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

    Neuronpedia

    INDEX
    Explanations

    `CharacterBody2D` or `Data` sources

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    то
    1.09
    ため
    1.08
    ко
    1.01
    加え
    0.94
    مان
    0.94
    ше
    0.90
    ι
    0.89
    хо
    0.88
    僻
    0.88
    мон
    0.88
    POSITIVE LOGITS
    as
    1.16
    enna
    1.09
    ord
    1.09
    ice
    0.99
    tsp
    0.94
    ab
    0.92
    s
    0.91
    asd
    0.91
    ts
    0.90
     gaps
    0.90
    Act Density 0.003%

    No Known Activations