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

    Neuronpedia

    INDEX
    Explanations

    variable names and media types

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     একটা
    0.99
     कुछ
    0.79
     거의
    0.79
     شو
    0.79
     পরিষ্কার
    0.78
     കുറ
    0.78
    كل
    0.78
     것도
    0.77
    것도
    0.76
     짧
    0.75
    POSITIVE LOGITS
     H
    0.86
    《
    0.85
     An
    0.85
    টি
    0.85
     trillion
    0.84
     Ar
    0.84
     extraordinaire
    0.82
     belonging
    0.81
     Innovation
    0.80
     X
    0.79
    Act Density 0.369%

    No Known Activations