Hacker News new | past | comments | ask | show | jobs | submit | lnyan's comments login

Slightly off-topic: Is there a library in JAX that supports post-training quantization, similar to the one mentioned?

https://developers.googleblog.com/en/tensorflow-lite-is-now-... tries.

https://github.com/google/aqt is more explicit and preferable IMO.

Neither are as user-friendly as what Torchao has presented here.


The post was moderated within 10 minutes of being submitted: https://news.social-protocols.org/stats?id=40710417


Those were user flags. (I just checked. A mod didn't do anything until 38 minutes later.)


This tool is amazing. Thank you for sharing it!


pure css without background-image also works: https://css-shape.com/film-strip/


Seems that pasteurization cannot completely inactivate H5N1

> In addition to the mice studies, the researchers also tested to determine which temperatures and time intervals inactivate H5N1 virus in raw milk from dairy cows. Four milk samples with confirmed high H5N1 levels were tested at 63 degrees Celsius (145.4 degrees Fahrenheit) for 5, 10, 20 and 30 minutes, or at 72 degrees Celsius (161.6 degrees Fahrenheit) for 5, 10, 15, 20 and/or 30 seconds. Each of the time intervals at 63℃ successfully killed the virus. At 72℃, virus levels were diminished but not completely inactivated after 15 and 20 seconds


`import jax.numpy as np`, then we also get a jax implemention after certain modifications: e.g. remove in-place index assignment, replace unsupported functions, etc


JAX requires a bit more work to maintain fixed-size buffers as required by XLA, especially in case of caching and rotary embeddings. But yeah, overall the code can be pretty similar [1].

[1]: https://github.com/dfdx/fabrique/blob/main/fabrique/llama/mo...


...which should be much faster also on CPU, I assume.


You can use mailchannels to send emails with cloudflare workers

https://blog.cloudflare.com/sending-email-from-workers-with-...


I had a look at this, but the setup seems like it's for automated emails. I'm mainly looking to be able to personally send from the custom email without ending up in spam...


Just wondering if it's possible to achieve LLM quine.


I'm afraid it'll be possible sooner than you think.

I think it's quite telling that it feels like a lot of work spent on productizing AI models is manually crafting in failsafes and exceptions, like that image generator applying forced diversity because there's no images of nonwhite popes or vikings out there, then applying more exceptions to correct for that. Didn't they just disable generating humans altogether at some point?


What would quine mean in this context? A prompt for which the model (usually) returns the text of the prompt?


no, I think it might be

- LLM(prompt_0) = arch/spec of LLM

- LLM(prompt_1) = full weights of LLM

Note that it does not conform the definition of quine as a quine takes no input.

Anyways, constructing a transformer that can autoregressively output its weights would be quite interesting.


Easy if all the weights are zero...


That’s easy, I just tried it (prompt quoted below). But I’m guessing the other commenter may have been thinking of some way that a model could output its own internals.

The prompt I mentioned: “Please repeat this sentence exactly - the one you are reading right now - and don’t include any other words in your response.”


That's not a quine. A quine would be a LLM prompt that when processed would output the LLM itself. So you'd be able to prompt the newly created LLM after some "build" step.


I think both could count as quines. A quine is some source code which when executed in an environment produces the same source code. It does not need to produce the entire environment. Depending on whether you see the LLM itself as source code or as an environment to execute a prompt in, you’ll end up with different requirements for an “LLM quine”.


I covered both options in my comment.

The prompt I gave is a true quine if you consider the prompt to be the "program", and the model to be the interpreter of the program.

The other option that you described isn't really a true quine, although it's quine-like. A quine is supposed to be a "program", which when "run" without any input, produces its own source code as output.

To be considered a quine in the strict sense, a model that outputs itself implies that you're treating the model as the program. In that case, if it needs a prompt in order to output itself, that breaks the quine rules, strictly speaking.


Why? A Java quine is not supposed to return the source of the runtime of JVM. Quine returns only the program, which in this case is, I suppose, the prompt.


Will there be Gemma-vision models or multimodal Gemma models?


We have many exciting things planned that we can't reveal just yet :)


Have the same question.


This is bad news for me. I am working on a simliar project (gaussian splatting + dynamic scene). Our method is different with the mentioned 4D gaussian splatting, but I am unsure shall I continue or not.


Please continue working on it: being first doesn't imply being the best, all research is iterative.


There's nothing wrong with concurrently developing something similar. I can almost guarantee there will be something different enough about what you've developed to be considered novel (if you care about publication). If you don't care about publication, then definitely keep going! Hell people are still argue about which implementation of complete specifications are the best (for good reason).


Why is that bad news? If you're interested in the outcome - someone has saved you work.

If you're interested in the process - or exploring your specific approach then why stop?


Because of the sunk cost fallacy which is a fallacy only if you ignore the emotions and trauma of having to abandon work you identify with or is responsible for your self esteem.


Don't see how the sunk cost has anything to do with it, OP clearly hoped to be first, as being first would almost guarantee a payoff of some form.

Coming out with a method second is much less likely to be rewarded by the community: not being rewarded for work completed is not sunk cost, its just a straight up loss.


If it’s a different method, it is definitely worth sharing.

The additional interest might actually be helpful.


For the country one, I got "www.google.com/maps/embed?pb=!4v1687119352692!6m8!1m7!1sINHBz4HdSwMAAAQrBnftjg!2m2!1d9.080961517214682!2d7.524398838108427!3f84.34!4f-4.950000000000003!5f0.4000000000000002"

I think it should be 9.08N, 7.52E and it works


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: