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

    Neuronpedia

    INDEX
    Explanations

    reduce or reduced

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Increase
    0.50
     aumento
    0.50
     INCREASE
    0.48
     Expense
    0.47
     increase
    0.46
    เพิ่ม
    0.46
     incremento
    0.45
    בוד
    0.45
    Increase
    0.45
     Composition
    0.44
    POSITIVE LOGITS
    reduced
    1.00
     reduced
    0.98
     reduziert
    0.91
    Reduced
    0.89
     reduz
    0.87
     умень
    0.86
    reduce
    0.84
     Reduced
    0.83
     mengurangi
    0.81
     сни
    0.80
    Act Density 0.033%

    No Known Activations