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

    Neuronpedia

    INDEX
    Explanations

    gentleman, sailor, limerick

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     berbasis
    0.57
     maximale
    0.56
     globale
    0.54
     langfrist
    0.54
     DeFi
    0.53
     ReLU
    0.52
     weltweit
    0.52
     utilizzo
    0.52
     emoz
    0.52
     Nutzung
    0.51
    POSITIVE LOGITS
     tavern
    0.60
     robbers
    0.56
     spinster
    0.54
     gentleman
    0.53
     seamen
    0.52
     newspap
    0.51
     waistcoat
    0.51
     sailor
    0.50
     drunkenness
    0.50
     pigeons
    0.50
    Act Density 0.068%

    No Known Activations