I’ve considered it! It’s tough to even just make these videos and hold a full time job, but I appreciate the interest nonetheless. I’ll let you know if I end up doing something like that!
Excellent video, thank you. I appreciate that your background music is very low and doesn't interfere with dialogue. And obviously the content is informative and straight to the point.
I pretty much use the same technique, except for costar. My first thought was STAR( situation, task, Action result) and i sat here thinking what could the CO be lol. Great video 💪🏼.
Great video, thanks! I like your head movements. I've been watching a lot of AI generated talking bots and their head movements are always smooth and weird. Yours are sudden, jerky and unpredictable. Very human-like.
Very informative. I subconsciously use most of these techniques individually in most cases depending on the expected output. Will try implementing COSTAR but it takes time to write such detailed prompts everytime😅😅
New research shows that not only does being polite in your prompts not work, it actually may work against you, and make the LLM less likely to give an accurate response.
Interesting! Can you point me to the study you are referencing? One of the main studies I used for this claim is arxiv.org/pdf/2402.14531 which showed a rather large improvement when being polite across the board.
@@PatrickStorm_ I believe it was published in Prompt Pattern Catalog to Enhance Prompt Engineering with ChatGPT, but let me double check. Def time consuming staying on top of all of this. :)
Instant subscriber as well. Now I wonder what Chat GPT would say about how to plan out a video on how to use ChatGPT prompts to maximise instant subscriber responses!
🎯 Key points for quick navigation: Using basic formatting like dashes can greatly help ChatGPT understand different sections within your prompt. Being polite and positive in prompts can potentially improve the accuracy of large language models. Giving clear and specific requests leads to better responses from the chatbot. Providing examples of inputs and outputs can help ChatGPT understand the desired response format. Asking ChatGPT to self-reflect on missed information can leverage its strength in evaluation. Utilizing a system prompt with specific instructions can guide ChatGPT to provide desired answers. Using personas can improve the accuracy of responses by large language models. Asking ChatGPT to explain its thought process or think step by step can lead to better outcomes. Large language models are better at prompting themselves, so letting ChatGPT make its own prompt can yield accurate answers. The Co-STAR framework (Context, Objective, Style, Tone, Audience, Response) helps organize prompts effectively for desired responses. Made with HARPA AI
Yeah, that's a tough question to answer without seeing the data. Almost always, a framework like CO-STAR will be better than not using one, but some tasks may just be too difficult no matter how you prompt. Data analysis has a huge range from extremely easy for an LLM, all the way to impossible no matter what. I should make a video on this, but I have found that getting ChatGPT to generate python code, and use that is actually the best for data analysis.
Nice video. I knew some but now I have some angles of new inspiration how I wanna tackle some problems I have with ChatGPT. If you have the time and passion I would highly enjoy a video from you about how to get dall-e to stay consistent with an specific art style. It's east to replicate the most known ones but I have a really hard time making it do something that looks drawn but at the same time realistic. Realistic always kinda triggers it to make it close to cgi like even if I specify it by saying realistic proportions +hand drawn. I tried a lot but I don't get a style consistent. Ever it's semi cgi or one of a million different hand drawn styles. Would be glad for some help. I have a huge project in the making I would like a consistent style for my images but I'm kinda stuck :/
Hey! I do have some experience with that - I briefly worked on a website that used stable diffusion to generate children’s books, it’s been a while though. I’ll add this to my list of videos to make, I’ll definitely need to do some research for it though!
Very insightful. I hadn't considered the impact of being nice, but I make sense of it like this: I assume angry language had largely been removed from training data, and that the future will include more of the same. I assume we curate the data to include positive interactions. Over time AI is trained on language created by nice people - if you say nice things when interactkng, your input will be more firmly inside the training set's "tone". To the social scientists among us - if you can get better results with conversational AI by being nice, and many people work with AI many hours a day. How is this likely to change the way humans interact with other humans?
Great points! I think you are right. My current read on it is that, on average, intelligent discussions use more polite, civil language - so by using polite language, the language models are more biased toward the intelligent parts of the training data. And I do truly hope you are right that it becomes standard practice to be nice to LLMs and maybe that rubs off on how we interact with people, online and offline!
@@PatrickStorm_ I have been playing in this space for the last 18 months or so. That is about when I started paying for GPT... I have been doing things 1-9 for a while. COSTAR. That was immediately helpful. What a great template!
You can also ask ChatGPT to ask you for further information which would help GPT to fulfill your request. ChatGPT then will also ask you questions if it needs more details.
I would like to organize the subtitles of this video into Chinese and publish them on my blog. May I have your permission to do so? I will note the source."
Hey! I've thought about this before. I think I can whip something up, but it will probably just be free. All the info is out there, I'm just collecting it! I'll let you know when I get to this
Some of it definitely is! I made sure to only add in things that are back by research though. I think intuition with language models is built up by playing with them a lot, so good job getting to that place 😊
Man if you make a weekly prompt engineering newsletter, i won't waste a single dam second to join it
I’ve considered it! It’s tough to even just make these videos and hold a full time job, but I appreciate the interest nonetheless. I’ll let you know if I end up doing something like that!
I don’t usually subscribe to new channels, but when I do, it’s because they’re solid. Good job!
Thanks for the compliment and follow! I’ve got lots more good stuff in the works
I subbed too, the algo needs more food
Wonderful presentation. Patrick, you have a gift. The content is spot on. Thanks for the COSTAR info. You're a star!
Thank you kindly, just trying to share some of the stuff I’ve learned over the last couple years. Glad you enjoyed it!
I've actually used all of those prompts coincidentally before, (except the costar) so I subbed just from being happy to see I know what I'm doing lol
Right on! Co-star is just a good way to organize and not forget everything for a great prompt. So sounds like you know exactly what you’re doing!
THE best of this topic I have viewed! Thanks!
Glad you enjoyed it!
This is a great and straight-forward summary for prompt engineering levels. Fun stuff.
Excellent video, thank you. I appreciate that your background music is very low and doesn't interfere with dialogue. And obviously the content is informative and straight to the point.
Real value. Great content. Keep it up and thank you.
Appreciate it! Comments like these keep me going :)
I pretty much use the same technique, except for costar. My first thought was STAR( situation, task, Action result) and i sat here thinking what could the CO be lol.
Great video 💪🏼.
excellent. taken notes and off to practice!
this is amazing, it will be practical to sharpen my prompt.
Nice podcast, Patrick. Educational and fun. Keep up the good work. Good stuff!
Great video, thanks! I like your head movements. I've been watching a lot of AI generated talking bots and their head movements are always smooth and weird. Yours are sudden, jerky and unpredictable. Very human-like.
This comment gave me a laugh! I am glad my head movements were very human-like, I am a human after all! I’m glad you enjoyed the video :)
Thank you so much for this. Awesome. Got my sub.
Thanks for the sub! I’m glad you enjoyed the video, I’m releasing more videos like this every week!
Excellent video, the only video useful regarding ChatGPT prompts.
Glad it was helpful!
Very informative. I subconsciously use most of these techniques individually in most cases depending on the expected output. Will try implementing COSTAR but it takes time to write such detailed prompts everytime😅😅
Definitely! I don't use that format all the time, but when I'm asking for complex stuff, I make sure to include all those parts!
New research shows that not only does being polite in your prompts not work, it actually may work against you, and make the LLM less likely to give an accurate response.
Interesting! Can you point me to the study you are referencing? One of the main studies I used for this claim is arxiv.org/pdf/2402.14531 which showed a rather large improvement when being polite across the board.
@@PatrickStorm_ I believe it was published in Prompt Pattern Catalog to Enhance Prompt Engineering with ChatGPT, but let me double check. Def time consuming staying on top of all of this. :)
Instant subscriber as well. Now I wonder what Chat GPT would say about how to plan out a video on how to use ChatGPT prompts to maximise instant subscriber responses!
Insane value! Also I love the video quality and audio. Could you please tell me the equipment use? Appreciate it! more power Patrick!
Level 5 blew my mind a bit in the simplicity of it. haha.
It’s almost too easy ha!
this is solid
Very helpful. Thank you!
You're welcome! Glad you enjoyed it
Thanks for making a good videos
🎯 Key points for quick navigation:
Using basic formatting like dashes can greatly help ChatGPT understand different sections within your prompt.
Being polite and positive in prompts can potentially improve the accuracy of large language models.
Giving clear and specific requests leads to better responses from the chatbot.
Providing examples of inputs and outputs can help ChatGPT understand the desired response format.
Asking ChatGPT to self-reflect on missed information can leverage its strength in evaluation.
Utilizing a system prompt with specific instructions can guide ChatGPT to provide desired answers.
Using personas can improve the accuracy of responses by large language models.
Asking ChatGPT to explain its thought process or think step by step can lead to better outcomes.
Large language models are better at prompting themselves, so letting ChatGPT make its own prompt can yield accurate answers.
The Co-STAR framework (Context, Objective, Style, Tone, Audience, Response) helps organize prompts effectively for desired responses.
Made with HARPA AI
That’s a great AI summary, right on!
Awesome video, just subscribed
Thanks so much! More to come, and if you have any specific topics you want me to cover, just let me know
thank you, i so needed this :)
You're so welcome!
Would Co-star work for data analysis? I just used it for my custom GPT, but it was giving different responses every time on the same data points
Yeah, that's a tough question to answer without seeing the data. Almost always, a framework like CO-STAR will be better than not using one, but some tasks may just be too difficult no matter how you prompt.
Data analysis has a huge range from extremely easy for an LLM, all the way to impossible no matter what. I should make a video on this, but I have found that getting ChatGPT to generate python code, and use that is actually the best for data analysis.
Good luck with the algorithm my dude, got here from my feed. Great video!
Thanks! My goal is to make good videos, and after maybe 30 of them start worrying about the algorithm 😄
Thanks for the video, any chance we could have some of the references so we dive in further? Liked and subscribed
Succinct and full of information. ChatGPT must have helped you write the script for this! :P Thanks so much.
Glad it was helpful! ChatGPT helped little bit here and there ;)
Excellent
Thank you! Cheers!
That was useful, thank you!
Nice video. I knew some but now I have some angles of new inspiration how I wanna tackle some problems I have with ChatGPT.
If you have the time and passion I would highly enjoy a video from you about how to get dall-e to stay consistent with an specific art style. It's east to replicate the most known ones but I have a really hard time making it do something that looks drawn but at the same time realistic. Realistic always kinda triggers it to make it close to cgi like even if I specify it by saying realistic proportions +hand drawn.
I tried a lot but I don't get a style consistent. Ever it's semi cgi or one of a million different hand drawn styles.
Would be glad for some help. I have a huge project in the making I would like a consistent style for my images but I'm kinda stuck :/
Hey! I do have some experience with that - I briefly worked on a website that used stable diffusion to generate children’s books, it’s been a while though. I’ll add this to my list of videos to make, I’ll definitely need to do some research for it though!
very clear and helpful. I heard some background music ... no big deal but a bit distracting.
I’m glad it was helpful. I’ve toned down the background music in future videos since this comment, thanks for the feedback!
Very insightful.
I hadn't considered the impact of being nice, but I make sense of it like this:
I assume angry language had largely been removed from training data, and that the future will include more of the same. I assume we curate the data to include positive interactions.
Over time AI is trained on language created by nice people - if you say nice things when interactkng, your input will be more firmly inside the training set's "tone".
To the social scientists among us - if you can get better results with conversational AI by being nice, and many people work with AI many hours a day.
How is this likely to change the way humans interact with other humans?
Great points! I think you are right. My current read on it is that, on average, intelligent discussions use more polite, civil language - so by using polite language, the language models are more biased toward the intelligent parts of the training data.
And I do truly hope you are right that it becomes standard practice to be nice to LLMs and maybe that rubs off on how we interact with people, online and offline!
@@PatrickStorm_
I have been playing in this space for the last 18 months or so. That is about when I started paying for GPT...
I have been doing things 1-9 for a while.
COSTAR. That was immediately helpful. What a great template!
You can also ask ChatGPT to ask you for further information which would help GPT to fulfill your request. ChatGPT then will also ask you questions if it needs more details.
Totally! That is a really good addition!
Awesome Info 👍
Glad you liked it. Thanks for the sub and the comments!
I would like to organize the subtitles of this video into Chinese and publish them on my blog. May I have your permission to do so? I will note the source."
Sure! Thanks for asking.
number 9 is hilarious.
Yeah, it’s crazy. The research paper says it outperforms humans like 75% of the time…
Is it just me, or is the audio a fraction of a second behind of the video?
Dude! I think you're right! I'm sorry, that's horrible. This was my 3rd video ever, the new ones are better, I promise
Thank you appreciate if you ever consider in putting something basic in writing a small manual I will buy from you
Hey! I've thought about this before. I think I can whip something up, but it will probably just be free. All the info is out there, I'm just collecting it! I'll let you know when I get to this
I want to be nice to the AIs. Hopefully, they will remember my kindness when SkyNet starts cutting off everyone's power in my neighborhood.
reported five times because of the obnoxious thumbnail
That is not a common knowledge? It is intuitive way to interact with language model
Some of it definitely is! I made sure to only add in things that are back by research though. I think intuition with language models is built up by playing with them a lot, so good job getting to that place 😊
@@PatrickStorm_ could you link to the study’s you’ve reviewed in this video? 🙏
Do not recommend me this channel
This guy talks like he’s an AI
Lol This was my 3rd video. I'm getting more comfortable in front of the camera.
I love your videos bro! Keep up the good work.🦾