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

    Neuronpedia

    INDEX
    Explanations

    asking "what" or "where"

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     whether
    0.65
    whether
    0.62
    Whether
    0.58
     Whether
    0.57
    Was
    0.53
     Was
    0.53
     WHETHER
    0.52
     apakah
    0.50
    Did
    0.49
    did
    0.47
    POSITIVE LOGITS
    可以使用
    0.53
     चुनौ
    0.48
     می‌توان
    0.45
     สามารถ
    0.44
     accompl
    0.43
     کیسے
    0.42
    具备
    0.42
     পাকসেন
    0.41
    豈
    0.41
     feasible
    0.40
    Act Density 0.034%

    No Known Activations