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

    Neuronpedia

    INDEX
    Explanations

    Code, transcripts, medical

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     above
    -0.63
    SOUNDBITE
    -0.61
     one
    -0.60
     lower
    -0.59
     but
    -0.58
     many
    -0.58
     the
    -0.56
     and
    -0.55
     another
    -0.55
     to
    -0.54
    POSITIVE LOGITS
    protoimpl
    0.69
    monary
    0.65
     transfieras
    0.63
     muualla
    0.59
    IntoConstraints
    0.53
     المعيارى
    0.52
     oprot
    0.51
     esternos
    0.50
     فريبيس
    0.49
    ?')
    0.48
    Act Density 0.013%

    No Known Activations