CHOOSESecurity & safety
Jev Token-Level Safety Monitor
idhant gulati@idhantgulati
1:15Jev functions as a token-level safety monitor for language models, providing near-instant, reliable, and cost-effective observability of model outputs by classifying each token or sequence as safe, refusal, sensitive, or unsafe.
so @typesafeai's Jev can be used as a token-level safety monitor for language models. it gives near-instant, reliable, and cheap monitoring / observability for model outputs (this costed <$0.01 !!)
quick demo of Jev monitoring deepseek v4.1 flash output tokens on the fly, classifying each token / seq one safe, refusal, sensitive, or unsafe