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

    Neuronpedia

    INDEX
    Explanations

    main and main character

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     المره
    0.92
    ্ম্ম
    0.92
     正規
    0.90
    مة
    0.90
     últimos
    0.89
     поводу
    0.88
     mínimo
    0.88
    None
    0.86
    比
    0.85
     множество
    0.85
    POSITIVE LOGITS
    テナンス
    2.06
    stays
    2.03
    tained
    1.92
    AxisAlignment
    1.75
    AxisSize
    1.58
    tains
    1.50
    stay
    1.46
    tenir
    1.44
    taining
    1.43
    ifest
    1.39
    Act Density 0.247%

    No Known Activations