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

    Neuronpedia

    INDEX
    Explanations
    No Explanations Found
    New Auto-Interp
    Negative Logits
    hazard
    -0.71
     Rez
    -0.67
     grave
    -0.66
     descend
    -0.63
     griev
    -0.63
     summ
    -0.63
    âĿ
    -0.61
     appropri
    -0.60
     testament
    -0.59
     exemplary
    -0.59
    POSITIVE LOGITS
    ãĢı
    0.78
    esson
    0.74
     Yao
    0.72
     Butterfly
    0.70
    ollar
    0.69
    ories
    0.68
     Corpus
    0.65
    ausible
    0.64
    itored
    0.63
    2200
    0.63
    Act Density 0.000%

    No Known Activations

    This feature has no known activations.