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

    Neuronpedia

    INDEX
    Explanations

    abstract concepts and states

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    های
    0.64
     an
    0.63
     foolproof
    0.62
    时间和
    0.61
    ционный
    0.60
     a
    0.57
    +)$
    0.55
     juicy
    0.55
    َب
    0.55
    lbrakk
    0.55
    POSITIVE LOGITS
     .
    0.86
    which
    0.84
    as
    0.78
    。
    0.77
     which
    0.74
    ．
    0.73
     They
    0.72
    ™.
    0.72
    These
    0.71
    .*
    0.71
    Act Density 0.008%

    No Known Activations