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

    Neuronpedia

    INDEX
    Explanations

    parenthetical phrases

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    F
    0.63
    인
    0.62
     대한
    0.59
     ihe
    0.57
     noma
    0.57
    J
    0.57
    Ш
    0.57
    L
    0.57
    Ц
    0.57
     Aan
    0.56
    POSITIVE LOGITS
    на
    0.76
    em
    0.64
    ας
    0.62
    ان
    0.60
    ți
    0.59
    at
    0.58
    vole
    0.57
    .'”
    0.57
    age
    0.57
    is
    0.56
    Act Density 1.347%

    No Known Activations