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

    Neuronpedia

    INDEX
    Explanations

    specific names

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     could
    0.66
     you
    0.63
     your
    0.63
     some
    0.63
     might
    0.63
     really
    0.62
     thinks
    0.61
     *
    0.61
     more
    0.59
      
    0.59
    POSITIVE LOGITS
     Luftwaffe
    0.83
    ポケモン
    0.76
     Minecraft
    0.73
     Pokémon
    0.71
    WWE
    0.71
     Arabidopsis
    0.70
    Minecraft
    0.70
     Warhammer
    0.69
    Pokemon
    0.68
     tokamak
    0.68
    Act Density 0.188%

    No Known Activations