DETECTSecurity & safety
TypeSafe AI Jev Model Comparison
H3x@h3xdigest
This project compares the TypeSafe AI Jev model against Mistral Moderation 2 using a dataset of labeled hate-speech comments.
We got waitlist access to @typesafeai Jev. We wanted to see how it holds up against our current @MistralAI pipeline.
We compared TypeSafe's Jev model with Mistral Moderation 2 on 391 labeled hate-speech comments provided by @TobiasHuch
[1]