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

    Neuronpedia

    INDEX
    Explanations

    beach or screen

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ات
    3.88
    s
    3.35
    ים
    3.23
    sit
    2.83
    en
    2.80
    ের
    2.77
    います
    2.68
    sr
    2.67
    sion
    2.64
    tale
    2.60
    POSITIVE LOGITS
    м
    3.84
    م
    3.37
    го
    2.70
    부터
    2.51
    тою
    2.40
    ので
    2.39
    서
    2.35
    ं
    2.33
    ق
    2.32
    なかなか
    2.32
    Act Density 0.118%

    No Known Activations