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

    Neuronpedia

    INDEX
    Explanations

    Transformer architecture

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     peaceful
    0.53
    izzazione
    0.51
     humane
    0.49
     serene
    0.49
    eating
    0.46
    candle
    0.46
    折り
    0.46
     flexible
    0.46
     foldable
    0.46
    enemy
    0.45
    POSITIVE LOGITS
     spawned
    0.46
    }))$
    0.46
     များ
    0.44
     مشترك
    0.43
    ómicos
    0.42
    佲
    0.42
    వ్
    0.42
    жнарод
    0.42
     свойства
    0.41
     appareils
    0.41
    Act Density 0.001%

    No Known Activations