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

    Neuronpedia

    INDEX
    Explanations

    names before @

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Dé
    1.06
     Ō
    1.01
     Cómo
    0.96
     Pré
    0.95
     También
    0.95
     Sé
    0.94
     Atlético
    0.92
     UserDefaults
    0.92
     Pokémon
    0.92
     İn
    0.91
    POSITIVE LOGITS
    @
    3.02
     @
    2.11
    ...@
    1.88
    .@
    1.85
    @[
    1.75
    @@
    1.74
    @(
    1.72
    \@
    1.72
    =@
    1.70
    (@
    1.63
    Act Density 0.028%

    No Known Activations