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

    Neuronpedia

    INDEX
    Explanations

    end of sentences doing

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    这也是
    0.99
    这是
    0.97
    เหล่านี้
    0.93
    這是
    0.92
     त्यासाठी
    0.91
    这就是
    0.87
     these
    0.86
     abbrevi
    0.86
    这是一个
    0.86
     cannot
    0.85
    POSITIVE LOGITS
    Doing
    0.90
     Doing
    0.85
    way
    0.73
    ]()
    0.72
     Zweiten
    0.72
    ząd
    0.72
    doing
    0.70
     doing
    0.70
    കാരം
    0.70
     그러면
    0.69
    Act Density 0.113%

    No Known Activations