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

    Neuronpedia

    INDEX
    Explanations

    text enclosed in quotation marks (") and unusual characters such as Ċ and Ļ in text sequences

    oai_token-act-pair · gpt-3.5-turbo

    quoted strings or characters in code

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    arius
    -0.75
    avorite
    -0.71
    ADRA
    -0.68
     pacing
    -0.67
     lapt
    -0.65
     lumber
    -0.63
     veter
    -0.63
     puberty
    -0.63
     plateau
    -0.62
     Bang
    -0.61
    POSITIVE LOGITS
    BuyableInstoreAndOnline
    1.11
    ./
    1.01
    Hello
    0.93
    wcsstore
    0.91
    default
    0.89
    normal
    0.85
    true
    0.83
    pers
    0.83
    SELECT
    0.83
    cheat
    0.81
    Act Density 0.048%

    No Known Activations