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

    Neuronpedia

    INDEX
    Explanations

    ten, range, quantities

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Fourth
    0.65
     第
    0.62
     pharetra
    0.60
    旲
    0.59
    第
    0.59
    Desde
    0.59
    Weekday
    0.58
     চতুর্থ
    0.57
     tweede
    0.57
    Double
    0.57
    POSITIVE LOGITS
     ten
    1.35
     six
    1.04
     TEN
    1.03
    ten
    1.02
     five
    1.01
     nine
    0.98
    TEN
    0.96
     десять
    0.96
    🔟
    0.92
    five
    0.90
    Act Density 0.375%

    No Known Activations