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

    Neuronpedia

    INDEX
    Explanations

    short, punctuated sentences

    oai_token-act-pair · gpt-3.5-turbo

    punctuation marks and sentence endings

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     tack
    -0.64
     bil
    -0.63
     submerged
    -0.63
     opponent
    -0.63
     nuts
    -0.61
     bonded
    -0.60
     criminal
    -0.60
     explo
    -0.59
     worker
    -0.58
     oun
    -0.58
    POSITIVE LOGITS
     Firstly
    1.04
     Whilst
    0.93
     Especially
    0.93
     Unless
    0.89
     Hopefully
    0.89
     Whether
    0.85
     Probably
    0.83
     Maybe
    0.83
     Literally
    0.83
     Assuming
    0.83
    Act Density 0.761%

    No Known Activations