News for hackers. Stuff that matters.
Re:Native JSON Output from GPT-4 by sanxiyn on Wednesday June 14, 2023 @04:09PM (#36331851)
For various reasons, token selection may be implemented as upweighting/downweighting instead of outright ban of invalid tokens. (Maybe it helps training?) Then the model could generate malformed JSON. I think it is premature to infer from "can generate malformed JSON" that OpenAI is not using token selection restriction.

There are no comments on this comment yet.