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

    Neuronpedia

    INDEX
    Explanations

    have/had followed by temporal, duration, or state indicators

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     been
    1.05
    been
    0.83
     BEEN
    0.82
     Been
    0.72
     ollut
    0.71
     sido
    0.68
     været
    0.66
     été
    0.63
    Been
    0.62
     olnud
    0.54
    POSITIVE LOGITS
    r
    0.64
    (
    0.60
     (
    0.59
    is
    0.46
    ergy
    0.46
     ，
    0.45
    ，
    0.45
    of
    0.44
    l
    0.42
     when
    0.42
    Act Density 0.057%

    No Known Activations