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

    Neuronpedia

    INDEX
    Explanations

    capacity and size measurements

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     ۔
    0.51
    iz
    0.50
     ر
    0.49
    lik
    0.47
    jší
    0.46
     র
    0.45
    или
    0.45
    یک
    0.45
     iliş
    0.44
     інші
    0.44
    POSITIVE LOGITS
    at
    0.73
    on
    0.61
    지
    0.56
    ad
    0.54
    em
    0.54
    中
    0.50
    ak
    0.50
    as
    0.49
    ной
    0.49
    in
    0.49
    Act Density 0.000%

    No Known Activations