Questions

Let's evaluate our understanding of BERT by trying to answer the following questions:

1. How does BERT differ from other embedding models?
2. What are the differences between the BERT-base and BERT-large models?
3. Define segment embedding.
4. How is BERT pre-trained?

5. How does the masked language modeling task work?
6. What is the 80-10-10% rule?
7. How does the NSP task work?