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

    Neuronpedia

    INDEX
    Explanations

    centered and using

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    hanahu
    0.39
    ̆
    0.39
    充满
    0.38
     необ
    0.37
    在一个
    0.37
    帶來
    0.37
    떻
    0.37
    elae
    0.37
    бто
    0.37
    țională
    0.36
    POSITIVE LOGITS
    以
    0.90
     centering
    0.79
     centered
    0.79
     using
    0.78
     centred
    0.76
    using
    0.72
    是以
    0.71
     以
    0.70
    centered
    0.69
     menjadikan
    0.69
    Act Density 0.025%

    No Known Activations