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

    Neuronpedia

    INDEX
    Explanations

    specific references or existing items

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    urger
    0.79
     ratio
    0.75
     ratios
    0.70
     poor
    0.70
     software
    0.69
     worst
    0.68
     timing
    0.67
     gef
    0.67
     low
    0.66
     rate
    0.66
    POSITIVE LOGITS
     Encycl
    1.16
     encycl
    1.10
    Geography
    1.08
     Realm
    1.04
    Expanding
    1.01
     Expanding
    1.01
    context
    1.00
    Knowledge
    1.00
    Globalization
    1.00
    Relevant
    0.99
    Act Density 0.950%

    No Known Activations