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

    Neuronpedia

    INDEX
    Explanations

    store data

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Lordships
    1.88
    шой
    1.81
    ﻄ
    1.66
    arı
    1.64
    知道
    1.63
     fric
    1.59
    ை
    1.55
    CRY
    1.54
    MATCH
    1.53
    േന
    1.51
    POSITIVE LOGITS
    ق
    2.75
    ع
    2.06
    ется
    2.03
    いた
    1.95
     Taille
    1.95
    على
    1.89
    علي
    1.88
    ため
    1.84
    illant
    1.83
    ようになりました
    1.82
    Act Density 0.165%

    No Known Activations