[asdfsa32]

And often -- without using code _openly_ or in thinking -- it is wrong because it is just a "guesstimate" at best due to the nature of how LLMs generate tokens.