Decode this: ENT1yWWhLL-Qz9mK

A Twitter snowflake ID is a 64-bit binary number, with 41 bits representing a timestamp, 10 bits representing a machine ID, and 12 bits representing a per-machine sequence number. The snowflake ID is generally serialized in decimal format and is used in the post URL.

For example,
https://twitter.com/i/status/1581674019991916545
Decoding 1581674019991916545:
{'creation_time_milliseconds': 1665935447105,
'datacenter_id': 10,
'machine_id': 337,
'sequence_id': 1,
'server_id': 17}

The filename of a Twitter photo also contains a base64-encoded snowflake ID. This can be decoded into a regular snowflake ID, although it contains three "extra" unknown bytes.
For example,
https://pbs.twimg.com/media/FdMR2WDaMAMPt2q?format=jpg
FdMR2WDaMAMPt2q -> 1572620320212922371

It is unknown if the filename of a Twitter video also contains a snowflake ID. You can see the URL contains it, but any off-platform video would would only contain the file name.
For example,
https://video.twimg.com/ext_tw_video/1564378065198882816/pu/vid/960x720/ENT1yWWhLL-Qz9mK.mp4

How would "ENT1yWWhLL-Qz9mK" be decoded into a snowflake, if it's possible?

Resources:
https://en.wikipedia.org/wiki/Snowflake_ID
https://discord.com/developers/docs/reference#snowflakes (good visualizations, twitter uses another epoch)
https://docs.google.com/document/d/1xVrPoNutyqTdQ04DXBEZW4ZW4A5RAQW2he7qIpTmG-M/edit
>>1263648