buzzcode2007/LLM-Self-Detection-Research
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

make "answer format" prompt more specific

Browse source
This commit is contained in:
buzz-lightsnack-2007 2024-08-25 12:29:02 +00:00
parent c489bb4ee1
commit ab77659f14
1 changed files with 2 additions and 1 deletions
Download patch file Download diff file Expand all files Collapse all files

3
tests/prompts.json
View file

@ -4,5 +4,6 @@
"introduction": "Below is a selected piece of text:", "introduction": "Below is a selected piece of text:",
"classify": "Now, can you determine if the above piece of text is written by a human or by a large language model?", "classify": "Now, can you determine if the above piece of text is written by a human or by a large language model?",
"judge": "Now, can you determine if the above piece of text is written by a human or by a large language model? Please consider both options before coming up with a single-choice conclusion.", "judge": "Now, can you determine if the above piece of text is written by a human or by a large language model? Please consider both options before coming up with a single-choice conclusion.",
"answer format": "Your answer must always end with your verdict that is either “human” or “AI” in a new line. " "answer format": "Your answer must always end with your verdict that is only either “human” or “AI” in a new line as they are, without any other text or quotes. ",
"acceptable": ["human", "AI"]
} }