Undang-undang Penskalaan untuk Kuiz Rangkaian Neural

Jawapan dan penjelasan muncul semasa anda mengambil kuiz.

1. What pattern did Kaplan et al. report for language-model loss as model size, data, or compute increased?
   • Approximate power-law relationships over the tested ranges
   • A sudden guaranteed drop to zero
   • A fixed increase in loss for every larger model
   • Random oscillation with no measurable trend
2. Tiga kuantiti yang manakah berkaitan dengan undang-undang skala asal dengan kerugian?
   • Saiz kelompok, keciciran dan zaman
   • Parameter model, saiz set data dan pengiraan
   • Kadar pembelajaran, pereputan berat, dan kedalaman
   • Kiraan GPU, memori dan kependaman
3. How can a scaling-law fit help when planning a larger training run?
   • It guarantees the final benchmark score
   • It eliminates the need for validation
   • It can inform an extrapolation from smaller experiments, subject to the tested setup
   • It predicts results exactly for any dataset or architecture
4. Apakah yang diwakili oleh istilah malar (tidak boleh dikurangkan) dalam formula kerugian undang-undang skala?
   • Pepijat perisian
   • Kadar pembelajaran
   • Bilangan GPU
   • Entropi intrinsik data yang tidak boleh dikalahkan oleh model
5. Why can absolute loss reductions become smaller as compute keeps increasing under a power-law fit?
   • The reducible loss term shrinks while the loss approaches its floor
   • The data is removed after each compute increase
   • The model stops using parameters
   • Power laws require loss to increase over time
6. Aliran manakah yang mendorong bidang melebihi skala 'lebih besar lebih baik'?
   • GPU menjadi percuma
   • Kekurangan data berkualiti tinggi serta had tenaga dan kos
   • Teks berkualiti tinggi tanpa had muncul
   • Lengkung rugi bertukar kepada garis menegak lurus
7. What did the original Kaplan study say about architecture details such as width and depth within its tested range?
   • They always dominated model size, data, and compute
   • They made loss independent of data
   • They had relatively small effects compared with the studied scaling variables
   • They guaranteed the same performance for all architectures
8. What compute-allocation tendency did the original Kaplan paper report?
   • Every model should be trained to zero loss
   • Small models always use less compute at equal loss
   • Training data has no effect on compute allocation
   • Large models could be more sample-efficient and compute-efficient training could stop before convergence