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

    Neuronpedia

    INDEX
    Explanations

    code

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     jc
    -0.06
    新型冠状
    -0.06
     얘기
    -0.06
    (tv
    -0.06
    stage
    -0.06
     llevar
    -0.06
     Donne
    -0.06
    긱
    -0.06
    asily
    -0.06
    /y
    -0.06
    POSITIVE LOGITS
    插入
    0.08
    反射
    0.08
    slice
    0.07
     usually
    0.07
    逼近
    0.07
     
↵
    0.07
    每一位
    0.07
    0.07
    遍及
    0.07
     affidavit
    0.07
    Act Density 0.006%

    No Known Activations