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

    Neuronpedia

    INDEX
    Explanations

    commit message titles

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    .
    0.77
    sided
    0.69
    references
    0.68
     terdapat
    0.67
    den
    0.67
     happening
    0.67
    ,
    0.65
    owned
    0.64
    surprisingly
    0.64
     where
    0.64
    POSITIVE LOGITS
    添加
    0.79
     添加
    0.74
     kasutada
    0.71
     revisar
    0.71
     добавить
    0.71
     trzech
    0.71
     Нужно
    0.70
     만들
    0.70
    请求
    0.69
    柟
    0.69
    Act Density 0.000%

    No Known Activations