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

    Neuronpedia

    INDEX
    Explanations

    involved in

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    DISABLE
    0.62
     Wasn
    0.61
    𝗳
    0.60
     हमारे
    0.60
     dotycz
    0.59
     îi
    0.59
     dotyczące
    0.58
    ರ್‌
    0.57
    যানী
    0.57
    Problem
    0.57
    POSITIVE LOGITS
    at
    1.56
    ت
    1.38
    in
    1.29
    m
    1.23
    т
    1.13
    r
    1.12
    त
    1.10
    t
    1.05
    w
    1.05
    n
    1.03
    Act Density 0.102%

    No Known Activations