Glossary of AI concepts, explained simply
3 concepts
AdvancedAn Anthropic interpretability technique that automatically translates a large language model's internal activation vectors into human-readable text, enabling pre-deployment alignment audits and detection of hidden biases or deceptive intent.
AdvancedA technique that injects synthetic vectors into a model's internal layers at inference time to directly shift its decision-making, enabling precision debiasing and behavioural control — but also capable of bypassing safety training without any jailbreak prompt.
AdvancedA class of techniques that directly read, steer, or couple the internal numerical representations of AI models rather than operating through text, enabling real-time alignment audits, bias detection, and token-free inter-model communication.
I can help you apply this concept to your business.
We use cookies to improve your experience. You can choose which types of cookies to allow.