The Encoder is one of the two main blocks of the original Transformer Architecture. Its primary purpose is to receive the pre-processed input tokens and convert them into Vector Embeddings that capture the semantic meaning of the input text.
Context:
- The transformer architecture predominantly consists of two main blocks the first is the encoder block.
- The main purpose of the encoder is actually to take in the input text and to convert those tokens into Vector Embeddings.
- The Bidirectional Encoder Representations from Transformers (BERT) model consists only of the encoder block.
