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

    Neuronpedia

    INDEX
    Explanations

    calling parent/base constructor/method

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    𝐬
    0.70
     an
    0.70
    ための
    0.68
     Medicines
    0.66
    ژ
    0.66
    사
    0.65
    falling
    0.63
    ffic
    0.63
    justice
    0.61
    უნქ
    0.61
    POSITIVE LOGITS
    ।
    0.80
     I
    0.77
    。
    0.71
     departe
    0.66
     exfoli
    0.56
     приве
    0.56
     مختلفة
    0.55
     ।
    0.54
    iš
    0.54
    ";
    0.54
    Act Density 0.000%

    No Known Activations