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

    Neuronpedia

    INDEX
    Explanations

    <

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Class
    0.73
    DOCUMENT
    0.72
     Persian
    0.72
     Class
    0.66
    Triplet
    0.66
    GEM
    0.64
    Document
    0.64
     Pamph
    0.63
    óch
    0.62
    <0xE7>
    0.61
    POSITIVE LOGITS
    main
    0.78
     main
    0.74
     चौरा
    0.69
     aside
    0.63
    ွေ
    0.58
     いき
    0.57
    人と
    0.57
     revenue
    0.56
    form
    0.55
     endDate
    0.55
    Act Density 0.016%

    No Known Activations