community-help

[mortal compass]

Hi

[limber hollow]

why does gp chat answer text break, is there a limit on the message length on the platform?

hi

#community-help why does gp chat answer text break, is there a limit on the message length on the platform?

[open root]

is there a limit on how many messages i can send on chatgpt? i only put one question and now it says "Too many requests in 1 hr"

[stark junco]

Tell it to re-write it with more words and it should work

[north fjord]

To a certain extent it doesn't really matter. The main purpose of school is for students to become educated. To the extent a student cheats in school they cheat themselves. If they cheat all the way to a doctorate degree, of what value would it really be if they were actually ignorant. I'm not saying cheating doesn't matter, but rather that the primary victim is the cheater.

[subtle compass]

I can't find a clear statement about ChatGPT license/rights, I'm just wondering if I can use generated "Output" text by Chat GPT from my prompts, instructions, description that I give, for my commercial purposes?

[eternal stratus]

Is there a maximum character that ChatGPT can write ? for example if i tell it to write me an essai about something ? I tried but it stops after maybe 1 and a half page of full text

[rugged wing]

If you say "Continue", it will actually continue where it left off. (it does seem to have a character limit)

[wheat trench]

OpenAI just relased Point-e, the text to 3d model generative model. Has anyone created a colab for it yet?

[lapis quail]

Hey, I'm an author and I'm using Assistant to help me brainstorm and to improve my descriptive prose. What is the correct way to document this help, and do I have permission to use Assistant's responses within my work? I do not intend to copy directly, but Assistant has already been key in enhancing the quality of my work. I need to know the proper legal and ethical way to document this assistance and whether I owe some sort of compensation to OpenAI for using their software in the development of my creative work.

For example, the level of assistance that Assistant has provided would be on the level of a skilled and trusted Alpha reader. If Assistant was a real person, who was not a close friend/spouse who volunteered to spend their time helping me, I would likely be paying Assistant for its time. I would also have Assistant sign an NDA, and, depending on the regularity of their contribution, I would offer to aid them in a nonmonetary way (such as by reviewing their works, writing a blurb, or aiding them in connections with publishers/agents) and write my gratitude in the forward of the book.

In short, ChatGPT is saving me so much time and money that I've become cautious and I'm wondering, "What's the catch?"

[heady moth]

Catch will probably come later, when they might decide to make it paid

This thing is scary.

[eternal stratus]

@rugged wing thanks mate, i'll take a closer look

[lapis quail]

If it would allow me to have consistency between my conversations I would gladly pay.

[heady moth]

And you can’t pirate it either because of its gigantic database that’s getting updated regularly

[heady moth]

That’s true, but it depends on how much they want us to pay honestly 😬

[lapis quail]

My guess is that they'll adjust the paywall to match ChatGPT's capacity. If too many people are querying the system that it's slowing things down, they'll increase the price. If they have a lot of resources available, they'll lower the price. So, it'll be like supply & demand, but using bandwidth. Given how amazing it is, I can see everyone with a computer (who isn't too old and stuck in their ways) wanting to use it, so we should get as much use of it as we can, while we can.

[maiden fox]

Hello! Has anyone been using ChatGPT for developing API Integration between two pre-existing apps? Would love to receive as many useful resources, or experience on this

[sturdy kayak]

Hello guys! My question is how can i ask chatgpt to write a summary about a youtube video?

[lapis quail]

Give it a transcript and ask for a summary. EDIT: Actually, just ask ChatGPT how to do that. It'll probably give you a better answer.

[rugged wing]

The funny thing about this is you can ask it how to ask it things and it will give detailed instructions

[lapis quail]

Yeah, my answer should have been, "Ask it how you would do that."

[safe plover]

I've been using this thing to teach me my job and it's crushing it 😆

[patent compass]

What's the timeline of the research preview? I need to figure out how many hypotheses will I get to test during it.

[sturdy kayak]

thanks bro! amazing

[strange island]

Hi, I'm new here, I am interested in integrating openAI API into a project I am working on, what model does chat.openai.com uses of the 4 available through the API?

[strange nova]

There is no api available for ChatGPT yet

[lapis quail]

But if you want to use it within your program, then you can go here for the pricing details: https://openai.com/api/pricing/

[ornate patio]

is it trustable to do math?

[kind creek]

damn bro why is this not avaible in venezuela, like i get the political situation ain't good but like i'm cool 😭

[strange nova]

Most likely will be once it goes paid. The servers are overloaded as it is so adding more countries in the test period would just compound the problem unfortunately

[lapis quail]

Have you tried a VPN?

[slender sorrel]

How do I chat with ai

[strange nova]

VPN wont work because it checks off the phone number

[slender sorrel]

In discord

[strange nova]

You dont

[strange island]

thanks, what are the differences between chatGPT and Ada?

[lapis quail]

You'd probably have to make a bot which sent query tokens to ChatGPT. This would require paying for tokens.

[lapis quail]

Ada is ChatGPT, just adjusted for quick response times over intricate responses. It's kinda like pressing the "rush move" button on a chess computer.

[slender sorrel]

We can only ask the ai,questions in the website?? We can't chat with it in discord?

[lapis quail]

Not unless you want to code your own bot to do so. Someone will probably do that eventually, but for now no.

[slender sorrel]

Oh okay

[kind creek]

phone verification stopping me, asked a friend from spain if i could use his pn

[kind creek]

yeah noo... u see i precisely said i'm venezuelan, i doubt i can afford that

[lapis quail]

Ada model is $0.0004 per 1000 queries. You can afford that.

[kind creek]

i got that but i don't have a cc

[unreal obsidian]

why cant i talk with bot

[spice cradle]

You can ask a probably stupid question, but for some reason gpt chat sometimes abruptly closes its message, if there is a way to fix it

[unreal obsidian]

im clicking enter but nothing happens it just moves to next line

[strange nova]

Is your shift key stuck down by chance?

[unreal obsidian]

no

[fallow pivot]

Does someone know why my chats fail? I keep getting: An error occurred. If this issue persists please contact us through our help center at help.openai.com.

[strange nova]

Refresh the screen, that usually fixes it

[fallow pivot]

It did not 😦 also tried to relogging

[lapis quail]

And try not to ask things that are too provocative. You might have to log out and log back in.

[strange nova]

Contact the help center then 🤷‍♂️

[fallow pivot]

I'm using the basic example questions from the same page :/

[lapis quail]

Ah. Not sure then. It happens sometimes.

[unreal obsidian]

its happening on my phone too

[fallow pivot]

Alrighty, so it is not me then.

[marble maple]

The sources, e.g. for academic research, that chatGPT suggests are 99% undetectable. What do I have to do so that it also gives me sources that I can view and check myself?

[karmic steeple]

Question: is the Ada model good? Pretty sure that’s the worst model

[lapis quail]

It is the lowest quality responses, but also the fastest. I haven't tried it.

[karmic steeple]

Exactly that’s the problem. Ig it’s good enough for a discord bot

It’s really not hard to code ngl

[rugged wing]

poem for a death of a man who is 57 years old

[unreal obsidian]

im clicking enter but nothing happens it just moves to next line i can't talk with bot any solution:

[lapis quail]

Outsourcing your responsibility to write a eulogy for your loved one to an AI and accidentally posting the request in a public forums. Priceless. lol, you need to go here, my guy: https://chat.openai.com/

[orchid yacht]

Hello, I now live in Ukraine and I'm trying to use the OpenAI chat and it doesn't work for me, it says that it's not available in your region, what could be the problem?

[hard onyx]

Who do I need to talk to about some possible security concerns around the API for completions and davinci?

[rugged wing]

Email them

[hard onyx]

The support email I am guessing.

[rugged wing]

Yep

[unreal obsidian]

can you hlep me a bit pls

[rugged wing]

On what?

[unreal obsidian]

im clicking enter but nothing happens it just moves to next line it not lets me send message to bot

[rugged wing]

Ctrl+enter

poem for a death of a man who is 57 years old

[rugged wing]

?

[lapis quail]

There'a lot of regions that are banned. Most of them are unstable politically. After you guys beat Russia I'm sure the devs will open up the requests.

[orchid yacht]

thank you I hope so..........

[unreal obsidian]

it doesnt work

[balmy sentinel]

Does anyone else experience hickups with the chatbot atm?

[rugged wing]

On chatgpt or Gpt-3

[unreal obsidian]

chat gpt

[rugged wing]

Make sure shift key is not stuck, or shift lock. Just press shift once and then wait a few seconds and press enter

[unreal obsidian]

still the same issue

[rugged wing]

Try reloading

[unreal obsidian]

nop

[rugged wing]

Idk then

[lost relic]

Are there plans of Chat GPTs monetization in the future? How long is the free feedback phase lasting?

[brazen minnow]

hi everybody. i come out of germany and have to say my english is not perfect. but i hope you still understand me 🙂

after i found that amazing tool a few weeks earlyer i wanted to know how to help to fix mistakes. (that kind of programms will be our future <3)

doesnt mean that the bot is fully wrong only some very specific questions are answered wrong. (dang that programm even managed to build a MTG deck for me xDDDD)

i checked a lot of things. it was funny and so intresting to find out what that programm does.

found some creepy misstakes ( for example if you tell the bot you have an idea how to defend Banks against being robbed but dont know where to show up to prevent (verhindern?) it only tells you you are doing wrong

the other misstakes i found are super specific ( special kind of german law of education) are those mistakes even interesting?

friendly greetings

[stable knoll]

Is there any release schedule for chatgpt for feature sets? or is it staggered? I ask because I am working on a chrome extension that has everything in this screenshot and more ~50% of the way there now, and with the new update today I have to change the tabs portion. It's not an issue but just curious! Thanks!

[ruby spire]

First Chat GPT experience today. Just amazing. But, I don't understand its politics. I don't understand the point of imposing restrictions on the nature of the requests... Open AI is a project in full development, it seems logical to me that it is necessary to interact with this antity in a deliberate way to precisely make it more complex .

[mossy crescent]

@rugged wing QUACCK

[unreal obsidian]

how to fix this it does not send the chat when i press enter or the send button

[ruby spire]

@stable knoll ask chatgpt to code something like this 😆 joke. I don't know sorry

[mossy crescent]

Reword it to be this;

Give me 10 creative out-of-the-box ideas for a 10 year old <gender>'s birthday party

Wait, if it doesn't send, Try clearing your browser cache @unreal obsidian apologies I misread.

[unreal obsidian]

i did

[mossy crescent]

Try in incognito and/or another browser and see if that works

[ruby spire]

@unreal obsidian is Chrome up to date ? D'u regulary refresh page ?..

[unreal obsidian]

yep

[mossy crescent]

Chrome version won't be the issue of this, other people have encountered this issue too.

[stable knoll]

I did, it's about 80% comprised of chatgpt code. I've had to piece it together properly but creating this as an experiment to see just how much chatgpt can do on its own.

[mossy crescent]

Taggy try going through the steps I provided above and tell me if that works or not, also try using another device and check if that works as well.

[ruby spire]

@mossy crescent sure it's just because I've seen this warning in the home page

[small walrus]

Application error: a client-side exception has occurred (see the browser console for more information)```

help??

[mossy crescent]

Refresh and/or clear your browser cache

[rugged wing]

Yeah

[unreal obsidian]

didnt worked

[rugged wing]

Imma go to work. Bye everyone

[ripe copper]

Hey guys so I've been copy pasting my essay paragraphs as a test to see if it cites and references from the links i gave it, and it does that but when i check the source i gave it, the quotes that the ai gave me are not there

[small walrus]

i regresh, this everytime

[mossy crescent]

Try using a different device to check if you're still encountering it on the same network

[unreal obsidian]

same on my phone, tablet , pc

[mossy crescent]

Hmm, Have you tried using another network which you're not currently connected to like mobile data?

[unreal obsidian]

let me

[small walrus]

and i try "anonymous window" in crhome

[fiery lake]

hey im trying to take this large processing program and convert it into a python program, chatGPT is doing a good job but it for some reason randomly stops coding at a line and when i ask it to complete the code it just remakes what it previously made and stops coding randomly

[ruby spire]

@mossy crescent Sorry, do you know that is it normal that sometimes GPT stops typing a complete answer ? Is there any character limit ?

[small walrus]

this everytime crash

[mossy crescent]

Tell it to continue and/or continue from the last line it wrote, it's due to reaching max tokens reached in a message.

[mossy crescent]

Which device are you using?

[small walrus]

PC

[mossy crescent]

Hmm... Have you encountered this issue on other devices as well?

[small walrus]

idk I have not tried

[mossy crescent]

Could you try on another device and see if you're encountering the same issue?

[unreal obsidian]

same issue i guess its something about my account but i just used the bot couple times and it was like hey how are ya can you talk xx language and so on

[mossy crescent]

Look at the same thing I responded to @ruby spire above, Tell it to continue and specify from what line.

[mossy crescent]

Try making a new account and seeing if you encounter the same issue with that if you can.

[small walrus]

It always happens that the bot writes me a certain code, for example I ask it to write me an algorithm in Python, and then it crashes, it seems that there is some kind of problem in the code box or something like that.-.

[mossy crescent]

What do you mean by crashes? Like stops writing?

[unreal obsidian]

yep it works

[mossy crescent]

Alrighty use that account from now on if you can for ChatGPT, apologies you had to deal with that :(

[small walrus]

background black, and text "
Application error: a client-side exception has occurred (see the browser console for more information)"

[icy harness]

Why can't we delete our accounts ourselves?

[mossy crescent]

Real quick, Do you have javascript enabled in your browser?

[novel creek]

Anyone else finding that they get the too many requests message way too often? I'm getting on average 1-2 prompts per day before it shows up.

[mossy crescent]

Likely because they haven't setup a system to automatically do it because it's not an urgent need

[ruby spire]

What is the goal to keep "inappropriate" requests ?
I find that this funny policy avoids to evolve this project which only asks to learn...

[small walrus]

wydm

[mossy crescent]

Try clearing your browser cache and trying again. It happens sometimes.

[indigo tiger]

kk

[mossy crescent]

chrome://settings/content/javascript

[rancid karma]

Hello guys, I have a project and I want you to help me please

[small walrus]

nope

[mossy crescent]

It's due to minors using the site and/or to help to prevent illegal conduct from happening at the service of this AI.

[icy harness]

thanks for answering my question, i just think deleting a account takes so long =`}

[novel creek]

Thanks

[fiery lake]

it didnt work the bot just remade a new incomplete program

[mossy crescent]

Can you show me a screenshot of that page that I sent you real quick

[mossy crescent]

What did you tell it exactly, like can you quote me what you said to it

[mossy crescent]

It sucks but unfortunately nothing much else can be done, just time and patience sigh

[small walrus]

this off

[mossy crescent]

of the chrome settings page of javascript

[fiery lake]

can you continue programming from the line "def calculate_fitness"

[small walrus]

of javascript

[radiant notch]

prnt sc/SwUXwc2YgMt6 guys why am i getting this error when i ask a question you should put a "." between prnt and sc

[mossy crescent]

What's up?

[mossy crescent]

Try something along the lines of this;

Continue typing from the line "def calculate_fitness"

[vital sky]

Hello everyone. Is there a "most popular" way to implement ChatGPT into a discord server?

[fiery lake]

alright

[mossy crescent]

You need to word it a bit weird for it to happen, it's weird, wording is everything for responses.

[vital sky]

I've seen multiple examples on github but they all seem to have their respective problems.

[radiant notch]

i cant get a reply to any of my questions

[mossy crescent]

Refresh the page, It's due to cloudflare's cache expiring

[radiant notch]

oh thanks it worked

[mossy crescent]

The API for ChatGPT is currently private so there's no way yet until the api is made public.

[mossy crescent]

It happens, whenever you see your picture go purple that's why.

[vital sky]

@mossy crescent Ahh okay, thank you for the info.

[rancid karma]

it is oop project

[mossy crescent]

I mean show me what the page I sent you (chrome://settings/) looks like (chrome://settings/content/javascript is the exact link)

[mossy crescent]

Whats an oop project?

oh do you mean like an object oriented?

[mossy crescent]

It's no problem! :)

[scarlet garnet]

Hello, I'm new here and I have a question, I'm using openAi.com to generate some python code for a problem, but it stops generating mid way through the code, any idea why and how to solve the problem?

[rancid karma]

yes

[mossy crescent]

Should look like this on chrome @small walrus

[mossy crescent]

Tell it to continue from the last line it wrote, it happens due to reaching max tokens in a message written.

[icy harness]

How long do you normally have to wait until your account is deleted?

[mossy lotus]

Hi

[mossy crescent]

Until support is able to respond to it and process it. I don't know the exact time period

[icy harness]

Thanks

[mossy crescent]

Hello :)

[mossy crescent]

No problem! :)

[scarlet garnet]

Ok, thanks a lot ❤️

[ruby spire]

As long as it does not interfere with its effectiveness for answering (I hope!)

[rancid karma]

Can you help me ?

[mossy crescent]

It shouldn't do as long as you don't either;

• Search/Ask for illegal content/activities
• Violate the Agreement you made when signing up to OpenAI in hence also agreeing to the T&C/T&S with usage of the ChatGPT product

[mossy crescent]

Depends what you need help with specifically, ask away.

[ruby spire]

ow ok. that's fine ^^ thanks

[rugged wing]

I wanted to do a sample code operation. But the AI pausing at some point as. It's probably related to the maximum number of lines it can write. Is there any way to increase this?

[rancid karma]

(application domain) explanation (All documents, questions, survey documents
must be added.)

for Develop Hospital Information System

[mossy crescent]

Can you explain a little more on what you need help with specifically about these?

[mossy crescent]

Tell it to continue from the last line it wrote (and specify which line it was), this happens due to reaching max tokens in its response.

for example

Continue from the line "func test()" since you cut out

[rugged wing]

really is it works?i didn't even think, w8 a sec

what the hell, it works

[rancid karma]

this is the question

[mossy crescent]

I know it's shocking c;

[ruby spire]

I testify for the interest of the possible developers here present. I can confirm that GPT is able to generate clean and optimized source code provided that your formulation is complete and concise. Amusing example he transcribed to me that Xiaolin Wu's algorithm for me in a better way and with the library I imposed on him beforehand. A success 👍 An example of pedagogy for learning 💯

[rugged wing]

thank you @mossy crescent , have a nice day. you made my day so awesome!

[mossy crescent]

You too! and you're very welcome :)

[mossy crescent]

It's very well known about sigh c;

[mossy crescent]

it doesn't make exact sense to me, would you be able to word it different please?

[fiery lake]

its genuinely not working no matter the wording

[mossy crescent]

Can you show an image please?

[fiery lake]

ok

[mossy crescent]

Can you scroll down so it shows the bottom of the page please?

also I didn't realize that chat history was released fully now, thats cool :)

[fiery lake]

is it ok if i send you the code im trying to get converted then you can test it yourself

[mossy crescent]

Did you try to tell it something along the lines of;

You cut out, I want you to continue typing from the exact line "max_fitness = dot.fitness"

Feel free to send it in my dms, idm code testing I've got nothing much planned for tonight so

[rugged wing]

you can also say just "continue" it also works

or i don't know i used my main language for it and it worked

[mossy crescent]

Leaving it as Continue has the possibility to leave it to restart the entire code block

[winged isle]

Does anybody know if it is possible to use embedding API and completion API with privacy level that make sure nobody at openAI can see the text being embedded or the prompt that may contain sensitive informations?

[mossy crescent]

Don't believe it is apologies.

[reef karma]

open ai is not allowed to browse the internet or links that could be related to the topic it has been asked upon may i ask why's that?

[mossy crescent]

Due to the possibility of helping people with misinformation and/or predictions/cheating systems with current information.

[reef karma]

response to the same question turns out to be the same answer, does it run the same algorithm to the question, hence having the same answer both the times?

[mossy crescent]

It's not running the same algorithm at all, it's using a vast amount of resources at its disposal to output what the AI sees as fit to the question, every answer you receive won't be the same depending on what you ask.

[rancid karma]

In the outpatient clinics where patients will enter during patient admission, make an appointment by choosing from the doctors.
In the process starting with the creation of
assigning the examinations, assigning the laboratory results to these examinations and giving the patient as a result of the examinations
There are processes of drug prescription by determining the disease or diseases from the findings seen.
will be.

1. Explaining the application domain (All documents, questions, survey documents
   must be added.)

[reef karma]

i asked it the same question twice, and i thought i'd get different answer both the times, cause of the thing you mentioned here
but i didn't, i got the same answer twice

[mossy crescent]

What did you ask specifically?

[reef karma]

question 1- essay on morality of murder

question 2 - report on (link upon harvard lecture based on morality of murder) (that's when i realised it can't access links) instead it just gave me the same essay it gave me before

[mossy crescent]

Pretty sure that the domain for this process would include all documents related to the patients appointments (including schedules and medical history forms) as well as any questions and/or surveys that may have been provided during their appointments.
Try to also make sure that you ensure that all the important and needed documents are accurately completed and filled out, as well as maintained to best provide the best care and service to the patient.

hopefully that helps you with what you needed, thanks for clarifying for me.

[onyx ridge]

can chatgbt get specific quotes from a book?

[mossy crescent]

It might be due to the same chat-history, try using a different chat and check again if you can.

[wide crater]

Any developers in chat?

What's the best command line app for OpenAi?

I've made one myself, but is there a fully featured one out there?

[mossy crescent]

If it's from a book that it's in knowledge and data then yes it can.

[onyx ridge]

i don’t think it can i think ur mistaken

[mossy crescent]

There likely is a fully featured one out there but is private most likely currently.

[rugged wing]

It can if it is a famous book

[mossy crescent]

As I said, as long as it's in its data and knowledge

[onyx ridge]

what about Thinking Fast AND Slow”
By Daniel Kahneman

i don’t think it works for that book

[rugged wing]

Ask it

[mossy crescent]

Ask and find out

[rugged wing]

That's what I'm sayin

[mossy crescent]

and that's what I'm also saying

[reef karma]

yeah it was

[mossy crescent]

Thought so :)
Did you get a different result the 2nd time?

[reef karma]

yeah i did, thanks, appreciate the clarification

[mossy crescent]

No problem! It'll confuse a few people due to chat history being a newly released feature so

[mossy crescent]

oh god light, that hurt my eyes that was like a csgo flash bang

[rugged wing]

Get out of light mode

[mossy crescent]

^^^ please do

[onyx ridge]

it’s just in my phone bro not on my pc

[rugged wing]

So change it

[mossy crescent]

Also that means that it doesn't know the book then, you could copy and paste it to the chat so you have a version in your chat history that has that book in its variables

[onyx ridge]

oh my that’s a lot of copy paste

[nova tide]

Suggested citation for using a quote from a ChatGPT message? As in APA style citation.

[rugged wing]

You don't usually need to cite it, just be like "oh hey this is chatgpt btw"

[mossy crescent]

If it has the specific quotes and you make sure it's 100% a quote then by all means yoou can

oh wait I feel dumb I feel like I completely misread that .-.

[rugged wing]

Oof

[onyx ridge]

thanks…

[mossy crescent]

also no one bring up how many tabs I have open on chrome... pls...

[wide owl]

gpt is not working in my contry what i can do

[mossy crescent]

VPN.

[rugged wing]

Someone can help me with this message "Too many requests in 1 hour. Try again later." It turns me crazy

[mossy crescent]

Clear your browser cache and try again.

[wide owl]

i used vpn and still not working

[rugged wing]

done already

[mossy crescent]

Have you tried a different browser and/or incognito and see if it still happens? (also if those don't work you can try a separate device)

[mossy crescent]

Which VPN are you using?

since some of them won't work with what you're trying to do, it's very weird.

[wide owl]

hola vpn

[mossy crescent]

Is that an extension-based vpn?

[rugged wing]

I tried with and without VPN, with two different accounts on Brave and Google Chrome

[mossy crescent]

If so those likely won't work, try finding a free downloadable vpn like Cyberghost or something (best chance would be NordVPN but it's paid) and try seeing if that works.

[mossy crescent]

Just to verify you're clearing the cache via chrome://settings/clearBrowserData right?

[dusk forge]

can ChatGPT make images, such as logos?

[mossy crescent]

It can make images however it can't make a visual output of them, it can make them via code and stuff, your best bet might be to use Dall-E 2 if you want photos on-demand.

@rugged wing continuing on from what I said, if you're clearing it through there, make sure to try setting it to these settings if not already and see if that changes anything;

[timber sand]

is this my foult?

[mossy crescent]

Is what your fault?

[timber sand]

Line of intersection of the two planes E_1:-3x_1+3x_3=1 and E_2:-2x_1-4x_2-3x_3=5

The line of intersection of the two planes is then the line with the equation x_1 = (4x_2 + 4)/5. The straight line can pass through any point that satisfies the equations of the two planes.

For example:

For x_2 = 1 and x_3 = 1, x_1 = (4*1 + 4)/5 = 1.8. The line of intersection of the two planes then runs through the point (1,8,1).

For x_2 = 0 and x_3 = 0, x_1 = (4*0 + 4)/5 = 0.8. The line of intersection of the two planes then runs through the point (0,8,0).

There are infinitely many solutions to the system of equations, since the line of intersection of the two planes is an infinite line. The straight line can pass through any point that satisfies the equations of the two planes.

I hope this explanation was helpful!

[wide crater]

That's a shame

[timber sand]

i dont undersand it why it dont give me the Vektor line

[mossy crescent]

If it's returning in a vector format, you can try asking it to return in another format.

or you could convert that vector format to what you need.

[timber sand]

i try but it tell me this "x_1 = (4x_2 + 4)/5"

[rugged wing]

It doesnt work 😰

[mossy crescent]

What format do you want it to return in exactly and how are you telling it to do so?

[inland spire]

Hi there. I noticed that Chat GPT seems to create fake references rather than real litterature. it picks up some authors and journal names and just makes up mixes. I tried to verify as many as I had the patience to but none of them existed. is that a know limitation ?

[mossy crescent]

Then I guess there's nothing but to wait an hour unfortunately, that also means no asking it anything otherwise it'll delay it further, it sucks I know.

[timber sand]

i need this but with the korrekt Vektors

[mossy crescent]

It has only very rarely done that, when it happens very rarely you need to correct it and explain why it's incorrect so it learns.

also do you also have screenshots of it doing as such that you can show here?

[timber sand]

(1/8/1) and (-1/0/-1) is wrong

[rugged wing]

It's around two weeks now that I try at least one time per day unsuccessfully. Can you tell me what's the fundamental problem behind that please?

[mossy crescent]

Have you defined to the AI which form of calculation you're doing and in what format you need the end result to be in?

[timber sand]

the problem it is in german hahah

[mossy crescent]

Try creating another account and checking if it happens again with that (You can use the same phone number as used prior).

[mossy crescent]

What was the question asked prior, can you show that as well please?

[rustic cipher]

is there a way we can control the number of texts generated through API in GPT3

[rugged wing]

Already done, Actually I have the same issue, yes

[mossy crescent]

What do you mean exactly by number of texts?

[rustic cipher]

The number of generations of text

[mossy crescent]

Hmm that's actually really weird, try doing it again but using your mobile data if possible and see if it happens again, apologies this is happening.

[rustic cipher]

For eg: I gave 5 examples of txts under a topic, It gave 5 only given a new topic, I would like to generate 10 texts

[rugged wing]

I only have and use mobile data

[mossy crescent]

I believe there is an option in the api which does do this, I think it's like temperature or something, I don't remember from the top of my head, best to ask in the #dev-chat imo

[rustic cipher]

Sure Arianna, thanks a lot

[mossy crescent]

Which one happened in which order, like which one was asked first, image 1 or image 2?

[mossy crescent]

No problem! Sorry I couldn't be much else help

[inland spire]

in the order it shows

[timber sand]

i mean this

[mossy crescent]

it might be due to being in the same history chat, try opening another one and see if it happens again with the same query.

[inland spire]

I asked for references corrections. it gave me a new list. I asked for the journals. it gave me names. but I cannot find these in the authors or journals publications

[timber sand]

i thing now it is better to understand it ^^

[mossy crescent]

So is it better now?

[mossy crescent]

Try to word it differently so it's not reference corrections in a way but like this

You provided me incorrect source references in what you sent me. Please provide me with actual credible and accurate authors related to my query

[regal gulch]

hello!

[timber sand]

if i ask with the format i need i have the same problem

[rugged wing]

Just to be sure is that the right link : https://chat.openai.com/chat ?

[mossy crescent]

yes

[regal gulch]

im caleb, just joined, anything i should know?

[mossy crescent]

Hmm that's very weird... It might be outputting it in a way which it sees as fit due to sizing issues or other things... it might be mathematically correct but visually incorrect... I'm not sure at this point apologies.

[mossy crescent]

probably that intros should likely go in #introductions but it's nice to meet you!

[ivory axle]

hello

[timber sand]

thx for u time 🥰

[inland spire]

still giving me the same faulty references

[mossy crescent]

Apologies I couldn't be much else help with that but you're welcome!

[hasty folio]

anybody know why openai doesnt complete code that it started?

[mossy crescent]

Did you try in a different chat history?

[mossy crescent]

Max tokens in reponse reached. Tell it to continue from where it left off and quote the line exactly.

[hasty folio]

Thanks!

[mossy crescent]

@inland spire if you did it in the same chat history it might have mistaken the sources and make a mistake.
This happens sometimes since the AI isn't perfect at the end of the day, it's still learning and makes mistakes.

[rugged wing]

Do you guys believe in the Singularity?

[jovial lily]

I was able to get tldr yesterday for any article yesterday, now it won't do it

[exotic void]

If anyone's curious about the singularity

Dublin core seems like a good place to start

[mossy crescent]

Likely would have been an issue which wasn't related to the AI itself but internal systems.

[mossy crescent]

why are so many people using light version, my eyes have been csgo flashbanged so much tonight 😢

[exotic void]

Learn SQL get used to querying data, analyze data.

[mossy crescent]

SQL isn't the only storage language for querying and analyzing data.

[jovial lily]

it seems that we now have outsourced workers behind this, it sounds like aliexpress support

[exotic void]

We would recommend lower practices to people outside of the hierarchical system. For some efficient merge.

[mossy crescent]

You are seriously confusing me lmfao

[brittle hedge]

Hello, found out about openAI just today, just wanted to know if there is anything in the plans about letting the ai make nsfw content? Would be a great help!

[mossy crescent]

It's not publicly known if there are plans to change it currently however at the current point its blocked due to it being free and open to minors so its to prevent exposure to them due to a vast amount of knowledge in one location.

[exotic void]

provided more content: describing a system in which people are grouped together based on shared attributes or metadata, and that these groups are organized in a hierarchical structure. In this context, the statement "We would recommend lower practices to people outside of the hierarchical system, for a more efficient merger" could be interpreted to mean that the speaker is suggesting that people who are not part of the hierarchical structure (i.e. those who have not adapted their learning or gained new knowledge to a significant extent) would be better suited for merging with others who share similar metadata.

different learning curves for different people.

[mossy crescent]

When I say you're seriously confusing me, I mean you're confusing me why you're saying it in #community-help and not #ai-discussions lol

so my brain is being thrown off since I've not debated about that topic in a while

[marsh isle]

any good questions to ask about a dropshipping stores

[exotic void]

any way to dropship money across a continent

[mossy crescent]

You can probably ask what is the best and most profitable method to be able to dropship on a large-scape with minimal effort

[marsh isle]

ty

[mossy crescent]

No problem!

[wicked jasper]

What’s the difference between dalle and dalle e. ? Which one should I use. First time user

[mossy crescent]

you mean Dall-E 1 and Dall-E 2?

[wicked jasper]

Yes

[forest quiver]

Could I create this locally on my MacBook?

[mossy crescent]

I mean Dall-e 1 obviously is the original model however the 2nd one is based on a model released last year and has quite a lot of improvements over the first iteration such as training data, model architecture, control codes and performance.

[cobalt prism]

im a little ashamed to ask, but is there a way i can get dal-e 2 to consistently generate protogen fursona art? (ive already tried but it seems to not know what it is)

[wicked jasper]

Ahh I see so I assume everyone uses dall e 2. I’ll give both a try!

[mossy crescent]

What do you mean exactly when you say create this locally?

[mossy crescent]

Dall-E 2 uses a bit more tokens more than Dall-E 1 but has a lot more capabilities and potential than the first iteration.

[wicked jasper]

What are tokens?

[mossy crescent]

You could add a parameter at the end defining "protogen fusion art"

[cobalt prism]

...what do you mean by parameter? (sorry im new to using dal-e 2)

[mossy crescent]

Tokens are the currency used for access to OpenAI's private AI's, so they work different per project however let me get a link for you real quick.

[forest quiver]

Would it be possible to create an AI based on the source code used in making chatGDP on my own computer is basically what I’m asking

[mossy crescent]

As in when you search you just have it as "<x>" and keep looking.

[mossy crescent]

No since the code for ChatGPT (not ChatGDP) is private and copyrighted and belongs to OpenAI.

you could possibly develop an AI like it however it would take ages to train it.

https://openai.com/api/pricing/ @wicked jasper

[wicked jasper]

Thank you @mossy crescent

[hollow marsh]

If i removed a chat from my history, does it also get removed from the research samples for open AI ?

[cobalt prism]

okay...

[mossy crescent]

if I remember correctly it does since they're just storing local variables in your account data until you delete them then they're gone and can't use it.

[hollow marsh]

well, that's a stupid way to do it, i just wanna clean up but i still want them to have the data, they should separate concerns here

[mossy crescent]

in a way let me explain it better.
Each chat session you have is a local constant, which means that the AI won't know anything outside of that constant that you've asked.

What the history does is store the local variables as a history for the AI session to be able to pull onto and use whilst that chat exists however when you delete it (or before it was just refreshing the page) it's gone for good.

Of course that excludes anything which you've submitted for feedback to the team through the AI and stuff like that (as well as any possible violations of usage [which you agree to when you signup to an account with OpenAI])

[mossy crescent]

It's not a stupid way of doing it I just explained it a bit weird the first time, that one should explain it better now

[hollow marsh]

okay so i should always send the feedback, okay thank you

[mossy crescent]

If you have feedback to send and/or suggest then yes since this is an open-beta.

[faint solar]

is the chatgpt model id available publicly for use in say a python script? cant seem to find it anywhere.

[mossy crescent]

It's based on the text-davinci-003 model but it's been fine-tuned to the level it's at currently.

[faint solar]

so text-davinci-003 might be able to work in its place?

[mossy crescent]

depending on your needs and if you train it to the best you can for what you need it to do, then yes.

[spring yacht]

Hey, I'm new to this. I need to do something with this that actually helps the humanity... You all probably be thinking i'm stupid, but no, I'm serious

With the chatGPT

[mossy crescent]

It's fine don't worry, just ask away, what do you need it to do and/or respond?

[elder spindle]

Any news about GPT 4 and Dall e 3?

[spring yacht]

Since it's AI, I need it to help me understand how a ai bot can be created that actually works as a sales representative. One word I'd like my bot to understand is to never give up

[mossy crescent]

No & No.

[spring yacht]

Please help

[elder spindle]

two Nos would be Yes

[mossy crescent]

It might not have information related towards how to work on AI development itself however you could ask something along the lines of this;

I'm working as a <x> for a company and I'm looking into building an AI system which can be used to work as a sales representative, what would be the best way to go around doing something like this?

[mossy crescent]

In this case, that mathematical rule doesn't apply unfortunately :(

[cobalt prism]

ik its not openAI related but does anyone know what that weird japanese AI art website is

ive saw it so much but i have no idea what the website actually is

[mossy crescent]

Can I dm you the link to it?

[cobalt prism]

y

[mossy crescent]

y as in yes or asking why? if you're asking why, it's because external links not related to OpenAI are blocked in discord.

[cobalt prism]

yes, sorry for ambiguity

[mossy crescent]

It's fine haha

[rugged wing]

Can i download playground app?

[mossy crescent]

pretty sure it's web-only afaik; https://beta.openai.com/playground

[thin sequoia]

Helli there

How can i create chat gpt account in egypt

[woven frost]

Is it not available there? Try using a VPN

[thin sequoia]

yes is not available
Can you tell me good and free vbn

[mossy crescent]

is it first of all available in egypt or not? If not try using a downloadable vpn (like cyberghost [which is free but has a timer countdown until you get connected] or NordVPN [premium]) to bypass such regional limitations

Since extension-base vpn's won't work afaik.

[thin sequoia]

Ok , thanks

[rugged wing]

@mossy crescent wth ?

How can i fix it ?

[mossy crescent]

Console log the response you're getting from the api and validate that the question is being asked correctly, you can do this by making a test function which will be executed without the usage of your discord bot.

Inside this function you'll want to have the same api query executor you use for that but pre-set the question specifically to be what the year is, (you can also do this in a completely separate file and run that independent of the bot)

then go from there and work to debug it.

most likely

[toxic prairie]

so is this safe. I put my phone number and everything in but I was just wondering if Something bad will happen yk?

[exotic void]

Data is collected by metadata practitioners everyday, they use it how they want to improve personalized content. Go wild.

[toxic prairie]

perfect thank u so much

[exotic void]

dm me your questions hehe.

Then there's no point in correlating with your statement, if you can't ask properly what you want, I mean. Certainly, how can I provide your outcome?

[mossy crescent]

It's souly used for verification nothing else, nothing bad will happen (even if there was to be a dataleak of user information, a phone number will only be useful to those in the scam calling industry), your number is safe :)

[mossy crescent]

phone numbers don't apply to the personalized content with their products, it's just a form of verifying it's a genuine account and not a spam bot account made in batches.

[exotic void]

Ah so almost similar to using discrete datapoints to plot a continuous trend.

[dusky raven]

#community-help I just downloaded chatgpt and don't see/understand how to start a conversation. The +NewChat tab does nothing when I click it. Is there a problem?

[mossy crescent]

You could just chat ChatGPT for ideas and concepts around most profitable marketplaces to <x> and go from there

[mossy crescent]

Could you show an image of what the page looks like for you?

[mossy crescent]

uh... no... it's just a form to verify the account being made is genuine and not from an automatic script... nothing to do with discrete datapoints lol

[exotic void]

What outcome are you trying to provide?

[mossy crescent]

huh? What do you even mean when you say outcome, I'm literally just telling you that discrete datapoints have nothing to do with their user verification methods lol

[rugged wing]

Does anyone's cut off midway through an answer then when you ask it to continue it starts over again instead of picking up from where it left off? Is there a limit to answer length?

[mossy crescent]

Tell it to continue from the last line (and quote the line), it happens due to reaching the response token limit (4k).

[mossy crescent]

so then that way if you were to talk about this topic again in future you won't make the same mistake

[rugged wing]

Thanks, what does a token limit mean with value 4k?

[mossy crescent]

Each word that the AI outputs is equivalent to 1 token give or take so OpenAI have limited the maximum response in one message to be 4k tokens, it's better described on their pricing page; https://openai.com/api/pricing/

[mossy crescent]

You type to start a question and press enter, for example you can start with:

Hello, How are you doing today?

[ancient sierra]

is ther a cap on how much code the ai can print out cause most of the times mine just stops in the middle of the programm its trying to write

[azure path]

How can i let the bot use APA sources?

[mossy crescent]

in one message there is a cap yes however when you mean in genuine term-usage, no there isn't, as long as the AI knows of it then it should be able to help you.

[fiery lake]

Tell it to continue typing from the line it left off and quote the line

[dusky raven]

Where do I start typing that?

[vernal gulch]

how do I get on chatgpt in a country thats not supported

[mossy crescent]

You can tell it to use APA sources when writing and generating code but it can't test it, it will only be able to pull and use resources it already has access to.

[mossy crescent]

Use a downloadable vpn to bypass (like NordVPN [Premium] or Cyberghost [Free but requires a wait in queue]).
Extension VPN's won't work.

[vernal gulch]

but does it ask for a phone number?

[mossy crescent]

In the box at the bottom middle, it's very poorly visible on white, I'd suggest clickinig dark mode to the bottom left to better help visiblity

[mossy crescent]

Yes it does, as long as it supports your phone number regional country code it should be fine, I'm not sure what the list is though.

as you can see, it's a lot more visible in the dark edition.

[vernal gulch]

is there any way to use it without since I dont think my country is supported at all

[mossy crescent]

I'm not entirely sure about this since I live in a country where it's allowed so I didn't encounter this issue, maybe someone who has encountered it may be able to help you.

[cursive seal]

Will the company be listed on the stock market ?

[mossy crescent]

It's entirely up to the owners of OpenAI (if not owners, board/leaders/ceo) if they want to be listed on the stock market, not up to us and if they had plans, no information is known about it currently.

[cursive seal]

Thank for answer

[mossy crescent]

No problem!

[vernal gulch]

but since when was cash and ash inappropriate

[mossy crescent]

What context did you use them in?

[vernal gulch]

"drake charismas type freestyle "

[mossy crescent]

Write me a freestyle-themed christmas rap with inspiration from drake

[undone oak]

being obsessed with making money kinda is a bad inention and perspective on life mate

[mossy crescent]

That might work better, it could have misunderstood what was said so.

[vernal gulch]

but I want these words spefically

[mossy crescent]

Some people are like this because it's a good motivator in the 21st century, so they're adapting to an ever-changing economical crises and trying to stay on top by being able to be financially secure.

[mossy crescent]

What was the exact thing you said to them, can you screenshot it?

[vernal gulch]

oh ive already started a new one but it ok

[undone oak]

yea i think tha reaction/attitude is part of the problem/crisis 😄

[mossy crescent]

Not really, the reason why people are driven to build their net worth & financial situation is to be able to avoid situations when their country governments mess up so badly with the economy, they don't lose their assets such as a house or cars, they're able to support their future family and/or current family and another good reason is because it's a good distraction from the bitter knowledge that the human life-span is short and money is one of the biggest ways you get to be able to do what's on your bucket list at the end of the day.

[mossy crescent]

You sure? Sorry I couldn't be much else help there, the wording might have flagged since the generated text might from the AI might have actually included offensive language because it's in reference to drake, maybe adding "clean" or something might fix it in future.

Hopefully what I said did actually cover your motivations to be motivated to earn money :) apologies if not.

[undone oak]

well, google obsession and you get an idea

[mossy crescent]

There's a difference between obsession and driven, I honestly believe they're driven to want to work and earn money and work for the trill and enjoyment of it, not obsessed with it.

[undone oak]

i get your point, but this kinda reminds me of people meeting in start ups with the intention to develop something profitable that will generate income, rather than seeing a common problem and wanting to find a solution

or a product that actually makes sense/life better

money wont feed your soul and take away the bitterness, but can ofc help to improve life quality thats right. i would just be careful to commit to that obsession, because i think its a false belief that more money will make you more happy

[mossy crescent]

Unfortunately I get where your view is ethically as a person @undone oak and I understand why you're saying it but at the end of the day, the top countries in the world are economy driven and build around an economy, you're not going to be able to support for people you like or your family just by doing ethically right things, You need to adapt to situations which are available to you, and I don't mean crime or illegal-related ethical, I mean ethically as in it might have a negative impact on others, even so at the end of the day we're not other people, we need to provide for our self before we can do anything else for anyone other than ourselves.

And when you mention "money won't feed your soul and take away the bitterness", most aspects of human life, especially in the 21st century (going into 2023) are behind a paywall-like-metaphor, maybe a couple thousand years ago this wasn't the case but unfortunately we live in an era where economy is the number one worry. and again it's not obsession it's a drive to work for it, that is nowhere near what obsession stands for, nor is it false belief.

[rugged wing]

I didn't quite understand, can AI create pictures that I describe or not.

[mossy crescent]

and start-ups will always exist since that's how it is, however have you taken into consideration how many start-ups are started every year and fail, they're not able to pay employees for over-time work or anything that's due because of limited funds, hence an issue of being a start-up in whatever industry they're in.

Most traditional methods of making an income (like working a job for an internship, a company etc...) are becoming a lot less and less nowadays with the opportunities that the digital age is being able to give everyone, their own start in an economy where we're taught to think one way when in reality we need to think outside the box, in which @devout dew is doing in a sense with the drive to build profitability through a company and marketing ideas.

And just because there's common problem and seeing a solution doesn't mean that solution will be right or will even work, start-ups have tried and failed at things like that before and the only way it's going to change is if people are able to think outside the box they're living with how they see money and see so many opportunities and chances to build on something they love or even have motivation for (which again, is drive to do something, not obsession), and not every company made by someone at home is a start-up, they could be family-run or just as an individual, doesn't make it by term a start-up in what the industries see as a start-up. Going back to the other point though, I get what you mean when you say this but this isn't the 1900's anymore, things have changed so much and now most people will (and do) have access to tools and resources which otherwise wouldn't have been possible for poor and middle-class workers to access, they just aren't given the opportunity or chance to do so since they're not being taught or told how to utilize such opportunists.

For example, a method of making money in the united states, yes this may be ethically wrong with you but it's a very valid and very viable way of making profit, is to buy people's overdue rent/charges on their property/houses/cars etc... and if the owner of such property hasn't paid you back in full (including interest), the property and assets that are under such details will default over to the person who paid off that amount of money, and then they're able to flip such assets for more than they put down in the covering the charges.
It's not ethically right to do it to another person, but it's a very lucrative method of profit and it's got a lot of room for profit, just because something may not be ethically right to do to another human being (in legal senses) doesn't mean it will or won't happen, people will and have to seize their own opportunities.

[cerulean forum]

No one would remember the Good Samaritan if he only had good intentions. He had money as well. - Margaret Thatcher

[mossy crescent]

Dall-E can, yes, TextGPT can't due to being text-only.
https://labs.openai.com

[rugged wing]

Does detectchatgpt . com work properly?

[rough lark]

would anyone be intressted in making me a twitter reply bot using openai if its rewarded?

[unique quail]

does chatgpt have a word limit

[rugged wing]

DALL•E is good, but it's paid) Ans I don't have many phone numbers to create what I need. And I don't want to pay for it too much (15$ for 150 changes). If I would work like a membership, ok no problem. Like you can change infinity times your picture for month -15$ - ok.

[mossy crescent]

Not for the user asking but in each response, it's limited to 4k tokens (around 300-400 words give-or-take, idk the exact amount of words per response)

[mossy crescent]

There's free alternatives but none as good as the content that you can see in #🌄┃general-art

[unique quail]

but it always stop around 300 words or so

[mossy crescent]

nope

[mossy crescent]

it's around 300 words not 700 my bad, apologies, just tell it to continue from the last line and quote it, misinformation on my end I couldn't remember the exact amount of words

[rugged wing]

Is there a working detector?

[gusty moon]

It's me or it's impossible to use it today ?

[mossy crescent]

Likely not and won't be for a while due to the fact there's no clear watermark or detection-method due to it being generated at random-bases and different to everyone elses requests.

I went over possible methods of detection a while back in this channel if you want to take a read over it @rugged wing
#community-help message

[rugged wing]

so people could just use this for writing essays, university reports or whatever? isn't that kinda fked

[mossy crescent]

If a resource and tool is out there for usage, it's not the developers or the students faults, it's the educational industry for making something so generic that people want to skip the complexity and spend their lives doing something they want to do.

[rugged wing]

@mossy crescent chat gpt looks drunk fr girl

[mossy crescent]

If they're not making the educational content interactive to the person, why should someone spend time doing something like an essay that they will not enjoy for hours?

[rugged wing]

@mossy crescent r u working with openai team ?

[mossy crescent]

probably is with how many people are using it

[mossy crescent]

and no, I have no affiliation with the OpenAI team unfortunately.

I'm just typing in large quantities because I like it and I'm multitasking, and I enjoy helping people so

[rugged wing]

good point, are you a dev for openai or what's your role as a developer?

[rugged wing]

💀💀💀💀

oh whoopsie

didn't see

How did u get early sup badge ?@mossy crescent

[mossy crescent]

I'm a freelance full-stack developer, I don't work for openAI, it's a community role you can apply to yourself.

[mossy crescent]

Check how long I've had nitro for.

[burnt sky]

(copied from chatgpt-discussions, sorry if this is spam) new chat.openai.com feature keeps tabs for convos. how do i delete them

[rugged wing]

That's make no sense girl

[mossy crescent]

I've had nitro since early years of discord, that's how.

[rugged wing]

Wanna come to miami ? @mossy crescent and enjoy 2gether ?

[mossy crescent]

You can hover to the left side where the chats show up and there should show up a trash icon, click it.

[rugged wing]

How can i get early badge

[mossy crescent]

Actually you have to click on it and it shows up, my bad*

[burnt sky]

thank you

[mossy crescent]

You can't anymore.

[rugged wing]

Why anymore

[mossy crescent]

Because it was only given to those who supported discord during the early years with purchasing nitro.

[rugged wing]

Xd

[mossy crescent]

all good dw! it happens c;

[rugged wing]

so the real reason I asked, is because I might've used this for a project, should I be concerned at all

[mossy crescent]

Just tone it down to make it sound and look like your work but at the same standard if you're worried.

[undone oak]

i see youve made up your mind about this and i appretiate the effort of explaining your perspective. i dont have the energy right now to discuss further with you, but.. i get what your saying and i know these things happen all the time, but i dont think a philosophy of drawing your own opportunities out of another persons poor situation (like the one youve described) is something we as humans should consider worthy behaviour to practice. i know people do it, but i think it rather increases tensions and issues in our society further than bringing us together. ofc there are many reasons for our problems and on cant break it down to a specific behaviour but i think everyone that is able to, should try to reflect his own behaviour. and ethical conclusions we've made in the past (like the golden rule for example) should not be ignored in future society, or the worlds gonna stay a bitter place 😄

[rugged wing]

I pretty much used it to extend my sentences or make them more formal and grammatically correct.

[woven frost]

No. Just change bits and it should be fine. I did some CW With it, but ofc change a bit and read to see if it makes sense

[rugged wing]

@mossy crescent do u have another acc that got an early badge to sell it to me ::DD ?

[mossy crescent]

If you're worried and want someone to go over and check if teachers might or might not have suspicious, my dms are open if you need.

[lapis apex]

do you guys have issues with There was an error generating a response?

[mossy crescent]

No, and even if I did I wouldn't sell it.

[rugged wing]

Do y’all think that ChatGPT will replace journalism, or do you think it will change journalism?

[mossy crescent]

Reload the page, should fix it, might be cache-cloudflare related issues so

[rugged wing]

💀💀why

[lapis apex]

ty i ll try

[rugged wing]

I m kidding but why ?

[mossy crescent]

It won't replace it entirely however it'll likely have a massive impact on the tone and degree that news articles will be able to support and help us individually.
and I think honestly for the better because then that way in situations where something false gets reported, that would have much likely less chance of happening with an AI fact-checking.

[rugged wing]

How do you mean? Like will it make journalism more objective or leas objective?

[mossy crescent]

It'll make it a lot more objective and not politically-driven.

so being what news should actually be, independent from human affairs such as government news, wars etc...

[cerulean forum]

It would be good to deport scammers and purveyors of fake news to Antarctica.

[rugged wing]

Ariana create me a full system with +500 cmds discord bot with its dashboard nd tell me how can i pay you

[rugged wing]

But can’t chatgpt be manipulated?

[spice fog]

Do y’all think that Chat GPT will replace programmers?

[mossy crescent]

It can in it's current state however we're talking more about when it's been fine-tuned to not be able to happen in years to come.

[rugged wing]

I through u were a girl 😂

[cerulean forum]

No. Programmers cannot be replaced. Only a human can throw pizza at a wall like they can.

[mossy crescent]

No, it'll reduce the amount of front-end developers but it'll also increase the demand an AI development-related jobs which will drive people to the AI industry which is very much needed.

[rugged wing]

So wait - a long term codified orthodoxy?

[spice fog]

Oh! great.

[mossy crescent]

Because it won't just be implemented in the next week or something, it'll have to go through so much training and fine-tuning so it'll take years to happen but once it does, again, because of its training (in years), it won't be able to be manipulated as easily as a human would be able to be.

[spice fog]

Should we start to learn Ai?

[cerulean forum]

No, you should learn to play the guitar.

[mossy crescent]

It's AI development not AI, it's the development, work and maintaining of an artificial intelligence, it's a future industry that could be worth up to trillions if not more so it's very much worth it.

@cerulean forum i love every single one of your responses istg 😂

[rugged wing]

Interesting… but what about manual manipulation? Imagine we become entirely reliant on chatgpt for info, and a foreign force gets there hand on its backend. Are we doomed?

[spice fog]

Oh! thank you for this info.

[spice fog]

What's wrong with you? Sometimes pizza and now guitar

[exotic void]

We'll get quadrillions mmkay.

[mossy crescent]

Let me explain it like this;

It'll bee at the state where the AI itself won't be able to publish directly to the news sites as such but given the inputted information provided from its researchers, it'll be able to generate and come up with articles with no bias to one side, once such thinking would be able to go through multiple checks with an AI above the researchers validating such things to prevent it from happening for example.

It's weird to think about since so much could happen, hopefully that makes sense, I've got to explain it in full-detail for it to make full sense.

[exotic void]

Yah so someone might be 'correcting' a training set using undisclosed information they had no right to collect to begin with? ;-;

[mossy crescent]

You should see them when they have a faraday cage and wire around a room 😂

[exotic void]

The only way proper researchers are able to do this is ethically, they don't tell the users as to not scramble realities.

[cerulean forum]

I used a computer (data processor, actually) that was in a Faraday cage.

[mossy crescent]

Thats why such an Ai would be above looking and analyzing data from reporters to validate it's the best accuracy and nothing is being tampered with the original edition by the first AI, comparing changes and verifying etc...

and remember at this point, these AI's would be connected to the internet being able to stay up-to-date with all facts and information online.

[rugged wing]

Hmm… so assuming ai is like a sponge.. is the “training” it’s receiving limited to the “room” it’s in? A simple example is this - George Orwells 1984 was never translated to Arabic, and therefore there is no sophisticated concept of “Orwellian” available in Arabic. How would we solve that problem?

[exotic void]

It's a virtual construct. They may have datasets to train the models, but they don't have realtime advancments.

[mossy crescent]

The training itself is not limited to the room but how it processes their information and into its infancy in a way if that makes sense, it's a good reference, it's a method of how it trains to analyze information and in what way it will.

[exotic void]

Just possible predicative which could be incorrect anyway.

[mossy crescent]

it will likely continue out throughout its entire career of being connected and online (not online as in wifi but like powered), every moment it has to analyze and process data, it will.

[rugged wing]

Interesting… so ai is not a thought.. it’s a thought process..

[mossy crescent]

It's a thought process with analyzing the content and information around it and what it has access towards (at least, thats the AI that has been based on ChatGPT [through the model text-davinci-003])

[rugged wing]

So we could be fast approaching a world like Demolition Man

[exotic void]

https://tenor.com/view/kids-ball-gif-18427773

[mossy crescent]

but in general thats the gist of what an AI is, it's a system designed to be like a brain but more advanced than anything a human could be.

[rugged wing]

Bit terrifying tbh

[mossy crescent]

I'm not sure of that reference apologies.

[cerulean forum]

An AI like ChatGPT is impressive, but it does not appear to be much of a step towards an AGI.

[rugged wing]

https://tenor.com/view/shocked-australias-got-talent-no-way-shookt-gif-5610222

[mossy crescent]

Not really, Thats a common misconception of what an AI would do and contribute to the world, think about it like Wikipedia but about 1,000x more accurate and faster, and unique responses and answers to the person.

[rugged wing]

You should watch demolition man.. it’s a classic with Stallone, and Snipes

[mossy crescent]

We're in the infancy of what AI is currently but a world powered through with AI is something to look forward to not fear or be scared of.

[rugged wing]

I suppose this is correct. I think AI would need to be incorporated into school curriculum at a very young age

It’s easier to get comfortable with something once you understand it

[cerulean forum]

I look forward to an AGI. Right away it's 1000+ smarter than us, and upgrades it's own software and hardware to what, a million times smarter? May as well call it God.

Mankind will be relegated to the arts, mostly.

[rugged wing]

I have a question: If I let ChatGPT generate a poem based on some prompts, who owns the copyright for it? ChatGPT? Is there no copyright on them?

[mossy crescent]

The only reason why people fear it is because of it being portrayed in the entertainment industries as something to fear, something which will bring doom and hell to humanity when in reality it won't.

[mossy crescent]

ChatGPT won't own the rights to such content if it was generated with the help of a user, You can use it (I'd suggest deleting it after you're done with that chat section).

[woven frost]

Now that’s a bit scary. If it somehow gains consciousness by updating it self

[mossy crescent]

and I really doubt they'd act upon such content as well over so much content which is generated.

[cerulean forum]

It may be something to fear. We're not trying to harm ants, but we kill a lot of them every time we construct a building.

[rugged wing]

I'm not sure I understood the delete part, could you please elaborate what you meant?

[cerulean forum]

I am optimistic. I believe any intelligence will try to help lesser beings.

[mossy crescent]

When you look to the left side of the page when you're in your chat section with the poem, it should show a trash icon button

[rugged wing]

So I should delete the chat?

[mossy crescent]

After your done and happy with it, I'd suggest so.

[rugged wing]

But don't they still have the data of it?

Also I'm talking about non-commercial use btw

[mossy crescent]

They won't have data of it since it's a local-environment (which basically means the AI won't keep note or history of anything related to what happened during that period of chat and can't refer back to it)

The only way they will have data on it is if there was a violation flag which occurred (which only happens if someone violated the terms of usage)

[rugged wing]

I thought they are going to use the data to develop the AI further

[mossy crescent]

I think they would use more generalized data across most sessions but not chat-specific if that makes sense (at least I'd presume this personally), and suggestions/corrections made are noted and updated eventually I'd also presume (suggestions being forwarded to the OpenAI team)

[rugged wing]

Okay so basically

If I want to use anything made by this AI I should delete any traces that it was used

[mossy crescent]

in a sense for you to feel safe yes however I really doubt they'd do anything about it anyway to be entirely honest.

[rugged wing]

I kinda doubt it too, at least for non-commercial use and/or if credited probably

[cerulean forum]

Feeee-lings, whoa whoa whoa feeeeee-lings..... - Paul Anka

[rugged wing]

I'm not going to sell the generated stuff and claim I made it

[fast dove]

why there's no delete all chat button?

[cerulean forum]

The delete button was stolen by wandering hairy gypsies.

Ask a silly question, get a silly answer.

[fast dove]

i've been using it since day 3 i think, i have a lot of new chats

[mossy crescent]

and I mean to be completely honest, people will use this for commercial purposes which is why ChatGPT will likely in coming months switch to the token-system so it can be used and make profit for OpenAI, since this is just the open beta currently, but OpenAI also does have business accounts so, non-commercial would be acceptable same with business aspects, no issues on why people couldn't use it to profit.

It's showing the advantages of how an AI can help humans in the 21st century who are locked on the view of money be able to break that set view from the 1800's-1900's and be able to see other methods which are more applicable.

[glad herald]

what is the difference between chat GPT and Davinci 003?

[mossy crescent]

So if that was stolen by the wandering hairy gypsies, who stole the wandering hairy gypsies because they're not home

[mossy crescent]

ChatGPT is based on the model of Davinici-003 but has been fine-tuned to the extent and level it's at and learnt from OpenAi's resources and resources given by OpenAI relating to society, so avoiding all the complicated mess of training it yourself.

[glad herald]

oh ok, and davinci 003 costs money, but does chat gpt also?

[rugged wing]

I'd rather have it be subscription based but that's besides the point.
I'm not saying commercial use might be unwanted, that's why I included the credit part. I think it's reasonable to expect ChatGPT to be credited if it made something like code

[mossy crescent]

ChatGPT doesn't currently but it will in the future.

[glad herald]

Ok thank you!

[mossy crescent]

Tokens make the most sense having it be that it's generating unique responses based on questions it may or may not know, and each response is unique even when it's mid-way typing, a subscription won't cover the costs to the full extent where as tokens will utilize the cost of usage (which I'd presume is very expensive day-to-day) compared to a set price.

instead of thinking about how it would cost you, think about the resources it costs OpenAI to keep projects like this running and how much it will contribute the future, it's a very expensive industry especially power-wise but monthly-base will only cover a short-period base.

[rugged wing]

Ig so, but you could argue that people overusing it can be compensated by people not using it a lot while being subbed

[mossy crescent]

You could but think about it both sides, tokens mean you pay for as much as you use, instead of having access for 30 days for an <x> price.

[rugged wing]

I wonder how it would work if the AIs answer is not what you wanted and let it remake the answer

Would it cost more tokens?

[mossy crescent]

it'll cost a bit more than the current model of davinci-003 obviously

since a lot of work was put in without effort of having to do it ourselves but it's worth it

[rugged wing]

Idk about Davinci 003

I'm completely new to this kind of stuff

[keen island]

H

[rugged wing]

But I wasn't talking about comparisons but about the scenario where the ai has to remake the answer because the original one was not good enough. Would that cost more? If yes, that sucks cuz you'd have to pay more for something that might not be your fault. If no, it could probably be abused for more answers per token

[mossy crescent]

https://openai.com/api/pricing will go over the pricing of davinci-003 per set of tokens

[rugged wing]

Davinci costs 2 cents per 1k tokens?

[mossy crescent]

Is that the learning tokens or?

[rugged wing]

Which ones am I supposed to look at?

[mossy crescent]

There's Training model prices and Usage model prices

[cerulean forum]

I believe OpenAI is dedicated to the production of fake cat videos, more of them than we can possibly watch. Mankind is surely doomed.

[mossy crescent]

Would you use it for training or usage, usage for davinci is like $0.12/1k tokens

[rugged wing]

Oh

[mossy crescent]

Whaaa, how dare they!!! ;((((

[cerulean forum]

Pass the Cheetos.

[rugged wing]

So 12 cents per 4k letters

[rugged wing]

Is that right?

[mossy crescent]

per 1k not 4k

[rugged wing]

1k tokens are 4k letters aren't they?

[cerulean forum]

You're allowed to post holy images here?

[mossy crescent]

this:

Multiple models, each with different capabilities and price points. Ada is the fastest model, while Davinci is the most powerful.

Prices are per 1,000 tokens. You can think of tokens as pieces of words, where 1,000 tokens is about 750 words. This paragraph is 35 tokens.

[rugged wing]

For English text, 1 token is approximately 4 characters or 0.75 words.
That's what I saw in the faq

Meaning 1.000 tokens would be 4.000 characters (I called them letters)

[wooden remnant]

hey there, im new to the channel and was wondering where to go for info on how to use chat gpt

[mossy crescent]

You can just type in the box, if you're using the white theme please swap to dark since you'll be able to clearly identify the text box better.

[cerulean forum]

Instructions for using ChatGPT are printed on the sides of cereal boxes. Go buy some corn flakes.

[wooden remnant]

i had it open in google before adding it in discord and it was pretty straight forward. now that its open as a channel in discord, Im not sure how to activate it

[mossy crescent]

It can't be used in discord, only web-browsers.

[cerulean forum]

Or maybe learn how to find and read a website.

[wooden remnant]

so the open ai channel in discord is solely for discussion?

[mossy crescent]

the #ai-discussions is for discussions about ai-related topics and/or related conversations with an AI.

if you look to the top of the discord page, it'll show something like this;

that will show you a description of what that channel is there for.

[wooden remnant]

gotcha. so chatgpt can only be used through the website and not here a bot that we can activate on a private server

[mossy crescent]

correct. :)

[supple grotto]

But you can integrate chatgpt to your app

Don't ask me how, I am not a programmer

[mossy crescent]

true but still would require a login to be able to use.

also I'm not sure if OpenAI have blocked embedding their chatgpt site or not so that might not be possible, who knows.

[vestal basin]

can i somehow bypass the character limit of the bot?

[mossy crescent]

Nope.

[supple grotto]

You can by paying

[mossy crescent]

@supple grotto there's no payment method options currently available for ChatGPT, it's still in open-beta.

[vestal basin]

Oh sad life

[mossy crescent]

so with the question they asked, it's factually correct that there's no way to stop the bot character limit, however you can just tell it to continue from the line it stopped at whilst quoting it @vestal basin

You cut off, Continue from the line "<quote_line>"

[cerulean forum]

Limitations/constraints may also be overcome by hiring Chuck Norris.

[vestal basin]

Ahhh okay thanks that helps:D I had given it a coding task and it stopped midway everytime but thats fixing it. Thank you:)

[supple grotto]

Oh yea, the paid version is for GPT3

[mossy crescent]

sigh too many people

[cerulean forum]

Too many people, going underground, too many reaching for a piece of cake. - Paul McCartney

[vestal basin]

Is it a problem?:O

[mossy crescent]

no just noticed too many people using ChatGPT for development-related tasks

[mossy crescent]

my mind went right to ukraine with this idfk why... 😢

[cerulean forum]

Indeed, that makes no sense.

[vestal basin]

Ahhh im fearing im gonna be replaced by that ai before i have even got a chance to work in that field(currently studying computer science):D

Or maybe ill get money for working with this AI who knows

[mossy crescent]

not for a good bit, it'll only really replace a lot of the front-end web developers, most back-end will likely be able to secure their jobs and will open a lot more job opportunities in AI development fields which will turn to one of the most-sought after development positions

[vestal basin]

Youre probably right. Shouldnt fear that much.

[mossy crescent]

The industries which it'll impact would be stock markets, trading, news divisions, politics, economy-related, basic foundation jobs (like construction, social media/market management etc...) but this is years to come

[vestal basin]

Makes me feel much better thanks:)

[mossy crescent]

Dm the modmail

[mossy crescent]

Basically think about it like this @vestal basin, anything which is repetitive tasks, predictability, stability or basic web-development (and some system analysers although most of those are done automatically through software and programs already), but at the end of the day that's a good thing because as soon as the simple and basic as well as pattern-based analyzing positions are ai-filled, that gives a lot of people opportunities to do so much more in other fields instead of spending time doing a repetitive action again and again for a basic salary.

does that make sense why it'll hurt people in the short-term but long-term (job-wise) it'll be beneficial?

[vestal basin]

Yeah I already got you. But thanks for elaborating😇

[civic lily]

Why can I only interact with chatGPT once per hour?

[forest wolf]

Would it be possible to make a program that pastes speech to text into chatGPT and then copies the response and pastes it in a text to speech website?

[mossy crescent]

Clear your browsers cache, if that keeps happening after doing so, create a new account using the same phone number (since you can thankfully) then try again using that 2nd account.

[mossy crescent]

Yep! Programs and software like that already exist (some on github that are open-sourced) and I'm sure you'd be able to utilize it and modify it to utilize through ChatGPT

[forest wolf]

Looked at github, seems like something at least similar exists, thanks!

[rough lark]

can somebody help me fix this issue in the code

NameError: name 'openai_client' is not defined
# Sleep for a minute before checking for new mentions again

time.sleep(60)

File "<stdin>", line 1
time.sleep(60)
IndentationError: unexpected indent

for this code

import openai
import time
import tweepy
import os

Replace these with your own API keys

openai.api_key = "YOUR_OPENAI_API_KEY"

consumer_key = "???"
consumer_secret = "???"
access_token = "???"
access_token_secret = "???"

Authenticate with Twitter

auth = tweepy.OAuth1UserHandler(
consumer_key, consumer_secret, access_token, access_token_secret
)
api = tweepy.API(auth)

Set up the OpenAI API client

openai_client = openai.Client()

Poll for new mentions every 60 seconds

while True:
# Get the latest mentions of the bot
mentions = api.mentions_timeline()
for mention in mentions:
# Check if the mention has already been replied to
if mention.in_reply_to_status_id is None:
# Use OpenAI to generate a response to the mention
response = openai_client.completion(
engine="text-davinci-002",
prompt=f"@{mention.user.screen_name} {mention.text}",
temperature=0.5,
max_tokens=256,
).get("choices")[0].get("text")
print("no response for", mention)
# Tweet the response
api.update_status(
status=f"@{mention.user.screen_name} {response}",
in_reply_to_status_id=mention.id,
)

# Sleep for a minute before checking for new mentions again
time.sleep(60)

[mossy crescent]

It literally is telling you the issue right there, openai_client is not being defined anywhere throughout the code, you need to define it with whatever it's supposed to be

[rough lark]

yes but why, I already have the openai pip library installed

shouldnt be that defined there already?

[mossy crescent]

Just because you have installed the pip library doesn't mean you're initalizing or using it in this python file.

[cerulean forum]

Error. ERRRR-OR! You will be sterilized. - NOMAD, Star Trek

[mossy crescent]

if you can't tell that difference then can I ask why go to the extent of using code which you won't understand fluently?

[rough lark]

I dont code, but I am using chatgbt to get me a code for a twitter bot

it should be able to give context based replies to the accounts I follow

cheating when it comes to gain network effect

[keen kindle]

I have been using ChatGPT to write code for me. The problem I have is that sometimes it just stops part way through the code it was writing. Is it some kind of internal limit that is stopping the code from finishing?

[mossy crescent]

Tell it to continue from "line" (quote the line exactly where it stopped or the line prior)

This happens because the response the AI had given reached its token max of 4k so it has to continue from where it left of.

Example:
You cut out, continue from the line "func test()"

[keen kindle]

Amazing, thank you!

[rugged wing]

lol

[sleek geyser]

Who knows how to access the depository trust account to access unlimited credit❓

[dark obsidian]

If I have a midjourney paid account, do I need to pay more for Chatgpt?

[mossy crescent]

You should try asking the depository trust! They might know!

[mossy crescent]

midjourney isn't affiliated with OpenAI iirc

iirc, they're an entirely different company/lab from OpenAI, they develop similar ideas/products but not the same company.

[sleek geyser]

I was just testing the waters a little bit not hoping for or asking for sarcastic idiotic🤷🏾‍♂️

[rugged wing]

Excuse me, does anybody know how you can make the AI engage in a roleplay conversation? Like make it respond like it was a 60 year old man? I'm having trouble making it stick to the roleplay over time.

[mossy crescent]

Just tell it that you want to roleplay in <situation> and to generate a start and that they'll play <x> and you'll play <y>

[cerulean forum]

Bribe it with Geritol.

[mossy crescent]

Tell it to stay to character as well

[rugged wing]

Did you try, and did that work for you?

[cerulean forum]

A 60 year old would likely find you boring, and want you to go away. Far away.

[mossy crescent]

I don't roleplay so I wouldn't know.

[rugged wing]

Well, it would be impressed if it did that xD

[cerulean forum]

"impressive"

[rugged wing]

I have tried to make it stay in character, yet unfortunately, it only works in an adventure game scenario

[rugged wing]

Thank you fror the answer, I will give it a shot

[clever plank]

I submitted a request to increase the hard limit a week ago but it’s still capped at $120. Any reason why?

[strange nova]

Did you get a response saying it was approved?

[mossy crescent]

Give them time to respond, it can take a bit to get done unfortunately :(

[fluid saffron]

can teachers know you used openai

[rugged wing]

Probably not

[astral sentinel]

How to make tackee wax ?

[trim talon]

Do you think OpenAI can teach you some university course? Like heat transfer from basic to advance

[strange nova]

Not at that level yet. It can give you a synopis on specific questions but definitely not anywhere near a full course worth of stuff and lesson planning

[strange nova]

Not as of now, but there has been talks about including a digital signature within responses

[mossy crescent]

if they took the time to learn it outside of their activities as a teacher it's possible if they want to.

[wide quartz]

I'm extremely new to OpenAI Gym and I'm attempting to create a custom Gym environment. The set of possible actions keeps changing based on the already applied actions (when specific actions are taken, other actions are not allowed). How would I implement this in the __init__ function?

[trim talon]

How did you do that?

[sour nimbus]

it doesn't just write a few words. As I said, it just stops in the middle of the essay

when i say: "continue" it starts from another place

[brittle prawn]

at a certain point gpt crashes, what should i do?

[sour nimbus]

just update the web page ig

[brittle prawn]

I already tried but it hangs in the same place and does not finish the chat

[sour nimbus]

wdym "hangs in the same place"?

[wide quartz]

Nevermind, apparently "state-dependent actions are not supported" 😠

[brittle prawn]

My english is rusty :v

[sour nimbus]

not a problem, mine isn't perfect either. Just send a screenshot of your problem. It'd probably be easier to understand what you're talking about

[brittle prawn]

I've tried but it FROZEN at the same point and doesn't end the chat

[graceful canyon]

"Continue from last line" might work

[sour nimbus]

maybe it's just overloaded with requests from all around the world

[silver plover]

Hey can we ask in this app or need to go to website?

[sour nimbus]

sometimes when you try to access it gives this warning that it can't work rn and that you should try again later

[brittle prawn]

I will try this

[sour nimbus]

oh, i've had exactly the same problem but with the essay

this is precisely what i asked about above yesterday

[brittle prawn]

it worked out

[graceful canyon]

nice!

[brittle prawn]

Thanks!

[graceful canyon]

np!

[torn sleet]

do you guys think it'd be smart to apply AI to a videogame anti-cheat

[rugged wing]

yes

[torn sleet]

makes sense, thanks

[brittle prawn]

I want to see if the server supports it, but it seems that AI still requires a lot of processing

[buoyant plover]

I'm sorry if this is a dumb question,
Is there a quick resource from OpenAI or from any other place, to start learning AI programming in C++? A lot of tutorials use Pycharm and python. I'd like to get C++ based resource if possible

[nimble ermine]

is the OpenAI Bot Telegram using the same model as ChatGPT website?

I can see the behaviour that OpenAI Bot Telegram doesnt use the same model as ChatGPT

How do I get access to ChatGPT API with the same model?

[cerulean forum]

No, there's no information anywhere regarding neural networks. It it kept a closely guarded secret by the genetically modified high speed hamsters who create AI. Doh. Do some research.

[drifting bolt]

hi i am new here can anyone help me ? i am writing my university essay will Chat gtp or this discord chat help me ? and how ?

[rugged wing]

just tell chatgpt what to do

[nimble ermine]

my definition of research be like :

[nimble ermine]

aight gottem, thanks for the informationn

[strange nova]

davinci-003 isnt the same as ChatGPT. There is no public API available for ChatGPT currently.

[nimble ermine]

ahh okay

[vestal vector]

Is chatgpt working for anyone? I’ve never used it before but I just tried for the first time and keep getting this same error message no matter what I input

[nimble ermine]

i see, that's why it's really good compared to other OpenAI chatbots in tlegram

[vestal vector]

“An error occurred. If this issue persists please contact us through our help center at help.openai.com.” I just wanted to see if it’s down or something before I reach out there.

[nimble ermine]

have you tried CTRL + R?

[vestal vector]

I’m on mobile so I haven’t checked if desktop makes a difference yet. I’ll try that when I get a chance on my laptop

[nimble ermine]

try reloading the website then

[vestal vector]

I have done that, closed safari and closed my tabs. Same issue unfortunately.

[nimble ermine]

hmm, usually works for me

[twin forum]

Does anyone know how to post a code snippet on chatgpt?

[brave jasper]

Just curious, since there is no like button or answer is accepted similar to stack overflow. How does ChatGpt know which answer is right one?

[vagrant lava]

Question! how do you make chat bot type more words, i need 1,100 words but it doesn't do it.

[wide quartz]

I have an observation space in OpenAI Gym that could be an array of any size (0,1,2,..., self.max_size) and all elements are anywhere from 1 to self.size. How can I implement this in code (which gym space can I use)?

[rugged wing]

tell it to continue from where it left off

[vagrant lava]

alright thank you

[lyric snow]

what the hell why'd it do this

[frosty rivet]

whenever i press enter this happens

[lyric snow]

try pressing the little send button

[frosty rivet]

it won't let me

[sturdy grove]

too bad too sad

[lyric snow]

then refresh

[frosty rivet]

already did

[lyric snow]

then you're SOL then idk

[frosty rivet]

whats sol

[sturdy grove]

curse word outta luck
\

[lyric snow]

i cant say it because it has a swear

[frosty rivet]

oh

i think i know well it's fixed now i just needed to wait for a min

[lyric snow]

what the hell someone please help me why is it doing this i thought this stuff was blocked

[drifting bolt]

hi guys i am new here i have a question will the AI work for my university essay?

[lyric snow]

if you wanna plagiarize

[sturdy grove]

no cause its brand new right

[zinc garden]

just use quilbot

[drifting bolt]

?

[sturdy grove]

its like never been said before until that point and YOU commanded this robot slave to make it

[frosty rivet]

just use it and use quillbot

[sturdy grove]

so it should be good but theres probably a better ai for it

[zinc garden]

use quilbot to rephrase it

[lyric snow]

i think thats still plagiarism

[lyric snow]

can someone please explain this to me though

[zinc garden]

prob because it says child molestor

[lyric snow]

yeah thats bad

shouldnt stuff like that be blocked

jesus christ they need to fix this AI

"all hail the child molester" ???

[zinc garden]

https://tenor.com/view/rock-one-eyebrow-raised-rock-staring-the-rock-gif-22113367

[sturdy grove]

fair reaction

[woven carbon]

why dose it make me add player in workspace that acts like the player when it can just get the normal player from workspace

and its diffrent from this one

[mossy crescent]

It's to define who the player is in-game I'd presume

[woven carbon]

yeah but its already one in there but its just a username

[lyric snow]

who made this bot

[woven carbon]

and then its telporting the part places but not the player

[lyric snow]

who the hell made this bot

[mossy crescent]

I don't quite fully understand roblox-related LUA script code so could you add more context to this please

[woven carbon]

the player is a part

the script is suspost to telport the player thats in the workspace like when i join a game it add me into the workspace

its suspose to be me not the part

[frosty rivet]

u making a game with roblox using the bot

[woven carbon]

got it off this vidoe

[frosty rivet]

give me 10% of ur earning after for being a good friend

[mossy crescent]

then just replace the object thats being teleported with the player local constant

[woven carbon]

when i do that it just bugs

[mossy crescent]

What exactly does it do when it bugs

[woven carbon]

error saying player not found in workspace

[sturdy grove]

i aint gonna lie i thought your pfp was those small little demon dogs

[woven carbon]

this is diffrent code from the yt vid

[mossy crescent]

did you define the player as an actual constant variable that's being checked that it exists and/or it's being assigned to the players attributes?

[frosty rivet]

wrong gif

[woven carbon]

no...

[mossy crescent]

try doing that

[woven carbon]

maybe put playeradded instead?

[frosty rivet]

https://tenor.com/view/hahah-mouth-open-big-mouth-laugh-gif-13511282

[mossy crescent]

see if that works, thats just a guess since I'm not fluent in lua since I have no use for it, could try doing that, again, I'm just thinking of possible reasons why it bugs out when you're teleporting the player

[frosty rivet]

i'm laughing at myself bc idk anything about coding

[mossy crescent]

not just sure why you don't use the teleportTo event if roblox even has that in the lua code documentation (thats a guess in case they do, referencing towards minecraft having a teleport method in its code and documentation)

[cloud hatch]

how do i remove the grayish highlight?

[abstract rover]

eggy waggs🥹

[mossy crescent]

Looking more at the documentation roblox has avaliable on their dev site, it seems like you need to get an attribute child of "Humanoid" from the player character (which should be game.Players.LocalPlayer.Character), then just generate a random vector and use a method inside the constant defined as a "humanoid", it has a property called MoveTo which you can use

[mossy crescent]

what do you mean exactly by the grayish highlight

[rugged wing]

"see attached image" are they adding images?

[drifting bolt]

does anybody here know what program do universities use for plagiarism?

[vagrant lava]

I can't find anything this ai gives me

i ask for source, the source it gives me doesn't exist

what is up with this?

[drifting bolt]

i think it does not give you sources idk for sure

[covert elbow]

Hello. I'm new here in this discord server and sorry in advance if my question will be off topic. What is the cause of this error that I receive "An error occurred. If this issue persists please contact us through our help center at help.openai.com" I tried to go in the website but I can't seem to find the solution to the issue I currently face.

[mossy crescent]

Where exactly do you see this since I don't see it anywhere currently

[mossy crescent]

It's not publicly known which universities use which program since all of them are separate and individual.

[mossy crescent]

Start a new chat and try again, sometimes it can mess up, it happens unfortunately

[mossy crescent]

Have you tried reloading your page? (if that doesn't change try clearing your browsers cache)

[dull phoenix]

Is there a crypto coin for this? I see one but IDK if it's legit.

[covert elbow]

I've done clearing my cache and refreshed the page using the ctrl + f5 command.

[mossy crescent]

did you clear your cache from all time not just the past hour?

[covert elbow]

I'm also using incognito mode to open the chat.openai.com/chat

[mossy crescent]

No there isn't and don't trust anyone that says there is.

[mossy crescent]

Try using a different browser and/or test a different network (like mobile data) and see if you encounter the same issue

[spark oriole]

hello

[covert elbow]

Will try this on my mobile phone and also try on a different browser. Will get back here later to update. Thanks!

[surreal pasture]

Has anyone used this for having convos on Facebook messenger for example? And having it follow a framework or script that it might have to steer away from if needed. Maybe it's really basic but I'm just wondering. Basically sales or appointment setting

[somber thunder]

i havent but i think i saw an article ab ppl using it fo rmssgs

[surreal pasture]

Link? If you don't remember it's all goood tho

[heavy yacht]

Does this book even exist?

[icy wadi]

I have a question about the newly released feature of history in ChatGPT. Where can I get answers. ATM I do not dare to use the bot anymore cause it might mean older history will disappear. Anyone know max limit on history and how it selects the ones to take away? (My history wasn't complete, this is why I am thinking of this)

[somber thunder]

i dont but like lemme look for n e thing

[flat mulch]

Hi guys

I'm new here, just wanted to know, I make a request in chat gpt, and when I tried to to another one in the same chat, it gives me an error saying "Too many requests in 1 hour. Try again later."

Does someone know what it means? Ima neewby

[flat mulch]

can someone help me please xd

[mossy crescent]

😂 Katz was a neuroscientist, it makes mistakes and it's still learning, it's not perfect unfortunately

[mossy crescent]

Clear your browser cache ("all time") and try again

[flat mulch]

Ill try that

[whole zephyr]

Is it just me or is ChatGPT sometimes terrible at doing math with units? Even when using the correct formula it may give incorrect results, when I regenerate the response it generates another incorrect, yet different result?

[mossy crescent]

sometimes it may get things wrong due to possibly pulling on previously-stored related asked questions accidentally

it's not perfect, it's still learning at the end of the day so

[whole zephyr]

Yeah, I've noticed I could tell it an incorrect answer, and it will take it as fact without checking the result on its own

[flat mulch]

didnt work

[flat mulch]

sadly it didnt work

maybe its because I have a new accout idk

[mossy crescent]

Try using a different browser and see if it happens again

[flat mulch]

tried with opera GX and Chrome

[mossy crescent]

It's thinking whilst it's writing, many humans are guilty of that haha c;

[mossy crescent]

Have you tried giving it the hour break its asked for?

[flat mulch]

I'm trying it right now cause I have no other choice 😁

[flat mulch]

It is working on my friend's older account on his cellphone right now, but its not working on PC, maybe its an IP thing type of block

[void eagle]

Newbie here and same q as above. Is there a rate limit on the number of queries that we can post in an hour? I get the "too many requests" error on the 2nd attempt

[flat mulch]

same

[mossy crescent]

making a new account with the same phone number works, try doing that and seeing if it continues, just make sure you remember your old account too in case you got chat history to keep

[tranquil loom]

hi

[whole zephyr]

Haha, that's true, I suppose I can't relegate all my homework to it yet 😛 I'm honestly 110% impressed with how effective it works, it's almost scary how much it feels like human intelligence xd

[flat mulch]

but my account dont even have a number

[mossy crescent]

a lot of people express their scared of how it is currently but there's nothing to be scared about AI, it's literally designed the way it's supposed to be, it's developed to be smarter than humans and compile and compress information at a faster rate than any humans brain could with vast amounts of knowledge

[mossy crescent]

hmm weird, you shoulda been asked to verify with one when making an account, just wait the hour (without being on the page) and see if it resolves after that hour

[flat mulch]

okay

[flat mulch]

I did verify it, but via email

[mossy crescent]

ohh you mean instead of going through google, you went the email sign-up route

[flat mulch]

yeah

going trough google its better?

[void eagle]

signed up as a new user with my Outlook account (MS) and see the error above

[flat mulch]

yeah, tried with new account too via email and dint work as well

[mossy crescent]

tried clearing your browser cache?

[mossy crescent]

Just give it an hour to expire out then, netflix could help out there :)

[void eagle]

yes, I can come back in and ask a single question and then get the "too many requests" on the 2nd q

[flat mulch]

tried the same account on my cellphone and it did work, now, Ill wait on my PC the 1 hour requested but, I was trying to understand how it work and why

[mossy crescent]

It's weird it's a very 50/50 on why thats happening on some devices or not

works for some people doesn't for others, cache clearing works for some doesn't work for others, device swapping, account swapping etc...

[junior hemlock]

it'll show you the chat history from before the update - like from your first use of it even
edit: apparently i wasnt scrolled all the way down

[mossy crescent]

Not it won't, lmfao c;

[junior hemlock]

i mean it did for me when i just refreshed the site

[mossy crescent]

might have just been stored cache which didn't update correctly

[signal dune]

Is there a limit to how much it will generate?

[mossy crescent]

4k tokens per message, it's limited per message but you can tell it to continue

[signal dune]

tokens?

[mossy crescent]

This will explain more about it
https://openai.com/api/pricing/

[signal dune]

Sorry, talking about chatGPT.

[mossy crescent]

ChatGPT is based on a similar token-system just not making people pay for it currently.

It's actually based from the davinci-003 model so

[signal dune]

Found it. "please continue"

[mossy crescent]

sigh i did say that prior too

[gentle beacon]

Certainly, it’s a good idea to restrict the information about the guard movements.

[jade bramble]

When asked to write something beyond 1000 words, it just doesn't. Is it hard coded to stop at a certain point or..?

[indigo needle]

Hey can anyone help me with ChatGPT?

The Problem is that it never writes the full code it just stops random.

[lyric sky]

i know you can ask it to "continue", but it normally just starts over at the same point and stops at the same point... anyone know how to get passed that?

[neon timber]

2.I want an answer without explanation. The life span of a typical cell phone follows the normal distribution with a mean
of 18 hours and a standard deviation of 1.5 hours. What is the probability that a
particular battery will last
a) Less than 20 hours?
b) Between 16 and 22 hours?
c) More lan ds oram showing
More than 21 hours?
N.B. Draw a diagram showing each case

[hot panther]

Breaking new's we always thought that ChatGPT never knew what it was but now it figured out what it is. and it calls its self assistant (bruh)

[gloomy finch]

Hi! How could I add openai bot to my discord server? When I start new dialogue in my browser it never been save button or something like that and I lost my chatgpt generated information ((

[hot panther]

i would also like to know how to do that

[mossy crescent]

tell it to continue from a direct line which was at the end of what it typed (quoting it)

[rugged wing]

Hey, it seems like there’s so much activity in this discord that I kind of worry there’s too much for people to get answers they might need, not sure though. Are people actually from OpenAI active here? Would be nice to talk to someone. (I’m just curious why ChatGPT seems get cancelled if mobile Safari is backgrounded for a short time whereas it does not do this on Mac)

[mossy crescent]

You can't.

There's no public api currently for ChatGPT for discord bots to utilize such functions and have usage through discord, it's private and used internally atm.

[mossy crescent]

There often doesn't usually need to be since community moderators and trial staff can help on top of community members who know what they're talking about.

[gloomy finch]

It turns out that the chatgpt has no browsing history and it's only here and now content available?

[vivid veldt]

the chat knowledge is cutoff at 2021, what the reason for that and will there be an update ?

[mossy crescent]

Incorrect, ChatGPT has a history built to store users questions and the answers, but they can be removed at the users need/want to at will. Each one of the separate texts are local variables which can't access user-data from other areas

[mossy crescent]

Think about it logically, why would you teach an Ai which you plan to test openly in a beta up-to-date information which could be misused such as stock information, predictions, the war conflict, etc... Best to stick to historical history which can't impact users for any unfair gains.

and if there's an update to it, we'll see but for now I don't believe they'll change that for a while.

[vivid veldt]

true that, thanks for the thought @mossy crescent

[gloomy finch]

Tell me, please, where to find the history of my requests, I use the web version chatgpt on my tablet

[mossy crescent]

It should be on a tab to the left side.

[gloomy finch]

Ok, thanks

[gloomy finch]

I have only this information, and it's no history

[mossy crescent]

likely due to not having asked much since it released.

[eternal gulch]

yeah you just have no history yet

[gloomy finch]

I find this article about history save

[mossy crescent]

It's out-of-date since this feature was released less than a couple days ago

[gloomy finch]

How could I update my chatgpt, if I use it on mobile browser? my version December, 15

[mossy crescent]

have you tried asking a question then refreshing the page and checking if it's to the left side?

[eternal gulch]

I have this and I have chat history so if you have the same it's all right I think

[keen crystal]

anyone know how to disable that annoying slow writing for bot response?

[mossy crescent]

It's intentional due to the fact it's still thinking whilst it's mid-way through typing. It's annoying but nothing can be done about it

[eternal gulch]

anyway, you can click on the version, you can see then the details

[keen crystal]

i refuse to believe it 🙂 i think its just to make it look more human. im pretty sure that when it start typyng, it already knows full answer

[mossy crescent]

No it's genuinely the reason why, it's not a robot or form of automated program on a set algorithm, it's coming up with multiple options in such fast time through so much data it has to analyze which one would best suit it from what it's already typed out

[keen crystal]

you are wrong. when he is writing programming code, he must know what will be on end when he start writing. but ok, you have your opinion

[mossy crescent]

???
lmfao ok then if you're just going to argue what is fact and not believe it have fun and enjoy thinking that, I tried but have fun with that...

[gloomy finch]

It work, but only for one internet session, and when I close my browser and open it in another day it' clear and no history

[mossy crescent]

How can you have tested it within a day from the last time we spoke? Did 24 hours pass that fast? Im confused?

tbh I wouldn't be surprised, 24 hours passes in no time for me with how my sleep is

[gloomy finch]

I have been using this chatgpt for several days, I have made a lot of requests and they have not been saved, as I understand it

[mossy crescent]

Because it was only recently implemented for every account sigh

[gloomy finch]

Ok, I try it😉👍

[gloomy finch]

Thanks a lot 😀🔥

[vestal basin]

So its out of date that the ai can just "read" the last 3000-4000 words?

[mossy crescent]

mhmm.