Yes because in most contexts it has seen "caveman" talk the conversations haven't been about rigorously explained maths/science/computing/etc... so it is less l

Yes because in most contexts it has seen "caveman" talk the conversations haven't been about rigorously explained maths/science/computing/etc... so it is less likely to predict that output.