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

    Neuronpedia

    INDEX
    Explanations

    in their context

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     is
    0.82
     are
    0.63
     +
    0.61
     è
    0.59
    0.58
     was
    0.58
     or
    0.56
    0
    0.55
     to
    0.55
    q
    0.54
    POSITIVE LOGITS
     timeliness
    0.97
    <unused933>
    0.96
    <unused1840>
    0.95
    <unused1855>
    0.93
    <unused323>
    0.93
    <unused282>
    0.92
    <unused652>
    0.92
    <unused2019>
    0.92
    <unused514>
    0.92
    <unused2048>
    0.91
    Act Density 4.998%

    No Known Activations