Ask HN: What's everyone's experience with AI evaluation/testing tools?

1 points by atmospherejeans 1y ago ↗ HN

Hi everyone, I was wondering what experiences people have with AI evaluators. I’ve mostly used the AI models as they are and haven’t done a lot of fine-tuning or evaluating.

What tools do people use, if any? And what benefits/issues do people have while using them?

2 comments

[ 3.3 ms ] story [ 14.2 ms ] thread

[dead] Daniel_Sums 1y ago ↗

[dead]

atmospherejeans 1y ago ↗

Thanks for the detailed response! Have you used any of those tools yourself? Were there any that made the evaluation process easier, or were they all relatively complex in their own way?