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

    Neuronpedia

    INDEX
    Explanations

    parentheses or percentages followed by descriptions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    وم
    0.54
    ان
    0.49
    段时间
    0.47
     От
    0.46
    ма
    0.46
    𝑮
    0.45
    اریخ
    0.45
    ৩
    0.44
    两
    0.44
    厢
    0.43
    POSITIVE LOGITS
    (
    0.54
     be
    0.52
    EN
    0.44
    y
    0.44
    D
    0.43
    ate
    0.42
     vý
    0.42
     regrettable
    0.42
    지
    0.41
    en
    0.41
    Act Density 0.001%

    No Known Activations