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

    Neuronpedia

    INDEX
    Explanations

    percentage increase

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    wich
    0.41
     немно
    0.40
     점점
    0.39
    嬗
    0.38
    paces
    0.38
     faiblement
    0.37
     слегка
    0.36
     faintly
    0.35
     Kellogg
    0.35
    ล็อต
    0.35
    POSITIVE LOGITS
     doubling
    1.10
     percentage
    1.07
     increase
    1.04
    increase
    1.02
     दोगु
    1.01
     doubled
    0.96
     yüzde
    0.96
     %
    0.94
    Increase
    0.93
     percentuale
    0.93
    Act Density 0.035%

    No Known Activations