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

    Neuronpedia

    INDEX
    Explanations

    example usage

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     drafted
    0.88
     reviewed
    0.85
     ஏற்ற
    0.80
     lò
    0.80
    गे
    0.79
     Christendom
    0.79
    reviewed
    0.78
    Ду
    0.77
    પણે
    0.77
    Quiet
    0.76
    POSITIVE LOGITS
     demonstrations
    1.05
     holdings
    1.04
     usage
    1.00
     trajectories
    0.97
     Usage
    0.97
     usages
    0.94
    usages
    0.93
    __:
    0.90
     demonstration
    0.89
    用法
    0.89
    Act Density 0.067%

    No Known Activations