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

    Neuronpedia

    INDEX
    Explanations

    mapping inputs to variables

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    怸
    0.66
    ı
    0.61
    Excelente
    0.54
    identité
    0.53
     Técnica
    0.52
    コロナ
    0.52
    polarized
    0.51
    Revenir
    0.51
    Después
    0.50
    फना
    0.50
    POSITIVE LOGITS
     ва
    0.54
     iti
    0.54
     నీ
    0.54
     ahor
    0.53
     не
    0.52
     ov
    0.52
     לא
    0.52
     vase
    0.52
     הב
    0.52
     אפ
    0.52
    Act Density 0.001%

    No Known Activations