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

    Neuronpedia

    INDEX
    Explanations

    legitimacy and validity of concerns

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     excelente
    0.44
     excellent
    0.42
     отлич
    0.41
     rnd
    0.38
     excelentes
    0.37
    excellent
    0.36
     excellente
    0.36
     தருக
    0.35
    绚
    0.35
    ácil
    0.35
    POSITIVE LOGITS
    是否
    1.05
     appropriateness
    0.96
     validity
    0.94
     legality
    0.94
     whether
    0.93
    会不会
    0.93
     adequacy
    0.93
     legitimacy
    0.89
    是否存在
    0.88
    是否有
    0.87
    Act Density 0.038%

    No Known Activations