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

    Neuronpedia

    INDEX
    Explanations

    list entries and starting points

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ↵
    1.75
    '
    1.65
    ע
    1.49
    ാ
    1.45
     on
    1.45
    ان
    1.42
    ক
    1.41
    ية
    1.36
    .
    1.34
    িন
    1.33
    POSITIVE LOGITS
    entry
    1.29
    the
    1.05
    test
    1.05
    and
    1.02
    line
    1.00
    list
    1.00
    docs
    0.95
    ris
    0.95
     and
    0.94
    does
    0.93
    Act Density 0.009%

    No Known Activations