Dan on Nostr: the impressive bit isn't the model itself, but *how* they trained it. It's the ...
the impressive bit isn't the model itself, but *how* they trained it. It's the training technique that's blowing the markets collective mind.
Luckily that research is totally open source so we'll be able to apply the same training techniques to our open source models, achieving the same massive efficiency gains but with less CCP fuckery
Published at
2025-01-28 01:15:36Event JSON
{
"id": "99bd6e59004f41f590504a761f31390ac98fb553b1b2dbcccd5022661932d7b2",
"pubkey": "40bdcc08888d98d49d03fb93bffd1cc46d9b8f187d08c9fc81b4adf0ad00fd2c",
"created_at": 1738026936,
"kind": 1,
"tags": [
[
"p",
"c07a2ea48b6753d11ad29d622925cb48bab48a8f38e954e85aec46953a0752a2",
"",
"mention"
],
[
"p",
"3f770d65d3a764a9c5cb503ae123e62ec7598ad035d836e2a810f3877a745b24",
""
],
[
"e",
"8ea82027169ef05d7aac369ce8dc1da2504d16a14ae22833629bc38db80cea1a",
"wss://relay.nostrplebs.com",
"root",
"3f770d65d3a764a9c5cb503ae123e62ec7598ad035d836e2a810f3877a745b24"
]
],
"content": "the impressive bit isn't the model itself, but *how* they trained it. It's the training technique that's blowing the markets collective mind. \n\nLuckily that research is totally open source so we'll be able to apply the same training techniques to our open source models, achieving the same massive efficiency gains but with less CCP fuckery",
"sig": "d1e4d0106e6380417d4cd882470f5ddfd573f44dcb86b546fbeab9c4d0f18d6b6d875b9a821f856e279bfaab1c35409778185edbbe32e3a5dcfc4e34b072bf36"
}
