Alamun (tokens) sune sasannin da LJJan suna sarrafa rubutu a ciki (sassan kalmomi), kuma cikaken rukunin jiya (context window) shine mafi girman adadin rubutu (alamun) da LJJ zai iya tunani a jiya gida. Fahimtar su yana da mahimmanci don amfani da LJJan da tsari, sarrafa kuɗi, da jajere da iyakokin su.
Menene alamun (tokens)
TOKEN → the unit LLMs process text in (not words/characters, but PIECES):
→ text is split into tokens (roughly ~4 characters or ~0.75 words each in English)
→ e.g. 'unbelievable' might be 3 tokens; common words are often 1 token
→ the model processes and generates token by token
→ LLMs work in tokens (input and output are measured in tokens)
