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

    Neuronpedia

    INDEX
    Explanations

    hyphens/delimiters

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     news
    -0.07
     Agents
    -0.07
    REDENTIAL
    -0.06
    _CE
    -0.06
    前の
    -0.06
    ानम
    -0.06
    етом
    -0.06
    ρωση
    -0.06
    231
    -0.06
    idente
    -0.06
    POSITIVE LOGITS
    CORE
    0.07
     ADD
    0.07
    ，更
    0.07
     pancre
    0.06
     такой
    0.06
    кры
    0.06
    �
    0.06
    ーナ
    0.06
    Lazy
    0.06
    =UTF
    0.06
    Act Density 0.127%

    No Known Activations