Modèlu attention bu néew quiz

Sooy def quiz bi, tontu yi ak leeral yi dañuy feeñ.

1. Lan moo waral bàyyi xel ci sa bopp lu seer lool ci toppalante yu gudd?
   • Njëg li dafay màgg ak kaare guddaayu toppalante bi
   • Mënul jëfandikoo ay GPU
   • Du dox ludul ci kàddu yu gàtt
   • Soxlawul benn memory
2. Luy palanteer buy gliise (local)?
   • Token bu nekk dafay jaar ci base de done
   • Token bu nekk dafay toppatoo dëkkandoo yi ko jege
   • Jetons duñu musa bàyyi xel ci dara
   • Token bu nekk dafay toppatoo token yépp ñaari yoon
3. Lan mooy njariñu 'token global' ci lu néew luñu koy bàyyi xel?
   • ngir yeexal model bi
   • Ngir denc nataal
   • nekk hub yuy toppatoo lépp te lépp di toppatoo
   • Ngir wecci vokabileer bi
4. Ban model moo firndeel ni boole bàyyi xel ci anam wu bari, palanteer, ak bàyyi xel ci àdduna bi dafay wéy di fësal xel mi yépp?
   • Resnet
   • Word2Vec
   • AlexNet
   • Picc mu Mag
5. Ci gàttal, naka lay demee ba limu lëkkaloo yi di yokku ci barab bu ñu defar bu baax?
   • Lu tollu ci ligneer ak guddaayu toppalante
   • Quadratiquement ak guddaayu toppalante
   • Du aju ci guddaay bi
   • Exponentiellement ak guddaayu toppalante
6. What do strided or dilated sparse-attention patterns allow?
   • xayma leen ba noppi def leen nul
   • Skip ahead to reach distant context with fewer attention links
   • Yonnee leen ci beneen serwër
   • Daf leen di denc ba fàww
7. How does Longformer combine local and global attention?
   • Every token attends to every other token
   • Only random edges are allowed
   • Most tokens use a sliding local window, while selected global tokens connect across the sequence
   • Each token attends only to the final token
8. What is the goal of learned, content-based sparse attention?
   • A full dense connection to every token
   • A rule for removing all future tokens
   • A learned vocabulary for each attention head
   • Avoid spending attention compute and memory on content unrelated to the current query