Kediska Feejignaanta Su'aalo-badan

Jawaabaha iyo sharraxaadaha waxay soo baxayaan markaad qaadanayso kediska.

1. Waa maxay dareenka su'aalo badan oo la wadaaga dhammaan madaxda dareenka?
   • Hal furayaal iyo qiimayaal
   • Odoroska weydiinta
   • Heerkulka softmax
   • Miisaanka lakabka wax soo saarka
2. Waa maxay cilada aasaasiga ah ee MQA ay wax ka qabato inta lagu jiro jiilka autoregressive?
   • Wax-soo-saarka xisaabinta-dhibcaha sabbaynaya
   • Xaddiga xusuusta ee soo dejinta KV cache
   • Tirada waayihii tababarka
   • cabbirka erayada Tokenizer
3. Qiyaastii qodobkee ayaa MQA u dhimaya kaydka KV?
   • Marka la eego dhererka isku xigxiga
   • Xajmiga Dufcaddii
   • By tirada madax dareenka
   • Ma beddelayso cabbirka kaydka KV
4. Waa maxay ganacsiga ugu weyn ee isticmaalka MQA?
   • Tilmaan aad u gaabis ah
   • Hadda ka dib ma isticmaali karto madaxyo badan
   • Waxay u baahan tahay xog tababar oo aad u badan
   • Hoos u dhac yar oo tayada model ah
5. Kala duwanaanshiyaha u dhexeeya heerka feejignaanta madax-badan iyo MQA?
   • Feejignaan-kooxeed
   • Silbadh-Daaqadda Feejignaanta
   • Fiiro gaar ah
   • Fiiro gaar ah
6. Yaa soo jeediyay feejignaan su'aalo badan?
   • Geoffrey Hinton
   • Noam Shazeer
   • Yann LeCun
   • Andrej Karpathy
7. In Multi-Query Attention, what remains separate for each attention head?
   • The shared key projection
   • The shared value projection
   • The query projection
   • The token sequence itself
8. How does Grouped-Query Attention differ from standard MQA?
   • It shares query projections but gives every head separate keys and values
   • It removes attention and replaces it with a recurrent layer
   • It groups output vocabulary tokens instead of attention heads
   • It shares keys and values within several groups of query heads rather than across all heads at once