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

    Neuronpedia

    INDEX
    Explanations

    custom dataset or code example

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    provide
    0.79
     arqué
    0.79
    skog
    0.78
    は何
    0.77
     dampen
    0.76
    ১২শ
    0.75
    ᅬ
    0.75
    pulumi
    0.75
    slash
    0.73
     হইয়া
    0.73
    POSITIVE LOGITS
    个
    0.83
    鸽
    0.80
     desserts
    0.75
    들이
    0.75
     उत्तम
    0.75
     Romans
    0.74
     INDIA
    0.73
     WTF
    0.71
     ϒ
    0.68
    ased
    0.68
    Act Density 0.001%

    No Known Activations