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

    Neuronpedia

    INDEX
    Explanations

    language identification

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    <0xE7>
    0.85
    <0xE5>
    0.81
    லைப்
    0.76
    ־
    0.73
    ங்களைத்
    0.70
    <unused143>
    0.69
    —
    0.67
     जाएं
    0.66
    <unused2157>
    0.65
    ‒
    0.64
    POSITIVE LOGITS
    또한
    2.46
     occuring
    2.22
     alot
    2.16
     loosing
    2.15
     ।
    2.13
     라는
    2.11
     를
    2.08
     التى
    2.04
     میتوان
    2.03
     কারনে
    2.03
    Act Density 3.909%

    No Known Activations