INDEX
    Explanations

    The neuron consistently activates on occurrences of the word “another.”

    New Auto-Interp
    Negative Logits
    App
    -0.07
    reation
    -0.07
     agency
    -0.06
    Books
    -0.06
    bruary
    -0.06
    ween
    -0.06
    _ROUTE
    -0.06
     Payments
    -0.06
    /application
    -0.06
    resents
    -0.06
    POSITIVE LOGITS
    .seconds
    0.07
     uží
    0.06
     lifted
    0.06
     didSelectRowAtIndexPath
    0.06
     poster
    0.06
     relent
    0.06
     다른
    0.06
     λίγ
    0.06
    0.06
    ;'
    0.06
    Act Density 0.035%

    No Known Activations