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

    Neuronpedia

    INDEX
    Explanations

    smoothie

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ções
    1.59
     dismay
    1.59
    ay
    1.53
            
    1.52
    ர்
    1.42
     scar
    1.39
     trasera
    1.39
    ñas
    1.37
     penal
    1.30
    ן
    1.30
    POSITIVE LOGITS
    tı
    2.13
    tır
    1.90
    thed
    1.77
    tions
    1.76
    تها
    1.70
    ารย์
    1.69
    t
    1.67
    tional
    1.64
    تش
    1.59
    truck
    1.57
    Act Density 0.001%

    No Known Activations