Generating One Token at a Time is a project focused on exploring how large language models generate text step by step. The process involves predicting one token at a time, where each new token is determined based on the context established by the preceding tokens.
AtharvaDomale/Generating-One-Token-at-a-Time
Folders and files
| Name | Name | Last commit date | ||
|---|---|---|---|---|