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

    Neuronpedia

    INDEX
    Explanations

    replace bracketed information

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    できる
    0.40
     言っ
    0.39
     вычисли
    0.39
    也會
    0.38
     ALSO
    0.38
     ALWAYS
    0.38
    मुळे
    0.37
    膺
    0.36
    ージ
    0.35
    ामुळे
    0.35
    POSITIVE LOGITS
     Variables
    0.42
    لە
    0.42
     informasi
    0.41
    变量
    0.41
    रोना
    0.40
     vào
    0.40
    🚐
    0.39
    variables
    0.39
    Those
    0.39
    Variables
    0.39
    Act Density 0.001%

    No Known Activations