A specific phenomenon when the when model discrepancy on Validation dataset is not close to the to model discrepancy on Training dataset and means that a given model realization has "remembered" the Training dataset but can not accurately predict on the data points outside the Training dataset.
...