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

    Neuronpedia

    INDEX
    Explanations

    Rape, Initialization, Strategy, Default

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     hydride
    0.76
     التحكم
    0.72
    halle
    0.71
    hin
    0.71
     FLOW
    0.69
    stile
    0.69
    HexString
    0.67
     homicide
    0.67
    FLOW
    0.67
     vintage
    0.66
    POSITIVE LOGITS
     إن
    0.62
     اعتراض
    0.57
    긋
    0.57
    Leod
    0.56
    Strategy
    0.56
    тах
    0.55
    Attr
    0.55
    쳥
    0.54
    ಗ
    0.54
    ائی
    0.54
    Act Density 0.047%

    No Known Activations