Putting neural networks under the microscope
01 February 2019
Rob Matheson | MIT News

Researchers pinpoint the “neurons” in machine-learning systems that capture specific linguistic features during language-processing tasks.