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

    Neuronpedia

    INDEX
    Explanations

    sentences ending in a period

    oai_token-act-pair · gpt-3.5-turbo

    sentences that provide information or summarize thoughts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     criminal
    -0.75
     elector
    -0.74
     explo
    -0.73
     defender
    -0.73
     opposing
    -0.69
     opponent
    -0.69
     accus
    -0.68
     ransom
    -0.68
     contested
    -0.68
     heir
    -0.68
    POSITIVE LOGITS
     Thankfully
    1.20
     Luckily
    1.18
     Needless
    1.17
     Eventually
    1.14
     My
    1.13
     Unfortunately
    1.12
     ðŁĻĤ
    1.11
     Then
    1.11
     Sadly
    1.10
     Fortunately
    1.09
    Act Density 0.694%

    No Known Activations