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

    Neuronpedia

    INDEX
    Explanations

    core programming logic

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    تين
    1.11
    تان
    1.10
     (
    1.05
    ted
    1.04
    td
    0.99
    تری
    0.98
    تنا
    0.98
    t
    0.97
    تها
    0.95
    ts
    0.95
    POSITIVE LOGITS
    й
    1.19
    ம்
    1.09
    いる
    1.02
    아
    0.98
    ר
    0.96
    ك
    0.93
    나
    0.92
    на
    0.91
    am
    0.91
    ক
    0.91
    Act Density 0.014%

    No Known Activations