Some details are sent.
Others make you recognizable.
When a browser connects to a website, the website's server receives your network address and request information. The page can also inspect many browser-visible properties, subject to browser rules and privacy settings. Each item may be ordinary on its own. In combination, the signals can make one browser stand out.
Cookies and other storage can keep identifiers between visits. Tracking companies may use them to connect activity across sites when the browser allows it. Browsers now partition or block some cross-site storage, but the exact protections depend on the browser, settings, exceptions, and extensions in use.
Network + referrer
A website's server sees the network address connecting to it. Depending on the browser and link, the request may also include a referrer, browser details, and language preferences. A web page cannot discover your public IP by itself without sending a request to some server.
Cookies + “supercookies”
Cookies are only one place websites can keep state. Local storage, IndexedDB, caches, and service workers can also retain data for a site. “Supercookie” is a catch-all term for tracking that uses multiple or less-visible storage paths to keep an identifier around; cache validators and other browser behaviors have also been abused as tracking channels.
Browser + device clues
Screen size, time zone, language, graphics information, available fonts, hardware hints, and feature support can contribute to a browser fingerprint. Some details are reduced, masked, or unavailable; the combination can still be distinctive.
Location + sensors
Websites can ask for location, camera, microphone, notifications, and other access. A permission already granted can reveal more. This scan reads permission status where the browser permits it; it never opens a camera, microphone, or location sensor.