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

    Neuronpedia

    INDEX
    Explanations

    written descriptions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     (
    0.42
     cunning
    0.35
    സ്ഥ
    0.34
     per
    0.34
     Too
    0.33
    বিএন
    0.33
     gastronomic
    0.33
     misfortunes
    0.33
    ării
    0.33
     Bezos
    0.33
    POSITIVE LOGITS
    লিখিত
    0.46
    written
    0.44
     written
    0.40
    變成
    0.40
     вместо
    0.39
    गतान
    0.39
    smoothed
    0.39
    rendered
    0.38
    变成了
    0.38
    Written
    0.38
    Act Density 0.131%

    No Known Activations