As the model gets smarter, you need to tell it less
It goes the opposite way too. Small models need minimal harness like pi or they get overwhelmed by long system message and tool descriptions
As the black box gets bigger, you need to stop expressing yourself even more.
Perhaps, but then the value of each thing you tell it goes to.
It goes the opposite way too. Small models need minimal harness like pi or they get overwhelmed by long system message and tool descriptions