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

    Neuronpedia

    INDEX
    Explanations

    code parsing

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     itſelf
    -1.41
    been
    -1.38
     myſelf
    -1.27
     been
    -1.24
     BEEN
    -1.23
     Efq
    -1.22
    Been
    -1.16
     Monfieur
    -1.16
     Cæsar
    -1.16
     themſelves
    -1.15
    POSITIVE LOGITS
     in
    0.88
    ,
    0.79
     the
    0.75
    <eos>
    0.72
     
    0.69
     a
    0.69
     at
    0.68
     on
    0.67
     und
    0.66
     or
    0.65
    Act Density 0.008%

    No Known Activations