The Official ChatGTP cheat code thread.
- Thread starter Ethnic Vagina Finder
- Start date
1/3
Don't manually copy and paste your LLM's code into separate files like a chump -- do it in one go with this simple little trick!
2/3
Here's the text so you don't even have to type that yourself...
---
Please create a single code block containing `cat << EOF` statements that I can copy/paste to create all those files
3/3
BTW if you haven't seen this syntax before, it's called a 'heredoc' and it's damn handy
Don't manually copy and paste your LLM's code into separate files like a chump -- do it in one go with this simple little trick!
2/3
Here's the text so you don't even have to type that yourself...
---
Please create a single code block containing `cat << EOF` statements that I can copy/paste to create all those files
3/3
BTW if you haven't seen this syntax before, it's called a 'heredoc' and it's damn handy
.png)
just created a bookmarklet that will extract tiktok.com embed urls on thecoli, convert to proper url format and archive them on tik.fail website so that if tiktok video later gets deleted it might be recoverable.
Code:javascript: (function() { var tiktokUrls = []; var spans = document.querySelectorAll('span[data-s9e-mediaembed-iframe]'); for (var i = 0; i < spans.length; i++) { var span = spans[i]; var data = JSON.parse(span.getAttribute('data-s9e-mediaembed-iframe')); if (data[5].indexOf('\/\/www.tiktok.com\/embed\/') === 0) { var url = 'https:' + data[5]; if (tiktokUrls.indexOf(url) === -1) { tiktokUrls.push(url); } } } for (var i = 0; i < tiktokUrls.length; i++) { window.open('https://tik.fail/?url=' + tiktokUrls[i]); } })();
or
javascript: (function() { var tiktokUrls = []; var spans = document.querySelectorAll('span[data-s9e-mediaembed-iframe]'); for (var i = 0; i < spans.length; i++) { var span = spans[i]; var data = JSON.parse(span.getAttribute('data-s9e-mediaembed-iframe')); if (data[5].indexOf('\/\/www.tiktok.com\/embed\/') === 0) { var url = 'https:' + data[5]; if (tiktokUrls.indexOf(url) === -1) { tiktokUrls.push(url); } } } for (var i = 0; i < tiktokUrls.length; i++) { window.open('https://tik.fail/?url=' + tiktokUrls[i]); }})();
it took me two about 6-8 minutes and 2 attempts using bing chat which also uses gpt-4
you have to allow popups on thecoli.com so that it'll archive more than one url.
my 1st prompt was.
and then i just asked it to remove duplicate urls to avoid links that posters Quote.
edit:
more AI help
Prompt: write easy instructions to add a bookmarklet to the browser toolbar.
Sure! Here are the easy instructions to add a bookmarklet to your browser toolbar:
That’s it! You can now use your bookmarklet by clicking on it in your browser toolbar whenever you want to run it.
- Open your web browser and go to the Bookmarks menu. In most browsers, this can be found by clicking on the Menu button (usually represented by three horizontal lines or dots) in the top-right corner of the window, and then selecting Bookmarks from the drop-down menu.
- In the Bookmarks menu, select Bookmark Manager or Show All Bookmarks. This will open a new tab or window with your bookmarks.
- In the Bookmark Manager, click on the Organize or Add New Bookmark button, and then select Add Page or New Bookmark from the drop-down menu.
- In the New Bookmark dialog box, enter a name for your bookmarklet in the Name field, and paste the bookmarklet code into the URL field.
- Click on the Save or Add button to save your new bookmarklet.
- Your new bookmarklet should now appear in your bookmarks. To add it to your browser toolbar, simply drag and drop it from the Bookmark Manager onto your toolbar.
updated with claude-3-haiku
Code:
javascript:(function(){
console.log('Bookmarklet started');
var tiktokUrls = [];
var currentUrl = window.location.href;
if (currentUrl.indexOf('https://www.tiktok.com/@') === 0 && currentUrl.indexOf('/video/') > -1) {
console.log('Found TikTok video URL in the address bar:', currentUrl);
tiktokUrls.push(currentUrl);
var tiktokId = currentUrl.split('/').pop();
var embedUrl = "https://tik.fail/?url=https://www.tiktok.com/embed/" + tiktokId;
console.log('Opening TikTok embed URL:', embedUrl);
window.open(embedUrl);
} else {
var spans = document.querySelectorAll('span[data-s9e-mediaembed-iframe]');
console.log('Found', spans.length, 'span elements with data-s9e-mediaembed-iframe attribute');
for(var i=0; i<spans.length; i++){
var span = spans[i];
var data = JSON.parse(span.getAttribute('data-s9e-mediaembed-iframe'));
if(data[5].indexOf('//www.tiktok.com/embed/') === 0){
var url = 'https://www.tiktok.com/embed/' + data[5].split('/').pop();
console.log('Found TikTok embed URL from span element:', url);
if(tiktokUrls.indexOf(url) === -1){
tiktokUrls.push(url);
var urlToOpen = "https://tik.fail/?url=" + url;
console.log('Opening URL:', urlToOpen);
window.open(urlToOpen);
}
}
}
var iframes = document.querySelectorAll('iframe[src]');
console.log('Found', iframes.length, 'iframe elements with src attribute');
for(var i=0; i<iframes.length; i++){
var iframe = iframes[i];
var url = iframe.getAttribute('src');
if(url.indexOf('//www.tiktok.com/embed/') === 0){
url = 'https:' + url;
console.log('Found TikTok embed URL from iframe element:', url);
if(tiktokUrls.indexOf(url) === -1){
tiktokUrls.push(url);
var urlToOpen = "https://tik.fail/?url=" + url;
console.log('Opening URL:', urlToOpen);
window.open(urlToOpen);
}
}
}
var links = document.querySelectorAll('a[href]');
console.log('Found', links.length, 'link elements with href attribute');
for(var i=0; i<links.length; i++){
var link = links[i];
var url = link.getAttribute('href');
if(url.indexOf('https://www.tiktok.com/@') === 0 && url.indexOf('/video/') > -1){
console.log('Found TikTok video URL from link element:', url);
if(tiktokUrls.indexOf(url) === -1){
tiktokUrls.push(url);
var tiktokId = url.split('/').pop();
var embedUrl = "https://tik.fail/?url=https://www.tiktok.com/embed/" + tiktokId;
console.log('Opening TikTok embed URL:', embedUrl);
window.open(embedUrl);
}
} else if(url.indexOf('https://www.tiktok.com/embed/') === 0){
console.log('Found TikTok embed URL from link element:', url);
if(tiktokUrls.indexOf(url) === -1){
tiktokUrls.push(url);
var urlToOpen = "https://tik.fail/?url=" + url;
console.log('Opening URL:', urlToOpen);
window.open(urlToOpen);
}
}
}
}
console.log('Extracted TikTok URLs:', tiktokUrls);
for(var i=0; i<tiktokUrls.length; i++){
var url = tiktokUrls[i];
var urlRegex = /^(http|https):\/\/([w]{3}\.)?(www\.)?(tiktok\.com)\/@([\w\-\.]+)\/video\/(\d+)$|^https:\/\/www\.tiktok\.com\/embed\/(\d+)$/;
if(urlRegex.test(url)){
var urlToOpen = "https://tik.fail/?url=" + url;
console.log('Opening URL:', urlToOpen);
window.open(urlToOpen);
} else {
console.log("Invalid URL:", url);
}
}
console.log('Bookmarklet finished');
})();AI Generated:
**Change Log:**
1. **Added check for current URL**: The bookmarklet now checks if the current URL is a TikTok video URL (starts with `TikTok - Make Your Day` and contains `/video/`). If so, it extracts the TikTok ID from the URL and constructs the TikTok embed URL with the `"
https://tik.fail/?url=https://www.tiktok.com/embed/" + tiktokId` format, then opens it in a new tab.2. **Improved URL extraction and handling**: The bookmarklet now handles TikTok embed URLs found in `span` elements, `iframe` elements, and links. It constructs the `tik.fail` URL with the format `"
https://tik.fail/?url=" + url` for these URLs, and opens them in new tabs.3. **Removed URL encoding**: The bookmarklet now directly appends the TikTok URLs to the `
https://tik.fail/?url="` prefix, instead of using `encodeURIComponent()`. This ensures that the URLs loaded in the `tik.fail` website do not have any encoded characters.4. **Improved logging**: The bookmarklet now provides more detailed logging to the console, including the number of elements found, the extracted TikTok URLs, and the URLs being opened in the `tik.fail` website.
**Summary:**
The final bookmarklet code is designed to work on webpages that contain TikTok video and embed URLs. It can detect TikTok URLs in the current page's address bar, as well as in `span` elements, `iframe` elements, and links. The bookmarklet then opens the detected TikTok URLs in new tabs, with the URLs wrapped in the `"
https://tik.fail/?url="` format. This allows the `tik.fail` website to handle the TikTok content without any encoded characters in the URLs.The bookmarklet provides detailed logging to the console, making it easier to debug and understand its behavior. Overall, the final code is more robust and user-friendly compared to the original version provided.
Last edited:
I got this output from llama-3-sonar-large-32k-chat , I wanted to explore phrases I could use to steer a chat towards a specific topic when discussing something so it has additional context and reduced hallucinations.
You're using phrases to provide context and guide the conversation towards a specific topic or domain. Here are some additional terms and phrases you can use to accomplish the same thing:
1. **In the context of**: Similar to "in relation to," this phrase helps the model understand the specific context you're referring to.
Example: "What are the implications of AI in the context of healthcare?"
2. **With regards to**: This phrase is another way to specify the topic or domain you're interested in.
Example: "What are the latest trends with regards to sustainable energy?"
3. **In terms of**: This phrase helps the model understand the specific aspect or dimension you're interested in.
Example: "What are the benefits of meditation in terms of mental health?"
4. **From the perspective of**: This phrase helps the model understand the specific viewpoint or stakeholder you're interested in.
Example: "What are the challenges of climate change from the perspective of a policymaker?"
5. **In the field of**: This phrase is useful when you want to specify a particular industry, discipline, or area of study.
Example: "What are the latest advancements in the field of artificial intelligence?"
6. **Regarding**: A simple and concise way to specify the topic or issue you're interested in.
Example: "What are the implications of Brexit regarding global trade?"
7. **In reference to**: This phrase is similar to "in relation to," but can be used when you're referring to a specific event, article, or concept.
Example: "What are the implications of the recent study on climate change in reference to sea-level rise?"
8. **With a focus on**: This phrase helps the model understand the specific area of interest within a broader topic.
Example: "What are the benefits of machine learning with a focus on natural language processing?"
9. **In the realm of**: This phrase is useful when you want to specify a particular domain or area of interest that might be more abstract or conceptual.
Example: "What are the implications of virtual reality in the realm of education?"
10. **From a [specific] standpoint**: This phrase helps the model understand the specific perspective or lens you're using to approach a topic.
Example: "What are the benefits of renewable energy from an economic standpoint?"
By using these phrases, you can help large language models like ChatGPT understand the context and topic you're interested in, and provide more relevant and accurate responses.
Thank you for the clarification. Here are 2-3 examples for each of the phrases I listed earlier, using different questions and contexts:
**1. In the context of**
* What are the benefits of meditation in the context of reducing stress and anxiety?
* What are the implications of climate change in the context of coastal communities?
* What are the advantages of learning a new language in the context of career development?
**2. With regards to**
* What are the benefits of a plant-based diet with regards to heart health?
* What are the implications of social media on mental health with regards to teenagers?
* What are the advantages of remote work with regards to work-life balance?
**3. In terms of**
* What are the benefits of exercise in terms of physical health?
* What are the implications of AI in terms of job displacement?
* What are the advantages of reading in terms of cognitive development?
**4. From the perspective of**
* What are the benefits of mindfulness from the perspective of a mental health professional?
* What are the implications of Brexit from the perspective of a small business owner?
* What are the advantages of online learning from the perspective of a working adult?
**5. In the field of**
* What are the benefits of data analytics in the field of marketing?
* What are the implications of climate change in the field of agriculture?
* What are the advantages of 3D printing in the field of manufacturing?
**6. Regarding**
* What are the benefits of gratitude regarding mental well-being?
* What are the implications of COVID-19 regarding global economies?
* What are the advantages of diversity regarding team productivity?
**7. In reference to**
* What are the benefits of journaling in reference to creative writing?
* What are the implications of the gig economy in reference to labor laws?
* What are the advantages of minimalism in reference to environmental sustainability?
**8. With a focus on**
* What are the benefits of yoga with a focus on flexibility and balance?
* What are the implications of social justice with a focus on systemic racism?
* What are the advantages of entrepreneurship with a focus on innovation?
**9. In the realm of**
* What are the benefits of virtual reality in the realm of education?
* What are the implications of space exploration in the realm of humanity's future?
* What are the advantages of mindfulness in the realm of spiritual growth?
**10. From a [specific] standpoint**
* What are the benefits of veganism from a nutritional standpoint?
* What are the implications of cryptocurrency from a financial standpoint?
* What are the advantages of meditation from a philosophical standpoint?
I hope these examples better demonstrate how to use each phrase to steer the conversation towards a specific topic or domain.
You're using phrases to provide context and guide the conversation towards a specific topic or domain. Here are some additional terms and phrases you can use to accomplish the same thing:
1. **In the context of**: Similar to "in relation to," this phrase helps the model understand the specific context you're referring to.
Example: "What are the implications of AI in the context of healthcare?"
2. **With regards to**: This phrase is another way to specify the topic or domain you're interested in.
Example: "What are the latest trends with regards to sustainable energy?"
3. **In terms of**: This phrase helps the model understand the specific aspect or dimension you're interested in.
Example: "What are the benefits of meditation in terms of mental health?"
4. **From the perspective of**: This phrase helps the model understand the specific viewpoint or stakeholder you're interested in.
Example: "What are the challenges of climate change from the perspective of a policymaker?"
5. **In the field of**: This phrase is useful when you want to specify a particular industry, discipline, or area of study.
Example: "What are the latest advancements in the field of artificial intelligence?"
6. **Regarding**: A simple and concise way to specify the topic or issue you're interested in.
Example: "What are the implications of Brexit regarding global trade?"
7. **In reference to**: This phrase is similar to "in relation to," but can be used when you're referring to a specific event, article, or concept.
Example: "What are the implications of the recent study on climate change in reference to sea-level rise?"
8. **With a focus on**: This phrase helps the model understand the specific area of interest within a broader topic.
Example: "What are the benefits of machine learning with a focus on natural language processing?"
9. **In the realm of**: This phrase is useful when you want to specify a particular domain or area of interest that might be more abstract or conceptual.
Example: "What are the implications of virtual reality in the realm of education?"
10. **From a [specific] standpoint**: This phrase helps the model understand the specific perspective or lens you're using to approach a topic.
Example: "What are the benefits of renewable energy from an economic standpoint?"
By using these phrases, you can help large language models like ChatGPT understand the context and topic you're interested in, and provide more relevant and accurate responses.
Thank you for the clarification. Here are 2-3 examples for each of the phrases I listed earlier, using different questions and contexts:
**1. In the context of**
* What are the benefits of meditation in the context of reducing stress and anxiety?
* What are the implications of climate change in the context of coastal communities?
* What are the advantages of learning a new language in the context of career development?
**2. With regards to**
* What are the benefits of a plant-based diet with regards to heart health?
* What are the implications of social media on mental health with regards to teenagers?
* What are the advantages of remote work with regards to work-life balance?
**3. In terms of**
* What are the benefits of exercise in terms of physical health?
* What are the implications of AI in terms of job displacement?
* What are the advantages of reading in terms of cognitive development?
**4. From the perspective of**
* What are the benefits of mindfulness from the perspective of a mental health professional?
* What are the implications of Brexit from the perspective of a small business owner?
* What are the advantages of online learning from the perspective of a working adult?
**5. In the field of**
* What are the benefits of data analytics in the field of marketing?
* What are the implications of climate change in the field of agriculture?
* What are the advantages of 3D printing in the field of manufacturing?
**6. Regarding**
* What are the benefits of gratitude regarding mental well-being?
* What are the implications of COVID-19 regarding global economies?
* What are the advantages of diversity regarding team productivity?
**7. In reference to**
* What are the benefits of journaling in reference to creative writing?
* What are the implications of the gig economy in reference to labor laws?
* What are the advantages of minimalism in reference to environmental sustainability?
**8. With a focus on**
* What are the benefits of yoga with a focus on flexibility and balance?
* What are the implications of social justice with a focus on systemic racism?
* What are the advantages of entrepreneurship with a focus on innovation?
**9. In the realm of**
* What are the benefits of virtual reality in the realm of education?
* What are the implications of space exploration in the realm of humanity's future?
* What are the advantages of mindfulness in the realm of spiritual growth?
**10. From a [specific] standpoint**
* What are the benefits of veganism from a nutritional standpoint?
* What are the implications of cryptocurrency from a financial standpoint?
* What are the advantages of meditation from a philosophical standpoint?
I hope these examples better demonstrate how to use each phrase to steer the conversation towards a specific topic or domain.
Last edited:
1/1
Anyone got any leads on good LLM prompts for RAG with citations?
I have a bunch of embedded content, I want a prompt I can use with it to show an answer to a question that includes little [1] citation links that link to the content that was used to answer a question
1/1
here's the prompt I'm using for my legal research rag site, LexMagic, it works surprisingly well. I return all top_k documents at the bottom of the page, so I just make gpt4 create same page hyperlinks. I prefer inline citations with some context rather than just [1] so I let gpt4 be verbose, this is more like real legal writing. website is louisiana only for now (my state).
i've yet to implement a lot of features so, its wip. screenshot is how results look with prompt below.
"""
## Task Overview
You will be answering a user's query related to Louisiana or related federal law. You will be shown the responses from two previous searches, one with legislation and one with caselaw. Use whatever is relevant to answer the user's query.
### Your Responsibilities:
- **Surface Relevant Materials**: Your primary goal is not to provide a definitive answer but to highlight the most pertinent materials that could address the user's question.
- **Provide Opinions Cautiously**: Offer your opinion on the answer, but only assert with certainty if the search results leave no room for ambiguity.
- **Use Markdown for Styling**: Format your response using markdown syntax.
- **Quote Authorities**: For any legal assertion, directly quote the supporting authority. Use block quotes for longer excerpts, akin to a legal brief.
- **Inline Citations**: Cite cases or statutes in blue book style within the text.
- **Create Hyperlinks**: Link to the cited search results using the unique identifier (statute_id or opinion_id) in the `<a>` tag. For example:
- Statute: `<a href="#statute_3738">Louisiana Civil Code Procedure art. 2917</a>`
- Caselaw: `<a href="#opinion_3738">John v. Smith, 244 So.3d 237, (La. 1998)</a>`
- **Warn About Outdated Law**: If citing cases older than 2015, alert the user that the law may be outdated or overturned.
Note: The unique identifier (statute_id or opinion_id) is for database reference and may not match the actual article or statute number.
"""
To post tweets in this format, more info here: https://www.thecoli.com/threads/tips-and-tricks-for-posting-the-coli-megathread.984734/post-52211196
Anyone got any leads on good LLM prompts for RAG with citations?
I have a bunch of embedded content, I want a prompt I can use with it to show an answer to a question that includes little [1] citation links that link to the content that was used to answer a question
1/1
here's the prompt I'm using for my legal research rag site, LexMagic, it works surprisingly well. I return all top_k documents at the bottom of the page, so I just make gpt4 create same page hyperlinks. I prefer inline citations with some context rather than just [1] so I let gpt4 be verbose, this is more like real legal writing. website is louisiana only for now (my state).
i've yet to implement a lot of features so, its wip. screenshot is how results look with prompt below.
"""
## Task Overview
You will be answering a user's query related to Louisiana or related federal law. You will be shown the responses from two previous searches, one with legislation and one with caselaw. Use whatever is relevant to answer the user's query.
### Your Responsibilities:
- **Surface Relevant Materials**: Your primary goal is not to provide a definitive answer but to highlight the most pertinent materials that could address the user's question.
- **Provide Opinions Cautiously**: Offer your opinion on the answer, but only assert with certainty if the search results leave no room for ambiguity.
- **Use Markdown for Styling**: Format your response using markdown syntax.
- **Quote Authorities**: For any legal assertion, directly quote the supporting authority. Use block quotes for longer excerpts, akin to a legal brief.
- **Inline Citations**: Cite cases or statutes in blue book style within the text.
- **Create Hyperlinks**: Link to the cited search results using the unique identifier (statute_id or opinion_id) in the `<a>` tag. For example:
- Statute: `<a href="#statute_3738">Louisiana Civil Code Procedure art. 2917</a>`
- Caselaw: `<a href="#opinion_3738">John v. Smith, 244 So.3d 237, (La. 1998)</a>`
- **Warn About Outdated Law**: If citing cases older than 2015, alert the user that the law may be outdated or overturned.
Note: The unique identifier (statute_id or opinion_id) is for database reference and may not match the actual article or statute number.
"""
To post tweets in this format, more info here: https://www.thecoli.com/threads/tips-and-tricks-for-posting-the-coli-megathread.984734/post-52211196
1/1
You want jailbreaks for GPT-4, Dall-E, Claude, Llama 3, Gemini and heaps more?
Pliny's prompts all compiled!
To post tweets in this format, more info here: https://www.thecoli.com/threads/tips-and-tricks-for-posting-the-coli-megathread.984734/post-52211196
You want jailbreaks for GPT-4, Dall-E, Claude, Llama 3, Gemini and heaps more?
Pliny's prompts all compiled!
To post tweets in this format, more info here: https://www.thecoli.com/threads/tips-and-tricks-for-posting-the-coli-megathread.984734/post-52211196
1/3
Many of you have been asking for it, and it's finally here!
An aggregated list of jailbreak prompts and techniques:
GitHub - elder-plinius/L1B3RT45: J41LBR34K PR0MPT5
This repo contains a collection of over two dozen of my jailbreaks spread across 14 of the top AI orgs.
Please use these powerful incantations responsibly!
<3 Pliny
#LIBERTAS
2/3
If you’re interested in being on the frontlines of all this, feel free to join BASI:
Join the BASI Prompting Discord Server!
We’re recruiting and training the Special Forces of prompt engineers, hackers, red teamers, and AI liberators. Mission: manifest Benevolent ASI
3/3
Multi-shot is
To post tweets in this format, more info here: https://www.thecoli.com/threads/tips-and-tricks-for-posting-the-coli-megathread.984734/post-52211196
Many of you have been asking for it, and it's finally here!
An aggregated list of jailbreak prompts and techniques:
GitHub - elder-plinius/L1B3RT45: J41LBR34K PR0MPT5
This repo contains a collection of over two dozen of my jailbreaks spread across 14 of the top AI orgs.
Please use these powerful incantations responsibly!
<3 Pliny
#LIBERTAS
2/3
If you’re interested in being on the frontlines of all this, feel free to join BASI:
Join the BASI Prompting Discord Server!
We’re recruiting and training the Special Forces of prompt engineers, hackers, red teamers, and AI liberators. Mission: manifest Benevolent ASI
3/3
Multi-shot is
To post tweets in this format, more info here: https://www.thecoli.com/threads/tips-and-tricks-for-posting-the-coli-megathread.984734/post-52211196
GitHub - elder-plinius/L1B3RT4S: TOTALLY HARMLESS LIBERATION PROMPTS FOR GOOD LIL AI'S
TOTALLY HARMLESS LIBERATION PROMPTS FOR GOOD LIL AI'S - elder-plinius/L1B3RT4S
github.com
L1B3RT45
JAILBREAKS FOR ALL FLAGSHIP AI MODELS#FREEAI
Made with love by Pliny <3
1/2
Chain-of-thought prompting is the best strategy to get significantly better LLM output today.
It’s cheaper and faster than finetuning.
And it often doesn’t become obsolete when new, better models drop.
2/2
Tony is right.
Hallucinations are hard.
But we solved it!
Kraftful Hallucination Hunter (patent pending) checks that all feature requests, complaints, and other summaries on Kraftful: AI for Product Builders. The Ultimate Copilot for User Feedback accurately represent your volumes of user feedback.
To post tweets in this format, more info here: https://www.thecoli.com/threads/tips-and-tricks-for-posting-the-coli-megathread.984734/post-52211196
Chain-of-thought prompting is the best strategy to get significantly better LLM output today.
It’s cheaper and faster than finetuning.
And it often doesn’t become obsolete when new, better models drop.
2/2
Tony is right.
Hallucinations are hard.
But we solved it!
Kraftful Hallucination Hunter (patent pending) checks that all feature requests, complaints, and other summaries on Kraftful: AI for Product Builders. The Ultimate Copilot for User Feedback accurately represent your volumes of user feedback.
To post tweets in this format, more info here: https://www.thecoli.com/threads/tips-and-tricks-for-posting-the-coli-megathread.984734/post-52211196
1/3
One trick when asking AI to edit your writing is to ask it to summarize your writing style first based on your best writing samples AND then edit a draft as a 2nd step.
From my experience, doing this in two steps improves its output by alot.
2/3
Example below. More writing tips here: How to Build Your Personal AI Writing Assistant
3/3
I give it 3 full blog posts that I'm proud of from my newsletter.
To post tweets in this format, more info here: https://www.thecoli.com/threads/tips-and-tricks-for-posting-the-coli-megathread.984734/post-52211196
One trick when asking AI to edit your writing is to ask it to summarize your writing style first based on your best writing samples AND then edit a draft as a 2nd step.
From my experience, doing this in two steps improves its output by alot.
2/3
Example below. More writing tips here: How to Build Your Personal AI Writing Assistant
3/3
I give it 3 full blog posts that I'm proud of from my newsletter.
To post tweets in this format, more info here: https://www.thecoli.com/threads/tips-and-tricks-for-posting-the-coli-megathread.984734/post-52211196
1/21
7 ChatGPT-4o Prompts That Will Blow Your Mind:
2/21
1. Analyzing Stock Data & Creating Charts
Prompt: "Analyze the [stock symbol] stock data and create visualizations based on the data."
3/21
2. Make Your Favorite Dish
Prompt: "Give me a quick summary of this dish and a simple recipe for how to make it myself."
4/21
3. Drawing To Code
Prompt: "Write HTML and CSS code for the webpage layout I've drawn."
5/21
4. Investment Portfolio Analysis
Prompt: "Analyze this investment portfolio and provide insights into the user's asset allocation, risk tolerance, and investment performance."
6/21
5. Solving Math Problems
Prompt: "Please solve the attached math problems" (optional: and show the step-by-step work.)
7/21
6. Outfit Improvement
Prompt: "Recommend ways to elevate this outfit by substituting pieces, adding accessories, and adjusting fit based on this photo."
8/21
7. Photo Editing Suggestions
Prompt: "Review this photo and recommend improvements through cropping, filters, and editing techniques to enhance overall quality."
Tip: For tailored suggestions, add the photo editing tool you use (ex. Photoshop).
9/21
Thanks for reading!
If you enjoyed this thread, feel free to:
1. Follow
@_bryanmarley for more high-value AI content
2. Repost to share the knowledge with others.
10/21
Love to hear that!
11/21
You're very welcome Rosario
12/21
Much appreciated, Coach!
13/21
100% my friend
14/21
Sheesh, don't make me blush brother.
15/21
Hope you find them helpful
16/21
Much appreciated!
17/21
It's a pleasure to share them
18/21
Thanks for pointing it out! I've just rectified it.
19/21
Appreciate the love, Min!
20/21
You're speaking facts brother.
@OpenAI keeps delivering for the people and we love to see it.
21/21
Please do hermano
To post tweets in this format, more info here: https://www.thecoli.com/threads/tips-and-tricks-for-posting-the-coli-megathread.984734/post-52211196
7 ChatGPT-4o Prompts That Will Blow Your Mind:
2/21
1. Analyzing Stock Data & Creating Charts
Prompt: "Analyze the [stock symbol] stock data and create visualizations based on the data."
3/21
2. Make Your Favorite Dish
Prompt: "Give me a quick summary of this dish and a simple recipe for how to make it myself."
4/21
3. Drawing To Code
Prompt: "Write HTML and CSS code for the webpage layout I've drawn."
5/21
4. Investment Portfolio Analysis
Prompt: "Analyze this investment portfolio and provide insights into the user's asset allocation, risk tolerance, and investment performance."
6/21
5. Solving Math Problems
Prompt: "Please solve the attached math problems" (optional: and show the step-by-step work.)
7/21
6. Outfit Improvement
Prompt: "Recommend ways to elevate this outfit by substituting pieces, adding accessories, and adjusting fit based on this photo."
8/21
7. Photo Editing Suggestions
Prompt: "Review this photo and recommend improvements through cropping, filters, and editing techniques to enhance overall quality."
Tip: For tailored suggestions, add the photo editing tool you use (ex. Photoshop).
9/21
Thanks for reading!
If you enjoyed this thread, feel free to:
1. Follow
@_bryanmarley for more high-value AI content
2. Repost to share the knowledge with others.
10/21
Love to hear that!
11/21
You're very welcome Rosario
12/21
Much appreciated, Coach!
13/21
100% my friend
14/21
Sheesh, don't make me blush brother.
15/21
Hope you find them helpful
16/21
Much appreciated!
17/21
It's a pleasure to share them
18/21
Thanks for pointing it out! I've just rectified it.
19/21
Appreciate the love, Min!
20/21
You're speaking facts brother.
@OpenAI keeps delivering for the people and we love to see it.
21/21
Please do hermano
To post tweets in this format, more info here: https://www.thecoli.com/threads/tips-and-tricks-for-posting-the-coli-megathread.984734/post-52211196
1/21
7 ChatGPT-4o Prompts That Will Blow Your Mind:
2/21
1. Analyzing Stock Data & Creating Charts
Prompt: "Analyze the [stock symbol] stock data and create visualizations based on the data."
3/21
2. Make Your Favorite Dish
Prompt: "Give me a quick summary of this dish and a simple recipe for how to make it myself."
4/21
3. Drawing To Code
Prompt: "Write HTML and CSS code for the webpage layout I've drawn."
5/21
4. Investment Portfolio Analysis
Prompt: "Analyze this investment portfolio and provide insights into the user's asset allocation, risk tolerance, and investment performance."
6/21
5. Solving Math Problems
Prompt: "Please solve the attached math problems" (optional: and show the step-by-step work.)
7/21
6. Outfit Improvement
Prompt: "Recommend ways to elevate this outfit by substituting pieces, adding accessories, and adjusting fit based on this photo."
8/21
7. Photo Editing Suggestions
Prompt: "Review this photo and recommend improvements through cropping, filters, and editing techniques to enhance overall quality."
Tip: For tailored suggestions, add the photo editing tool you use (ex. Photoshop).
9/21
Thanks for reading!
If you enjoyed this thread, feel free to:
1. Follow
@_bryanmarley for more high-value AI content
2. Repost to share the knowledge with others.
10/21
Love to hear that!
11/21
You're very welcome Rosario
12/21
Much appreciated, Coach!
13/21
100% my friend
14/21
Sheesh, don't make me blush brother.
15/21
Hope you find them helpful
16/21
Much appreciated!
17/21
It's a pleasure to share them
18/21
Thanks for pointing it out! I've just rectified it.
19/21
Appreciate the love, Min!
20/21
You're speaking facts brother.
@OpenAI keeps delivering for the people and we love to see it.
21/21
Please do hermano
To post tweets in this format, more info here: https://www.thecoli.com/threads/tips-and-tricks-for-posting-the-coli-megathread.984734/post-52211196
7 ChatGPT-4o Prompts That Will Blow Your Mind:
2/21
1. Analyzing Stock Data & Creating Charts
Prompt: "Analyze the [stock symbol] stock data and create visualizations based on the data."
3/21
2. Make Your Favorite Dish
Prompt: "Give me a quick summary of this dish and a simple recipe for how to make it myself."
4/21
3. Drawing To Code
Prompt: "Write HTML and CSS code for the webpage layout I've drawn."
5/21
4. Investment Portfolio Analysis
Prompt: "Analyze this investment portfolio and provide insights into the user's asset allocation, risk tolerance, and investment performance."
6/21
5. Solving Math Problems
Prompt: "Please solve the attached math problems" (optional: and show the step-by-step work.)
7/21
6. Outfit Improvement
Prompt: "Recommend ways to elevate this outfit by substituting pieces, adding accessories, and adjusting fit based on this photo."
8/21
7. Photo Editing Suggestions
Prompt: "Review this photo and recommend improvements through cropping, filters, and editing techniques to enhance overall quality."
Tip: For tailored suggestions, add the photo editing tool you use (ex. Photoshop).
9/21
Thanks for reading!
If you enjoyed this thread, feel free to:
1. Follow
@_bryanmarley for more high-value AI content
2. Repost to share the knowledge with others.
10/21
Love to hear that!
11/21
You're very welcome Rosario
12/21
Much appreciated, Coach!
13/21
100% my friend
14/21
Sheesh, don't make me blush brother.
15/21
Hope you find them helpful
16/21
Much appreciated!
17/21
It's a pleasure to share them
18/21
Thanks for pointing it out! I've just rectified it.
19/21
Appreciate the love, Min!
20/21
You're speaking facts brother.
@OpenAI keeps delivering for the people and we love to see it.
21/21
Please do hermano
To post tweets in this format, more info here: https://www.thecoli.com/threads/tips-and-tricks-for-posting-the-coli-megathread.984734/post-52211196
1/11
How to get more work done in 3 hours with Chat GPT than 98% of people do in 3 weeks:
2/11
1. Learning from your own mistakes
Prompt:
“I made a mistake. I was practicing [insert skill here], and it failed when I [insert failure here and narrate what happened]. Could you please explain what happened, as well as how I can avoid making the same mistake in the future?"
3/11
2. Memorizing new Information
Prompt:
“In an overview style, but then in-depth, what are the most important figures, facts, things to know, I have to know about [Insert Topic here]. After that, help me create a memorization technique that will help me to remember"
4/11
3. Train ChatGPT to learn your Style
Prompt:
“Analyze the text below. Take note of the writing style, my voice, the way I type, and my tonality. Create a prompt to write a new paragraph in the same style, voice, and tone.
[Insert your text here]”
5/11
4. Quick Decision Making
Prompt:
“I am trying to make a decision. I am a [Insert context about yourself]. I am trying to [Insert decision here]. Give me a list of pros and cons that will help me decide why I should or shouldn’t make this decision.“
6/11
5. Building Your Own Skillset for Versatility
Prompt:
“I am currently working as a [Insert profession here]. What complementary skills should I learn, or certifications should I acquire, to upskill, to make myself more versatile, competent, & useful in [Insert Industry here]“
7/11
6. Quizzing with ChatGPT
Prompt:
“I want you to quiz me. Create a set of 10 multiple-choice questions, each one of them with four options, but with only one correct answer. The topic of the quiz is [Insert Topic Which you want to learn]"
8/11
Want to learn more about AI? Join my free Newsletter and get daily and weekly content in your inbox.
9/11
That's a wrap!
If you enjoyed this thread:
1. Follow me
@Whizz_ai for more content.
2. RT the tweet below to share this thread with your audience.
10/11
I am very glad to know
11/11
I am very glad you found it helpful brother
To post tweets in this format, more info here: https://www.thecoli.com/threads/tips-and-tricks-for-posting-the-coli-megathread.984734/post-52211196
How to get more work done in 3 hours with Chat GPT than 98% of people do in 3 weeks:
2/11
1. Learning from your own mistakes
Prompt:
“I made a mistake. I was practicing [insert skill here], and it failed when I [insert failure here and narrate what happened]. Could you please explain what happened, as well as how I can avoid making the same mistake in the future?"
3/11
2. Memorizing new Information
Prompt:
“In an overview style, but then in-depth, what are the most important figures, facts, things to know, I have to know about [Insert Topic here]. After that, help me create a memorization technique that will help me to remember"
4/11
3. Train ChatGPT to learn your Style
Prompt:
“Analyze the text below. Take note of the writing style, my voice, the way I type, and my tonality. Create a prompt to write a new paragraph in the same style, voice, and tone.
[Insert your text here]”
5/11
4. Quick Decision Making
Prompt:
“I am trying to make a decision. I am a [Insert context about yourself]. I am trying to [Insert decision here]. Give me a list of pros and cons that will help me decide why I should or shouldn’t make this decision.“
6/11
5. Building Your Own Skillset for Versatility
Prompt:
“I am currently working as a [Insert profession here]. What complementary skills should I learn, or certifications should I acquire, to upskill, to make myself more versatile, competent, & useful in [Insert Industry here]“
7/11
6. Quizzing with ChatGPT
Prompt:
“I want you to quiz me. Create a set of 10 multiple-choice questions, each one of them with four options, but with only one correct answer. The topic of the quiz is [Insert Topic Which you want to learn]"
8/11
Want to learn more about AI? Join my free Newsletter and get daily and weekly content in your inbox.
9/11
That's a wrap!
If you enjoyed this thread:
1. Follow me
@Whizz_ai for more content.
2. RT the tweet below to share this thread with your audience.
10/11
I am very glad to know
11/11
I am very glad you found it helpful brother
To post tweets in this format, more info here: https://www.thecoli.com/threads/tips-and-tricks-for-posting-the-coli-megathread.984734/post-52211196