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

    Neuronpedia

    INDEX
    Explanations

    was beautifully, were incredibly, were clean

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     contextos
    0.28
     semantics
    0.28
     kontek
    0.28
     uglify
    0.28
     unamb
    0.28
    让你
    0.27
     игре
    0.27
    损伤
    0.27
    损害
    0.27
     죽
    0.27
    POSITIVE LOGITS
     welcoming
    0.30
     ourselves
    0.29
     vardı
    0.29
     buss
    0.29
     delici
    0.28
     our
    0.28
     tijdens
    0.28
     delicious
    0.28
     cleanliness
    0.28
    fresh
    0.27
    Act Density 0.006%

    No Known Activations