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

    Neuronpedia

    INDEX
    Explanations

    increase or decrease in metrics

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    costs
    0.54
     부담
    0.49
     risques
    0.49
     risco
    0.48
     burden
    0.48
     burdens
    0.48
     risiko
    0.47
    リスク
    0.47
    Costs
    0.47
     rischio
    0.46
    POSITIVE LOGITS
     efficiency
    0.63
     productivity
    0.59
    效率
    0.58
     revenue
    0.57
     efficiencies
    0.57
    ประสิทธิภาพ
    0.57
     savings
    0.55
     Efficiency
    0.52
     transparency
    0.52
     Revenue
    0.52
    Act Density 0.073%

    No Known Activations