Token value that indicates the encoding used to decode the source bytes into text. The first token returned by tokenize.tokenize() will always be an ENCODING token.
The official docs include a CPython implementation detail for this entry.
token/ENCODING
ENCODING