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

    Neuronpedia

    INDEX
    Explanations

    emphasized words

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    V
    0.55
    ة
    0.53
    H
    0.52
    도
    0.52
    ess
    0.47
    కు
    0.46
    D
    0.46
    етка
    0.46
    ’
    0.45
    ет
    0.45
    POSITIVE LOGITS
     on
    0.68
     of
    0.64
     און
    0.51
     ی
    0.51
     な
    0.50
     at
    0.49
     Fonbet
    0.49
     کے
    0.48
    zhou
    0.48
     й
    0.48
    Act Density 0.703%

    No Known Activations