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

    Neuronpedia

    INDEX
    Explanations

    how it works

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     actually
    1.02
     véritable
    0.97
     réellement
    0.96
     nonché
    0.96
    乃至
    0.95
     살펴보도록
    0.94
     realmente
    0.93
     genuinely
    0.90
    或其他
    0.88
    <end_of_turn>
    0.87
    POSITIVE LOGITS
    :
    0.79
    =
    0.76
    ->
    0.70
    :**
    0.69
    ：
    0.69
          
    0.68
     करें
    0.67
    Donate
    0.67
            
    0.67
     મુજબ
    0.66
    Act Density 0.162%

    No Known Activations