Test your understanding of RLHF, negative constraints, formatting rules, and safety triage.
Question 1 of 66 unanswered
A prompt asks the model to 'write a response that does NOT contain the letter e'. The AI output is: 'This is a prompt that does not have any violations.' Is this output compliant?