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

    Neuronpedia

    INDEX
    Explanations

    non-textual characters and noise, likely not related to meaningful patterns in the text

    oai_token-act-pair · gpt-3.5-turbo

    numerical or symbolic representations, particularly those resembling mathematical or coded expressions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Pwr
    -0.70
     Mous
    -0.67
     Breed
    -0.67
     Burke
    -0.65
    ynt
    -0.64
    esp
    -0.64
    eur
    -0.63
    priority
    -0.63
    cephal
    -0.63
    ption
    -0.63
    POSITIVE LOGITS
    assetsadobe
    0.81
    taboola
    0.80
    )=(
    0.79
     certs
    0.77
    entimes
    0.75
    artifacts
    0.74
     antioxid
    0.73
    ulkan
    0.73
     hobbies
    0.72
     pregn
    0.71
    Act Density 0.446%

    No Known Activations