Sheria za Kuongeza Maswali ya Mitandao ya Neural

Majibu na maelezo huonekana unapojibu maswali.

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. Je, ni viwango gani vitatu ambavyo sheria za awali za kuongeza viwango zinahusiana na hasara?
   • Ukubwa wa kundi, kuacha shule, na nyakati
   • Vigezo vya mfano, saizi ya seti ya data na hesabu
   • Kiwango cha kujifunza, kuoza kwa uzito, na kina
   • Hesabu ya GPU, kumbukumbu, na muda wa kusubiri
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. Neno la mara kwa mara (lisiloweza kupunguzwa) katika fomula ya upotezaji wa sheria ya kuongeza inawakilisha nini?
   • Hitilafu ya programu
   • Kiwango cha kujifunza
   • Idadi ya GPU
   • Utangulizi wa asili wa data ambayo hakuna modeli inayoweza kushinda
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. Je, ni mwelekeo gani unaosukuma uga zaidi ya upanuzi rahisi wa 'kubwa ni bora'?
   • GPU kuwa huru
   • Uhaba wa data ya ubora wa juu pamoja na vikomo vya nishati na gharama
   • Maandishi ya ubora wa juu yasiyo na kikomo yanaonekana
   • Mikondo ya upotevu inayogeuka kuwa mistari ya wima iliyonyooka
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