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

    Neuronpedia

    INDEX
    Explanations

    learns or generates systematically

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    しやすい
    0.49
     분명
    0.45
    ਦ
    0.45
     mutlaka
    0.44
     duidelijk
    0.42
     Mudah
    0.42
    やすい
    0.41
     jelas
    0.41
    結局
    0.41
    เสมอ
    0.41
    POSITIVE LOGITS
     literally
    0.89
     selectively
    0.85
     essentially
    0.83
     dynamically
    0.81
     chemically
    0.81
     literalmente
    0.80
     electronically
    0.79
     mathematically
    0.79
     digitally
    0.78
     systematically
    0.78
    Act Density 0.166%

    No Known Activations