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

    Neuronpedia

    INDEX
    Explanations

    opening phrases

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    TestCase
    -0.85
     Carnaval
    -0.85
     بہ
    -0.84
     Fläche
    -0.80
    必须要
    -0.78
     uska
    -0.77
    TERM
    -0.77
     تناول
    -0.77
    🍢
    -0.77
     isotopic
    -0.76
    POSITIVE LOGITS
     attention
    1.05
     day
    0.95
     داد
    0.90
     "
    0.90
     ahead
    0.90
     local
    0.89
     predicted
    0.88
     afternoon
    0.85
     правил
    0.83
    estial
    0.82
    Act Density 0.003%

    No Known Activations