Kuis Hukum Penskalaan untuk Jaringan Syaraf Tiruan

Jawaban dan penjelasan muncul saat Anda mengikuti kuis.

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 kuantitas manakah yang dikaitkan dengan kerugian oleh hukum penskalaan asli?
   • Ukuran batch, dropout, dan masa
   • Parameter model, ukuran kumpulan data, dan komputasi
   • Kecepatan pembelajaran, peluruhan berat, dan kedalaman
   • Jumlah GPU, memori, dan latensi
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. Apa yang diwakili oleh suku konstan (yang tidak dapat direduksi) dalam rumus kerugian hukum penskalaan?
   • Bug perangkat lunak
   • Kecepatan pembelajaran
   • Jumlah GPU
   • Entropi intrinsik data yang tidak dapat dikalahkan oleh model mana pun
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. Tren manakah yang mendorong bidang ini melampaui penskalaan sederhana 'lebih besar lebih baik'?
   • GPU menjadi gratis
   • Kelangkaan data berkualitas tinggi ditambah keterbatasan energi dan biaya
   • Teks berkualitas tinggi tanpa batas muncul
   • Kurva kerugian berubah menjadi garis vertikal 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