the bear? Prime numbers are typically defined as numbers with no positive divisors other than one and the number in question.
Post reply on HN