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

    Neuronpedia

    INDEX
    Explanations

    qualities, locations, and categories

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Когда
    0.49
    ándole
    0.47
     Когда
    0.44
    Quando
    0.44
    Cuando
    0.44
    比如说
    0.44
    querdo
    0.43
    Kalau
    0.43
    Lorsque
    0.41
    Então
    0.41
    POSITIVE LOGITS
     (
    0.51
     vetted
    0.48
     ALL
    0.46
     ASAP
    0.46
    /
    0.46
     within
    0.46
     API
    0.46
     consistently
    0.46
     proactively
    0.45
     included
    0.44
    Act Density 1.048%

    No Known Activations