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

    Neuronpedia

    INDEX
    Explanations

    builds upon

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     wobei
    0.38
    ’.
    0.38
     각종
    0.37
     quedando
    0.34
     disertai
    0.34
    ™.
    0.33
     během
    0.33
    %.
    0.32
    然后在
    0.32
    参数向量
    0.32
    POSITIVE LOGITS
     solely
    0.86
     primarily
    0.83
     mainly
    0.79
     squarely
    0.78
     directly
    0.77
    primarily
    0.75
     largely
    0.72
     principally
    0.71
     specifically
    0.70
     entirely
    0.70
    Act Density 0.742%

    No Known Activations