Learn Before
Language Model Token Selection
Based on the provided scenario, describe the two crucial, sequential actions the model must perform on the reduced set of candidates to determine the final output token.
0
1
Tags
Ch.5 Inference - Foundations of Large Language Models
Foundations of Large Language Models
Foundations of Large Language Models Course
Computing Sciences
Application in Bloom's Taxonomy
Cognitive Psychology
Psychology
Social Science
Empirical Science
Science
Related
Language Model Token Selection
A language model has calculated the probabilities for the next possible tokens in a sequence. The five most likely tokens are: 'the' (0.4), 'a' (0.2), 'on' (0.1), 'in' (0.05), and 'at' (0.05). If the model uses a selection process where only the top 3 candidates are considered (k=3), what will be the new, renormalized probability distribution for the tokens that are ultimately sampled from?
A language model is generating text and has just identified the 'k' most likely next tokens from a large vocabulary. Arrange the following actions in the correct chronological order to complete the generation of the single next token.