Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    solutions, usage

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     the
    -1.40
     a
    -1.13
     an
    -0.93
     those
    -0.84
     your
    -0.82
     various
    -0.81
     some
    -0.79
     our
    -0.77
     their
    -0.75
     what
    -0.69
    POSITIVE LOGITS
    .
    1.02
     in
    0.82
     because
    0.70
     while
    0.70
     during
    0.68
    ;
    0.67
    ,
    0.67
     with
    0.66
     for
    0.66
     throughout
    0.64
    Act Density 0.063%

    No Known Activations