Top Answers

1. Why Bert transformer uses [CLS] token for classification instead of average over all tokens?
   23 votes
2. what the difference between att_mask and key_padding_mask in MultiHeadAttnetion
   17 votes
3. Efficient metrics evaluation in PyTorch
   13 votes
4. How to get immediate next word probability using GPT2 model?
   13 votes
5. BERT sentence embeddings from transformers
   12 votes