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

    Neuronpedia

    INDEX
    Explanations

    otherwise

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     donc
    0.48
     nên
    0.46
     deci
    0.40
    ロップ
    0.39
    だから
    0.38
     tehát
    0.38
     pertanto
    0.37
     więc
    0.37
     generations
    0.37
     pouv
    0.37
    POSITIVE LOGITS
    这将
    1.01
     Otherwise
    0.91
     इससे
    0.90
    否则
    0.90
    इससे
    0.90
     Dadurch
    0.88
     그러면
    0.84
     अन्यथा
    0.84
     otherwise
    0.83
    otherwise
    0.83
    Act Density 0.037%

    No Known Activations