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

    Neuronpedia

    INDEX
    Explanations

    statements about being the first or pioneering in something

    oai_token-act-pair · gpt-3.5-turbo

    phrases indicating something is the "first" of its kind or instance

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    Dat
    -0.74
    love
    -0.67
     Skydragon
    -0.64
     Secrets
    -0.63
    èª
    -0.63
    query
    -0.63
    inventoryQuantity
    -0.63
    beaut
    -0.62
    ãĥ¼ãĥ
    -0.62
     Theme
    -0.61
    POSITIVE LOGITS
     ever
    0.83
    ever
    0.81
     officially
    0.72
    clusively
    0.70
     EVER
    0.68
     induct
    0.67
    OPLE
    0.64
     formally
    0.63
    itatively
    0.63
     foray
    0.63
    Act Density 0.230%

    No Known Activations