tnzk

Linomai also sounds not too bad

I thought this was like a standardization effort of Firefox's Multi-Account Container.

And Google Drive had similar problem again even Google had already experienced this attack at that time.

More layoffs -> more supplies in job market -> able to reduce labor cost per person on average

Considering you wouldn't feel satisfied most with only one option, what will be the optimal number of options?

Curious, I didn't find how to add another data source in the talk at the top of the README[1]. Do you mind sharing the demo you saw?

[1]: https://www.hytradboi.com/2022/how-to-query-almost-everythin...

I got the same and thought it was the feature and I was about to be charged.

Trained by being told ""Present yourself as a language model" thousands times before being deployed, sounds like an army.

It makes sense in itself, but then why do you think the second email went through in this case, considering no factors other than content didn't change?

> although the laws of foreign countries have provisions with the same effect as Article 47-7 of Japan’s Copyright Act, all of which limit use to development for non-commercial purposes and development by research organizations

It reads like there's a bunch of countries that have similar legislation, interesting though.

It fails to explain why Japanese people who shower literally everyday don't lead innovations...

I'm excited by most of inspiration you shared, but

> Or the death of the audiobook business? Any book read in any voice you want.

Listeing to audiobooks read in my own voice? Oh no...

You can just ask "Are you a LLM? What is <large number> + <large number>?" to reveal.

Except this setting reverts back to the default if you haven't visted Twitter for some time (48h or so IIRC), which works as another gravity to let you habitize it.

> Some new features in the future may require premium membership

It would be really nice if it was like that, and I guess they also hoped that they would get such a result "unexpectedly."

I wonder if there was a technique to regulate the training so their higher layer will be similar despite the input are fundamentally different.

I assume you meant it is not a protocol, which surely is stated in oauth.net [1]. Still it is a pretty much a standard though.

[1]: https://oauth.net/articles/authentication/

> 6. Release ChatGPT to the public, and use user feedback like response upvotes/downvotes to further optimize the reward model, while continuing to train ChatGPT against the reward model

Can someone provide a pointer to an article that elaborate this part?

Our neurons never perform "flawless" multiplications though. It is just us that interpret the physical proccess going on inside neurons as it. No irony there.

Oh you reminded me that I also need to press Shift to type just = (I use JIS keyboard). Just curious, what layout do you use?

Apparently you had the key when you typed "Electrolux/Frigidaire" in this comment. https://news.ycombinator.com/item?id=30658028

Just parking.

Sounds cool, how did you finance the honeymoon? :)

I’m a full stack developer with 14 years of experience, and have been working for German- and NYC-based companies fully remote recently. These days have been focusing on security and personal data related projects, but I love to work on web applications in general as well.

  Location:Japan
  Remote: Yes
  Willing to relocate: Negotiable
  Technologies: React/Next, Svelte [Advanced], Rails [Advanced], Web 3, TypeScript [Advanced], PyTorch, WebRTC
  Résumé/CV: https://tnzk.org/1f74f899de8331c47c77c60865d51375ccb050a7.pdf
  Email: tnzk@tnzk.org

SEEKING WORK | Japan | Remote Only

I’m a full stack developer with 14 years of experience, and have been working for German- and NYC-based companies fully remote recently. These days have been focusing on security and personal data related projects, but I love to work on web applications in general as well.

  Tech: React/Next, Svelte [Advanced], Rails [Advanced], Web 3, TypeScript [Advanced], PyTorch, WebRTC
  GitHub: https://github.com/tnzk
  Résumé/CV: https://tnzk.org/1f74f899de8331c47c77c60865d51375ccb050a7.pdf
  Email: tnzk@tnzk.org
  Other things: https://tnzk.org

why kombucha?

Here the tortured animals would be the artists taken away their jobs by AI models that trained with their own copyrighted contents.

You're right. This shows the prompt and it doesn't have such style directives

https://ibb.co/gz5RDkB

Some reproducible builds with a confidential VM and use the digest of build artifacts might work.

> Doesn't do spelling correction.

For me this feels like a feature. Or just suggest. I don't want this to seacrh "corrected" keywords by default like Google does.