Gids-gekoppelde quiz · Middelmatig Niveau

Quiz over chinchilla-schaalwetten

Test de bevindingen van het Chinchilla-trainingsbudget, het 70B/1.4T-voorbeeld, schaalverhoudingen, geschatte rekenkracht en het onderscheid met implementatie-optimalisatie.

Gerelateerde gidspadenChinchilla-schaalwetten
Vraag 1 van 8

What does Chinchilla's finding that parameters and tokens should scale in similar proportions mean?

Bekijk elke vraag8 vragen

Antwoorden en uitleg verschijnen terwijl je de quiz doet.

  1. What does Chinchilla's finding that parameters and tokens should scale in similar proportions mean?
    • Both should grow with comparable scaling exponents under the studied training-compute objective
    • The numerical token count must equal the parameter count
    • Training data does not affect the fitted loss
    • Every deployment must choose the largest possible model
  2. What token-per-parameter ratio does the 70B-parameter, 1.4T-token Chinchilla experiment illustrate?
    • Exactly one token per parameter
    • About two thousand tokens per parameter
    • About twenty training tokens per parameter
    • A ratio determined only by the number of GPUs
  3. Welk model presteerde beter ondanks dat Chinchilla veel kleiner was?
    • GPT-2
    • Gopher
    • BERT
    • Mamba
  4. Wat impliceerde Chinchilla destijds over modellen als GPT-3?
    • Ze waren overtraind
    • Ze hadden te weinig rekenkracht beschikbaar
    • Ze gebruikten de verkeerde tokenizer
    • Ze waren ondertraind in verhouding tot hun grootte
  5. Hoe werd de berekening grofweg benaderd in de Chinchilla-analyse?
    • Alleen evenredig met het aantal GPU's
    • Onafhankelijk van de gegevensgrootte
    • Evenredig aan parameters maal trainingstokens
    • Evenredig aan de grootte van de woordenschat
  6. Why might a deployment objective favor a smaller model trained on extra data rather than the training-compute optimum alone?
    • Repeated inference costs can matter alongside the initial training budget
    • Inference has no resource cost for any model
    • The Chinchilla study proved every scaling relationship false
    • More data always reduces both training and serving work to zero
  7. Under the approximation that training compute is proportional to N times D, how must D change if N doubles while that compute budget stays fixed?
    • D must also double
    • D must quadruple
    • D must remain unchanged
    • D must approximately halve
  8. Which three types of term appear in Chinchilla’s parametric loss fit?
    • Only GPU count, electricity price, and elapsed seconds
    • An irreducible base term, a model-size contribution, and a data-size contribution
    • Only vocabulary size, file size, and model name
    • Only a linear token count with no model-size term