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

    Neuronpedia

    INDEX
    Explanations

    belongs in, currently running

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Used
    0.89
     used
    0.85
     Used
    0.85
    used
    0.79
     digunakan
    0.75
     USED
    0.74
     !==
    0.73
     mistaken
    0.71
    用来
    0.71
    用於
    0.67
    POSITIVE LOGITS
     thuộc
    1.05
     belongs
    0.90
    屬於
    0.85
     belong
    0.81
    belongs
    0.81
    属于
    0.78
    categoria
    0.77
    belong
    0.77
     Under
    0.72
     erled
    0.71
    Act Density 0.807%

    No Known Activations