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

    Neuronpedia

    INDEX
    Explanations

    technical expert, error rate, joke please

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     von
    0.37
    و
    0.37
     of
    0.36
    ologous
    0.35
     dụ
    0.34
    ított
    0.34
     ammon
    0.33
    吲
    0.33
    peč
    0.33
    ൊക്കെ
    0.33
    POSITIVE LOGITS
     Klinik
    0.40
     woes
    0.38
     मैं
    0.36
    terminology
    0.35
     Kerala
    0.35
     berkualitas
    0.34
     Let
    0.34
     “,
    0.34
     เริ่ม
    0.34
    ៦
    0.33
    Act Density 0.148%

    No Known Activations