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

    Neuronpedia

    INDEX
    Explanations

    foreign characters

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     fara
    0.59
     sunt
    0.56
     este
    0.55
     među
    0.53
     din
    0.53
     fiind
    0.52
    теле
    0.52
     הן
    0.52
     mentre
    0.51
     trebuie
    0.51
    POSITIVE LOGITS
     sco
    0.47
     गर्म
    0.46
    萛
    0.46
     rid
    0.46
    読み
    0.46
     пере
    0.45
    आत्म
    0.45
    確認
    0.44
    罢
    0.44
     ሥ
    0.44
    Act Density 0.001%

    No Known Activations