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

    Neuronpedia

    INDEX
    Explanations

    what a superlative is

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    <0x9B>
    0.52
    Тер
    0.48
    Ве
    0.47
    кает
    0.47
    闶
    0.46
    ύν
    0.44
    மாண
    0.44
    ЕН
    0.43
    Free
    0.43
    ási
    0.43
    POSITIVE LOGITS
     incidente
    0.50
     discoveries
    0.49
     如
    0.47
     của
    0.46
     instrumento
    0.45
     anecd
    0.44
     graines
    0.43
     descubrimiento
    0.43
     intact
    0.43
     exemplo
    0.43
    Act Density 0.004%

    No Known Activations