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

    Neuronpedia

    INDEX
    Explanations

    alongside others

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     is
    1.20
    2
    0.94
    &&
    0.80
    ната
    0.75
    лены
    0.75
     But
    0.74
    𝟐
    0.74
    ды
    0.73
    larını
    0.73
     тема
    0.73
    POSITIVE LOGITS
    '
    1.65
    (
    1.27
    पी
    1.16
    ing
    1.16
    y
    1.13
    -
    1.13
    ج
    1.13
    માં
    1.12
    i
    1.11
    ou
    1.09
    Act Density 0.002%

    No Known Activations