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

    Neuronpedia

    INDEX
    Explanations

    text within square brackets followed by a high number

    oai_token-act-pair · gpt-3.5-turbo

    punctuation marks or brackets within a text

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    onite
    -0.78
    Ń·
    -0.75
     unwanted
    -0.74
    oe
    -0.69
     SERV
    -0.63
    ciating
    -0.63
     fatig
    -0.61
    hement
    -0.61
    ignt
    -0.61
    İĭ
    -0.61
    POSITIVE LOGITS
    âĨij
    0.74
    TPS
    0.73
     ARTICLE
    0.73
     IMAGES
    0.72
    eous
    0.71
     Cheong
    0.70
     largeDownload
    0.69
    âĨ
    0.68
    REDACTED
    0.66
     ISI
    0.64
    Act Density 0.042%

    No Known Activations