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

    Neuronpedia

    INDEX
    Explanations

    commands preceded by structure

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    allenges
    0.78
     halfCanvas
    0.75
    დესაც
    0.75
    大量的
    0.72
    ousands
    0.72
    戰爭
    0.72
    !!!");
    0.68
    高品質
    0.68
    敵人
    0.66
     izgrad
    0.66
    POSITIVE LOGITS
     or
    1.02
     또는
    0.98
     eller
    0.97
     atau
    0.91
     but
    0.88
     หรือ
    0.86
     without
    0.84
     বা
    0.82
     või
    0.81
     или
    0.81
    Act Density 0.000%

    No Known Activations