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

Location: Tampa, Florida

Remote: Yes

Willing to relocate: No

Technologies: C++, C, Assembly, COBOL, C#, Java, Javascript, Image Processing, Cross Platform Development, SDK Development, Mainframe Development

Résumé/CV: https://docs.google.com/document/d/1mQLv89cN536thPuK1DOttAAH...

Email: rabiega.dan AT gmail.com

I have eight years of development experience focused on developing well documented, high performance code, making it accessible to other developers, and debugging difficult issues.


I've seen several varieties of this idea several times, and it always seems pretty straw-mannish to me.

No one is asserting that computers themselves should have rights. But their users certainly do. If it is legal for me, a human person, to do something inside my own brain, why would it not also be legal for me to do a rough approximation of the same thing inside my computer?


The luddite argument is understandable for sure, but it is not in the advantage of progress to stall the development of ai with pseudo copyright concerns that are really just disguising the idea that such progress will displace existing people who currently derive benefit from the status quo.


Why do AI grifters love the word "luddite" so much?


How about the idea that these artists were given no choice in how their personal IP was used? Can I just steal your valuable work in the name of my own technical ‘progress?’


Discord's text search isn't exact, which can be a big problem sometimes. For instance, I was in a discord for a game mod recently which had questions asked about the terms 'element' and 'elemental', which were two very different topics. You simply can't narrow your search down to just one of those. Even searching for "elemental" brought up every usage of "element".


Yes, that's exactly it. The search is greedy and even partial hits (or hits with the word order not matching your search, etc.) will show up.


BLS calls U-3 the "official unemployment rate" I guess, but they publish it in a table with 5 other measures that count different things. I'm not entirely sure it's fair to blame them for the fact that most people just focus on that one measurement.


The fault is with the US newsmedia, which very consistently always talks ONLY about the U-3 number, and in my years, has never acknowledged that their are other numbers worth discussing.


If I understand the pricing correctly, that seems prohibitive. The models go up to 2k tokens of input, so wouldn't that mean that it costs $.16 per output token?


I'm confused as well, what's a 'token' and how many do I need?

Some FAQ items:

What does 2M tokens equal in terms of number of documents/books/etc?

This is roughly equivalent to 3,000 pages of text. As a point of reference, Shakespeare’s entire collection is ~900,000 words or 1.2M tokens.

Will the API be general public access starting 10/1?

No, we will still be in limited private beta.

How are the number of tokens per each subscription tier calculated?

The number of tokens per tier includes both the prompt and completion tokens.

How are tokens differentiated across engines?

These token limits assume all tokens are generated by davinci. We will be sharing a reference legend for other engines soon.

What will fine-tuning cost? Is it offered as part of this pricing?

Fine-tuning is currently only available for the Scale pricing tier.

Like, this doesn't tell me how much project X or project Y will really cost.


AI Dungeon is a pretty great use case. I'm not sure if it'll survive this pricing, though.


You get GPT-3 either way now, I believe. Dragon, the one locked to subscribers, is GPT-3 XL or something like that?

Relevant Developer Tweet: https://twitter.com/nickwalton00/status/1284842455188164609


Interesting, in the game settings it still refers to non-dragon as GPT-2. Thought it doesn't specifically say Griffon. It's a little confusing.

> Note: on custom prompts, the very first generation is generated with GPT-2 instead of Dragon (but every generation after will use Dragon).


Ah, that's referring to something else. By generation it means the first output string of every scenario.

IIRC I heard someone mention that OpenAI made them do that because people were trying to end-run around the GPT-3 access restrictions?


Maybe not by itself, but maybe useful in conjunction with eye tracking?


But eye tracking is a huge problem in itself, and, to my knowledge, it's not solved at all. I type this as I sit in front of my "15 notebook, and I need barely move my eyes to see the whole screen.

I think that figuring out where exactly I look on my screen is impossible problem to solve, since I always clearly see at least 3/4 of the whole screen. You can only guess where exactly on this 3/4 property of my screen I'm really focused.


I have tested several eye tracking solutions on VR platforms, including Tobii and FOVE. Both solutions worked well enough for me to play games at the same speed I would with other control methods. Although it may be that eye tracking from that close of a distance works better than from a laptop screen distance, I haven't tried the Tobii laptop eye tracking solution so I can't say.


I don't understand what you are saying here. It seems to me like you would expect hunger to increase over time after eating. Thus putting the part of your fast with the highest hunger intensity into your sleep time seems considerably easier than intentionally being awake for it?

I.E: If I skip dinner earlier I am effectively adding hours of minimal hunger experience but if I skip breakfast I am adding hours of maximal hunger experience.


> It seems to me like you would expect hunger to increase over time after eating.

This actually isn't the case at all. Thirst works this way: the longer you don't drink, the thirstier you get. But hunger tends to subside after a while if you don't eat, and comes back at the next normal meal time.


I have not read the study, but when I heard this on the radio the researcher specifically said that the weight loss was because of the reduction in calories.

The part that is interesting is that by instructing people to have their last meal earlier they naturally reduced their caloric intake without being instructed to do so.


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

Search: