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

    Neuronpedia

    INDEX
    Explanations

    discrete fourier transform

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    watchlist
    0.41
    ppins
    0.38
    dynam
    0.37
    jans
    0.37
    seems
    0.37
    这也是
    0.37
    triple
    0.36
    tans
    0.36
    !==
    0.36
    Interview
    0.36
    POSITIVE LOGITS
     때
    0.47
     அதை
    0.43
     可以
    0.42
    陘
    0.42
    但是
    0.40
     createTask
    0.40
    疋
    0.40
    เพื่อ
    0.38
     _
    0.37
     Pounds
    0.37
    Act Density 0.001%

    No Known Activations