3 ms·
That's not how it works. LLMs were only trained on text, so this is new data it's never seen before. There's no train-test leakage.
by Legend2440 3y ago
That's not how it works. LLMs were only trained on text, so this is new data it's never seen before. There's no train-test leakage.
- darig 3y ago[dead]
- nuancebydefault 3y agoYou are right but I am sure there are patterns in the weights that (accidentally or not) predict patterns commonly seen in media.