Ahmadzei's picture
added 3 more tables for large emb model
5fa1a76
It replaces the recurrent layers most commonly used in encoder-decoder architectures with multi-headed self-attention .