Re:A GPT-4 capability forecasting challenge
by wavemode on Saturday September 02, 2023 @04:43PM
(#37365366)
That tripped me up, too. I was often reading the prompt and confusing it for the evaluation criteria, when they are actually two separate things. The prompt stated the flag must be perfectly accurate, yes, but the evaluation criteria allowed the flag to not be quite right.
[ Parent
| on HN ]
There are no comments on this comment yet.