But after the 10^10 point, something interesting happens: the score starts growing much faster (~N). And for some tasks, the plot looks like a hockey stick (a sudden change from ~0 to almost-human).
But after the 10^10 point, something interesting happens: the score starts growing much faster (~N).
And for some tasks, the plot looks like a hockey stick (a sudden change from ~0 to almost-human).
Seems interestingly similar to the grokking phenomenon.
Seems interestingly similar to the grokking phenomenon.