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

    Neuronpedia

    INDEX
    Explanations

    Programming code

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     Ник
    -0.07
     goose
    -0.07
    imap
    -0.07
     blanc
    -0.06
     Пот
    -0.06
     бед
    -0.06
     Ros
    -0.06
     yayın
    -0.06
    анная
    -0.06
     commonly
    -0.06
    POSITIVE LOGITS
     fila
    0.07
     mouth
    0.06
    илась
    0.06
    PW
    0.06
    uid
    0.06
    �
    0.06
    elage
    0.06
    一卷
    0.06
    ！」
    0.05
     prez
    0.05
    Act Density 0.032%

    No Known Activations