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

    Neuronpedia

    INDEX
    Explanations

    explanations and examples

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ountains
    0.72
    زاده
    0.69
    ETARY
    0.66
    טע
    0.65
    ANCED
    0.64
    酿
    0.64
     blisters
    0.64
     Plants
    0.63
     doorbell
    0.62
     powered
    0.62
    POSITIVE LOGITS
    亘
    0.66
    ook
    0.65
    uto
    0.64
     kri
    0.63
     тех
    0.62
     والت
    0.60
    shall
    0.60
    రి
    0.60
     আসবেন
    0.59
     одного
    0.58
    Act Density 0.002%

    No Known Activations