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

    Neuronpedia

    INDEX
    Explanations

    numerical values and calculations

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Goals
    0.94
    itäten
    0.86
     pptn
    0.83
     늘
    0.80
    getInfo
    0.79
    的情况
    0.78
    searchResults
    0.77
     Curves
    0.77
     Inputs
    0.77
     Versus
    0.77
    POSITIVE LOGITS
     value
    1.37
     sum
    1.12
    value
    1.09
     значение
    1.07
     valeur
    1.01
     difference
    0.98
     значення
    0.93
     approximate
    0.92
     absolute
    0.92
     waarde
    0.90
    Act Density 0.742%

    No Known Activations