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

    Neuronpedia

    INDEX
    Explanations

    acknowledge user intent

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     what
    0.99
    what
    0.95
     என்ன
    0.84
     WHAT
    0.84
    What
    0.80
    WHAT
    0.79
     What
    0.78
     Fears
    0.78
     fears
    0.77
     frightening
    0.77
    POSITIVE LOGITS
     regular
    0.56
     еле
    0.54
     previstas
    0.53
    Regular
    0.51
    今年も
    0.50
     элементы
    0.50
    annies
    0.50
     Regular
    0.49
     exercitation
    0.48
     belle
    0.48
    Act Density 0.107%

    No Known Activations