Mastery card
I can explain Attention
How a model decides which words matter most for the next word.
Attention is a mechanism that lets a model assign importance weights across tokens in a sequence when computing representations.
Sticky trick
Attention = spotlight. The model shines light on relevant bits.
Take into the room
In plain English, what is Attention?
Or open the share card 🔦
Or start your own 5-word trail from home.