Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Perspective Filter Toxicity #4594

Open
CGBLR opened this issue Apr 15, 2024 · 3 comments
Open

Perspective Filter Toxicity #4594

CGBLR opened this issue Apr 15, 2024 · 3 comments

Comments

@CGBLR
Copy link

CGBLR commented Apr 15, 2024

Can we set multiple filters from the Coral dashboard?
If yes, should they be comma and space separated like this -- TOXICITY, INSULT, PROFANITY, THREAT?
image

@losowsky
Copy link
Member

Interesting. Would you set a different threshold for each filter?

I will also note that the more models you send it to, the slower it would be to return a response before Coral can publish. We've already had occasional issues with some of their speed of response for a single model.

@CGBLR
Copy link
Author

CGBLR commented Apr 16, 2024

Ideally, yes for the threshold. But a single threshold for multiple attributes should also be fine.
I'm not sure if we could set multiple attributes from the coral dashboard. Could you please confirm?

@losowsky
Copy link
Member

losowsky commented Apr 16, 2024

You can't currently, as we're concerned about the speed of response for multiple attributes. Can you detail what you're trying to solve for? The general Toxicity model includes all of their other filters afaik.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

No branches or pull requests

2 participants