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

    Neuronpedia

    INDEX
    Explanations

    primarily

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    व
    0.67
    м
    0.57
    et
    0.55
     as
    0.54
     más
    0.54
    ある
    0.52
    多个
    0.52
    不止
    0.52
    विल
    0.48
     many
    0.47
    POSITIVE LOGITS
    mainly
    0.59
     primarily
    0.57
     mainly
    0.57
    sächlich
    0.56
     ব্যবহৃত
    0.55
    mostly
    0.55
     Mainly
    0.54
    primarily
    0.54
    stays
    0.52
     principalement
    0.52
    Act Density 0.051%

    No Known Activations