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

    Neuronpedia

    INDEX
    Explanations

    past tense auxiliary verbs, especially forms of "to have" and "to be"

    oai_token-act-pair · gemini-2.0-flash

    Legal documents

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    <bos>
    -1.02
     étoient
    -0.93
     avoient
    -0.92
     sauvages
    -0.90
     tasche
    -0.85
     métallique
    -0.84
     ögon
    -0.82
     découver
    -0.80
     asiatique
    -0.78
     Wikiseite
    -0.77
    POSITIVE LOGITS
     latter
    0.61
     last
    0.61
     previously
    0.61
    /
    0.59
    ->
    0.58
     extra
    0.57
     <<<<<<<<<<<<<<
    0.56
     otherwise
    0.56
     fris
    0.56
     etc
    0.55
    Act Density 1.259%

    No Known Activations