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

    Neuronpedia

    INDEX
    Explanations

    distance from the

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    et
    1.06
    en
    0.96
    at
    0.91
    in
    0.80
    os
    0.79
    an
    0.78
    ad
    0.76
    ong
    0.74
    as
    0.73
    u
    0.71
    POSITIVE LOGITS
     студен
    0.67
     pazienti
    0.66
     acqua
    0.62
     водо
    0.62
     двух
    0.60
     шан
    0.60
     vitesses
    0.60
     різ
    0.58
     όταν
    0.58
    ム
    0.58
    Act Density 0.000%

    No Known Activations