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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating an alternative or additional option

    oai_token-act-pair · gpt-3.5-turbo

    sentences that convey an ending or conclusion

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     purse
    -0.76
     robber
    -0.69
    hoe
    -0.66
     ambush
    -0.65
     ransom
    -0.63
     assassin
    -0.60
    essed
    -0.59
    CVE
    -0.58
     allowance
    -0.57
     exclusively
    -0.57
    POSITIVE LOGITS
     Whether
    0.98
     Regardless
    0.98
     Aside
    0.97
     Instead
    0.97
     Especially
    0.97
     Besides
    0.95
     Whereas
    0.95
     Thankfully
    0.95
     Thanks
    0.95
     Fortunately
    0.94
    Act Density 0.627%

    No Known Activations