Event JSON
{
"id": "68fd4a42880b0203e214f40f447d28f28ebb6ed4c3eeea2a2e43b062a2b81122",
"pubkey": "9fec72d579baaa772af9e71e638b529215721ace6e0f8320725ecbf9f77f85b1",
"created_at": 1736862314,
"kind": 1,
"tags": [
[
"e",
"c5576144027c30d18afb06b872e2f60b78f427e9fc735f0ea09b7c4e1abf5133",
"wss://relay.lumina.rocks",
"root"
],
[
"e",
"1d98ed6c1243f948e86b4c5ed958bae3a7321ecbf95ce6e34dfcb32722563fdf",
"wss://relay.lumina.rocks",
"reply"
],
[
"p",
"9fec72d579baaa772af9e71e638b529215721ace6e0f8320725ecbf9f77f85b1"
],
[
"p",
"ae1008d23930b776c18092f6eab41e4b09fcf3f03f3641b1b4e6ee3aa166d760"
],
[
"r",
"wss://e.nos.lol/"
],
[
"r",
"wss://nos.lol/"
],
[
"r",
"wss://nostr.land/"
],
[
"r",
"wss://nostr.mom/"
],
[
"r",
"wss://nostr.oxtr.dev/"
],
[
"r",
"wss://nostr.wine/"
],
[
"r",
"wss://nostr21.com/"
],
[
"r",
"wss://nostrue.com/"
],
[
"r",
"wss://relay.0xchat.com/"
],
[
"r",
"wss://relay.damus.io/"
],
[
"r",
"wss://relay.nos.social/"
],
[
"r",
"wss://relay.nostr.band/"
],
[
"r",
"wss://relay.snort.social/"
],
[
"r",
"wss://relay.primal.net/"
]
],
"content": "Download all the notes. \nTake the \"content\" field from the notes and change the name to \"text\":\n\nPreviously:\n{\"id\":\".....................\", \"pubkey\": \".................\", \"content\": \"gm, pv, bitcoin fixes this!\", .......}\n{\"id\":\".....................\", \"pubkey\": \".................\", \"content\": \"second note\", .......}\n\n\nConverted into jsonl file:\n{\"text\": \"gm, pv, bitcoin fixes this!\" }\n{\"text\": \"second note\" }\n\nUsed Unsloth and ms-swift to train. Unsloth needed to convert from base to instruct. This is a little advanced. If you don't want to do that and just start with instruct model, you can use ms-swift or llama-factory.\n\nYou will do lora, pretraining. I used 32 as lora rank but you can choose another number.",
"sig": "55650ee7b168e19c3784ba399bfed4d45460fdc3a2a73cdb64dfba84ee5d6aaafa422e035bf3961590b8211d6e876b648490440b87893e4141c9c3cc57ec1b7b"
}