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

    Neuronpedia

    INDEX
    Explanations

    support for capabilities

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    <0xE3>
    0.46
     än
    0.45
     و
    0.44
    ية
    0.44
    had
    0.44
    （
    0.44
     که
    0.42
    也就是
    0.42
    ioid
    0.40
    ellä
    0.39
    POSITIVE LOGITS
     UIView
    0.53
     darstellen
    0.50
     кеңселер
    0.49
     VIEW
    0.48
     кеңсе
    0.48
     szolg
    0.48
     감사합니다
    0.48
     restitu
    0.47
     አገልግሎ
    0.46
     représentant
    0.45
    Act Density 0.001%

    No Known Activations