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

    Neuronpedia

    INDEX
    Explanations

    personal pronouns followed by verbs indicating a comparison or a sense of exceeding something

    oai_token-act-pair · gpt-3.5-turbo

    phrases expressing personal experiences or opinions that involve counting, understanding limits, or acknowledging difficulties

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    istries
    -0.71
     Firm
    -0.61
     largeDownload
    -0.60
    lance
    -0.57
    steen
    -0.56
     Channel
    -0.56
     Parameters
    -0.56
     Eleanor
    -0.55
     Environment
    -0.54
     Commando
    -0.54
    POSITIVE LOGITS
    average
    0.85
    usual
    0.82
     usual
    0.79
    iably
    0.74
     barg
    0.71
    adequ
    0.70
     realizes
    0.69
     ever
    0.68
     sum
    0.67
     admit
    0.67
    Act Density 0.202%

    No Known Activations