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

    Neuronpedia

    INDEX
    Explanations

    phrases describing specific actions or methods of doing something

    oai_token-act-pair · gpt-3.5-turbo

    phrases that describe various methods or ways to achieve something

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ukong
    -0.70
    notations
    -0.66
    ventures
    -0.61
    essor
    -0.59
    iannopoulos
    -0.59
    irens
    -0.58
    Versions
    -0.58
    eor
    -0.58
    atur
    -0.58
    eus
    -0.57
    POSITIVE LOGITS
     to
    1.16
     through
    0.98
     simply
    0.83
     probably
    0.82
     undoubtedly
    0.82
     via
    0.80
     by
    0.70
     usually
    0.70
     TO
    0.68
     thru
    0.68
    Act Density 0.084%

    No Known Activations