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

    Neuronpedia

    INDEX
    Explanations

    try something new or again

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    tour
    0.53
    screen
    0.52
    sufficient
    0.51
    inien
    0.47
    </strong>
    0.46
    rap
    0.46
    serve
    0.46
     Serve
    0.46
    ー
    0.45
    sym
    0.45
    POSITIVE LOGITS
     approaches
    0.75
    試
    0.68
     unsuccessfully
    0.65
    不同的
    0.64
     experimenting
    0.64
     thử
    0.61
     Approaches
    0.61
    尝试
    0.61
     hairstyles
    0.61
     alternatives
    0.60
    Act Density 0.058%

    No Known Activations