Superposition
Noun · AI & Machine Learning
Definitions
A phenomenon in which multiple concepts or features appear to be represented in overlapping ways within the same model components rather than each having a clean dedicated slot. Superposition is discussed in interpretability because it makes internal behavior harder to disentangle.
In plain English: Different concepts being packed into overlapping internal model representations.
Example: "The interpretability study suggested superposition was causing the neuron to respond to several unrelated-looking concepts at once."