• 0 Posts
  • 74 Comments
Joined 1 year ago
cake
Cake day: July 3rd, 2023

help-circle

  • What’s funny to me is, the agile approach seems like it’s a much better fit for open-source, non-commercial software development.

    The corporate world and is management practices based around quarters and deadlines can’t seem to see how anything could get done without deadlines, but that’s usually less of a factor with open-source. People laugh at “scrum masters” because in a corporate environment, all the scrum stuff tends to be mostly performative. But it seems to me that open-source projects with multiple contributors already kind of work in an agile manner.
















  • I don’t like the idea of restricting the model’s corpus further. Rather, I think it would be good if it used a bigger corpus, but added the date of origin for each element as further context.

    Separately, I think it could be good to train another LLM to recognize biases in various content, and then use that to add further context for the main LLM when it ingests that content. I’m not sure how to avoid bias in that second LLM, though. Maybe complete lack of bias is an unattainable ideal that you can only approach without ever reaching it.