Ikibazo gihuza ibibazo · Hagati Urwego

Ikibazo-cyo Kwiga

Ongera usubiremo imyiyerekano yihuse, uburemere bwikitegererezo, imipaka yubushakashatsi bwimbitse, urugero-gutondekanya ingaruka, hamwe ningengo yimiterere yinyigisho nkeya.

Inzira zijyanye no kuyoboraMu Kwiga
Ikibazo 1 Bya 8

Ni izihe mpinduka zifatika imbere yicyitegererezo mugihe cyo kwiga?

Reba buri kibazo8 ibibazo

Ibisubizo nibisobanuro bigaragara mugihe ufata ikibazo.

  1. Ni izihe mpinduka zifatika imbere yicyitegererezo mugihe cyo kwiga?
    • Ibipimo by'icyitegererezo bivugururwa n'intambwe nkeya
    • Agace gato ka adapter gatozwa ingero
    • Icyitegererezo gifata mu mutwe burundu umurimo mushya
    • Nta kintu na kimwe mu bipimo gihinduka; gusa ikibazo kiyobora ibisohoka
  2. Ikibazo gikubiyemo ingero nyinshi zakozwe mbere yikibazo nyacyo cyitwa:
    • Zero-kurasa
    • Kubaza bike
    • Kuringaniza neza
    • Kwiga gushimangira
  3. What does GPT-3's reported improvement in few-shot performance with scale establish?
    • A universal rule that all models below one hundred billion parameters cannot use examples
    • A requirement for internet access during every prompt
    • An observed trend for the evaluated models and tasks
    • A separate emergency training phase
  4. How should induction-head research be related to broad in-context learning?
    • As evidence about a studied mechanism, with limitations on broader conclusions
    • As proof that every task uses only one circuit
    • As a permanent database of every demonstration
    • As a replacement for all task evaluation
  5. How should an application compare repeated few-shot prompting with fine-tuning?
    • Assume fine-tuning is always the least expensive choice
    • Measure task quality and the relevant training, request, and maintenance costs
    • Assume adding unlimited prompt examples has no cost
    • Decide solely by which approach has the longer name
  6. A few-shot ticket prompt demonstrates only refund and billing labels. What do those examples provide for the next ticket?
    • An automatic update to the model's stored weights
    • A guarantee that every next ticket fits one of the labels
    • A permanent change to all future conversations
    • An in-prompt task and label pattern that can guide classification
  7. Why might the same few-shot classification examples produce different results after reordering?
    • The model can be sensitive to the order of examples in its context
    • Reordering automatically retrains every parameter
    • The label texts must disappear
    • The model's vocabulary changes with each permutation
  8. What resource tradeoff arises when more demonstrations are added to an in-context prompt?
    • They use no input tokens because they are examples
    • They make the context limit disappear
    • They consume more supported context and per-request processing
    • They permanently reduce the model's parameter count