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

    Neuronpedia

    INDEX
    Explanations

    indentation and formatting

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     can
    1.21
     on
    1.16
    f
    1.07
    ร
    1.02
     it
    0.96
    ی
    0.96
    い
    0.92
    ים
    0.89
    یت
    0.86
    บ
    0.79
    POSITIVE LOGITS
    el
    0.80
    n
    0.80
     MAKEINTRESOURCE
    0.73
     ιδια
    0.73
    кола
    0.72
    of
    0.72
     доби
    0.71
    EN
    0.71
    ви
    0.71
     вій
    0.71
    Act Density 0.002%

    No Known Activations