Kuiz berkaitan panduan · Sederhana Tahap

Kuiz Undang-undang Penskalaan Chinchilla

Uji penemuan belanjawan latihan Chinchilla, contoh 70B/1.4T, perkadaran penskalaan, pengiraan anggaran dan perbezaan daripada pengoptimuman penggunaan.

Laluan panduan yang berkaitanUndang-undang Penskalaan Chinchilla
soalan 1 daripada 8

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

Pratonton setiap soalan8 soalan

Jawapan dan penjelasan muncul semasa anda mengambil kuiz.

  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. Model manakah yang Chinchilla mengatasi prestasi walaupun jauh lebih kecil?
    • GPT-2
    • Gopher
    • BERT
    • Mamba
  4. Apakah yang dimaksudkan oleh Chinchilla tentang model seperti GPT-3 pada masa itu?
    • Mereka terlalu terlatih
    • Mereka mempunyai terlalu sedikit pengiraan yang tersedia
    • Mereka menggunakan tokenizer yang salah
    • Mereka kurang terlatih berbanding saiz mereka
  5. Secara kasar bagaimana pengiraan dianggarkan dalam analisis Chinchilla?
    • Berkadar dengan bilangan GPU sahaja
    • Bebas daripada saiz data
    • Berkadar dengan parameter kali token latihan
    • Berkadar dengan saiz perbendaharaan kata
  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