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

    Neuronpedia

    INDEX
    Explanations

    suite numbers in addresses

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ת
    1.05
    ↵
    0.90
    u
    0.89
     pubblica
    0.86
    ין
    0.84
    ה
    0.84
    ان
    0.82
    an
    0.81
    a
    0.80
    ,
    0.80
    POSITIVE LOGITS
     be
    1.19
     an
    1.06
    ö
    0.80
     at
    0.80
    ెస్
    0.75
     a
    0.73
    awt
    0.73
     ২০
    0.72
     \
    0.71
    ất
    0.69
    Act Density 0.002%

    No Known Activations