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

    Neuronpedia

    INDEX
    Explanations

    what's happening

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ↵
    1.27
    ↵↵
    0.97
     He
    0.82
     he
    0.79
     Movies
    0.79
    he
    0.78
    А
    0.77
     Movie
    0.77
     जिला
    0.77
     он
    0.77
    POSITIVE LOGITS
    <unused462>
    1.68
    <unused294>
    1.67
    <unused1840>
    1.66
    <unused482>
    1.65
    <unused271>
    1.64
    <unused647>
    1.64
    <unused1855>
    1.63
    <unused429>
    1.58
    <unused231>
    1.58
    <unused426>
    1.58
    Act Density 0.000%

    No Known Activations