Content pfp
Content
@
0 reply
0 recast
0 reaction

David Furlong pfp
David Furlong
@df
the brightest minds couldn't solve it - they just added an if statement at the start of the model in o1 to finally solve this scenario (joking)
3 replies
0 recast
18 reactions

Stephan pfp
Stephan
@stephancill
on a serious note this is what makes me skeptical of measuring AI capabilities by the same classic problems/benchmarks - when the problem becomes a meme it potentially pollutes the pretraining data, so the model potentially trains on test data ask it how many "r"s are in strawberrrry
0 reply
0 recast
0 reaction