FlashcardBeginner
Next-Token Prediction
LLMs work by repeatedly predicting the most likely next token given the preceding text.
Question
How does an LLM generate text at its core?
Click to reveal answer
LLMs work by repeatedly predicting the most likely next token given the preceding text.
How does an LLM generate text at its core?