@seanhart
Is it just me, or does GPT4 respond better if questioned about content that might contain an error, without pointing out the error? Inevitably, if I accuse it of an error, it gets tied in logic loops much more quickly (flustered?)
1 reply
0 recast
0 reaction