#The Prompt Index - A place for inspiration and collaboration

96 messages · Page 1 of 1 (latest)

tiny mason
#

Hi everyone 👋

I just wanted to share out of the love for AI, The Prompt Index is a fast growing database full to the brim of amazing prompts made by amazingly skilled people (some from here!) currently we have around 550 prompts.

It’s completely free, there’s no paywall (why would you pay for prompts!) AI should be accessible to absolutely everyone.

Not only that but I created an image bank for Dalle, Midjourney and Stable Diffusion where you can get inspiration and see how each image was made.

Some features include:

Upvote system
Save your favourite prompts
A weekly newsletter showcasing the best.
Free text search for prompts and image prompts
A learn section with weekly blogs and how to’s

And lots more coming…

I’m reaching out to also ask for feedback as sometimes that’s hard to get. So any feedback is welcome 🙏

Thanks for taking the time to read and happy prompting 🙂

uneven dew
#

Nice job!yes

tiny mason
#

Appreciate that, thank you very much 🙂

sick matrix
#

Thank you , will try to give feedback as I can openai_logo

tiny mason
unborn lion
#

Good job, thanks

alpine ingot
#

Buzz

mortal mango
#

Good Job where is the link or file?

tiny mason
tiny mason
tiny mason
tiny mason
#

Latest blog has just gone up. Which looks at a study by a team of researchers at Johns Hopkins University who basically figured out how to slightly improve the accuracy of outputs and reduce hallucinations.

tiny mason
#

AI tools database now live, along with our reading corner with amazing articles on Midjourney parameters, prompting techniques such as anchoring to get more accurate results and Tree of Thoughts. Feedback welcome 🙂 p.s still no paywalls

paper flint
#

Just found this. Other than the hot pink color, fantastic job! Keep it up and thanks for your efforts!

tiny mason
dusty mango
#

Yea lmao ngl the pink kinda makes it look like a dating website

tiny mason
sick matrix
#

Where to find the index?

tiny mason
proper field
#

Certainly, Tom! Here are 10 popular adjectives that can be used to tap into my knowledge and capabilities:

  1. Informative: I can provide detailed information on a wide range of topics.
  2. Analytical: I can analyze data, trends, and provide insights.
  3. Versatile: I can assist with various tasks, from research to writing.
  4. Instructive: I can provide step-by-step instructions and guidance.
  5. Creative: I can help generate creative ideas and content.
  6. Efficient: I can save you time by quickly delivering information.
  7. Accurate: I strive to provide accurate and reliable information.
  8. Resourceful: I can find solutions and resources for your questions.
  9. Adaptable: I can adjust to different topics and requirements.
  10. Engaging: I aim to keep our conversations interesting and enjoyable.
    That simple!
ornate vale
#

Thanks. Would you consider making a dark mode? Or just get rid of that shocking pink background on everything? It would make a lot easier to explore!

tiny mason
proper field
#

Excellent Index, awesome you shared this. Thanks 🤝

tiny mason
acoustic cave
#

Great site!Would you consider letting the users comment?

tiny mason
acoustic cave
tiny mason
acoustic cave
tiny mason
#

Sure

fleet umbra
#

where i can find it?

tiny mason
#

You just have to give us a google “The Prompt Index” which we rank number 1 for and also “Prompt Database”

modern narwhal
tiny mason
twin herald
#

Mother lord that colors XD do please give options that are a little more color blind friendly. I have partial red blindness and purple is messing with my eyes a lot.
Cool project though.

tiny mason
#

I'm sorry my dude, i'll get onto the colours asap. Which colour is best for colour blindness?

tiny mason
#

Feel free to join our telegram and get involved in the conversation along with help with prompts, resources and much more just search ChatGPTMastermind

ornate vale
tiny mason
tiny mason
#

Latest blog “Demystifying Tokens: A Beginners Guide To Understanding AI Building Blocks” has just gone up 🙂

ornate vale
#

Looking forward to reading it when I can read the text. 🙂

tiny mason
twin herald
tiny mason
twin herald
#

(still to white but I don't have the knowledge to change every element do dark gray to see how it would look)

twin herald
#

No probs. Great resource 🥹

tiny mason
tiny mason
#

Latest blog is up, looking at how emotion can improve outputs. Based on a research paper by some folks at Microsoft. Has some interesting key takeaways from your prompts.

crimson cargo
#

this is very helpful!

tiny mason
tiny mason
#

Latest blog is up looking at how the AI safeguarding is not as high of a standard when using less resourced langauges such as Zulu or other less commonly used Dialects.

void sequoia
#

Thank you, very appreciated 🙏

tiny mason
tiny mason
#

Just a reminder, if you have an account (which is free) you can make your own prompt library where you can save all your favourite prompts 🙂

spare orchid
#

Thank you for the callout in your newsletter.

tiny mason
tiny mason
#

Latest blog is up. Feedback welcome

ornate vale
#

Are there any good prompts for the voice feature? I mean prompts that lend themselves to improving the voice experience? I'm finding that creating interactive quizzes where it asks you a question and do you answer and it gives you feedback can be very useful.

tiny mason
#

Voice input? Hmm not specifically for the voice feature but there will be some prompts that can improve the quiz feature.

tiny mason
#

The latest blog is up, here’s what I could fit in

PsyCoT: A Window into the Future of AI That Can Perceive Our Personalities and Moods

It’s based of a recent, Interesting research paper by Yang et al

Here it is:

Personality detection involves identifying an individual's personality traits purely from their written text. This capability has many potential applications, like improving human-computer interaction. However, traditional machine learning models struggle to capture the nuances required for accurate personality detection from texts.

Researchers from Sun Yat-sen University propose a new method called PsyCoT that takes inspiration from psychological questionnaires to significantly boost the performance of large language models (LLMs) like GPT-3.5 for zero-shot personality detection.

The PsyCoT Approach:
PsyCoT treats personality detection as a multi-step reasoning task, similar to a person completing a personality questionnaire. It has the LLM take on the role of an AI assistant specializing in text analysis.

Over multiple dialogue turns, the assistant is prompted to rate relevant questionnaire items based on the input text, with access to previous ratings. After rating all items, the assistant makes an overall personality judgment. This mimics the human process of reasoning through a questionnaire to evaluate someone's personality. The items act as an effective chain-of-thought, enhancing the LLM's capability for nuanced personality inference.

Key Takeaways:
PsyCoT prompts the LLM to rate questionnaire items in a multi-turn dialogue, mimicking how humans complete personality tests. This acts as an effective chain-of-thought.

Full blog on The Prompt Index

rustic magnet
#

How do I get to the URL for this project?

tiny mason
#

Just search for it on google

tiny mason
#

Latest blog is up, based on the serious vulnerabilities around the new GPTs.

tough atlas
#

This has been a thing all year round

tiny mason
#

Q* rumour has been confirmed by Sam as a leak as well

torpid remnant
#

So i made a Scientific Method in NLP chain of though to train gpts on the actual scientific method, which hasn’t been done this way so far : ##### COPY/Paste the code below to PDF and upload to your Knowledge Base. #####


The document outlines a structured workflow for applying the scientific method in Natural Language Processing (NLP), encompassing several key CoT stages:

Observation: This initial phase involves observing phenomena or data that sparks curiosity, laying the groundwork for further investigation.

Question: Following the observation, a relevant question is formulated to explain the observation or solve a problem.

Hypothesis: A testable prediction or educated guess is proposed to explain the question arising from the observation.

Experiment: The hypothesis is tested through controlled, repeatable procedures designed to collect data.

Analysis: Data collected from the experiment is analyzed, often using statistical methods, to determine if the results support or refute the hypothesis.

Conclusion: Interpretation of the data analysis results occurs. If results align with the hypothesis, it is supported; otherwise, it may be revised or rejected.

Communication: Results are communicated, often through publishing in scientific journals, presenting at conferences, or other forms of sharing. This step is crucial for peer review and contributing to scientific knowledge.

Reiteration: The scientific method is iterative, with the steps being repeated to refine hypotheses, explore new questions, and build upon previous research.

This approach reflects a dynamic, continuous process of scientific inquiry, with each stage contributing to expanding knowledge and understanding of the natural world. The document emphasizes maintaining a balance between scientific rigor and articulate, engaging communication.


#

The bot can then be reinforced with these 2 images

calm fossil
#

Just to verify is this the right place to signup ?

tiny mason
white aurora
#

@tiny mason Great!!!

#

@tiny mason How to access that wonderful database?

tiny mason
tiny mason
somber canyon
#

@tiny mason where can I see the prompt index?

tiny mason
sacred turtle
tiny mason
foggy hill
#

игяя

silent hull
#

hi

bright copper
#

@tiny mason @tiny mason @tiny mason @tiny mason 😋

junior hedge
#

buzz

vernal vapor
#

Legend

tiny mason
tiny mason
tiny mason
tiny mason