3 ms·
The technology that allows an LLM to "see" images and video is completely different though. It's not what is being trained on common crawl.
by ryaniscool 2y ago
The technology that allows an LLM to "see" images and video is completely different though. It's not what is being trained on common crawl.
- brrrrrm 2y agonot really. embeddings are embeddings. check out llava