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

    Neuronpedia

    INDEX
    Explanations

    hope for better

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ੍ਹ
    0.79
     dẫn
    0.75
    decoration
    0.75
     memungkinkan
    0.75
    Include
    0.74
     acondicionado
    0.73
     disarmament
    0.73
     אשר
    0.72
    รู้สึก
    0.70
    millimeters
    0.69
    POSITIVE LOGITS
    ा
    0.80
    ols
    0.75
     chinois
    0.70
     मित्त
    0.70
    ۰
    0.69
     народа
    0.68
    క్కువ
    0.67
     Moshe
    0.66
     {}'.
    0.65
    asshi
    0.64
    Act Density 0.001%

    No Known Activations