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

    Neuronpedia

    INDEX
    Explanations

    common elements and significant issues

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     the
    0.69
     The
    0.68
    te
    0.66
    जफ्
    0.66
    粈
    0.64
    수의
    0.63
    The
    0.63
     Vys
    0.63
    ']*
    0.62
     পাবার
    0.62
    POSITIVE LOGITS
    9
    0.79
     आहे
    0.78
    7
    0.77
    0.76
    3
    0.74
     .
    0.72
    5
    0.70
    4
    0.70
    2
    0.69
     {
    0.64
    Act Density 0.447%

    No Known Activations