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

    Neuronpedia

    INDEX
    Explanations

    Left curly bracket

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    进入了
    -0.08
    .sample
    -0.07
    创造了
    -0.07
     Nay
    -0.07
    pciones
    -0.07
    我们要
    -0.07
     jealous
    -0.07
    וו
    -0.07
     reign
    -0.07
    .results
    -0.06
    POSITIVE LOGITS
    搪
    0.08
     Jeb
    0.07
    (NUM
    0.07
    貧
    0.07
    安东
    0.07
     postpon
    0.07
    ท
    0.07
    opcode
    0.06
     Fuj
    0.06
    oric
    0.06
    Act Density 0.003%

    No Known Activations