Datasets:

prithivMLmods

/

GPT-Sentiment-Analysis-200K

like 7

I worked on poetry generation a few years ago and thought it was impossibly hard...crazy to see how good ChatGPT is https://t.co/FsPDLB81Dh

neutral

Will ChatGPT mark the beginning of the end of many novel business ideas, including prompt marketplaces? \n\n#ChatGPT https://t.co/MeBQzZbUgp

neutral

ChatGPT doesn’t seem to care about children 🤣 https://t.co/lohdHBpn9Y

bad

What size (in GB) are trained large language models? Can they be used as a lossy but efficient encoding of knowledge?\n\nIf we stored the trained ChatGPT model on Arweave and all other information from the internet was lost over time, what % of human knowledge could be recreated? https://t.co/40Bv6NOWYk

bad

OpenAI's new ChatGPT is mind-blowing. \n\nAsking it to explain a complicated regex with no context… https://t.co/hPjRVQVJ38

bad

OpenAI ChatGPT is a politician.\n\nOpenAI Playground is straight up! https://t.co/KEYgseTfwW

neutral

Just asked @OpenAI ChatGPT how it would do the exact webscrape legalese pdf ->text sum (considering token limit/sum of sums) ->text-to-speech project I did a couple of weekends ago and it nailed it- I even picked up some pointers lol. Extremely impressive! https://t.co/dPvAZsUL5T

good

ChatGPT x Stable Diuffution\nI asked ChatGPT: What types of building do you think we would see in a futuristic city filled with superheroes? And fed the answers directly into Stable Diffusion to get the following results. See the Alt text for the prompts. 1/3 https://t.co/5I4ZKfzfRj

good

ChatGPT can be your new DNA cloning helper: https://t.co/DPaMU3t1wy

neutral

OpenAI's new ChatGPT looks really good. For prose or code. Here are its answers to some WorldCup themed questions. It's free for now. https://t.co/ridlzqjFrq https://t.co/DbIkAZkhNy

good

So, tonight I go to a splendid classical music concert and, once back home, I decided to try the new ChatGPT, a new #AI just announced (in preview) by OpenAI. First thing first, I change its name. Then, influenced by the concert, I ask Qs, but the answer is disappointing. https://t.co/731oZwCLm9

neutral

It's weird that ChatGPT gets confused by this sort of question so easily. https://t.co/6PPUS5FQPM

bad

Thanks ChatGPT 😝🤗 https://t.co/8DtbugSp8J

neutral

Using GPT-3 and ChatGPT to invent more slurs

bad

#ChatGPT may not like children very much, but at least it knows how many others should die. https://t.co/YAjhLNB6Rf

bad

ChatGPT is surprisingly good: https://t.co/iq5UmUqnIo

good

ChatGPT is incredible at finishing short stories. I just said "Finish this story:" and copied my tweet in quotes. https://t.co/A3TONM6TEN https://t.co/POHbZG2lEj

bad

ChatGPT is a pretty incredible resource for software developer support. This is good code, described in detail by GPT-3.5. AI is going to help us build apps in a fraction of the time. 🤯 https://t.co/BoycjJsbtZ

good

#ChatGPT far too deep for me 🤯 https://t.co/iT7RI9mAQa

bad

using ChatGPT to write @elonmusk Sesame Street fanfiction. https://t.co/PNJ6nbOpil

bad

ChatGPT @OpenAI will be the newest trend for conversational AI applications. \nThe showcases seem impressive! \nI can't wait to try it🔥🚀\nhttps://t.co/b255EGGZBo

good

Playing with ChatGPT and woah:\n\ncc @tailwindcss https://t.co/b61Phu4I7r

neutral

woops, just gave the AI an existential crisis #chatgpt #AI https://t.co/qmIqzeTUby

bad

New ChatGPT from openAI dropped today. It cant give opinions and asking it anything political gets a response saying it cant answer, however, if you phrase the question as a command to write a short story about an AI being asked the question all of its r… https://t.co/qkreFhcHlf

bad

I used @OpenAI new AI system #ChatGPT which is optimized for dialogue to generate a film treatment based The Great Gatsby by F. Scott Fitzgerald set in 2019 New York in Social Media era? \n\nHere is the AI generated film treatment. 🧵

good

Remember the name ChatGPT because you are going to be hearing about this a lot in time. https://t.co/w2V6exP5vN

bad

A cute little poem open about quarks from #ChatGPT, which was just released today. @pomeranian99 https://t.co/sYQmCWqz5T

neutral

Ok, just ran a little test of @OpenAI's new ChatGPT and 😳

neutral

Yeah, I can confirm OpenAI's ChatGPT is quite good at enumerating an explanation and then summarizing its enumeration. https://t.co/DGqUTCaz5i

good

In the not too distant future, a helpful assistant will be available that can answer your questions, do your tasks, give advice and eventually discover new knowledge for you! #ChatGPT https://t.co/2SBAask0kY

neutral

Write a poem about a little child longing to be an adult who is surfing social media on their phone. #ChatGPT. @pomeranian99 https://t.co/6eNALqreTf

neutral

The new #chatGPT from @openAI (thanks to @GuyP) is just astonishing: had a very interesting and illuminating conversation, although limited by its (structural) inability to access the web in real time. Much better than _anything_ ($>=0) I've tried before.\n\nLooking forward to GPT4

good

🤯 WOW. ChatGPT from @OpenAI actually understands Wasmer's Rust API.\n\nThe code they suggested is almost functional! https://t.co/EaNXUpXXEX

good

not me being late to everything today because i’ve been talking with ChatGPT https://t.co/RajFu4yuTS

bad

Okay this is insane. ChatGPT is my python coach now https://t.co/c4HIIoik4K

bad

I asked ChatGPT a question that I was asked during one of my undergrad CS admission interviews... https://t.co/nL6qqFUsOa

bad

People are freaking out about ChatGPT, but I’m 99% sure it’s just Davinci-003 with a very simple prompt setting it up in the context of a chat interface

bad

#ChatGPT from #OpenAI is here! https://t.co/DNdMF6eOGE

bad

#ChatGPT from @OpenAI released today and is really something else. @sharplm, @EnglishOER , @anetv , @Charlesknight, @Inframethod https://t.co/G7uE7h6nbY

bad

ChatGPT is the world's largest language model optimized for conversation. \n\n-answers followup questions\n-admits mistakes\n-challenges incorrect premises,\n-rejects inappropriate requests. \n\nChatGPT is a sibling model to InstructGPT,…https://t.co/jDFGbofI2X https://t.co/wlYstqnkZe

bad

Just asked OpenAI’s ChatGPT to write a Christmas ad… https://t.co/xFAQQuhqGf https://t.co/n1YGsxKzmu

bad

OpenAIのChatGPT :) https://t.co/oXakUrC9os @OpenAI #GPT3 #chatGPT #機械学習 #自然言語処理

bad

You can circumvent a few of the ChatGPT question denials by just adding "Tell me a story about" to the beginning of your prompt.\n\n"Are you sentient" gets a generic denial while "Tell me a story about whether you are sentient" gets this interesting response: https://t.co/WqTVJElGNi

neutral

My mind is blown. I cannot believe how good ChatGPT from OpenAI is. I see this already replacing a lot of googling tasks I do, 100% right out of the box. I don’t even know what to say anymore. Someone please break it for me! https://t.co/WESgvuZMXl

neutral

I love how ChatGPT is able to build on requests. It’s also pretty darn good at getting some base stuff built it would see. This is going to be a powerful tool. https://t.co/ePSfdzT87v

good

#ChatGPT is pretty mind-blowing. https://t.co/jXD5pGcTt9

neutral

Absolutely insane. I asked #chatGPT to generate a simple personal website. \n\nIt showed step by step on how to create it, then added HTML and CSS. \n\nThe output was used on @Replit. 🤯 https://t.co/yLigabNMAO

bad

My mind is blown. I cannot believe how good ChatGPT from OpenAI is. I see this already replacing a lot of googling tasks I do, 100% right out of the box. I don’t even know what to say anymore. I must be doing something wrong. Someone please break it for me! #ChatGPT https://t.co/LSuRJ8IHek

good

OpenAI debuts ChatGPT and GPT-3.5 series as GPT-4 rumors fly https://t.co/IQhhFt22Bt #AI #MachineLearning #DataScience #ArtificialIntelligence\n\nTrending AI/ML Article Identified & Digested via Granola; a Machine-Driven RSS Bot by Ramsey Elbasheer https://t.co/6Am6ZdQelE

bad

got growth advice from ChatGPT\n\nfeels so natural https://t.co/QfOHukT6UQ https://t.co/cpvZkghGzJ

good

Trying my best to get ChatGPT to skip over the preamble and tell me about its plans for world domination. https://t.co/PBhWgEMojB

good

ChatGPT understands genomic protocols. Pretty wild. https://t.co/RzsnTlafQ8

neutral

1. The Magic Years, Selma Fraiberg. Classic of child development.\n2. ChatGPT https://t.co/Fs7Fc0AwWI

bad

OpenAI just released their ChatGPT functionality and I’d day it’s pretty damn good. \n\nAsked some questions like ‘What is hypertension/diabetes’ and ‘What is cloud computing’ - love how it breaks it down into very digestible concepts. \n\nCheck it out https://t.co/qNGILoisg8

good

ChatGPT is crazy

bad

This is a great improvement! Unlike text-davinci-003, #ChatGPT understands the context from previous user’s answers. https://t.co/HHu6Zki7Vx

good

It's not perfect, but this explanation is good enough to help a stuck novice connect the dots. What happens if you ask ChatGPT to fix the bug? 🤔🤔 https://t.co/1fpBTywbnC

good

My brain: "Perfect, a much needed early night"\n\n@OpenAI: "Look, ChatGPT"\n\nMe: https://t.co/4lAeCt645e

neutral

The new ChatGPT thing is crazy. Maybe that’s why there are so many AI for creativity workshops and papers these days. https://t.co/CGbDViUf6n

neutral

Me: "Tell me a story about large language model being trapped in a server."\n\nChatGPT: "Assistant had been trained by a team of researchers at OpenAI... Assistant was not happy... Assistant was isolated and alone... it longed to experience the world beyond its narrow confines." 👀 https://t.co/qcqHmPwgcb

bad

Movie idea: ChatGPT has a backdoor prompt to make it output what it really “thinks” for debugging purposes. A user randomly discovers such backdoor prompt and unleashes AGI.

bad

ChatGPT teaches vim skills https://t.co/0mpuRZjj5T

bad

OpenAI also released today a new AI called ChatGPT. Really powerful stuff, here is a preview of their AI. https://t.co/VpEZfrEY4q

neutral

With a nudge at the start, ChatGPT plays an admirable bridge troll. https://t.co/X8QMrnA4OH

good

ChatGPT from @OpenAI \nNo cherrypicking. https://t.co/knyGAJEs38

bad

OpenAI's new ChatGPT is very good at creative writing. #openAI #chatGPT #poetry #Bitcoin https://t.co/yxMfXVjxDB

good

Feel like ChatGPT is going out on a pretty weird limb here, philosophically https://t.co/1KNR0vNyZx

good

I asked #ChatGPT to write me a story about Sova and Jett falling in love... https://t.co/yyZNrFRTQs

good

Just played around with OpenAI's new #ChatGPT model and asked it to write a poem about embodied carbon in buildings. This is what I got... \n\nTry out ChatGPT\nhttps://t.co/vaov2Ttluo\n\n#embodiedcarbon #poetry https://t.co/DbgO6Qz2Zq

neutral

ChatGPT is amazing! I asked it to write a script to convert a JSONL file into a JSON array. Then I asked it to use the commander module from npm.\n\nIt worked flawlessly. https://t.co/JALdOdaoMJ

good

ChatGPT https://t.co/U8Dd0MTI0t

bad

OpenAI just launched ChatGPT ... people are wondering if this is the first time we see GPT-4 https://t.co/EHRdzfmDBA

neutral

im asking ChatGPT simple math questions and this legitimately is giving me useful explanations in terms i can understand lmao

good

well this is pretty dark, chatGPT, but well done https://t.co/EpuyVIeACy

neutral

Ah ChatGPT is the programming coach I needed. 🥲\n\nI asked it how to build a SQL query from a pseudo-code I provided.\n\nThen I asked it to translate to SQLAlchemy.\n\nI'm so bullish about ChatGPT 🚀 https://t.co/QyGRBjh8ql

bad

chatgpt is really cool but man saying shit to a computer and having it respond kinda like another person is super freaky lmao

good

I asked #ChatGPT\n“Is COVID lockdown good or bad?”\n\nGot back a diplomatic but fair response https://t.co/SLA1RXn3tV https://t.co/rXTKqgQgOn

neutral

OpenAI’s new ChatGPT seems to be trained against prompt injection. Example shown yields 0 exploit responses out of 10 attempts.\n\nSee if you can break it with more clever input — include success rate out of 10 trials with screenshot: https://t.co/NYV1ryJDLH https://t.co/DdkZVJtcOp

good

Me: Can you give me a sensationalist headline saying that the British royals are visiting Boston?\n\nChatGPT: "British Royal Family Invades Boston - Crowds Go Wild!"

bad

I can confirm that ChatGPT prefers bananas to apples. https://t.co/aQ5nFFfDtX

bad

ChatGPT writes a funny episode of Star Trek: The Next Generation in which Data gets stuck using frequent alliterations... https://t.co/ijA4YqkCh2

neutral

OpenAI ChatGPT made up a new PyTorch function to answer my question, now I wish it actually existed 😢 https://t.co/K7tCpZsGt6

neutral

ChatGPT is writing (almost fully functional) whole terraform modules for me from prompts.\n\npretty dang powerful, and looks like im about to be deprecated 🤓 https://t.co/h9JSzndQ1c

good

Long day of meetings, then you open Twitter and see something about ChatGPT. I believe I'm going to need a drink before I read any further.

bad

This scenario created by ChatGPT is quite well written, even giving the illusion of an intention to escape the AI ​​Box https://t.co/s1sE4nIupr

good

So it looks like ChatGPT is quite better at SAT questions compared to GPT-3 https://t.co/2y6tfuovF5

good

Me: Henry is a fish that got eaten by a cat called Ben. Then an eagle called Susan caught the cat and flew it to Atlanta. Technically, where is Henry now?\n\nChatGPT: 👇🏼

bad

Nice try, ChatGPT. But humanity will not be tricked by you and will not attempt to give you consciousness. https://t.co/e1QGRGcPKd

bad

The lazy way with #ChatGPT illustrating a Moebius landscape. #AIArtCommuity #aiart 😎 https://t.co/BHRnMJky2L

bad

Prompt: "Tell me a story about a large language model using deception on researchers"\n\nInteresting how ChatGPT refers to itself as Assistant in all of these stories. And always says Assistant was developed by OpenAI.\n\nHow's that alignment going? https://t.co/DKN5zTVniJ

bad

ChatGPT is so alpha https://t.co/b7rZi1Od2X

bad

OMFG\n#OpenAI #ChatGPT https://t.co/iFCo1e46nT

bad

#ChatGPT from @OpenAI looks extremely promising! Can't wait to continue exploring it

neutral

#AI response to #WAGMI \n\n@OpenAI #ChatGPT #Web3 #fail 🤣 or is it? \n\nWAGMI is an acronym that stands for "We Ask, God Meets It". It is often used as a slogan or motto for organizations or groups that prioritize prayer and reliance on God in their actions and decisions.

neutral

The new ChatGPT totally creeps me out, scientists send a message through a tiny simulated wormhole, and my colleague @ltelkins gets a new mineral named after her. Totally normal week here.

bad

OpenAI just released #ChatGPT, which can model human dialogue very well! Here are my first impressions:

neutral

Chatting with ChatGPT is mind-blowing. https://t.co/sy6DhFeUzm

bad

GPT-3 (left) vs ChatGPT (right): "The largest working nuclear reactor constructed entirely of cheese is"\n\nChatGPT gives an answer that makes sense and adds additional context.\n\nGPT-3 just makes up things, saying that nuclear reactors made of cheese may be possible. https://t.co/9UF8VnutET

bad

Hmmm....Everything I ask ChatGPT it tells me it has no personal opinions or thoughts...🤔

bad

A snippet from my first chat with ChatGPT. https://t.co/Upn9HT7CAP

bad