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

    Neuronpedia

    INDEX
    Explanations

    python testing tools

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    飫
    0.42
    calculation
    0.41
    Smoothing
    0.40
    smoothing
    0.39
    ímav
    0.39
     অনিবার
    0.38
     calculation
    0.38
    計算
    0.38
    FDMA
    0.38
    calculations
    0.37
    POSITIVE LOGITS
     test
    1.99
     Test
    1.99
     tests
    1.99
     Tests
    1.99
     Testing
    1.90
     testing
    1.89
    测试
    1.88
    Tests
    1.85
    Test
    1.84
     TESTS
    1.82
    Act Density 0.013%

    No Known Activations