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

    Neuronpedia

    INDEX
    Explanations

    italic styling

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ducible
    0.48
     episódio
    0.44
     rufis
    0.44
     licencia
    0.44
     antisemit
    0.44
     whakar
    0.44
     Poké
    0.44
     enig
    0.42
     arbitr
    0.42
     féidir
    0.42
    POSITIVE LOGITS
    with
    0.54
    especially
    0.54
    from
    0.53
    f
    0.52
    b
    0.50
    at
    0.48
    gary
    0.48
    man
    0.47
    MAN
    0.47
    WITH
    0.47
    Act Density 0.002%

    No Known Activations