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

    Neuronpedia

    INDEX
    Explanations

    varied writing styles

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    cedures
    -0.07
     feeds
    -0.07
     backups
    -0.06
     Algorithm
    -0.06
     Pills
    -0.06
     même
    -0.06
     grinder
    -0.06
    normally
    -0.06
     pump
    -0.06
     nowhere
    -0.06
    POSITIVE LOGITS
    ::__
    0.07
    :"+
    0.07
     اسر
    0.06
     تف
    0.06
    Startup
    0.06
    trajectory
    0.06
    buie
    0.06
     小
    0.06
    /dc
    0.06
    ленных
    0.06
    Act Density 0.175%

    No Known Activations