The neuron consistently lights up on predicate-style tokens in formal/specification contexts—things like boolean literals (true, false), relation terms (adjacent), status or property words (complete, optimal, set, signal) that appear in precise “if … then” or definition statements.
