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

    Neuronpedia

    INDEX
    Explanations

    JSON type keys

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     redundancies
    0.38
    aminan
    0.37
    Duck
    0.35
     sanctuaries
    0.35
    redund
    0.34
     duck
    0.34
     reducir
    0.34
     presentación
    0.34
     ruido
    0.34
    shapes
    0.34
    POSITIVE LOGITS
     </
    0.35
     ওয়ে
    0.33
     Marty
    0.33
     Range
    0.33
     ranges
    0.33
     meh
    0.32
    朴
    0.32
     Тар
    0.32
     লীগের
    0.31
     Ranges
    0.31
    Act Density 0.001%

    No Known Activations