Haha. Love it. Trying not to exaggerate the impact of simple techniques, but in this case, you're right. It can be mind-boggling how much a simple tweak can change the output from ChatGPT.
LLMS
-
8 ChatGPT mistakes to avoid: optimization tips
By
–
Summary 8 ChatGPT mistakes to avoid • Not being specific about your goals
• Not asking it to reduce its output
• Mixing topics in a single chat
• Asking only 1 thing at a time
• Prompting in the negative
• Not giving examples
• Asking it to do math
• Not iterating -
Managing ChatGPT’s Output: Tips for Getting Concise Responses
By
–
Not asking it to reduce its output ChatGPT might give you way too much information. Did you know you can ask it to help with that? Ask the assistant to reduce, remove, compile, or rewrite.
-
Providing Examples to ChatGPT for Better Output Quality
By
–
Not giving examples of what you want You can give ChatGPT full examples of the kind of output you're looking for. Word your prompt carefully to have it replicate the aspects of the example that you like.
-
Language Models Fake Math: Why They Sound Convincing But Fail
By
–
Asking a language model to do math The answer will sound convincing. Sometimes parts of it are correct. But, for the most part, it's faking it — just predicting the next word. The math will likely be wrong. (If you have math problems, check out WolframAlpha instead.)
-
ChatGPT Output Specificity: Removing Pretext and Context
By
–
🔸 Not being specific about your output goals
— Rob Lennon 🗯 | AI Whisperer (@thatroblennon) 5 janvier 2023
ChatGPT has been trained to prefer certain outputs.
You can request a change.
A quick hack:
Try asking it to "remove pre-text and context" and "only return" certain results: pic.twitter.com/qd2PIXEXqKNot being specific about your output goals ChatGPT has been trained to prefer certain outputs. You can request a change. A quick hack: Try asking it to "remove pre-text and context" and "only return" certain results:
-
8 Common ChatGPT Mistakes Limiting Your Results
By
–
Most new ChatGPT users are making simple mistakes. (And they don't realize results could be TWICE as good.) 8 problems with your AI prompts to stop right now:
-
ICML2023 LLM Detection Policy False Positives Analysis
By
–
Bringing it back to our community, I wonder how many false positives (and false negatives!) there will be for #ICML2023's ban on LLM-generated text
-
Fine-tuned Language Models for Text Rephrasing Projects
By
–
I'm sure we'll see lots more projects pop up that let you "rephrase" an input text in all kinds of ways using fine-tuned language models. Many of them will be open source.