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

    Neuronpedia

    INDEX
    Explanations

    'f' character or starts of acronyms

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     of
    1.40
    field
    1.12
    y
    1.08
    ви
    1.03
    ert
    1.02
    I
    0.99
    ff
    0.98
    ur
    0.98
    ü
    0.96
    on
    0.94
    POSITIVE LOGITS
    ud
    1.21
    ج
    1.20
    ק
    1.19
    ’
    1.17
    কে
    1.13
    很
    1.13
    ب
    1.03
    ↵↵
    0.99
     представ
    0.98
    ോട്ടോ
    0.97
    Act Density 0.702%

    No Known Activations