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

    Neuronpedia

    INDEX
    Explanations

    highlighting specific conditions or examples

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    :
    1.15
     a
    1.02
    ;
    1.00
     as
    0.98
    O
    0.95
    ing
    0.90
     of
    0.88
    <h3>
    0.77
    :>
    0.76
    E
    0.76
    POSITIVE LOGITS
    on
    0.95
    на
    0.84
     besonders
    0.82
    p
    0.82
     Particularly
    0.77
    s
    0.76
    in
    0.75
    ात
    0.73
    де
    0.72
    h
    0.72
    Act Density 0.067%

    No Known Activations