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

    Neuronpedia

    INDEX
    Explanations

    plain descriptions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     подобных
    0.85
     подобные
    0.83
     другими
    0.80
     другим
    0.73
    жению
    0.72
     другого
    0.67
    ElementException
    0.67
     toch
    0.66
     некоторым
    0.65
     подоб
    0.65
    POSITIVE LOGITS
     simple
    1.10
     straightforward
    1.02
     literal
    1.01
     trivial
    0.98
     formal
    0.97
     عادي
    0.97
     mundane
    0.94
     outright
    0.92
     ranging
    0.91
    ranging
    0.91
    Act Density 1.032%

    No Known Activations