Every story and comment from the Hacker News API, in a layout borrowed from 2001.
A comment on Building LLM Applications for Production. Parent comment
Useful answer - fine tune on large training set, set temperature to 0, monitor token probability and highlight risk when probability < some threshold.
There are no comments on this comment yet.
Useful answer - fine tune on large training set, set temperature to 0, monitor token probability and highlight risk when probability < some threshold.