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

    Neuronpedia

    INDEX
    Explanations

    words that, when stemmed, end with "ure" or "ar"

    oai_token-act-pair · gemini-2.0-flash

    word suffixes

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     Anſ
    -0.91
    Autoritní
    -0.90
     Monfieur
    -0.82
    <bos>
    -0.80
     Shakspeare
    -0.77
     Theſe
    -0.77
     дописавши
    -0.74
     незавершена
    -0.70
     Houſe
    -0.68
    ".
    -0.67
    POSITIVE LOGITS
    SerializeField
    0.63
     K
    0.61
     o
    0.61
    gewiesen
    0.61
     Ar
    0.59
     O
    0.58
     N
    0.57
     Gu
    0.56
     R
    0.56
     C
    0.56
    Act Density 0.819%

    No Known Activations