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

    Neuronpedia

    INDEX
    Explanations

    code execution and output

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     must
    0.90
     less
    0.84
     constant
    0.84
    ’’
    0.83
     n
    0.82
     refers
    0.81
     higher
    0.81
     /
    0.80
    ‘’
    0.79
     seek
    0.79
    POSITIVE LOGITS
    Debug
    1.31
    ---------------
    1.27
    Before
    1.21
    Using
    1.21
    Número
    1.20
    Original
    1.20
    Setelah
    1.18
    Starting
    1.16
    Taking
    1.16
    After
    1.16
    Act Density 0.173%

    No Known Activations