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

    Neuronpedia

    INDEX
    Explanations

    a father, a doctor, a project

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     pouquinho
    0.30
    льнявыя
    0.29
     diamètre
    0.28
     filmpje
    0.27
     whiche
    0.27
     wodurch
    0.27
     яе
    0.26
     carénés
    0.26
     interessant
    0.26
     মিজান
    0.26
    POSITIVE LOGITS
    0.45
    :
    0.39
    ,
    0.30
     and
    0.26
     the
    0.25
    ↵↵
    0.24
     Chicago
    0.24
    $
    0.24
     political
    0.24
     $
    0.24
    Act Density 0.706%

    No Known Activations