← Back to context Comment by nonethewiser 8 hours ago I wonder what hooks they have in place to be able to configure safeguards at runtime. 2 comments nonethewiser Reply aleksiy123 7 hours ago Probably a mix of heuristics, keywords and simple ml model.Then maybe a second gate with a lightweight llm?Edit: actually Gcp, azure, and OpenAI all have paid apis that you can also use.But I don’t think they go into details about the exact implementation https://redteams.ai/topics/defense-mitigation/guardrails-arc... ryoshu 7 hours ago When we do these it's a fine-tuned classifier, generally a BERT class model. Works quite well when you sanitize input and output with low latency/cost.
aleksiy123 7 hours ago Probably a mix of heuristics, keywords and simple ml model.Then maybe a second gate with a lightweight llm?Edit: actually Gcp, azure, and OpenAI all have paid apis that you can also use.But I don’t think they go into details about the exact implementation https://redteams.ai/topics/defense-mitigation/guardrails-arc... ryoshu 7 hours ago When we do these it's a fine-tuned classifier, generally a BERT class model. Works quite well when you sanitize input and output with low latency/cost.
ryoshu 7 hours ago When we do these it's a fine-tuned classifier, generally a BERT class model. Works quite well when you sanitize input and output with low latency/cost.
Probably a mix of heuristics, keywords and simple ml model.
Then maybe a second gate with a lightweight llm?
Edit: actually Gcp, azure, and OpenAI all have paid apis that you can also use.
But I don’t think they go into details about the exact implementation https://redteams.ai/topics/defense-mitigation/guardrails-arc...
When we do these it's a fine-tuned classifier, generally a BERT class model. Works quite well when you sanitize input and output with low latency/cost.