Crazy how fast your channel is growing!! I've watched wayyyy too many "AI/GPT" vids in the past week and you're BY FAR the best in terms of pacing, info, critical thinking, aesthetic and value. Thanks for the info and inspiration!
Ok, so I don't know if it was a coincidence or not, but I just asked chat to rewrite one of my paragraphs on a post I'm working on, but to write it with a temperature of 1. Wow-it totally gave me what I wanted without any additional tweaking! Thanks for the video!
Good video. I've been using ChatGPT extensively for Japanese to English and English to Japanese communication and translation. I've figured out the prompting on my own. It's been an invaluable tool except for the time it tried to kill me by suggesting I add bleach to my vinegar and salt solution I was using to patina metal parts.
I would just like to add that, although Open AI Playground is free, it is only free for 3 months. After that, you have to pay. So that is why I use ChatGPT instead now.
Thank you…Thank you…Thank you… I’m sure others will agree.. Been trawling for weeks (hours)… And truthfully.. BEST introduction to GPT so far… Phew what a relief..
As an AI language model, I don't have a physical temperature setting that can be adjusted. My purpose is to provide information and answer questions based on the text data I have been trained on. If you have any questions or need assistance, please feel free to ask, and I'll do my best to provide accurate and helpful information.
As a Python developer, creator of EmpathyBot ai, and a person generally interested in ai, I really love your videos and learn much from you! A caring request: Please get more vitamin D and/or Sunlight if possible in your region.
I have a curious question that is even more fundaMENTAL than your astutely asking "how can we know about ChatGPT being wrong". How can we know when Ai is being programmed to provide misleading information? Or when is an AI Criminati Master engineering It to cheat, lie and obfuscate? Etc….
I have watched several videos on ChatGPT, and I must say that this one stands out as the best. However, having a background in software engineering, I have observed that most of the content generated by ChatGPT tends to have errors. It is important to note that ChatGPT cannot replace human employees, and businesses that rely on it to develop back-end code may face challenges.
I was actually surprised at the sort of mistakes it was making when I asked it to solve physics problems. Mistakes I absolutely would not expect an AI to make. Like I asked it to solve a relativity problem and it gave me completely wrong answers. Looking through the steps it took, I saw that it used 15/16 as the length contraction factor for an object moving at 1/4 the speed of light. The real answer is sqrt(15/16). It dropped the square root. THAT is a mistake I'd expect a human to make, not an AI. And it's not like it didn't know there was supposed to be a square root there, the problem involved an object moving at 1/4 the speed of light, an object moving at 1/2 the speed of light and an object moving at 3/4 of the speed of light, and it had the square root for the other 2 objects, it just left it off for one of the 3. And it made other mistakes for the other two objects. Its calculations were just a complete mess.
The answer saying invest but not mentioning work wasn’t wrong, it was just assuming the user was smart enough to know they had to work. The “better” response was just a dumbed down version.
I have been wanting to jump in figure out this whole AI thing but it is completely overwhelming. I haven't done anything at all with it yet, but today I discovered you. I am going to stick with you. Everything that you say makes sense and is understandable. You also have an amazingly calming effect when you talk. I intend to watch every single video that you have posted. Thank you so much.
MAny can talk. Few can educate. You have skill and artistry in this area, and if I might add, in decorating your room/studio. Classsy. You don't see that much any more. Nice pacing. Keep em' coming.
Crazy how fast your channel is growing!! I've watched wayyyy too many "AI/GPT" vids in the past week and you're BY FAR the best in terms of pacing, info, critical thinking, aesthetic and value. Thanks for the info and inspiration!
Ok, so I don't know if it was a coincidence or not, but I just asked chat to rewrite one of my paragraphs on a post I'm working on, but to write it with a temperature of 1. Wow-it totally gave me what I wanted without any additional tweaking! Thanks for the video!
She makes wrong interpretation at 7.42, question was how to get rich and not how to earn income.
wow!
Good video. I've been using ChatGPT extensively for Japanese to English and English to Japanese communication and translation. I've figured out the prompting on my own. It's been an invaluable tool except for the time it tried to kill me by suggesting I add bleach to my vinegar and salt solution I was using to patina metal parts.
I had no idea about Playground ‘temperature’, so great to know and great tips!
I would just like to add that, although Open AI Playground is free, it is only free for 3 months. After that, you have to pay. So that is why I use ChatGPT instead now.
Thank you…Thank you…Thank you…
I’m sure others will agree..
Been trawling for weeks (hours)… And truthfully..
BEST introduction to GPT so far… Phew what a relief..
She said Dices. Can we trust her opinion on language.. wake up sheeple
from gpt 4
As an AI language model, I don't have a physical temperature setting that can be adjusted. My purpose is to provide information and answer questions based on the text data I have been trained on. If you have any questions or need assistance, please feel free to ask, and I'll do my best to provide accurate and helpful information.
lol
As a Python developer, creator of EmpathyBot ai, and a person generally interested in ai, I really love your videos and learn much from you! A caring request: Please get more vitamin D and/or Sunlight if possible in your region.
It has been made politically biased e.g. Ukraine war – Russia should not defend itself from NATO or EU expansionism,
Totally awesome video! Real inspiration for me in my work, too.
also ive heard of open ai revoking people's api keys for using them in ways they dont like
I have a curious question that is even more fundaMENTAL than your astutely asking "how can we know about ChatGPT being wrong".
How can we know when Ai is being programmed to provide misleading information?
Or when is an AI Criminati Master engineering It to cheat, lie and obfuscate? Etc….
you lost me at alligator cat
It will start to seem "normal" unless huge new developments in AI come along in a few months or years.
Ask chatgpt for its references and it gets awkward
I have watched several videos on ChatGPT, and I must say that this one stands out as the best. However, having a background in software engineering, I have observed that most of the content generated by ChatGPT tends to have errors. It is important to note that ChatGPT cannot replace human employees, and businesses that rely on it to develop back-end code may face challenges.
Hello , I tried to sign up for ChatGPT web site , but I got a message " Sign up is unavailable ,please try again later " Any idea why please ?🤔🤔🤔
Awesome! TY
EXELLENT video! 👍🏾👍🏾
Interesting
You can't control chatgpt, use it as how like and enjoy his answers. That is. Chatgpt say this.
I was actually surprised at the sort of mistakes it was making when I asked it to solve physics problems. Mistakes I absolutely would not expect an AI to make. Like I asked it to solve a relativity problem and it gave me completely wrong answers. Looking through the steps it took, I saw that it used 15/16 as the length contraction factor for an object moving at 1/4 the speed of light. The real answer is sqrt(15/16). It dropped the square root. THAT is a mistake I'd expect a human to make, not an AI. And it's not like it didn't know there was supposed to be a square root there, the problem involved an object moving at 1/4 the speed of light, an object moving at 1/2 the speed of light and an object moving at 3/4 of the speed of light, and it had the square root for the other 2 objects, it just left it off for one of the 3. And it made other mistakes for the other two objects. Its calculations were just a complete mess.
You are so smart and very attractive. You gave a very enjoyable class, thank you so much!
Looks like Google is not the same for everybody. I for instance don't get "Use over time for: …." ;-(
If you use ChatGPT (the plus version), can you control temperature? Or only inside the playground?
this video was awesome
8:00 I just asked chatgpt and its running with 0.7. Or is this information new? I mean the video is two month old now.
The answer saying invest but not mentioning work wasn’t wrong, it was just assuming the user was smart enough to know they had to work. The “better” response was just a dumbed down version.
i get plus and im sad i cant use playground
I have been wanting to jump in figure out this whole AI thing but it is completely overwhelming. I haven't done anything at all with it yet, but today I discovered you. I am going to stick with you. Everything that you say makes sense and is understandable. You also have an amazingly calming effect when you talk. I intend to watch every single video that you have posted. Thank you so much.
fascinating, intelligent, articulate and well presented, thank you.
Great presentation of chatGPT keep up the good work 🙂
Really great information. Smart girl, doesn't hurt that she's cute too. No disrespect intended.
👍🤩
Its not great AI is a leftist leaning dangerous app.
Thanks!
MAny can talk. Few can educate. You have skill and artistry in this area, and if I might add, in decorating your room/studio. Classsy. You don't see that much any more. Nice pacing. Keep em' coming.
Comments are closed.