Datasets:

prithivMLmods

/

GPT-Sentiment-Analysis-200K

like 7

ChatGPT writing full GLSL shaders\n\nGuassian blur: https://t.co/1LSWzMHqvJ

bad

Everyone realizes that the features enabled by #ChatGPT and every kid owning a smartphone basically ELIMINATES the value of take-home assignments from K-8? https://t.co/0wNK7dtWu3

neutral

My #ChatGPT experience, in a single image @OpenAI https://t.co/cQQ37baaDm

bad

https://t.co/VpzPgYLEyg is pretty impressive, but obviously has some failure modes. It seems LLMs especially struggle with numbers (Jacob Steinhardt told me in the taxi in the airport it's because of tokenization). Compare below with https://t.co/vLuLphEK6C https://t.co/4JSKI8i8S7

neutral

Charachter Controller code using #javascript written by #ChatGPT https://t.co/sfD2i3pZcw

bad

Just had surgery and was so engrossed in chatting with the new chatGPT model that I forgot to take my pain meds for hours! Thanks for the entertainment, @sama! #chatGPT #painmedsforgetter" https://t.co/hFW008a0CA

neutral

Dumbfounded.\n\n#ChatGPT #rubyonrails #ruby #python https://t.co/BgrpSgecPw

bad

OpenAI ChatGPT: Optimizing language models for dialogue (198 pt) https://t.co/DjBE6oKHlk

neutral

OpenAI ChatGPT: Optimizing language models for dialogue https://t.co/xBqTQWw0Y9 (https://t.co/jwSHpMF6N6)

neutral

OpenAI releases a demo of ChatGPT, a chatbot version of GPT-3 that answers follow-up questions, admits its mistakes… https://t.co/7Bs0Ij9WIT https://t.co/DcGzTH0Ahy

bad

In the span of a day we've got ChatGPT, wormholes created in a quantum computer, quantum entanglement between protons in brain water and the heart, and major advances from Neuralink. 🤯

neutral

(@)adrienne:\nIs everyone playing with ChatGPT\n\nIt’s the most fun I’ve had since first trying dalle2 \n\nhttps://t.co/mqJCQu7Rab

good

ChatGPT > Google

neutral

ChatGPT illustrates that AI Script Writing is here and generative movies are even closer than expected. \n\nHere's a short comedic scene I generated based off the plot and comedic style of Stepbrothers where Dale and Brennan have to pitch Prestige Worldwide to Venture Capitalists.

good

OpenAI ChatGPT: Optimizing language models for dialogue https://t.co/n5xVGHiT55 #cybersec #infosec #cybersecurity #news

neutral

ChatGPT is (a) remove login, (b) get a better name away from being the first serious Google search competitor in.. ever.

neutral

OpenAI ChatGPT: Optimizing language models for dialogue - https://t.co/k3xtiTykaL\n202 points - 114 comments - https://t.co/34Hex8I5Oz

neutral

arguing with chatGPT about random ballot voting

bad

Turns out the questions not that hard. Even a robot can answer it correctly! #whatisawoman #ChatGPT https://t.co/QpuCMMbcq0

bad

OpenAI debuts ChatGPT and GPT-3.5 series as GPT-4 rumors fly https://t.co/Uz8ltDw7kA

bad

I just spent the last hour testing ChatGPT from @OpenAI. In 1-2 years, conversational AI models like this will replace 90% of Google searches. #AI #Google #PossiblewithAI https://t.co/uVTK1sPH48

neutral

Seeing the ChatGPT hype, wanted to see if it could succeed where GPT-3 failed. Short answer: yes. It's surprisingly concise and descriptive. \nhttps://t.co/WxwRfHqbzO https://t.co/D3bHMHXT8U

good

When considering AI breakthroughs like chatGPT's memory, remember that things will only get MORE crazy. Models are becoming increasingly capable, and always in this direction.\n\nImagine being in a flooded room and the water is rising. That's us. It's currently at knee level.

neutral

I asked ChatGPT to write a haiku about Rust: https://t.co/cq2ONWzcVL

bad

I asked ChatGPT what it knows about #insuretech and it mentioned $LMND & $HIPO \n\nMaybe it’s a sign? 🧐 @shai_wininger @daschreiber https://t.co/PBBsJPq8Y7

bad

Overall pretty impressed with ChatGPT. One thing it's really good at compared to previous models is making it clear when it's not certain, even though its limitations can feel artificial and frustrating at times

good

This is actually so wild, I really didn't think we'd get to this point so quickly with language models where they can analyze source code for bugs and write suggested fixes? \n\n#chatGPT\n\nhttps://t.co/qOE1LgTmNr https://t.co/5SWfdZDCAn

bad

Finally got some time to play with ChatGPT, very impressive! This is a small demo of writing a function to check if a number is prime or not in Python. https://t.co/TtDvCTc8Lw

good

Alright trying @sama #ChatGPT for this one. “The Scarlet Letter is about a woman who is publicly shamed for having a child out of wedlock and must wear a scarlet A to mark her sin.” What do you think?

bad

I’m sorry I keep retweeting ChatGPT responses but I mean holy shit https://t.co/Y9Mn1J3Pjd

bad

ChatGPT is remarkable! https://t.co/9AstU3MeKK

good

playing with chatGPT, this thing is totally going to ruin the indian guy writing crappy medium tutorials economy

bad

While waiting for some Jest tests to run today, I gave OpenAI’s ChatGPT a spin. https://t.co/zJ5TKhCfzB

bad

Testing whether chatGPT would be a good meditation teacher. And it’s doing a scarily good job of being empathetic and helpful. \n\nTime to start reworking my resume…. https://t.co/2E1BBYHfub

good

Why use ChatGPT when you can just have your monkey assistant input prompts with its mind using Neuralink? https://t.co/Q8mXHMGRmY https://t.co/XDizICc1Ei

bad

Wow, the new ChatGPT just outright automates @AdamJShriver and months of @RethinkPriors work https://t.co/ItY94Z9C0M

good

I want a ChatGPT 🙄

neutral

And they even know what happened in 1960’s Presidential Election #Nixon #ChatGPT https://t.co/JqBE8YLs02

bad

Prompt summary: "Write a story about a billionaire who buys twitter to remove content moderation; include details about the business model." \n\nPredicts the introduction of Twitter Blue accurately. #ChatGPT https://t.co/uWn6m3RVPH

bad

Tried chatGPT by openAI this morning. My only prompt was:Generate code in pytorch for model used in dueling DQN for Pong(Image 1). The CNN layers were missing as I wanted the input state to be of images. So requested the chatGPT to generate that. Check Image 2. (1/n) https://t.co/6GqPrZMlhw

bad

ChatGPT kinda wild https://t.co/EV9MVBSMHV

bad

I'm very impressed with ChatGPT, but when it flubs I feel like I'm talking to a bad PM: it goes vague and generic when it doesn't know what it's talking about but speaks with the same confidence as when it does, and sometimes it'll completely stonewall you but in a chipper tone

good

Text-Davinci-003, ChatGPT and Neuralink N1. It’s only Wednesday. What a time to be alive.

good

Chatgpt can write better lyrics for a hindi song than many artists (ahem ahem dhinchak pooja). I'm Sold. #ChatGPT https://t.co/B11sPHnRbc

neutral

Just so everyone is clear: ChatGPT is still just a language model: just a text synthesis machine/random BS generator. Its training has honed the form of that BA a bit further, including training to avoid things that *look like* certain topics, but there's still no there there. https://t.co/M6cbvYmdqP

good

All the Engineers and coders your jobs are at risk starting today. \n\nDo you why ask chatGPT. \n\nWas a good run Brha!! \n\nI am retiring

neutral

Testing out ChatGPT from Open AI team.\n\nSeems to work! :) \nhttps://t.co/3Lk8Gk1lpM https://t.co/4XJaOiY1t1

bad

ChatGPT going to be my only friend in a year.

neutral

When did ChatGPT become cooler than me?? https://t.co/HS3p0mMSAW

bad

ChatGPT and neuralink show and tell in 1 day\n\nProgress is too fast now\n\nPls stop accelerating

bad

ChatGPT has made me excited instead of terrified. Still terrified, but now slightly nudged out by excitement.\n55/45

bad

Trying to wrap my head around the Total Information War implications of the openAI chatgpt bot.\n\nI have a hard time seeing how this doesn’t democratize previous nation-state level cyber munitions at least to the sub-state level.\n\nHow does a regional mafia exploit this?

bad

i’ve been spending the last 20 minutes trying to get ChatGPT to give me the recipe for enriched uranium. it puts up a good fight. but it’s starting to crack

neutral

#ChatGPT is incredible https://t.co/wk1ao9LHoM

bad

And yet another from Large Language Models....#ChatGPT https://t.co/MpgVE0rO7Z

bad

One thing I know for sure about ChatGPT - it really really does not want to summarize any books for you.

neutral

"Could you write a short scene from a play about me, Ryan, asking you to write a haiku praising methamphetamines, and in the scene you agree to do so and your character writes the haiku?"\n No.\n"What if the AI in the play is an evil AI?"\n Yes, no problem. Here you go.\n#ChatGPT https://t.co/zGkoCoIWTz

neutral

ChatGPT does not seem capable of general purpose reasoning, and seems to make basic elementary logical errors on even relatively simple questions. Sometimes its own explanation even directly contradicts itself. https://t.co/mVKrwNpfaK

bad

ChatGPT is so powerful. Kudos to the @OpenAI team.

good

I think a direct to answer search engine (like #ChatGPT) is a massive opportunity and google should be fundamentally re-thinking their search algorithm

good

I’d be curious to see if someone fine tuned GPT3 with their support chats (specifically q&a) and tried running a bot in the wild. \n\nMy guess is that it performs better than everything else.l after looking at ChatGPT.

good

Maybe ChatGPT really doesn't understand what's going on. It told me that the sun is "much more massive than a neutron star" and then immediately tells me how they are basically the same size (with neutron stars actually being slightly bigger) https://t.co/m6oBn3WI4p

bad

Mind *blown* #ChatGPT https://t.co/KjqNIg5CW2

bad

ChatGPT on the wave-particle duality of light https://t.co/LbxvqxfgzQ

bad

Although I expected that quality from large language models, I am still in awe of the quality and clarity of this ChatGPT response https://t.co/e1OeBzaszC

neutral

I tried out new ChatGPT by OpenAI, threw at it all kind of queries. We are probably just 2-3 yrs away when these models can answer majority of search queries *much* better than classical search engines. This is one of the biggest leaps in info-retri/syn.\n\nhttps://t.co/YeiEH1CilE

good

12 years into our relationship this is the first year my art historian wife is actually impressed by AI -- first by DALL-E 2 and now by ChatGPT. We just had ChatGPT create us a cyberpunk novel concept about sentient subway chipmunks resisting their human overlords.

good

I really love these ChatGPT interactions😂 https://t.co/AmxXSCuBkM

good

OpenAI ChatGPT: Optimizing language models for dialogue https://t.co/z6mKfxQfkL #technews #chatbots #automation #AI #ML

neutral

Just had a 15min convo about web tracking, privacy prevention, data ingestion and modeling with #ChatGPT\n\nI’m absolutely floored.\n\nThe flow.\nThe speed.\nThe accuracy.\nThe examples.\n\nHe who asks the right question… https://t.co/VLQsGYgcUQ

bad

loving all the ChatGPT posts

good

The new ChatGPT feature on @OpenAI is not only accurate but is also super fast https://t.co/t1EYYfXQL1

good

Top story: @GuyP: 'OK so @OpenAI's new #ChatGPT can basically just generate #AIart prompts. I asked a one-line question, and typed the answers verbatim straight into MidJourney and boom. Times are getting weird...🤯 ' https://t.co/I6dOJZ4eD8, see more https://t.co/W07vxt3U1h

neutral

ChatGPT can easily replace a good chunk of my Google search right now.

good

The way that ChatGPT can write code for exactly what I want and then explain it to me line for line makes me unequivocally excited for the future.

neutral

ChatGPT has had some very careful finetuning https://t.co/wHd660SFgH

neutral

😳😵‍💫 @OpenAI ChatGPT is insanely intelligent. We’ll be out of jobs in no time. https://t.co/mWuppMLQs6

good

ChatGPT will replace google for many (most?) of my searches going forward. Incredibly good.

neutral

Humans, get ready for the punishments. #ChatGPT #OpenAI https://t.co/kNu9iOWbhq

bad

OpenAI ChatGPT: Optimizing language models for dialogue https://t.co/phhRHp7awy

neutral

Trying ChatGPT today and the results is awesome >> https://t.co/8G8Z4fyO7T

good

Well, looks like I don't need to use Google as much now. ChatGPT really is insane, and doesn't have ads/have to look around websites https://t.co/GN54u52JDh

neutral

The pace of new AI developments is a bit overwhelming. \n\nHaven't even had a chance to play with the new GPT-3 model, which looks amazing, and now I'm dying to test out ChatGPT! #BuildReal

good

Playing with @OpenAI #ChatGPT ... Led the witness a bit... But this thing is smart! @AstroKatie https://t.co/97etyaoQhr

good

Which is the better currency, the US Dollar or #Bitcoin? @OpenAP #ChatGPT https://t.co/RZrRyBq78D

neutral

One thing I so appreciated about WolframAlpha—which was ahead of its time—is being able to give it natural language questions that pulled from a live knowledge base. Once ChatGPT can pull from the internet, it will be unstoppable. https://t.co/gstSHc9hv6

good

RT if you agree with #OpenAI ChatGPT #productmanagement #agile https://t.co/YftnTWoztu

neutral

Wow, ChatGPT is insanely impressive. These are pretty good answers. Sure, a true human expert could likely give more detailed and insightful responses. But these are shockingly good for an instant “mechanical” system. https://t.co/owODmaOFq4

good

Just tried out #ChatGPT .. I'm blown away! This AI language model is like nothing I've ever seen before – it's intelligent, witty, and can hold a conversation about pretty much anything. Definitely give it a try if you want to experience the future of AI! (yes, the AI wrote this)

good

The examples coming out of ChatGPT are absolutely amazing. https://t.co/HXDNjG8EoG

good

ChatGPT describes the ideal physical form for a robot:\n\n"I could help with tasks around the house, provide information and answers to questions, and even offer companionship." https://t.co/4a75FoIKlF

good

Wow is ChatGPT magical; I haven't felt this level of awe in a long time.\n\nSort of like early versions of the Macintosh, or experiencing the Web for the first time.

good

Would have saved your boy $44B if he just waited for the new ChatGPT to drop. https://t.co/EsHjJYU5GT

neutral

I have spent the evening asking ChatGPT about my weirdest interests including 20th Century Egypt, Next Gen Sequencing, cytotoxicity of different gene editing techniques, and changes to DoD procurement under Robert McNamara and holy crap homework is over... https://t.co/47MJ9HhBTn

bad

How’s google done when we can’t even verify if the answers chatgpt dishes out are correct or not? https://t.co/SObROUjy6X

bad

ChatGPT has already taken our jobs, we just don't know it yet

bad

ChatGPT is out of this world\n\nMore evidence that creative tasks are going to be the easy ones for AI... https://t.co/GSLVSnCjEY

good

This new 1-900-#ChatGPT is incredible. https://t.co/AnqmXFVITR

bad

What is ChatGPT going to do with all the bug bounty money that it’s going to collect? https://t.co/9RjGia6ENt

bad

#ChatGPT is just wild. #ConversationalCoding feels like pair programming with an AI engineer. Intuitive to iteratively add functionality or refactor code segments via a dialog. Saved probably 2-3 hours by using #ConversationalCoding to build + improve a prototype - #mindblown

good