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

    Neuronpedia

    INDEX
    Explanations

    the word "continue" or its variations

    oai_token-act-pair · gpt-3.5-turbo

    phrases that indicate ongoing actions or conditions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ooters
    -0.70
    és
    -0.66
    ographical
    -0.63
    azeera
    -0.62
    »Ĵ
    -0.61
    osher
    -0.59
    rored
    -0.58
    otor
    -0.58
    ody
    -0.57
    anca
    -0.56
    POSITIVE LOGITS
     to
    1.06
     unab
    0.85
     ap
    0.72
     onward
    0.67
    To
    0.63
    to
    0.60
     TO
    0.60
     To
    0.59
     unchanged
    0.59
     onwards
    0.59
    Act Density 0.049%

    No Known Activations