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

    Neuronpedia

    INDEX
    Explanations

    how scientists

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    幅度
    0.51
     المناطق
    0.49
    anea
    0.46
     നിരവധി
    0.46
    blos
    0.46
    按摩
    0.46
    一提
    0.45
    नं
    0.45
    gaxModule
    0.43
     மேற்கொள்ள
    0.43
    POSITIVE LOGITS
    ی
    0.56
    at
    0.54
    in
    0.50
    ോദ
    0.48
    s
    0.44
     died
    0.43
    ...")
    0.42
    ”).
    0.42
    ic
    0.42
    .")
    0.42
    Act Density 0.002%

    No Known Activations