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

    Neuronpedia

    INDEX
    Explanations

    defining characteristics or core components

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    いたり
    0.97
     atau
    0.91
     можли
    0.88
     возмо
    0.88
     nebo
    0.87
     tertentu
    0.84
    possible
    0.84
    Possible
    0.83
    ったり
    0.83
    something
    0.82
    POSITIVE LOGITS
    用于
    1.01
     discussed
    1.01
    用於
    1.01
     differentiating
    0.99
     defining
    0.99
     explaining
    0.97
     determining
    0.97
     mentioned
    0.96
     shaping
    0.95
     characterizing
    0.94
    Act Density 0.167%

    No Known Activations