Event JSON
{
"id": "5b63aaf80e9a6436af49015963bbec8f72f921661faf721a1bf3c6f08055dcf1",
"pubkey": "6ec01da149886b4554de8f3326868c4a40600ca01ce6d1ae1496c649d3800d87",
"created_at": 1703207575,
"kind": 1,
"tags": [
[
"p",
"640e31bbcbf503b8aec585063c5e8bd7255942dd9f66bac7053eb25e509d67a3",
"wss://relay.mostr.pub"
],
[
"p",
"e780dfc7f1fe58b31b0ca8d26d5f5bcb2e610fe70d8eb0db1594d4bf8a9db5ec",
"wss://relay.mostr.pub"
],
[
"e",
"3cf792c6ef3a947bb3eff66de1301637e05737b7b0c69db9b3e7219aa5de4d9f",
"wss://relay.mostr.pub",
"reply"
],
[
"proxy",
"https://mastodon.social/users/ratkins/statuses/111621411662797912",
"activitypub"
]
],
"content": "nostr:npub1vs8rrw7t75pm3tk9s5rrch5t6uj4jskanant43c986e9u5yav73sek54zy I wonder if we’ll learn more about architectures/training regimens which will tell us that smaller, better quality data sets are a viable way forward—train on, say, Getty’s library rather than random shit pulled from Instagram or AAP’s news library rather than weeb subreddits.",
"sig": "b34a5937507d7a018168c981d3be679ec9594181218291c1195c45b10fd272a8448a53113e53f52a42c16c09530a697bfe5ba9faea843d90489e5d3ab64f0a14"
}
