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

    Neuronpedia

    INDEX
    Explanations

    categorized by level or degree

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    amsung
    0.37
     Samsung
    0.34
     Handsome
    0.34
    beautiful
    0.34
    చిన
    0.33
     samsung
    0.33
    特殊的
    0.33
    WeekDates
    0.32
     తప్ప
    0.32
    cucumber
    0.32
    POSITIVE LOGITS
     complexity
    1.13
     level
    1.08
     degree
    1.08
     intensity
    1.05
     severity
    1.04
    程度
    1.00
    复杂度
    0.98
     sophistication
    0.97
     levels
    0.96
    degree
    0.91
    Act Density 0.122%

    No Known Activations