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

    Neuronpedia

    INDEX
    Explanations

    with permission

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    There
    1.00
    They
    0.97
     There
    0.91
    It
    0.88
     They
    0.83
     بشكل
    0.82
     они
    0.79
     برای
    0.79
     ketika
    0.79
    下旬
    0.78
    POSITIVE LOGITS
     regards
    2.36
     regard
    2.24
     respect
    1.79
    standing
    1.67
     impunity
    1.56
     whom
    1.54
    drawn
    1.50
     respecto
    1.42
    inthe
    1.37
    respect
    1.31
    Act Density 0.375%

    No Known Activations