5 ms·

Except that's literally how LLMs work. Small changes to the prompt or training can greatly affect its output.

• alganet 1y ago
  [flagged]
  • Trasmatta 1y ago
    I legitimately don't understand what your point is. Are you saying that they actually did this intentionally?
    • alganet 1y ago
      [flagged]