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

    Neuronpedia

    INDEX
    Explanations

    programming tests

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    套装
    -0.74
    ỗng
    -0.73
    等人
    -0.71
    ровано
    -0.71
    integrity
    -0.69
     destacar
    -0.69
    pubs
    -0.68
     говорить
    -0.68
    Namara
    -0.68
    Accessory
    -0.68
    POSITIVE LOGITS
     tests
    1.50
     test
    1.45
    Test
    1.34
    测试
    1.20
     Test
    1.16
     Tests
    1.13
    テスト
    1.11
     testing
    1.02
    TEST
    1.00
     TESTS
    1.00
    Act Density 0.004%

    No Known Activations