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

    Neuronpedia

    INDEX
    Explanations

    filler, padding, positions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    isSuccessful
    -0.80
    bringen
    -0.77
    cimientos
    -0.75
     committed
    -0.74
    ⏟
    -0.72
     死
    -0.71
    Hardcover
    -0.69
     enseignants
    -0.69
    χη
    -0.68
    BIÉN
    -0.68
    POSITIVE LOGITS
    positions
    1.03
     positions
    1.02
     FILLER
    1.01
     Filler
    0.98
     padding
    0.98
    Positions
    0.95
     relleno
    0.95
     Positions
    0.93
     Picture
    0.91
    Filler
    0.90
    Act Density 0.052%

    No Known Activations