Thus the training data didn't just contain text, but rather text where each passage is tagged and attributed to a particular user.
This aspect of the training data was critical in creating the illusion of talking to another person.
An LLM doesn't just predict the next text. It predicts the next text that might come from another user. You need to hard code this in to make it work well.
Leave it out and there is no conversation.
myrmepropagandist
npub1cp…uhrnq
2026-05-01 10:15:38 UTC
in reply to nevent1q…nmzs
Author Public Key
npub1cp2pgntkzkpqa23rytnchggwzywyggvst9yzkgd6w8j349ef7s9shuhrnqSeen on
wss://relay.ditto.pubPublished at
2026-05-01 10:15:38 UTCEvent JSON
{
"id": "724c32ea75c162b57ae8d53552e9b05342574caee8845a56ac94ad2d77e7d2e4",
"pubkey": "c054144d7615820eaa2322e78ba10e111c44219059482b21ba71e51a9729f40b",
"created_at": 1777630538,
"kind": 1,
"tags": [
[
"e",
"8eb48d90e305e840cdab35fdd1899c4b2c53410a60eb34c4534aebf9b28d28c9",
"wss://relay.ditto.pub",
"reply"
],
[
"proxy",
"https://sauropods.win/users/futurebird/statuses/116498794998354046",
"activitypub"
],
[
"client",
"Mostr",
"31990:6be38f8c63df7dbf84db7ec4a6e6fbbd8d19dca3b980efad18585c46f04b26f9:mostr",
"wss://relay.ditto.pub"
]
],
"content": "Thus the training data didn't just contain text, but rather text where each passage is tagged and attributed to a particular user. \n\nThis aspect of the training data was critical in creating the illusion of talking to another person. \n\nAn LLM doesn't just predict the next text. It predicts the next text that might come from another user. You need to hard code this in to make it work well. \n\nLeave it out and there is no conversation.",
"sig": "f3a6f30a6cb9b6edbd64ce9000fb8edd112b5f43e8ec8bb94367605f1810a575ac049c8eaf1c0b863ccad9539086d0fddbefaa4001230a55a9e1ae9b9cf7a84e"
}