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

    Neuronpedia

    INDEX
    Explanations

    hexadecimal numbers and defines in C code

    oai_token-act-pair · gemini-2.0-flash

    ff

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     three
    -1.13
    three
    -1.06
     four
    -1.03
    Three
    -1.00
     trois
    -0.98
    Four
    -0.97
     Three
    -0.90
     तीन
    -0.89
     Four
    -0.89
     tiga
    -0.89
    POSITIVE LOGITS
     milliers
    0.65
    SequentialGroup
    0.63
     thousands
    0.55
    reathe
    0.54
    กุ
    0.54
    thousands
    0.52
     MaterialApp
    0.52
     diritto
    0.51
     InputDecoration
    0.50
    chanter
    0.50
    Act Density 0.526%

    No Known Activations