Big endian means the most significant values are stored first.
Little endian means the least significant values are stored first.
They are about byte order not bit order. In all cases, the most significant bits within each byte come first.
x86 is little endian.
So take a UTF-8 Byte Order Mark (BOM). It stores the unicode value u+feff.
In a big-endian system, it is stored feff.
In a little-endian system, it is stored fffe.
But in UTF-8 it is stored efbbbf. In UTF-8 you only have individual bytes, so endianness has no effect.