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

    Neuronpedia

    INDEX
    Explanations

    more, any, or, half

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ricted
    0.98
    ك
    0.96
    tered
    0.93
    एल
    0.82
    ンの
    0.81
     가진
    0.80
    ა
    0.80
    ಂಡ್
    0.80
    س
    0.80
    tering
    0.79
    POSITIVE LOGITS
     orchards
    0.85
     eben
    0.85
    ના
    0.84
     améric
    0.84
    습니다
    0.84
    }{
    0.83
     importantes
    0.82
     batas
    0.81
     casks
    0.80
     ανθρώ
    0.79
    Act Density 0.099%

    No Known Activations