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

    Neuronpedia

    INDEX
    Explanations

    low-stakes or no-stakes

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    a
    0.80
    te
    0.73
    一定的
    0.69
     attended
    0.68
    高的
    0.68
    at
    0.67
    y
    0.66
    ate
    0.66
     analysed
    0.66
    ष्क
    0.65
    POSITIVE LOGITS
     Hanging
    0.86
     Fiesta
    0.85
    льний
    0.83
     Ceux
    0.82
     topo
    0.82
     Анастасия
    0.81
     Reli
    0.80
     lollipop
    0.80
     Babel
    0.80
     IUP
    0.80
    Act Density 0.000%

    No Known Activations