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

    Neuronpedia

    INDEX
    Explanations

    word pairs with a heavy emphasis on the second word

    oai_token-act-pair · gpt-3.5-turbo

    punctuation marks, specifically periods

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    thal
    -0.81
    adi
    -0.75
    iber
    -0.73
     pit
    -0.73
     overboard
    -0.72
    onga
    -0.72
    ascus
    -0.71
    oshenko
    -0.71
    itled
    -0.71
     transition
    -0.70
    POSITIVE LOGITS
     Secondly
    0.98
     Doctors
    0.96
     Whatever
    0.94
    tumblr
    0.93
    txt
    0.92
     However
    0.91
     Feel
    0.91
     Whenever
    0.90
     Whether
    0.90
     But
    0.89
    Act Density 0.970%

    No Known Activations