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

    Neuronpedia

    INDEX
    Explanations

    had been fading

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Raised
    0.69
    अनुसूचित
    0.67
    DialogOpen
    0.67
    hoog
    0.66
    ové
    0.66
    ুব্ধ
    0.66
     обуча
    0.64
    ungkinkan
    0.64
     çıkarm
    0.64
    тбол
    0.62
    POSITIVE LOGITS
     fading
    3.20
     decay
    3.08
     fade
    2.92
     fades
    2.76
     decaying
    2.74
     decays
    2.71
     disintegr
    2.70
     disappearing
    2.65
     dying
    2.58
     faded
    2.58
    Act Density 0.771%

    No Known Activations