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

    Neuronpedia

    INDEX
    Explanations

    saving code to run

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    .;
    1.42
     .;
    1.37
    .?
    1.30
    .:
    1.29
    .,
    1.26
     Specifically
    1.24
    。“
    1.23
    .,"
    1.22
     (“
    1.19
     blah
    1.19
    POSITIVE LOGITS
    ין
    0.72
    е
    0.71
    વ
    0.71
    ف
    0.70
    ب
    0.68
    Neck
    0.68
    read
    0.68
    MAN
    0.67
    lfloor
    0.66
    populate
    0.65
    Act Density 0.049%

    No Known Activations