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

    Neuronpedia

    INDEX
    Explanations

    math problems and foreign languages

    np_max-act-logits · gemini-2.0-flashTriggered by @bettyfeng2003
    New Auto-Interp
    Negative Logits
     Developed
    -0.07
     unicode
    -0.07
    Benefits
    -0.07
    出てくる
    -0.07
    .MustCompile
    -0.07
    合规
    -0.07
    compare
    -0.07
    _TH
    -0.06
    _ser
    -0.06
    -pr
    -0.06
    POSITIVE LOGITS
    asket
    0.07
    יחוד
    0.07
     Gerät
    0.07
    .gamma
    0.07
    /fwlink
    0.07
    ckså
    0.07
    "--
    0.07
    (ids
    0.07
    aj
    0.07
    脑子
    0.07
    Act Density 0.140%

    No Known Activations