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

    Neuronpedia

    INDEX
    Explanations

    Creating a model response

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     ente
    0.42
    堵
    0.42
    敘
    0.41
     designee
    0.38
     acesso
    0.38
    (&:
    0.38
     cumulative
    0.37
     Saga
    0.37
     नवे
    0.37
     passag
    0.37
    POSITIVE LOGITS
     Creating
    0.42
    <h1>
    0.41
     здесь
    0.41
    Creating
    0.40
    这里的
    0.40
     here
    0.39
     aquí
    0.39
     yep
    0.38
     Freizeit
    0.38
    Here
    0.38
    Act Density 0.017%

    No Known Activations