Tim

Please take a look at #133 and check whether #133 is sufficient already.

Related to: - #126 - #223

If this PR doesn't add a new eval you can remove the eval specific fields from the PR description.

Could you explain how this automated test can find bias in the answer the LLM gives?

Please fill out the PR template with some details on this change, thank you!

Something seems to be wrong with the commits in this PR, could you please have a look @Eric0101?

Please fill out the fields in the PR template.

Please add some context to this issue, thanks!

ChatGPT (GPT-3) result is, to phrase it friendly, interesting 😄 

Related to: - #55