11 votes

Fine-tuning an LLM to write docs like it's 1995

1 comment

  1. cqns
    Link
    The obvious, OBVIOUS takeaway, etc, etc, etc. I could've predicted (pun not intentional) where this was ultimately going. The moment that LLMs are no longer wanted by the masses is the moment...

    The reassuring takeaway is that such a model can never replace a human tech writer, only augment them. The fine-tuned models have the same lack of judgement as their non-tuned siblings, and they need abundant steering. Fabrice will have to wait.

    The obvious, OBVIOUS takeaway, etc, etc, etc. I could've predicted (pun not intentional) where this was ultimately going. The moment that LLMs are no longer wanted by the masses is the moment where we're collectively going to have a new renaissance period. I could only hope this is sooner than later.

    1 vote