Goodfire says its activation monitors use signals from inside a model to detect undesired behaviors, catching more cases, running faster and costing less than text-based monitors. Baseten customers can use them to monitor for prompt injection, actions outside policy, sensitive data exposure and cyber misuse.
Part of a thread · earlier post
Our activation monitors detect undesired behaviors using signals from inside a model, so they catch more, run faster, and cost less than text-based monitors.
Baseten customers can monitor for:
- Prompt injection
- Actions outside policy
- Sensitive data exposure
- Cyber misuse
Source: Goodfire · x.comPublished · added here
