[Minor] Add in trainable intervention based on binary mask intervention

[frankaging]

Description

Add in a baseline for trainable interventions. It learns a binary mask over representation, and the mask is used to choose the dimension to do intervention.

Testing Done

A tutorial is included.

Checklist:

• [x] My PR title strictly follows the format: [Your Priority] Your Title
• [x] I have attached the testing log above
• [x] I provide enough comments to my code
• [x] I have changed documentations
• [x] I have added tests for my changes