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

    Neuronpedia

    INDEX
    Explanations

    sameness

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    -
    0.70
    (
    0.68
     to
    0.65
     there
    0.64
    \
    0.61
    there
    0.60
    '
    0.57
    िव
    0.57
     and
    0.57
    ่
    0.56
    POSITIVE LOGITS
     continúa
    0.73
    ры
    0.73
     diligent
    0.72
     publicada
    0.71
     Cualquier
    0.69
     说
    0.69
     tireless
    0.68
    رك
    0.67
     Тол
    0.67
     jóvenes
    0.67
    Act Density 0.008%

    No Known Activations