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

    Neuronpedia

    INDEX
    Explanations

    short phrases or sentences that end with specific punctuation marks like period and colon

    oai_token-act-pair · gpt-3.5-turbo

    punctuation and sentence structure

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     eleph
    -0.89
     earthqu
    -0.81
     pione
    -0.80
    Þ
    -0.78
    ¥ŀ
    -0.77
     neighb
    -0.76
     undet
    -0.76
     experienced
    -0.73
    aditional
    -0.73
     oun
    -0.72
    POSITIVE LOGITS
    Anyway
    1.95
    Nevertheless
    1.89
    Nonetheless
    1.85
    Still
    1.84
    Advertisement
    1.79
    Moreover
    1.77
    Similarly
    1.75
    Regardless
    1.75
    Likewise
    1.74
    But
    1.71
    Act Density 0.498%

    No Known Activations