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

    Neuronpedia

    INDEX
    Explanations

    duration and effort

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ಾ
    0.70
     vivent
    0.68
    ޓ
    0.66
     affez
    0.65
     abbiamo
    0.64
     militares
    0.64
    त
    0.64
     nourrir
    0.64
     auront
    0.63
    т
    0.63
    POSITIVE LOGITS
    时间和
    0.85
     времени
    0.79
     시간을
    0.79
    時間の
    0.75
     Periods
    0.75
    ⏱
    0.74
    時間が
    0.73
     সময়
    0.72
    ⏰
    0.72
    쩝
    0.72
    Act Density 0.859%

    No Known Activations