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

    Neuronpedia

    INDEX
    Explanations

    xml

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    (Book
    -0.08
    拜登
    -0.07
     teste
    -0.07
     nisi
    -0.07
    两天
    -0.06
     sede
    -0.06
    为抓手
    -0.06
    焦
    -0.06
    ONO
    -0.06
     <<=
    -0.06
    POSITIVE LOGITS
     Influence
    0.07
    _Y
    0.07
     eliminates
    0.07
     Explos
    0.07
    br
    0.07
    sharp
    0.06
    lock
    0.06
    绵
    0.06
     serialized
    0.06
    极少
    0.06
    Act Density 0.000%

    No Known Activations