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

    Neuronpedia

    INDEX
    Explanations

    Join or structure relation

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Callback
    0.45
    loé
    0.40
    丽
    0.40
    Jersey
    0.40
    Dio
    0.39
    Meta
    0.39
     डाइऑक्साइड
    0.39
    وید
    0.38
    滩
    0.38
     irány
    0.38
    POSITIVE LOGITS
     Join
    0.59
     join
    0.55
    Join
    0.53
     point
    0.52
     returning
    0.50
     proceeding
    0.49
    join
    0.48
     Joint
    0.48
     Point
    0.47
     public
    0.47
    Act Density 0.002%

    No Known Activations