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

    Neuronpedia

    INDEX
    Explanations

    controversial or critical statements

    oai_token-act-pair · gpt-3.5-turbo

    punctuation and formatting-related elements in the text

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    adden
    -0.60
    tones
    -0.55
    boro
    -0.54
    heit
    -0.49
    ici
    -0.48
     tongue
    -0.48
    ildo
    -0.48
     inclusion
    -0.47
    packages
    -0.47
    ell
    -0.47
    POSITIVE LOGITS
    Nevertheless
    0.70
    Nonetheless
    0.69
    âĢ¢âĢ¢
    0.67
    Meanwhile
    0.67
    Still
    0.66
    Finally
    0.65
    Likewise
    0.64
    Attempts
    0.64
    Similarly
    0.63
    ccording
    0.62
    Act Density 0.980%

    No Known Activations