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

    Neuronpedia

    INDEX
    Explanations

    is connected to a state

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    oises
    0.41
    Wherever
    0.38
    ন
    0.38
    はなく
    0.37
    ोष
    0.36
     невозможно
    0.35
    熵
    0.35
    Ỳ
    0.35
    ంశ
    0.34
    Фі
    0.34
    POSITIVE LOGITS
     valid
    0.63
    valid
    0.62
     alive
    0.62
     eligible
    0.61
     locked
    0.60
     connected
    0.57
     feasible
    0.57
     gült
    0.56
     ready
    0.55
     docked
    0.55
    Act Density 0.019%

    No Known Activations