Maybe an advantage of this setup is that you ensure that your bus factor is high and you're constantly testing it to make sure it stays high? Kind of like how Netflix uses ChaosMonkey.
Bleeding stops when the uterus shrinks back down so the huge open wound left behind by the placenta becomes a small wound. Oxytocin makes that happen, and you get that by just holding your baby. I don’t know how eating the placenta would contribute.
If it generates “I ate” and the next word can be “a” or “an”, then it will just generate one or the other based on how often they appear after “I ate”. It hasn’t decided by this point what it has eaten. After it has generated the next token, for example “I ate an”, then its next token is now limited to food items that fit the grammatical structure of this sentence so far. Now it can decide: did I eat an apple? An orange? An awesome steak? etc
I’ve discovered TVP recently. Very cheap and dense source of protein. My partner likes to joke that I’m eating dog food, and I’m starting to think that she might be right considering that dog food keeps coming up when I search for TVP.
It does not. If you enforce 0/0=1, then you end up in a situation where you can prove any two numbers are equal to each other and you end up with a useless system, so we do not allow for that.
e.g. 0=0*2 -> 0/0 = (0/0)2 -> 1=12 -> 1=2
If you get into calculus though, you’ll have ways to deal with this to some extent using limits.
We’re not on a space ship though. We’re on Earth, so what happens on this planet matters. You may care more about not having leap years, but the majority of us care about knowing approximately what the weather will look like at a given point in time and how much sunlight to expect, since those things actually affect our daily lives, whereas an extra day in a given month does not.