arXiv cs.CL
· Papers
Off-Axis, On Purpose: Where a Transformer Computes Concepts and Why it Does So
arXiv:2608.10251v1 Announce Type: new Abstract: A transformer's answer lives on one axis: the direction its unembedding reads. Its intermediate states largely do not, and that off-axis position is usually treated as an obstacle to interpretation. We show it is functional. A 12-layer model computes in two phases. Throug