Chapter 14, Interpreting Black Box Transformer Models
- BERTViz only shows the output of the last layer of the BERT model. (True/False)
False. BERTViz displays the outputs of all the layers.
- BERTViz shows the attention heads of each layer of a BERT model. (True/False)
True.
- BERTViz shows how the tokens relate to each other. (True/False)
True.
- LIT shows the inner workings of the attention heads like BERTViz. (True/False)
False. However, LIT makes non-probing predictions.
- Probing is a way for an algorithm to predict language representations. (True/False)
True.
- NER is a probing task. (True/False)
True.
- PCA and UMAP are non-probing tasks. (True/False)
True.
- LIME is model-agnostic. (True/False)
True.
- Transformers deepen the relationships of the tokens layer by layer. (True/False)
True.
-
...
