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

    Neuronpedia

    INDEX
    Explanations

    2023 turning into 2024

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     five
    0.85
     Five
    0.80
     fifth
    0.78
     पांच
    0.76
     eight
    0.74
    five
    0.73
     fives
    0.72
     six
    0.72
     Sixty
    0.71
    0.70
    POSITIVE LOGITS
    cInfo
    0.73
     طلع
    0.69
    IDENCE
    0.66
    াস্ত
    0.65
    該
    0.64
    <unused1689>
    0.63
    ocrat
    0.63
    쟁
    0.62
    𝘂
    0.62
    <unused604>
    0.62
    Act Density 0.152%

    No Known Activations