4 ms·
very different initial approaches to training dataset curation, to SD's detriment.
by napier 3y ago
very different initial approaches to training dataset curation, to SD's detriment.
- GaggiX 3y agoCan you be more explicit about what you mean? Also, the biggest difference is probably just the fact that MJ is using a much larger model than SD.