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

    Neuronpedia

    INDEX
    Explanations

    pronouns

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    ThroughAttribute
    -0.75
    fır
    -0.61
     certeza
    -0.60
    usermodel
    -0.59
    hésite
    -0.59
    pounds
    -0.58
    Примі
    -0.56
    勁
    -0.56
    umumkan
    -0.55
     mía
    -0.53
    POSITIVE LOGITS
     should
    0.76
     shouldn
    0.74
    Should
    0.73
    Shouldn
    0.69
     Should
    0.68
    should
    0.66
     devraient
    0.59
     Shouldn
    0.56
     shouldnt
    0.55
    InjectAttribute
    0.54
    Act Density 0.004%

    No Known Activations