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

    Neuronpedia

    INDEX
    Explanations

    responsiveness and scaling

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     contaminated
    0.83
     ইলেকট্র
    0.79
    munition
    0.78
     ammunition
    0.77
     polluted
    0.75
    絶縁
    0.73
     इलेक्ट्र
    0.73
     unarmed
    0.73
     impurities
    0.72
    Immun
    0.72
    POSITIVE LOGITS
     adjust
    1.24
     readjust
    1.22
     adjusting
    1.17
     resizing
    1.11
    adjust
    1.09
    Adjust
    1.08
     adjusts
    1.08
    调整
    1.07
     adjustments
    1.07
     Adjust
    1.07
    Act Density 0.243%

    No Known Activations