Sensitivity is defined as:

Sensitivity can be considered as a measure of how good the model is at predicting the positive outcome. As you can see, it is obtained by computing the ratio between the number of positives predicted from the model that were actually positive, and the number of positives that the model should have predicted.