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

    Neuronpedia

    INDEX
    Explanations

    disruptions after certain words, Douglas Adams

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    '
    0.51
    ↵↵
    0.47
    Type
    0.44
    hyde
    0.43
    byshire
    0.43
    <
    0.42
    ant
    0.41
    ib
    0.39
    äft
    0.39
    eta
    0.39
    POSITIVE LOGITS
     système
    0.55
     consecuencia
    0.54
     ಮೆ
    0.54
    캘
    0.53
     يق
    0.53
     nome
    0.52
     забезпе
    0.52
     சுத்தம்
    0.52
     பிரச்
    0.51
     tabela
    0.51
    Act Density 0.002%

    No Known Activations