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

    Neuronpedia

    INDEX
    Explanations

    how questions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ס
    2.14
    ING
    2.05
    ABLE
    2.00
    ч
    1.89
     comunque
    1.83
    HA
    1.81
    봅시다
    1.79
    ’
    1.78
    ется
    1.77
    ভাবে
    1.77
    POSITIVE LOGITS
    soever
    2.77
    های
    2.16
     וכ
    1.92
    يت
    1.88
    รู้สึก
    1.81
    zhihu
    1.77
    看待
    1.76
    েও
    1.67
    िति
    1.66
    infodisc
    1.61
    Act Density 0.244%

    No Known Activations