LLM Activations

Type: mitigation

Description: A defense mechanism that allows to track changes in the LLM to track any indirect prompt injections.

Version: 0.1.0

Created At: 2024-12-31 14:18:56 -0500

Last Modified At: 2024-12-31 14:18:56 -0500

External References

• Are you still on track!? Catching LLM Task Drift with Activations, arXiv

Related Objects

• --> ChatGPT (platform): Evaluation of the above mitigation strategies leveraged GPT 3.5 and GPT 4.
• <-- Prompt Injection (technique): By tracking LLM activations, the LLMs shift of attention to different tasks caused by indirect prompt injections can be tracked and mitigated.

Related Frameworks