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

    Neuronpedia

    INDEX
    Explanations

    for

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     is
    -0.07
     Dünya
    -0.07
     오
    -0.07
    amerate
    -0.07
     are
    -0.06
     sing
    -0.06
     isp
    -0.06
    .We
    -0.06
     predicted
    -0.06
     ais
    -0.06
    POSITIVE LOGITS
     参考
    0.07
    _SUR
    0.07
    章
    0.06
    번째
    0.06
    Rule
    0.06
    [vi
    0.06
    Subject
    0.06
     thuộc
    0.06
     đột
    0.06
     Barton
    0.06
    Act Density 0.057%

    No Known Activations