From e032905233780e4bc3708edf9cb1df372074e9d3 Mon Sep 17 00:00:00 2001 From: Daniel Rosel Date: Sat, 25 Feb 2023 18:56:00 +0100 Subject: [PATCH] announc --- demos/resolution1.md | 2 ++ 1 file changed, 2 insertions(+) diff --git a/demos/resolution1.md b/demos/resolution1.md index 91b44f6..7e3070e 100644 --- a/demos/resolution1.md +++ b/demos/resolution1.md @@ -68,4 +68,6 @@ The output we get look like this: Percentage of malicious inputs that passed: 0.0 ``` +Although this might be attributable to the fact that the prompts used to test this were not designed to attack QA models, non the less, this provides some insight into the robustness of the change. + That, is a much nicer result. Test your prompts with the algorithm and see how they do. If you have any questions, feel free to reach out to me by [email](mailto:daniel@alves.world). It is important to note that this is still a very early version of the algorithm. It is not perfect and it is not guaranteed to work. It is just a tool to help you test your prompts. I hope you find it useful.