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

    Neuronpedia

    INDEX
    Explanations

    main

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    None
    0.94
     máximo
    0.89
     正規
    0.89
     mínimo
    0.87
    مة
    0.86
    м
    0.86
     visok
    0.84
     минима
    0.83
     поводу
    0.82
    比
    0.82
    POSITIVE LOGITS
    テナンス
    2.07
    stays
    1.95
    tained
    1.89
    AxisAlignment
    1.67
    AxisSize
    1.60
    tains
    1.47
    ifest
    1.42
    taining
    1.42
    tenir
    1.40
    stay
    1.37
    Act Density 0.077%

    No Known Activations