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

    Neuronpedia

    INDEX
    Explanations

    eng

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     once
    -0.26
    deaux
    -0.25
     mer
    -0.24
    çijķ
    -0.24
    ushima
    -0.24
    åĪ¶
    -0.23
     mental
    -0.23
    è¯Ŀé¢ĺ
    -0.23
    uParam
    -0.23
    para
    -0.23
    POSITIVE LOGITS
    owing
    0.33
    icularly
    0.27
    ictions
    0.27
    ordering
    0.25
     BLOCK
    0.24
    _Blue
    0.24
     lul
    0.24
    _Matrix
    0.24
    åħ«å¹´
    0.24
    åĻ©
    0.23
    Act Density 0.038%

    No Known Activations