Back to questions
QuestionHardMachine LearningTechnical interview · Sber / GigaChat

KV cache during generation of LLM

What is a KV cache, how does it eliminate the recalculation of past tokens during autoregression generation and how much memory does it take?

Answer as you would in an interview

Answer aloud or write down a few key points. The text field is optional; the review opens after your attempt.

Loading
KV cache during generation of LLM — interview question — ML Mentor