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

    Neuronpedia

    INDEX
    Explanations

    mixed categories

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    に興味
    -1.07
     to
    -0.91
     telefono
    -0.90
    звичай
    -0.89
     preferências
    -0.89
     definitiva
    -0.89
     düz
    -0.86
    瘜
    -0.82
     материалов
    -0.81
    Mã
    -0.81
    POSITIVE LOGITS
    mixing
    1.08
     Mixed
    0.94
     hemel
    0.93
    雑
    0.92
    стреча
    0.90
     filmp
    0.87
    젝
    0.87
    mixed
    0.86
    艿
    0.85
     accessoires
    0.85
    Act Density 0.022%

    No Known Activations