2024-12-14 13:12:33
Myself and many others have tried identical prompt but never had this answer. The image posted might be manipulated. [Link] NEEDS_MORE_RATINGS(3-0-10) Author
2024-12-14 13:36:48
NNN. Large language models are inherently non-deteriministic and the same question may give different results. Additionally, there are many different versions and models used for chatgpt that can give different results. NEEDS_MORE_RATINGS(10-0-1) Author
2024-12-15 08:20:28
This post is misleading. LLMs generate responses based on log probabilities, leading to some randomness and creative variety. Repeating the same question often yields different answers. For more stable outputs, use the API and set the temperature close to 0. [Link] NEEDS_MORE_RATINGS(0-0-1) Author