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

    Neuronpedia

    INDEX
    Explanations

    information

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     domać
    0.66
     razy
    0.61
     iht
    0.59
     oh
    0.59
    ók
    0.58
     अर्थात
    0.57
     tygod
    0.57
     रिक्वायरमेंट
    0.57
     Oh
    0.56
     hydrochloric
    0.56
    POSITIVE LOGITS
    RB
    0.55
    ()}
    0.54
    LB
    0.54
    ファ
    0.53
    "-"
    0.52
     সমাবে
    0.52
    тена
    0.52
    imiento
    0.51
     ')
    0.51
     матрицы
    0.51
    Act Density 0.000%

    No Known Activations