vinni2

It’s ironic that the hotter Europe gets, the wetter and colder Norway gets in summer.

> extract the contents directly out of the weights

If you can successfully demonstrate that then yes it is a copyright infringement and successfully doing that would be worthy of NeurIPS or ACL paper.

How can you be sure that what-if analysis by the LLMs are correct or plausible?

It’s incredibly optimistic and egotistical to even think that humans will be around that long.

> If a calculator works great 99% of the time you could not use that calculator to build a bridge.

But if the alternative is doing calculations by hand (writing code manually) there is a higher chance of making mistakes.

Just like calculations are double checked while building bridges unit tests and code reviews should catch bugs introduced by LLM written code.

Was it with search on or with parametric knowledge.

I hope can buy audiobooks from the iOS audible app soon.

Isn’t this the main idea behind https://lastexam.ai/

Perfect. Let the students submit AI solutions and let the uni assess using AI. Let’s all go on vacation and let the AI sort it out.

Care to share the source?

It’s getting there.

I struggled to find the source the fact that $1T is lost to scammers. From the FBI internet crime report in 2023 Americans lost about 12.5 Billion dollars [0]. That is far from 1T$.

Another source reports cost if cybercrime is reported to reach 9.5T$ in 2024 but again no source provided for that number [1]

[0] https://www.ic3.gov/AnnualReport/Reports/2023_IC3Report.pdf#...

[1] https://www.esentire.com/resources/library/2023-official-cyb...

For me the interesting part is they were able to recover data from iPad after 5 years being under water.

Can you please provide evidence for your claims and be factual?

Did you find out your cousin was gay through 23 and me or from him?

Can you share some of your wisdom on setting up a scalable inference infrastructure?

Depends on what these PhDs are supposed to do. Also is this an average Phd or a brilliant PhD level? There is a huge spectrum of PhDs out there. I highly doubt these phd level models are able to solve any problems in a creative way or discover new things other than regurgitating the knowledge they are trained on.

Why don’t they focus on delivering packages which are increasing in volume. They make us pick up packages from supermarkets and complain that there are not enough letters to deliver. I don’t know what obvious point I am missing.

I don’t know about Denmark but in Norway there is a discussion to pick up letters from designated supermarkets from where packages are already picked up.

I like this backlash. Europeans should show that they have values and that their market matters.

It is an interesting problem to train an LLM to make it aware only the things I know and with only my knowledge and beliefs stored in the weights. Not to be corrupted by the knowledge from the rest of the world. For new things it can have access to the web search and learn under my supervision. I.e, what data should be used to update it’s weights.

My parents and grandparents are non techies and don’t have a lot of experience but they can make whatsapp calls easily. It takes bit of getting used to though.

It’s a pity original link was not picked up.

https://news.ycombinator.com/item?id=43187724

I have to cover my face walking by Teslas. Driving by a Tesla is harder to avoid.

I turned off auto update let’s see how long it lasts.

And yet their stock price is all time high!

They can hire Homer Simpson.

Using x.com as source doesn’t give me confidence. I want to see official documents.