6 ms·
Just like for llms for the base model you need quantity not quality. It just needs to learn how to correctly predict the next frame.
by leumon 1y ago
Just like for llms for the base model you need quantity not quality. It just needs to learn how to correctly predict the next frame.