quiz bu lëkkale ak tegtal · Dëgër Niveau

Quiz buy yokk cache (CAG)

Saytu sa xam-xam ci defar cache-augmented: naka lañuy njëkka sarse ak jëfandikoowaat cache KV yi, reset cache yi ci diggante laaj yi, ak yamaleg dayo corpus.

Yooni njiit yu jëm ciCache buñ yokk
Laaj 1 de 8

Lan la CAG di jëfandikoo ci barabu seet bu laaj bu nekk?

Xoolaatal laaj bu nekk8 laaj

Sooy def quiz bi, tontu yi ak leeral yi dañuy feeñ.

  1. Lan la CAG di jëfandikoo ci barabu seet bu laaj bu nekk?
    • A keyword search engine
    • Fine-tuning the model on the documents
    • A new vector index per question
    • Preloading the whole knowledge base into the KV cache once and reusing it
  2. Luy cache KV ci biir trafo?
    • A database of user accounts
    • The stored attention keys and values computed for tokens the model has already read
    • A list of document URLs
    • The model's trained weights
  3. Lan moo waral sistem CAG di dagg cache bi ginaaw tontu bu nekk?
    • To save the answer for training
    • To delete the documents permanently
    • To restore the documents-only state so the next query does not see the previous question and answer
    • To switch to a different model
  4. Lan mooy gàttal bu gëna mag ci CAG?
    • The entire corpus must fit in the model's context window
    • It requires a vector database
    • It cannot answer questions quickly
    • Dafay soppi diisaayu model bi
  5. Lan moo waral CAG di gaaw ci tontu moo gën ñu teg këyit yi ci bépp laaj?
    • It skips generating output tokens
    • The model only needs to compute the new question tokens on top of the cached prefix
    • Dafay jëfandikoo model bu gëna ndaw
    • It retrieves fewer documents
  6. Which knowledge base is the best fit for CAG?
    • A news archive updated every hour
    • Millions of customer records with per-user permissions
    • An entire web crawl
    • A small, stable product manual that rarely changes
  7. Which statement about CAG is a misconception?
    • CAG removes retrieval errors because all documents are present
    • Changing documents requires rebuilding the cache
    • CAG fine-tunes the model's weights on the documents
    • Hosted prompt caching is related to the idea
  8. Which techniques reduce the memory needed for a large KV cache?
    • Grouped-query attention and cache quantization
    • Yokk yeneen njiit yuy bàyyi xel
    • Using longer prompts
    • Raising the sampling temperature