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

    Neuronpedia

    INDEX
    Explanations

    references to measurements of agreement or consistency

    oai_token-act-pair · gpt-4.1-nano
    New Auto-Interp
    Negative Logits
     averaged
    -1.34
     ranged
    -1.25
    averaged
    -0.98
     averaging
    -0.92
     ranges
    -0.90
     averages
    -0.87
    ranges
    -0.77
     lasted
    -0.75
    Aver
    -0.75
     Ranges
    -0.74
    POSITIVE LOGITS
     ranging
    1.46
    ranging
    1.14
    \{\\
    0.60
    WEBPACK
    0.56
    RectangleBorder
    0.56
    zbęd
    0.52
     تضيفلها
    0.52
     allant
    0.51
    0.51
     hObject
    0.50
    Act Density 0.002%

    No Known Activations