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

    Neuronpedia

    INDEX
    Explanations

    phrases related to quality or uniqueness

    oai_token-act-pair · gpt-3.5-turbo

    isolated instances of the end-of-text token

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    farious
    -0.78
    geist
    -0.67
    ems
    -0.63
    arettes
    -0.62
    agra
    -0.62
    vertisement
    -0.61
    vation
    -0.60
    pecting
    -0.59
    olesc
    -0.58
    ãĥ¼ãĥĨ
    -0.57
    POSITIVE LOGITS
     infographic
    0.75
     curated
    0.71
     versatile
    0.69
     customizable
    0.68
     flavorful
    0.66
     interactive
    0.63
     FREE
    0.63
     crochet
    0.62
     awesome
    0.61
     amazing
    0.60
    Act Density 0.929%

    No Known Activations