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

    Neuronpedia

    INDEX
    Explanations

    even important

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    安全
    0.49
    地说
    0.47
    重新
    0.47
    ｇ
    0.46
    ებით
    0.46
    ﺳ
    0.46
    一个
    0.45
    说
    0.44
     collectively
    0.44
    组织的
    0.43
    POSITIVE LOGITS
     austerity
    0.50
     används
    0.48
     options
    0.46
    琨
    0.46
     individu
    0.44
     individuelle
    0.44
     haunts
    0.44
    ampton
    0.44
    -'
    0.44
     customizations
    0.44
    Act Density 0.004%

    No Known Activations