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

    Neuronpedia

    INDEX
    Explanations

    conversation gradually pushed

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    cip
    1.24
    器
    1.19
    k
    1.16
    }".
    1.15
    意义
    1.14
    cwd
    1.14
     चारा
    1.14
    oce
    1.12
    cars
    1.12
    pw
    1.12
    POSITIVE LOGITS
     Möglichkeit
    1.23
     hinzuge
    1.18
     indistinct
    1.17
     Paraná
    1.13
     Lebih
    1.13
     unbedingt
    1.12
     Förderung
    1.11
     જેમ
    1.11
    েলের
    1.10
     оказался
    1.07
    Act Density 0.001%

    No Known Activations