Same idea here? Larger models do a better job forgetting their training data and dropping their semantic priors. Perhaps another way of thinking through this is

Same idea here? Larger models do a better job forgetting their training data and dropping their semantic priors. Perhaps another way of thinking through this is that larger models learn new information and drop old information faster. https://arxiv.org/abs/2303.03846 https://arxiv.org/abs/2303.03846 Isn't that interesting? The idea of "mental liquidity", or "strong opinions weakly held"? https://news.ycombinator.com/item?id=36280772 https://news.ycombinator.com/item?id=36280772