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

    Neuronpedia

    INDEX
    Explanations

    city names

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     Amiens
    -1.10
     Barcelone
    -1.03
     Parigi
    -1.02
     Avignon
    -1.00
     Monfieur
    -0.97
     Rimini
    -0.96
     Rouen
    -0.95
     Bucharest
    -0.94
     Moscou
    -0.93
     Ancona
    -0.93
    POSITIVE LOGITS
    oye
    0.57
     Gran
    0.56
     Region
    0.54
     Pal
    0.53
     dest
    0.53
     El
    0.51
     Il
    0.51
     La
    0.50
     Ver
    0.50
     Bar
    0.50
    Act Density 0.121%

    No Known Activations