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

    Neuronpedia

    INDEX
    Explanations

    foreign/special characters

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     argument
    0.55
     earn
    0.55
     app
    0.52
     earned
    0.52
     argued
    0.50
     Argument
    0.50
     arg
    0.49
     iOS
    0.48
    "
    0.47
     CLOSE
    0.47
    POSITIVE LOGITS
    ۰۰
    0.52
    وری
    0.47
    ատ
    0.47
    skih
    0.47
    ʏ
    0.47
     সতর্ক
    0.46
     ٹینس
    0.46
    ₀
    0.45
    ués
    0.45
    ậ
    0.45
    Act Density 0.002%

    No Known Activations