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

    Neuronpedia

    INDEX
    Explanations

    what does it want / need / do

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     কাকে
    0.53
     Importance
    0.45
     Have
    0.45
     如何
    0.40
     كيفية
    0.39
    Have
    0.37
    டக்க
    0.36
     importance
    0.36
     फु
    0.36
    Можно
    0.36
    POSITIVE LOGITS
     करेगा
    0.95
     đều
    0.83
     хочет
    0.83
     сможет
    0.82
    都能
    0.81
     любит
    0.80
     رکھتا
    0.80
     रखता
    0.79
     miał
    0.75
     κάνει
    0.75
    Act Density 0.015%

    No Known Activations