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

    Neuronpedia

    INDEX
    Explanations

    performs or behaves

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    щи
    0.38
     জড়িত
    0.37
    ective
    0.37
     versuchen
    0.35
     toxic
    0.35
    anians
    0.35
     시작하겠습니다
    0.35
    してください
    0.34
    ectl
    0.34
     existence
    0.34
    POSITIVE LOGITS
     lend
    0.92
     lends
    0.91
     performs
    0.89
     behave
    0.83
     perform
    0.82
     behaves
    0.82
     fares
    0.81
     fared
    0.80
     Performs
    0.79
     fare
    0.77
    Act Density 0.098%

    No Known Activations