[–] 2 points 3 months ago

Lower temperature is not necessarily better quality. At 0 it will get very repetitive, it's for classification tasks, not for prose.

I suppose the optimal temperature depends on the model and the task, and I don't think it's very sensitive. Varying temp might even give better results, who knows.

  • source
  • parent
  • context
  • [–] 18 points 3 months ago (4 children)

    By being just a little weird, but in a pattern, and not to you. A pattern like: every n tokens raise the temperature (randomness) for one token. Then to dectect AI, you tokenize and calculate how expected each next token is. Then you try fitting the pattern to that.

  • source
  • parent
  • context
  • view more: next ›