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

    Neuronpedia

    INDEX
    Explanations

    unpacking explanations

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ב
    1.49
    υ
    1.41
    ateboard
    1.34
    끔
    1.29
    க்கு
    1.27
    에
    1.25
    ä
    1.23
    claim
    1.23
    的方式
    1.19
    łą
    1.18
    POSITIVE LOGITS
    ﺭ
    1.62
    एर
    1.51
     jewels
    1.48
     vacancies
    1.47
    wolves
    1.45
    VER
    1.39
    нд
    1.37
     соответствии
    1.36
     sorrows
    1.36
    ராம
    1.34
    Act Density 0.001%

    No Known Activations