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

    Neuronpedia

    INDEX
    Explanations

    LMS, acronyms ending in PS

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    t
    3.27
    to
    3.23
    tions
    3.23
    tion
    3.03
    the
    2.97
    table
    2.89
    ti
    2.83
    tr
    2.75
    times
    2.67
    takes
    2.64
    POSITIVE LOGITS
    ن
    2.80
    itating
    2.30
    ה
    2.25
    س
    2.23
    ם
    2.19
    ̌
    2.16
    ول
    2.13
    ح
    2.13
    بي
    2.11
    ك
    2.03
    Act Density 0.689%

    No Known Activations