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

    Neuronpedia

    INDEX
    Explanations

    that

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     dtype
    -0.07
    .relu
    -0.07
    �
    -0.07
     Olympic
    -0.07
    fft
    -0.07
    ㉡
    -0.07
    熻
    -0.06
     tg
    -0.06
     hasher
    -0.06
    	meta
    -0.06
    POSITIVE LOGITS
    桌面
    0.08
     数
    0.07
    ston
    0.07
     ville
    0.07
     Fall
    0.07
    ?↵
    0.06
    ''.
    0.06
    带
    0.06
    ubbles
    0.06
    /";↵
    0.06
    Act Density 0.017%

    No Known Activations