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

    Neuronpedia

    INDEX
    Explanations

    given

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    🇹
    0.73
    。
    0.72
    "],
    0.67
    ли
    0.66
    ।
    0.65
    "
    0.63
    right
    0.61
    actual
    0.61
     outp
    0.61
    shed
    0.61
    POSITIVE LOGITS
     diferencia
    0.91
    et
    0.91
    ógica
    0.89
     differenza
    0.88
    で
    0.87
    ņa
    0.86
     manera
    0.84
     Dünya
    0.84
     Imágenes
    0.84
     están
    0.84
    Act Density 0.000%

    No Known Activations