Greedy decoding
Always take the single most likely next token. Deterministic and locally optimal, which is exactly why it tends to loop and produce flat text.
Companion explanation in Step by Token, chapter 7.
Where this term gets built
Continue
Always take the single most likely next token. Deterministic and locally optimal, which is exactly why it tends to loop and produce flat text.
Companion explanation in Step by Token, chapter 7.
Continue