INDEX
    Explanations

    The neuron responds to occurrences of the root “intersect” (e.g. “intersection,” “intersecting,” “intersects”).

    New Auto-Interp
    Negative Logits
    prog
    -0.07
    _d
    -0.07
     Wales
    -0.07
    modal
    -0.06
    .d
    -0.06
     cola
    -0.06
    La
    -0.06
     d
    -0.06
    Code
    -0.06
    -0.06
    POSITIVE LOGITS
     intersect
    0.12
     intersection
    0.11
     intersects
    0.10
     Intersection
    0.09
     intersections
    0.09
    intersect
    0.08
    _intersection
    0.08
    Intersection
    0.08
    intersection
    0.08
     Rencontres
    0.08
    Act Density 0.004%

    No Known Activations