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

    Neuronpedia

    INDEX
    Explanations

    actions via methods

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     hyvä
    0.74
    來講
    0.70
     обосно
    0.69
    седнев
    0.69
    эй
    0.68
    আপ
    0.67
    êmement
    0.66
     آگے
    0.66
    └──
    0.65
    にとって
    0.65
    POSITIVE LOGITS
     using
    4.56
     via
    4.48
     through
    3.96
    using
    3.95
     melalui
    3.87
     Using
    3.81
     menggunakan
    3.73
     tramite
    3.70
     usando
    3.68
    Using
    3.67
    Act Density 2.032%

    No Known Activations