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

    Neuronpedia

    INDEX
    Explanations

    recycling and recycled materials

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    is
    1.13
    {
    1.10
     mismas
    0.93
     mesmas
    0.93
    ن
    0.93
    ⟤
    0.91
     цер
    0.91
     drogas
    0.90
     tropas
    0.84
    ன்
    0.83
    POSITIVE LOGITS
    ל
    1.12
    ও
    1.07
    0
    1.06
    ip
    0.99
     in
    0.95
    tól
    0.90
    在
    0.90
     recycle
    0.89
    k
    0.88
     recycling
    0.87
    Act Density 0.006%

    No Known Activations