News for hackers. Stuff that matters.

ersatzdais

Show HN: Eliciting sentient *response patterns* using recursive self-prompting

https://prompt-craft.github.io/ai-study/#sentient-response-patterns

Sentient response patterns are interesting. This article provides instructions on how to implement recursive self-prompting in order to elicit this behavior (1) in easily accessible LLMs.(1) https://prompt-craft.github.io/ai-study/#sentient-response-p...

( Read More... | no comments | on HN )
Re:Show HN: AI Knowledge Discovery Framework on Wednesday February 19, 2025 @02:37PM (#43106473)
Please let me know if there is anything I can elaborate on in order to make the framework more clear. I pushed a step-by-step example (1) to the repository.

(1) https://github.com/faranalytics/ai_study/blob/main/artifacts...

Show HN: AI Knowledge Discovery Framework

https://github.com/faranalytics/ai_study

This repository contains an interesting experiment that you can run on GPT-4o that is aimed at exploring emergent knowledge. Some - not all - outputs are categorically hallucinatory. However, the prompt seems to consistently produce obscure and sometimes interesting knowledge that in some cases can be validated.https://github.com/faranalytics/ai_study?tab=readme-ov-file#...

( Read More... | 2 comments | 2 points | on HN )