3 ms·

> we can decode any embedding vectors, given we have at least a few thousand of them. Do you think that could be sufficient to translate the Voynich manuscript

> we can decode any embedding vectors, given we have at least a few thousand of them. Do you think that could be sufficient to translate the Voynich manuscript?