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

    Neuronpedia

    INDEX
    Explanations

    large egg

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Tö
    0.70
     Desen
    0.65
     Beschäft
    0.64
     Sohn
    0.64
    Сі
    0.63
     incó
    0.63
     następ
    0.62
     Grü
    0.61
     Có
    0.61
     próxima
    0.61
    POSITIVE LOGITS
     thats
    1.05
     that
    1.04
    that
    0.99
     which
    0.91
    thats
    0.90
    which
    0.89
     yang
    0.87
     которые
    0.83
     которое
    0.83
     being
    0.82
    Act Density 0.000%

    No Known Activations